From Claude.Marche at lri.fr Mon Jul 17 11:11:56 2006 From: Claude.Marche at lri.fr (Claude Marche) Date: Mon Jul 17 11:11:59 2006 Subject: [icfp-discuss] small vs large teams Message-ID: <17595.43196.415081.165334@serveur9-10.lri.fr> Dear all, I've just noticed the new rule of this year, stating that teams of more than 4 members cannot win the 1st place. Why is it so? Not only I don't understand why there is such a rule, but it may cause several problems/complaints/... How the judges can check if a team is not of more than 4 ? I participated to the last 3 competitions, within a team where there were not more that 4 programmers, but several other people gave ideas, such as about good strategies to follow. As far as I understand this cannot be considered a small team. So it means that to compete for the first place this year, we must not discuss with anybody except the programmers. But if we do it anyway, nobody would know it... Anyway, we would like to participate to this contest for the fun, and we don't see any point why we should not talk about it to anybody. But knowing that we cannot win the first place even before knowing the task, is really disappointing. Any comments ? - Claude, from The Caml Riders team -- | Claude March? | mailto:Claude.Marche@lri.fr | | LRI - B?t. 490 | http://www.lri.fr/~marche/ | | Universit? de Paris-Sud | phoneto: +33 1 69 15 64 85 | | F-91405 ORSAY Cedex | faxto: +33 1 69 15 65 86 | From james at grayproductions.net Mon Jul 17 11:30:58 2006 From: james at grayproductions.net (James Edward Gray II) Date: Mon Jul 17 11:31:03 2006 Subject: [icfp-discuss] small vs large teams In-Reply-To: <17595.43196.415081.165334@serveur9-10.lri.fr> References: <17595.43196.415081.165334@serveur9-10.lri.fr> Message-ID: On Jul 17, 2006, at 10:11 AM, Claude Marche wrote: > How the judges can check if a team is not of more than 4 ? I imagine we're on the honor system for that. (Just guessing.) > Anyway, we would like to participate to this contest for the fun, > and we don't see any point why we should not talk about it to > anybody. But knowing that we cannot win the first place even before > knowing the task, is really disappointing. > > Any comments ? I'm pretty sure most of us doing this for fun, aren't too worried about winning first prize. ;) Seriously, this contest has been won in the past by nine man teams from MIT including PHDed professors that invented their own programming language. That's pretty different from my team of four friends who have given it our best these last two years but are in no danger of taking first anytime soon. :) I welcome the separation of those two groups. Heck, I wish it was a bigger separation. Don't avoid talking to anyone you want. That's just silly. If it lands you in the big group category, well, then my team wishes you all the luck at nailing second. ;) James Edward Gray II From tom7 at cs.cmu.edu Mon Jul 17 11:37:42 2006 From: tom7 at cs.cmu.edu (Tom Murphy) Date: Mon Jul 17 11:32:16 2006 Subject: [icfp-discuss] small vs large teams In-Reply-To: <17595.43196.415081.165334@serveur9-10.lri.fr> References: <17595.43196.415081.165334@serveur9-10.lri.fr> Message-ID: <44BBAEC6.6060407@cs.cmu.edu> Claude, We've participated in the contest for several years, and I understand your concern. It's not a decision we made lightly. Unfortunately because the task hasn't been announced yet, it's difficult to give our reasoning now. If the rule doesn't make sense after you start on the task, I'll be happy to explain the reasons more openly at that point! We cannot check the size of a team, but it would be considered dishonest (that is, cheating) to have more members than you say you have. There are other forms of cheating we can't check too, so some parts of the contest rely on the honor of its participants. Since most teams do not win first place, my advice is just to pick a team size that you will have fun participating with. If you like, you can always start in the small team division and then change your mind after the contest begins. Tom Claude Marche wrote: > Dear all, > > I've just noticed the new rule of this year, stating that teams of > more than 4 members cannot win the 1st place. Why is it so? > Not only I don't understand why there is such a rule, but it may cause > several problems/complaints/... > > How the judges can check if a team is not of more than 4 ? > > I participated to the last 3 competitions, within a team where there > were not more that 4 programmers, but several other people gave ideas, > such as about good strategies to follow. As far as I understand this > cannot be considered a small team. So it means that to compete for the > first place this year, we must not discuss with anybody except the > programmers. But if we do it anyway, nobody would know it... Anyway, > we would like to participate to this contest for the fun, and we don't > see any point why we should not talk about it to anybody. But knowing > that we cannot win the first place even before knowing the task, is > really disappointing. > > Any comments ? > > - Claude, from The Caml Riders team From rob.hunter at gmail.com Mon Jul 17 13:12:45 2006 From: rob.hunter at gmail.com (Rob Hunter) Date: Mon Jul 17 13:12:49 2006 Subject: [icfp-discuss] small vs large teams In-Reply-To: <44BBAEC6.6060407@cs.cmu.edu> References: <17595.43196.415081.165334@serveur9-10.lri.fr> <44BBAEC6.6060407@cs.cmu.edu> Message-ID: I too am suspicious of the new rule. My two cents is that I've always found it a very appealing part of the contest that there were essentially _no rules_. The contest is to solve the problem as best as you can, and beyond that use whatever tools you deem best--you choose the best programming languages, the best hardware, and of course the best people. Larger teams have a large number of brains and more typists, but small teams have fewer communication burdens. Which is better? Let's have a look at the history: Last year's first prize went to a Haskell team of two. 2004's first place was a team of four. 2003's first place was a team of one! 2002's first place was a team of three. 2001's first place was a team of two. 2000's was a team of four. I have no data on 1999. It isn't until we get to 1998 (the very first contest) that we have a first place team that would be considered a "large team" (eight members). The evidence does not seem to motivate any need for this new rule (in fact, I would argue, quite the opposite!). Furthermore, it doesn't seem too late to change it. Claude's points are valid too: It's certainly great to be able to talk with people about ideas (that's the academic spirit after all) without worrying about being disqualified from a sweet first place victory. Thanks for listening. --rob On 7/17/06, Tom Murphy wrote: > Claude, > > We've participated in the contest for several years, and I understand > your concern. It's not a decision we made lightly. Unfortunately because > the task hasn't been announced yet, it's difficult to give our reasoning > now. If the rule doesn't make sense after you start on the task, I'll be > happy to explain the reasons more openly at that point! > > We cannot check the size of a team, but it would be considered dishonest > (that is, cheating) to have more members than you say you have. There > are other forms of cheating we can't check too, so some parts of the > contest rely on the honor of its participants. > > Since most teams do not win first place, my advice is just to pick a > team size that you will have fun participating with. If you like, you > can always start in the small team division and then change your mind > after the contest begins. > > Tom > > Claude Marche wrote: > > Dear all, > > > > I've just noticed the new rule of this year, stating that teams of > > more than 4 members cannot win the 1st place. Why is it so? > > Not only I don't understand why there is such a rule, but it may cause > > several problems/complaints/... > > > > How the judges can check if a team is not of more than 4 ? > > > > I participated to the last 3 competitions, within a team where there > > were not more that 4 programmers, but several other people gave ideas, > > such as about good strategies to follow. As far as I understand this > > cannot be considered a small team. So it means that to compete for the > > first place this year, we must not discuss with anybody except the > > programmers. But if we do it anyway, nobody would know it... Anyway, > > we would like to participate to this contest for the fun, and we don't > > see any point why we should not talk about it to anybody. But knowing > > that we cannot win the first place even before knowing the task, is > > really disappointing. > > > > Any comments ? > > > > - Claude, from The Caml Riders team > > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > From tom7 at cs.cmu.edu Mon Jul 17 14:13:42 2006 From: tom7 at cs.cmu.edu (Tom Murphy) Date: Mon Jul 17 14:13:45 2006 Subject: [icfp-discuss] small vs large teams In-Reply-To: Message-ID: I agree that unlimited team size is a nice feature of the contest, which is why we do allow teams to be any size. The small/large divisions exist mainly so that a team can compare itself to similar teams--the first place rule will only affect a tiny number of teams, if any. While you're right that past contests have usually been dominated by small teams, we believe that this year's contest is very different. It's not really feasible to change the rules four days before the contest (nor would it be fair to those that have registered over the past month), so why not try it with us and we can all see how it plays out? Tom > I too am suspicious of the new rule. My two cents is that I've always > found it a very appealing part of the contest that there were > essentially _no rules_. The contest is to solve the problem as best > as you can, and beyond that use whatever tools you deem best--you > choose the best programming languages, the best hardware, and of > course the best people. > > Larger teams have a large number of brains and more typists, but small > teams have fewer communication burdens. Which is better? Let's have > a look at the history: > > Last year's first prize went to a Haskell team of two. 2004's first > place was a team of four. 2003's first place was a team of one! > 2002's first place was a team of three. 2001's first place was a team > of two. 2000's was a team of four. I have no data on 1999. It isn't > until we get to 1998 (the very first contest) that we have a first > place team that would be considered a "large team" (eight members). > > The evidence does not seem to motivate any need for this new rule (in > fact, I would argue, quite the opposite!). Furthermore, it doesn't > seem too late to change it. Claude's points are valid too: It's > certainly great to be able to talk with people about ideas (that's the > academic spirit after all) without worrying about being disqualified > from a sweet first place victory. > > Thanks for listening. > --rob > > On 7/17/06, Tom Murphy wrote: > > Claude, > > > > We've participated in the contest for several years, and I understand > > your concern. It's not a decision we made lightly. Unfortunately because > > the task hasn't been announced yet, it's difficult to give our reasoning > > now. If the rule doesn't make sense after you start on the task, I'll be > > happy to explain the reasons more openly at that point! > > > > We cannot check the size of a team, but it would be considered dishonest > > (that is, cheating) to have more members than you say you have. There > > are other forms of cheating we can't check too, so some parts of the > > contest rely on the honor of its participants. > > > > Since most teams do not win first place, my advice is just to pick a > > team size that you will have fun participating with. If you like, you > > can always start in the small team division and then change your mind > > after the contest begins. > > > > Tom > > From rob.hunter at gmail.com Mon Jul 17 14:31:58 2006 From: rob.hunter at gmail.com (Rob Hunter) Date: Mon Jul 17 14:32:00 2006 Subject: [icfp-discuss] small vs large teams In-Reply-To: References: Message-ID: I'll definitely be there to try it! And thanks for your thoughtful responses. --rob On 7/17/06, Tom Murphy wrote: > > I agree that unlimited team size is a nice feature of the contest, which > is why we do allow teams to be any size. The small/large divisions exist > mainly so that a team can compare itself to similar teams--the first place > rule will only affect a tiny number of teams, if any. While you're right > that past contests have usually been dominated by small teams, we believe > that this year's contest is very different. It's not really feasible to > change the rules four days before the contest (nor would it be fair to > those that have registered over the past month), so why not try it with us > and we can all see how it plays out? > > Tom > > > I too am suspicious of the new rule. My two cents is that I've always > > found it a very appealing part of the contest that there were > > essentially _no rules_. The contest is to solve the problem as best > > as you can, and beyond that use whatever tools you deem best--you > > choose the best programming languages, the best hardware, and of > > course the best people. > > > > Larger teams have a large number of brains and more typists, but small > > teams have fewer communication burdens. Which is better? Let's have > > a look at the history: > > > > Last year's first prize went to a Haskell team of two. 2004's first > > place was a team of four. 2003's first place was a team of one! > > 2002's first place was a team of three. 2001's first place was a team > > of two. 2000's was a team of four. I have no data on 1999. It isn't > > until we get to 1998 (the very first contest) that we have a first > > place team that would be considered a "large team" (eight members). > > > > The evidence does not seem to motivate any need for this new rule (in > > fact, I would argue, quite the opposite!). Furthermore, it doesn't > > seem too late to change it. Claude's points are valid too: It's > > certainly great to be able to talk with people about ideas (that's the > > academic spirit after all) without worrying about being disqualified > > from a sweet first place victory. > > > > Thanks for listening. > > --rob > > > > On 7/17/06, Tom Murphy wrote: > > > Claude, > > > > > > We've participated in the contest for several years, and I understand > > > your concern. It's not a decision we made lightly. Unfortunately because > > > the task hasn't been announced yet, it's difficult to give our reasoning > > > now. If the rule doesn't make sense after you start on the task, I'll be > > > happy to explain the reasons more openly at that point! > > > > > > We cannot check the size of a team, but it would be considered dishonest > > > (that is, cheating) to have more members than you say you have. There > > > are other forms of cheating we can't check too, so some parts of the > > > contest rely on the honor of its participants. > > > > > > Since most teams do not win first place, my advice is just to pick a > > > team size that you will have fun participating with. If you like, you > > > can always start in the small team division and then change your mind > > > after the contest begins. > > > > > > Tom > > > > > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > From Claude.Marche at lri.fr Tue Jul 18 09:42:09 2006 From: Claude.Marche at lri.fr (Claude Marche) Date: Tue Jul 18 09:42:12 2006 Subject: [icfp-discuss] small vs large teams In-Reply-To: <44BBAEC6.6060407@cs.cmu.edu> References: <17595.43196.415081.165334@serveur9-10.lri.fr> <44BBAEC6.6060407@cs.cmu.edu> Message-ID: <17596.58673.90804.877606@serveur9-10.lri.fr> >>>>> "Tom" == Tom Murphy writes: Tom> Claude, Tom> We've participated in the contest for several years, and I understand Tom> your concern. It's not a decision we made lightly. Unfortunately because Tom> the task hasn't been announced yet, it's difficult to give our reasoning Tom> now. If the rule doesn't make sense after you start on the task, I'll be Tom> happy to explain the reasons more openly at that point! OK, thanks for the answer. So we'll wait until Friday to understand why computational archeolinguistics is easier to study by a group of at least five members ;-) - Claude -- | Claude March? | mailto:Claude.Marche@lri.fr | | LRI - B?t. 490 | http://www.lri.fr/~marche/ | | Universit? de Paris-Sud | phoneto: +33 1 69 15 64 85 | | F-91405 ORSAY Cedex | faxto: +33 1 69 15 65 86 | From james at grayproductions.net Tue Jul 18 11:37:17 2006 From: james at grayproductions.net (James Edward Gray II) Date: Tue Jul 18 11:37:23 2006 Subject: [icfp-discuss] Materials Available When Message-ID: <425D52AC-7CF8-40FB-8468-C15B33EF8E34@grayproductions.net> I have noticed that the site says, "A few days prior to the start of the contest, we will make some required materials available on this website." I was just curious about when we should expect those to appear? I was taught a "few" is three or more and today is three days before, by my count. ;) (I'm kidding around here, in case it's not obvious. I'm really more curious than impatient.) James Edward Gray II P.S. While we are bugging you with all these pre-contest curiosities, let me take this chance to thank the contest organizers for all the hard work they are doing, have done, and will do putting this together. It has to be an amazing amount of work to pull something like this off. You guys are our heroes! From ccasingh at andrew.cmu.edu Tue Jul 18 11:55:30 2006 From: ccasingh at andrew.cmu.edu (Big Chris) Date: Tue Jul 18 11:55:39 2006 Subject: [icfp-discuss] Materials Available When In-Reply-To: <425D52AC-7CF8-40FB-8468-C15B33EF8E34@grayproductions.net> References: <425D52AC-7CF8-40FB-8468-C15B33EF8E34@grayproductions.net> Message-ID: James, > I have noticed that the site says, "A few days prior to the start of the > contest, we will make some required materials available on this website." I > was just curious about when we should expect those to appear? We plan to make some materials available tomorrow. Look for information on the website in the afternoon. We will also send out an announcement on the icfpcontest-announce list. > I was taught a "few" is three or more and today is three days before, by my > count. ;) (I'm kidding around here, in case it's not obvious. I'm really > more curious than impatient.) :) > P.S. While we are bugging you with all these pre-contest curiosities, let me > take this chance to thank the contest organizers for all the hard work they > are doing, have done, and will do putting this together. It has to be an > amazing amount of work to pull something like this off. You guys are our > heroes! Thanks for the kind words. It's a lot of work, but we've been enjoying it quite a bit. We hope you all will enjoy it as much as we have already! --Chris From chad.harrington at gmail.com Tue Jul 18 17:50:28 2006 From: chad.harrington at gmail.com (Chad Harrington) Date: Tue Jul 18 17:50:31 2006 Subject: [icfp-discuss] Materials Available When In-Reply-To: <425D52AC-7CF8-40FB-8468-C15B33EF8E34@grayproductions.net> References: <425D52AC-7CF8-40FB-8468-C15B33EF8E34@grayproductions.net> Message-ID: <5f680c10607181450xc81a875p5af0994a9242e1ee@mail.gmail.com> On 7/18/06, James Edward Gray II wrote: > > > P.S. While we are bugging you with all these pre-contest curiosities, > let me take this chance to thank the contest organizers for all the > hard work they are doing, have done, and will do putting this > together. It has to be an amazing amount of work to pull something > like this off. You guys are our heroes! I'd like to second that. If past contests are an accurate indicator, there will be various vocal individuals who are unhappy with one aspect of the contest or other. Please take comfort that the majority of the participants are very grateful for the time you have put in (and will put in) to provide this great service. It is especially noble of you, since you get no compensation for your efforts. Hats off to the organizers! Chad Harrington _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > -- Chad Harrington chad.harrington@gmail.com -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060718/485344a5/attachment.html From flyfree at gmail.com Tue Jul 18 23:25:55 2006 From: flyfree at gmail.com (Zhenzhong Xu) Date: Tue Jul 18 23:25:58 2006 Subject: [icfp-discuss] Pre-contest Language statistics Message-ID: <7b2bf4900607182025g1c4b2167i2e9710d31f01f3ec@mail.gmail.com> I am really getting excited about this year's contest. I am also curious how is this year's language statistics going to turn out. I think we should do a pre-contest poll on which language you'll most likely to use before you know *any information* about the contest. Personally, I just started learning Haskell about 3 weeks ago, I am not sure I would be able to use haskell for this year's contest, mostly like I'll use C#. (unfortunately C# has never been a popular choice for ICFP contests, but anyway let's see how it turns out this year). If you'd like, just post the language you'll mostly likely to use here. Yay, go go .net developers! let's proof we are not stinky after all ^_^ Zhenzhong Xu From geepokey at yahoo.com Wed Jul 19 01:42:57 2006 From: geepokey at yahoo.com (Gee Pokey) Date: Wed Jul 19 01:43:00 2006 Subject: [icfp-discuss] Pre-contest Language statistics In-Reply-To: Message-ID: <20060719054257.88880.qmail@web37215.mail.mud.yahoo.com> We are planning to use a combination of Python and C++ - depending on the performance requirements of the solution. My python world simulator for the Ant contest ran ~1000 times slower than the C ones. :/ -Gumby of Canivsar From chad.harrington at gmail.com Wed Jul 19 03:06:49 2006 From: chad.harrington at gmail.com (Chad Harrington) Date: Wed Jul 19 03:06:50 2006 Subject: [icfp-discuss] Pre-contest Language statistics In-Reply-To: <7b2bf4900607182025g1c4b2167i2e9710d31f01f3ec@mail.gmail.com> References: <7b2bf4900607182025g1c4b2167i2e9710d31f01f3ec@mail.gmail.com> Message-ID: <5f680c10607190006i7feacb45x4ba42cab30903d75@mail.gmail.com> On 7/18/06, Zhenzhong Xu wrote: > > > If you'd like, just post the language you'll mostly likely to use here. I'll be using Python with a possibility of C if necessary. -- Chad Harrington chad.harrington@gmail.com -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060719/9da42977/attachment.html From clive at xtra.co.nz Wed Jul 19 03:39:55 2006 From: clive at xtra.co.nz (Clive Gifford) Date: Wed Jul 19 03:40:51 2006 Subject: [icfp-discuss] Pre-contest Language statistics References: <7b2bf4900607182025g1c4b2167i2e9710d31f01f3ec@mail.gmail.com> <5f680c10607190006i7feacb45x4ba42cab30903d75@mail.gmail.com> Message-ID: <001701c6ab06$8c8cd0b0$0301a8c0@Owhango1> Python (+ helpers) is looking good so far... :-) I'm also planning to use Python primarily, plus Pyrex and perhaps C for performance critical extensions, etc. Clive From christophe.poucet at gmail.com Wed Jul 19 03:50:13 2006 From: christophe.poucet at gmail.com (Christophe Poucet) Date: Wed Jul 19 03:50:19 2006 Subject: [icfp-discuss] Pre-contest Language statistics In-Reply-To: <001701c6ab06$8c8cd0b0$0301a8c0@Owhango1> References: <7b2bf4900607182025g1c4b2167i2e9710d31f01f3ec@mail.gmail.com> <5f680c10607190006i7feacb45x4ba42cab30903d75@mail.gmail.com> <001701c6ab06$8c8cd0b0$0301a8c0@Owhango1> Message-ID: <44BDE435.30002@gmail.com> We're planning to use Haskell for the task :) Cheers Clive Gifford wrote: > Python (+ helpers) is looking good so far... :-) > > I'm also planning to use Python primarily, plus Pyrex and perhaps C for > performance critical extensions, etc. > > Clive > > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > From galimberti at gmail.com Wed Jul 19 07:15:13 2006 From: galimberti at gmail.com (seba galimba) Date: Wed Jul 19 07:15:21 2006 Subject: [icfp-discuss] Pre-contest Language statistics In-Reply-To: <44BDE435.30002@gmail.com> References: <7b2bf4900607182025g1c4b2167i2e9710d31f01f3ec@mail.gmail.com> <5f680c10607190006i7feacb45x4ba42cab30903d75@mail.gmail.com> <001701c6ab06$8c8cd0b0$0301a8c0@Owhango1> <44BDE435.30002@gmail.com> Message-ID: <2e5a41ff0607190415v718c9649nd8fec170bcfe7c58@mail.gmail.com> HASKELL !!! =^p On 7/19/06, Christophe Poucet wrote: > > We're planning to use Haskell for the task :) > > Cheers > > Clive Gifford wrote: > > Python (+ helpers) is looking good so far... :-) > > > > I'm also planning to use Python primarily, plus Pyrex and perhaps C for > > performance critical extensions, etc. > > > > Clive > > > > _______________________________________________ > > icfpcontest-discuss mailing list > > icfpcontest-discuss@lists.andrew.cmu.edu > > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > > > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060719/4dc7b818/attachment.html From janto at sun.ac.za Wed Jul 19 07:36:52 2006 From: janto at sun.ac.za (Janto Dreijer) Date: Wed Jul 19 07:36:55 2006 Subject: [icfp-discuss] Pre-contest Language statistics References: <7b2bf4900607182025g1c4b2167i2e9710d31f01f3ec@mail.gmail.com> Message-ID: <013001c6ab27$a15e8bc0$5edde892@sun.ac.za> Can we please not do this directly on the mailing list. It's going to generate a lot of messages and swamp other threads. It's also going to be difficult to interpret the results into anything meaningful. Is there a chance the organisers might find time to create a basic survey/poll on the website? (My apologies if this is the second time you get this message. I messed up my subscription.) ----- Original Message ----- From: "Zhenzhong Xu" To: Sent: Wednesday, July 19, 2006 5:25 AM Subject: [icfp-discuss] Pre-contest Language statistics >I am really getting excited about this year's contest. I am also > curious how is this year's language statistics going to turn out. I > think we should do a pre-contest poll on which language you'll most > likely to use before you know *any information* about the contest. > > Personally, I just started learning Haskell about 3 weeks ago, I am > not sure I would be able to use haskell for this year's contest, > mostly like I'll use C#. (unfortunately C# has never been a popular > choice for ICFP contests, but anyway let's see how it turns out this > year). > > If you'd like, just post the language you'll mostly likely to use here. > > Yay, go go .net developers! let's proof we are not stinky after all ^_^ > > Zhenzhong Xu > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > > From kheiron at citromail.hu Wed Jul 19 07:55:52 2006 From: kheiron at citromail.hu (=?ISO-8859-2?Q?Endrey_M=E1rk?=) Date: Wed Jul 19 07:55:56 2006 Subject: [icfp-discuss] Pre-contest Language statistics Message-ID: <20060719115552.26746.qmail@server02.citromail.hu> I shall use Haskell, I have used it for three years. By the way, the first functional language in my life (which took me to Haskell and made me love it) was pure combinatory logic (CL). Three years ago, I implemented a CL interpreter (in C++), so that I be able to play with it. The biggest program I have ever written in combinatory logic is a quine (self-replicating program) ``which writes out its own sourcecode'' (i.e. which evaluates to its own term representation). Now I use Haskell for almost everything -- except for some mathematical logic funny things like quine, metacircular interpreter etc., such things I try to do in pure CL. Endrey Mark _________________________________________ H?rkeres?.hu - Mindig friss h?rek, toplist?k, szt?rpletyk?k. A legfontosabb 70 h?rforr?s k?zel 2.500 cikke naponta!Ha egy lapon akarsz mindent ?ttekinteni KLIKK IDE! - http://www.hirkereso.hu/ -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060719/2e660ffb/attachment.html From johannes at albapasser.de Wed Jul 19 08:04:54 2006 From: johannes at albapasser.de (Johannes) Date: Wed Jul 19 08:04:59 2006 Subject: [icfp-discuss] Pre-contest Language statistics In-Reply-To: <20060719115552.26746.qmail@server02.citromail.hu> References: <20060719115552.26746.qmail@server02.citromail.hu> Message-ID: <44BE1FE6.6080202@albapasser.de> We will use Haskell, but I think the choice of language is largely irrelevant (as in any software project above a certain size). What counts is to have a clear model of the problem and a good plan for implementation and testing. Which we don't have, at this moment :-) I agree with Claude: the "large team cannot win" rule looks rather strange. It cannot be checked, and I don't see how it is motivated. From kheiron at citromail.hu Wed Jul 19 08:26:48 2006 From: kheiron at citromail.hu (=?ISO-8859-2?Q?Endrey_M=E1rk?=) Date: Wed Jul 19 08:26:51 2006 Subject: [icfp-discuss] Sorry Message-ID: <20060719122648.4634.qmail@server02.citromail.hu> Sorry for my incidental disregarding Your ask. It came while I was writing my answer, so I could read it only after I sent my letter. Endrey Mark-- Eredeti ?zenet --Felad?: Janto Dreijer <janto@sun.ac.za>C?mzett: Discussion of the 2006 ICFP Programming Contest <icfpcontest-discuss@lists.andrew.cmu.edu>M?solat: Elk?ldve: 13:37T?ma: Re: [icfp-discuss] Pre-contest Language statisticsCan we please not do this directly on the mailing list. It's going to generate a lot of messages and swamp other threads. It's also going to be difficult to interpret the results into anything meaningful. Is there a chance the organisers might find time to create a basic survey/poll on the website? (My apologies if this is the second time you get this message. I messed up my subscription.) _________________________________________ H?rkeres?.hu - Mindig friss h?rek, toplist?k, szt?rpletyk?k. A legfontosabb 70 h?rforr?s k?zel 2.500 cikke naponta!Ha egy lapon akarsz mindent ?ttekinteni KLIKK IDE! - http://www.hirkereso.hu/ -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060719/4fb87a2b/attachment.html From james at grayproductions.net Wed Jul 19 09:10:42 2006 From: james at grayproductions.net (James Edward Gray II) Date: Wed Jul 19 09:10:47 2006 Subject: [icfp-discuss] Pre-contest Language statistics In-Reply-To: <7b2bf4900607182025g1c4b2167i2e9710d31f01f3ec@mail.gmail.com> References: <7b2bf4900607182025g1c4b2167i2e9710d31f01f3ec@mail.gmail.com> Message-ID: On Jul 18, 2006, at 10:25 PM, Zhenzhong Xu wrote: > If you'd like, just post the language you'll mostly likely to use > here. Our two main programmers are Rubyists, so it's Ruby for us. Another member of the team knows a little Ruby and may be able to help out here and there. We have one Perl guy on the team too, but I imagine he'll be more of an idea man. James Edward Gray II From james at grayproductions.net Wed Jul 19 09:13:01 2006 From: james at grayproductions.net (James Edward Gray II) Date: Wed Jul 19 09:13:05 2006 Subject: [icfp-discuss] Pre-contest Language statistics In-Reply-To: <013001c6ab27$a15e8bc0$5edde892@sun.ac.za> References: <7b2bf4900607182025g1c4b2167i2e9710d31f01f3ec@mail.gmail.com> <013001c6ab27$a15e8bc0$5edde892@sun.ac.za> Message-ID: <72E4FA13-B882-4B95-A342-FD2F78E1B5AF@grayproductions.net> On Jul 19, 2006, at 6:36 AM, Janto Dreijer wrote: > Can we please not do this directly on the mailing list. This list is for general discussion about the contest, so this seems very on topic to me. Any important messages should come across the announcement list, and thus not be "swamped." James Edward Gray II From james at grayproductions.net Wed Jul 19 09:15:24 2006 From: james at grayproductions.net (James Edward Gray II) Date: Wed Jul 19 09:15:29 2006 Subject: [icfp-discuss] Pre-contest Language statistics In-Reply-To: <44BE1FE6.6080202@albapasser.de> References: <20060719115552.26746.qmail@server02.citromail.hu> <44BE1FE6.6080202@albapasser.de> Message-ID: <250A426D-51FB-4CEF-99EA-ACF3E2FD70D1@grayproductions.net> On Jul 19, 2006, at 7:04 AM, Johannes wrote: > but I think the choice of language > is largely irrelevant (as in any software project above > a certain size). What counts is to have a clear model > of the problem and a good plan for implementation and testing. Wise words. James Edward Gray II From jjmaes at gmail.com Wed Jul 19 09:30:11 2006 From: jjmaes at gmail.com (James Darwin Maes The First (a javatarian)) Date: Wed Jul 19 09:30:14 2006 Subject: [icfp-discuss] Pre-contest Language statistics In-Reply-To: <250A426D-51FB-4CEF-99EA-ACF3E2FD70D1@grayproductions.net> References: <20060719115552.26746.qmail@server02.citromail.hu> <44BE1FE6.6080202@albapasser.de> <250A426D-51FB-4CEF-99EA-ACF3E2FD70D1@grayproductions.net> Message-ID: Right on ol'chap! Well done indeed. -JDM (the first) - an Al-Gore-Rhythm On 7/19/06, James Edward Gray II wrote: > > On Jul 19, 2006, at 7:04 AM, Johannes wrote: > > > but I think the choice of language > > is largely irrelevant (as in any software project above > > a certain size). What counts is to have a clear model > > of the problem and a good plan for implementation and testing. > > Wise words. > > James Edward Gray II > > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060719/6e046603/attachment.html From muranushi at gmail.com Wed Jul 19 21:41:27 2006 From: muranushi at gmail.com (Takayuki Muranushi) Date: Wed Jul 19 21:41:30 2006 Subject: [icfp-discuss] Contest Materials added in 19 Jul Message-ID: You may have noticed that contest material codex.umz had been added on the contest site. However, I couldn't find out what the file format ".umz" is. Is this original file format for this contest? I'm also not sure about what "The codex" exactly means. It translates "an old handwritten book." Maybe this is a story related flavour, or is it some sort of technical term? Does anybody have some clue? One possible answer is --- those stay mystery until the contest begins :) Takayuki Muranushi From flyfree at gmail.com Wed Jul 19 21:49:28 2006 From: flyfree at gmail.com (Zhenzhong Xu) Date: Wed Jul 19 21:49:33 2006 Subject: [icfp-discuss] Contest Materials added in 19 Jul In-Reply-To: References: Message-ID: <7b2bf4900607191849v6172d85dm4519be1eab7c7990@mail.gmail.com> I am guessing we will be looking at a "decode a alien language" task. Z. On 7/19/06, Takayuki Muranushi wrote: > You may have noticed that contest material codex.umz had been added on > the contest site. > > However, I couldn't find out what the file format ".umz" is. Is this > original file format for this contest? > I'm also not sure about what "The codex" exactly means. It translates > "an old handwritten book." Maybe this is a story related flavour, or > is it some sort of technical term? > > Does anybody have some clue? > One possible answer is --- those stay mystery until the contest begins :) > > Takayuki Muranushi > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > From james at grayproductions.net Wed Jul 19 22:37:16 2006 From: james at grayproductions.net (James Edward Gray II) Date: Wed Jul 19 22:37:22 2006 Subject: [icfp-discuss] Contest Materials added in 19 Jul In-Reply-To: References: Message-ID: <1E306FAD-C742-4E64-8965-B50D424CABD3@grayproductions.net> On Jul 19, 2006, at 8:41 PM, Takayuki Muranushi wrote: > You may have noticed that contest material codex.umz had been added on > the contest site. When I heard they were releasing a file early, I had two guesses at the reasoning: * It was software we needed time to get running in our environment * It was huge and they were worried about download sizes Now that I've seen it, I'm revising my guesswork: I think it's a teaser. ;) James Edward Gray II P.S. Whatever the intention, it's surely evil! I have a deadline at work tomorrow I would surely be fired for not making no matter how much I tried to explain about needing to decipher "the codex"... :D From Charles.Bouillaguet at crans.org Wed Jul 19 23:01:42 2006 From: Charles.Bouillaguet at crans.org (Charles Bouillaguet) Date: Wed Jul 19 23:08:10 2006 Subject: [icfp-discuss] Pre-contest Language statistics In-Reply-To: <7b2bf4900607182025g1c4b2167i2e9710d31f01f3ec@mail.gmail.com> References: <7b2bf4900607182025g1c4b2167i2e9710d31f01f3ec@mail.gmail.com> Message-ID: <1153364503.9342.10.camel@zblup> Ocaml ! On Tue, 2006-07-18 at 22:25 -0500, Zhenzhong Xu wrote: > I am really getting excited about this year's contest. I am also > curious how is this year's language statistics going to turn out. I > think we should do a pre-contest poll on which language you'll most > likely to use before you know *any information* about the contest. From clive at xtra.co.nz Wed Jul 19 23:09:44 2006 From: clive at xtra.co.nz (Clive Gifford) Date: Wed Jul 19 23:10:42 2006 Subject: [icfp-discuss] Preliminary analysis reveals flaw in the codex... References: <1E306FAD-C742-4E64-8965-B50D424CABD3@grayproductions.net> Message-ID: <002301c6aba9$f5f4ed50$0301a8c0@Owhango1> I've done some preliminary analysis and believe the file is probably incorrect. After all, it doesn't include the string "xyzzy"... and clearly this *should* have in there somewhere... Waiting now for the updated version to be posted. :-) From tony.gies at gmail.com Wed Jul 19 23:22:06 2006 From: tony.gies at gmail.com (Tony Gies) Date: Wed Jul 19 23:22:09 2006 Subject: [icfp-discuss] Preliminary analysis reveals flaw in the codex... In-Reply-To: <002301c6aba9$f5f4ed50$0301a8c0@Owhango1> References: <1E306FAD-C742-4E64-8965-B50D424CABD3@grayproductions.net> <002301c6aba9$f5f4ed50$0301a8c0@Owhango1> Message-ID: <31d4b4690607192022l60bd562dic40e2377f302f5b3@mail.gmail.com> On 7/19/06, Clive Gifford wrote: > I've done some preliminary analysis and believe the file is probably > incorrect. After all, it doesn't include the string "xyzzy"... and clearly > this *should* have in there somewhere... Also missing are "etaoin shrdlu", "plugh". I, for one, expected better of the organizers. tgies From clive at xtra.co.nz Wed Jul 19 23:40:31 2006 From: clive at xtra.co.nz (Clive Gifford) Date: Wed Jul 19 23:41:33 2006 Subject: [icfp-discuss] Preliminary analysis reveals flaw in the codex... References: <1E306FAD-C742-4E64-8965-B50D424CABD3@grayproductions.net><002301c6aba9$f5f4ed50$0301a8c0@Owhango1> <31d4b4690607192022l60bd562dic40e2377f302f5b3@mail.gmail.com> Message-ID: <003001c6abae$462a3bf0$0301a8c0@Owhango1> You're absolutely right Tony. Thanks for pointing that out. I'm now considering changing from using Python to using ETA instead. I'd recommend as many other contestants as possible consider using ETA also. See http://www.miketaylor.org.uk/tech/eta/doc/ for more details. ----- Original Message ----- From: "Tony Gies" To: "Discussion of the 2006 ICFP Programming Contest" Sent: Thursday, July 20, 2006 3:22 PM Subject: Re: [icfp-discuss] Preliminary analysis reveals flaw in the codex... On 7/19/06, Clive Gifford wrote: > I've done some preliminary analysis and believe the file is probably > incorrect. After all, it doesn't include the string "xyzzy"... and clearly > this *should* have in there somewhere... Also missing are "etaoin shrdlu", "plugh". I, for one, expected better of the organizers. tgies _______________________________________________ icfpcontest-discuss mailing list icfpcontest-discuss@lists.andrew.cmu.edu https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss From jackdied at jackdied.com Thu Jul 20 00:27:16 2006 From: jackdied at jackdied.com (Jack Diederich) Date: Thu Jul 20 00:27:20 2006 Subject: [icfp-discuss] Pre-contest Language statistics Message-ID: <20060720042716.GI2540@performancedrivers.com> Python. Pure-python as a first pass and then rewrite the slow part as a C extension. In past years the "slow part" has been the "board" state object so this means rewriting the board in C while keeping the logic and state manipulation in python. Psycho (python JIT) can help but tends to segfault when doing closures and other functional stuff. Pyrex works OK but a hand crafted C extension isn't much harder (an extra 100 lines of C) and is faster and less buggy in my experience. That said it is very strange to see so many answers of "python" this year. In the past it has been a minority language - not invisible, just a minority. -jackdied nb, the codex file isn't a zip, gzip, bz2, or gpg file but it does offer up some da vinci code-ish stuff if you run "strings" on it. I wouldn't read too much into that, as an easter egg I left a binary cron job running at CDNow after I quit that was named "mail_credit_card_info" and "strings" only produced the output "running strings on this won't buy you anything" Fun stuff, it produced different egg output depending on who ran it. From dufresnep at fastmail.fm Thu Jul 20 01:44:17 2006 From: dufresnep at fastmail.fm (Paul Dufresne) Date: Thu Jul 20 01:44:25 2006 Subject: [icfp-discuss] Oztracon team found the secret 3 letters... Message-ID: <1153374257.25075.266450420@webmail.messagingengine.com> that are beginning by C. But let's wait Friday to know more, about what we found. -- http://www.fastmail.fm - Access all of your messages and folders wherever you are From niklas.broberg at gmail.com Thu Jul 20 02:12:32 2006 From: niklas.broberg at gmail.com (Niklas Broberg) Date: Thu Jul 20 02:12:35 2006 Subject: [icfp-discuss] Contest Materials added in 19 Jul In-Reply-To: <7b2bf4900607191849v6172d85dm4519be1eab7c7990@mail.gmail.com> References: <7b2bf4900607191849v6172d85dm4519be1eab7c7990@mail.gmail.com> Message-ID: On 7/20/06, Zhenzhong Xu wrote: > I am guessing we will be looking at a "decode a alien language" task. The file does include, in plain text, the words "roswell" and "area51" near the end, so it must be "alien" in the ET sense! :-) However, just decoding a text seems too simple a task (not simple as in easy), how do you determine which team did the "best" decoding? Perhaps it's a source code for an alien programming language, and we are to create something that makes it produce some "sensible" output? /Niklas From aguilar.james at gmail.com Thu Jul 20 02:20:20 2006 From: aguilar.james at gmail.com (James Aguilar) Date: Thu Jul 20 02:20:23 2006 Subject: [icfp-discuss] Contest Materials added in 19 Jul In-Reply-To: References: <7b2bf4900607191849v6172d85dm4519be1eab7c7990@mail.gmail.com> Message-ID: On 7/19/06, Niklas Broberg wrote: > > The file does include, in plain text, the words "roswell" and "area51" > near the end, so it must be "alien" in the ET sense! :-) Also "i love bees." file(1) does not know the format. Of course. However, just decoding a text seems too simple a task (not simple as > in easy), how do you determine which team did the "best" decoding? > Perhaps it's a source code for an alien programming language, and we > are to create something that makes it produce some "sensible" output? Agreed. Or something. We have sufficiently little information that we could not possibly make progress at the current juncture. Although, alien? The graphic on the main page makes me think more of ancient scroll (wizard and all), as does the word "codex." Could it be a rosetta stone of sorts? But I'd be surprised if our code was not to compete against other people's code at some point. -- James -- [?] James Aguilar [@] 18100 NE 95th St. #RR3088 [@] Redmond, WA 98052 [#] 314 494 0450 [!] There's a song that is sung by the saints who [!] have come and have gone from under the sun -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060719/bbed844d/attachment.html From dranger at gmail.com Thu Jul 20 06:25:46 2006 From: dranger at gmail.com (Stephen Dranger) Date: Thu Jul 20 06:25:48 2006 Subject: [icfp-discuss] Oztracon team found the secret 3 letters... Message-ID: <159b8eea0607200325u5e78180dx20746fa09216f2c4@mail.gmail.com> After reading through the file they gave us, I'm convinced our challenge will be to take a faded note written by the Knights Templar and extrapolate a global conspiracy from it. --Stephen that are beginning by C. > But let's wait Friday to know more, about what we found. > > -- > http://www.fastmail.fm - Access all of your messages and folders > wherever you are > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060720/3756d594/attachment.html From nelsoneci at gmail.com Thu Jul 20 13:41:41 2006 From: nelsoneci at gmail.com (Nelson Castillo) Date: Thu Jul 20 13:41:43 2006 Subject: [icfp-discuss] Contest Materials added in 19 Jul In-Reply-To: References: <7b2bf4900607191849v6172d85dm4519be1eab7c7990@mail.gmail.com> Message-ID: <2accc2ff0607201041y3fb9763av2d1adc4019f913d3@mail.gmail.com> > rosetta stone of sorts? But I'd be surprised if our code was not to compete > against other people's code at some point. GIF89a ? -- http://arhuaco.org/ From aguilar.james at gmail.com Thu Jul 20 13:45:59 2006 From: aguilar.james at gmail.com (James Aguilar) Date: Thu Jul 20 13:46:01 2006 Subject: [icfp-discuss] Contest Materials added in 19 Jul In-Reply-To: <2accc2ff0607201041y3fb9763av2d1adc4019f913d3@mail.gmail.com> References: <7b2bf4900607191849v6172d85dm4519be1eab7c7990@mail.gmail.com> <2accc2ff0607201041y3fb9763av2d1adc4019f913d3@mail.gmail.com> Message-ID: On 7/20/06, Nelson Castillo wrote: > > > rosetta stone of sorts? But I'd be surprised if our code was not to > compete > > against other people's code at some point. > > GIF89a ? Mu? -- [?] James Aguilar [@] 18100 NE 95th St. #RR3088 [@] Redmond, WA 98052 [#] 314 494 0450 [!] There's a song that is sung by the saints who [!] have come and have gone from under the sun -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060720/aadae458/attachment.html From dufresnep at fastmail.fm Thu Jul 20 14:08:07 2006 From: dufresnep at fastmail.fm (Paul Dufresne) Date: Thu Jul 20 14:08:08 2006 Subject: [icfp-discuss] Contest Materials added in 19 Jul In-Reply-To: <2accc2ff0607201041y3fb9763av2d1adc4019f913d3@mail.gmail.com> References: <7b2bf4900607191849v6172d85dm4519be1eab7c7990@mail.gmail.com> <2accc2ff0607201041y3fb9763av2d1adc4019f913d3@mail.gmail.com> Message-ID: <1153418887.13247.266500088@webmail.messagingengine.com> > GIF89a ? :-) yes. But the image is elsewhere too! And the message is very clear: The extra-terrestrials are using Call-by-Value in their programming language(s). But don't tell this aloud, this may be a very big secret! --Oztracon team. -- http://www.fastmail.fm - Choose from over 50 domains or use your own From blo.b at infonie.fr Thu Jul 20 14:25:22 2006 From: blo.b at infonie.fr (Laurent Vaucher) Date: Thu Jul 20 14:25:35 2006 Subject: [icfp-discuss] Contest Materials added in 19 Jul In-Reply-To: <1153418887.13247.266500088@webmail.messagingengine.com> References: <7b2bf4900607191849v6172d85dm4519be1eab7c7990@mail.gmail.com> <2accc2ff0607201041y3fb9763av2d1adc4019f913d3@mail.gmail.com> <1153418887.13247.266500088@webmail.messagingengine.com> Message-ID: <37783591-EEF8-4D57-A6DF-51B01E426DB8@infonie.fr> > And the message is very clear: > The extra-terrestrials are using Call-by-Value in their programming > language(s). Don' t you think they speak a rare colombian language? https://www.ethnologue.com/14/show_language.asp?code=CBV From aguilar.james at gmail.com Thu Jul 20 14:31:03 2006 From: aguilar.james at gmail.com (James Aguilar) Date: Thu Jul 20 14:31:07 2006 Subject: [icfp-discuss] Contest Materials added in 19 Jul In-Reply-To: <37783591-EEF8-4D57-A6DF-51B01E426DB8@infonie.fr> References: <7b2bf4900607191849v6172d85dm4519be1eab7c7990@mail.gmail.com> <2accc2ff0607201041y3fb9763av2d1adc4019f913d3@mail.gmail.com> <1153418887.13247.266500088@webmail.messagingengine.com> <37783591-EEF8-4D57-A6DF-51B01E426DB8@infonie.fr> Message-ID: On 7/20/06, Laurent Vaucher wrote: > Don' t you think they speak a rare colombian language? > > https://www.ethnologue.com/14/show_language.asp?code=CBV I can see that I am already at a pretty big disadvantage in this contest. :P You guys are damn smart to have figured out this stuff already. -- [?] James Aguilar [@] 18100 NE 95th St. #RR3088 / Redmond, WA 98052 [#] 314 494 0450 [!] In the wind and the rain, my darling, say goodbye -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060720/40bc7748/attachment.html From katolazzus at katolaz.homeunix.net Thu Jul 20 15:00:17 2006 From: katolazzus at katolaz.homeunix.net (KatolaZ) Date: Thu Jul 20 15:00:21 2006 Subject: [icfp-discuss] Material....:-)) Message-ID: <20060720190017.GB18430@wontolla> Hi. These are results of randomness tests to "The Codex": Total bytes: 2346960 bit 0: 9387756 perc: 49.999553 bit 1: 9387924 perc: 50.000447 HalfNibble 0: 4702717 perc: 25.046853 HalfNibble 1: 4685039 perc: 24.952699 HalfNibble 2: 4685039 perc: 24.952699 HalfNibble 3: 4702885 perc: 25.047748 Frequency test: 0.001503 (should be lower than [2.7055:10.8276] ) Serial test: 1.00000000 (should be lower than [4.6052:13.8155]) Poker test (m=4): 135.62797150 (should be lower than [22.3071:37.6973]) AHAHAHAHAHAH!!!!! So it has been obtained by a not-so-accurate random number generator, apart of the clear message that everyone can see beyond the text itself ;-) "...And I will call it Abulafia; Abu for friends. And my Abulafia will be more careful and clever than yours. The problem is finding all the permutations of The Name of God, isn't it ? Well, look on this manual: I've a small program, written in BASIC, which permutes all the sequences of four letters. It seems to fit well for IHVH!..." ...... From "The Pendulum of Foucault", U. Eco ....... From ij7b3u502 at sneakemail.com Thu Jul 20 15:12:50 2006 From: ij7b3u502 at sneakemail.com (vbzoli) Date: Thu Jul 20 15:12:54 2006 Subject: [icfp-discuss] Contest Materials added in 19 Jul Message-ID: <24064-82324@sneakemail.com> Oops! I discovered clear connection with Hungarian history (found string "rakoczi"): http://en.wikipedia.org/wiki/Ferenc_II_Rákóczi :o)) VOROSBARANYI Zoltan From blo.b at infonie.fr Thu Jul 20 15:18:11 2006 From: blo.b at infonie.fr (Laurent Vaucher) Date: Thu Jul 20 15:18:19 2006 Subject: [icfp-discuss] Material....:-)) In-Reply-To: <20060720190017.GB18430@wontolla> References: <20060720190017.GB18430@wontolla> Message-ID: <37A58407-07CF-4E50-B2AB-393ED26DA6F9@infonie.fr> And what do you make of Raimundus Lullus? He seemed very fond of designing paper 'machines' to generate all the truths of christianity. http://en.wikipedia.org/wiki/Ramon_Lull Oh, and by the way, did you know? Paul is dead. From jjmaes at gmail.com Thu Jul 20 15:21:13 2006 From: jjmaes at gmail.com (James Darwin Maes The First (a javatarian)) Date: Thu Jul 20 15:21:16 2006 Subject: [icfp-discuss] Material....:-)) In-Reply-To: <37A58407-07CF-4E50-B2AB-393ED26DA6F9@infonie.fr> References: <20060720190017.GB18430@wontolla> <37A58407-07CF-4E50-B2AB-393ED26DA6F9@infonie.fr> Message-ID: Also... "societas eruditorum" right after the embedded gif http://en.wikipedia.org/wiki/Societas_Eruditorum On 7/20/06, Laurent Vaucher wrote: > > And what do you make of Raimundus Lullus? He seemed very fond of > designing paper 'machines' to generate all the truths of christianity. > > http://en.wikipedia.org/wiki/Ramon_Lull > > > Oh, and by the way, did you know? Paul is dead. > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060720/14b011ad/attachment-0001.html From katolazzus at katolaz.homeunix.net Thu Jul 20 15:32:38 2006 From: katolazzus at katolaz.homeunix.net (Katolaz) Date: Thu Jul 20 15:32:43 2006 Subject: [me@katolaz.homeunix.net: Re: [icfp-discuss] Material....:-))] Message-ID: <20060720193238.GA18765@wontolla> On Thu, Jul 20, 2006 at 02:21:13PM -0500, James Darwin Maes The First (a javatarian) wrote: > Also... > > "societas eruditorum" right after the embedded gif > > > http://en.wikipedia.org/wiki/Societas_Eruditorum > "ignoti et quasi occulti"... From katolazzus at katolaz.homeunix.net Thu Jul 20 15:33:15 2006 From: katolazzus at katolaz.homeunix.net (Katolaz) Date: Thu Jul 20 15:33:18 2006 Subject: [me@katolaz.homeunix.net: [me@katolaz.homeunix.net: Re: [icfp-discuss] Material....:-))]] Message-ID: <20060720193315.GB18765@wontolla> On Thu, Jul 20, 2006 at 09:18:11PM +0200, Laurent Vaucher wrote: > And what do you make of Raimundus Lullus? He seemed very fond of > designing paper 'machines' to generate all the truths of christianity. > > http://en.wikipedia.org/wiki/Ramon_Lull > > > Oh, and by the way, did you know? Paul is dead. Yes, all of us knows :-) And I suppose that you also know the real story of the Count "Tsarogy", also known as Count "Welldone", also known as Francis "Rakoczy", also known as "Surmount", also known as the Count fof Saint Germain..... I suppose also that all of you note the reference to "anbaric and telluric currents"... but don't be silly, the message is really CLEAR. And members of "tycon mismatch", who are the most relevant expertises in the field, understood it ! From Charles.Bouillaguet at crans.org Thu Jul 20 15:37:02 2006 From: Charles.Bouillaguet at crans.org (Charles Bouillaguet) Date: Thu Jul 20 15:37:07 2006 Subject: [icfp-discuss] Contest Materials added in 19 Jul In-Reply-To: <1153418887.13247.266500088@webmail.messagingengine.com> References: <7b2bf4900607191849v6172d85dm4519be1eab7c7990@mail.gmail.com> <2accc2ff0607201041y3fb9763av2d1adc4019f913d3@mail.gmail.com> <1153418887.13247.266500088@webmail.messagingengine.com> Message-ID: <1153424222.12052.1.camel@zblup> On Thu, 2006-07-20 at 14:08 -0400, Paul Dufresne wrote: > > GIF89a ? > :-) yes. > But the image is elsewhere too! > And the message is very clear: > The extra-terrestrials are using Call-by-Value in their programming > language(s). You can find the gif image at offset 2340188. It's 623 bytes long. you can extract it with : tail -c 2340188 codex.umz | head -c 623 > cbv.gif good luck, folks :) -- Entirely Non-Serious team From jjmaes at gmail.com Thu Jul 20 15:43:26 2006 From: jjmaes at gmail.com (James Darwin Maes The First (a javatarian)) Date: Thu Jul 20 15:43:30 2006 Subject: [icfp-discuss] Contest Materials added in 19 Jul In-Reply-To: <1153424222.12052.1.camel@zblup> References: <7b2bf4900607191849v6172d85dm4519be1eab7c7990@mail.gmail.com> <2accc2ff0607201041y3fb9763av2d1adc4019f913d3@mail.gmail.com> <1153418887.13247.266500088@webmail.messagingengine.com> <1153424222.12052.1.camel@zblup> Message-ID: I am sure this is all just to keep us busy until tomorrow morning. That being said, here are a quick list of the strings found (via a strings command matched against a linux words file plus a little looking around) gnoti et quasi occulti well done daed si luap Evan Chan was murdered fnord tsarogy apply MKULTRA abracadabra eval eval so dark the con of man dust solomon rakoczi novus ordo seclorum tell uric plbndetibh u ou rvcofalt ea templar tycon mismatch apply raimundus societas eruditorum ADE anbaric Tga'G abulafi a bad wolf lambda roswell area51 i love bees __42__ lullus surmount currents On 7/20/06, Charles Bouillaguet wrote: > > On Thu, 2006-07-20 at 14:08 -0400, Paul Dufresne wrote: > > > GIF89a ? > > :-) yes. > > But the image is elsewhere too! > > And the message is very clear: > > The extra-terrestrials are using Call-by-Value in their programming > > language(s). > > You can find the gif image at offset 2340188. It's 623 bytes long. > > you can extract it with : > > tail -c 2340188 codex.umz | head -c 623 > cbv.gif > > good luck, folks :) > > -- > Entirely Non-Serious team > > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060720/dbc2fcb1/attachment.html From flyfree at gmail.com Thu Jul 20 15:43:59 2006 From: flyfree at gmail.com (Zhenzhong Xu) Date: Thu Jul 20 15:44:02 2006 Subject: [icfp-discuss] Contest Materials added in 19 Jul In-Reply-To: <1153424222.12052.1.camel@zblup> References: <7b2bf4900607191849v6172d85dm4519be1eab7c7990@mail.gmail.com> <2accc2ff0607201041y3fb9763av2d1adc4019f913d3@mail.gmail.com> <1153418887.13247.266500088@webmail.messagingengine.com> <1153424222.12052.1.camel@zblup> Message-ID: <7b2bf4900607201243o51160e7sc6795879ce636fa9@mail.gmail.com> and what's up with this http://juliansanchez.com/notes/archives/2002/11/evan_chan_was_m.php On 7/20/06, Charles Bouillaguet wrote: > On Thu, 2006-07-20 at 14:08 -0400, Paul Dufresne wrote: > > > GIF89a ? > > :-) yes. > > But the image is elsewhere too! > > And the message is very clear: > > The extra-terrestrials are using Call-by-Value in their programming > > language(s). > > You can find the gif image at offset 2340188. It's 623 bytes long. > > you can extract it with : > > tail -c 2340188 codex.umz | head -c 623 > cbv.gif > > good luck, folks :) > > -- > Entirely Non-Serious team > > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > From katolazzus at katolaz.homeunix.net Thu Jul 20 15:48:52 2006 From: katolazzus at katolaz.homeunix.net (Katolaz) Date: Thu Jul 20 15:48:55 2006 Subject: [katolazzus@katolaz.homeunix.net: Re: [icfp-discuss] Material....:-))] Message-ID: <20060720194852.GD18765@wontolla> On Thu, Jul 20, 2006 at 09:18:11PM +0200, Laurent Vaucher wrote: > And what do you make of Raimundus Lullus? He seemed very fond of > designing paper 'machines' to generate all the truths of christianity. > > http://en.wikipedia.org/wiki/Ramon_Lull > > > Oh, and by the way, did you know? Paul is dead. Also Chan, since it was murdered! Perhaps he had anything to do with MKULTRA ???? From Charles.Bouillaguet at crans.org Thu Jul 20 15:58:03 2006 From: Charles.Bouillaguet at crans.org (Charles Bouillaguet) Date: Thu Jul 20 15:58:18 2006 Subject: [icfp-discuss] Contest Materials added in 19 Jul In-Reply-To: <1153424222.12052.1.camel@zblup> References: <7b2bf4900607191849v6172d85dm4519be1eab7c7990@mail.gmail.com> <2accc2ff0607201041y3fb9763av2d1adc4019f913d3@mail.gmail.com> <1153418887.13247.266500088@webmail.messagingengine.com> <1153424222.12052.1.camel@zblup> Message-ID: <1153425485.12052.4.camel@zblup> On Thu, 2006-07-20 at 15:37 -0400, Charles Bouillaguet wrote: > On Thu, 2006-07-20 at 14:08 -0400, Paul Dufresne wrote: > > > GIF89a ? > > :-) yes. > > But the image is elsewhere too! > > And the message is very clear: > > The extra-terrestrials are using Call-by-Value in their programming > > language(s). > > You can find the gif image at offset 2340188. It's 623 bytes long. my mistake, the offset really is 6772 -sorry- From blo.b at infonie.fr Thu Jul 20 16:10:03 2006 From: blo.b at infonie.fr (Laurent Vaucher) Date: Thu Jul 20 16:10:11 2006 Subject: [icfp-discuss] Contest Materials added in 19 Jul In-Reply-To: References: <7b2bf4900607191849v6172d85dm4519be1eab7c7990@mail.gmail.com> <2accc2ff0607201041y3fb9763av2d1adc4019f913d3@mail.gmail.com> <1153418887.13247.266500088@webmail.messagingengine.com> <1153424222.12052.1.camel@zblup> Message-ID: > That being said, here are a quick list of the strings found (via a > strings command matched against a linux words file plus a little > looking around) You missed the mighty CBV, at the end, between area51 and i love bees. CBV, which is the rot13 transform of POI!!! I will say no more because there might be malignant ears in the vicinity. From niklas.broberg at gmail.com Thu Jul 20 16:29:11 2006 From: niklas.broberg at gmail.com (Niklas Broberg) Date: Thu Jul 20 16:29:14 2006 Subject: [icfp-discuss] Contest Materials added in 19 Jul In-Reply-To: References: <7b2bf4900607191849v6172d85dm4519be1eab7c7990@mail.gmail.com> <2accc2ff0607201041y3fb9763av2d1adc4019f913d3@mail.gmail.com> <1153418887.13247.266500088@webmail.messagingengine.com> <1153424222.12052.1.camel@zblup> Message-ID: I strongly believe the word "fnord" has been given too little attention around here. http://en.wikipedia.org/wiki/Fnord /Niklas From aguilar.james at gmail.com Thu Jul 20 16:30:24 2006 From: aguilar.james at gmail.com (James Aguilar) Date: Thu Jul 20 16:30:29 2006 Subject: [me@katolaz.homeunix.net: [me@katolaz.homeunix.net: Re: [icfp-discuss] Material....:-))]] In-Reply-To: <20060720193315.GB18765@wontolla> References: <20060720193315.GB18765@wontolla> Message-ID: Katolaz, Would it be possible for you to fix the subject lines of your emails? They result in these huge messes (See the subject line of this mail -- everything after the Re: is yours). Thanks. Yours, James Aguilar -- [?] James Aguilar [@] 18100 NE 95th St. #RR3088 / Redmond, WA 98052 [#] 314 494 0450 [!] In the wind and the rain, my darling, say goodbye -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060720/9cba0e0b/attachment.html From jjmaes at gmail.com Thu Jul 20 16:39:01 2006 From: jjmaes at gmail.com (James Darwin Maes The First (a javatarian)) Date: Thu Jul 20 16:39:03 2006 Subject: [me@katolaz.homeunix.net: [me@katolaz.homeunix.net: Re: [icfp-discuss] Material....:-))]] In-Reply-To: References: <20060720193315.GB18765@wontolla> Message-ID: wow guy who hates bad email subject lines On 7/20/06, James Aguilar wrote: > > Katolaz, > > Would it be possible for you to fix the subject lines of your emails? > They result in these huge messes (See the subject line of this mail -- > everything after the Re: is yours). Thanks. > > Yours, > James Aguilar > > > -- > [?] James Aguilar > [@] 18100 NE 95th St. #RR3088 / Redmond, WA 98052 > [#] 314 494 0450 > [!] In the wind and the rain, my darling, say goodbye > > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > > > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060720/2114a593/attachment.html From aguilar.james at gmail.com Thu Jul 20 16:43:32 2006 From: aguilar.james at gmail.com (James Aguilar) Date: Thu Jul 20 16:43:35 2006 Subject: [me@katolaz.homeunix.net: [me@katolaz.homeunix.net: Re: [icfp-discuss] Material....:-))]] In-Reply-To: References: <20060720193315.GB18765@wontolla> Message-ID: On 7/20/06, James Darwin Maes The First (a javatarian) wrote: > > wow > > guy who hates bad email subject lines > Oops, didn't mean that to go to everyone. All the same. He doesn't have to do it; I'm fine with filtering out his emails if it's too much of an imposition. ;) Yours, James Aguilar -- [?] James Aguilar [@] 18100 NE 95th St. #RR3088 / Redmond, WA 98052 [#] 314 494 0450 [!] In the wind and the rain, my darling, say goodbye -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060720/36c27ffa/attachment.html From dranger at gmail.com Thu Jul 20 20:10:43 2006 From: dranger at gmail.com (Stephen Dranger) Date: Thu Jul 20 20:10:46 2006 Subject: [me@katolaz.homeunix.net: [me@katolaz.homeunix.net: Re: [icfp-discuss] Material....:-))]] In-Reply-To: <20060720193315.GB18765@wontolla> References: <20060720193315.GB18765@wontolla> Message-ID: <159b8eea0607201710x6db8f692h9566b920de7e4dba@mail.gmail.com> Clearly Tom is the Count of Saint Germain, getting us to do his dirty work for him, exposing the location of the telluric currents. --Stephen On 7/21/06, Katolaz wrote: > > On Thu, Jul 20, 2006 at 09:18:11PM +0200, Laurent Vaucher wrote: > > And what do you make of Raimundus Lullus? He seemed very fond of > > designing paper 'machines' to generate all the truths of christianity. > > > > http://en.wikipedia.org/wiki/Ramon_Lull > > > > > > Oh, and by the way, did you know? Paul is dead. > > Yes, all of us knows :-) > > And I suppose that you also know the real story of the Count "Tsarogy", > also known as Count "Welldone", also known as Francis "Rakoczy", also > known as "Surmount", also known as the Count fof Saint Germain..... > > I suppose also that all of you note the reference to "anbaric and > telluric currents"... but don't be silly, the message is really CLEAR. > > And members of "tycon mismatch", who are the most relevant expertises > in the field, understood it ! > > > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060721/e54c567e/attachment.html From dranger at gmail.com Thu Jul 20 20:12:08 2006 From: dranger at gmail.com (Stephen Dranger) Date: Thu Jul 20 20:12:10 2006 Subject: [icfp-discuss] Contest Materials added in 19 Jul In-Reply-To: <1153424222.12052.1.camel@zblup> References: <7b2bf4900607191849v6172d85dm4519be1eab7c7990@mail.gmail.com> <2accc2ff0607201041y3fb9763av2d1adc4019f913d3@mail.gmail.com> <1153418887.13247.266500088@webmail.messagingengine.com> <1153424222.12052.1.camel@zblup> Message-ID: <159b8eea0607201712j5db590ajc4b3bb36278a51bb@mail.gmail.com> On 7/21/06, Charles Bouillaguet wrote: > > On Thu, 2006-07-20 at 14:08 -0400, Paul Dufresne wrote: > > > GIF89a ? > > :-) yes. > > But the image is elsewhere too! > > And the message is very clear: > > The extra-terrestrials are using Call-by-Value in their programming > > language(s). > > You can find the gif image at offset 2340188. It's 623 bytes long. > > you can extract it with : > > tail -c 2340188 codex.umz | head -c 623 > cbv.gif > > good luck, folks :) Yes, but did you find the *other* CBV image? --Stephen Pretend Robot Pants -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060721/c69d4770/attachment.html From daniel at erlfinsys.net Fri Jul 21 02:49:45 2006 From: daniel at erlfinsys.net (Danie Schutte) Date: Fri Jul 21 02:52:36 2006 Subject: [icfp-discuss] Fwd: Small teams Message-ID: <200607210849.45744.daniel@erlfinsys.net> just a quick very silly remark a small team consist of 0 to 4 members - 0 contributing members - any such teams out there -- no answers will constitute a positive response :) -- Managing Director Erlang Financial Systems (Pty) Ltd Mobile: +27 84 468 3138 Phone : +11 235 6500 ext 6801 Fax : +11 235 6690 From ashish_frnds at yahoo.co.in Fri Jul 21 02:59:31 2006 From: ashish_frnds at yahoo.co.in (ashish yadav) Date: Fri Jul 21 02:59:36 2006 Subject: [icfp-discuss] plz help!!! Message-ID: <20060721065931.72462.qmail@web8606.mail.in.yahoo.com> hi ppl.. i m participating in this contest first time.. and i don't know about about the umz file required to be downloaded prior to contest.. i mean what it is and.. ad how to see the contents.. thaks in advance.. --------------------------------- Find out what India is talking about on Yahoo! Answers India. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060721/b2113342/attachment-0001.html From rob.hunter at gmail.com Fri Jul 21 03:06:33 2006 From: rob.hunter at gmail.com (Rob Hunter) Date: Fri Jul 21 03:06:35 2006 Subject: [icfp-discuss] plz help!!! In-Reply-To: <20060721065931.72462.qmail@web8606.mail.in.yahoo.com> References: <20060721065931.72462.qmail@web8606.mail.in.yahoo.com> Message-ID: The umz file is just a teaser of some sort. The contest organizers haven't given any more information. Check the archives of this list for info on what people have discovered about the file so far, but I don't believe you are supposed to know yet what to do with the file. All should be revealed tomorrow though when the contest begins. --rob On 7/20/06, ashish yadav wrote: > > hi ppl.. > i m participating in this contest first time.. > and i don't know about about the umz file required to be downloaded prior to > contest.. > i mean what it is and.. ad how to see the contents.. > thaks in advance.. > > > > ________________________________ > Find out what India is talking about on Yahoo! Answers India. > > > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > > > From ashish_frnds at yahoo.co.in Fri Jul 21 03:21:27 2006 From: ashish_frnds at yahoo.co.in (ashish yadav) Date: Fri Jul 21 03:33:09 2006 Subject: [icfp-discuss] plz help!!! In-Reply-To: Message-ID: <20060721072127.17354.qmail@web8601.mail.in.yahoo.com> thanks.. for ur help.. then i have to wait till tomorrow i guess.. Rob Hunter wrote: The umz file is just a teaser of some sort. The contest organizers haven't given any more information. Check the archives of this list for info on what people have discovered about the file so far, but I don't believe you are supposed to know yet what to do with the file. All should be revealed tomorrow though when the contest begins. --rob On 7/20/06, ashish yadav wrote: > > hi ppl.. > i m participating in this contest first time.. > and i don't know about about the umz file required to be downloaded prior to > contest.. > i mean what it is and.. ad how to see the contents.. > thaks in advance.. > > > > ________________________________ > Find out what India is talking about on Yahoo! Answers India. > > > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > > > _______________________________________________ icfpcontest-discuss mailing list icfpcontest-discuss@lists.andrew.cmu.edu https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss --------------------------------- Find out what India is talking about on Yahoo! Answers India. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060721/957b61ba/attachment.html From traude at gmx.de Fri Jul 21 03:50:41 2006 From: traude at gmx.de (Traude Manz) Date: Fri Jul 21 03:50:45 2006 Subject: [icfp-discuss] Fwd: Small teams In-Reply-To: <200607210849.45744.daniel@erlfinsys.net> References: <200607210849.45744.daniel@erlfinsys.net> Message-ID: <20060721075041.84200@gmx.net> Yes, I registered the team HaveFun0 with 0 participants because I will not be able to submit due to time constraints this weekend, but I don't want to miss all the email exchanges :-) Traude -------- Original-Nachricht -------- Datum: Fri, 21 Jul 2006 08:49:45 +0200 Von: Danie Schutte An: Betreff: [icfp-discuss] Fwd: Small teams > > just a quick very silly remark > a small team consist of 0 to 4 members - 0 contributing members - any > such teams out there -- no answers will constitute a positive response :) > > -- > Managing Director > Erlang Financial Systems (Pty) Ltd > > Mobile: +27 84 468 3138 > Phone : +11 235 6500 ext 6801 > Fax : +11 235 6690 > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss -- Echte DSL-Flatrate dauerhaft f?r 0,- Euro*! "Feel free" mit GMX DSL! http://www.gmx.net/de/go/dsl From roshan.r at gmail.com Fri Jul 21 05:23:57 2006 From: roshan.r at gmail.com (Roshan Rammohan) Date: Fri Jul 21 05:24:01 2006 Subject: [icfp-discuss] Fwd: Small teams In-Reply-To: <20060721075041.84200@gmx.net> References: <200607210849.45744.daniel@erlfinsys.net> <20060721075041.84200@gmx.net> Message-ID: <44C09D2D.6030402@gmail.com> Hehe.. Im a noob team of one soon to change team strength to 0 for similar reasons. But the emails have been quite entertaining :) Traude Manz wrote: > Yes, I registered the team HaveFun0 with 0 participants > because I will not be able to submit due to time constraints > this weekend, but I don't want to miss all the email exchanges :-) > > Traude > > -------- Original-Nachricht -------- > Datum: Fri, 21 Jul 2006 08:49:45 +0200 > Von: Danie Schutte > An: > Betreff: [icfp-discuss] Fwd: Small teams > > >> just a quick very silly remark >> a small team consist of 0 to 4 members - 0 contributing members - any >> such teams out there -- no answers will constitute a positive response :) >> >> -- >> Managing Director >> Erlang Financial Systems (Pty) Ltd >> >> Mobile: +27 84 468 3138 >> Phone : +11 235 6500 ext 6801 >> Fax : +11 235 6690 >> _______________________________________________ >> icfpcontest-discuss mailing list >> icfpcontest-discuss@lists.andrew.cmu.edu >> https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss >> > > From icfp_06 at schlagsei.de Fri Jul 21 06:22:59 2006 From: icfp_06 at schlagsei.de (=?ISO-8859-1?Q?Daniel_Schr=F6ter?=) Date: Fri Jul 21 06:23:03 2006 Subject: [icfp-discuss] Contest Materials added in 19 Jul In-Reply-To: <159b8eea0607201712j5db590ajc4b3bb36278a51bb@mail.gmail.com> References: <7b2bf4900607191849v6172d85dm4519be1eab7c7990@mail.gmail.com> <2accc2ff0607201041y3fb9763av2d1adc4019f913d3@mail.gmail.com> <1153418887.13247.266500088@webmail.messagingengine.com> <1153424222.12052.1.camel@zblup> <159b8eea0607201712j5db590ajc4b3bb36278a51bb@mail.gmail.com> Message-ID: <44C0AB03.5010608@schlagsei.de> Stephen Dranger wrote: > On 7/21/06, *Charles Bouillaguet* > wrote: > > On Thu, 2006-07-20 at 14:08 -0400, Paul Dufresne wrote: > > > GIF89a ? > > :-) yes. > > But the image is elsewhere too! > > And the message is very clear: > > The extra-terrestrials are using Call-by-Value in their programming > > language(s). > > You can find the gif image at offset 2340188. It's 623 bytes long. > > you can extract it with : > > tail -c 2340188 codex.umz | head -c 623 > cbv.gif > > good luck, folks :) > > > Yes, but did you find the *other* CBV image? > --Stephen > Pretend Robot Pants > ------------------------------------------------------------------------ > > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > Do you mean the one at 1576? 4096 Bytes of grayscale codes? 64 Bytes per line? greetz Daniel From dufresnep at fastmail.fm Fri Jul 21 07:47:59 2006 From: dufresnep at fastmail.fm (Paul Dufresne) Date: Fri Jul 21 07:48:06 2006 Subject: [icfp-discuss] Contest Materials added in 19 Jul In-Reply-To: <44C0AB03.5010608@schlagsei.de> References: <7b2bf4900607191849v6172d85dm4519be1eab7c7990@mail.gmail.com> <2accc2ff0607201041y3fb9763av2d1adc4019f913d3@mail.gmail.com> <1153418887.13247.266500088@webmail.messagingengine.com> <1153424222.12052.1.camel@zblup> <159b8eea0607201712j5db590ajc4b3bb36278a51bb@mail.gmail.com> <44C0AB03.5010608@schlagsei.de> Message-ID: <1153482479.18416.266556413@webmail.messagingengine.com> > Do you mean the one at 1576? 4096 Bytes of grayscale codes? 64 Bytes per > line? That is indeed the one, I, was talking about when I said "the image is elsewhere too". Let's just hope they won't say: "Hey, all the task description of this year is in this file, you just have to find it!" -- http://www.fastmail.fm - I mean, what is it about a decent email service? From davidad at mit.edu Fri Jul 21 08:32:11 2006 From: davidad at mit.edu (David Dalrymple) Date: Fri Jul 21 08:32:15 2006 Subject: [icfp-discuss] Contest Materials added in 19 Jul In-Reply-To: <1153482479.18416.266556413@webmail.messagingengine.com> References: <7b2bf4900607191849v6172d85dm4519be1eab7c7990@mail.gmail.com> <2accc2ff0607201041y3fb9763av2d1adc4019f913d3@mail.gmail.com> <1153418887.13247.266500088@webmail.messagingengine.com> <1153424222.12052.1.camel@zblup> <159b8eea0607201712j5db590ajc4b3bb36278a51bb@mail.gmail.com> <44C0AB03.5010608@schlagsei.de> <1153482479.18416.266556413@webmail.messagingengine.com> Message-ID: <504b2780607210532s4d439c20tf798bf5988cf165c@mail.gmail.com> > Let's just hope they won't say: > "Hey, all the task description of this year is in this file, you just > have to find it!" Heh, yes. I'm sort of hoping for the opposite -- "Hope you all had fun with that random binary file we made! Now, here's the task: ..." -- but I don't think that's likely either. --David From franke.daniel at gmail.com Fri Jul 21 08:46:06 2006 From: franke.daniel at gmail.com (Daniel Franke) Date: Fri Jul 21 08:46:08 2006 Subject: [icfp-discuss] some more speculations at slashdot Message-ID: <640ad44b0607210546u484a92f6h49d6d9c7c57df0bc@mail.gmail.com> Hi all. Since I just ran into them: some additional conspiracy theories may be found at slashdot: http://it.slashdot.org/article.pl?sid=06/07/20/1629201 Something not yet mentioned on this list: >>It turns out "plbndetibh u ou rvcofalt ea" is an anagram of "unfavorable botched tulip"<< Let's wait for illumination. Fnord. From abhimanyu.seth at gmail.com Fri Jul 21 08:49:11 2006 From: abhimanyu.seth at gmail.com (Abhimanyu Seth) Date: Fri Jul 21 08:49:12 2006 Subject: [icfp-discuss] Contest Materials added in 19 Jul In-Reply-To: References: <7b2bf4900607191849v6172d85dm4519be1eab7c7990@mail.gmail.com> <2accc2ff0607201041y3fb9763av2d1adc4019f913d3@mail.gmail.com> <1153418887.13247.266500088@webmail.messagingengine.com> <1153424222.12052.1.camel@zblup> Message-ID: <591d95910607210549j10c57861uede8be03136afc9a@mail.gmail.com> Also, regarding well done daed si luap, "daed si luap" is the reverse of Paul is dead. Any idea who this Paul is? On 7/21/06, James Darwin Maes The First (a javatarian) wrote: > > I am sure this is all just to keep us busy until tomorrow morning. > > That being said, here are a quick list of the strings found (via a strings > command matched against a linux words file plus a little looking around) > > > gnoti et quasi occulti > > well done daed si luap > > Evan Chan was murdered > > fnord > > tsarogy > > apply > > MKULTRA > > abracadabra > > eval > > eval so dark the con of man > > dust solomon > > rakoczi > > novus ordo seclorum > > tell uric > > plbndetibh u ou rvcofalt ea > > templar > > tycon mismatch > > apply > > raimundus > > societas eruditorum > > ADE > > anbaric > > Tga'G > > abulafi a bad wolf > > lambda > > roswell > > area51 > > i love bees > > __42__ > > lullus > > surmount > > currents > > > > On 7/20/06, Charles Bouillaguet wrote: > > > > On Thu, 2006-07-20 at 14:08 -0400, Paul Dufresne wrote: > > > > GIF89a ? > > > :-) yes. > > > But the image is elsewhere too! > > > And the message is very clear: > > > The extra-terrestrials are using Call-by-Value in their programming > > > language(s). > > > > You can find the gif image at offset 2340188. It's 623 bytes long. > > > > you can extract it with : > > > > tail -c 2340188 codex.umz | head -c 623 > cbv.gif > > > > good luck, folks :) > > > > -- > > Entirely Non-Serious team > > > > _______________________________________________ > > icfpcontest-discuss mailing list > > icfpcontest-discuss@lists.andrew.cmu.edu > > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > > > > > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > > > -- Regards, Abhimanyu -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060721/e941689c/attachment.html From johannes at albapasser.de Fri Jul 21 08:54:41 2006 From: johannes at albapasser.de (Johannes) Date: Fri Jul 21 08:54:51 2006 Subject: [icfp-discuss] Contest Materials added in 19 Jul In-Reply-To: <591d95910607210549j10c57861uede8be03136afc9a@mail.gmail.com> References: <7b2bf4900607191849v6172d85dm4519be1eab7c7990@mail.gmail.com> <2accc2ff0607201041y3fb9763av2d1adc4019f913d3@mail.gmail.com> <1153418887.13247.266500088@webmail.messagingengine.com> <1153424222.12052.1.camel@zblup> <591d95910607210549j10c57861uede8be03136afc9a@mail.gmail.com> Message-ID: <44C0CE91.6050508@albapasser.de> > Paul is dead. Any idea who this Paul is? http://en.wikipedia.org/wiki/Paul_is_dead From davidad at mit.edu Fri Jul 21 08:56:09 2006 From: davidad at mit.edu (David Dalrymple) Date: Fri Jul 21 08:56:11 2006 Subject: [icfp-discuss] Contest Materials added in 19 Jul In-Reply-To: <591d95910607210549j10c57861uede8be03136afc9a@mail.gmail.com> References: <7b2bf4900607191849v6172d85dm4519be1eab7c7990@mail.gmail.com> <2accc2ff0607201041y3fb9763av2d1adc4019f913d3@mail.gmail.com> <1153418887.13247.266500088@webmail.messagingengine.com> <1153424222.12052.1.camel@zblup> <591d95910607210549j10c57861uede8be03136afc9a@mail.gmail.com> Message-ID: <504b2780607210556j52f1cd23i3297a0ef556c89c1@mail.gmail.com> Paul Graham? Just the first name that comes to mind for "Paul" in an FP context... --David On 7/21/06, Abhimanyu Seth wrote: > Also, regarding well done daed si luap, "daed si luap" is the reverse of > Paul is dead. Any idea who this Paul is? > > On 7/21/06, James Darwin Maes The First (a javatarian) > wrote: > > > > > > I am sure this is all just to keep us busy until tomorrow morning. > > > > That being said, here are a quick list of the strings found (via a strings > command matched against a linux words file plus a little looking around) > > > > > > > > gnoti et quasi occulti > > > > well done daed si luap > > > > Evan Chan was murdered > > > > fnord > > > > tsarogy > > > > apply > > > > MKULTRA > > > > abracadabra > > > > eval > > > > eval so dark the con of man > > > > dust solomon > > > > rakoczi > > > > novus ordo seclorum > > > > tell uric > > > > plbndetibh u ou rvcofalt ea > > > > templar > > > > tycon mismatch > > > > apply > > > > raimundus > > > > societas eruditorum > > > > ADE > > > > anbaric > > > > Tga'G > > > > abulafi a bad wolf > > > > lambda > > > > roswell > > > > area51 > > > > i love bees > > > > __42__ > > > > lullus > > > > surmount > > > > currents > > > > > > > > > > On 7/20/06, Charles Bouillaguet < Charles.Bouillaguet@crans.org> wrote: > > > > > > > > > On Thu, 2006-07-20 at 14:08 -0400, Paul Dufresne wrote: > > > > GIF89a ? > > > :-) yes. > > > > > But the image is elsewhere too! > > > > > And the message is very clear: > > > The extra-terrestrials are using Call-by-Value in their programming > > > language(s). > > > > You can find the gif image at offset 2340188. It's 623 bytes long. > > > > you can extract it with : > > > > tail -c 2340188 codex.umz | head -c 623 > cbv.gif > > > > good luck, folks :) > > > > > > -- > > Entirely Non-Serious team > > > > > > _______________________________________________ > > icfpcontest-discuss mailing list > > icfpcontest-discuss@lists.andrew.cmu.edu > > > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > > > > > > From chad.harrington at gmail.com Fri Jul 21 09:16:32 2006 From: chad.harrington at gmail.com (Chad Harrington) Date: Fri Jul 21 09:16:34 2006 Subject: [icfp-discuss] Contest Materials added in 19 Jul In-Reply-To: <44C0AB03.5010608@schlagsei.de> References: <7b2bf4900607191849v6172d85dm4519be1eab7c7990@mail.gmail.com> <2accc2ff0607201041y3fb9763av2d1adc4019f913d3@mail.gmail.com> <1153418887.13247.266500088@webmail.messagingengine.com> <1153424222.12052.1.camel@zblup> <159b8eea0607201712j5db590ajc4b3bb36278a51bb@mail.gmail.com> <44C0AB03.5010608@schlagsei.de> Message-ID: <5f680c10607210616r57e9c9c8j1e218d617be11da9@mail.gmail.com> On 7/21/06, Daniel Schr?ter wrote: > > Do you mean the one at 1576? 4096 Bytes of grayscale codes? 64 Bytes per > line? > > greetz > Daniel I am curious to know how you smart folks figured out the existence of the the two images. It looked like random bytes to me. Anyone want to share? -- Chad Harrington chad.harrington@gmail.com -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060721/aa4517ea/attachment.html From johannes at albapasser.de Fri Jul 21 09:18:53 2006 From: johannes at albapasser.de (Johannes) Date: Fri Jul 21 09:18:58 2006 Subject: [icfp-discuss] Contest Materials added in 19 Jul In-Reply-To: <5f680c10607210616r57e9c9c8j1e218d617be11da9@mail.gmail.com> References: <7b2bf4900607191849v6172d85dm4519be1eab7c7990@mail.gmail.com> <2accc2ff0607201041y3fb9763av2d1adc4019f913d3@mail.gmail.com> <1153418887.13247.266500088@webmail.messagingengine.com> <1153424222.12052.1.camel@zblup> <159b8eea0607201712j5db590ajc4b3bb36278a51bb@mail.gmail.com> <44C0AB03.5010608@schlagsei.de> <5f680c10607210616r57e9c9c8j1e218d617be11da9@mail.gmail.com> Message-ID: <44C0D43D.9040104@albapasser.de> > I am curious to know how you smart folks figured out the existence of the > the two images. It looked like random bytes to me. Anyone want to share? GIF89 occurs as a substring. If you view the whole file as .pgm (e. g. by preprending the three lines P5, 1024 2000, 255) you see a regular pattern near the beginning From davidad at mit.edu Fri Jul 21 09:20:04 2006 From: davidad at mit.edu (David Dalrymple) Date: Fri Jul 21 09:20:07 2006 Subject: [icfp-discuss] Contest Materials added in 19 Jul In-Reply-To: <5f680c10607210616r57e9c9c8j1e218d617be11da9@mail.gmail.com> References: <7b2bf4900607191849v6172d85dm4519be1eab7c7990@mail.gmail.com> <2accc2ff0607201041y3fb9763av2d1adc4019f913d3@mail.gmail.com> <1153418887.13247.266500088@webmail.messagingengine.com> <1153424222.12052.1.camel@zblup> <159b8eea0607201712j5db590ajc4b3bb36278a51bb@mail.gmail.com> <44C0AB03.5010608@schlagsei.de> <5f680c10607210616r57e9c9c8j1e218d617be11da9@mail.gmail.com> Message-ID: <504b2780607210620r45123e87ta8bfb92d0c0febc8@mail.gmail.com> > I am curious to know how you smart folks figured out the existence of the > the two images. It looked like random bytes to me. Anyone want to share? The first one was rather obvious once you saw "GIF" in `strings` output, but I don't know about this one at 1576... --David From franke.daniel at gmail.com Fri Jul 21 09:23:56 2006 From: franke.daniel at gmail.com (Daniel Franke) Date: Fri Jul 21 09:23:58 2006 Subject: [icfp-discuss] Re: Contest Materials added in 19 Jul In-Reply-To: <5f680c10607210616r57e9c9c8j1e218d617be11da9@mail.gmail.com> References: <7b2bf4900607191849v6172d85dm4519be1eab7c7990@mail.gmail.com> <2accc2ff0607201041y3fb9763av2d1adc4019f913d3@mail.gmail.com> <1153418887.13247.266500088@webmail.messagingengine.com> <1153424222.12052.1.camel@zblup> <159b8eea0607201712j5db590ajc4b3bb36278a51bb@mail.gmail.com> <44C0AB03.5010608@schlagsei.de> <5f680c10607210616r57e9c9c8j1e218d617be11da9@mail.gmail.com> Message-ID: <640ad44b0607210623q79ee11ddqab6a623e2710d9c2@mail.gmail.com> 2006/7/21, Chad Harrington : > On 7/21/06, Daniel Schr?ter wrote: > > > > Do you mean the one at 1576? 4096 Bytes of grayscale codes? 64 Bytes per > > line? > > I am curious to know how you smart folks figured out the existence of the > the two images. It looked like random bytes to me. Anyone want to share? Open the file in your favourite hex editor, for example. Daniel (another one) From dranger at gmail.com Fri Jul 21 09:27:12 2006 From: dranger at gmail.com (Stephen Dranger) Date: Fri Jul 21 09:27:15 2006 Subject: [icfp-discuss] Contest Materials added in 19 Jul In-Reply-To: <504b2780607210620r45123e87ta8bfb92d0c0febc8@mail.gmail.com> References: <2accc2ff0607201041y3fb9763av2d1adc4019f913d3@mail.gmail.com> <1153418887.13247.266500088@webmail.messagingengine.com> <1153424222.12052.1.camel@zblup> <159b8eea0607201712j5db590ajc4b3bb36278a51bb@mail.gmail.com> <44C0AB03.5010608@schlagsei.de> <5f680c10607210616r57e9c9c8j1e218d617be11da9@mail.gmail.com> <504b2780607210620r45123e87ta8bfb92d0c0febc8@mail.gmail.com> Message-ID: <159b8eea0607210627j53f1368k5c8d060335673b9@mail.gmail.com> On 7/21/06, David Dalrymple wrote: > > > I am curious to know how you smart folks figured out the existence of > the > > the two images. It looked like random bytes to me. Anyone want to > share? > > The first one was rather obvious once you saw "GIF" in `strings` > output, but I don't know about this one at 1576... You must not be a fan of ASCII Art then. --Stephen -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060721/77a49cb4/attachment.html From VAUCHER at fermat.eu Fri Jul 21 07:50:05 2006 From: VAUCHER at fermat.eu (VAUCHER Laurent) Date: Fri Jul 21 09:33:45 2006 Subject: [icfp-discuss] Contest Materials added in 19 Jul Message-ID: <200607211350644.SM01464@037f27dd816a40c> Cerebral Blood Volume ? Cannabivarin (http://en.wikipedia.org/wiki/Cannabivarin) From VAUCHER at fermat.eu Fri Jul 21 08:56:06 2006 From: VAUCHER at fermat.eu (VAUCHER Laurent) Date: Fri Jul 21 09:33:45 2006 Subject: [icfp-discuss] Contest Materials added in 19 Jul Message-ID: <55DDB08CC9CD2941A70E8D626789A2C901FF9E3E@ec8l7ljvo9h5dde.hosting.exch> McCartney. ________________________________ De : icfpcontest-discuss-bounces@lists.andrew.cmu.edu [mailto:icfpcontest-discuss-bounces@lists.andrew.cmu.edu] De la part de Abhimanyu Seth Envoy? : 21 July 2006 14:49 ? : Discussion of the 2006 ICFP Programming Contest Objet : Re: [icfp-discuss] Contest Materials added in 19 Jul Also, regarding well done daed si luap, "daed si luap" is the reverse of Paul is dead. Any idea who this Paul is? On 7/21/06, James Darwin Maes The First (a javatarian) wrote: I am sure this is all just to keep us busy until tomorrow morning. That being said, here are a quick list of the strings found (via a strings command matched against a linux words file plus a little looking around) gnoti et quasi occulti well done daed si luap Evan Chan was murdered fnord tsarogy apply MKULTRA abracadabra eval eval so dark the con of man dust solomon rakoczi novus ordo seclorum tell uric plbndetibh u ou rvcofalt ea templar tycon mismatch apply raimundus societas eruditorum ADE anbaric Tga'G abulafi a bad wolf lambda roswell area51 i love bees __42__ lullus surmount currents On 7/20/06, Charles Bouillaguet < Charles.Bouillaguet@crans.org > wrote: On Thu, 2006-07-20 at 14:08 -0400, Paul Dufresne wrote: > > GIF89a ? > :-) yes. > But the image is elsewhere too! > And the message is very clear: > The extra-terrestrials are using Call-by-Value in their programming > language(s). You can find the gif image at offset 2340188. It's 623 bytes long. you can extract it with : tail -c 2340188 codex.umz | head -c 623 > cbv.gif good luck, folks :) -- Entirely Non-Serious team _______________________________________________ icfpcontest-discuss mailing list icfpcontest-discuss@lists.andrew.cmu.edu https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss _______________________________________________ icfpcontest-discuss mailing list icfpcontest-discuss@lists.andrew.cmu.edu https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss -- Regards, Abhimanyu -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060721/7dcb8703/attachment.html From nattfodd at gmail.com Fri Jul 21 11:29:54 2006 From: nattfodd at gmail.com (Alexandre Buisse) Date: Fri Jul 21 11:29:57 2006 Subject: [icfp-discuss] IRC channel Message-ID: <41b037ed0607210829p4bec9894pa64f4b0648732f2@mail.gmail.com> Hi, I just wanted to know if people were already hanging on some IRC channel. I seem to remember #icfpc on freenode from last year but there are only two persons there (including me). So, where are you all hiding? /Alexandre -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060721/8c8e718f/attachment.html From christophe.poucet at gmail.com Fri Jul 21 11:31:39 2006 From: christophe.poucet at gmail.com (Christophe Poucet) Date: Fri Jul 21 11:31:42 2006 Subject: [icfp-discuss] IRC channel In-Reply-To: <41b037ed0607210829p4bec9894pa64f4b0648732f2@mail.gmail.com> References: <41b037ed0607210829p4bec9894pa64f4b0648732f2@mail.gmail.com> Message-ID: <2b0051e00607210831v46e2f62bp6d688b17ba71ec7@mail.gmail.com> Hello, We have a semi-official channel going on freenode at the moment. It's somewhat haskell-oriented. The name is #oasis On 7/21/06, Alexandre Buisse wrote: > > Hi, > I just wanted to know if people were already hanging on some IRC channel. > I seem to remember #icfpc on freenode from last year but there are only two > persons there (including me). > > So, where are you all hiding? > > /Alexandre > > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > > > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060721/52f91faf/attachment.html From christophe.poucet at gmail.com Fri Jul 21 11:32:08 2006 From: christophe.poucet at gmail.com (Christophe Poucet) Date: Fri Jul 21 11:32:11 2006 Subject: [icfp-discuss] IRC channel In-Reply-To: <2b0051e00607210831v46e2f62bp6d688b17ba71ec7@mail.gmail.com> References: <41b037ed0607210829p4bec9894pa64f4b0648732f2@mail.gmail.com> <2b0051e00607210831v46e2f62bp6d688b17ba71ec7@mail.gmail.com> Message-ID: <2b0051e00607210832r3d0ef605p213212bc09a65d2d@mail.gmail.com> Semi-official as far as unofficial things go :P I'm in no way affiliated to the contest On 7/21/06, Christophe Poucet wrote: > > Hello, > > We have a semi-official channel going on freenode at the moment. It's > somewhat haskell-oriented. The name is #oasis > > On 7/21/06, Alexandre Buisse wrote: > > > Hi, > I just wanted to know if people were already hanging on some IRC channel. > I seem to remember #icfpc on freenode from last year but there are only two > persons there (including me). > > So, where are you all hiding? > > /Alexandre > > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > > > > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060721/ebe1e901/attachment-0001.html From flyfree at gmail.com Fri Jul 21 11:32:12 2006 From: flyfree at gmail.com (Zhenzhong Xu) Date: Fri Jul 21 11:32:14 2006 Subject: [icfp-discuss] IRC channel In-Reply-To: <41b037ed0607210829p4bec9894pa64f4b0648732f2@mail.gmail.com> References: <41b037ed0607210829p4bec9894pa64f4b0648732f2@mail.gmail.com> Message-ID: <7b2bf4900607210832g3547a5a2o2b58c43c694ca93f@mail.gmail.com> can you provide the freenode server information again? I can't seem to be able to find it. Z. On 7/21/06, Alexandre Buisse wrote: > Hi, > I just wanted to know if people were already hanging on some IRC channel. I > seem to remember #icfpc on freenode from last year but there are only two > persons there (including me). > > So, where are you all hiding? > > /Alexandre > > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > > > From gregory.t.brown at gmail.com Fri Jul 21 11:32:35 2006 From: gregory.t.brown at gmail.com (Gregory Brown) Date: Fri Jul 21 11:32:35 2006 Subject: [icfp-discuss] IRC channel In-Reply-To: <41b037ed0607210829p4bec9894pa64f4b0648732f2@mail.gmail.com> References: <41b037ed0607210829p4bec9894pa64f4b0648732f2@mail.gmail.com> Message-ID: On 7/21/06, Alexandre Buisse wrote: > Hi, > I just wanted to know if people were already hanging on some IRC channel. I > seem to remember #icfpc on freenode from last year but there are only two > persons there (including me). > > So, where are you all hiding? As long as there isn't already a channel going, I say this one is fine. I've logged in as sandal From flyfree at gmail.com Fri Jul 21 11:32:48 2006 From: flyfree at gmail.com (Zhenzhong Xu) Date: Fri Jul 21 11:32:49 2006 Subject: [icfp-discuss] IRC channel In-Reply-To: <7b2bf4900607210832g3547a5a2o2b58c43c694ca93f@mail.gmail.com> References: <41b037ed0607210829p4bec9894pa64f4b0648732f2@mail.gmail.com> <7b2bf4900607210832g3547a5a2o2b58c43c694ca93f@mail.gmail.com> Message-ID: <7b2bf4900607210832sd52a898mf473997af645f226@mail.gmail.com> Or is the organizer planning to have an "official" channe;? On 7/21/06, Zhenzhong Xu wrote: > can you provide the freenode server information again? I can't seem to > be able to find it. > > Z. > > On 7/21/06, Alexandre Buisse wrote: > > Hi, > > I just wanted to know if people were already hanging on some IRC channel. I > > seem to remember #icfpc on freenode from last year but there are only two > > persons there (including me). > > > > So, where are you all hiding? > > > > /Alexandre > > > > _______________________________________________ > > icfpcontest-discuss mailing list > > icfpcontest-discuss@lists.andrew.cmu.edu > > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > > > > > > > From christophe.poucet at gmail.com Fri Jul 21 11:37:41 2006 From: christophe.poucet at gmail.com (Christophe Poucet) Date: Fri Jul 21 11:37:43 2006 Subject: [icfp-discuss] IRC channel In-Reply-To: <7b2bf4900607210832g3547a5a2o2b58c43c694ca93f@mail.gmail.com> References: <41b037ed0607210829p4bec9894pa64f4b0648732f2@mail.gmail.com> <7b2bf4900607210832g3547a5a2o2b58c43c694ca93f@mail.gmail.com> Message-ID: <2b0051e00607210837j411a92b6xa2cb89a4240fdcf1@mail.gmail.com> Hi, The server information is: irc.freenode.net The channel is: #oasis (the reason for the name is that it was originally our team channel but we've migrated to a private channel) On 7/21/06, Zhenzhong Xu wrote: > > can you provide the freenode server information again? I can't seem to > be able to find it. > > Z. > > On 7/21/06, Alexandre Buisse wrote: > > Hi, > > I just wanted to know if people were already hanging on some IRC > channel. I > > seem to remember #icfpc on freenode from last year but there are only > two > > persons there (including me). > > > > So, where are you all hiding? > > > > /Alexandre > > > > _______________________________________________ > > icfpcontest-discuss mailing list > > icfpcontest-discuss@lists.andrew.cmu.edu > > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > > > > > > > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060721/3ad0712a/attachment.html From aguilar.james at gmail.com Fri Jul 21 12:47:10 2006 From: aguilar.james at gmail.com (James Aguilar) Date: Fri Jul 21 12:47:11 2006 Subject: [icfp-discuss] Virtual machine specification Message-ID: Hey all, Has anyone been able to figure out if the codex is just one giant program or a series of smaller ones? Do we have to search for entry points or should a complete machine simply be able to run the entire program without stopping? Yours, James Aguilar -- [?] James Aguilar [@] 18100 NE 95th St. #RR3088 / Redmond, WA 98052 [#] 314 494 0450 [!] In the wind and the rain, my darling, say goodbye -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060721/970a67ae/attachment.html From muranushi at gmail.com Fri Jul 21 13:14:21 2006 From: muranushi at gmail.com (Takayuki Muranushi) Date: Fri Jul 21 13:14:24 2006 Subject: [icfp-discuss] Virtual machine specification In-Reply-To: References: Message-ID: I find a few statements ambiguous on the Specification ... what do you think? Ambiguity 1 #5. Division. The register A receives the value in register B divided by the value in register C, if any, where each quantity is treated treated as an unsigned 32 bit number. what does "if any" mean? "if any" does not appear in addition nor multiplication. It may mean if there is any integer that equals B/C, then put it into A. otherwise do nothing if there is any mathematical value for B/C (or, unless C!=0) put it into A. or other possibilities ... Ambiguity 2. #8. Allocation. A new array is created with a capacity of platters commensurate to the value in the register C. This new array is initialized entirely with platters holding the value 0. A bit pattern not consisting of exclusively the 0 bit, and that identifies no other active allocated array, is placed in the B register. How do you know the identifier of created array? Is it written in resister A? Ambiguity 3 #13. Orthography. The value indicated is loaded into the register A forthwith. You know, the constant has only 25 bits. What happens to higher bits "VUTSRQP" of resister A? Do they remain unchanged or set to 0, or set to the 25th bit of the constant .... May all the contestants do well, Takayuki Muranushi From aguilar.james at gmail.com Fri Jul 21 13:18:54 2006 From: aguilar.james at gmail.com (James Aguilar) Date: Fri Jul 21 13:18:57 2006 Subject: [icfp-discuss] Virtual machine specification In-Reply-To: References: Message-ID: On 7/21/06, Takayuki Muranushi wrote: > > I find a few statements ambiguous on the Specification ... what do you > think? > > Ambiguity 1 > #5. Division. > The register A receives the value in register B > divided by the value in register C, if any, where > each quantity is treated treated as an unsigned 32 > bit number. > what does "if any" mean? "if any" does not appear in addition nor > multiplication. > It may mean > if there is any integer that equals B/C, then put it into A. > otherwise do nothing > if there is any mathematical value for B/C (or, unless C!=0) put it into > A. > or other possibilities ... Haven't got an answer for you on this one. Ambiguity 2. > #8. Allocation. > > A new array is created with a capacity of platters > commensurate to the value in the register C. This > new array is initialized entirely with platters > holding the value 0. A bit pattern not consisting of > exclusively the 0 bit, and that identifies no other > active allocated array, is placed in the B register. > How do you know the identifier of created array? Is it written in resister > A? You can call the array whatever you want, as long as it's not the same as another identified array. The "name" is placed into register B so that the machine can read it. Ambiguity 3 > #13. Orthography. > The value indicated is loaded into the register A > forthwith. > You know, the constant has only 25 bits. What happens to higher bits > "VUTSRQP" of resister A? Do they remain unchanged or set to 0, or set > to the 25th bit of the constant .... I would assume they are zeroed, but this also appears ambiguous. Yours, James Aguilar -- [?] James Aguilar [@] 18100 NE 95th St. #RR3088 / Redmond, WA 98052 [#] 314 494 0450 [!] In the wind and the rain, my darling, say goodbye -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060721/b17c1c93/attachment.html From aguilar.james at gmail.com Fri Jul 21 13:27:01 2006 From: aguilar.james at gmail.com (James Aguilar) Date: Fri Jul 21 13:27:04 2006 Subject: [icfp-discuss] Virtual machine specification In-Reply-To: References: Message-ID: More ambiguities: The '0' array shall be the most sublime choice for loading, and shall be handled with the utmost velocity. What in the world does that mean? #11. Input. The universal machine waits for input on the console. When input arrives, the register C is loaded with the input, which must be between and including 0 and 255. If the end of input has been signaled, then the register C is endowed with a uniform value pattern where every place is pregnant with the 1 bit. What if our console requires a newline before end of input? Should we use a redirected file with no newline, or is the newline expected? Yours, James Aguilar -- [?] James Aguilar [@] 18100 NE 95th St. #RR3088 / Redmond, WA 98052 [#] 314 494 0450 [!] In the wind and the rain, my darling, say goodbye -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060721/da10db22/attachment.html From crary at cs.cmu.edu Fri Jul 21 13:29:39 2006 From: crary at cs.cmu.edu (Karl Crary) Date: Fri Jul 21 13:29:41 2006 Subject: [icfp-discuss] Virtual machine specification In-Reply-To: References: Message-ID: <44C10F03.3090204@cs.cmu.edu> Takayuki Muranushi wrote: > I find a few statements ambiguous on the Specification ... what do you > think? > > Ambiguity 1 > #5. Division. > The register A receives the value in register B > divided by the value in register C, if any, where > each quantity is treated treated as an unsigned 32 > bit number. > what does "if any" mean? "if any" does not appear in addition nor > multiplication. > It may mean > if there is any integer that equals B/C, then put it into A. > otherwise do nothing > if there is any mathematical value for B/C (or, unless C!=0) put it > into A. > or other possibilities ... The phrase "if any" refers to the possibility of division-by-zero. As the spec states at the end, the behavior in the event of division-by-zero is undefined. > Ambiguity 2. > #8. Allocation. > > A new array is created with a capacity of platters > commensurate to the value in the register C. This > new array is initialized entirely with platters > holding the value 0. A bit pattern not consisting of > exclusively the 0 bit, and that identifies no other > active allocated array, is placed in the B register. > How do you know the identifier of created array? Is it written in > resister A? The UM may create any identifier, as long as it is not zero. > Ambiguity 3 > #13. Orthography. > The value indicated is loaded into the register A > forthwith. > You know, the constant has only 25 bits. What happens to higher bits > "VUTSRQP" of resister A? Do they remain unchanged or set to 0, or set > to the 25th bit of the constant .... The 25 bits in the instruciton specify a 32-bit constant. The spec states at the beginning that all numbers are unsigned. Karl Crary (for the contest organizers) From Alain.Frisch at inria.fr Fri Jul 21 13:53:36 2006 From: Alain.Frisch at inria.fr (Alain Frisch) Date: Fri Jul 21 13:53:39 2006 Subject: [icfp-discuss] Virtual machine specification In-Reply-To: <44C10F03.3090204@cs.cmu.edu> References: <44C10F03.3090204@cs.cmu.edu> Message-ID: <44C114A0.6070201@inria.fr> Karl Crary wrote: > The UM may create any identifier, as long as it is not zero. I guess the integer put in B is the identifier of the newly allocated array. The spec does not say it explicitly... -- Alain From davidad at mit.edu Fri Jul 21 14:04:51 2006 From: davidad at mit.edu (David Dalrymple) Date: Fri Jul 21 14:04:54 2006 Subject: [icfp-discuss] Virtual machine specification In-Reply-To: <44C114A0.6070201@inria.fr> References: <44C10F03.3090204@cs.cmu.edu> <44C114A0.6070201@inria.fr> Message-ID: <504b2780607211104g584036afs929d4d72be3dcccd@mail.gmail.com> > I guess the integer put in B is the identifier of the newly allocated > array. The spec does not say it explicitly... Very true. It could be a random number completely unassociated with the newly allocated array and still be in line with the explicit spec so long as it doesn't reference a previously extant array... --David From per.gustafsson at it.uu.se Fri Jul 21 14:03:12 2006 From: per.gustafsson at it.uu.se (Per Gustafsson) Date: Fri Jul 21 14:27:43 2006 Subject: [icfp-discuss] Virtual machine specification In-Reply-To: References: Message-ID: <44C116E0.3040204@it.uu.se> James Aguilar wrote: > More ambiguities: > > The '0' array shall be the most sublime choice for > loading, and shall be handled with the utmost > velocity. > > What in the world does that mean? I suppose that you should not duplicate the array if B contains '0' but only change the execution finger > > #11. Input. > > The universal machine waits for input on the console. > When input arrives, the register C is loaded with the > input, which must be between and including 0 and 255. > If the end of input has been signaled, then the > register C is endowed with a uniform value pattern > where every place is pregnant with the 1 bit. > > What if our console requires a newline before end of input? Should we > use a redirected file with no newline, or is the newline expected? > > Yours, > James Aguilar > > -- > [?] James Aguilar > [@] 18100 NE 95th St. #RR3088 / Redmond, WA 98052 > [#] 314 494 0450 > [!] In the wind and the rain, my darling, say goodbye > >------------------------------------------------------------------------ > >_______________________________________________ >icfpcontest-discuss mailing list >icfpcontest-discuss@lists.andrew.cmu.edu >https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > > Per From franke.daniel at gmail.com Fri Jul 21 15:28:02 2006 From: franke.daniel at gmail.com (Daniel Franke) Date: Fri Jul 21 15:28:11 2006 Subject: [icfp-discuss] so many registers ... Message-ID: <200607212128.02420.franke.daniel@gmail.com> A non-native speaker seeks for confirmation: There are registers and registers: -> Eight distinct general-purpose registers, capable of holding one platter each. -> Each Standard Operator performs an errand using three registers, called A, B, and C. So, if operation 0 states: -> The register A receives the value in register B, unless the register C contains 0. It is meant: "The general purpose register with the number specified by operand A of this platter is set to the value of the general purpose register named by operand B if and only if the general purpose register pointed to by operand C does not equal 0" This correct? Thanks in advance for replies. Daniel From dankna at gmail.com Fri Jul 21 15:30:12 2006 From: dankna at gmail.com (Dan Knapp) Date: Fri Jul 21 15:30:14 2006 Subject: [icfp-discuss] so many registers ... In-Reply-To: <200607212128.02420.franke.daniel@gmail.com> References: <200607212128.02420.franke.daniel@gmail.com> Message-ID: <5961bfde0607211230h72f3f655xe7e413131ceac6ab@mail.gmail.com> As I understand it, yes, you have it right. On 7/21/06, Daniel Franke wrote: > > > A non-native speaker seeks for confirmation: > > There are registers and registers: > > -> Eight distinct general-purpose registers, capable of holding one > platter each. > > -> Each Standard Operator performs an errand using three registers, > called A, B, and C. > > > So, if operation 0 states: > > -> The register A receives the value in register B, > unless the register C contains 0. > > It is meant: > "The general purpose register with the number specified by operand A of > this > platter is set to the value of the general purpose register named by > operand B if and only if the general purpose register pointed to by > operand C does not equal 0" > > This correct? > Thanks in advance for replies. > > Daniel > > > > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > -- Dan Knapp -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060721/8a4a1710/attachment.html From blo.b at infonie.fr Fri Jul 21 15:31:16 2006 From: blo.b at infonie.fr (Laurent Vaucher) Date: Fri Jul 21 15:31:20 2006 Subject: [icfp-discuss] so many registers ... In-Reply-To: <200607212128.02420.franke.daniel@gmail.com> References: <200607212128.02420.franke.daniel@gmail.com> Message-ID: <1669C81C-BB65-4BC3-84B6-B722F94A1285@infonie.fr> > "The general purpose register with the number specified by > operand A of this > platter is set to the value of the general purpose register > named by > operand B if and only if the general purpose register pointed > to by > operand C does not equal 0" > > This correct? That's how I understand it too. From rvandam00 at gmail.com Fri Jul 21 15:31:50 2006 From: rvandam00 at gmail.com (Robert Van Dam) Date: Fri Jul 21 15:31:52 2006 Subject: [icfp-discuss] so many registers ... In-Reply-To: <200607212128.02420.franke.daniel@gmail.com> References: <200607212128.02420.franke.daniel@gmail.com> Message-ID: <1815f00607211231n56b6dcb3ua743a89c5930c15f@mail.gmail.com> I'm a native enlish speaker and it took me a while to figure out. Yes, you are correct (I think ;). Rob On 7/21/06, Daniel Franke wrote: > > A non-native speaker seeks for confirmation: > > There are registers and registers: > > -> Eight distinct general-purpose registers, capable of holding one > platter each. > > -> Each Standard Operator performs an errand using three registers, > called A, B, and C. > > > So, if operation 0 states: > > -> The register A receives the value in register B, > unless the register C contains 0. > > It is meant: > "The general purpose register with the number specified by operand A of this > platter is set to the value of the general purpose register named by > operand B if and only if the general purpose register pointed to by > operand C does not equal 0" > > This correct? > Thanks in advance for replies. > > Daniel > > > > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > From jacobdp at gmail.com Fri Jul 21 15:34:27 2006 From: jacobdp at gmail.com (Jacob Potter) Date: Fri Jul 21 15:48:40 2006 Subject: [icfp-discuss] so many registers ... In-Reply-To: <200607212128.02420.franke.daniel@gmail.com> References: <200607212128.02420.franke.daniel@gmail.com> Message-ID: On 7/21/06, Daniel Franke wrote: > It is meant: > "The general purpose register with the number specified by operand A of this > platter is set to the value of the general purpose register named by > operand B if and only if the general purpose register pointed to by > operand C does not equal 0" Yep. For example, an instruction with Operator = 0, A = 1, B = 2, C = 3 means "set register 1 equal to the sum of register 2 + register 3". - Jacob From davidad at mit.edu Fri Jul 21 15:51:40 2006 From: davidad at mit.edu (David Dalrymple) Date: Fri Jul 21 15:51:41 2006 Subject: [icfp-discuss] so many registers ... In-Reply-To: References: <200607212128.02420.franke.daniel@gmail.com> Message-ID: <504b2780607211251u301e4dc3sbe69652eec9b41c6@mail.gmail.com> > Yep. For example, an instruction with Operator = 0, A = 1, B = 2, C = > 3 means "set register 1 equal to the sum of register 2 + register 3". You mean operator = 3, right? (Operator 0 = conditional move) --David From katolazzus at katolaz.homeunix.net Fri Jul 21 16:03:34 2006 From: katolazzus at katolaz.homeunix.net (Katolaz) Date: Fri Jul 21 16:03:38 2006 Subject: [me@katolaz.homeunix.net: [me@katolaz.homeunix.net: Re: [icfp-discuss] Material....:-))]] In-Reply-To: References: <20060720193315.GB18765@wontolla> Message-ID: <20060721200334.GA25300@wontolla> On Thu, Jul 20, 2006 at 03:39:01PM -0500, James Darwin Maes The First (a javatarian) wrote: > wow > > guy who hates bad email subject lines > > :-) Also I hate bad subjects. It would be better if someone could fix the Mailman configuration for the list, allowing the Reply-To to point directly to the list, instead of the sender :-) HND From james at grayproductions.net Fri Jul 21 16:17:06 2006 From: james at grayproductions.net (James Edward Gray II) Date: Fri Jul 21 16:17:11 2006 Subject: [icfp-discuss] Error Condition Message-ID: <2FDC4DE7-0E11-414E-A1F8-87B80CBF0DF0@grayproductions.net> Would it also be a machine failure if the Allocation operator was called when all 32 bit integer names for arrays are taken? James Edward Gray II From tom7 at cs.cmu.edu Fri Jul 21 16:21:37 2006 From: tom7 at cs.cmu.edu (Tom Murphy VII) Date: Fri Jul 21 16:21:36 2006 Subject: [icfp-discuss] Error Condition In-Reply-To: <2FDC4DE7-0E11-414E-A1F8-87B80CBF0DF0@grayproductions.net> References: <2FDC4DE7-0E11-414E-A1F8-87B80CBF0DF0@grayproductions.net> Message-ID: <44C13751.5020901@cs.cmu.edu> Yes, but I don't think you'll need to worry about this. Tom > Would it also be a machine failure if the Allocation operator was > called when all 32 bit integer names for arrays are taken? > > James Edward Gray II > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > From rvandam00 at gmail.com Fri Jul 21 16:31:23 2006 From: rvandam00 at gmail.com (Robert Van Dam) Date: Fri Jul 21 16:31:24 2006 Subject: [icfp-discuss] ASCII input/output Message-ID: <1815f00607211331k3f05fe44r3ccae9187d7c1f8f@mail.gmail.com> Unless you're on an old MS-DOS machine still using the IBM codepage with smileys and playing cards, it might be difficult to display some characters in a '1x1 character resolution console'. Is there any particular way in which we should go about displaying \000, etc in one character? Or perhaps the machine just won't output such characters? (that seems unlikely but...) Rob From aguilar.james at gmail.com Fri Jul 21 16:34:00 2006 From: aguilar.james at gmail.com (James Aguilar) Date: Fri Jul 21 16:34:02 2006 Subject: [icfp-discuss] ASCII input/output In-Reply-To: <1815f00607211331k3f05fe44r3ccae9187d7c1f8f@mail.gmail.com> References: <1815f00607211331k3f05fe44r3ccae9187d7c1f8f@mail.gmail.com> Message-ID: On 7/21/06, Robert Van Dam wrote: > > Is there any > particular way in which we should go about displaying \000, etc in one > character? I'd be willing to hazard a guess that the machine does not ever output such characters. :P -- [?] James Aguilar [@] 18100 NE 95th St. #RR3088 / Redmond, WA 98052 [#] 314 494 0450 [!] In the wind and the rain, my darling, say goodbye -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060721/186442b4/attachment.html From crary at cs.cmu.edu Fri Jul 21 16:40:30 2006 From: crary at cs.cmu.edu (Karl Crary) Date: Fri Jul 21 16:40:31 2006 Subject: [icfp-discuss] ASCII input/output In-Reply-To: <1815f00607211331k3f05fe44r3ccae9187d7c1f8f@mail.gmail.com> References: <1815f00607211331k3f05fe44r3ccae9187d7c1f8f@mail.gmail.com> Message-ID: <44C13BBE.5070500@cs.cmu.edu> Your implementation should handle all characters from 0 to 255 uniformly. It is important that you do not do any special processing on unprintable characters. Karl Crary (for the organizers) Robert Van Dam wrote: > Unless you're on an old MS-DOS machine still using the IBM codepage > with smileys and playing cards, it might be difficult to display some > characters in a '1x1 character resolution console'. Is there any > particular way in which we should go about displaying \000, etc in one > character? Or perhaps the machine just won't output such characters? > (that seems unlikely but...) > > Rob > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > From azul at freaks-unidos.net Fri Jul 21 17:07:39 2006 From: azul at freaks-unidos.net (Alejandro Forero Cuervo) Date: Fri Jul 21 16:52:54 2006 Subject: [icfp-discuss] Virtual machine specification In-Reply-To: References: Message-ID: <20060721210739.GQ15374@bachue.com> The spec reads: > The execution finger is placed to indicate the platter of this array > that is described by the offset given in C, where the value 0 > denotes the first platter, 1 the second, et cetera. This means that the execution finger is placed on the Cth platter of the new array, right? Or should it be set to its current value (in the original array) plus the offset C? We find the first to be more logical but we noticed that if we use the second interpretation, we get a "funny" result... Alejo, still trying to figure out what our team's wrong assumption is... http://azul.freaks-unidos.net/ From jacob at durbatuluk.us Fri Jul 21 16:53:07 2006 From: jacob at durbatuluk.us (Jacob Potter) Date: Fri Jul 21 16:53:09 2006 Subject: [icfp-discuss] so many registers ... In-Reply-To: <504b2780607211251u301e4dc3sbe69652eec9b41c6@mail.gmail.com> References: <200607212128.02420.franke.daniel@gmail.com> <504b2780607211251u301e4dc3sbe69652eec9b41c6@mail.gmail.com> Message-ID: On 7/21/06, David Dalrymple wrote: > > You mean operator = 3, right? (Operator 0 = conditional move) D'oh! From popiel at wolfskeep.com Fri Jul 21 17:12:04 2006 From: popiel at wolfskeep.com (T. Alexander Popiel) Date: Fri Jul 21 17:12:06 2006 Subject: [icfp-discuss] Virtual machine specification In-Reply-To: <20060721210739.GQ15374@bachue.com> References: <20060721210739.GQ15374@bachue.com> Message-ID: <20060721211204.3BC122DDB9@cashew.wolfskeep.com> In message: <20060721210739.GQ15374@bachue.com> Alejandro Forero Cuervo writes: > >This means that the execution finger is placed on the Cth platter of >the new array, right? Or should it be set to its current value (in >the original array) plus the offset C? We find the first to be more >logical but we noticed that if we use the second interpretation, we >get a "funny" result... That "funny" result would be a failure of the self-check, I think. It's even trying to tell you what you did wrong. I just wish that the self-checks were a little more thorough (I'm having issues with integer math, myself...) In any case, the first interpretation (with C as an absolute address into the new array on LOADPROG) seems to be more useful. - Alex From gus-icfp at guxx.net Fri Jul 21 17:17:43 2006 From: gus-icfp at guxx.net (G.Schoepp) Date: Fri Jul 21 17:17:47 2006 Subject: [icfp-discuss] LOAD and array duplication Message-ID: <44C14477.9080301@guxx.net> Hi, an instruction LOAD A:3,B:6,C:0 with register content Reg0 = 13 ... Reg6 = 0 should duplicate array no 0, replace array no 0 by array no 0, and set the "execution finger" to 13. Is this right? Guido From Alain.Frisch at inria.fr Fri Jul 21 17:18:56 2006 From: Alain.Frisch at inria.fr (Alain Frisch) Date: Fri Jul 21 17:19:02 2006 Subject: [icfp-discuss] spirit of the contest? Message-ID: <44C144C0.9090905@inria.fr> Hello, The rules say: # All participants must adhere to the spirit of the contest. For example, teams that attempt to forge publications or manipulate the behavior of the codex will be disqualified. In order to comply with this rule, we would like some better definition of the spirit of the contest. Is it ok, e.g. to reverse-engineer the UMIX OS in order to disable the password check? -- Alain From aguilar.james at gmail.com Fri Jul 21 17:19:41 2006 From: aguilar.james at gmail.com (James Aguilar) Date: Fri Jul 21 17:19:45 2006 Subject: [icfp-discuss] Virtual machine specification In-Reply-To: <20060721211204.3BC122DDB9@cashew.wolfskeep.com> References: <20060721210739.GQ15374@bachue.com> <20060721211204.3BC122DDB9@cashew.wolfskeep.com> Message-ID: All, Another question: if we do LoadProgram then should the finger be advanced afterward this cycle as it would be on any other instruction? Or should it stay at the loaded position? Yours, James Aguilar -- [?] James Aguilar [@] 18100 NE 95th St. #RR3088 / Redmond, WA 98052 [#] 314 494 0450 [!] In the wind and the rain, my darling, say goodbye -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060721/496f2f5c/attachment-0001.html From aguilar.james at gmail.com Fri Jul 21 17:23:29 2006 From: aguilar.james at gmail.com (James Aguilar) Date: Fri Jul 21 17:23:31 2006 Subject: [icfp-discuss] LOAD and array duplication In-Reply-To: <44C14477.9080301@guxx.net> References: <44C14477.9080301@guxx.net> Message-ID: On 7/21/06, G.Schoepp wrote: > > Hi, > > an instruction > LOAD A:3,B:6,C:0 > with register content > Reg0 = 13 > ... > Reg6 = 0 > should duplicate array no 0, replace array no 0 by array no 0, and set > the "execution finger" to 13. > > Is this right? Guido, Only thing is, what is the point of duplicating the array? If it's at location zero, you're just going to overwrite the previous one anyway, because there is no particular place to put it. I think you should just shift the finger. Yours, James Aguilar -- [?] James Aguilar [@] 18100 NE 95th St. #RR3088 / Redmond, WA 98052 [#] 314 494 0450 [!] In the wind and the rain, my darling, say goodbye -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060721/a0d283bc/attachment.html From bartoschek at gmx.de Fri Jul 21 17:24:30 2006 From: bartoschek at gmx.de (Christoph Bartoschek) Date: Fri Jul 21 17:24:33 2006 Subject: [icfp-discuss] Two errors in the specification Message-ID: <200607212324.30515.bartoschek@gmx.de> Hi, I see two errors: 1. The section "Behavior" never states when an operator should be executed. 2. In "Load Program" one should set the finger to a new offset. But before the operator is discharged one has to increment finger. But as far as I see it one should not increment the finger if one gets "Load Program". Christoph From popiel at wolfskeep.com Fri Jul 21 17:26:55 2006 From: popiel at wolfskeep.com (T. Alexander Popiel) Date: Fri Jul 21 17:27:01 2006 Subject: [icfp-discuss] LOAD and array duplication In-Reply-To: <44C14477.9080301@guxx.net> References: <44C14477.9080301@guxx.net> Message-ID: <20060721212655.B82562DDB9@cashew.wolfskeep.com> In message: <44C14477.9080301@guxx.net> "G.Schoepp" writes: >Hi, > >an instruction > LOAD A:3,B:6,C:0 >with register content > Reg0 = 13 > ... > Reg6 = 0 >should duplicate array no 0, replace array no 0 by array no 0, and set >the "execution finger" to 13. > >Is this right? I believe so. - Alex From azul at freaks-unidos.net Fri Jul 21 17:45:28 2006 From: azul at freaks-unidos.net (Alejandro Forero Cuervo) Date: Fri Jul 21 17:30:38 2006 Subject: [icfp-discuss] Virtual machine specification In-Reply-To: References: <20060721210739.GQ15374@bachue.com> <20060721211204.3BC122DDB9@cashew.wolfskeep.com> Message-ID: <20060721214528.GR15374@bachue.com> > Another question: if we do LoadProgram then should the finger be advanced > afterward this cycle as it would be on any other instruction? Or should it > stay at the loaded position? I think it should not be advanced. We are advancing it *before* we execute the instruction (to make up for this). By the way, the spec reads: > The execution finger is placed to indicate the platter of this array > that is described by the offset given in C, where the value 0 > denotes the first platter, 1 the second, et cetera. I think it would be far more clear to read "the offset given by register C". This would have saved us quite a while. *sigh Alejo. http://azul.freaks-unidos.net/ From crary at cs.cmu.edu Fri Jul 21 17:32:15 2006 From: crary at cs.cmu.edu (Karl Crary) Date: Fri Jul 21 17:32:16 2006 Subject: [icfp-discuss] Virtual machine specification In-Reply-To: References: <20060721210739.GQ15374@bachue.com> <20060721211204.3BC122DDB9@cashew.wolfskeep.com> Message-ID: <44C147DF.3050608@cs.cmu.edu> The specification covers this point; the execution finger is advanced before the instruction is executed. Therefore, after the instruction, the finger will be at the loaded position. Karl Crary (for the organizers) James Aguilar wrote: > All, > > Another question: if we do LoadProgram then should the finger be > advanced afterward this cycle as it would be on any other > instruction? Or should it stay at the loaded position? > > Yours, > James Aguilar > > -- > [?] James Aguilar > [@] 18100 NE 95th St. #RR3088 / Redmond, WA 98052 > [#] 314 494 0450 > [!] In the wind and the rain, my darling, say goodbye > ------------------------------------------------------------------------ > > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > From iicfp at yahoo.com Fri Jul 21 16:46:49 2006 From: iicfp at yahoo.com (icfp icfp) Date: Fri Jul 21 17:32:47 2006 Subject: [icfp-discuss] Registration problems Message-ID: <20060721204649.79495.qmail@web55705.mail.re3.yahoo.com> I have been trying to register for the past hour for the contest, team name iicfp, this email. I receive a success web page, but then do not receive the promised email with the encryption key. This email address works fine. This is handicapping me, since the simulator is done and has passed the self-test, and I cannot proceed. Anyone there who can fix this? __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060721/5bc88e30/attachment.html From tom7 at cs.cmu.edu Fri Jul 21 17:32:55 2006 From: tom7 at cs.cmu.edu (Tom Murphy VII) Date: Fri Jul 21 17:32:54 2006 Subject: [icfp-discuss] spirit of the contest? In-Reply-To: <44C144C0.9090905@inria.fr> References: <44C144C0.9090905@inria.fr> Message-ID: <44C14807.6080703@cs.cmu.edu> Changing the behavior of the Codex is specifically disallowed. We need to be able to reproduce your results in the original copy of the codex with a compliant version of the universal machine. That said, since you wrote the universal machine, it would be okay to observe the behavior of UMIX--without changing it--in order to learn what it does. In our opinion it will be easier to solve the problems without any reverse-engineering of the UM binaries. Tom >Hello, > >The rules say: > ># All participants must adhere to the spirit of the contest. For >example, teams that attempt to forge publications or manipulate the >behavior of the codex will be disqualified. > > >In order to comply with this rule, we would like some better definition >of the spirit of the contest. Is it ok, e.g. to reverse-engineer the >UMIX OS in order to disable the password check? > > > >-- Alain >_______________________________________________ >icfpcontest-discuss mailing list >icfpcontest-discuss@lists.andrew.cmu.edu >https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > > > From aguilar.james at gmail.com Fri Jul 21 17:34:38 2006 From: aguilar.james at gmail.com (James Aguilar) Date: Fri Jul 21 17:34:41 2006 Subject: [icfp-discuss] Virtual machine specification In-Reply-To: <44C147DF.3050608@cs.cmu.edu> References: <20060721210739.GQ15374@bachue.com> <20060721211204.3BC122DDB9@cashew.wolfskeep.com> <44C147DF.3050608@cs.cmu.edu> Message-ID: On 7/21/06, Karl Crary wrote: > > The specification covers this point; the execution finger is advanced > before the instruction is executed. Therefore, after the instruction, > the finger will be at the loaded position. > > Karl Crary > (for the organizers) Oh, I see. "Discharged," means "executed," not "gotten rid of." -- James -- [?] James Aguilar [@] 18100 NE 95th St. #RR3088 / Redmond, WA 98052 [#] 314 494 0450 [!] In the wind and the rain, my darling, say goodbye -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060721/22b8349d/attachment.html From rwh at cs.cmu.edu Fri Jul 21 17:38:36 2006 From: rwh at cs.cmu.edu (Robert Harper) Date: Fri Jul 21 17:38:51 2006 Subject: [icfp-discuss] spirit of the contest? In-Reply-To: <44C144C0.9090905@inria.fr> References: <44C144C0.9090905@inria.fr> Message-ID: I would say that this is not in the spirit of the contest. Bob Harper On Jul 21, 2006, at 5:18 PM, Alain Frisch wrote: > Hello, > > The rules say: > > # All participants must adhere to the spirit of the contest. For > example, teams that attempt to forge publications or manipulate the > behavior of the codex will be disqualified. > > > In order to comply with this rule, we would like some better > definition > of the spirit of the contest. Is it ok, e.g. to reverse-engineer the > UMIX OS in order to disable the password check? > > > > -- Alain > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss From iicfp at yahoo.com Fri Jul 21 17:45:33 2006 From: iicfp at yahoo.com (icfp icfp) Date: Fri Jul 21 17:47:13 2006 Subject: [icfp-discuss] Registration problems In-Reply-To: <20060721204649.79495.qmail@web55705.mail.re3.yahoo.com> Message-ID: <20060721214534.36526.qmail@web55710.mail.re3.yahoo.com> Still the same problem. No encryption key received. icfp icfp wrote: I have been trying to register for the past hour for the contest, team name iicfp, this email. I receive a success web page, but then do not receive the promised email with the encryption key. This email address works fine. This is handicapping me, since the simulator is done and has passed the self-test, and I cannot proceed. Anyone there who can fix this? __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com _______________________________________________ icfpcontest-discuss mailing list icfpcontest-discuss@lists.andrew.cmu.edu https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss --------------------------------- Yahoo! Messenger with Voice. Make PC-to-Phone Calls to the US (and 30+ countries) for 2?/min or less. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060721/b82f00c1/attachment.html From crary at cs.cmu.edu Fri Jul 21 17:59:03 2006 From: crary at cs.cmu.edu (Karl Crary) Date: Fri Jul 21 17:59:04 2006 Subject: [icfp-discuss] Registration problems In-Reply-To: <20060721214534.36526.qmail@web55710.mail.re3.yahoo.com> References: <20060721214534.36526.qmail@web55710.mail.re3.yahoo.com> Message-ID: <44C14E27.7060506@cs.cmu.edu> I believe the problem is on your end. We have sent you multiple emails with your team information. Karl Crary icfp icfp wrote: > Still the same problem. No encryption key received. > > */icfp icfp /* wrote: > > I have been trying to register for the past hour for the contest, > team name iicfp, this email. I receive a success web page, but > then do not receive the promised email with the encryption key. > This email address works fine. This is handicapping me, since the > simulator is done and has passed the self-test, and I cannot > proceed. > > Anyone there who can fix this? > __________________________________________________ > Do You Yahoo!? > Tired of spam? Yahoo! Mail has the best spam protection around > http://mail.yahoo.com _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > > From franke.daniel at gmail.com Fri Jul 21 18:00:49 2006 From: franke.daniel at gmail.com (Daniel Franke) Date: Fri Jul 21 18:00:57 2006 Subject: [icfp-discuss] "LOADPROG off (high)" Message-ID: <200607220000.49273.franke.daniel@gmail.com> After I got the registers sorted out (btw, thanks Guido), I got the message above and a halt instruction - WTF? Someone seen this before? Regards Daniel P.S. I'm not asking how to work around this, just whether someone hit this wall as well ... P.P.S. Thanks again for all the answers regarding the registers. From ij7b3u502 at sneakemail.com Fri Jul 21 18:14:16 2006 From: ij7b3u502 at sneakemail.com (vbzoli) Date: Fri Jul 21 18:14:17 2006 Subject: [icfp-discuss] Division Message-ID: <25512-98643@sneakemail.com> Division should truncate its result? e.g. 3 / 2 => 1? VOROSBARANYI Zoltan From popiel at wolfskeep.com Fri Jul 21 18:15:31 2006 From: popiel at wolfskeep.com (T. Alexander Popiel) Date: Fri Jul 21 18:15:32 2006 Subject: [icfp-discuss] "LOADPROG off (high)" In-Reply-To: <200607220000.49273.franke.daniel@gmail.com> References: <200607220000.49273.franke.daniel@gmail.com> Message-ID: <20060721221531.8ADD42DDC3@cashew.wolfskeep.com> In message: <200607220000.49273.franke.daniel@gmail.com> Daniel Franke writes: > >After I got the registers sorted out (btw, thanks Guido), >I got the message above and a halt instruction - WTF? > >Someone seen this before? Yes, it's been seen. :-) It's one of the self-check failures. Read through the conversation regarding the sequence of execution. - Alex From franke.daniel at gmail.com Fri Jul 21 18:32:37 2006 From: franke.daniel at gmail.com (Daniel Franke) Date: Fri Jul 21 18:32:47 2006 Subject: [icfp-discuss] "LOADPROG off (high)" In-Reply-To: <20060721221531.8ADD42DDC3@cashew.wolfskeep.com> References: <200607220000.49273.franke.daniel@gmail.com> <20060721221531.8ADD42DDC3@cashew.wolfskeep.com> Message-ID: <200607220032.37259.franke.daniel@gmail.com> On Saturday 22 July 2006 00:15, T. Alexander Popiel wrote: > Yes, it's been seen. :-) It's one of the self-check failures. > Read through the conversation regarding the sequence of execution. Argl. The same procedure as every year: it's _almost_ right. Thanks for the pointer. Daniel From crary at cs.cmu.edu Fri Jul 21 18:36:44 2006 From: crary at cs.cmu.edu (Karl Crary) Date: Fri Jul 21 18:36:46 2006 Subject: [icfp-discuss] Division In-Reply-To: <25512-98643@sneakemail.com> References: <25512-98643@sneakemail.com> Message-ID: <44C156FC.3040903@cs.cmu.edu> Yes. Karl Crary vbzoli wrote: > Division should truncate its result? > > e.g. > > 3 / 2 => 1? > > > VOROSBARANYI Zoltan > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > > From dokondr at gmail.com Fri Jul 21 18:42:58 2006 From: dokondr at gmail.com (Dmitri O.Kondratiev) Date: Fri Jul 21 18:58:10 2006 Subject: [icfp-discuss] Registration: How to get a decryption key? Message-ID: <53396d9e0607211542t6e3f1fb6m238cd5a513c4c784@mail.gmail.com> According to registration instructions: * Register to receive a decryption key. * Download the codex and the UM specification. * Implement the UM and run the codex. * Supply the decryption key you received from us. * Obtain publications and submit them. I should get a "decryption key" after registration. All I got was a e-mail without any key. How to get the key ? Thanks, Dima From anton at zorgameuh.org Fri Jul 21 19:06:19 2006 From: anton at zorgameuh.org (Antonin ENFRUN) Date: Fri Jul 21 19:12:29 2006 Subject: [icfp-discuss] How many UM cycles to decode the codex ? Message-ID: <5E04F8A2-2470-4C2E-B517-A5D59B5CA77C@zorgameuh.org> Could someone tell me approx. how many UM cycles are needed to decode the codex ? So I'll know if mine is dead slow or dead wrong (400 Mega cycles, and decryption is not done). Thanks. From gs at guxx.net Fri Jul 21 19:18:04 2006 From: gs at guxx.net (G.Schoepp) Date: Fri Jul 21 19:19:07 2006 Subject: [icfp-discuss] "LOADPROG off (high)" In-Reply-To: <200607220032.37259.franke.daniel@gmail.com> References: <200607220000.49273.franke.daniel@gmail.com> <20060721221531.8ADD42DDC3@cashew.wolfskeep.com> <200607220032.37259.franke.daniel@gmail.com> Message-ID: <44C160AC.6040106@guxx.net> Daniel Franke schrieb: > On Saturday 22 July 2006 00:15, T. Alexander Popiel wrote: >> Yes, it's been seen. :-) It's one of the self-check failures. >> Read through the conversation regarding the sequence of execution. > > Argl. The same procedure as every year: it's _almost_ right. > Thanks for the pointer. I got another self-check failure: "reg 6 not zero" Guido From andy_zyx at yahoo.com Fri Jul 21 19:22:39 2006 From: andy_zyx at yahoo.com (Andy) Date: Fri Jul 21 19:22:42 2006 Subject: [icfp-discuss] How to use the decryption key on the output ? Message-ID: <20060721232239.98889.qmail@web50715.mail.yahoo.com> Hi, Started looking at this problem and have made some progress(I guess) in understanding it. :) I was wondering what is meant by supplying the decryption key to obtain the publication ? As in how to go about using it on the output generated by the UM on the codex ? Does anybody know ? Thanks Andy __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com From triska at gmx.at Fri Jul 21 19:41:35 2006 From: triska at gmx.at (Markus Triska) Date: Fri Jul 21 19:36:39 2006 Subject: [icfp-discuss] How to use the decryption key on the output ? In-Reply-To: <20060721232239.98889.qmail@web50715.mail.yahoo.com> References: <20060721232239.98889.qmail@web50715.mail.yahoo.com> Message-ID: <17601.26159.541136.980617@localhost.localdomain> Andy writes: > how to go about using it on the output generated by the UM on the codex ? The program asks for it when run, and you must enter it to proceed. Markus Triska. From christophe.poucet at gmail.com Fri Jul 21 19:40:36 2006 From: christophe.poucet at gmail.com (Christophe Poucet) Date: Fri Jul 21 19:40:39 2006 Subject: [icfp-discuss] Regarding scoring Message-ID: <44C165F4.6010801@gmail.com> Can certain parts give more or less points depending on solution? Cheers -- Christophe Poucet Ph.D. Student DESICS - DDT Phone:+32 16 28 87 20 E-mail: Christophe (dot) Poucet (at) imec (dot) be IMEC vzw ? Register of Legal Entities Leuven VAT BE 0425.260.668 ? Kapeldreef 75, B-3001 Leuven, Belgium ? http://www.imec.be -------------------------------------------------------------------------------- From aguilar.james at gmail.com Fri Jul 21 19:45:18 2006 From: aguilar.james at gmail.com (James Aguilar) Date: Fri Jul 21 19:45:23 2006 Subject: [icfp-discuss] How to use the decryption key on the output ? In-Reply-To: <17601.26159.541136.980617@localhost.localdomain> References: <20060721232239.98889.qmail@web50715.mail.yahoo.com> <17601.26159.541136.980617@localhost.localdomain> Message-ID: On 7/21/06, Markus Triska wrote: > > Andy writes: > > > how to go about using it on the output generated by the UM on the codex > ? > > The program asks for it when run, and you must enter it to proceed. OK, I got this, but how are we to signal EOF without first adding a \n? -- [?] James Aguilar [@] 18100 NE 95th St. #RR3088 / Redmond, WA 98052 [#] 314 494 0450 [!] In the wind and the rain, my darling, say goodbye -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060721/b1c64337/attachment.html From tom7 at cs.cmu.edu Fri Jul 21 19:52:04 2006 From: tom7 at cs.cmu.edu (Tom Murphy VII) Date: Fri Jul 21 19:52:03 2006 Subject: [icfp-discuss] Regarding scoring In-Reply-To: <44C165F4.6010801@gmail.com> References: <44C165F4.6010801@gmail.com> Message-ID: <44C168A4.3040408@cs.cmu.edu> Yes. >Can certain parts give more or less points depending on solution? > >Cheers > > From triska at gmx.at Fri Jul 21 20:01:34 2006 From: triska at gmx.at (Markus Triska) Date: Fri Jul 21 19:56:38 2006 Subject: [icfp-discuss] How to use the decryption key on the output ? In-Reply-To: References: <20060721232239.98889.qmail@web50715.mail.yahoo.com> <17601.26159.541136.980617@localhost.localdomain> Message-ID: <17601.27358.309467.817912@localhost.localdomain> James Aguilar writes: > > > OK, I got this, but how are we to signal EOF without first adding a \n? > Ctrl + d Markus Triska. From andy_zyx at yahoo.com Fri Jul 21 19:56:51 2006 From: andy_zyx at yahoo.com (Andy) Date: Fri Jul 21 19:56:54 2006 Subject: [icfp-discuss] How to use the decryption key on the output ? In-Reply-To: <17601.26159.541136.980617@localhost.localdomain> Message-ID: <20060721235651.93589.qmail@web50713.mail.yahoo.com> Thanks Markus. Andy. --- Markus Triska wrote: > Andy writes: > > > how to go about using it on the output generated by the UM on the codex ? > > The program asks for it when run, and you must enter it to proceed. > > Markus Triska. > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com From popiel at wolfskeep.com Fri Jul 21 20:09:24 2006 From: popiel at wolfskeep.com (T. Alexander Popiel) Date: Fri Jul 21 20:09:27 2006 Subject: [icfp-discuss] How many UM cycles to decode the codex ? In-Reply-To: <5E04F8A2-2470-4C2E-B517-A5D59B5CA77C@zorgameuh.org> References: <5E04F8A2-2470-4C2E-B517-A5D59B5CA77C@zorgameuh.org> Message-ID: <20060722000924.6DDB92DDC3@cashew.wolfskeep.com> In message: <5E04F8A2-2470-4C2E-B517-A5D59B5CA77C@zorgameuh.org> Antonin ENFRUN writes: >Could someone tell me approx. how many UM cycles are needed to decode >the codex ? >So I'll know if mine is dead slow or dead wrong (400 Mega cycles, and >decryption is not done). My implementation took about 1.8 gigacycles to get the useful output from the original codex. Stuff after that... well, let's just say I'm trying to speed up my implementation. - Alex From dinko.tenev at gmail.com Fri Jul 21 20:32:04 2006 From: dinko.tenev at gmail.com (Dinko Tenev) Date: Fri Jul 21 20:32:07 2006 Subject: [icfp-discuss] How to use the decryption key on the output ? In-Reply-To: <17601.26159.541136.980617@localhost.localdomain> References: <20060721232239.98889.qmail@web50715.mail.yahoo.com> <17601.26159.541136.980617@localhost.localdomain> Message-ID: My key is rejected, no matter what I try. The last attempt was to put the key (and just the key, no terminating newline or anything) into a file and feed the file to the program, but that failed too. Could it be that the key I got was somehow mangled? I see unusual characters in it. Cheers, Dinko On 7/22/06, Markus Triska wrote: > Andy writes: > > > how to go about using it on the output generated by the UM on the codex ? > > The program asks for it when run, and you must enter it to proceed. > > Markus Triska. From crary at cs.cmu.edu Fri Jul 21 20:37:28 2006 From: crary at cs.cmu.edu (Karl Crary) Date: Fri Jul 21 20:37:30 2006 Subject: [icfp-discuss] How to use the decryption key on the output ? In-Reply-To: References: <20060721232239.98889.qmail@web50715.mail.yahoo.com> <17601.26159.541136.980617@localhost.localdomain> Message-ID: <44C17348.3070302@cs.cmu.edu> It sounds like you have a bug in your UM implementation. Karl Crary Dinko Tenev wrote: > My key is rejected, no matter what I try. The last attempt was to put > the key (and just the key, no terminating newline or anything) into a > file and feed the file to the program, but that failed too. Could it > be that the key I got was somehow mangled? I see unusual characters > in it. > > Cheers, > > Dinko > > > On 7/22/06, Markus Triska wrote: >> Andy writes: >> >> > how to go about using it on the output generated by the UM on the >> codex ? >> >> The program asks for it when run, and you must enter it to proceed. >> >> Markus Triska. > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > From jmk at amoebaville.net Fri Jul 21 20:40:22 2006 From: jmk at amoebaville.net (Justin Koser) Date: Fri Jul 21 20:40:28 2006 Subject: [icfp-discuss] How to use the decryption key on the output ? In-Reply-To: <44C17348.3070302@cs.cmu.edu> References: <20060721232239.98889.qmail@web50715.mail.yahoo.com> <17601.26159.541136.980617@localhost.localdomain> <44C17348.3070302@cs.cmu.edu> Message-ID: <44C173F6.6020707@amoebaville.net> Karl Crary wrote: > Dinko Tenev wrote: >> My key is rejected, no matter what I try. The last attempt was to put >> the key (and just the key, no terminating newline or anything) into a >> file and feed the file to the program, but that failed too. Could it >> be that the key I got was somehow mangled? I see unusual characters >> in it. > > It sounds like you have a bug in your UM implementation. Agreed. And it sounds like I have the same bug. /sigh -- Justin From druffer at worldnet.att.net Fri Jul 21 20:49:23 2006 From: druffer at worldnet.att.net (Dennis Ruffer) Date: Fri Jul 21 20:49:27 2006 Subject: [icfp-discuss] How to use the decryption key on the output ? In-Reply-To: <44C173F6.6020707@amoebaville.net> References: <20060721232239.98889.qmail@web50715.mail.yahoo.com> <17601.26159.541136.980617@localhost.localdomain> <44C17348.3070302@cs.cmu.edu> <44C173F6.6020707@amoebaville.net> Message-ID: <740AFEEC-3F5E-4E9B-959A-557F54A9F759@worldnet.att.net> On Jul 21, 2006, at 8:40 PM, Justin Koser wrote: > Karl Crary wrote: > > Dinko Tenev wrote: > >> My key is rejected, no matter what I try. The last attempt was > to put > >> the key (and just the key, no terminating newline or anything) > into a > >> file and feed the file to the program, but that failed too. > Could it > >> be that the key I got was somehow mangled? I see unusual > characters > >> in it. > > >> It sounds like you have a bug in your UM implementation. > > Agreed. And it sounds like I have the same bug. /sigh Has anyone gotten past the "wrong key" error yet? DaR From rvandam00 at gmail.com Fri Jul 21 20:53:59 2006 From: rvandam00 at gmail.com (Robert Van Dam) Date: Fri Jul 21 20:54:02 2006 Subject: [icfp-discuss] How to use the decryption key on the output ? In-Reply-To: <740AFEEC-3F5E-4E9B-959A-557F54A9F759@worldnet.att.net> References: <20060721232239.98889.qmail@web50715.mail.yahoo.com> <17601.26159.541136.980617@localhost.localdomain> <44C17348.3070302@cs.cmu.edu> <44C173F6.6020707@amoebaville.net> <740AFEEC-3F5E-4E9B-959A-557F54A9F759@worldnet.att.net> Message-ID: <1815f00607211753o3c53293g4be321e635ae13d5@mail.gmail.com> Considering the scores (don't worry, I'm not up there), I would say, definitely. Rob On 7/21/06, Dennis Ruffer wrote: > On Jul 21, 2006, at 8:40 PM, Justin Koser wrote: > > Karl Crary wrote: > > > Dinko Tenev wrote: > > >> My key is rejected, no matter what I try. The last attempt was > > to put > > >> the key (and just the key, no terminating newline or anything) > > into a > > >> file and feed the file to the program, but that failed too. > > Could it > > >> be that the key I got was somehow mangled? I see unusual > > characters > > >> in it. > > > > >> It sounds like you have a bug in your UM implementation. > > > > Agreed. And it sounds like I have the same bug. /sigh > > Has anyone gotten past the "wrong key" error yet? > > DaR > > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > From shawnyar217 at yahoo.com Fri Jul 21 20:55:30 2006 From: shawnyar217 at yahoo.com (Shawn Yarbrough) Date: Fri Jul 21 20:55:33 2006 Subject: [icfp-discuss] How to use the decryption key on the output ? In-Reply-To: <44C173F6.6020707@amoebaville.net> Message-ID: <20060722005530.94096.qmail@web53406.mail.yahoo.com> Don't feel too bad. I have yet to get the UM to output anything at all. If I hand decode the first several instructions, they look like an illegal program to me. Same if I reverse the byte significance. Obviously I'm missing something critical about the operators. Shawn Justin Koser wrote: Karl Crary wrote: > Dinko Tenev wrote: >> My key is rejected, no matter what I try. The last attempt was to put >> the key (and just the key, no terminating newline or anything) into a >> file and feed the file to the program, but that failed too. Could it >> be that the key I got was somehow mangled? I see unusual characters >> in it. > > It sounds like you have a bug in your UM implementation. Agreed. And it sounds like I have the same bug. /sigh -- Justin _______________________________________________ icfpcontest-discuss mailing list icfpcontest-discuss@lists.andrew.cmu.edu https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss --------------------------------- How low will we go? Check out Yahoo! Messenger?s low PC-to-Phone call rates. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060721/f176a9fc/attachment.html From gus-icfp at guxx.net Fri Jul 21 21:02:28 2006 From: gus-icfp at guxx.net (G.Schoepp) Date: Fri Jul 21 21:02:31 2006 Subject: [icfp-discuss] Jippie: UM program follows colon... Message-ID: <44C17924.8040504@guxx.net> Got the first step. Time for a nap! Guido From jmk at amoebaville.net Fri Jul 21 21:07:13 2006 From: jmk at amoebaville.net (Justin Koser) Date: Fri Jul 21 21:07:18 2006 Subject: [icfp-discuss] How to use the decryption key on the output ? In-Reply-To: <20060722005530.94096.qmail@web53406.mail.yahoo.com> References: <20060722005530.94096.qmail@web53406.mail.yahoo.com> Message-ID: <44C17A41.4070401@amoebaville.net> Shawn Yarbrough wrote: > Don't feel too bad. I have yet to get the UM to output anything at > all. If I hand decode the first several instructions, they look like an > illegal program to me. Same if I reverse the byte significance. > Obviously I'm missing something critical about the operators. They're not illegal -- keep in mind that for most of the instructions, some bits are ignored. I'm sure the organizers made use of that property to disguise the code before the spec came out. Incidentally, I'm not even getting an invalid key message... just segfaulting. Time to go through the array code again, eh? Cheers, Justin From gus-icfp at guxx.net Fri Jul 21 22:11:29 2006 From: gus-icfp at guxx.net (G.Schoepp) Date: Fri Jul 21 22:11:28 2006 Subject: [icfp-discuss] Login to UMIX? Message-ID: <44C18951.8000303@guxx.net> Hi, I got the UMIX running but can't login with guest/guest. Should this work (so it's a bug in my UM)? Or do I have to hack into the UMIX? Guido From crary at cs.cmu.edu Fri Jul 21 22:28:21 2006 From: crary at cs.cmu.edu (Karl Crary) Date: Fri Jul 21 22:28:24 2006 Subject: [icfp-discuss] Login to UMIX? In-Reply-To: <44C18951.8000303@guxx.net> References: <44C18951.8000303@guxx.net> Message-ID: <44C18D45.9090402@cs.cmu.edu> If you log in as guest, it should not prompt you for a password. It sounds like there may be something wrong with your implementation of input. Karl Crary G.Schoepp wrote: > Hi, > > I got the UMIX running but can't login with guest/guest. Should this > work (so it's a bug in my UM)? Or do I have to hack into the UMIX? > > Guido > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > From wei.hoo at gmail.com Fri Jul 21 23:10:12 2006 From: wei.hoo at gmail.com (Wei Hu) Date: Fri Jul 21 23:10:16 2006 Subject: [icfp-discuss] A question regarding Operation #8 Message-ID: <71fd12e60607212010h762e276cxbc3264f3f16eeb3b@mail.gmail.com> Hi All, I had a hard time understanding the description of operation #8, in particular the following sentence: A bit pattern not consisting of exclusively the 0 bit, and that identifies no other active allocated array, is placed in the B register. Can anyone kindly explain it to me? Thanks, Wei From wei.hoo at gmail.com Fri Jul 21 23:13:19 2006 From: wei.hoo at gmail.com (Wei Hu) Date: Fri Jul 21 23:13:24 2006 Subject: [icfp-discuss] Re: A question regarding Operation #8 In-Reply-To: <71fd12e60607212010h762e276cxbc3264f3f16eeb3b@mail.gmail.com> References: <71fd12e60607212010h762e276cxbc3264f3f16eeb3b@mail.gmail.com> Message-ID: <71fd12e60607212013m613c63d4qf94bc426c6d901c5@mail.gmail.com> Also I don't see which register identifies the newly allocated array? On 7/21/06, Wei Hu wrote: > Hi All, > > I had a hard time understanding the description of operation #8, in > particular the following sentence: > > A bit pattern not consisting of exclusively the 0 bit, and that > identifies no other active allocated array, is placed in the B > register. > > Can anyone kindly explain it to me? > > Thanks, > Wei > From wei.hoo at gmail.com Fri Jul 21 23:17:27 2006 From: wei.hoo at gmail.com (Wei Hu) Date: Fri Jul 21 23:17:30 2006 Subject: [icfp-discuss] Re: A question regarding Operation #8 In-Reply-To: <71fd12e60607212013m613c63d4qf94bc426c6d901c5@mail.gmail.com> References: <71fd12e60607212010h762e276cxbc3264f3f16eeb3b@mail.gmail.com> <71fd12e60607212013m613c63d4qf94bc426c6d901c5@mail.gmail.com> Message-ID: <71fd12e60607212017n6d2b45efs8376c59c035080b3@mail.gmail.com> Sorry, please disregard my email. I figured out the meaning now. Thanks, Wei On 7/21/06, Wei Hu wrote: > Also I don't see which register identifies the newly allocated array? > > On 7/21/06, Wei Hu wrote: > > Hi All, > > > > I had a hard time understanding the description of operation #8, in > > particular the following sentence: > > > > A bit pattern not consisting of exclusively the 0 bit, and that > > identifies no other active allocated array, is placed in the B > > register. > > > > Can anyone kindly explain it to me? > > > > Thanks, > > Wei > > > From jm at gaillourdet.net Fri Jul 21 22:14:10 2006 From: jm at gaillourdet.net (Jean-Marie Gaillourdet) Date: Fri Jul 21 23:20:17 2006 Subject: [icfp-discuss] Login to UMIX? In-Reply-To: <44C18951.8000303@guxx.net> References: <44C18951.8000303@guxx.net> Message-ID: <95E42D50-B9C4-4E05-95EE-07E945EDF53B@gaillourdet.net> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi, On 22.07.2006, at 04:11, G.Schoepp wrote: > Hi, > > I got the UMIX running but can't login with guest/guest. Should > this work (so it's a bug in my UM)? Or do I have to hack into the > UMIX? No, I was able to login. See Alain Frisch's question. Jean-Marie -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.1 (Darwin) iD8DBQFEwYn2NIUNP/I5YOgRAgIRAJ9BslTvEqnn3DSOXYeapkHQlCSAkQCfZUHk XFoBna09vVT/EdhRD3bsaGw= =JtNe -----END PGP SIGNATURE----- From andy_zyx at yahoo.com Fri Jul 21 23:40:15 2006 From: andy_zyx at yahoo.com (Andy) Date: Fri Jul 21 23:40:17 2006 Subject: [icfp-discuss] Couple of doubts ! In-Reply-To: <71fd12e60607212017n6d2b45efs8376c59c035080b3@mail.gmail.com> Message-ID: <20060722034015.30719.qmail@web50701.mail.yahoo.com> Hi All, This maybe way off but am getting confused with couple of things: [1] What are "program" scrolls. Is it the codex being read ? [2] Other than array '0', am not seeing how the other arrays [ in the collection of arrays ] is used [ unless and until they are being overwritten on the array '0' and the execution finger traverses across it ]. Am I right ? Thanks, Andy --- Wei Hu wrote: > Sorry, please disregard my email. > I figured out the meaning now. > > Thanks, > Wei > > On 7/21/06, Wei Hu wrote: > > Also I don't see which register identifies the newly allocated array? > > > > On 7/21/06, Wei Hu wrote: > > > Hi All, > > > > > > I had a hard time understanding the description of operation #8, in > > > particular the following sentence: > > > > > > A bit pattern not consisting of exclusively the 0 bit, and that > > > identifies no other active allocated array, is placed in the B > > > register. > > > > > > Can anyone kindly explain it to me? > > > > > > Thanks, > > > Wei > > > > > > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com From tom7 at cs.cmu.edu Sat Jul 22 00:05:59 2006 From: tom7 at cs.cmu.edu (Tom Murphy VII) Date: Sat Jul 22 00:06:00 2006 Subject: [icfp-discuss] Couple of doubts ! In-Reply-To: <20060722034015.30719.qmail@web50701.mail.yahoo.com> References: <20060722034015.30719.qmail@web50701.mail.yahoo.com> Message-ID: <44C1A427.3010205@cs.cmu.edu> 1. Yes, the codex is a program scroll. 2. The array amendment and array index operations also write to and read from allocated arrays. Tom >Hi All, > >This maybe way off but am getting confused with couple of things: >[1] What are "program" scrolls. Is it the codex being read ? >[2] Other than array '0', am not seeing how the other arrays [ in the collection of arrays ] is >used [ unless and until they are being overwritten on the array '0' and the execution finger >traverses across it ]. Am I right ? > >Thanks, >Andy > > > From trivik at gmail.com Sat Jul 22 03:06:19 2006 From: trivik at gmail.com (saccade) Date: Sat Jul 22 03:06:21 2006 Subject: [icfp-discuss] register vs. register number Message-ID: In the descriptions of operators #1, #2 and #12 should the occurrences of "B", "C" and "C" be replaced by "register B", "register C" and "register C", or what was intended is what is written? Trivik From trivik at gmail.com Sat Jul 22 03:07:03 2006 From: trivik at gmail.com (saccade) Date: Sat Jul 22 03:07:06 2006 Subject: [icfp-discuss] operation #12 (load program) Message-ID: Does operation #12 lead to any net new active array? Trivik From blo.b at infonie.fr Sat Jul 22 03:07:38 2006 From: blo.b at infonie.fr (Laurent Vaucher) Date: Sat Jul 22 03:07:41 2006 Subject: [icfp-discuss] Virtual machine specification In-Reply-To: <20060721214528.GR15374@bachue.com> References: <20060721210739.GQ15374@bachue.com> <20060721211204.3BC122DDB9@cashew.wolfskeep.com> <20060721214528.GR15374@bachue.com> Message-ID: <03EC92AF-8EBF-49FC-AE96-179E4FBC2749@infonie.fr> My UM processed the initial codex in about 5 minutes. Considering that the execution of the second program still hasn' t produced anything after 10 hours, is it possible that it's wrong ? From flyfree at gmail.com Sat Jul 22 03:08:55 2006 From: flyfree at gmail.com (Zhenzhong Xu) Date: Sat Jul 22 03:08:58 2006 Subject: [icfp-discuss] register vs. register number In-Reply-To: References: Message-ID: <7b2bf4900607220008i77344ca5v6469c3a47971a445@mail.gmail.com> yes. On 7/22/06, saccade wrote: > In the descriptions of operators #1, #2 and #12 should the occurrences > of "B", "C" and "C" be replaced by "register B", "register C" and > "register C", or what was intended is what is written? > > Trivik > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > From aguilar.james at gmail.com Sat Jul 22 03:10:02 2006 From: aguilar.james at gmail.com (James Aguilar) Date: Sat Jul 22 03:10:05 2006 Subject: [icfp-discuss] register vs. register number In-Reply-To: References: Message-ID: On 7/22/06, saccade wrote: > > In the descriptions of operators #1, #2 and #12 should the occurrences > of "B", "C" and "C" be replaced by "register B", "register C" and > "register C", or what was intended is what is written? I had this question too. Currently, I've implemented it as the former, but I'm not sure because I haven't had a successful run yet. -- James -- [?] James Aguilar [@] 18100 NE 95th St. #RR3088 / Redmond, WA 98052 [#] 314 494 0450 [!] In the wind and the rain, my darling, say goodbye -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060722/c2b3d128/attachment-0001.html From aguilar.james at gmail.com Sat Jul 22 03:11:14 2006 From: aguilar.james at gmail.com (James Aguilar) Date: Sat Jul 22 03:11:17 2006 Subject: [icfp-discuss] register vs. register number In-Reply-To: <7b2bf4900607220008i77344ca5v6469c3a47971a445@mail.gmail.com> References: <7b2bf4900607220008i77344ca5v6469c3a47971a445@mail.gmail.com> Message-ID: On 7/22/06, Zhenzhong Xu wrote: > > yes. There were two questions there, so which are you saying yes to? -- [?] James Aguilar [@] 18100 NE 95th St. #RR3088 / Redmond, WA 98052 [#] 314 494 0450 [!] In the wind and the rain, my darling, say goodbye -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060722/c1de1e76/attachment.html From drl at cs.cmu.edu Sat Jul 22 03:13:23 2006 From: drl at cs.cmu.edu (Dan Licata) Date: Sat Jul 22 03:13:56 2006 Subject: [icfp-discuss] register vs. register number In-Reply-To: References: Message-ID: <20060722071323.GE26438@cs.cmu.edu> Yes. A capital letter, such as B, denotes the index of a register (3 bits), and the description always refers to the contents of the register at that index. -Dan On Jul22, James Aguilar wrote: > On 7/22/06, saccade wrote: > > > >In the descriptions of operators #1, #2 and #12 should the occurrences > >of "B", "C" and "C" be replaced by "register B", "register C" and > >"register C", or what was intended is what is written? > > > I had this question too. Currently, I've implemented it as the former, but > I'm not sure because I haven't had a successful run yet. > > -- James > > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss From flyfree at gmail.com Sat Jul 22 03:14:31 2006 From: flyfree at gmail.com (Zhenzhong Xu) Date: Sat Jul 22 03:14:34 2006 Subject: [icfp-discuss] register vs. register number In-Reply-To: References: <7b2bf4900607220008i77344ca5v6469c3a47971a445@mail.gmail.com> Message-ID: <7b2bf4900607220014s589439faw4f6e545daa574861@mail.gmail.com> --In the descriptions of operators #1, #2 and #12 should the occurrences --of "B", "C" and "C" be replaced by "register B", "register C" and --"register C", or what was intended is what is written? yes. some of the terms are ambigous, but i've referred all A,B,C as the values in register[A], register[B],register[C] On 7/22/06, James Aguilar wrote: > On 7/22/06, Zhenzhong Xu wrote: > > yes. > > There were two questions there, so which are you saying yes to? > > -- > [?] James Aguilar > [@] 18100 NE 95th St. #RR3088 / Redmond, WA 98052 > [#] 314 494 0450 > [!] In the wind and the rain, my darling, say goodbye > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > > > From drl at cs.cmu.edu Sat Jul 22 03:22:32 2006 From: drl at cs.cmu.edu (Dan Licata) Date: Sat Jul 22 03:22:59 2006 Subject: [icfp-discuss] Virtual machine specification In-Reply-To: <03EC92AF-8EBF-49FC-AE96-179E4FBC2749@infonie.fr> References: <20060721210739.GQ15374@bachue.com> <20060721211204.3BC122DDB9@cashew.wolfskeep.com> <20060721214528.GR15374@bachue.com> <03EC92AF-8EBF-49FC-AE96-179E4FBC2749@infonie.fr> Message-ID: <20060722072232.GG26438@cs.cmu.edu> The 5 minutes is reasonable. On our UM, which is of comparable speed, the second binary produces output within a second. -Dan On Jul22, Laurent Vaucher wrote: > My UM processed the initial codex in about 5 minutes. Considering > that the execution of the second program still hasn' t produced > anything after 10 hours, is it possible that it's wrong ? > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > From bluelive at xs4all.nl Sat Jul 22 03:30:43 2006 From: bluelive at xs4all.nl (Bart van der Werf) Date: Sat Jul 22 03:30:48 2006 Subject: [icfp-discuss] How to use the decryption key on the output ? In-Reply-To: <20060721232239.98889.qmail@web50715.mail.yahoo.com> Message-ID: <001401c6ad60$be243f30$4101a8c0@andromeda> No clue, im running into a problem where the programs wants more inputdata after providing the key, I have no clue what I should provide. Grt,Bart -----Original Message----- From: icfpcontest-discuss-bounces@lists.andrew.cmu.edu [mailto:icfpcontest-discuss-bounces@lists.andrew.cmu.edu] On Behalf Of Andy Sent: Saturday, July 22, 2006 1:23 AM To: icfpcontest-discuss@lists.andrew.cmu.edu Subject: [icfp-discuss] How to use the decryption key on the output ? Hi, Started looking at this problem and have made some progress(I guess) in understanding it. :) I was wondering what is meant by supplying the decryption key to obtain the publication ? As in how to go about using it on the output generated by the UM on the codex ? Does anybody know ? Thanks Andy __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com _______________________________________________ icfpcontest-discuss mailing list icfpcontest-discuss@lists.andrew.cmu.edu https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss From flyfree at gmail.com Sat Jul 22 03:33:28 2006 From: flyfree at gmail.com (Zhenzhong Xu) Date: Sat Jul 22 03:33:31 2006 Subject: [icfp-discuss] How to use the decryption key on the output ? In-Reply-To: <001401c6ad60$be243f30$4101a8c0@andromeda> References: <20060721232239.98889.qmail@web50715.mail.yahoo.com> <001401c6ad60$be243f30$4101a8c0@andromeda> Message-ID: <7b2bf4900607220033y6ceab1bcuce70fbab0226341d@mail.gmail.com> > No clue, im running into a problem where the programs wants more > inputdata after providing the key, I have no clue what I should provide. > check your #12 loadprogram Z. > Grt,Bart > > > -----Original Message----- > From: icfpcontest-discuss-bounces@lists.andrew.cmu.edu > [mailto:icfpcontest-discuss-bounces@lists.andrew.cmu.edu] On Behalf Of > Andy > Sent: Saturday, July 22, 2006 1:23 AM > To: icfpcontest-discuss@lists.andrew.cmu.edu > Subject: [icfp-discuss] How to use the decryption key on the output ? > > > Hi, > > Started looking at this problem and have made some progress(I guess) in > understanding it. :) > I was wondering what is meant by supplying the decryption key to obtain > the publication ? As in how to go about using it on the output generated > by the UM on the codex ? > > Does anybody know ? > > Thanks > Andy > > > > __________________________________________________ > Do You Yahoo!? > Tired of spam? Yahoo! Mail has the best spam protection around > http://mail.yahoo.com > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > From drl at cs.cmu.edu Sat Jul 22 03:35:33 2006 From: drl at cs.cmu.edu (Dan Licata) Date: Sat Jul 22 03:35:46 2006 Subject: [icfp-discuss] How to use the decryption key on the output ? In-Reply-To: <001401c6ad60$be243f30$4101a8c0@andromeda> References: <20060721232239.98889.qmail@web50715.mail.yahoo.com> <001401c6ad60$be243f30$4101a8c0@andromeda> Message-ID: <20060722073533.GI26438@cs.cmu.edu> On Jul22, Bart van der Werf wrote: > No clue, im running into a problem where the programs wants more > inputdata after providing the key, I have no clue what I should provide. > > Grt,Bart Depending on how you implemented your UM, you may need to give the UM a newline before it reads the input. If your UM is correct, then, after entering the key, your UM should print some lines and then print a menu prompting you for the next input. -Dan > -----Original Message----- > From: icfpcontest-discuss-bounces@lists.andrew.cmu.edu > [mailto:icfpcontest-discuss-bounces@lists.andrew.cmu.edu] On Behalf Of > Andy > Sent: Saturday, July 22, 2006 1:23 AM > To: icfpcontest-discuss@lists.andrew.cmu.edu > Subject: [icfp-discuss] How to use the decryption key on the output ? > > > Hi, > > Started looking at this problem and have made some progress(I guess) in > understanding it. :) > I was wondering what is meant by supplying the decryption key to obtain > the publication ? As in how to go about using it on the output generated > by the UM on the codex ? > > Does anybody know ? > > Thanks > Andy > > > > __________________________________________________ > Do You Yahoo!? > Tired of spam? Yahoo! Mail has the best spam protection around > http://mail.yahoo.com > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > From blo.b at infonie.fr Sat Jul 22 03:42:13 2006 From: blo.b at infonie.fr (Laurent Vaucher) Date: Sat Jul 22 03:42:21 2006 Subject: [icfp-discuss] How to use the decryption key on the output ? In-Reply-To: <20060722073533.GI26438@cs.cmu.edu> References: <20060721232239.98889.qmail@web50715.mail.yahoo.com> <001401c6ad60$be243f30$4101a8c0@andromeda> <20060722073533.GI26438@cs.cmu.edu> Message-ID: <323D6D24-F1C4-4925-BA07-8CC58F81491C@infonie.fr> > Depending on how you implemented your UM, you may need to give the > UM a > newline before it reads the input. Can the newline be a problem? Because I still can't see why my UM runs the codex perfectly but doesn't make any progress on the next program. From drochom at googlemail.com Sat Jul 22 04:01:13 2006 From: drochom at googlemail.com (przemek drochomirecki) Date: Sat Jul 22 04:03:27 2006 Subject: [icfp-discuss] #8. Allocation. Message-ID: <8daef39f0607220101vef45f4eye62d70c2f2ce522a@mail.gmail.com> Hi, Is it possible that '0' array is allocating? Cheers Przemek -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060722/0514cff7/attachment-0001.html From franke.daniel at gmail.com Sat Jul 22 04:12:17 2006 From: franke.daniel at gmail.com (Daniel Franke) Date: Sat Jul 22 04:12:31 2006 Subject: [icfp-discuss] #8. Allocation. In-Reply-To: <8daef39f0607220101vef45f4eye62d70c2f2ce522a@mail.gmail.com> References: <8daef39f0607220101vef45f4eye62d70c2f2ce522a@mail.gmail.com> Message-ID: <200607221012.17225.franke.daniel@gmail.com> > Is it possible that '0' array is allocating? That's me as well ... or trying to access general purpose register 119. If I don't mind accessing the later, my machine passes the self test *g* Regards Daniel From blo.b at infonie.fr Sat Jul 22 04:16:42 2006 From: blo.b at infonie.fr (Laurent Vaucher) Date: Sat Jul 22 04:16:47 2006 Subject: [icfp-discuss] Two small hits for Java programmers out there In-Reply-To: <323D6D24-F1C4-4925-BA07-8CC58F81491C@infonie.fr> References: <20060721232239.98889.qmail@web50715.mail.yahoo.com> <001401c6ad60$be243f30$4101a8c0@andromeda> <20060722073533.GI26438@cs.cmu.edu> <323D6D24-F1C4-4925-BA07-8CC58F81491C@infonie.fr> Message-ID: <7A0926F0-CF04-4947-9DC0-7B09B20AE27E@infonie.fr> Having solved the intro problem, I feel in a generous mood. The two things that prevented me from getting there sooner are the following : - unsigned 32 bits integer arithmetic is hard in Java. I mean, Java does not have *unsigned* integer types. You'll have to juggle a bit (with longs and bitwise operations) to get it right. - don't use PrintStream or PrintWriter.print() to produce output: it has the nasty property of trying to convert character values to a given encoding, which is why my second program ran in circles for ten hours... And congratulations to the CMU POP team for this ongoing contest. Laurent. From samuel at acid-code.ch Sat Jul 22 04:19:27 2006 From: samuel at acid-code.ch (Samuel Burri) Date: Sat Jul 22 04:18:26 2006 Subject: [icfp-discuss] #8. Allocation. In-Reply-To: <200607221012.17225.franke.daniel@gmail.com> References: <8daef39f0607220101vef45f4eye62d70c2f2ce522a@mail.gmail.com> <200607221012.17225.franke.daniel@gmail.com> Message-ID: <61986.84.226.35.108.1153556367.squirrel@serioussam.selfip.org> How can #119 be accessed with a three bit address? Just wondering... By the way: How do you mean "'0' array is allocating"? Sam > >> Is it possible that '0' array is allocating? > > That's me as well ... or trying to access general purpose register 119. > If I don't mind accessing the later, my machine passes the self test *g* > > Regards > Daniel > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > > From bluelive at xs4all.nl Sat Jul 22 04:23:12 2006 From: bluelive at xs4all.nl (Bart van der Werf) Date: Sat Jul 22 04:23:16 2006 Subject: [icfp-discuss] #8. Allocation. In-Reply-To: <200607221012.17225.franke.daniel@gmail.com> Message-ID: <000501c6ad68$12ea3220$4101a8c0@andromeda> > > > Is it possible that '0' array is allocating? > > That's me as well ... or trying to access general purpose register 119. > If I don't mind accessing the later, my machine passes the self test > *g* > How does the value 119 fit into a 3 bit uint ? Grt, Bart From franke.daniel at gmail.com Sat Jul 22 04:28:50 2006 From: franke.daniel at gmail.com (Daniel Franke) Date: Sat Jul 22 04:29:00 2006 Subject: [icfp-discuss] #8. Allocation. In-Reply-To: <61986.84.226.35.108.1153556367.squirrel@serioussam.selfip.org> References: <8daef39f0607220101vef45f4eye62d70c2f2ce522a@mail.gmail.com> <200607221012.17225.franke.daniel@gmail.com> <61986.84.226.35.108.1153556367.squirrel@serioussam.selfip.org> Message-ID: <200607221028.50879.franke.daniel@gmail.com> On Saturday 22 July 2006 10:19, Samuel Burri wrote: > > > Is it possible that '0' array is allocating? > > > > That's me as well ... or trying to access general purpose register 119. > > If I don't mind accessing the later, my machine passes the self test *g* > > How can #119 be accessed with a three bit address? > Just wondering... > By the way: How do you mean "'0' array is allocating"? That's the value within the register. That's my question from *err* many hours ago. The distinction of registers and registers. Operation #1, Operands: A=1, B=0, C=6 Register values: 179 119 4 53 0 0 0 0 "The register A receives the value stored at offset in register C in the array identified by B." That is: "the array identified by B" -> 179 "offset in register C" -> 0 "The register A" -> 119 There you go. Write to register 119. I didn't confuse them AGAIN? From samuel at acid-code.ch Sat Jul 22 04:38:00 2006 From: samuel at acid-code.ch (Samuel Burri) Date: Sat Jul 22 04:36:58 2006 Subject: [icfp-discuss] #8. Allocation. In-Reply-To: <200607221028.50879.franke.daniel@gmail.com> References: <8daef39f0607220101vef45f4eye62d70c2f2ce522a@mail.gmail.com> <200607221012.17225.franke.daniel@gmail.com> <61986.84.226.35.108.1153556367.squirrel@serioussam.selfip.org> <200607221028.50879.franke.daniel@gmail.com> Message-ID: <62727.84.226.35.108.1153557480.squirrel@serioussam.selfip.org> > On Saturday 22 July 2006 10:19, Samuel Burri wrote: >> > > Is it possible that '0' array is allocating? >> > >> > That's me as well ... or trying to access general purpose register >> 119. >> > If I don't mind accessing the later, my machine passes the self test >> *g* >> >> How can #119 be accessed with a three bit address? >> Just wondering... >> By the way: How do you mean "'0' array is allocating"? > > That's the value within the register. That's my question from *err* many > hours > ago. The distinction of registers and registers. > > Operation #1, Operands: A=1, B=0, C=6 > Register values: 179 119 4 53 0 0 0 0 > > "The register A receives the value stored at offset > in register C in the array identified by B." > > That is: > "the array identified by B" -> 179 > "offset in register C" -> 0 > "The register A" -> 119 > > There you go. Write to register 119. > > > > I didn't confuse them AGAIN? I would say, that register a is the register #1 which currently contains the 119 From drochom at googlemail.com Sat Jul 22 04:39:55 2006 From: drochom at googlemail.com (przemek drochomirecki) Date: Sat Jul 22 04:39:59 2006 Subject: [icfp-discuss] #8. Allocation. In-Reply-To: <61986.84.226.35.108.1153556367.squirrel@serioussam.selfip.org> References: <8daef39f0607220101vef45f4eye62d70c2f2ce522a@mail.gmail.com> <200607221012.17225.franke.daniel@gmail.com> <61986.84.226.35.108.1153556367.squirrel@serioussam.selfip.org> Message-ID: <8daef39f0607220139u69f25d59k73ad8b53e7e8a05a@mail.gmail.com> #8. Allocation. A new array is created with a capacity of platters commensurate to the value in the register C. This new array is initialized entirely with platters holding the value 0. A bit pattern not consisting of exclusively the 0 bit, and that identifies no other active allocated array, is placed in the B register. I mean: Identifier of newly created array is 0. (The value of register B is equal 0) P. On 7/22/06, Samuel Burri wrote: > > How can #119 be accessed with a three bit address? > Just wondering... > By the way: How do you mean "'0' array is allocating"? > > Sam > > > > >> Is it possible that '0' array is allocating? > > > > That's me as well ... or trying to access general purpose register 119. > > If I don't mind accessing the later, my machine passes the self test *g* > > > > Regards > > Daniel > > _______________________________________________ > > icfpcontest-discuss mailing list > > icfpcontest-discuss@lists.andrew.cmu.edu > > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > > > > > > > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060722/2caf2618/attachment.html From wei.hoo at gmail.com Sat Jul 22 04:42:55 2006 From: wei.hoo at gmail.com (Wei Hu) Date: Sat Jul 22 04:42:58 2006 Subject: [icfp-discuss] #8. Allocation. In-Reply-To: <8daef39f0607220101vef45f4eye62d70c2f2ce522a@mail.gmail.com> References: <8daef39f0607220101vef45f4eye62d70c2f2ce522a@mail.gmail.com> Message-ID: <71fd12e60607220142q166e4553rf87ae2fc79d149f9@mail.gmail.com> I feel frustrated as I eyeballed the mysterious code but failed to figure out the reason. I ran into the problem of allocating '0' array too. At execution finger 0x20c, the instruction 0x89119e4a is fetched, so it tries to allocate array_reg1, while reg1 was zeroed out by an earlier instruction, which means it's going to allocate array0 #%@#$.. There must be some problem in my code, because allocating '0' array doesn't make sense to me... Otherwise we'll be executing operation #0 until we run out of the array boundary.. Plus in the problem description it's stated that the identifier doesn't refer to an active array. On 7/22/06, przemek drochomirecki wrote: > Hi, > > Is it possible that '0' array is allocating? > > Cheers > Przemek > > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > > > From trivik at gmail.com Sat Jul 22 05:06:15 2006 From: trivik at gmail.com (saccade) Date: Sat Jul 22 05:13:29 2006 Subject: [icfp-discuss] Re: operation #12 (load program) In-Reply-To: References: Message-ID: > Does operation #12 lead to any net new active array? clarification of teh question: the um-spec says: "#12. Load Program. The array identified by the B register is duplicated and the duplicate shall replace the '0' array, regardless of size." by "replace the '0' array" does one mean 1. start treating the duplicate as the program? (so there will be no way to reference any platter within the the program array but the '0' array would be unchanged) or 2. the duplicate's 32-bit identifier should be set to 0 ? From samuel at acid-code.ch Sat Jul 22 05:14:35 2006 From: samuel at acid-code.ch (Samuel Burri) Date: Sat Jul 22 05:13:32 2006 Subject: [Re: [icfp-discuss] #8. Allocation.] Message-ID: <50900.84.226.35.108.1153559675.squirrel@serioussam.selfip.org> Whereas I say, that in reg[b] must be placed an identifier for the newly allocated array, so that it can be accessed by operations 1,2, etc. The array must have enough entries to store variables as indicated by reg[c]. Obviously an array identifier can't be 0 (bit pattern consisting of exclusively the value 0) But I also admit that it's quite hard to understand for people who learned english in school. (Like me as an example) Sam > #8. Allocation. > > A new array is created with a capacity of platters > commensurate to the value in the register C. This > new array is initialized entirely with platters > holding the value 0. A bit pattern not consisting of > exclusively the 0 bit, and that identifies no other > active allocated array, is placed in the B register. > > I mean: > Identifier of newly created array is 0. (The value of register B is equal > 0) > > P. > > On 7/22/06, Samuel Burri wrote: >> >> How can #119 be accessed with a three bit address? >> Just wondering... >> By the way: How do you mean "'0' array is allocating"? >> >> Sam >> >> > >> >> Is it possible that '0' array is allocating? >> > >> > That's me as well ... or trying to access general purpose register >> 119. >> > If I don't mind accessing the later, my machine passes the self test >> *g* >> > >> > Regards >> > Daniel >> > _______________________________________________ >> > icfpcontest-discuss mailing list >> > icfpcontest-discuss@lists.andrew.cmu.edu >> > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss >> > >> > >> >> >> _______________________________________________ >> icfpcontest-discuss mailing list >> icfpcontest-discuss@lists.andrew.cmu.edu >> https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss >> > From Tiphaine.Turpin at irisa.fr Sat Jul 22 05:17:23 2006 From: Tiphaine.Turpin at irisa.fr (Tiphaine Turpin) Date: Sat Jul 22 05:23:06 2006 Subject: [icfp-discuss] Two small hits for Java programmers out there In-Reply-To: <7A0926F0-CF04-4947-9DC0-7B09B20AE27E@infonie.fr> References: <20060721232239.98889.qmail@web50715.mail.yahoo.com> <001401c6ad60$be243f30$4101a8c0@andromeda> <20060722073533.GI26438@cs.cmu.edu> <323D6D24-F1C4-4925-BA07-8CC58F81491C@infonie.fr> <7A0926F0-CF04-4947-9DC0-7B09B20AE27E@infonie.fr> Message-ID: <44C1ED23.8000807@irisa.fr> Laurent Vaucher wrote: > Having solved the intro problem, I feel in a generous mood. The two > things that prevented me from getting there sooner are the following : > - unsigned 32 bits integer arithmetic is hard in Java. I mean, Java > does not have *unsigned* integer types. You'll have to juggle a bit > (with longs and bitwise operations) to get it right. We had the same problem in ocaml (another typed language :-) and the only solution we found was using external C functions (just spent 3 hours to solve our problems with the interfacing :-(. The Makakapuss Team > - don't use PrintStream or PrintWriter.print() to produce output: it > has the nasty property of trying to convert character values to a > given encoding, which is why my second program ran in circles for ten > hours... > > And congratulations to the CMU POP team for this ongoing contest. > > > Laurent. > > > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss From samuel at acid-code.ch Sat Jul 22 05:24:55 2006 From: samuel at acid-code.ch (Samuel Burri) Date: Sat Jul 22 05:23:55 2006 Subject: [icfp-discuss] Re: operation #12 (load program) In-Reply-To: References: Message-ID: <50997.84.226.35.108.1153560295.squirrel@serioussam.selfip.org> I think the '0' should be replaced by a copy of the array identified by the value in reg[b]. Of course if reg[b] is zero only the execution finger should be changed. >> Does operation #12 lead to any net new active array? > clarification of teh question: > > the um-spec says: > "#12. Load Program. The array identified by the B register is > duplicated and the duplicate shall replace the '0' array, regardless > of size." > > by "replace the '0' array" does one mean > > 1. start treating the duplicate as the program? (so there will be no > way to reference any platter within the the program array but the '0' > array would be unchanged) > > or > > 2. the duplicate's 32-bit identifier should be set to 0 ? > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > > From franke.daniel at gmail.com Sat Jul 22 05:28:06 2006 From: franke.daniel at gmail.com (Daniel Franke) Date: Sat Jul 22 05:28:31 2006 Subject: [Re: [icfp-discuss] #8. Allocation.] In-Reply-To: <50900.84.226.35.108.1153559675.squirrel@serioussam.selfip.org> References: <50900.84.226.35.108.1153559675.squirrel@serioussam.selfip.org> Message-ID: <200607221128.06999.franke.daniel@gmail.com> On Saturday 22 July 2006 11:14, Samuel Burri wrote: > Whereas I say, that in reg[b] must be placed an identifier for the newly > allocated array, so that it can be accessed by operations 1,2, etc. The > array must have enough entries to store variables as indicated by reg[c]. > Obviously an array identifier can't be 0 (bit pattern consisting of > exclusively the value 0) > But I also admit that it's quite hard to understand for people who learned > english in school. (Like me as an example) *swearwords censored by higher entities* The register pointed to by operand B does not HOLD an array descriptor, but one is supposed to PLACE one there? *ouch* Ok, now it works. Thanks Samuel! From asynth at gmail.com Sat Jul 22 05:31:38 2006 From: asynth at gmail.com (James McCartney) Date: Sat Jul 22 05:31:40 2006 Subject: [icfp-discuss] I get an illegal instruction during LOADING: Message-ID: I enter my decription key. It says: decrypting... ok LOADING: and then I get an Array Index instruction at element 1 of a 1 element array and my virtual machine exits. I assume arrays are zero based because there is a zero array access in the self check. So element 1 should be illegal, correct. I guess I still have a bug in the machine and I'm just executing bad code. -- --- james mccartney From 0zu31 at gmx.de Sat Jul 22 05:22:47 2006 From: 0zu31 at gmx.de (Mechthild 'Rejis' Czapp) Date: Sat Jul 22 05:47:36 2006 Subject: [icfp-discuss] Understanding the text Message-ID: <200607221122.47790.0zu31@gmx.de> Hi fellow competitors, I am having problems understanding the text and the US-Americans and Brits in my team* couldn't help me: * "#2. Array Amendment" this increases the size of the array? or is it just an assign? * Do the arrays need to be consecutive? * "If the end of input has been signaled, then the register C is endowed with a uniform value pattern where every place is pregnant with the 1 bit." This means maxint, right? * "The '0' array shall be the most sublime choice for loading, and shall be handled with the utmost velocity." This means what? Thank you in advance and excuse my horrible English! Yours sincerely, Mechthild *a bunch of wannabes named after the IRC-channel, we meet in, so nothing special From samuel at acid-code.ch Sat Jul 22 05:53:54 2006 From: samuel at acid-code.ch (Samuel Burri) Date: Sat Jul 22 05:52:52 2006 Subject: [icfp-discuss] Understanding the text In-Reply-To: <200607221122.47790.0zu31@gmx.de> References: <200607221122.47790.0zu31@gmx.de> Message-ID: <51592.84.226.35.108.1153562034.squirrel@serioussam.selfip.org> Hi there #2 means just an assing to the designated position. As long as arrays can be accessed with indices from zero to size-1 they don't need to be conscutive. Altough I think that helps. And a value of all ones (~0) is maxint or 0xFFFFFFFF. With utmost velocity just means that its not wise to selfassign the '0'-array. An adjustment of the 'execution finger' suffices in the latter case. Just my two cents. Sam > Hi fellow competitors, > > I am having problems understanding the text and the US-Americans and Brits > in > my team* couldn't help me: > * "#2. Array Amendment" this increases the size of the array? or is it > just an > assign? > * Do the arrays need to be consecutive? > * "If the end of input has been signaled, then the register C is endowed > with > a uniform value pattern where every place is pregnant with the 1 bit." > This > means maxint, right? > * "The '0' array shall be the most sublime choice for loading, and shall > be > handled with the utmost velocity." This means what? > > Thank you in advance and excuse my horrible English! > > Yours sincerely, > Mechthild > > *a bunch of wannabes named after the IRC-channel, we meet in, so nothing > special > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > > From rafe at csse.unimelb.edu.au Sat Jul 22 06:04:14 2006 From: rafe at csse.unimelb.edu.au (Ralph Becket) Date: Sat Jul 22 06:04:21 2006 Subject: [icfp-discuss] Re: I get an illegal instruction during LOADING: Message-ID: <20060722100414.GD16251@ceres.cs.mu.oz.au> I get a similar problem. Here is the output of my program with some tracing turned on: $ time ./um64 codex.umz < key * allocated array 1 of size 4 * deleted array 1 * allocated array 2 of size 4 * deleted array 2 * allocated array 3 of size 138 * allocated array 4 of size 138 * deleted array 3 * deleted array 4 * allocated array 5 of size 0 * deleted array 5 self-check succeeded! enter decryption key: decrypting... * allocated array 6 of size 584445 ok * loaded array 6; executing from 0 LOADING: * allocated array 7 of size 7 * allocated array 8 of size 986508 * allocated array 9 of size 986508 ! array index error: array 7 has size 7; index was 7 a = 187 b = 7 c = 433194207 d = 1 e = 7 f = 182 g = 0 h = 48 ! 129175640 instructions executed real 0m5.734s user 0m5.713s sys 0m0.016s My guess is that decryption is not working correctly under my interpreter. However, three of us have spent much time going over the same 150 lines of code implementing the fetch-execute cycle (the bulk of which is for sanity checking and tracing) with a fine-tooth comb, but fail to see where the problem lies. We've had identical results from - a 32 bit C program - a 64 bit C program - implementing modulo reduction as ((x) % (1L<<32)) - implementing modulo reduction as ((x) & ((1L<<32) - 1)) - implementing modulo reduction as ((x) & 0xffffffff) - a Mercury program using arbitrary precision integer arithmetic. Gah! -- Ralph From asynth at gmail.com Sat Jul 22 06:13:25 2006 From: asynth at gmail.com (James McCartney) Date: Sat Jul 22 06:13:27 2006 Subject: [icfp-discuss] Re: I get an illegal instruction during LOADING: In-Reply-To: <20060722100414.GD16251@ceres.cs.mu.oz.au> References: <20060722100414.GD16251@ceres.cs.mu.oz.au> Message-ID: On 7/22/06, Ralph Becket wrote: > LOADING: * allocated array 7 of size 7 I get exactly the same result here if I allocate my array IDs consecutively like you are doing. If I use a stack to reuse the abandoned IDs, then I get array #1 allocated to size 1 and it then tries to access element 1 of array #1. -- --- james mccartney From franke.daniel at gmail.com Sat Jul 22 06:17:43 2006 From: franke.daniel at gmail.com (Daniel Franke) Date: Sat Jul 22 06:17:54 2006 Subject: [icfp-discuss] ascii or binary key? Message-ID: <200607221217.43732.franke.daniel@gmail.com> Is the key to be interpreted literally or as binary, e.g. if there is something as "\a" in there, two input characters, '\' and 'a' or a single input 0x7 ? Thanks Daniel From probst at imm.dtu.dk Sat Jul 22 06:24:07 2006 From: probst at imm.dtu.dk (Christian W. Probst) Date: Sat Jul 22 06:24:20 2006 Subject: [icfp-discuss] ascii or binary key? In-Reply-To: <200607221217.43732.franke.daniel@gmail.com> References: <200607221217.43732.franke.daniel@gmail.com> Message-ID: <44C1FCC7.5050307@imm.dtu.dk> Daniel Franke wrote: > Is the key to be interpreted literally or as binary, e.g. if there is > something as "\a" in there, two input characters, '\' and 'a' or a single > input 0x7 ? two characters /chris > > Thanks > Daniel > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > -- <><> Christian W. Probst * Technical University of Denmark <><> Assistant Professor * www.imm.dtu.dk/~probst <><> phone +45 45 25 75 12 * fax +45 45 93 00 74 From popiel at wolfskeep.com Sat Jul 22 06:46:39 2006 From: popiel at wolfskeep.com (T. Alexander Popiel) Date: Sat Jul 22 06:46:41 2006 Subject: [icfp-discuss] How to use the decryption key on the output ? In-Reply-To: <323D6D24-F1C4-4925-BA07-8CC58F81491C@infonie.fr> References: <20060721232239.98889.qmail@web50715.mail.yahoo.com> <001401c6ad60$be243f30$4101a8c0@andromeda> <20060722073533.GI26438@cs.cmu.edu> <323D6D24-F1C4-4925-BA07-8CC58F81491C@infonie.fr> Message-ID: <20060722104639.1F5F12DDB5@cashew.wolfskeep.com> In message: <323D6D24-F1C4-4925-BA07-8CC58F81491C@infonie.fr> Laurent Vaucher writes: >> Depending on how you implemented your UM, you may need to give the >> UM a >> newline before it reads the input. > > Can the newline be a problem? Because I still can't see why my UM >runs the codex perfectly but doesn't make any progress on the next >program. I would suggest that your memory management routines need optimization. The initial codex only allocates about a dozen arrays; the following codex allocates and deallocates thousands, or possibly millions, depending on how long you interact with it. I was having unpleasantly slow response until I optimized the heck out of operations #8 and #9... - Alex From Alain.Frisch at inria.fr Sat Jul 22 06:49:33 2006 From: Alain.Frisch at inria.fr (Alain Frisch) Date: Sat Jul 22 06:49:37 2006 Subject: [icfp-discuss] Problem with the XML normalizer Message-ID: <44C202BD.1070506@inria.fr> Hello, I've a problem with the advise tool (user hmonk), for the xml task. =============================================== Initial Term: SNF (Seq (Tag Bold (Tag Maj (Tag Bold (Tag Emph A)))) (Tag Emph (Seq (Tag Emph (Tag Maj B)) (Seq (Tag Bold A) (Tag Bold B))))) Applying advice... Final Term: ((Seq ((Tag Bold) ((Tag Maj) ((Tag Emph) A)))) ((Tag Emph) ((Seq ((Tag Emph) ((Tag Maj) B))) ((Tag Bold) ((Seq A) B))))) Total Steps: 47 INCORRECT! Final term is not in short normal form. =============================================== I don't see why the final term is not in normal form according to the spec. Does somebody see what's going wrong? Alain From samuel at acid-code.ch Sat Jul 22 06:53:26 2006 From: samuel at acid-code.ch (Samuel Burri) Date: Sat Jul 22 06:52:27 2006 Subject: [icfp-discuss] Login fails Message-ID: <54076.84.226.35.108.1153565606.squirrel@serioussam.selfip.org> Hi there fellows Just a few questions, thought that maybe someone could help me. When I run the code obtained after decrypting the codex it takes like forever (>10m on an 2GHz Pentium-M) when finaly after a series of blank lines the login shows up. But yet I can't login with guest. It just prompts for a password and then it continues forever. It would be very kind if someone could help me. Maybe just pointing out some clues I might have overlooked. I think there must be some fundamental error in my UM yet I can't find it. Just for the login. How am I supposed to supply the username? Does there have to be an endline-character after the username or is an EOF expected? It just isn't fun to wait every time +10m and then fail with the login... Thanks all Sam From niklas.broberg at gmail.com Sat Jul 22 07:15:47 2006 From: niklas.broberg at gmail.com (Niklas Broberg) Date: Sat Jul 22 07:15:52 2006 Subject: [icfp-discuss] Scoring question Message-ID: Some problems obviously gives a different amount of score depending on how "well" you do them. What happens if I submit a publication for an accomplished task, and then find a better way to solve it for a higher score. If I submit the second publication afterwards, will my score increase that number, or is it the first submitted score that counts forevermore? Thanks, /Niklas, finger on the Submit button From tom7 at cs.cmu.edu Sat Jul 22 07:50:27 2006 From: tom7 at cs.cmu.edu (Tom Murphy) Date: Sat Jul 22 07:44:39 2006 Subject: [icfp-discuss] Scoring question In-Reply-To: References: Message-ID: <44C21103.4030201@cs.cmu.edu> If you find a better publication, submit it, and it will be reflected in your score (replacing your previous best). Tom Niklas Broberg wrote: > Some problems obviously gives a different amount of score depending on > how "well" you do them. What happens if I submit a publication for an > accomplished task, and then find a better way to solve it for a higher > score. If I submit the second publication afterwards, will my score > increase that number, or is it the first submitted score that counts > forevermore? > > Thanks, > > /Niklas, finger on the Submit button > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > From tom7 at cs.cmu.edu Sat Jul 22 07:51:27 2006 From: tom7 at cs.cmu.edu (Tom Murphy) Date: Sat Jul 22 07:45:38 2006 Subject: [icfp-discuss] Login fails In-Reply-To: <54076.84.226.35.108.1153565606.squirrel@serioussam.selfip.org> References: <54076.84.226.35.108.1153565606.squirrel@serioussam.selfip.org> Message-ID: <44C2113F.5000501@cs.cmu.edu> If it takes 10 minutes to get to the prompt, the next phases will probably be unusably slow. It takes less than one second for the prompt on my 1GHz Pentium M. The prompt just expects you to type guest and press enter. Tom Samuel Burri wrote: > Hi there fellows > > Just a few questions, thought that maybe someone could help me. > > When I run the code obtained after decrypting the codex it takes like > forever (>10m on an 2GHz Pentium-M) when finaly after a series of blank > lines the login shows up. > > But yet I can't login with guest. It just prompts for a password and then > it continues forever. It would be very kind if someone could help me. > Maybe just pointing out some clues I might have overlooked. I think there > must be some fundamental error in my UM yet I can't find it. > > Just for the login. How am I supposed to supply the username? Does there > have to be an endline-character after the username or is an EOF expected? > > It just isn't fun to wait every time +10m and then fail with the login... > > Thanks all > > Sam > > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > From drochom at googlemail.com Sat Jul 22 07:51:35 2006 From: drochom at googlemail.com (przemek drochomirecki) Date: Sat Jul 22 07:51:39 2006 Subject: [icfp-discuss] Scoring question In-Reply-To: <44C21103.4030201@cs.cmu.edu> References: <44C21103.4030201@cs.cmu.edu> Message-ID: <8daef39f0607220451p4aff2ef0k6f77604387570e54@mail.gmail.com> Does anyone have problems with submitting publications? IS XXXXXXXX.LOG=XXX@XXXXXXXXX..... A VALID PUBLICATION? cheers Przemek On 7/22/06, Tom Murphy wrote: > > If you find a better publication, submit it, and it will be reflected in > your score (replacing your previous best). > > Tom > > Niklas Broberg wrote: > > Some problems obviously gives a different amount of score depending on > > how "well" you do them. What happens if I submit a publication for an > > accomplished task, and then find a better way to solve it for a higher > > score. If I submit the second publication afterwards, will my score > > increase that number, or is it the first submitted score that counts > > forevermore? > > > > Thanks, > > > > /Niklas, finger on the Submit button > > _______________________________________________ > > icfpcontest-discuss mailing list > > icfpcontest-discuss@lists.andrew.cmu.edu > > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > > > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060722/ef9a10ae/attachment-0001.html From tom7 at cs.cmu.edu Sat Jul 22 08:00:30 2006 From: tom7 at cs.cmu.edu (Tom Murphy) Date: Sat Jul 22 07:54:38 2006 Subject: [icfp-discuss] Scoring question In-Reply-To: <8daef39f0607220451p4aff2ef0k6f77604387570e54@mail.gmail.com> References: <44C21103.4030201@cs.cmu.edu> <8daef39f0607220451p4aff2ef0k6f77604387570e54@mail.gmail.com> Message-ID: <44C2135E.8050503@cs.cmu.edu> If you have a publication that you think should be accepted, but isn't, please do email the organizers list. Tom przemek drochomirecki wrote: > Does anyone have problems with submitting publications? > > IS XXXXXXXX.LOG=XXX@XXXXXXXXX..... A VALID PUBLICATION? > > cheers > Przemek > From gus-icfp at guxx.net Sat Jul 22 08:00:51 2006 From: gus-icfp at guxx.net (G.Schoepp) Date: Sat Jul 22 08:00:50 2006 Subject: [icfp-discuss] Login fails In-Reply-To: <54076.84.226.35.108.1153565606.squirrel@serioussam.selfip.org> References: <54076.84.226.35.108.1153565606.squirrel@serioussam.selfip.org> Message-ID: <44C21373.10006@guxx.net> Samuel Burri schrieb: > Just a few questions, thought that maybe someone could help me. > > When I run the code obtained after decrypting the codex it takes like > forever (>10m on an 2GHz Pentium-M) when finaly after a series of blank > lines the login shows up. On my 2GHz Athlon (Windows 2000) it takes 10 secs until the 15MB UMIX (the decrypted codex) is loaded and another 5 secs until the login prompt appears. > But yet I can't login with guest. It just prompts for a password and then > it continues forever. [...] > Just for the login. How am I supposed to supply the username? Does > there have to be an endline-character after the username or is an > EOF expected? If it asks for a password something is wrong with your UM. In my case I forgot to add a newline (0xA) after my input. Guido From drochom at googlemail.com Sat Jul 22 08:02:41 2006 From: drochom at googlemail.com (przemek drochomirecki) Date: Sat Jul 22 08:02:43 2006 Subject: [icfp-discuss] submitting publication issue Message-ID: <8daef39f0607220502q1ac86951k97da3769ee4f9981@mail.gmail.com> hi, i have 8 different publications but i receive the same error all the time " error: invalid publication". do u have any idea what is wrong? thx in adv. przemek -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060722/098c9525/attachment.html From niklas.broberg at gmail.com Sat Jul 22 08:04:09 2006 From: niklas.broberg at gmail.com (Niklas Broberg) Date: Sat Jul 22 08:04:12 2006 Subject: [icfp-discuss] Scoring question In-Reply-To: <44C21103.4030201@cs.cmu.edu> References: <44C21103.4030201@cs.cmu.edu> Message-ID: > If you find a better publication, submit it, and it will be reflected in > your score (replacing your previous best). Great, thanks a lot! /Niklas goes to push those buttons From finnw at sucs.org Sat Jul 22 08:07:58 2006 From: finnw at sucs.org (Finn Wilcox) Date: Sat Jul 22 08:08:08 2006 Subject: [icfp-discuss] Scoring question In-Reply-To: <8daef39f0607220451p4aff2ef0k6f77604387570e54@mail.gmail.com> References: <44C21103.4030201@cs.cmu.edu> <8daef39f0607220451p4aff2ef0k6f77604387570e54@mail.gmail.com> Message-ID: On Sat, 22 Jul 2006, przemek drochomirecki wrote: > Does anyone have problems with submitting publications? > > IS XXXXXXXX.LOG=XXX@XXXXXXXXX..... A VALID PUBLICATION? > > cheers > Przemek > Check for leading/trailing spaces. You need to remove them. Finn From gs at guxx.net Sat Jul 22 08:13:29 2006 From: gs at guxx.net (G.Schoepp) Date: Sat Jul 22 08:13:28 2006 Subject: [icfp-discuss] High array count Message-ID: <44C21669.20000@guxx.net> Hi, after loading UMIX and logging in the number of arrays of the platters (the UM memory) is 649,826! Is this normal? Or does it indicate an error in reusing the arrays? Guido From gus-icfp at guxx.net Sat Jul 22 08:13:50 2006 From: gus-icfp at guxx.net (G.Schoepp) Date: Sat Jul 22 08:13:49 2006 Subject: [icfp-discuss] High array count Message-ID: <44C2167E.8050506@guxx.net> Hi, after loading UMIX and logging in the number of arrays of the platters (the UM memory) is 649,826! Is this normal? Or does it indicate an error in reusing the arrays? Guido From jjmaes at gmail.com Sat Jul 22 08:14:41 2006 From: jjmaes at gmail.com (James Darwin Maes The First (a javatarian)) Date: Sat Jul 22 08:14:44 2006 Subject: [icfp-discuss] Two small hits for Java programmers out there In-Reply-To: <7A0926F0-CF04-4947-9DC0-7B09B20AE27E@infonie.fr> References: <20060721232239.98889.qmail@web50715.mail.yahoo.com> <001401c6ad60$be243f30$4101a8c0@andromeda> <20060722073533.GI26438@cs.cmu.edu> <323D6D24-F1C4-4925-BA07-8CC58F81491C@infonie.fr> <7A0926F0-CF04-4947-9DC0-7B09B20AE27E@infonie.fr> Message-ID: Ouch! That Hurt! you hit me! Judges, he hit me! Mommy!!!!! On 7/22/06, Laurent Vaucher wrote: > > Having solved the intro problem, I feel in a generous mood. The two > things that prevented me from getting there sooner are the following : > - unsigned 32 bits integer arithmetic is hard in Java. I mean, Java > does not have *unsigned* integer types. You'll have to juggle a bit > (with longs and bitwise operations) to get it right. > - don't use PrintStream or PrintWriter.print() to produce output: it > has the nasty property of trying to convert character values to a > given encoding, which is why my second program ran in circles for ten > hours... > > And congratulations to the CMU POP team for this ongoing contest. > > > Laurent. > > > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060722/4a45525d/attachment.html From tom7 at cs.cmu.edu Sat Jul 22 08:32:27 2006 From: tom7 at cs.cmu.edu (Tom Murphy) Date: Sat Jul 22 08:26:20 2006 Subject: [icfp-discuss] High array count In-Reply-To: <44C2167E.8050506@guxx.net> References: <44C2167E.8050506@guxx.net> Message-ID: <44C21ADB.6040609@cs.cmu.edu> This definitely sounds within reason. Sand was very plentiful in 19100. Tom G.Schoepp wrote: > Hi, > > after loading UMIX and logging in the number of arrays of the platters > (the UM memory) is 649,826! > Is this normal? Or does it indicate an error in reusing the arrays? > > Guido > > > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > From contest.crackers at gmail.com Sat Jul 22 08:31:33 2006 From: contest.crackers at gmail.com (Contest Crackers) Date: Sat Jul 22 08:31:36 2006 Subject: [icfp-discuss] All we get is a menu to dump UM data Message-ID: <472ee7770607220531p2e975d33qa54420834c8859ce@mail.gmail.com> Hi, After providing our decryption key, all we get (after waiting for about 5 minutes) is this. decrypting... ok LOADING: 9876543210 == CBV ARCHIVE == VOLUME ID 7 Choose a command: p) dump UM data x) exit ? INPUT: For the dump option (p) there is huge garbled dump. So this would be a bug in our UM? Has anyone else seen this? We tried tweaking some of our "strategies", it doesn't seem to matter. - "Contest Crackers" From kilop84 at gmail.com Sat Jul 22 08:31:48 2006 From: kilop84 at gmail.com (Oge) Date: Sat Jul 22 08:31:50 2006 Subject: [icfp-discuss] byte 26 Message-ID: <13690c9c0607220531v27e41e3dof5ce1484865eadd7@mail.gmail.com> Hi, I'm still decoding the codex and I am using Ruby. When I try to read in the codex file, I only receive bytes up to where there exists byte 26 (Ctrl Z). Have any of you encountered such a problem, and if so, how did you work around it? Oge. From tom7 at cs.cmu.edu Sat Jul 22 08:40:51 2006 From: tom7 at cs.cmu.edu (Tom Murphy) Date: Sat Jul 22 08:34:42 2006 Subject: [icfp-discuss] All we get is a menu to dump UM data In-Reply-To: <472ee7770607220531p2e975d33qa54420834c8859ce@mail.gmail.com> References: <472ee7770607220531p2e975d33qa54420834c8859ce@mail.gmail.com> Message-ID: <44C21CD3.9010404@cs.cmu.edu> Looks like you're on the right track! We suggest saving the data to a file, having a look at it, and then doing whatever you tend to do with mysterious UM data from the ICFP organizers... Tom Contest Crackers wrote: > Hi, > > After providing our decryption key, all we get (after waiting for > about 5 minutes) is this. > > decrypting... > ok > LOADING: > 9876543210 > == CBV ARCHIVE == > VOLUME ID 7 > > Choose a command: > > p) dump UM data > x) exit > > ? > INPUT: > > For the dump option (p) there is huge garbled dump. > > So this would be a bug in our UM? Has anyone else seen this? We tried > tweaking some of our "strategies", it doesn't seem to matter. > > - "Contest Crackers" > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > From niklas.broberg at gmail.com Sat Jul 22 09:03:22 2006 From: niklas.broberg at gmail.com (Niklas Broberg) Date: Sat Jul 22 09:03:37 2006 Subject: [icfp-discuss] Problem with the XML normalizer In-Reply-To: <44C202BD.1070506@inria.fr> References: <44C202BD.1070506@inria.fr> Message-ID: Maj > Emph, so the first line should be (equiv to) (Tag Bold (Tag Emph (Tag Maj A))), you've got the last two the wrong way around. :-) /Niklas On 7/22/06, Alain Frisch wrote: > Hello, > > I've a problem with the advise tool (user hmonk), for the xml task. > > =============================================== > Initial Term: SNF (Seq (Tag Bold (Tag Maj (Tag Bold (Tag Emph A)))) > (Tag Emph (Seq (Tag Emph (Tag Maj B)) (Seq (Tag Bold A) (Tag Bold B))))) > Applying advice... > Final Term: ((Seq ((Tag Bold) ((Tag Maj) ((Tag Emph) A)))) ((Tag > Emph) ((Seq ((Tag Emph) ((Tag Maj) B))) ((Tag Bold) ((Seq A) B))))) > Total Steps: 47 > > INCORRECT! > Final term is not in short normal form. > =============================================== > > I don't see why the final term is not in normal form according to the > spec. Does somebody see what's going wrong? > > Alain > > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > From blo.b at infonie.fr Sat Jul 22 09:05:59 2006 From: blo.b at infonie.fr (Laurent Vaucher) Date: Sat Jul 22 09:06:23 2006 Subject: [icfp-discuss] Two small hints for Java programmers out there In-Reply-To: References: <20060721232239.98889.qmail@web50715.mail.yahoo.com> <001401c6ad60$be243f30$4101a8c0@andromeda> <20060722073533.GI26438@cs.cmu.edu> <323D6D24-F1C4-4925-BA07-8CC58F81491C@infonie.fr> <7A0926F0-CF04-4947-9DC0-7B09B20AE27E@infonie.fr> Message-ID: > Ouch! That Hurt! you hit me! Judges, he hit me! Mommy!!!!! Sorry, I corrected the subject :o) Hope it was not too painful. From papierfalter at yahoo.de Sat Jul 22 09:16:19 2006 From: papierfalter at yahoo.de (Wolfram Fischer) Date: Sat Jul 22 09:17:17 2006 Subject: [icfp-discuss] Not-And Message-ID: <44C22523.8000104@yahoo.de> First off all: Thanks for the contest! I've a question regarding the 'Not-And' Operator (English isn't my native language so I just want to verify my understanding). Does it work like this: A B RESULT 0 0 0 0 1 1 1 0 1 1 1 0 Thanks in advance and good luck to everyone! Wolf ___________________________________________________________ Telefonate ohne weitere Kosten vom PC zum PC: http://messenger.yahoo.de From tom7 at cs.cmu.edu Sat Jul 22 09:25:51 2006 From: tom7 at cs.cmu.edu (Tom Murphy) Date: Sat Jul 22 09:21:12 2006 Subject: [icfp-discuss] Not-And In-Reply-To: <44C22523.8000104@yahoo.de> References: <44C22523.8000104@yahoo.de> Message-ID: <44C2275F.2020302@cs.cmu.edu> Nope, not-and is bitwise and followed by bitwise not. The truth table looks like this: A B NAND(A, B) 0 0 1 0 1 1 1 0 1 1 1 0 Tom Wolfram Fischer wrote: > First off all: Thanks for the contest! > > I've a question regarding the 'Not-And' Operator (English isn't my > native language so I just want to verify my understanding). > > Does it work like this: > > A B RESULT > 0 0 0 > 0 1 1 > 1 0 1 > 1 1 0 > > Thanks in advance and good luck to everyone! > > Wolf > > > > ___________________________________________________________ Telefonate > ohne weitere Kosten vom PC zum PC: http://messenger.yahoo.de > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > From samuel at acid-code.ch Sat Jul 22 09:26:02 2006 From: samuel at acid-code.ch (Samuel Burri) Date: Sat Jul 22 09:25:27 2006 Subject: [icfp-discuss] Not-And In-Reply-To: <44C22523.8000104@yahoo.de> References: <44C22523.8000104@yahoo.de> Message-ID: <58102.84.226.35.108.1153574762.squirrel@serioussam.selfip.org> No, it works like > A B RESULT > 0 0 1 > 0 1 1 > 1 0 1 > 1 1 0 In the first line its 1 to result = ~(A & B) > First off all: Thanks for the contest! > > I've a question regarding the 'Not-And' Operator (English isn't my > native language so I just want to verify my understanding). > > Does it work like this: > > A B RESULT > 0 0 0 > 0 1 1 > 1 0 1 > 1 1 0 > > Thanks in advance and good luck to everyone! > > Wolf > > > > ___________________________________________________________ > Telefonate ohne weitere Kosten vom PC zum PC: http://messenger.yahoo.de > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > > From jjmaes at gmail.com Sat Jul 22 09:44:51 2006 From: jjmaes at gmail.com (James Darwin Maes The First (a javatarian)) Date: Sat Jul 22 09:45:31 2006 Subject: [icfp-discuss] Two small hints for Java programmers out there In-Reply-To: References: <20060721232239.98889.qmail@web50715.mail.yahoo.com> <001401c6ad60$be243f30$4101a8c0@andromeda> <20060722073533.GI26438@cs.cmu.edu> <323D6D24-F1C4-4925-BA07-8CC58F81491C@infonie.fr> <7A0926F0-CF04-4947-9DC0-7B09B20AE27E@infonie.fr> Message-ID: wow guy who corrects bad subject lines because of guy who makes bad subject line jokes made a subject line joke about his bad subject line. On 7/22/06, Laurent Vaucher wrote: > > > Ouch! That Hurt! you hit me! Judges, he hit me! Mommy!!!!! > > Sorry, I corrected the subject :o) Hope it was not too painful. > > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060722/677acf16/attachment-0001.html From papierfalter at yahoo.de Sat Jul 22 09:49:10 2006 From: papierfalter at yahoo.de (Wolfram Fischer) Date: Sat Jul 22 09:49:28 2006 Subject: [icfp-discuss] Not-And In-Reply-To: <44C2275F.2020302@cs.cmu.edu> References: <44C22523.8000104@yahoo.de> <44C2275F.2020302@cs.cmu.edu> Message-ID: <44C22CD6.8060303@yahoo.de> Looks like a NAND to me ^^. Thank you very much! Tom Murphy schrieb: > > Nope, not-and is bitwise and followed by bitwise not. The truth table > looks like this: > > A B NAND(A, B) > 0 0 1 > 0 1 1 > 1 0 1 > 1 1 0 > > Tom > > Wolfram Fischer wrote: >> First off all: Thanks for the contest! >> >> I've a question regarding the 'Not-And' Operator (English isn't my >> native language so I just want to verify my understanding). >> >> Does it work like this: >> >> A B RESULT >> 0 0 0 >> 0 1 1 >> 1 0 1 >> 1 1 0 >> >> Thanks in advance and good luck to everyone! >> >> Wolf >> >> >> ___________________________________________________________ >> Telefonate ohne weitere Kosten vom PC zum PC: http://messenger.yahoo.de >> _______________________________________________ >> icfpcontest-discuss mailing list >> icfpcontest-discuss@lists.andrew.cmu.edu >> https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss >> > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > ___________________________________________________________ Telefonate ohne weitere Kosten vom PC zum PC: http://messenger.yahoo.de From popiel at wolfskeep.com Sat Jul 22 10:26:06 2006 From: popiel at wolfskeep.com (T. Alexander Popiel) Date: Sat Jul 22 10:26:39 2006 Subject: [icfp-discuss] Login fails In-Reply-To: <44C21373.10006@guxx.net> References: <54076.84.226.35.108.1153565606.squirrel@serioussam.selfip.org> <44C21373.10006@guxx.net> Message-ID: <20060722142606.6C1132DDB5@cashew.wolfskeep.com> In message: <44C21373.10006@guxx.net> "G.Schoepp" writes: >Samuel Burri schrieb: >> Just a few questions, thought that maybe someone could help me. >> >> When I run the code obtained after decrypting the codex it takes like >> forever (>10m on an 2GHz Pentium-M) when finaly after a series of blank >> lines the login shows up. > >On my 2GHz Athlon (Windows 2000) it takes 10 secs until the 15MB UMIX >(the decrypted codex) is loaded and another 5 secs until the login >prompt appears. Well, if we're going to get into speed contests: CPU: AMD Sempron(tm) 2200+ (2994.99 bogomips) % (echo guest; echo logout) | time umo umix.umz 0.64user 0.12system 0:01.49elapsed 50%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (100major+14510minor)pagefaults 0swaps - Alex From dodnet at web.de Sat Jul 22 10:30:46 2006 From: dodnet at web.de (=?ISO-8859-15?Q?Pierre_Br=E4unig?=) Date: Sat Jul 22 10:29:51 2006 Subject: [icfp-discuss] UM wants decryption key twice Message-ID: <44C23696.3070106@web.de> Hi all, when I run my implementation of the UM it asks for the decryption key and starts decrypting. This takes about 20 secs then it says 'ok, self-check succeeded' and asks again for the decryption key. When I enter the same key again, it starts decrypting again and after a few seconds says 'wrong key'. Has anyone had the same problem? Maybe a wrong key? Any suggestions? Thanks for the contest - its fun so far :o) Pierre From davidad at mit.edu Sat Jul 22 10:34:02 2006 From: davidad at mit.edu (David Dalrymple) Date: Sat Jul 22 10:34:41 2006 Subject: [icfp-discuss] UM wants decryption key twice In-Reply-To: <44C23696.3070106@web.de> References: <44C23696.3070106@web.de> Message-ID: <504b2780607220734m5dd02968nccb7a948b35aa21@mail.gmail.com> I haven't even gotten quite that far yet but I recommend checking your instruction #12 to make sure it actually copies the array in the case that it's not 0. --David On 7/22/06, Pierre Br?unig wrote: > Hi all, > > when I run my implementation of the UM it asks for the decryption key > and starts decrypting. This takes about 20 secs then it says 'ok, > self-check succeeded' and asks again for the decryption key. When I > enter the same key again, it starts decrypting again and after a few > seconds says 'wrong key'. > Has anyone had the same problem? Maybe a wrong key? Any suggestions? > > Thanks for the contest - its fun so far :o) > Pierre > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > From contest.crackers at gmail.com Sat Jul 22 10:34:08 2006 From: contest.crackers at gmail.com (Contest Crackers) Date: Sat Jul 22 10:34:41 2006 Subject: [icfp-discuss] UM wants decryption key twice In-Reply-To: <44C23696.3070106@web.de> References: <44C23696.3070106@web.de> Message-ID: <472ee7770607220734u204ffbb7rc2af8ebe2b3ed48@mail.gmail.com> We saw this too. But it turned out to be a bug in our UM implementation of the Load operation. The problem went away when we fixed that. Thanks. On 7/22/06, Pierre Br?unig wrote: > Hi all, > > when I run my implementation of the UM it asks for the decryption key > and starts decrypting. This takes about 20 secs then it says 'ok, > self-check succeeded' and asks again for the decryption key. When I > enter the same key again, it starts decrypting again and after a few > seconds says 'wrong key'. > Has anyone had the same problem? Maybe a wrong key? Any suggestions? > > Thanks for the contest - its fun so far :o) > Pierre > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > From tom7 at cs.cmu.edu Sat Jul 22 10:41:42 2006 From: tom7 at cs.cmu.edu (Tom Murphy) Date: Sat Jul 22 10:36:49 2006 Subject: [icfp-discuss] UM wants decryption key twice In-Reply-To: <44C23696.3070106@web.de> References: <44C23696.3070106@web.de> Message-ID: <44C23926.5030803@cs.cmu.edu> After decrypting, the codex uses the Load Program instruction to jump to the decrypted code. If that instruction doesn't work properly, your UM might jump to the beginning of the codex again. Tom Pierre Br?unig wrote: > Hi all, > > when I run my implementation of the UM it asks for the decryption key > and starts decrypting. This takes about 20 secs then it says 'ok, > self-check succeeded' and asks again for the decryption key. When I > enter the same key again, it starts decrypting again and after a few > seconds says 'wrong key'. > Has anyone had the same problem? Maybe a wrong key? Any suggestions? > > Thanks for the contest - its fun so far :o) > Pierre > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > From contest.crackers at gmail.com Sat Jul 22 10:37:42 2006 From: contest.crackers at gmail.com (Contest Crackers) Date: Sat Jul 22 10:38:24 2006 Subject: [icfp-discuss] All we get is a menu to dump UM data In-Reply-To: <44C21CD3.9010404@cs.cmu.edu> References: <472ee7770607220531p2e975d33qa54420834c8859ce@mail.gmail.com> <44C21CD3.9010404@cs.cmu.edu> Message-ID: <472ee7770607220737t53a40116p6e22513ce7fdce3a@mail.gmail.com> Oh okay, thanks! The UM data however seems to be of a size in bytes that is not divisible by 4. (We have picked all the bytes after but not including the colon ":". Given that the platters all contain 32 bits, is this an error? Are we missing something? (Yeah, when we run it, it seeks input without having displayed anything. Still trying to figure out what input to provide). Thanks. - "Contest Crackers". On 7/22/06, Tom Murphy wrote: > Looks like you're on the right track! > We suggest saving the data to a file, having a look at it, and then > doing whatever you tend to do with mysterious UM data from the ICFP > organizers... > > Tom > > Contest Crackers wrote: > > Hi, > > > > After providing our decryption key, all we get (after waiting for > > about 5 minutes) is this. > > > > decrypting... > > ok > > LOADING: > > 9876543210 > > == CBV ARCHIVE == > > VOLUME ID 7 > > > > Choose a command: > > > > p) dump UM data > > x) exit > > > > ? > > INPUT: > > > > For the dump option (p) there is huge garbled dump. > > > > So this would be a bug in our UM? Has anyone else seen this? We tried > > tweaking some of our "strategies", it doesn't seem to matter. > > > > - "Contest Crackers" > > _______________________________________________ > > icfpcontest-discuss mailing list > > icfpcontest-discuss@lists.andrew.cmu.edu > > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > > > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > From ij7b3u502 at sneakemail.com Sat Jul 22 10:42:25 2006 From: ij7b3u502 at sneakemail.com (vbzoli) Date: Sat Jul 22 10:43:14 2006 Subject: [icfp-discuss] UM wants decryption key twice Message-ID: <654-94229@sneakemail.com> I had the same bug. Your array duplication code is buggy. Maybe it's copying the old array '0'. VOROSBARANYI Zoltan > [...] > When I > enter the same key again, it starts > decrypting again and after a few > seconds says 'wrong key'. From jmk at amoebaville.net Sat Jul 22 10:43:08 2006 From: jmk at amoebaville.net (Justin Koser) Date: Sat Jul 22 10:44:15 2006 Subject: [icfp-discuss] All we get is a menu to dump UM data In-Reply-To: <472ee7770607220737t53a40116p6e22513ce7fdce3a@mail.gmail.com> References: <472ee7770607220531p2e975d33qa54420834c8859ce@mail.gmail.com> <44C21CD3.9010404@cs.cmu.edu> <472ee7770607220737t53a40116p6e22513ce7fdce3a@mail.gmail.com> Message-ID: <44C2397C.9020804@amoebaville.net> Contest Crackers wrote: > The UM data however seems to be of a size in bytes that is not > divisible by 4. (We have picked all the bytes after but not including > the colon ":". Given that the platters all contain 32 bits, is this > an error? Are we missing something? I just checked, and that happened with me too, but my VM happens to ignore any odd bytes on the end and things run just fine. > (Yeah, when we run it, it seeks input without having displayed > anything. Still trying to figure out what input to provide). It should output something first... maybe it's just taking a long time to allocate arrays as previously discussed? -- Justin From shawnyar217 at yahoo.com Sat Jul 22 10:48:56 2006 From: shawnyar217 at yahoo.com (Shawn Yarbrough) Date: Sat Jul 22 10:49:35 2006 Subject: [icfp-discuss] Re: I get an illegal instruction during LOADING: In-Reply-To: <20060722100414.GD16251@ceres.cs.mu.oz.au> Message-ID: <20060722144856.21620.qmail@web53407.mail.yahoo.com> My trace agrees exactly with Ralph's trace -- every number. Too bad the codex doesn't self-test itself more. This array out-of-bounds error could be caused by anything that got corrupted earlier in the program. Shawn Ralph Becket wrote: I get a similar problem. Here is the output of my program with some tracing turned on: $ time ./um64 codex.umz < key * allocated array 1 of size 4 * deleted array 1 * allocated array 2 of size 4 * deleted array 2 * allocated array 3 of size 138 * allocated array 4 of size 138 * deleted array 3 * deleted array 4 * allocated array 5 of size 0 * deleted array 5 self-check succeeded! enter decryption key: decrypting... * allocated array 6 of size 584445 ok * loaded array 6; executing from 0 LOADING: * allocated array 7 of size 7 * allocated array 8 of size 986508 * allocated array 9 of size 986508 ! array index error: array 7 has size 7; index was 7 a = 187 b = 7 c = 433194207 d = 1 e = 7 f = 182 g = 0 h = 48 ! 129175640 instructions executed real 0m5.734s user 0m5.713s sys 0m0.016s My guess is that decryption is not working correctly under my interpreter. However, three of us have spent much time going over the same 150 lines of code implementing the fetch-execute cycle (the bulk of which is for sanity checking and tracing) with a fine-tooth comb, but fail to see where the problem lies. We've had identical results from - a 32 bit C program - a 64 bit C program - implementing modulo reduction as ((x) % (1L<<32)) - implementing modulo reduction as ((x) & ((1L<<32) - 1)) - implementing modulo reduction as ((x) & 0xffffffff) - a Mercury program using arbitrary precision integer arithmetic. Gah! -- Ralph _______________________________________________ icfpcontest-discuss mailing list icfpcontest-discuss@lists.andrew.cmu.edu https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss --------------------------------- Yahoo! Messenger with Voice. Make PC-to-Phone Calls to the US (and 30+ countries) for 2?/min or less. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060722/85f34b9b/attachment.html From samuel at acid-code.ch Sat Jul 22 10:50:58 2006 From: samuel at acid-code.ch (Samuel Burri) Date: Sat Jul 22 10:50:39 2006 Subject: [Re: [icfp-discuss] Login fails] Message-ID: <60338.84.226.35.108.1153579858.squirrel@serioussam.selfip.org> But how did you achieve that speed. I guess my memory allocation/deallocation is way too slow. But how do you improve this? Any suggestions? Whats an easy way to go? - Sam > In message: <44C21373.10006@guxx.net> > "G.Schoepp" writes: >>Samuel Burri schrieb: >>> Just a few questions, thought that maybe someone could help me. >>> >>> When I run the code obtained after decrypting the codex it takes like >>> forever (>10m on an 2GHz Pentium-M) when finaly after a series of blank >>> lines the login shows up. >> >>On my 2GHz Athlon (Windows 2000) it takes 10 secs until the 15MB UMIX >>(the decrypted codex) is loaded and another 5 secs until the login >>prompt appears. > > Well, if we're going to get into speed contests: > > CPU: AMD Sempron(tm) 2200+ (2994.99 bogomips) > > % (echo guest; echo logout) | time umo umix.umz > 0.64user 0.12system 0:01.49elapsed 50%CPU (0avgtext+0avgdata > 0maxresident)k > 0inputs+0outputs (100major+14510minor)pagefaults 0swaps > > - Alex > > From jmk at amoebaville.net Sat Jul 22 10:33:30 2006 From: jmk at amoebaville.net (Justin Koser) Date: Sat Jul 22 10:54:56 2006 Subject: [icfp-discuss] UM wants decryption key twice In-Reply-To: <44C23696.3070106@web.de> References: <44C23696.3070106@web.de> Message-ID: <44C2373A.8010606@amoebaville.net> Pierre Br?unig wrote: > when I run my implementation of the UM it asks for the decryption key > and starts decrypting. This takes about 20 secs then it says 'ok, > self-check succeeded' and asks again for the decryption key. When I > enter the same key again, it starts decrypting again and after a few > seconds says 'wrong key'. > Has anyone had the same problem? Maybe a wrong key? Any suggestions? Nah, sounds like a UM bug. The sequence of events should be: 1) self-check succeeded! 2) enter decryption key: 3) LOADING -- Justin From ij7b3u502 at sneakemail.com Sat Jul 22 10:54:58 2006 From: ij7b3u502 at sneakemail.com (vbzoli) Date: Sat Jul 22 10:56:02 2006 Subject: [icfp-discuss] All we get is a menu to dump UM data Message-ID: <11928-70109@sneakemail.com> The program allocates a really big number of small arrays. I had to rewrite the engine in C to get reasonable performance. VOROSBARANYI Zoltan From spoons at cs.cmu.edu Sat Jul 22 11:10:05 2006 From: spoons at cs.cmu.edu (Daniel Spoonhower) Date: Sat Jul 22 11:11:08 2006 Subject: [icfp-discuss] All we get is a menu to dump UM data In-Reply-To: <472ee7770607220737t53a40116p6e22513ce7fdce3a@mail.gmail.com> References: <472ee7770607220531p2e975d33qa54420834c8859ce@mail.gmail.com> <44C21CD3.9010404@cs.cmu.edu> <472ee7770607220737t53a40116p6e22513ce7fdce3a@mail.gmail.com> Message-ID: <44C23FCD.4070804@cs.cmu.edu> Crackers, The size of the dumped UM data should be 15901920 bytes. (This is all the bytes after the colon ":".) Did you accidentally add a character when you removed the prefix? Some editors will add an extra new line at end of the file. --djs Contest Crackers wrote: > Oh okay, thanks! > > The UM data however seems to be of a size in bytes that is not > divisible by 4. (We have picked all the bytes after but not including > the colon ":". Given that the platters all contain 32 bits, is this > an error? Are we missing something? > > (Yeah, when we run it, it seeks input without having displayed > anything. Still trying to figure out what input to provide). > > Thanks. > > - "Contest Crackers". > > On 7/22/06, Tom Murphy wrote: >> Looks like you're on the right track! >> We suggest saving the data to a file, having a look at it, and then >> doing whatever you tend to do with mysterious UM data from the ICFP >> organizers... >> >> Tom >> >> Contest Crackers wrote: >> > Hi, >> > >> > After providing our decryption key, all we get (after waiting for >> > about 5 minutes) is this. >> > >> > decrypting... >> > ok >> > LOADING: >> > 9876543210 >> > == CBV ARCHIVE == >> > VOLUME ID 7 >> > >> > Choose a command: >> > >> > p) dump UM data >> > x) exit >> > >> > ? >> > INPUT: >> > >> > For the dump option (p) there is huge garbled dump. >> > >> > So this would be a bug in our UM? Has anyone else seen this? We tried >> > tweaking some of our "strategies", it doesn't seem to matter. >> > >> > - "Contest Crackers" >> > _______________________________________________ >> > icfpcontest-discuss mailing list >> > icfpcontest-discuss@lists.andrew.cmu.edu >> > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss >> > >> _______________________________________________ >> icfpcontest-discuss mailing list >> icfpcontest-discuss@lists.andrew.cmu.edu >> https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss >> > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > From gs at guxx.net Sat Jul 22 11:11:12 2006 From: gs at guxx.net (G.Schoepp) Date: Sat Jul 22 11:12:32 2006 Subject: [Re: [icfp-discuss] Login fails] In-Reply-To: <60338.84.226.35.108.1153579858.squirrel@serioussam.selfip.org> References: <60338.84.226.35.108.1153579858.squirrel@serioussam.selfip.org> Message-ID: <44C24010.5040306@guxx.net> Samuel Burri schrieb: > But how did you achieve that speed. I guess my memory > allocation/deallocation is way too slow. But how do you improve this? > Any suggestions? Whats an easy way to go? The easy way depends on the programming language... In my case it was a collection class I used from the language's SDK for administering the arrays of platters. Compiling with the /bin/qbasic took about 30 minutes. I optimized this replacing the collection class with pure arrays. Now it takes 1 second. Guido From gus-icfp at guxx.net Sat Jul 22 11:11:42 2006 From: gus-icfp at guxx.net (gus) Date: Sat Jul 22 11:12:33 2006 Subject: [Re: [icfp-discuss] Login fails] In-Reply-To: <60338.84.226.35.108.1153579858.squirrel@serioussam.selfip.org> References: <60338.84.226.35.108.1153579858.squirrel@serioussam.selfip.org> Message-ID: <44C2402E.7050705@guxx.net> Samuel Burri schrieb: > But how did you achieve that speed. I guess my memory > allocation/deallocation is way too slow. But how do you improve this? > Any suggestions? Whats an easy way to go? The easy way depends on the programming language... In my case it was a collection class I used from the language's SDK for administering the arrays of platters. Compiling with the /bin/qbasic took about 30 minutes. I optimized this replacing the collection class with pure arrays. Now it takes 1 second. Guido From shawnyar217 at yahoo.com Sat Jul 22 11:23:16 2006 From: shawnyar217 at yahoo.com (Shawn Yarbrough) Date: Sat Jul 22 11:24:02 2006 Subject: [icfp-discuss] 32-bit division Message-ID: <20060722152316.7622.qmail@web53413.mail.yahoo.com> >From the spec for the division operator: The register A receives the value in register B divided by the value in register C, if any, where each quantity is treated treated as an unsigned 32 bit number. Anyone know why they specify unsigned 32-bit values again here? Platters are already unsigned 32-bit values. The "if any" is confusing, too. Register C always has a value in it, by definition, even if that value is zero or whatever. Shawn --------------------------------- Do you Yahoo!? Everyone is raving about the all-new Yahoo! Mail Beta. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060722/d188ff6a/attachment-0001.html From cwitty at gemini.newtonlabs.com Sat Jul 22 12:08:55 2006 From: cwitty at gemini.newtonlabs.com (Carl R. Witty) Date: Sat Jul 22 12:27:03 2006 Subject: [icfp-discuss] Re: I get an illegal instruction during LOADING: Message-ID: <200607221608.k6MG8t4E021590@gemini.newtonlabs.com> I had the exact same problem as Ralph and Shawn. Let me just give a hint: look very closely at the UM instruction that results in printing the diagnostic message "allocated array 7 of size 7"; once I did that, it all became clear and my program was working soon after. Carl From gus-icfp at guxx.net Sat Jul 22 12:52:43 2006 From: gus-icfp at guxx.net (gus) Date: Sat Jul 22 12:53:56 2006 Subject: [icfp-discuss] Scoreboard with ranking position Message-ID: <44C257DB.9010304@guxx.net> Hi, it would be nice if the team's current position will be shown in front of each line in the scoreboard. So everyone can track easily climbing up the ranking ... ;-) Guido From gus-icfp at guxx.net Sat Jul 22 12:57:07 2006 From: gus-icfp at guxx.net (gus) Date: Sat Jul 22 12:58:13 2006 Subject: [icfp-discuss] Logout and administrative privileges Message-ID: <44C258E3.2030506@guxx.net> Hi, is it correct that the "logout" command in UMIX will lead to a HALT? I'm wondering how to use the administrative privileges when you can't logout and login again. After the HALT I have to restart UMIX and so the privileges are gone... Guido From crary at cs.cmu.edu Sat Jul 22 13:06:10 2006 From: crary at cs.cmu.edu (Karl Crary) Date: Sat Jul 22 13:07:29 2006 Subject: [icfp-discuss] Logout and administrative privileges In-Reply-To: <44C258E3.2030506@guxx.net> References: <44C258E3.2030506@guxx.net> Message-ID: <44C25B02.2010407@cs.cmu.edu> Yes, that is correct. However, account passwords are the same in all runs, so that should not pose a problem. Karl Crary (for the organizers) gus wrote: > Hi, > > is it correct that the "logout" command in UMIX will lead to a HALT? > > I'm wondering how to use the administrative privileges when you can't > logout and login again. After the HALT I have to restart UMIX and so > the privileges are gone... > > Guido > > > > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > From asynth at gmail.com Sat Jul 22 13:15:46 2006 From: asynth at gmail.com (James McCartney) Date: Sat Jul 22 13:16:41 2006 Subject: [icfp-discuss] Re: I get an illegal instruction during LOADING: In-Reply-To: <200607221608.k6MG8t4E021590@gemini.newtonlabs.com> References: <200607221608.k6MG8t4E021590@gemini.newtonlabs.com> Message-ID: OK, Thanks! I'm past that now. On 7/22/06, Carl R. Witty wrote: > > I had the exact same problem as Ralph and Shawn. Let me just give a > hint: look very closely at the UM instruction that results in printing > the diagnostic message "allocated array 7 of size 7"; once I did that, > it all became clear and my program was working soon after. > > Carl > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > -- --- james mccartney -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060722/610f4b15/attachment.html From jmk at amoebaville.net Sat Jul 22 13:18:47 2006 From: jmk at amoebaville.net (Justin Koser) Date: Sat Jul 22 13:20:11 2006 Subject: [icfp-discuss] Best quote EVAR! Message-ID: <44C25DF7.4050107@amoebaville.net> "Advice is like snow---the softer it falls, the longer it dwells upon, and the deeper it sinks into the mind." --Coleridge Thanks for the advice. :) -- Justin From azul at freaks-unidos.net Sat Jul 22 13:43:44 2006 From: azul at freaks-unidos.net (Alejandro Forero Cuervo) Date: Sat Jul 22 13:30:02 2006 Subject: [icfp-discuss] Scoreboard with ranking position In-Reply-To: <44C257DB.9010304@guxx.net> References: <44C257DB.9010304@guxx.net> Message-ID: <20060722174344.GS15374@bachue.com> > it would be nice if the team's current position will be shown in front > of each line in the scoreboard. So everyone can track easily climbing up > the ranking ... ;-) Hahah, I was about to make this very suggestion! Alejo, team FUN, which we're certainly having. :-) http://azul.freaks-unidos.net/ From gus-icfp at guxx.net Sat Jul 22 13:47:30 2006 From: gus-icfp at guxx.net (gus) Date: Sat Jul 22 13:48:32 2006 Subject: [icfp-discuss] Joining a team / looking for teammate Message-ID: <44C264B2.7000300@guxx.net> Hi, my teammate has to work and unfortunately cannot help me anymore. So I'm thinking about joining another team or searching for another programmer. I currently have a stable (seems so) UM which is not too slow (compiling the BASIC program in 2 seconds). The UM is a Windows program written in Borland Delphi 7. I'm experienced in Java, Groovy, Perl and Delphi and have knowledge in Linux and Windows system administration. We already have collected 325 points (team "Rhinos"), solved the INTRO and BASIC tasks. Anyone interested? Mail to gus-icfp@guxx.net Guido From icfp2006 at vcxz.co.uk Sat Jul 22 14:03:56 2006 From: icfp2006 at vcxz.co.uk (Jon Colverson) Date: Sat Jul 22 14:04:55 2006 Subject: [icfp-discuss] Virtual machine specification Message-ID: <44C2688C.70208@vcxz.co.uk> I have a question about this part of the specification: "Before this operator is discharged, the execution finger shall be advanced to the next platter, if any." Could someone explain what is meant by "if any". If the execution finger is on the last platter of the program array and executes the instruction there, should the finger stay where it is, or should it be advanced "off the end" of the array? Thanks for your help. -- Jon From crary at cs.cmu.edu Sat Jul 22 14:25:12 2006 From: crary at cs.cmu.edu (Karl Crary) Date: Sat Jul 22 14:25:12 2006 Subject: [icfp-discuss] Virtual machine specification In-Reply-To: <44C2688C.70208@vcxz.co.uk> References: <44C2688C.70208@vcxz.co.uk> Message-ID: <44C26D88.4090305@cs.cmu.edu> This is covered in the discussion of "impossible behaviors" at the end of the spec. The behavior in this case is undefined. Karl Crary Jon Colverson wrote: > I have a question about this part of the specification: > > "Before this operator is discharged, the execution finger shall be > advanced to the next platter, if any." > > Could someone explain what is meant by "if any". If the execution finger > is on the last platter of the program array and executes the instruction > there, should the finger stay where it is, or should it be advanced "off > the end" of the array? > > Thanks for your help. > From leenstrab at comcast.net Sat Jul 22 14:40:14 2006 From: leenstrab at comcast.net (Bruce Leenstra) Date: Sat Jul 22 14:40:51 2006 Subject: [icfp-discuss] Joining a team / looking for teammate Message-ID: <01e801c6adbe$45934a30$6501a8c0@had1.or.comcast.net> Gus, Have you found another teammate or joined a team yet? You're way ahead of me (team of one) I currently have a UM written as a Delphi 7 console app. I ran it and got this message "endian" which implies some of the operators were working. so I tried to swap the order of each 4 bytes. but now it tries to do a bunch of loads without any allocates and aborts. (Anyone know the proper way to fix this, or did I do it right and something else is busted?) Bruce in Oregon -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060722/1c4c47fa/attachment.html From rinon.programming at gmail.com Sat Jul 22 15:08:38 2006 From: rinon.programming at gmail.com (Stephen Crane) Date: Sat Jul 22 15:08:39 2006 Subject: [icfp-discuss] How much memory are you guys using? Message-ID: How much memory are your UM's taking up during runtime? Just curious, cause our UM is taking up waaay too much memory (>700MB). - Stephen Crane B.C. Tech -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060722/12c79473/attachment.html From icfp_06 at schlagsei.de Sat Jul 22 15:26:05 2006 From: icfp_06 at schlagsei.de (=?ISO-8859-1?Q?Daniel_Schr=F6ter?=) Date: Sat Jul 22 15:25:58 2006 Subject: [icfp-discuss] How much memory are you guys using? In-Reply-To: References: Message-ID: <44C27BCD.9080108@schlagsei.de> I am currently recoding my UM, because of too much memory use. Now trying to store the arrays in FileStreams :) Daniel > How much memory are your UM's taking up during runtime? Just curious, > cause our UM is taking up waaay too much memory (>700MB). > > - Stephen Crane > B.C. Tech > ------------------------------------------------------------------------ > > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > > ------------------------------------------------------------------------ > > No virus found in this incoming message. > Checked by AVG Free Edition. > Version: 7.1.394 / Virus Database: 268.10.3/395 - Release Date: 21.07.2006 > From popiel at wolfskeep.com Sat Jul 22 15:28:26 2006 From: popiel at wolfskeep.com (T. Alexander Popiel) Date: Sat Jul 22 15:28:49 2006 Subject: [icfp-discuss] Login fails In-Reply-To: <20060722191834.394892DDB5@cashew.wolfskeep.com> References: <54076.84.226.35.108.1153565606.squirrel@serioussam.selfip.org> <44C21373.10006@guxx.net> <20060722142606.6C1132DDB5@cashew.wolfskeep.com> <59698.84.226.35.108.1153579097.squirrel@serioussam.selfip.org> <20060722191834.394892DDB5@cashew.wolfskeep.com> Message-ID: <20060722192826.002232DDB5@cashew.wolfskeep.com> In message: <59698.84.226.35.108.1153579097.squirrel@serioussam.selfip.org> "Samuel Burri" writes: >But how did you achieve that speed. I guess my memory >allocation/deallocation is way too slow. But how do you improve this? >Any suggestions? Whats an easy way to go? (only intermittently checking email... too much fun with the puzzles!) I achieved that speed in two ways: 1. I don't allocate the asked-for amount of memory; instead, I round it up to 16 words, 256 words, 4K words, or a multiple of 64K words. Thus, for the thousands and thousands of small allocations, I'm using consistently sized bits of memory. 2. For each size of memory above, I keep a separate free list. Allocation consists of checking the relevant free list, grabbing the top element if it exists, otherwise allocating a new one. Free is similarly fast; just stick it on the front of the appropriate free list. With those two optimizations (and a very small amount of nasty code for flipping array 0 around), the VM becomes pleasantly quick (assuming that the rest of the executor is decently fast, but my initial profile with running umix showed 99% of the time in allocate_array, so I didn't bother with further optimizations). - Alex From papierfalter at yahoo.de Sat Jul 22 15:29:50 2006 From: papierfalter at yahoo.de (Wolfram Fischer) Date: Sat Jul 22 15:31:08 2006 Subject: [icfp-discuss] UM-DUMP / Checksums Message-ID: <44C27CAE.102@yahoo.de> Hi Everyone, I'm not sure wether this it is possible to verify the UM-DUMP with Checksums? My Dump is exactly 15901920 Bytes (after dumping it under Linux - with XP I got to much characters - a known linefeed / cr issue?), but my um craches. Is the um-dump dependant on our encryption-keys? Anyway, I added my checksums as attachement and would be very thankfull if anyone could verify them if possible. Thanks in advance, Wolf -------------- next part -------------- 9f03f54ff249d681a39ac3625befaec60bbcde0e ldump.umz -------------- next part -------------- 57b9e3f788913e70ce1ad3861e74e226 ldump.umz From popiel at wolfskeep.com Sat Jul 22 15:32:07 2006 From: popiel at wolfskeep.com (T. Alexander Popiel) Date: Sat Jul 22 15:32:22 2006 Subject: [icfp-discuss] How much memory are you guys using? In-Reply-To: References: Message-ID: <20060722193207.046D22DDB5@cashew.wolfskeep.com> In message: "Stephen Crane" writes: > >How much memory are your UM's taking up during runtime? Just curious, cause >our UM is taking up waaay too much memory (>700MB). In the middle of playing adventure, my UM is taking 584M... a bit painful, but with such a constricted instruction set, sort of to be expected. - Alex From finnw at sucs.org Sat Jul 22 15:33:54 2006 From: finnw at sucs.org (Finn Wilcox) Date: Sat Jul 22 15:34:17 2006 Subject: [icfp-discuss] How much memory are you guys using? In-Reply-To: References: Message-ID: On Sat, 22 Jul 2006, Stephen Crane wrote: > How much memory are your UM's taking up during runtime? Just curious, cause > our UM is taking up waaay too much memory (>700MB). Mine is taking about the same (using C malloc to allocate arrays.) I doubt its possible to get it much lower. I guess they just assume all "real" hackers have >= 1Gb installed. Unfortunately I don't and most programs slow to a crawl after a couple of minutes. Strangely the amount of memory used is pretty much constant whatever program I'm running. Perhaps the OS has its own allocation routines that are tuned to use that amount. Maybe it can be tweaked... From wei.hoo at gmail.com Sat Jul 22 15:42:52 2006 From: wei.hoo at gmail.com (Wei Hu) Date: Sat Jul 22 15:43:17 2006 Subject: [icfp-discuss] How much memory are you guys using? In-Reply-To: References: Message-ID: <71fd12e60607221242t5e2c6920x4a4d4ffd64dafab@mail.gmail.com> > Mine is taking about the same (using C malloc to allocate arrays.) > I doubt its possible to get it much lower. > I guess they just assume all "real" hackers have >= 1Gb installed. I suppose they are expecting some tricks of lazy evaluation? However I ended up using C++ because of its convenient bit operations. I really look forward to seeing others' solutions in functional languages.. Do they mean to show lambda calculus can simulate Turing machine efficiently? > Unfortunately I don't and most programs slow to a crawl after a couple of > minutes. > Strangely the amount of memory used is pretty much constant whatever program > I'm running. Perhaps the OS has its own allocation routines that are tuned > to use that amount. Maybe it can be tweaked... > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > From popiel at wolfskeep.com Sat Jul 22 15:50:48 2006 From: popiel at wolfskeep.com (T. Alexander Popiel) Date: Sat Jul 22 15:51:09 2006 Subject: [icfp-discuss] How much memory are you guys using? In-Reply-To: <71fd12e60607221242t5e2c6920x4a4d4ffd64dafab@mail.gmail.com> References: <71fd12e60607221242t5e2c6920x4a4d4ffd64dafab@mail.gmail.com> Message-ID: <20060722195048.DA5A62DDB5@cashew.wolfskeep.com> In message: <71fd12e60607221242t5e2c6920x4a4d4ffd64dafab@mail.gmail.com> "Wei Hu" writes: >> Mine is taking about the same (using C malloc to allocate arrays.) >> I doubt its possible to get it much lower. >> I guess they just assume all "real" hackers have >= 1Gb installed. > >I suppose they are expecting some tricks of lazy evaluation? >However I ended up using C++ because of its convenient bit operations. >I really look forward to seeing others' solutions in functional languages.. >Do they mean to show lambda calculus can simulate Turing machine efficiently? Actually, I suspect that the VM is just to allow us all to run their binaries with confidence that they aren't doing anying even remotely nasty to our systems, since the only persistant I/O is to the console. This way we don't have to trust the contest administrators with our system security. Plus, it's a good weeder problem. If you have trouble implementing a 14-instruction VM, then you're likely to be completely frustrated by the real challenges once the system is up and running... Anyway, the puzzles inside UMIX are certainly more of the functional flavor. - Alex From finnw at sucs.org Sat Jul 22 16:06:10 2006 From: finnw at sucs.org (Finn Wilcox) Date: Sat Jul 22 16:06:28 2006 Subject: [icfp-discuss] How much memory are you guys using? In-Reply-To: <20060722195048.DA5A62DDB5@cashew.wolfskeep.com> References: <71fd12e60607221242t5e2c6920x4a4d4ffd64dafab@mail.gmail.com> <20060722195048.DA5A62DDB5@cashew.wolfskeep.com> Message-ID: On Sat, 22 Jul 2006, T. Alexander Popiel wrote: > Actually, I suspect that the VM is just to allow us all to run their > binaries with confidence that they aren't doing anying even remotely > nasty to our systems, since the only persistant I/O is to the console. > This way we don't have to trust the contest administrators with our > system security. This is not strictly true, considering that: 1. Some VM's will presumably be written in C 2. Some of the C VM's will represent array names using actual pointers for speed 3. Those same VM's will probably not check that pointers are valid before modifying the arrays. From davidad at mit.edu Sat Jul 22 16:08:27 2006 From: davidad at mit.edu (David Dalrymple) Date: Sat Jul 22 16:08:45 2006 Subject: [icfp-discuss] How much memory are you guys using? In-Reply-To: References: <71fd12e60607221242t5e2c6920x4a4d4ffd64dafab@mail.gmail.com> <20060722195048.DA5A62DDB5@cashew.wolfskeep.com> Message-ID: <504b2780607221308q67b88934sa843c162daf32a74@mail.gmail.com> Yep, all three of those describe my implementation. However, if I were particularly concerned about system security, I could fix any one of those three. --David On 7/22/06, Finn Wilcox wrote: > On Sat, 22 Jul 2006, T. Alexander Popiel wrote: > > > Actually, I suspect that the VM is just to allow us all to run their > > binaries with confidence that they aren't doing anying even remotely > > nasty to our systems, since the only persistant I/O is to the console. > > This way we don't have to trust the contest administrators with our > > system security. > > This is not strictly true, considering that: > 1. Some VM's will presumably be written in C > 2. Some of the C VM's will represent array names using actual pointers for speed > 3. Those same VM's will probably not check that pointers are valid before modifying the arrays. > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > From contest.crackers at gmail.com Sat Jul 22 16:10:47 2006 From: contest.crackers at gmail.com (Contest Crackers) Date: Sat Jul 22 16:11:09 2006 Subject: [icfp-discuss] All we get is a menu to dump UM data In-Reply-To: <44C23FCD.4070804@cs.cmu.edu> References: <472ee7770607220531p2e975d33qa54420834c8859ce@mail.gmail.com> <44C21CD3.9010404@cs.cmu.edu> <472ee7770607220737t53a40116p6e22513ce7fdce3a@mail.gmail.com> <44C23FCD.4070804@cs.cmu.edu> Message-ID: <472ee7770607221310x785ab53dva9d8886da53cb1ae@mail.gmail.com> Thanks! Even our size was different. Changing from printf("%c") or putchar, to fputc did the trick (we are using C/C++). Btw, the cksum of the program that gets dumped is 1309045862 15901894. Thanks once again. - "Contest Crackers". On 7/22/06, Daniel Spoonhower wrote: > Crackers, > > The size of the dumped UM data should be 15901920 bytes. (This is all the > bytes after the colon ":".) Did you accidentally add a character when you > removed the prefix? Some editors will add an extra new line at end of the file. > > --djs > > Contest Crackers wrote: > > Oh okay, thanks! > > > > The UM data however seems to be of a size in bytes that is not > > divisible by 4. (We have picked all the bytes after but not including > > the colon ":". Given that the platters all contain 32 bits, is this > > an error? Are we missing something? > > > > (Yeah, when we run it, it seeks input without having displayed > > anything. Still trying to figure out what input to provide). > > > > Thanks. > > > > - "Contest Crackers". > > > > On 7/22/06, Tom Murphy wrote: > >> Looks like you're on the right track! > >> We suggest saving the data to a file, having a look at it, and then > >> doing whatever you tend to do with mysterious UM data from the ICFP > >> organizers... > >> > >> Tom > >> > >> Contest Crackers wrote: > >> > Hi, > >> > > >> > After providing our decryption key, all we get (after waiting for > >> > about 5 minutes) is this. > >> > > >> > decrypting... > >> > ok > >> > LOADING: > >> > 9876543210 > >> > == CBV ARCHIVE == > >> > VOLUME ID 7 > >> > > >> > Choose a command: > >> > > >> > p) dump UM data > >> > x) exit > >> > > >> > ? > >> > INPUT: > >> > > >> > For the dump option (p) there is huge garbled dump. > >> > > >> > So this would be a bug in our UM? Has anyone else seen this? We tried > >> > tweaking some of our "strategies", it doesn't seem to matter. > >> > > >> > - "Contest Crackers" > >> > _______________________________________________ > >> > icfpcontest-discuss mailing list > >> > icfpcontest-discuss@lists.andrew.cmu.edu > >> > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > >> > > >> _______________________________________________ > >> icfpcontest-discuss mailing list > >> icfpcontest-discuss@lists.andrew.cmu.edu > >> https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > >> > > _______________________________________________ > > icfpcontest-discuss mailing list > > icfpcontest-discuss@lists.andrew.cmu.edu > > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > > > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > From popiel at wolfskeep.com Sat Jul 22 16:51:13 2006 From: popiel at wolfskeep.com (T. Alexander Popiel) Date: Sat Jul 22 16:51:34 2006 Subject: [icfp-discuss] How much memory are you guys using? In-Reply-To: References: <71fd12e60607221242t5e2c6920x4a4d4ffd64dafab@mail.gmail.com> <20060722195048.DA5A62DDB5@cashew.wolfskeep.com> Message-ID: <20060722205113.AF98C2DDB5@cashew.wolfskeep.com> In message: Finn Wilcox writes: >On Sat, 22 Jul 2006, T. Alexander Popiel wrote: > >> Actually, I suspect that the VM is just to allow us all to run their >> binaries with confidence that they aren't doing anying even remotely >> nasty to our systems, since the only persistant I/O is to the console. >> This way we don't have to trust the contest administrators with our >> system security. > >This is not strictly true, considering that: >1. Some VM's will presumably be written in C >2. Some of the C VM's will represent array names using actual pointers >for speed >3. Those same VM's will probably not check that pointers are valid >before modifying the arrays. Just because the VM _allows_ us to run their binaries with confidence doesn't mean that all people will take advantage of it. Honestly, though, bounds checking on the arrays is pretty cheap; in my case (since I'm using sequentially numbered arrays instead of pointers), it's an extra 2 compares per access (is the array index in bounds, and is the offset within that array within bounds). - Alex From bluelive at xs4all.nl Sat Jul 22 17:08:45 2006 From: bluelive at xs4all.nl (Bart van der Werf) Date: Sat Jul 22 17:09:04 2006 Subject: [icfp-discuss] How much memory are you guys using? In-Reply-To: <504b2780607221308q67b88934sa843c162daf32a74@mail.gmail.com> Message-ID: <000001c6add3$05131840$4101a8c0@andromeda> On 7/22/06, Finn Wilcox wrote: > On Sat, 22 Jul 2006, T. Alexander Popiel wrote: > > This is not strictly true, considering that: > 1. Some VM's will presumably be written in C > 2. Some of the C VM's will represent array names using actual pointers > for speed 3. Those same VM's will probably not check that pointers are > valid before modifying the arrays. I wrote mine in Delphi and have none fo these issues while till pumping nearly 50Mops/sec. Only using 32Meg of mem while doing the Black puzzle. Grt, Bart From iicfp at yahoo.com Fri Jul 21 16:43:27 2006 From: iicfp at yahoo.com (icfp icfp) Date: Sat Jul 22 17:09:56 2006 Subject: [icfp-discuss] Unable to register Message-ID: <20060721204328.74886.qmail@web55701.mail.re3.yahoo.com> I have been trying to register for the past hour for the contest, team name iicfp, this email. I receive a success web page, but then do not receive the promised email with the encryption key. This email address works fine. This is handicapping me, since the simulator is done and has passed the self-test, and I cannot proceed. Anyone there who can fix this? --------------------------------- Yahoo! Music Unlimited - Access over 1 million songs.Try it free. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060721/7270994b/attachment.html From andy_zyx at yahoo.com Sat Jul 22 18:30:49 2006 From: andy_zyx at yahoo.com (Andy) Date: Sat Jul 22 18:31:32 2006 Subject: [icfp-discuss] First instruction fails ! Message-ID: <20060722223049.8056.qmail@web50714.mail.yahoo.com> Hi, My first operator is always 15 and hence unable to run!. Did anyboy too get this initially ?. I think this is a FAIL status according to the desc. am using Java but I have spent enough making sure am handling the unsigned byte and in order A,B,C and D besides the ops and operands ? Any ideas why this could be happening ?. Thanks, Andy __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com From flicken-icfpcontest at flicken.net Sat Jul 22 18:43:42 2006 From: flicken-icfpcontest at flicken.net (flicken-icfpcontest@flicken.net) Date: Sat Jul 22 18:45:07 2006 Subject: [icfp-discuss] First instruction fails ! In-Reply-To: <20060722223049.8056.qmail@web50714.mail.yahoo.com> References: <20060722223049.8056.qmail@web50714.mail.yahoo.com> Message-ID: > My first operator is always 15 and hence unable to run!. Did anyboy too get this initially ?. > I think this is a FAIL status according to the desc. > am using Java but I have spent enough making sure am handling the unsigned byte and in order A,B,C > and D besides the ops and operands ? > > Any ideas why this could be happening ?. Re-check that you're doing signs correctly (Java is *very* tricky with this). I ran into the same difficulty. The first operator is a valid operator. - flicken From rvandam00 at gmail.com Sat Jul 22 19:17:16 2006 From: rvandam00 at gmail.com (Robert Van Dam) Date: Sat Jul 22 19:18:59 2006 Subject: [icfp-discuss] wrong key Message-ID: <1815f00607221617l2cb0bea5r9fa66dd81764901b@mail.gmail.com> Our VM seems to be working with the exception of getting the input in. I can enter all of the characters from the key just fine but when I'm done it says 'wrong key'. I'm pretty certain that the problem lies with the quirkiness of the input mechanism in the language we're using. I've seen talk on the discussion list already about newlines and I've tried several things but to no avail. One thing that I'm uncertain about: if the key is 30 characters long, should I input 30 or 31 bytes (ie w/ or w/o newline)? If it's 30 then it almost seems as if the 30th character is ignored because the register C is "endowed with a uniform value pattern where every place is pregnant with the 1 bit" at that point. Is any of this on track? Did someone else have the same problem and discover that it was an error in their VM? Rob From contest.crackers at gmail.com Sat Jul 22 19:22:43 2006 From: contest.crackers at gmail.com (Contest Crackers) Date: Sat Jul 22 19:24:17 2006 Subject: [icfp-discuss] wrong key In-Reply-To: <1815f00607221617l2cb0bea5r9fa66dd81764901b@mail.gmail.com> References: <1815f00607221617l2cb0bea5r9fa66dd81764901b@mail.gmail.com> Message-ID: <472ee7770607221622n1d2b44b2g85cf1891b9ae356f@mail.gmail.com> You need to input the key as is followed by the "end of input" character. This is Ctrl+D i.e. ascii 4. Hope this helps. On 7/22/06, Robert Van Dam wrote: > Our VM seems to be working with the exception of getting the input in. > I can enter all of the characters from the key just fine but when I'm > done it says 'wrong key'. I'm pretty certain that the problem lies > with the quirkiness of the input mechanism in the language we're > using. I've seen talk on the discussion list already about newlines > and I've tried several things but to no avail. One thing that I'm > uncertain about: if the key is 30 characters long, should I input 30 > or 31 bytes (ie w/ or w/o newline)? If it's 30 then it almost seems > as if the 30th character is ignored because the register C is "endowed > with a uniform value pattern where every place is pregnant with the 1 > bit" at that point. Is any of this on track? Did someone else have > the same problem and discover that it was an error in their VM? > > Rob > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > From rvandam00 at gmail.com Sat Jul 22 19:24:47 2006 From: rvandam00 at gmail.com (Robert Van Dam) Date: Sat Jul 22 19:26:04 2006 Subject: [icfp-discuss] First instruction fails ! In-Reply-To: <20060722223049.8056.qmail@web50714.mail.yahoo.com> References: <20060722223049.8056.qmail@web50714.mail.yahoo.com> Message-ID: <1815f00607221624o6c551fc1i20151e6c926bb8d7@mail.gmail.com> Well, since 15 is '1111' in binary I went and took a look at the codex in a hex editor and that string of 4 1's doesn't occur until the 25 byte so this is probably a problem unique to your code. Rob On 7/22/06, Andy wrote: > Hi, > > My first operator is always 15 and hence unable to run!. Did anyboy too get this initially ?. > I think this is a FAIL status according to the desc. > am using Java but I have spent enough making sure am handling the unsigned byte and in order A,B,C > and D besides the ops and operands ? > > Any ideas why this could be happening ?. > > Thanks, > Andy > > > > __________________________________________________ > Do You Yahoo!? > Tired of spam? Yahoo! Mail has the best spam protection around > http://mail.yahoo.com > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > From franke.daniel at gmail.com Sat Jul 22 19:25:27 2006 From: franke.daniel at gmail.com (Daniel Franke) Date: Sat Jul 22 19:26:55 2006 Subject: [icfp-discuss] wrong key In-Reply-To: <1815f00607221617l2cb0bea5r9fa66dd81764901b@mail.gmail.com> References: <1815f00607221617l2cb0bea5r9fa66dd81764901b@mail.gmail.com> Message-ID: <200607230125.27604.franke.daniel@gmail.com> On Sunday 23 July 2006 01:17, Robert Van Dam wrote: > Did someone else have > the same problem and discover that it was an error in their VM? I had. My problem was NotAnd. I also found, that the decoder will read a predefined number of bytes without checking for the "pregnant" byte. At least, I got away without it (I didn't at the login, but that's a different story). For me, "wrong key" meant: go and fix your arithmetics ... Good luck Daniel From popiel at wolfskeep.com Sat Jul 22 20:01:33 2006 From: popiel at wolfskeep.com (T. Alexander Popiel) Date: Sat Jul 22 20:01:35 2006 Subject: [icfp-discuss] wrong key In-Reply-To: <1815f00607221617l2cb0bea5r9fa66dd81764901b@mail.gmail.com> References: <1815f00607221617l2cb0bea5r9fa66dd81764901b@mail.gmail.com> Message-ID: <20060723000133.624C32DDB5@cashew.wolfskeep.com> In message: <1815f00607221617l2cb0bea5r9fa66dd81764901b@mail.gmail.com> "Robert Van Dam" writes: >Our VM seems to be working with the exception of getting the input in. > I can enter all of the characters from the key just fine but when I'm >done it says 'wrong key'. I'm pretty certain that the problem lies >with the quirkiness of the input mechanism in the language we're >using. I've seen talk on the discussion list already about newlines >and I've tried several things but to no avail. One thing that I'm >uncertain about: if the key is 30 characters long, should I input 30 >or 31 bytes (ie w/ or w/o newline)? If it's 30 then it almost seems >as if the 30th character is ignored because the register C is "endowed >with a uniform value pattern where every place is pregnant with the 1 >bit" at that point. Is any of this on track? Did someone else have >the same problem and discover that it was an error in their VM? More likely is that you're having some math errors with regard to integer overflow, and thus the decryption routines are failing. - Alex From rvandam00 at gmail.com Sat Jul 22 20:04:51 2006 From: rvandam00 at gmail.com (Robert Van Dam) Date: Sat Jul 22 20:04:54 2006 Subject: [icfp-discuss] wrong key In-Reply-To: <20060723000133.624C32DDB5@cashew.wolfskeep.com> References: <1815f00607221617l2cb0bea5r9fa66dd81764901b@mail.gmail.com> <20060723000133.624C32DDB5@cashew.wolfskeep.com> Message-ID: <1815f00607221704u3d08a149i76248edf268e0f5f@mail.gmail.com> Yes, thank you all for pointing me in that direction. Forgot to put parentheses before taking the modulo. :( On 7/22/06, T. Alexander Popiel wrote: > In message: <1815f00607221617l2cb0bea5r9fa66dd81764901b@mail.gmail.com> > "Robert Van Dam" writes: > >Our VM seems to be working with the exception of getting the input in. > > I can enter all of the characters from the key just fine but when I'm > >done it says 'wrong key'. I'm pretty certain that the problem lies > >with the quirkiness of the input mechanism in the language we're > >using. I've seen talk on the discussion list already about newlines > >and I've tried several things but to no avail. One thing that I'm > >uncertain about: if the key is 30 characters long, should I input 30 > >or 31 bytes (ie w/ or w/o newline)? If it's 30 then it almost seems > >as if the 30th character is ignored because the register C is "endowed > >with a uniform value pattern where every place is pregnant with the 1 > >bit" at that point. Is any of this on track? Did someone else have > >the same problem and discover that it was an error in their VM? > > More likely is that you're having some math errors with regard to > integer overflow, and thus the decryption routines are failing. > > - Alex > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > From iicfp at yahoo.com Sat Jul 22 18:42:13 2006 From: iicfp at yahoo.com (icfp icfp) Date: Sat Jul 22 21:13:26 2006 Subject: [icfp-discuss] Valid but useless dumped program Message-ID: <20060722224213.44326.qmail@web55703.mail.re3.yahoo.com> I seem to get a useless program as output from the first stage. The first instruction in the dumped program, starting after the colon, is: 11000011100100000000000000010000 (== reload and start from the beginning) which has the effect of just executing itself over and over in an infinite loop. Has anyone else run into this kind of issue? __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060722/bea162f7/attachment-0001.html From alexr at caltech.edu Sat Jul 22 21:37:49 2006 From: alexr at caltech.edu (Alex Roper) Date: Sat Jul 22 22:05:33 2006 Subject: [icfp-discuss] Solution to invalid dump Message-ID: <923b01ff0607221837p32b3efc8k1879ff77bc22dd37@mail.gmail.com> So I had a problem with invalid dump. I dumped the program under Linux, scp-ed it (for reasons I don't want to get into because it's embaressing) to Windows, did my hex editing (to eliminate the "Program follows colon" and such), then scp-ed it back. WinSCP treated it as a text file, and added in a bunch of gibberish (I think every [10] byte becomes [13][10]) You may run into this issue dumping it under windows, you may not. Be sure you're using binary output (ios::binary if you're using C++) to ensure most likely success. At any rate, as several messages have stated (Including one from the sponsors I believe), the correct size of the dump is 15,901,920 bytes. Good luck all, Alex From windenntw at gmail.com Sat Jul 22 21:16:04 2006 From: windenntw at gmail.com (Antonio Vargas) Date: Sat Jul 22 22:06:35 2006 Subject: [icfp-discuss] my progress so far... Message-ID: <69304d110607221816s6bbf2040nff1703f2e01bf5f@mail.gmail.com> After some hours chasing silly bugs in my UM implementation: self-check succeeded! enter decryption key: decrypting... ok LOADING: 98Segmentation fault $ it looks like it's running very fast (C implementation, all code in main function, ppc g4@1.25), but anyways it crashes between 8 and 7 (i suppose it's a decrypt/unpack loop, isn't it?) anyways off to bed, its 3.14am here and very tired... ps. i wonder how fast could this go via dynamic binary translation + COW for LOAD instruction using mmap+mprotect -- Greetz, Antonio Vargas aka winden of network http://network.amigascne.org/ windNOenSPAMntw@gmail.com thesameasabove@amigascne.org Every day, every year you have to work you have to study you have to scene. From andy_zyx at yahoo.com Sat Jul 22 22:29:30 2006 From: andy_zyx at yahoo.com (Andy) Date: Sat Jul 22 22:29:31 2006 Subject: [icfp-discuss] my progress so far... In-Reply-To: <69304d110607221816s6bbf2040nff1703f2e01bf5f@mail.gmail.com> Message-ID: <20060723022930.66898.qmail@web50706.mail.yahoo.com> Great.... I just completed my UM and its been running for the last 20 minutes. No news so far ! How long did it take you guys to hear anything from the run for the first time ? [I hope to see atleast "self-check succeeded" in the next 30 minutes ]... :( Andy --- Antonio Vargas wrote: > After some hours chasing silly bugs in my UM implementation: > > self-check succeeded! > enter decryption key: > decrypting... > ok > LOADING: 98Segmentation fault > $ > > it looks like it's running very fast (C implementation, all code in > main function, ppc g4@1.25), but anyways it crashes between 8 and 7 (i > suppose it's a decrypt/unpack loop, isn't it?) > > anyways off to bed, its 3.14am here and very tired... > > ps. i wonder how fast could this go via dynamic binary translation + > COW for LOAD instruction using mmap+mprotect > > -- > Greetz, Antonio Vargas aka winden of network > > http://network.amigascne.org/ > windNOenSPAMntw@gmail.com > thesameasabove@amigascne.org > > Every day, every year > you have to work > you have to study > you have to scene. > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com From wei.hoo at gmail.com Sat Jul 22 22:32:26 2006 From: wei.hoo at gmail.com (Wei Hu) Date: Sat Jul 22 22:32:28 2006 Subject: [icfp-discuss] my progress so far... In-Reply-To: <20060723022930.66898.qmail@web50706.mail.yahoo.com> References: <69304d110607221816s6bbf2040nff1703f2e01bf5f@mail.gmail.com> <20060723022930.66898.qmail@web50706.mail.yahoo.com> Message-ID: <71fd12e60607221932s570dc3b7xb845ae7dbc5022b5@mail.gmail.com> What language were you using? It shouldn't be that slow.. On 7/22/06, Andy wrote: > > Great.... > I just completed my UM and its been running for the last 20 minutes. No news so far ! > How long did it take you guys to hear anything from the run for the first time ? [I hope to see > atleast "self-check succeeded" in the next 30 minutes ]... :( > > Andy > > > > --- Antonio Vargas wrote: > > > After some hours chasing silly bugs in my UM implementation: > > > > self-check succeeded! > > enter decryption key: > > decrypting... > > ok > > LOADING: 98Segmentation fault > > $ > > > > it looks like it's running very fast (C implementation, all code in > > main function, ppc g4@1.25), but anyways it crashes between 8 and 7 (i > > suppose it's a decrypt/unpack loop, isn't it?) > > > > anyways off to bed, its 3.14am here and very tired... > > > > ps. i wonder how fast could this go via dynamic binary translation + > > COW for LOAD instruction using mmap+mprotect > > > > -- > > Greetz, Antonio Vargas aka winden of network > > > > http://network.amigascne.org/ > > windNOenSPAMntw@gmail.com > > thesameasabove@amigascne.org > > > > Every day, every year > > you have to work > > you have to study > > you have to scene. > > _______________________________________________ > > icfpcontest-discuss mailing list > > icfpcontest-discuss@lists.andrew.cmu.edu > > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > > > > > __________________________________________________ > Do You Yahoo!? > Tired of spam? Yahoo! Mail has the best spam protection around > http://mail.yahoo.com > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > From gregory.t.brown at gmail.com Sat Jul 22 22:38:40 2006 From: gregory.t.brown at gmail.com (Gregory Brown) Date: Sat Jul 22 22:38:42 2006 Subject: [icfp-discuss] my progress so far... In-Reply-To: <20060723022930.66898.qmail@web50706.mail.yahoo.com> References: <69304d110607221816s6bbf2040nff1703f2e01bf5f@mail.gmail.com> <20060723022930.66898.qmail@web50706.mail.yahoo.com> Message-ID: On 7/22/06, Andy wrote: > > Great.... > I just completed my UM and its been running for the last 20 minutes. No news so far ! > How long did it take you guys to hear anything from the run for the first time ? [I hope to see > atleast "self-check succeeded" in the next 30 minutes ]... :( Our Ruby implementation completed the dump in just under 2 hours (running on YARV). This probably is nowhere near fast enough for the 'real' challenges, so we're working on a C rewrite, but if it helps, even in Ruby, we got past the self check in a couple of seconds. From aguilar.james at gmail.com Sat Jul 22 22:39:02 2006 From: aguilar.james at gmail.com (James Aguilar) Date: Sat Jul 22 22:39:04 2006 Subject: [icfp-discuss] Trying to get sandmark working Message-ID: Gentlemen, I'm having a bit of trouble fixing my UM for sandmark. I've been hitting my head against it, and as a team of one, I'm probably just missing something obvious. For this reason, I'm enlisting you guys' help, if you're willing. My program halts with "Index in 0-array fail." My log is at the bottom of this mail. I believe that the LOADPROG is working, but something is wrong with ARRAY_IND. With all the NOT_AND_s, register 3 should be 0 by the end, but it ends up being 647168. But I've verified that the array load (appears to be) correct. The ORTHGRPY instruction loads 1414 into register 0, then the array is read at that location, giving 654321. This corresponds to the hex I see in the file (00 09 fb f1). Any advice? Pertinent sections of the log: COND_MOV if 0 r5 = 371 ========1100 0000000000000000000 000 110 101 LOADPROG arrays(0)[416] ========1101 0000000000000000010 110 000 110 ORTHGRPY r0 = 1414 ========0001 0000000000000000000 001 110 000 ARRY_IND r1 = arrays(0)[1414] (654321) ========1101 0100000010011111101 111 110 001 ORTHGRPY r2 = 7153 ========0110 0000000000000000000 100 001 001 NOT_AND_ r4 = (654321 ~& 654321) NOT_AND_ r1 = 00000000000010011111101111110001 NOT_AND_ r1 = 00000000000010011111101111110001 NOT_AND_ r4 = 11111111111101100000010000001110 ========0110 0000000000000000000 011 010 010 NOT_AND_ r3 = (7153 ~& 7153) NOT_AND_ r2 = 00000000000000000001101111110001 NOT_AND_ r2 = 00000000000000000001101111110001 NOT_AND_ r3 = 11111111111111111110010000001110 ========0110 0000000000000000000 011 100 011 NOT_AND_ r3 = (4294312974 ~& 4294960142) NOT_AND_ r4 = 11111111111101100000010000001110 NOT_AND_ r3 = 11111111111111111110010000001110 NOT_AND_ r3 = 00000000000010011111101111110001 ========0110 0000000000000000000 100 001 010 NOT_AND_ r4 = (654321 ~& 7153) NOT_AND_ r1 = 00000000000010011111101111110001 NOT_AND_ r2 = 00000000000000000001101111110001 NOT_AND_ r4 = 11111111111111111110010000001110 ========0110 0000000000000000000 011 100 011 NOT_AND_ r3 = (4294960142 ~& 654321) NOT_AND_ r4 = 11111111111111111110010000001110 NOT_AND_ r3 = 00000000000010011111101111110001 NOT_AND_ r3 = 11111111111101100001111111111111 ========0110 0000000000000000000 011 011 011 NOT_AND_ r3 = (4294320127 ~& 4294320127) NOT_AND_ r3 = 11111111111101100001111111111111 NOT_AND_ r3 = 11111111111101100001111111111111 NOT_AND_ r3 = 00000000000010011110000000000000 ========1101 1000000000000000100 010 011 000 ORTHGRPY r4 = 2200 ========1101 1010000000000000000 110 101 101 ORTHGRPY r5 = 429 ========0000 0000000000000000000 101 100 011 COND_MOV if 647168 r5 = 2200 ========1100 0000000000000000000 000 110 101 LOADPROG arrays(0)[2200] -- [?] James Aguilar [@] 18100 NE 95th St. #RR3088 / Redmond, WA 98052 [#] 314 494 0450 [!] In the wind and the rain, my darling, say goodbye -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060722/96eaaaab/attachment.html From aguilar.james at gmail.com Sat Jul 22 22:40:06 2006 From: aguilar.james at gmail.com (James Aguilar) Date: Sat Jul 22 22:40:09 2006 Subject: [icfp-discuss] my progress so far... In-Reply-To: <20060723022930.66898.qmail@web50706.mail.yahoo.com> References: <69304d110607221816s6bbf2040nff1703f2e01bf5f@mail.gmail.com> <20060723022930.66898.qmail@web50706.mail.yahoo.com> Message-ID: Andy, It takes mine about a fifth of a second to finish the self-check in Python, which is slow^maxx. -- [?] James Aguilar [@] 18100 NE 95th St. #RR3088 / Redmond, WA 98052 [#] 314 494 0450 [!] In the wind and the rain, my darling, say goodbye -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060722/47f258c5/attachment-0001.html From andy_zyx at yahoo.com Sat Jul 22 22:41:56 2006 From: andy_zyx at yahoo.com (Andy) Date: Sat Jul 22 22:41:57 2006 Subject: [icfp-discuss] First instruction fails ! In-Reply-To: Message-ID: <20060723024157.66377.qmail@web50701.mail.yahoo.com> Thanks flicken. I indeed missed out a one-off on shift. --- flicken-icfpcontest@flicken.net wrote: > > My first operator is always 15 and hence unable to run!. Did anyboy too get this initially ?. > > I think this is a FAIL status according to the desc. > > am using Java but I have spent enough making sure am handling the unsigned byte and in order > A,B,C > > and D besides the ops and operands ? > > > > Any ideas why this could be happening ?. > > Re-check that you're doing signs correctly (Java is *very* tricky with > this). I ran into the same difficulty. The first operator is a valid > operator. > > - flicken > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com From andy_zyx at yahoo.com Sat Jul 22 22:44:59 2006 From: andy_zyx at yahoo.com (Andy) Date: Sat Jul 22 22:45:00 2006 Subject: [icfp-discuss] my progress so far... In-Reply-To: <71fd12e60607221932s570dc3b7xb845ae7dbc5022b5@mail.gmail.com> Message-ID: <20060723024459.73121.qmail@web50706.mail.yahoo.com> Thanks guys. Am using java....[ I did have some memory out of error ] but going by what you all are saying...looks like I am way off to even get the "self-check" on time.... Its one hour already.....oh man ! Andy --- Wei Hu wrote: > What language were you using? > It shouldn't be that slow.. > > On 7/22/06, Andy wrote: > > > > Great.... > > I just completed my UM and its been running for the last 20 minutes. No news so far ! > > How long did it take you guys to hear anything from the run for the first time ? [I hope to > see > > atleast "self-check succeeded" in the next 30 minutes ]... :( > > > > Andy > > > > > > > > --- Antonio Vargas wrote: > > > > > After some hours chasing silly bugs in my UM implementation: > > > > > > self-check succeeded! > > > enter decryption key: > > > decrypting... > > > ok > > > LOADING: 98Segmentation fault > > > $ > > > > > > it looks like it's running very fast (C implementation, all code in > > > main function, ppc g4@1.25), but anyways it crashes between 8 and 7 (i > > > suppose it's a decrypt/unpack loop, isn't it?) > > > > > > anyways off to bed, its 3.14am here and very tired... > > > > > > ps. i wonder how fast could this go via dynamic binary translation + > > > COW for LOAD instruction using mmap+mprotect > > > > > > -- > > > Greetz, Antonio Vargas aka winden of network > > > > > > http://network.amigascne.org/ > > > windNOenSPAMntw@gmail.com > > > thesameasabove@amigascne.org > > > > > > Every day, every year > > > you have to work > > > you have to study > > > you have to scene. > > > _______________________________________________ > > > icfpcontest-discuss mailing list > > > icfpcontest-discuss@lists.andrew.cmu.edu > > > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > > > > > > > > > __________________________________________________ > > Do You Yahoo!? > > Tired of spam? Yahoo! Mail has the best spam protection around > > http://mail.yahoo.com > > _______________________________________________ > > icfpcontest-discuss mailing list > > icfpcontest-discuss@lists.andrew.cmu.edu > > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > > > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com From aguilar.james at gmail.com Sat Jul 22 23:15:53 2006 From: aguilar.james at gmail.com (James Aguilar) Date: Sat Jul 22 23:15:57 2006 Subject: [icfp-discuss] Re: Trying to get sandmark working In-Reply-To: References: Message-ID: On 7/22/06, James Aguilar wrote: > > ========1101 0100000010011111101 111 110 001 > ORTHGRPY r2 = 7153 > Guys, Nevermind. My problem is here. 7153 should be 654321. I'll fix it. -- [?] James Aguilar [@] 18100 NE 95th St. #RR3088 / Redmond, WA 98052 [#] 314 494 0450 [!] In the wind and the rain, my darling, say goodbye -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060722/ee7da7f1/attachment.html From aguilar.james at gmail.com Sat Jul 22 23:22:59 2006 From: aguilar.james at gmail.com (James Aguilar) Date: Sat Jul 22 23:23:01 2006 Subject: [icfp-discuss] Re: Trying to get sandmark working In-Reply-To: References: Message-ID: On 7/22/06, James Aguilar wrote: > > On 7/22/06, James Aguilar wrote: > > > > ========1101 0100000010011111101 111 110 001 > > ORTHGRPY r2 = 7153 > > > > Guys, > > Nevermind. My problem is here. 7153 should be 654321. I'll fix it. > I should stop spamming this list, but one last thing? What is the difference between 0x1FFF and 0x1FFFFFF? The right answer and the wrong answer, of course! -- [?] James Aguilar [@] 18100 NE 95th St. #RR3088 / Redmond, WA 98052 [#] 314 494 0450 [!] In the wind and the rain, my darling, say goodbye -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060722/ed037334/attachment.html From anand.chittu at gmail.com Sat Jul 22 23:55:25 2006 From: anand.chittu at gmail.com (Anand Chitipothu) Date: Sat Jul 22 23:55:24 2006 Subject: [icfp-discuss] All we get is a menu to dump UM data Message-ID: <474B33DE-4283-4B87-9D83-5160514871B7@gmail.com> I got the dump data and size is exactly 15901920 bytes as pointed in this list. But when i ran that program it doesnt do anything. When i enabled trace, i could see it is allocating and freeing many arrays. is there anything wrong with my program. thanks, anand From flyfree at gmail.com Sun Jul 23 00:00:33 2006 From: flyfree at gmail.com (Zhenzhong Xu) Date: Sun Jul 23 00:00:36 2006 Subject: [icfp-discuss] All we get is a menu to dump UM data In-Reply-To: <474B33DE-4283-4B87-9D83-5160514871B7@gmail.com> References: <474B33DE-4283-4B87-9D83-5160514871B7@gmail.com> Message-ID: <7b2bf4900607222100l146dd490qe07e3ba63a243df9@mail.gmail.com> looks like it's working :) might just be slow, at some point of my UM, it produces more than 1.8 millions of arrays in memory, which is pretty scary. Z. On 7/22/06, Anand Chitipothu wrote: > > I got the dump data and size is exactly 15901920 bytes as pointed in > this list. But when i ran that program it doesnt do anything. When i > enabled trace, i could see it is allocating and freeing many arrays. > > is there anything wrong with my program. > > thanks, > anand > > > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > From leenstrab at comcast.net Sun Jul 23 00:11:21 2006 From: leenstrab at comcast.net (Bruce Leenstra) Date: Sun Jul 23 00:11:24 2006 Subject: [icfp-discuss] Minimum memory requirements Message-ID: <024101c6ae0e$0e752050$6501a8c0@had1.or.comcast.net> What is the minimum system memory needed to run the decrypted dump? I have 1 GB and I'm getting an out_of_memory error: I am getting through the sandmark just fine. when I run the decrypted dump, i get the login:guest prompt, it prompts for password, (i suspect this is wrong) then I get an out_of_memory error. Is it my allocation/abandon routines, or am I actually running out of memory? Bruce -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060722/8e99c850/attachment.html From rvandam00 at gmail.com Sun Jul 23 00:12:44 2006 From: rvandam00 at gmail.com (Robert Van Dam) Date: Sun Jul 23 00:12:46 2006 Subject: [icfp-discuss] infinite LOAD loop Message-ID: <1815f00607222112t5a0fba0erf9c09cb64d45f6c0@mail.gmail.com> I get an infinite (or so it appears) LOAD loop on both the codex and the sandmark. Has anyone encountered the same and figured out what they did wrong? Or is it just a regular loop that I need to optimize? Rob From andy_zyx at yahoo.com Sun Jul 23 00:25:03 2006 From: andy_zyx at yahoo.com (Andy) Date: Sun Jul 23 00:25:05 2006 Subject: [icfp-discuss] my progress so far... In-Reply-To: Message-ID: <20060723042503.22217.qmail@web50707.mail.yahoo.com> I seem to be getting an infinite loop (which i figured could have been the reason for its ever-ending run). After the first few cycles, the UM seems to be going around on this sequence of ops: 0, 3 , 13, 13, 13, 13, 12 Any help ? Andy --- James Aguilar wrote: > Andy, > > It takes mine about a fifth of a second to finish the self-check in Python, > which is slow^maxx. > > -- > [?] James Aguilar > [@] 18100 NE 95th St. #RR3088 / Redmond, WA 98052 > [#] 314 494 0450 > [!] In the wind and the rain, my darling, say goodbye > > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com From anand.chittu at gmail.com Sun Jul 23 00:25:52 2006 From: anand.chittu at gmail.com (Anand Chitipothu) Date: Sun Jul 23 00:25:43 2006 Subject: [icfp-discuss] All we get is a menu to dump UM data In-Reply-To: <7b2bf4900607222100l146dd490qe07e3ba63a243df9@mail.gmail.com> References: <474B33DE-4283-4B87-9D83-5160514871B7@gmail.com> <7b2bf4900607222100l146dd490qe07e3ba63a243df9@mail.gmail.com> Message-ID: no. it just allocated 40576 arrays and halted. On 23-Jul-06, at 9:30 AM, Zhenzhong Xu wrote: > looks like it's working :) might just be slow, at some point of my UM, > it produces more than 1.8 millions of arrays in memory, which is > pretty scary. > > Z. > > On 7/22/06, Anand Chitipothu wrote: >> >> I got the dump data and size is exactly 15901920 bytes as pointed in >> this list. But when i ran that program it doesnt do anything. When i >> enabled trace, i could see it is allocating and freeing many arrays. >> >> is there anything wrong with my program. >> >> thanks, >> anand >> >> >> _______________________________________________ >> icfpcontest-discuss mailing list >> icfpcontest-discuss@lists.andrew.cmu.edu >> https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss >> > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss From jacobdp at gmail.com Sun Jul 23 00:26:57 2006 From: jacobdp at gmail.com (Jacob Potter) Date: Sun Jul 23 00:27:00 2006 Subject: [icfp-discuss] All we get is a menu to dump UM data In-Reply-To: References: <474B33DE-4283-4B87-9D83-5160514871B7@gmail.com> <7b2bf4900607222100l146dd490qe07e3ba63a243df9@mail.gmail.com> Message-ID: Something's wrong with your VM, then. Also consider optimizing the array allocation/free code, since it'll be getting called a lot. - Jacob On 7/23/06, Anand Chitipothu wrote: > > no. it just allocated 40576 arrays and halted. > > On 23-Jul-06, at 9:30 AM, Zhenzhong Xu wrote: > > > looks like it's working :) might just be slow, at some point of my UM, > > it produces more than 1.8 millions of arrays in memory, which is > > pretty scary. > > > > Z. > > > > On 7/22/06, Anand Chitipothu wrote: > >> > >> I got the dump data and size is exactly 15901920 bytes as pointed in > >> this list. But when i ran that program it doesnt do anything. When i > >> enabled trace, i could see it is allocating and freeing many arrays. > >> > >> is there anything wrong with my program. > >> > >> thanks, > >> anand > >> > >> > >> _______________________________________________ > >> icfpcontest-discuss mailing list > >> icfpcontest-discuss@lists.andrew.cmu.edu > >> https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > >> > > _______________________________________________ > > icfpcontest-discuss mailing list > > icfpcontest-discuss@lists.andrew.cmu.edu > > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > From davidad at mit.edu Sun Jul 23 00:28:39 2006 From: davidad at mit.edu (David Dalrymple) Date: Sun Jul 23 00:28:42 2006 Subject: [icfp-discuss] my progress so far... In-Reply-To: <20060723042503.22217.qmail@web50707.mail.yahoo.com> References: <20060723042503.22217.qmail@web50707.mail.yahoo.com> Message-ID: <504b2780607222128n122e5a5cle4336c9954be4da6@mail.gmail.com> Your implementation of op 12 is wrong - check where it's putting the finger... --David On 7/23/06, Andy wrote: > > I seem to be getting an infinite loop (which i figured could have been the reason for its > ever-ending run). After the first few cycles, the UM seems to be going around on this sequence of > ops: 0, 3 , 13, 13, 13, 13, 12 > > Any help ? > > Andy > > > --- James Aguilar wrote: > > > Andy, > > > > It takes mine about a fifth of a second to finish the self-check in Python, > > which is slow^maxx. > > > > -- > > [?] James Aguilar > > [@] 18100 NE 95th St. #RR3088 / Redmond, WA 98052 > > [#] 314 494 0450 > > [!] In the wind and the rain, my darling, say goodbye > > > _______________________________________________ > > icfpcontest-discuss mailing list > > icfpcontest-discuss@lists.andrew.cmu.edu > > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > > > > > __________________________________________________ > Do You Yahoo!? > Tired of spam? Yahoo! Mail has the best spam protection around > http://mail.yahoo.com > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > From aguilar.james at gmail.com Sun Jul 23 00:35:02 2006 From: aguilar.james at gmail.com (James Aguilar) Date: Sun Jul 23 00:35:07 2006 Subject: [icfp-discuss] infinite LOAD loop In-Reply-To: <1815f00607222112t5a0fba0erf9c09cb64d45f6c0@mail.gmail.com> References: <1815f00607222112t5a0fba0erf9c09cb64d45f6c0@mail.gmail.com> Message-ID: On 7/22/06, Robert Van Dam wrote: > > I get an infinite (or so it appears) LOAD loop on both the codex and > the sandmark. Has anyone encountered the same and figured out what > they did wrong? Or is it just a regular loop that I need to optimize? I see something similar, but the array ammendment stuff appears to be different every time, which leads me to believe that there's something going on there. -- [?] James Aguilar [@] 18100 NE 95th St. #RR3088 / Redmond, WA 98052 [#] 314 494 0450 [!] In the wind and the rain, my darling, say goodbye -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060722/880d214e/attachment.html From rvandam00 at gmail.com Sun Jul 23 00:45:12 2006 From: rvandam00 at gmail.com (Robert Van Dam) Date: Sun Jul 23 00:45:15 2006 Subject: [icfp-discuss] infinite LOAD loop In-Reply-To: References: <1815f00607222112t5a0fba0erf9c09cb64d45f6c0@mail.gmail.com> Message-ID: <1815f00607222145t6731e65k98c20781c2f2c6dc@mail.gmail.com> I realized it wasn't truly an infinite loop. I just had some crappy debugging stuff slowing it down so much that I figured it was stuck. Rob On 7/22/06, James Aguilar wrote: > On 7/22/06, Robert Van Dam wrote: > > > I get an infinite (or so it appears) LOAD loop on both the codex and > > the sandmark. Has anyone encountered the same and figured out what > > they did wrong? Or is it just a regular loop that I need to optimize? > > > I see something similar, but the array ammendment stuff appears to be > different every time, which leads me to believe that there's something going > on there. > > -- > [?] James Aguilar > [@] 18100 NE 95th St. #RR3088 / Redmond, WA 98052 > [#] 314 494 0450 > [!] In the wind and the rain, my darling, say goodbye > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > > > From aguilar.james at gmail.com Sun Jul 23 00:48:14 2006 From: aguilar.james at gmail.com (James Aguilar) Date: Sun Jul 23 00:48:17 2006 Subject: [icfp-discuss] infinite LOAD loop In-Reply-To: <1815f00607222145t6731e65k98c20781c2f2c6dc@mail.gmail.com> References: <1815f00607222112t5a0fba0erf9c09cb64d45f6c0@mail.gmail.com> <1815f00607222145t6731e65k98c20781c2f2c6dc@mail.gmail.com> Message-ID: On 7/22/06, Robert Van Dam wrote: > > I realized it wasn't truly an infinite loop. I just had some crappy > debugging stuff slowing it down so much that I figured it was stuck. Yes, mine is quite slow, I'm hoping that removing all of the logging and importing psycho will help a little, but I'm running low on optimism. Currently profiling to try and find bottlenecks before I go to the optimized versions. -- James -- [?] James Aguilar [@] 18100 NE 95th St. #RR3088 / Redmond, WA 98052 [#] 314 494 0450 [!] In the wind and the rain, my darling, say goodbye -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060722/3c67b340/attachment.html From davidad at mit.edu Sun Jul 23 01:10:57 2006 From: davidad at mit.edu (David Dalrymple) Date: Sun Jul 23 01:11:01 2006 Subject: [icfp-discuss] Who's got the fastest sandmark? Message-ID: <504b2780607222210m4b36741ak57ff53e3d6a7e3ca@mail.gmail.com> Just got mine to work - runs in 1m38s. Of course, that's on a 3Ghz Xeon server -- so maybe it would be fairer to do a ratio against SuperPI. On SuperPI calculating 2^20 digits on the server, I have 40s user time. So that's a calibrated SANDmark of 2.45. Can anyone beat that? --David From gus-icfp at guxx.net Sun Jul 23 01:39:23 2006 From: gus-icfp at guxx.net (gus) Date: Sun Jul 23 01:39:28 2006 Subject: [icfp-discuss] Scoreboard with ranking position In-Reply-To: <44C257DB.9010304@guxx.net> References: <44C257DB.9010304@guxx.net> Message-ID: <44C30B8B.3080405@guxx.net> gus schrieb: > it would be nice if the team's current position will be shown in front > of each line in the scoreboard. So everyone can track easily climbing up > the ranking ... ;-) Thanks for making my wish come true ;-) Guido From andy_zyx at yahoo.com Sun Jul 23 01:50:43 2006 From: andy_zyx at yahoo.com (Andy) Date: Sun Jul 23 01:50:46 2006 Subject: [icfp-discuss] my progress so far... In-Reply-To: <504b2780607222128n122e5a5cle4336c9954be4da6@mail.gmail.com> Message-ID: <20060723055043.52787.qmail@web50707.mail.yahoo.com> Corrected the op12...but now i get a "DIV error" message even though i neither have any div by zero situations nor overflow ... Andy --- David Dalrymple wrote: > Your implementation of op 12 is wrong - check where it's putting the finger... > > --David > > On 7/23/06, Andy wrote: > > > > I seem to be getting an infinite loop (which i figured could have been the reason for its > > ever-ending run). After the first few cycles, the UM seems to be going around on this sequence > of > > ops: 0, 3 , 13, 13, 13, 13, 12 > > > > Any help ? > > > > Andy > > > > > > --- James Aguilar wrote: > > > > > Andy, > > > > > > It takes mine about a fifth of a second to finish the self-check in Python, > > > which is slow^maxx. > > > > > > -- > > > [?] James Aguilar > > > [@] 18100 NE 95th St. #RR3088 / Redmond, WA 98052 > > > [#] 314 494 0450 > > > [!] In the wind and the rain, my darling, say goodbye > > > > _______________________________________________ > > > icfpcontest-discuss mailing list > > > icfpcontest-discuss@lists.andrew.cmu.edu > > > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > > > > > > > > > __________________________________________________ > > Do You Yahoo!? > > Tired of spam? Yahoo! Mail has the best spam protection around > > http://mail.yahoo.com > > _______________________________________________ > > icfpcontest-discuss mailing list > > icfpcontest-discuss@lists.andrew.cmu.edu > > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > > > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com From plakal at gmail.com Sun Jul 23 02:07:00 2006 From: plakal at gmail.com (Manoj Plakal) Date: Sun Jul 23 02:07:03 2006 Subject: [icfp-discuss] Who's got the fastest sandmark? In-Reply-To: <504b2780607222210m4b36741ak57ff53e3d6a7e3ca@mail.gmail.com> References: <504b2780607222210m4b36741ak57ff53e3d6a7e3ca@mail.gmail.com> Message-ID: <5230c2820607222307m46edfbadwdc84035b5a549578@mail.gmail.com> $ time ./dirty sandmark.umz > /dev/null gives me the following real elapsed times: - 1m 36 sec on a 2.8ghz pentium4, 3.5gb ram, linux 2.4.22 - 2m 22 sec on a mac mini core duo (1.66ghz dual-core pentium m), 2gb ram, macos x 10.4 dirty, aptly named, is our quick and dirty universal machine, written/debugged/operational in less than 2 hrs total wall clock time by two people in ~190 lines of grungy k&r c, with everything stuffed into main() plus some macros, no checking of any kind, compiled with gcc -O3/-fast. -manoj (team pipewrench) On 7/22/06, David Dalrymple wrote: > Just got mine to work - runs in 1m38s. > > Of course, that's on a 3Ghz Xeon server -- so maybe it would be fairer > to do a ratio against SuperPI. > > On SuperPI calculating 2^20 digits on the server, I have 40s user time. > > So that's a calibrated SANDmark of 2.45. Can anyone beat that? > > --David > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > From davidad at mit.edu Sun Jul 23 02:18:12 2006 From: davidad at mit.edu (David Dalrymple) Date: Sun Jul 23 02:18:15 2006 Subject: [icfp-discuss] Who's got the fastest sandmark? In-Reply-To: <5230c2820607222307m46edfbadwdc84035b5a549578@mail.gmail.com> References: <504b2780607222210m4b36741ak57ff53e3d6a7e3ca@mail.gmail.com> <5230c2820607222307m46edfbadwdc84035b5a549578@mail.gmail.com> Message-ID: <504b2780607222318x631b33bakc870e24b81c24850@mail.gmail.com> Heh, I didn't even think to use -O3. I now have 1m7s on my Xeon server. In your other departments I can't really compete - it took me more like 35 hours wall clock to do my UM, it's 271 lines of code, and I have some function calls in there. I have no checking of any kind either, though. --David On 7/23/06, Manoj Plakal wrote: > $ time ./dirty sandmark.umz > /dev/null > gives me the following real elapsed times: > - 1m 36 sec on a 2.8ghz pentium4, 3.5gb ram, linux 2.4.22 > - 2m 22 sec on a mac mini core duo (1.66ghz dual-core pentium m), 2gb > ram, macos x 10.4 > > dirty, aptly named, is our quick and dirty universal machine, > written/debugged/operational in less than 2 hrs total wall clock time > by two people in ~190 lines of grungy k&r c, with everything stuffed > into main() plus > some macros, no checking of any kind, compiled with gcc -O3/-fast. > > -manoj > (team pipewrench) > > > On 7/22/06, David Dalrymple wrote: > > Just got mine to work - runs in 1m38s. > > > > Of course, that's on a 3Ghz Xeon server -- so maybe it would be fairer > > to do a ratio against SuperPI. > > > > On SuperPI calculating 2^20 digits on the server, I have 40s user time. > > > > So that's a calibrated SANDmark of 2.45. Can anyone beat that? > > > > --David > > _______________________________________________ > > icfpcontest-discuss mailing list > > icfpcontest-discuss@lists.andrew.cmu.edu > > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > > > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > From rvandam00 at gmail.com Sun Jul 23 02:41:23 2006 From: rvandam00 at gmail.com (Robert Van Dam) Date: Sun Jul 23 02:41:27 2006 Subject: [icfp-discuss] Who's got the fastest sandmark? In-Reply-To: <504b2780607222318x631b33bakc870e24b81c24850@mail.gmail.com> References: <504b2780607222210m4b36741ak57ff53e3d6a7e3ca@mail.gmail.com> <5230c2820607222307m46edfbadwdc84035b5a549578@mail.gmail.com> <504b2780607222318x631b33bakc870e24b81c24850@mail.gmail.com> Message-ID: <1815f00607222341n1ba39a94s4055d21e7c88b3c0@mail.gmail.com> I'm quite curious to hear if anyone has gotten this to run within a reasonable amount of time using something other than C (perhaps even a functional language ;). Rob On 7/23/06, David Dalrymple wrote: > Heh, I didn't even think to use -O3. I now have 1m7s on my Xeon > server. In your other departments I can't really compete - it took me > more like 35 hours wall clock to do my UM, it's 271 lines of code, and > I have some function calls in there. I have no checking of any kind > either, though. > > --David > > On 7/23/06, Manoj Plakal wrote: > > $ time ./dirty sandmark.umz > /dev/null > > gives me the following real elapsed times: > > - 1m 36 sec on a 2.8ghz pentium4, 3.5gb ram, linux 2.4.22 > > - 2m 22 sec on a mac mini core duo (1.66ghz dual-core pentium m), 2gb > > ram, macos x 10.4 > > > > dirty, aptly named, is our quick and dirty universal machine, > > written/debugged/operational in less than 2 hrs total wall clock time > > by two people in ~190 lines of grungy k&r c, with everything stuffed > > into main() plus > > some macros, no checking of any kind, compiled with gcc -O3/-fast. > > > > -manoj > > (team pipewrench) > > > > > > On 7/22/06, David Dalrymple wrote: > > > Just got mine to work - runs in 1m38s. > > > > > > Of course, that's on a 3Ghz Xeon server -- so maybe it would be fairer > > > to do a ratio against SuperPI. > > > > > > On SuperPI calculating 2^20 digits on the server, I have 40s user time. > > > > > > So that's a calibrated SANDmark of 2.45. Can anyone beat that? > > > > > > --David > > > _______________________________________________ > > > icfpcontest-discuss mailing list > > > icfpcontest-discuss@lists.andrew.cmu.edu > > > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > > > > > _______________________________________________ > > icfpcontest-discuss mailing list > > icfpcontest-discuss@lists.andrew.cmu.edu > > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > > > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > From davidad at mit.edu Sun Jul 23 02:44:35 2006 From: davidad at mit.edu (David Dalrymple) Date: Sun Jul 23 02:44:38 2006 Subject: [icfp-discuss] Who's got the fastest sandmark? In-Reply-To: <1815f00607222341n1ba39a94s4055d21e7c88b3c0@mail.gmail.com> References: <504b2780607222210m4b36741ak57ff53e3d6a7e3ca@mail.gmail.com> <5230c2820607222307m46edfbadwdc84035b5a549578@mail.gmail.com> <504b2780607222318x631b33bakc870e24b81c24850@mail.gmail.com> <1815f00607222341n1ba39a94s4055d21e7c88b3c0@mail.gmail.com> Message-ID: <504b2780607222344m7438554bx2b5352094cdaef03@mail.gmail.com> I think that would be very unlikely. Other C-like languages might have a shot (Fortran, C++) but even they would be slower. ASM has a shot given unlimited time to code it :) I think OCaml is the best functional language for this but that would be too slow as well (that is just a guess- I haven't tried it). But if anyone can prove me wrong, I'd be interested... --David On 7/23/06, Robert Van Dam wrote: > I'm quite curious to hear if anyone has gotten this to run within a > reasonable amount of time using something other than C (perhaps even a > functional language ;). > > Rob > > On 7/23/06, David Dalrymple wrote: > > Heh, I didn't even think to use -O3. I now have 1m7s on my Xeon > > server. In your other departments I can't really compete - it took me > > more like 35 hours wall clock to do my UM, it's 271 lines of code, and > > I have some function calls in there. I have no checking of any kind > > either, though. > > > > --David > > > > On 7/23/06, Manoj Plakal wrote: > > > $ time ./dirty sandmark.umz > /dev/null > > > gives me the following real elapsed times: > > > - 1m 36 sec on a 2.8ghz pentium4, 3.5gb ram, linux 2.4.22 > > > - 2m 22 sec on a mac mini core duo (1.66ghz dual-core pentium m), 2gb > > > ram, macos x 10.4 > > > > > > dirty, aptly named, is our quick and dirty universal machine, > > > written/debugged/operational in less than 2 hrs total wall clock time > > > by two people in ~190 lines of grungy k&r c, with everything stuffed > > > into main() plus > > > some macros, no checking of any kind, compiled with gcc -O3/-fast. > > > > > > -manoj > > > (team pipewrench) > > > > > > > > > On 7/22/06, David Dalrymple wrote: > > > > Just got mine to work - runs in 1m38s. > > > > > > > > Of course, that's on a 3Ghz Xeon server -- so maybe it would be fairer > > > > to do a ratio against SuperPI. > > > > > > > > On SuperPI calculating 2^20 digits on the server, I have 40s user time. > > > > > > > > So that's a calibrated SANDmark of 2.45. Can anyone beat that? > > > > > > > > --David > > > > _______________________________________________ > > > > icfpcontest-discuss mailing list > > > > icfpcontest-discuss@lists.andrew.cmu.edu > > > > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > > > > > > > _______________________________________________ > > > icfpcontest-discuss mailing list > > > icfpcontest-discuss@lists.andrew.cmu.edu > > > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > > > > > _______________________________________________ > > icfpcontest-discuss mailing list > > icfpcontest-discuss@lists.andrew.cmu.edu > > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > > > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > From aguilar.james at gmail.com Sun Jul 23 02:51:47 2006 From: aguilar.james at gmail.com (James Aguilar) Date: Sun Jul 23 02:51:49 2006 Subject: [icfp-discuss] Who's got the fastest sandmark? In-Reply-To: <504b2780607222344m7438554bx2b5352094cdaef03@mail.gmail.com> References: <504b2780607222210m4b36741ak57ff53e3d6a7e3ca@mail.gmail.com> <5230c2820607222307m46edfbadwdc84035b5a549578@mail.gmail.com> <504b2780607222318x631b33bakc870e24b81c24850@mail.gmail.com> <1815f00607222341n1ba39a94s4055d21e7c88b3c0@mail.gmail.com> <504b2780607222344m7438554bx2b5352094cdaef03@mail.gmail.com> Message-ID: On 7/22/06, David Dalrymple wrote: > > I think that would be very unlikely. Other C-like languages might > have a shot (Fortran, C++) but even they would be slower. ASM has a > shot given unlimited time to code it :) I think OCaml is the best > functional language for this but that would be too slow as well (that > is just a guess- I haven't tried it). > > But if anyone can prove me wrong, I'd be interested... My python+psyco implementation would probably finish in about two hours. I let it get to test eighty before I stopped it. -- [?] James Aguilar [@] 18100 NE 95th St. #RR3088 / Redmond, WA 98052 [#] 314 494 0450 [!] In the wind and the rain, my darling, say goodbye -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060722/025b7bce/attachment-0001.html From colanderman at gmail.com Sun Jul 23 03:06:42 2006 From: colanderman at gmail.com (Chris King) Date: Sun Jul 23 03:06:45 2006 Subject: [icfp-discuss] Who's got the fastest sandmark? In-Reply-To: <504b2780607222344m7438554bx2b5352094cdaef03@mail.gmail.com> References: <504b2780607222210m4b36741ak57ff53e3d6a7e3ca@mail.gmail.com> <5230c2820607222307m46edfbadwdc84035b5a549578@mail.gmail.com> <504b2780607222318x631b33bakc870e24b81c24850@mail.gmail.com> <1815f00607222341n1ba39a94s4055d21e7c88b3c0@mail.gmail.com> <504b2780607222344m7438554bx2b5352094cdaef03@mail.gmail.com> Message-ID: <875c7e070607230006w45d4193cq5f4b7587a8a90fcd@mail.gmail.com> On 7/23/06, David Dalrymple wrote: > I think that would be very unlikely. Other C-like languages might > have a shot (Fortran, C++) but even they would be slower. ASM has a > shot given unlimited time to code it :) I think OCaml is the best > functional language for this but that would be too slow as well (that > is just a guess- I haven't tried it). My O'Caml implementation clocks about 10x slower than my C implementation, even with optimizations turned on. It suffers from massive indirection, since it uses library functions for the 32-bit integers and the arrays are unboxed. It also suffers from heavy memory thrashing due to all this overhead. My C implementation on the other hand gets 1m 59s on a 1.5GHz single-processor Mac Mini. - Chris King From anand.chittu at gmail.com Sun Jul 23 03:13:12 2006 From: anand.chittu at gmail.com (Anand Chitipothu) Date: Sun Jul 23 03:13:03 2006 Subject: [icfp-discuss] All we get is a menu to dump UM data In-Reply-To: References: <474B33DE-4283-4B87-9D83-5160514871B7@gmail.com> <7b2bf4900607222100l146dd490qe07e3ba63a243df9@mail.gmail.com> Message-ID: <4F7FA0AD-79DE-431D-BE93-C41BADA42E00@gmail.com> I am getting an error when running the UM dump. There is a LOADPROGRAM instruction, which gives the offset more than the array size. So my execution finger is going out of range. has anybody had the same problem? thanks anand On 23-Jul-06, at 9:56 AM, Jacob Potter wrote: > Something's wrong with your VM, then. Also consider optimizing the > array allocation/free code, since it'll be getting called a lot. > > - Jacob > > On 7/23/06, Anand Chitipothu wrote: >> >> no. it just allocated 40576 arrays and halted. >> >> On 23-Jul-06, at 9:30 AM, Zhenzhong Xu wrote: >> >> > looks like it's working :) might just be slow, at some point of >> my UM, >> > it produces more than 1.8 millions of arrays in memory, which is >> > pretty scary. >> > >> > Z. >> > >> > On 7/22/06, Anand Chitipothu wrote: >> >> >> >> I got the dump data and size is exactly 15901920 bytes as >> pointed in >> >> this list. But when i ran that program it doesnt do anything. >> When i >> >> enabled trace, i could see it is allocating and freeing many >> arrays. >> >> >> >> is there anything wrong with my program. >> >> >> >> thanks, >> >> anand >> >> >> >> >> >> _______________________________________________ >> >> icfpcontest-discuss mailing list >> >> icfpcontest-discuss@lists.andrew.cmu.edu >> >> https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss >> >> >> > _______________________________________________ >> > icfpcontest-discuss mailing list >> > icfpcontest-discuss@lists.andrew.cmu.edu >> > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss >> >> _______________________________________________ >> icfpcontest-discuss mailing list >> icfpcontest-discuss@lists.andrew.cmu.edu >> https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss >> > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss From blo.b at infonie.fr Sun Jul 23 03:54:13 2006 From: blo.b at infonie.fr (Laurent Vaucher) Date: Sun Jul 23 03:54:17 2006 Subject: [icfp-discuss] Who's got the fastest sandmark? In-Reply-To: <1815f00607222341n1ba39a94s4055d21e7c88b3c0@mail.gmail.com> References: <504b2780607222210m4b36741ak57ff53e3d6a7e3ca@mail.gmail.com> <5230c2820607222307m46edfbadwdc84035b5a549578@mail.gmail.com> <504b2780607222318x631b33bakc870e24b81c24850@mail.gmail.com> <1815f00607222341n1ba39a94s4055d21e7c88b3c0@mail.gmail.com> Message-ID: 21 minutes for sandmark UM in Java5 on iBook, MacOSX 10.4 That's not snappy but it seems alright for UMIX, most of the time (if array allocation/deallocation is not too bad). The qbasic programs are quite slow, though. From bluelive at xs4all.nl Sun Jul 23 04:02:50 2006 From: bluelive at xs4all.nl (Bart van der Werf) Date: Sun Jul 23 04:02:53 2006 Subject: [icfp-discuss] Who's got the fastest sandmark? In-Reply-To: <504b2780607222210m4b36741ak57ff53e3d6a7e3ca@mail.gmail.com> Message-ID: <003101c6ae2e$6493e320$4101a8c0@andromeda> 4Min:09Sec 529M on an amd 2600 (1.5ghz) -----Original Message----- From: icfpcontest-discuss-bounces@lists.andrew.cmu.edu [mailto:icfpcontest-discuss-bounces@lists.andrew.cmu.edu] On Behalf Of David Dalrymple Sent: Sunday, July 23, 2006 7:11 AM To: Discussion of the 2006 ICFP Programming Contest Subject: [icfp-discuss] Who's got the fastest sandmark? Just got mine to work - runs in 1m38s. Of course, that's on a 3Ghz Xeon server -- so maybe it would be fairer to do a ratio against SuperPI. On SuperPI calculating 2^20 digits on the server, I have 40s user time. So that's a calibrated SANDmark of 2.45. Can anyone beat that? --David _______________________________________________ icfpcontest-discuss mailing list icfpcontest-discuss@lists.andrew.cmu.edu https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss From mhx at iae.nl Sun Jul 23 03:45:23 2006 From: mhx at iae.nl (Marcel Hendrix) Date: Sun Jul 23 04:03:25 2006 Subject: [icfp-discuss] Who's got the fastest sandmark? Message-ID: <20060723074523.WKHH1331.amsfep20-int.chello.nl@smtp.chello.nl> "David Dalrymple" wrote Re: [icfp-discuss] Who's got the fastest sandmark? > Just got mine to work - runs in 1m38s. > Of course, that's on a 3Ghz Xeon server -- so maybe it would be fairer > to do a ratio against SuperPI. Sandmark: 4 minutes, 40 seconds on a PIV 3 GHz. Language used: Forth (iForth 2.0) All checks and all bells and whistles are on. (No arcane optimizations used). Not bad for a interpreter :-) State of the art Forths (VFX) could be up to twice faster than this timing. -marcel --- 09:31:33, July 23, 2006 trying to Allocate array of size 0.. trying to Abandon size 0 allocation.. trying to Allocate size 11.. trying Array Index on allocated array.. trying Amendment of allocated array.. checking Amendment of allocated array.. trying Alloc(a,a) and amending it.. comparing multiple allocations.. pointer arithmetic.. check old allocation.. simple tests ok! about to load program from some allocated array.. success. verifying that the array and its copy are the same... success. testing aliasing.. success. free after loadprog.. success. loadprog ok. == SANDmark 19106 beginning stress test / benchmark.. == 100. 12345678.09abcdef 99. 6d58165c.2948d58d 98. 0f63b9ed.1d9c4076 [ ... ] 6. a7b6666b.509e5338 5. d0e8236e.8b0e9826 4. 4d20f3ac.a25d05a8 3. 7c7394b2.476c1ee5 2. f3a52453.19cc755d 1. 2c80b43d.5646302f 0. a8d1619e.5540e6cf SANDmark complete. 09:36:14, July 23, 2006 ok FORTH> From bluelive at xs4all.nl Sun Jul 23 04:38:08 2006 From: bluelive at xs4all.nl (Bart van der Werf) Date: Sun Jul 23 04:38:12 2006 Subject: [icfp-discuss] Who's got the fastest sandmark? In-Reply-To: <20060723074523.WKHH1331.amsfep20-int.chello.nl@smtp.chello.nl> Message-ID: <000001c6ae33$534ae460$4101a8c0@andromeda> Nice, that places it in the range of C implementations :) -----Original Message----- From: icfpcontest-discuss-bounces@lists.andrew.cmu.edu [mailto:icfpcontest-discuss-bounces@lists.andrew.cmu.edu] On Behalf Of Marcel Hendrix Sent: Sunday, July 23, 2006 9:45 AM To: icfpcontest-discuss@lists.andrew.cmu.edu Subject: Re: [icfp-discuss] Who's got the fastest sandmark? "David Dalrymple" wrote Re: [icfp-discuss] Who's got the fastest sandmark? > Just got mine to work - runs in 1m38s. > Of course, that's on a 3Ghz Xeon server -- so maybe it would be fairer > to do a ratio against SuperPI. Sandmark: 4 minutes, 40 seconds on a PIV 3 GHz. Language used: Forth (iForth 2.0) All checks and all bells and whistles are on. (No arcane optimizations used). Not bad for a interpreter :-) State of the art Forths (VFX) could be up to twice faster than this timing. -marcel --- 09:31:33, July 23, 2006 trying to Allocate array of size 0.. trying to Abandon size 0 allocation.. trying to Allocate size 11.. trying Array Index on allocated array.. trying Amendment of allocated array.. checking Amendment of allocated array.. trying Alloc(a,a) and amending it.. comparing multiple allocations.. pointer arithmetic.. check old allocation.. simple tests ok! about to load program from some allocated array.. success. verifying that the array and its copy are the same... success. testing aliasing.. success. free after loadprog.. success. loadprog ok. == SANDmark 19106 beginning stress test / benchmark.. == 100. 12345678.09abcdef 99. 6d58165c.2948d58d 98. 0f63b9ed.1d9c4076 [ ... ] 6. a7b6666b.509e5338 5. d0e8236e.8b0e9826 4. 4d20f3ac.a25d05a8 3. 7c7394b2.476c1ee5 2. f3a52453.19cc755d 1. 2c80b43d.5646302f 0. a8d1619e.5540e6cf SANDmark complete. 09:36:14, July 23, 2006 ok FORTH> _______________________________________________ icfpcontest-discuss mailing list icfpcontest-discuss@lists.andrew.cmu.edu https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss From aguilar.james at gmail.com Sun Jul 23 05:02:11 2006 From: aguilar.james at gmail.com (James Aguilar) Date: Sun Jul 23 05:02:14 2006 Subject: [icfp-discuss] Who's got the fastest sandmark? In-Reply-To: <504b2780607222210m4b36741ak57ff53e3d6a7e3ca@mail.gmail.com> References: <504b2780607222210m4b36741ak57ff53e3d6a7e3ca@mail.gmail.com> Message-ID: On 7/22/06, David Dalrymple wrote: > > Just got mine to work - runs in 1m38s. > I'm not going to do a whole calibration thing, but I will say that Python+psyco was working but would have taken hours, so I redid it in C++ and got 3:38 on a 1.3 GHz Pentium M with 768 M of slooooooow RAM. :) -- James Aguilar -- [?] James Aguilar [@] 18100 NE 95th St. #RR3088 / Redmond, WA 98052 [#] 314 494 0450 [!] In the wind and the rain, my darling, say goodbye -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060723/970625e1/attachment.html From mhx at iae.nl Sun Jul 23 05:09:04 2006 From: mhx at iae.nl (Marcel Hendrix) Date: Sun Jul 23 05:09:20 2006 Subject: [icfp-discuss] Who's got the fastest sandmark? Message-ID: <20060723090904.TOXC1471.amsfep11-int.chello.nl@smtp.chello.nl> "Bart van der Werf" writes RE: [icfp-discuss] Who's got the fastest sandmark? > Nice, that places it in the range of C implementations :) I downloaded SuperPi 1.1e. On my machine it takes 49 seconds for 2^20 digits, so my calibrated SANDMARKrating is 280/49 = 5.71. -marcel From chris at chr-breitkopf.de Sun Jul 23 05:30:28 2006 From: chris at chr-breitkopf.de (Christoph Breitkopf) Date: Sun Jul 23 05:30:17 2006 Subject: [icfp-discuss] Who's got the fastest sandmark? In-Reply-To: <1815f00607222341n1ba39a94s4055d21e7c88b3c0@mail.gmail.com> References: <504b2780607222210m4b36741ak57ff53e3d6a7e3ca@mail.gmail.com> <5230c2820607222307m46edfbadwdc84035b5a549578@mail.gmail.com> <504b2780607222318x631b33bakc870e24b81c24850@mail.gmail.com> <1815f00607222341n1ba39a94s4055d21e7c88b3c0@mail.gmail.com> Message-ID: "Robert Van Dam" writes: > I'm quite curious to hear if anyone has gotten this to run within a > reasonable amount of time using something other than C (perhaps even a > functional language ;). If Java counts as "other than C" (it's not, really), then yes, in about one hour. Regards, Chris From chris at chr-breitkopf.de Sun Jul 23 06:02:11 2006 From: chris at chr-breitkopf.de (Christoph Breitkopf) Date: Sun Jul 23 06:02:01 2006 Subject: [icfp-discuss] Who's got the fastest sandmark? In-Reply-To: <504b2780607222210m4b36741ak57ff53e3d6a7e3ca@mail.gmail.com> References: <504b2780607222210m4b36741ak57ff53e3d6a7e3ca@mail.gmail.com> Message-ID: "David Dalrymple" writes: > Just got mine to work - runs in 1m38s. > > Of course, that's on a 3Ghz Xeon server -- so maybe it would be fairer > to do a ratio against SuperPI. Same time (1m37.946s) on a 2.533 GHz Pentium 4 :-) Plain interpreter in C using malloc()/free() even for small arrays. So I guess there's still some space for improvements. My older Java version takes 6m48s. Regards, Chris From phallien at yahoo.com Sun Jul 23 04:59:02 2006 From: phallien at yahoo.com (P. Hallien) Date: Sun Jul 23 06:14:27 2006 Subject: [icfp-discuss] UMIX Password Cracking Message-ID: <20060723085902.78478.qmail@web35508.mail.mud.yahoo.com> [ NOTE: Some spoilers may follow, depending on your current progress ] Gentlemen, I've been noodling around for some time now trying to crack the 'gardener' and 'bbarker' accounts on the UMIX system. The comment embedded in guest/code/hack.bas seemed to suggest that appending a two-digit decimal number after a dictionary word would be the way to go ... I tried this, along with as many permutations as I could think of, (e.g. using "I", "II", "IV", "X", "C", pre-pending the digits instead of appending them, etc.) , all without success. I'm hoping that someone could at least bound the scope of the effort that's required (or give me a hint as to the direction I should be heading) -- should I be using a different (larger/external) dictionary file? Thanks! P.Hallien __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com From AndreasRossberg at web.de Sun Jul 23 06:01:55 2006 From: AndreasRossberg at web.de (Andreas Rossberg) Date: Sun Jul 23 06:14:27 2006 Subject: [icfp-discuss] Who's got the fastest sandmark? Message-ID: <003401c6ae3f$07781240$15b2a8c0@wiko> Robert Van Dam wrote:> I'm quite curious to hear if anyone has gotten this to run within a> reasonable amount of time using something other than C (perhaps even a> functional language ;).Chris King wrote:>On 7/23/06, David Dalrymple wrote:>> I think that would be very unlikely. Other C-like languages might>> have a shot (Fortran, C++) but even they would be slower. ASM has a>> shot given unlimited time to code it :) I think OCaml is the best>> functional language for this but that would be too slow as well (that>> is just a guess- I haven't tried it).>>My O'Caml implementation clocks about 10x slower than my C>implementation, even with optimizations turned on. It suffers from>massive indirection, since it uses library functions for the 32-bit>integers and the arrays are unboxed. It also suffers from heavy>memory thrashing due to all this overhead.[You mean _boxed_?]Our straightforward implementation in Standard ML (110 lines) takesabout 2m30 to run SANDmark on a 2.2 GHz P4 machine. But indeed, weused the MLton compiler, because it uses unboxed 32 bit words.- Andreas From tom7 at cs.cmu.edu Sun Jul 23 06:20:50 2006 From: tom7 at cs.cmu.edu (Tom Murphy) Date: Sun Jul 23 06:20:51 2006 Subject: [icfp-discuss] UMIX Password Cracking In-Reply-To: <20060723085902.78478.qmail@web35508.mail.mud.yahoo.com> Message-ID: Only a few users have insecure passwords. You should try to work on the other puzzles in order to unlock more accounts... > [ NOTE: Some spoilers may follow, depending on your > current progress ] > > > Gentlemen, > > I've been noodling around for some time now trying > to crack the 'gardener' and 'bbarker' accounts on the > UMIX system. > > The comment embedded in guest/code/hack.bas seemed > to suggest that appending a two-digit decimal number > after a dictionary word would be the way to go ... > > I tried this, along with as many permutations as I > could think of, (e.g. using "I", "II", "IV", "X", "C", > pre-pending the digits instead of appending them, > etc.) , all without success. > > I'm hoping that someone could at least bound the > scope of the effort that's required (or give me a hint > as to the direction I should be heading) -- should I > be using a different (larger/external) dictionary > file? > > Thanks! > P.Hallien > > > __________________________________________________ > Do You Yahoo!? > Tired of spam? Yahoo! Mail has the best spam protection around > http://mail.yahoo.com > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > > [ NEW! : http://tom7.org/ ] [ OLD! : http://fonts.tom7.com/ ] From 0zu31 at gmx.de Sun Jul 23 06:23:51 2006 From: 0zu31 at gmx.de (Mechthild 'Rejis' Czapp) Date: Sun Jul 23 06:24:54 2006 Subject: [icfp-discuss] Am I going insane? codex fails even when tracing by hand Message-ID: <200607231223.51173.0zu31@gmx.de> Hi, My virtual machine fails after 5 instructions because it then stumbles upon an operation which is C4 5D D8 F0 in hex (which I so far interpreted as operation 12 with the parameters 3 6 0). The MD5 sum appears to be correct... what is the problem? Horrified in H?rth, Mechthild From aguilar.james at gmail.com Sun Jul 23 06:48:19 2006 From: aguilar.james at gmail.com (James Aguilar) Date: Sun Jul 23 06:48:23 2006 Subject: [icfp-discuss] Am I going insane? codex fails even when tracing by hand In-Reply-To: <200607231223.51173.0zu31@gmx.de> References: <200607231223.51173.0zu31@gmx.de> Message-ID: On 7/23/06, Mechthild 'Rejis' Czapp <0zu31@gmx.de> wrote: > > Hi, > > My virtual machine fails after 5 instructions because it then stumbles > upon an > operation which is C4 5D D8 F0 in hex (which I so far interpreted as > operation 12 with the parameters 3 6 0). The MD5 sum appears to be > correct... > what is the problem? At that point in time, the value in register 6 should be zero and the value in register 0 should be (some number later in the file. What about your machine fails? It should just jump to index C in array B. -- [?] James Aguilar [@] 18100 NE 95th St. #RR3088 / Redmond, WA 98052 [#] 314 494 0450 [!] In the wind and the rain, my darling, say goodbye -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060723/817b298d/attachment.html From samuel at acid-code.ch Sun Jul 23 07:10:07 2006 From: samuel at acid-code.ch (Samuel Burri) Date: Sun Jul 23 07:09:00 2006 Subject: [icfp-discuss] UMIX Password Cracking In-Reply-To: References: <20060723085902.78478.qmail@web35508.mail.mud.yahoo.com> Message-ID: <53835.85.2.216.161.1153653007.squirrel@serioussam.selfip.org> Plus you have to append decimal digits eg. asciicodes 48 through 57 > > Only a few users have insecure passwords. You should try to work on the > other puzzles in order to unlock more accounts... > > >> [ NOTE: Some spoilers may follow, depending on your >> current progress ] >> >> >> Gentlemen, >> >> I've been noodling around for some time now trying >> to crack the 'gardener' and 'bbarker' accounts on the >> UMIX system. >> >> The comment embedded in guest/code/hack.bas seemed >> to suggest that appending a two-digit decimal number >> after a dictionary word would be the way to go ... >> >> I tried this, along with as many permutations as I >> could think of, (e.g. using "I", "II", "IV", "X", "C", >> pre-pending the digits instead of appending them, >> etc.) , all without success. >> >> I'm hoping that someone could at least bound the >> scope of the effort that's required (or give me a hint >> as to the direction I should be heading) -- should I >> be using a different (larger/external) dictionary >> file? >> >> Thanks! >> P.Hallien >> >> >> __________________________________________________ >> Do You Yahoo!? >> Tired of spam? Yahoo! Mail has the best spam protection around >> http://mail.yahoo.com >> _______________________________________________ >> icfpcontest-discuss mailing list >> icfpcontest-discuss@lists.andrew.cmu.edu >> https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss >> >> > > > [ NEW! : http://tom7.org/ ] > [ OLD! : http://fonts.tom7.com/ ] > > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > > From icfp_06 at schlagsei.de Sun Jul 23 07:32:01 2006 From: icfp_06 at schlagsei.de (=?ISO-8859-15?Q?Daniel_Schr=F6ter?=) Date: Sun Jul 23 07:31:49 2006 Subject: [icfp-discuss] Sandmark Message-ID: <44C35E31.2030506@schlagsei.de> Hi! I spend 3 hours waiting for my UM to decrypt last night (Germany, so it was late :) ). It crahed with an OutOfMemory Exception, so i went to bed and thought: Too much for my 1.2Ghz, <1GB RAM machine ;) After getting up today, i found the sandmark.umz and tested it. And this is the result: Index in 0-array fail How can my UM run the codex.umz hours and hours, but be wrong?! greetz Daniel From jjmaes at gmail.com Sun Jul 23 07:32:20 2006 From: jjmaes at gmail.com (James Darwin Maes The First (a javatarian)) Date: Sun Jul 23 07:32:23 2006 Subject: [icfp-discuss] Who's got the fastest sandmark? In-Reply-To: <003401c6ae3f$07781240$15b2a8c0@wiko> References: <003401c6ae3f$07781240$15b2a8c0@wiko> Message-ID: 3m14sec with Java 1.5 on a Intel(R) Xeon(TM) CPU 2.80GHz On 7/23/06, Andreas Rossberg wrote: > > Robert Van Dam wrote:> I'm quite curious to hear if anyone has gotten this > to run within a> reasonable amount of time using something other than C > (perhaps even a> functional language ;).Chris King wrote:>On 7/23/06, > David > Dalrymple wrote:>> I think that would be very > unlikely. > Other C-like languages might>> have a shot (Fortran, C++) but even they > would be slower. ASM has a>> shot given unlimited time to code it :) I > think OCaml is the best>> functional language for this but that would be > too > slow as well (that>> is just a guess- I haven't tried it).>>My O'Caml > implementation clocks about 10x slower than my C>implementation, even with > optimizations turned on. It suffers from>massive indirection, since it > uses > library functions for the 32-bit>integers and the arrays are unboxed. It > also suffers from heavy>memory thrashing due to all this overhead.[You > mean > _boxed_?]Our straightforward implementation in Standard ML (110 lines) > takesabout 2m30 to run SANDmark on a 2.2 GHz P4 machine. But indeed, > weused > the MLton compiler, because it uses unboxed 32 bit words.- Andreas > > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060723/d62dabc4/attachment.html From dhebert at uci.edu Sun Jul 23 07:50:24 2006 From: dhebert at uci.edu (dhebert@uci.edu) Date: Sun Jul 23 07:50:26 2006 Subject: [icfp-discuss] Bug? Or just slow? Message-ID: <4356.71.106.157.231.1153655424.squirrel@webmail.uci.edu> Hey there everyone, My team's UM currently appears to decrypt the codex correctly, passes the SANDmark self tests, and lets us log into UMIX. However, the SANDmark stress test benchmarker didn't finish after 2 hours, and when I input code with UMODEM, it nevers seems to finish whatever it does after I input "STOP". Is it possible that there is still a bug in our UM? We are coding in C/C++ with a decent amount of direct memory accessing and most stuff runs pretty darn quick in it. Thanks for the help, -Daniel, a member of team B.C. Tech From nickie at softlab.ntua.gr Sun Jul 23 07:36:13 2006 From: nickie at softlab.ntua.gr (Nikolaos S. Papaspyrou) Date: Sun Jul 23 07:56:25 2006 Subject: [icfp-discuss] sandmark: testing aliasing.. Message-ID: <44C35F2D.1020308@softlab.ntua.gr> Hi there, We have a UM in OCaml that seems to be running perfectly. It decrypted the codex a long time ago, and everything works from that point (we have around 1000 points using it). After the sandmark was out, we tested our UM with the sandmark. It fails with: testing aliasing.. loadProg: array not copied (1) For speeding up instruction #12 we've done a couple of tricks, but we believe our UM is complying. Can we have some idea about what "testing aliasing" tests? Thanks, Nikos. -- Nikolaos S. Papaspyrou National Technical University of Athens School of Electrical & Computer Engineering Software Engineering Laboratory From spoons at cs.cmu.edu Sun Jul 23 08:08:09 2006 From: spoons at cs.cmu.edu (Daniel Spoonhower) Date: Sun Jul 23 08:08:16 2006 Subject: [icfp-discuss] sandmark: testing aliasing.. In-Reply-To: <44C35F2D.1020308@softlab.ntua.gr> References: <44C35F2D.1020308@softlab.ntua.gr> Message-ID: <44C366A9.1090709@cs.cmu.edu> The specification states that... The array identified by the B register is duplicated and the duplicate shall replace the '0' array, regardless of size. This test is checking whether or not you have aliased the array identified by the B register. --djs Nikolaos S. Papaspyrou wrote: > Hi there, > > We have a UM in OCaml that seems to be running perfectly. > It decrypted the codex a long time ago, and everything > works from that point (we have around 1000 points using it). > After the sandmark was out, we tested our UM with the > sandmark. It fails with: > > testing aliasing.. > loadProg: array not copied (1) > > For speeding up instruction #12 we've done a couple of > tricks, but we believe our UM is complying. > > Can we have some idea about what "testing aliasing" tests? > > Thanks, > Nikos. > > -- > Nikolaos S. Papaspyrou > National Technical University of Athens > School of Electrical & Computer Engineering > Software Engineering Laboratory > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > From nickie at softlab.ntua.gr Sun Jul 23 08:16:52 2006 From: nickie at softlab.ntua.gr (Nikolaos S. Papaspyrou) Date: Sun Jul 23 08:17:00 2006 Subject: [icfp-discuss] sandmark: testing aliasing.. In-Reply-To: <44C366A9.1090709@cs.cmu.edu> References: <44C35F2D.1020308@softlab.ntua.gr> <44C366A9.1090709@cs.cmu.edu> Message-ID: <44C368B4.7090708@softlab.ntua.gr> > The specification states that... > > The array identified by the B register is duplicated > and the duplicate shall replace the '0' array, > regardless of size. > > This test is checking whether or not you have aliased the array identified by > the B register. I can think of two ways to test whether our UM is just aliasing, instead of copying: 1) by using ammend in one of the two arrays and testing whether the other changes 2) by freeing the array identified by the B register Is there some other test that escapes us? It would be useful to know... By the way, I find it a little strange that we're seeing ants and adventure games and everything else, and still the sandmark fails because of a bug in our UM. Thanks, Nikos. -- Nikolaos S. Papaspyrou National Technical University of Athens School of Electrical & Computer Engineering Software Engineering Laboratory From spoons at cs.cmu.edu Sun Jul 23 08:27:20 2006 From: spoons at cs.cmu.edu (Daniel Spoonhower) Date: Sun Jul 23 08:27:24 2006 Subject: [icfp-discuss] sandmark: testing aliasing.. In-Reply-To: <44C368B4.7090708@softlab.ntua.gr> References: <44C35F2D.1020308@softlab.ntua.gr> <44C366A9.1090709@cs.cmu.edu> <44C368B4.7090708@softlab.ntua.gr> Message-ID: <44C36B28.7090205@cs.cmu.edu> > I can think of two ways to test whether our UM is just aliasing, > instead of copying: > > 1) by using ammend in one of the two arrays and testing whether > the other changes > 2) by freeing the array identified by the B register The test that is failing ("loadProg: array not copied (1)") is an example of #1. > By the way, I find it a little strange that we're seeing ants and > adventure games and everything else, and still the sandmark fails > because of a bug in our UM. The self-checks are engineered to test compliance with the specification (though they are still not exhaustive). On the other hand, the rest of the UMIX program was designed for speed and as a result might use some features of the UM rarely (if at all). --djs From clive at xtra.co.nz Sun Jul 23 08:36:02 2006 From: clive at xtra.co.nz (Clive Gifford) Date: Sun Jul 23 08:36:23 2006 Subject: [icfp-discuss] Who's got the fastest sandmark? References: <003401c6ae3f$07781240$15b2a8c0@wiko> Message-ID: <003501c6ae54$9287ff70$0301a8c0@Owhango1> Hah! The "Black Knights" are finally on the scoreboard... I finally decided to do what every fibre of my being was telling me do when I first saw the spec and that was write it in C! Instead I spent lots of time with Python (knew that would be too slow thought it would be fun to try it first) and Pyrex (= "Python with option C types") but even that was too slow, even after making it more and more C-like. Sometimes you just have to use the right tool for the job. My absolutely bare bones C version runs the sandmark in 7m 20s on a 500Mhz Pentuim 3 with 384MB RAM on Win2000 Pro. "C is the language of choice for discriminating hackers!" Sometimes you just have to use the right tool for the job!! And by the way, is there anybody out there working on a SLOWER machine than that? :-) Good night from New Zealand... Black Knights. From pauls101 at mindspring.com Sun Jul 23 09:27:02 2006 From: pauls101 at mindspring.com (Paul Sexton) Date: Sun Jul 23 09:27:08 2006 Subject: [icfp-discuss] sandmark: testing aliasing.. References: <44C35F2D.1020308@softlab.ntua.gr> <44C366A9.1090709@cs.cmu.edu><44C368B4.7090708@softlab.ntua.gr> <44C36B28.7090205@cs.cmu.edu> Message-ID: <000801c6ae5b$afa7bb20$6401a8c0@ACER> ----- Original Message ----- From: "Daniel Spoonhower" To: "Nikolaos S. Papaspyrou" Cc: "Discussion of the 2006 ICFP Programming Contest" Sent: Sunday, July 23, 2006 8:27 AM Subject: Re: [icfp-discuss] sandmark: testing aliasing.. > >> I can think of two ways to test whether our UM is just aliasing, >> instead of copying: >> >> 1) by using ammend in one of the two arrays and testing whether >> the other changes >> 2) by freeing the array identified by the B register > > The test that is failing ("loadProg: array not copied (1)") is an example > of #1. > >> By the way, I find it a little strange that we're seeing ants and >> adventure games and everything else, and still the sandmark fails >> because of a bug in our UM. > > The self-checks are engineered to test compliance with the specification > (though they are still not exhaustive). On the other hand, the rest of > the > UMIX program was designed for speed and as a result might use some > features of > the UM rarely (if at all). > One thing is worse "seeing ants and adventure games" and the sandmark failing, and that is the sandmark passing but the dumped program is still corrupt and won't run! I'm stuck in a cheap motel w/ a very cheap laptop and a copy of VC++ Express I grabbed on my way here, and now I'm running out of ideas to tweak my UM to get past whatever isn't quite right. From tom7 at cs.cmu.edu Sun Jul 23 09:41:46 2006 From: tom7 at cs.cmu.edu (Tom Murphy) Date: Sun Jul 23 09:41:48 2006 Subject: [icfp-discuss] sandmark: testing aliasing.. In-Reply-To: <000801c6ae5b$afa7bb20$6401a8c0@ACER> Message-ID: If SANDmark is passing, I'd take a second look at the Output instruction and the way that you're editing off the header of the UM data. Many teams have had trouble with annoying silent things about their environment or tools like \n characters being converted to \r\n or \r being dropped. Is the output data the correct size? Tom > > One thing is worse "seeing ants and adventure games" and the sandmark > failing, and that is the sandmark passing but the dumped program is still > corrupt and won't run! > > I'm stuck in a cheap motel w/ a very cheap laptop and a copy of VC++ Express > I grabbed on my way here, and now I'm running out of ideas to tweak my UM to > get past whatever isn't quite right. > > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > > [ NEW! : http://tom7.org/ ] [ OLD! : http://fonts.tom7.com/ ] From andy_zyx at yahoo.com Sun Jul 23 10:26:48 2006 From: andy_zyx at yahoo.com (Andy) Date: Sun Jul 23 10:26:50 2006 Subject: [icfp-discuss] DIV error Message-ID: <20060723142648.27105.qmail@web50701.mail.yahoo.com> Hi All, I completed coding the UM and tried running it and am getting the message "DIV error". I checked to make sure that there is div by zero or negaivtive value conditions ...anybody had this problem ? Thanks, Andy __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com From pocmatos at gmail.com Sun Jul 23 10:59:02 2006 From: pocmatos at gmail.com (Paulo J. Matos) Date: Sun Jul 23 10:59:05 2006 Subject: [icfp-discuss] Question on division Message-ID: <11b141710607230759o3336775aq5e04f2bc82590477@mail.gmail.com> Can someone confirm this please. This seems to be platter 405 in Sandmark: Division(R2, R0, R1) : R0 = 1, R1 = 4294967295, R2 <- 0 Should R2 have 0 after this division? Cheers, -- Paulo Jorge Matos - pocm at sat inesc-id pt Web: http://sat.inesc-id.pt/~pocm Computer and Software Engineering INESC-ID - SAT Group From windenntw at gmail.com Sun Jul 23 11:18:25 2006 From: windenntw at gmail.com (Antonio Vargas) Date: Sun Jul 23 11:18:28 2006 Subject: [icfp-discuss] Question on division In-Reply-To: <11b141710607230759o3336775aq5e04f2bc82590477@mail.gmail.com> References: <11b141710607230759o3336775aq5e04f2bc82590477@mail.gmail.com> Message-ID: <69304d110607230818u518276e2x6b6c8bb722dc8710@mail.gmail.com> Paulo, i've decided not to control if the divisor == 0 and have had no errors yes, so most probably its an earlier error that your are having a divisor == 0. On 7/23/06, Paulo J. Matos wrote: > Can someone confirm this please. > > This seems to be platter 405 in Sandmark: > > Division(R2, R0, R1) : R0 = 1, R1 = 4294967295, R2 <- 0 > > Should R2 have 0 after this division? > > Cheers, > > -- > Paulo Jorge Matos - pocm at sat inesc-id pt > Web: http://sat.inesc-id.pt/~pocm > Computer and Software Engineering > INESC-ID - SAT Group > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > -- Greetz, Antonio Vargas aka winden of network http://network.amigascne.org/ windNOenSPAMntw@gmail.com thesameasabove@amigascne.org Every day, every year you have to work you have to study you have to scene. From tom7 at cs.cmu.edu Sun Jul 23 11:31:43 2006 From: tom7 at cs.cmu.edu (Tom Murphy) Date: Sun Jul 23 11:31:45 2006 Subject: [icfp-discuss] Question on division In-Reply-To: <69304d110607230818u518276e2x6b6c8bb722dc8710@mail.gmail.com> Message-ID: I'm not sure about the offset, but SANDmark does test that 1 / 0xFFFFFFFF correctly computes 0, yes. It should never try to divide by zero. Tom > Paulo, i've decided not to control if the divisor == 0 and have had no > errors yes, so most probably its an earlier error that your are having > a divisor == 0. > > > On 7/23/06, Paulo J. Matos wrote: > > Can someone confirm this please. > > > > This seems to be platter 405 in Sandmark: > > > > Division(R2, R0, R1) : R0 = 1, R1 = 4294967295, R2 <- 0 > > > > Should R2 have 0 after this division? > > > > Cheers, > > > > -- > > Paulo Jorge Matos - pocm at sat inesc-id pt > > Web: http://sat.inesc-id.pt/~pocm > > Computer and Software Engineering > > INESC-ID - SAT Group > > _______________________________________________ > > icfpcontest-discuss mailing list > > icfpcontest-discuss@lists.andrew.cmu.edu > > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > > > > > -- > Greetz, Antonio Vargas aka winden of network > > http://network.amigascne.org/ > windNOenSPAMntw@gmail.com > thesameasabove@amigascne.org > > Every day, every year > you have to work > you have to study > you have to scene. > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > > [ NEW! : http://tom7.org/ ] [ OLD! : http://fonts.tom7.com/ ] From pocmatos at gmail.com Sun Jul 23 11:46:09 2006 From: pocmatos at gmail.com (Paulo J. Matos) Date: Sun Jul 23 11:46:13 2006 Subject: [icfp-discuss] Question on division In-Reply-To: References: <69304d110607230818u518276e2x6b6c8bb722dc8710@mail.gmail.com> Message-ID: <11b141710607230846n48dc5d0cm7eaae6ceb30a27ec@mail.gmail.com> On 23/07/06, Tom Murphy wrote: > > I'm not sure about the offset, but SANDmark does test that 1 / 0xFFFFFFFF > correctly computes 0, yes. Great, that's what I needed to know. Thanks! :-) >It should never try to divide by zero. > > Tom > > > Paulo, i've decided not to control if the divisor == 0 and have had no > > errors yes, so most probably its an earlier error that your are having > > a divisor == 0. > > > > > > On 7/23/06, Paulo J. Matos wrote: > > > Can someone confirm this please. > > > > > > This seems to be platter 405 in Sandmark: > > > > > > Division(R2, R0, R1) : R0 = 1, R1 = 4294967295, R2 <- 0 > > > > > > Should R2 have 0 after this division? > > > > > > Cheers, > > > > > > -- > > > Paulo Jorge Matos - pocm at sat inesc-id pt > > > Web: http://sat.inesc-id.pt/~pocm > > > Computer and Software Engineering > > > INESC-ID - SAT Group > > > _______________________________________________ > > > icfpcontest-discuss mailing list > > > icfpcontest-discuss@lists.andrew.cmu.edu > > > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > > > > > > > > > -- > > Greetz, Antonio Vargas aka winden of network > > > > http://network.amigascne.org/ > > windNOenSPAMntw@gmail.com > > thesameasabove@amigascne.org > > > > Every day, every year > > you have to work > > you have to study > > you have to scene. > > _______________________________________________ > > icfpcontest-discuss mailing list > > icfpcontest-discuss@lists.andrew.cmu.edu > > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > > > > > > > [ NEW! : http://tom7.org/ ] > [ OLD! : http://fonts.tom7.com/ ] > > -- Paulo Jorge Matos - pocm at sat inesc-id pt Web: http://sat.inesc-id.pt/~pocm Computer and Software Engineering INESC-ID - SAT Group From davidad at mit.edu Sun Jul 23 11:48:41 2006 From: davidad at mit.edu (David Dalrymple) Date: Sun Jul 23 11:48:44 2006 Subject: [icfp-discuss] DIV error In-Reply-To: <20060723142648.27105.qmail@web50701.mail.yahoo.com> References: <20060723142648.27105.qmail@web50701.mail.yahoo.com> Message-ID: <504b2780607230848vdbf1ce5qed1971ac39559a89@mail.gmail.com> Possibly truncating/rounding issue? --David On 7/23/06, Andy wrote: > Hi All, > > I completed coding the UM and tried running it and am getting the message "DIV error". I checked > to make sure that there is div by zero or negaivtive value conditions ...anybody had this problem > ? > > Thanks, > Andy > > > __________________________________________________ > Do You Yahoo!? > Tired of spam? Yahoo! Mail has the best spam protection around > http://mail.yahoo.com > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > From tom7 at cs.cmu.edu Sun Jul 23 11:53:02 2006 From: tom7 at cs.cmu.edu (Tom Murphy) Date: Sun Jul 23 11:53:03 2006 Subject: [icfp-discuss] DIV error In-Reply-To: <504b2780607230848vdbf1ce5qed1971ac39559a89@mail.gmail.com> Message-ID: Also, make sure that you are doing unsigned division. At this point it should be possible to single-step through the instructions and check that they are doing the correct thing... Tom > Possibly truncating/rounding issue? > > --David > > On 7/23/06, Andy wrote: > > Hi All, > > > > I completed coding the UM and tried running it and am getting the message "DIV error". I checked > > to make sure that there is div by zero or negaivtive value conditions ...anybody had this problem > > ? > > > > Thanks, > > Andy > > [ NEW! : http://tom7.org/ ] [ OLD! : http://fonts.tom7.com/ ] From windenntw at gmail.com Sun Jul 23 12:38:48 2006 From: windenntw at gmail.com (Antonio Vargas) Date: Sun Jul 23 12:38:52 2006 Subject: [icfp-discuss] Who's got the fastest sandmark? In-Reply-To: <003501c6ae54$9287ff70$0301a8c0@Owhango1> References: <003401c6ae3f$07781240$15b2a8c0@wiko> <003501c6ae54$9287ff70$0301a8c0@Owhango1> Message-ID: <69304d110607230938p6da165d1s35b138d6ee1f030f@mail.gmail.com> ... 0. a8d1619e.5540e6cf SANDmark complete. ABORT reached real 3m21.343s user 3m7.810s sys 0m1.320s output to console is not buffered, maybe that explains the >10 secs of system time.... G4@1.25, 1G ram, OSX 10.3, pure C implementation with gcc4, -O3 and -fstrict-aliasing -- Greetz, Antonio Vargas aka winden of network http://network.amigascne.org/ windNOenSPAMntw@gmail.com thesameasabove@amigascne.org Every day, every year you have to work you have to study you have to scene. From rinnert at hep.ph.liv.ac.uk Sun Jul 23 12:54:59 2006 From: rinnert at hep.ph.liv.ac.uk (Kurt Rinnert) Date: Sun Jul 23 12:56:20 2006 Subject: [icfp-discuss] Who's got the fastest sandmark? In-Reply-To: <69304d110607230938p6da165d1s35b138d6ee1f030f@mail.gmail.com> References: <003401c6ae3f$07781240$15b2a8c0@wiko> <003501c6ae54$9287ff70$0301a8c0@Owhango1> <69304d110607230938p6da165d1s35b138d6ee1f030f@mail.gmail.com> Message-ID: <20060723165459.GC5614@vulcan> [...] 1. 2c80b43d.5646302f 0. a8d1619e.5540e6cf real 2m48.250s user 2m46.326s sys 0m1.374s AMD Athlon(tm) 64 Processor 3000+ 1.8 GHz 2 GB RAM Linux 2.6.8-24.20 (SuSE 9.2) C++ implementation with gcc (GCC) 3.3.4 (pre 3.3.5 20040809), -O3 -fstrict-aliasing. Also non-buffered output -- after all it said "[...] is displayed on the console immediately [...]" On Sun, Jul 23, 2006 at 06:38:48PM +0200, Antonio Vargas wrote: > ... > 0. a8d1619e.5540e6cf > SANDmark complete. > ABORT reached > > real 3m21.343s > user 3m7.810s > sys 0m1.320s > > output to console is not buffered, maybe that explains the >10 secs of > system time.... > G4@1.25, 1G ram, OSX 10.3, pure C implementation with gcc4, -O3 and > -fstrict-aliasing > > > -- > Greetz, Antonio Vargas aka winden of network > > http://network.amigascne.org/ > windNOenSPAMntw@gmail.com > thesameasabove@amigascne.org > > Every day, every year > you have to work > you have to study > you have to scene. > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss -- Kurt Rinnert, University of Liverpool - Department of Physics From aguilar.james at gmail.com Sun Jul 23 13:31:18 2006 From: aguilar.james at gmail.com (James Aguilar) Date: Sun Jul 23 13:31:20 2006 Subject: [icfp-discuss] Sandmark In-Reply-To: <44C35E31.2030506@schlagsei.de> References: <44C35E31.2030506@schlagsei.de> Message-ID: On 7/23/06, Daniel Schr?ter wrote: > > Hi! > > I spend 3 hours waiting for my UM to decrypt last night (Germany, so it > was late :) ). > It crahed with an OutOfMemory Exception, so i went to bed and thought: > Too much for my 1.2Ghz, <1GB RAM machine ;) > > After getting up today, i found the sandmark.umz and tested it. > > And this is the result: > Index in 0-array fail > > How can my UM run the codex.umz hours and hours, but be wrong?! If your indexing on the zero array is wrong, you might just go to a wrong instruction, so there would be no guarantee that you would hit anything that would cause you to halt. In any case, the error you're talking about comes pretty close to the beginning of sandmark, which might mean that you really aren't getting that many instructions done before you crash. In any case, I have 1.3 GHz and 768 MB RAM, so you shouldn't be having too much trouble on that count as long as you're managing the memory intelligently. -- [?] James Aguilar [@] 18100 NE 95th St. #RR3088 / Redmond, WA 98052 [#] 314 494 0450 [!] In the wind and the rain, my darling, say goodbye -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060723/17508f3b/attachment.html From aguilar.james at gmail.com Sun Jul 23 13:34:02 2006 From: aguilar.james at gmail.com (James Aguilar) Date: Sun Jul 23 13:34:04 2006 Subject: [icfp-discuss] Bug? Or just slow? In-Reply-To: <4356.71.106.157.231.1153655424.squirrel@webmail.uci.edu> References: <4356.71.106.157.231.1153655424.squirrel@webmail.uci.edu> Message-ID: On 7/23/06, dhebert@uci.edu wrote: > > Hey there everyone, > > My team's UM currently appears to decrypt the codex correctly, passes > the SANDmark self tests, and lets us log into UMIX. However, the > SANDmark stress test benchmarker didn't finish after 2 hours, and when I > input code with UMODEM, it nevers seems to finish whatever it does after > I input "STOP". Is it possible that there is still a bug in our UM? We > are coding in C/C++ with a decent amount of direct memory accessing and > most stuff runs pretty darn quick in it. It's hard to know if it's a bug or a bottleneck without being able to compare against other things that do work. For instance, my UM takes 3 minutes to finish sandmark, but finishes the first part (up to "Loading prog ok" or whatever the last message is) in less than a second. So if you're seeing appropriate scaling there, maybe no problem. If not, maybe you've got a bug OR maybe you've got a bottleneck somewhere in operations that are specifically used a lot in these areas. Try using gprof to find out where you're spending all your time. -- [?] James Aguilar [@] 18100 NE 95th St. #RR3088 / Redmond, WA 98052 [#] 314 494 0450 [!] In the wind and the rain, my darling, say goodbye -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060723/51489a26/attachment.html From mhx at iae.nl Sun Jul 23 13:27:17 2006 From: mhx at iae.nl (Marcel Hendrix) Date: Sun Jul 23 13:34:35 2006 Subject: [icfp-discuss] Who's got the fastest sandmark? Message-ID: <20060723172717.UJBG1331.amsfep20-int.chello.nl@smtp.chello.nl> "Bart van der Werf" wrote RE: [icfp-discuss] Who's got the fastest sandmark? > Nice, that places it in the range of C implementations :) With a small tweak (turning off the cycle counter) the timing for Forth is now 3 minutes, 9 seconds, or a ( SuperPi[2^20] takes 49 s) 3.86 SANDMARK rating. wallclock 189 seconds Intel PIV HT processor 3 GHz 512 MB RAM Windows XP iForth 2.0 Forth interpreter. -marcel From andy_zyx at yahoo.com Sun Jul 23 13:56:14 2006 From: andy_zyx at yahoo.com (Andy) Date: Sun Jul 23 13:56:16 2006 Subject: [icfp-discuss] DIV error In-Reply-To: Message-ID: <20060723175614.58713.qmail@web50715.mail.yahoo.com> Thanks Tom. Is it worth to continue now that many have already submitted and in lead or the contest is of the nature where there is still a chance for the latecomers ? Just wondering :( Andy --- Tom Murphy wrote: > > Also, make sure that you are doing unsigned division. > At this point it should be possible to single-step through the > instructions and check that they are doing the correct thing... > > Tom > > > Possibly truncating/rounding issue? > > > > --David > > > > On 7/23/06, Andy wrote: > > > Hi All, > > > > > > I completed coding the UM and tried running it and am getting the message "DIV error". I > checked > > > to make sure that there is div by zero or negaivtive value conditions ...anybody had this > problem > > > ? > > > > > > Thanks, > > > Andy > > > > > [ NEW! : http://tom7.org/ ] > [ OLD! : http://fonts.tom7.com/ ] > > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com From me.dendik at gmail.com Sun Jul 23 14:07:52 2006 From: me.dendik at gmail.com (Danya) Date: Sun Jul 23 14:07:54 2006 Subject: [icfp-discuss] DIV error In-Reply-To: <20060723175614.58713.qmail@web50715.mail.yahoo.com> References: <20060723175614.58713.qmail@web50715.mail.yahoo.com> Message-ID: <8350b5a00607231107x207280ecl1dfb91e9098d10bb@mail.gmail.com> I'd tell, it's DEFINITELY, UNDOUBTABLY worth continuing! You'll fing a lot of fun discovering the easter eggs contest organizers have put into the tasks! And if you go on, good luck to you with that! Danya. On 7/23/06, Andy wrote: > Thanks Tom. > Is it worth to continue now that many have already submitted and in lead or the contest is of the > nature where there is still a chance for the latecomers ? Just wondering :( From aguilar.james at gmail.com Sun Jul 23 14:08:05 2006 From: aguilar.james at gmail.com (James Aguilar) Date: Sun Jul 23 14:08:07 2006 Subject: [icfp-discuss] DIV error In-Reply-To: <20060723175614.58713.qmail@web50715.mail.yahoo.com> References: <20060723175614.58713.qmail@web50715.mail.yahoo.com> Message-ID: On 7/23/06, Andy wrote: > > Thanks Tom. > Is it worth to continue now that many have already submitted and in lead > or the contest is of the > nature where there is still a chance for the latecomers ? Just wondering > :( It's highly unlikely you'll be able to win at this point. BUT, you should still do it because this contest is f'in amazing. -- James -- [?] James Aguilar [@] 18100 NE 95th St. #RR3088 / Redmond, WA 98052 [#] 314 494 0450 [!] In the wind and the rain, my darling, say goodbye -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060723/e341dab3/attachment-0001.html From micha-1 at fantasymail.de Sun Jul 23 14:12:53 2006 From: micha-1 at fantasymail.de (Michael Wohlwend) Date: Sun Jul 23 14:08:27 2006 Subject: [icfp-discuss] access denied for user Message-ID: <200607232012.54103.micha-1@fantasymail.de> Hi, I allways get the above message, when trying to log into UMIX. Is this part of the game? I tried 'guest' with and without password, teamname... nothing works :-( So if the first task is to find the right login... fine :-) if not, can someone give me a hint? cheers Michael From tom7 at cs.cmu.edu Sun Jul 23 14:08:43 2006 From: tom7 at cs.cmu.edu (Tom Murphy) Date: Sun Jul 23 14:08:44 2006 Subject: [icfp-discuss] DIV error In-Reply-To: <20060723175614.58713.qmail@web50715.mail.yahoo.com> Message-ID: Sure, I think it's worth continuing! I think you can do a lot in the day, and I think that the problems are really fun to solve even if you don't win. (But if it isn't fun, don't torture yourself!) Tom > Thanks Tom. > Is it worth to continue now that many have already submitted and in > lead or the contest is of the nature where there is still a chance for > the latecomers ? Just wondering :( > > Andy > > > > > --- Tom Murphy wrote: > > > > > Also, make sure that you are doing unsigned division. > > At this point it should be possible to single-step through the > > instructions and check that they are doing the correct thing... > > > > Tom > > > > > Possibly truncating/rounding issue? > > > > > > --David > > > > > > On 7/23/06, Andy wrote: > > > > Hi All, > > > > > > > > I completed coding the UM and tried running it and am getting the message "DIV error". I > > checked > > > > to make sure that there is div by zero or negaivtive value conditions ...anybody had this > > problem > > > > ? > > > > > > > > Thanks, > > > > Andy > > > > > > > > [ NEW! : http://tom7.org/ ] > > [ OLD! : http://fonts.tom7.com/ ] > > > > _______________________________________________ > > icfpcontest-discuss mailing list > > icfpcontest-discuss@lists.andrew.cmu.edu > > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > > > > > __________________________________________________ > Do You Yahoo!? > Tired of spam? Yahoo! Mail has the best spam protection around > http://mail.yahoo.com > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > > [ NEW! : http://tom7.org/ ] [ OLD! : http://fonts.tom7.com/ ] From kurt.rinnert at cern.ch Sun Jul 23 14:12:54 2006 From: kurt.rinnert at cern.ch (Kurt Rinnert) Date: Sun Jul 23 14:12:57 2006 Subject: [icfp-discuss] access denied for user In-Reply-To: <200607232012.54103.micha-1@fantasymail.de> References: <200607232012.54103.micha-1@fantasymail.de> Message-ID: <20060723181254.GA10442@vulcan> Hi, 'guest ' without password should do the job. Maybe you have a bug in your input operator? Or somewhere else... best, Kurt On Sun, Jul 23, 2006 at 08:12:53PM +0200, Michael Wohlwend wrote: > Hi, > > I allways get the above message, when trying to log into UMIX. > Is this part of the game? I tried 'guest' with and without password, > teamname... nothing works :-( > So if the first task is to find the right login... fine :-) if not, can > someone give me a hint? > > cheers > Michael > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss -- Kurt Rinnert, University of Liverpool - Department of Physics From leenstrab at comcast.net Sun Jul 23 14:13:20 2006 From: leenstrab at comcast.net (Bruce Leenstra) Date: Sun Jul 23 14:13:23 2006 Subject: [icfp-discuss] access denied for user Message-ID: <024f01c6ae83$ae0ffcf0$6501a8c0@had1.or.comcast.net> I am at the same place. It was giving a mem fault, but I fixed that. It's *always* prompted for a *guest* password. Help! -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060723/12472d72/attachment.html From andy_zyx at yahoo.com Sun Jul 23 14:16:49 2006 From: andy_zyx at yahoo.com (Andy) Date: Sun Jul 23 14:16:51 2006 Subject: [icfp-discuss] DIV error In-Reply-To: <20060723175614.58713.qmail@web50715.mail.yahoo.com> Message-ID: <20060723181649.18542.qmail@web50713.mail.yahoo.com> Its surely fun and a great contest organised by you guys....thats the reason I participated...but too bad I was not able to get it running atleast by y'day...but I guess would continue to see what unravels....:) --- Andy wrote: > Thanks Tom. > Is it worth to continue now that many have already submitted and in lead or the contest is of > the > nature where there is still a chance for the latecomers ? Just wondering :( > > Andy > > > > > --- Tom Murphy wrote: > > > > > Also, make sure that you are doing unsigned division. > > At this point it should be possible to single-step through the > > instructions and check that they are doing the correct thing... > > > > Tom > > > > > Possibly truncating/rounding issue? > > > > > > --David > > > > > > On 7/23/06, Andy wrote: > > > > Hi All, > > > > > > > > I completed coding the UM and tried running it and am getting the message "DIV error". I > > checked > > > > to make sure that there is div by zero or negaivtive value conditions ...anybody had this > > problem > > > > ? > > > > > > > > Thanks, > > > > Andy > > > > > > > > [ NEW! : http://tom7.org/ ] > > [ OLD! : http://fonts.tom7.com/ ] > > > > _______________________________________________ > > icfpcontest-discuss mailing list > > icfpcontest-discuss@lists.andrew.cmu.edu > > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > > > > > __________________________________________________ > Do You Yahoo!? > Tired of spam? Yahoo! Mail has the best spam protection around > http://mail.yahoo.com > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com From bluelive at xs4all.nl Sun Jul 23 14:20:08 2006 From: bluelive at xs4all.nl (Bart van der Werf) Date: Sun Jul 23 14:20:11 2006 Subject: [icfp-discuss] access denied for user In-Reply-To: <024f01c6ae83$ae0ffcf0$6501a8c0@had1.or.comcast.net> Message-ID: <001401c6ae84$a1374d70$4101a8c0@andromeda> you probably have a space or somethign other then -1 following it -----Original Message----- From: icfpcontest-discuss-bounces@lists.andrew.cmu.edu [mailto:icfpcontest-discuss-bounces@lists.andrew.cmu.edu] On Behalf Of Bruce Leenstra Sent: Sunday, July 23, 2006 8:13 PM To: icfpcontest-discuss@lists.andrew.cmu.edu Subject: [icfp-discuss] access denied for user I am at the same place. It was giving a mem fault, but I fixed that. It's *always* prompted for a *guest* password. Help! -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060723/e6f537b2/attachment.html From samuel at acid-code.ch Sun Jul 23 14:34:04 2006 From: samuel at acid-code.ch (Samuel Burri) Date: Sun Jul 23 14:32:56 2006 Subject: [icfp-discuss] access denied for user In-Reply-To: <001401c6ae84$a1374d70$4101a8c0@andromeda> References: <024f01c6ae83$ae0ffcf0$6501a8c0@had1.or.comcast.net> <001401c6ae84$a1374d70$4101a8c0@andromeda> Message-ID: <56018.85.2.216.161.1153679644.squirrel@serioussam.selfip.org> you have to supply guest + 0xA or 10 (ten) in decimal every line you send must be terminated with 0xA > you probably have a space or somethign other then -1 following it > > -----Original Message----- > From: icfpcontest-discuss-bounces@lists.andrew.cmu.edu > [mailto:icfpcontest-discuss-bounces@lists.andrew.cmu.edu] On Behalf Of > Bruce Leenstra > Sent: Sunday, July 23, 2006 8:13 PM > To: icfpcontest-discuss@lists.andrew.cmu.edu > Subject: [icfp-discuss] access denied for user > > > I am at the same place. It was giving a mem fault, but I fixed that. > It's *always* prompted for a *guest* password. > > Help! > > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > From ij7b3u502 at sneakemail.com Sun Jul 23 14:34:50 2006 From: ij7b3u502 at sneakemail.com (vbzoli) Date: Sun Jul 23 14:34:52 2006 Subject: [icfp-discuss] Who's got the fastest sandmark? Message-ID: <1307-89755@sneakemail.com> My results... on a 1.5GHz Celeron M laptop: real 1m19.109s user 1m15.105s sys 0m0.340s on an Opteron 244 box: real 1m41.522s user 1m38.178s sys 0m0.704s Simple C implementation with calloc/malloc/free. Circular buffer used for free array indices. Compiled with gcc -O6. And linux, of course. :o) -- vbzoli From popiel at wolfskeep.com Sun Jul 23 14:44:25 2006 From: popiel at wolfskeep.com (T. Alexander Popiel) Date: Sun Jul 23 14:44:27 2006 Subject: [icfp-discuss] sandmark: testing aliasing.. In-Reply-To: <44C368B4.7090708@softlab.ntua.gr> References: <44C35F2D.1020308@softlab.ntua.gr> <44C366A9.1090709@cs.cmu.edu> <44C368B4.7090708@softlab.ntua.gr> Message-ID: <20060723184425.2E2222DDCC@cashew.wolfskeep.com> In message: <44C368B4.7090708@softlab.ntua.gr> "Nikolaos S. Papaspyrou" writes: > >By the way, I find it a little strange that we're seeing ants and >adventure games and everything else, and still the sandmark fails >because of a bug in our UM. That just shows that the majority of UMIX isn't relying on the copy behaviour. - Alex From 0zu31 at gmx.de Sun Jul 23 14:45:14 2006 From: 0zu31 at gmx.de (Mechthild 'Rejis' Czapp) Date: Sun Jul 23 14:46:22 2006 Subject: [icfp-discuss] "The Universal machine failed" Message-ID: <200607232045.14870.0zu31@gmx.de> Any hints on what this message signifies? It appears when running the dumped UM-program. Chanceless in Cologne, Mechthild From micha-1 at fantasymail.de Sun Jul 23 14:51:49 2006 From: micha-1 at fantasymail.de (Michael Wohlwend) Date: Sun Jul 23 14:47:02 2006 Subject: [icfp-discuss] access denied for user In-Reply-To: <56018.85.2.216.161.1153679644.squirrel@serioussam.selfip.org> References: <024f01c6ae83$ae0ffcf0$6501a8c0@had1.or.comcast.net> <001401c6ae84$a1374d70$4101a8c0@andromeda> <56018.85.2.216.161.1153679644.squirrel@serioussam.selfip.org> Message-ID: <200607232051.49629.micha-1@fantasymail.de> Am Sonntag, 23. Juli 2006 20:34 schrieb Samuel Burri: > you have to supply guest + 0xA or 10 (ten) in decimal > every line you send must be terminated with 0xA that was it, thanks. I thought a (-1) should be at the end *grrr* :-) cheers, Michael From bartoschek at gmx.de Sun Jul 23 15:28:11 2006 From: bartoschek at gmx.de (Christoph Bartoschek) Date: Sun Jul 23 15:28:17 2006 Subject: [icfp-discuss] Raytrace.spec Message-ID: <200607232128.12373.bartoschek@gmx.de> Hi, the raytrace spec says that: Medium = Inr (Inl ()) But when I write: *========================* !send [(Inr (Inl ()), E)]! *========================* I get: [2D] Parse Error : A command didn't parse - the string send [(Inr (Inl ()), E)] is not a valid command. but when I write: *======================* !send [(Inr Inl (), E)]! *======================* the command succeddes. Who is correct? Christoph From crary at cs.cmu.edu Sun Jul 23 15:33:06 2006 From: crary at cs.cmu.edu (Karl Crary) Date: Sun Jul 23 15:33:06 2006 Subject: [icfp-discuss] Raytrace.spec In-Reply-To: <200607232128.12373.bartoschek@gmx.de> References: <200607232128.12373.bartoschek@gmx.de> Message-ID: <44C3CEF2.2090001@cs.cmu.edu> The error is in the spec. This is one of the errors corrected in the new codex; I'd recommend upgrading. Karl Crary Christoph Bartoschek wrote: > Hi, > > the raytrace spec says that: > > Medium = Inr (Inl ()) > > But when I write: > > *========================* > !send [(Inr (Inl ()), E)]! > *========================* > > I get: > > [2D] Parse Error : A command didn't parse - the string > send [(Inr (Inl ()), E)] > is not a valid command. > > > but when I write: > > *======================* > !send [(Inr Inl (), E)]! > *======================* > > the command succeddes. > > Who is correct? > > Christoph > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > > From z81 at backpath.croco.net Sun Jul 23 16:19:53 2006 From: z81 at backpath.croco.net (Andrey V. Stolyarov) Date: Sun Jul 23 16:18:15 2006 Subject: [icfp-discuss] basic In-Reply-To: <44C3CEF2.2090001@cs.cmu.edu> References: <200607232128.12373.bartoschek@gmx.de> <44C3CEF2.2090001@cs.cmu.edu> Message-ID: Hmmm, all the other puzzles are at least visible in the game, but I'm totally unable to understand what to do with BASIC in order to turn it into a puzzle. I've tried to run several programs with that /bin/qbasic, tried to drop it into some troubles (but it successfully shows respective error messages), needless to say I've run the hack.bas program (after removig all the crap nearby its end) and successfully hacked into howie and ohmega, but.. hmmm.. I still never seen the BASIC.XXX@ or whatever. Is it a part of the game to discover where this thing is hiding from me (if so, I'll keep trying) or is it another bug in my UM? -- Croco From bartoschek at gmx.de Sun Jul 23 16:29:22 2006 From: bartoschek at gmx.de (Christoph Bartoschek) Date: Sun Jul 23 16:29:29 2006 Subject: [icfp-discuss] Raytrace.spec In-Reply-To: <44C3CEF2.2090001@cs.cmu.edu> References: <200607232128.12373.bartoschek@gmx.de> <44C3CEF2.2090001@cs.cmu.edu> Message-ID: <200607232229.22471.bartoschek@gmx.de> Am Sonntag, 23. Juli 2006 21:33 schrieb Karl Crary: > The error is in the spec. This is one of the errors corrected in the > new codex; I'd recommend upgrading. I think I use the new codex. Christoph From nattfodd at gmail.com Sun Jul 23 16:32:18 2006 From: nattfodd at gmail.com (Alexandre Buisse) Date: Sun Jul 23 16:32:21 2006 Subject: [icfp-discuss] basic In-Reply-To: References: <200607232128.12373.bartoschek@gmx.de> <44C3CEF2.2090001@cs.cmu.edu> Message-ID: <41b037ed0607231332m53d20249sfbee3e81b7676e6e@mail.gmail.com> On 7/23/06, Andrey V. Stolyarov wrote: > > > > Hmmm, > > all the other puzzles are at least visible in the game, but I'm > totally unable to understand what to do with BASIC in order to > turn it into a puzzle. I've tried to run several programs with > that /bin/qbasic, tried to drop it into some troubles (but it > successfully shows respective error messages), needless to say > I've run the hack.bas program (after removig all the crap nearby > its end) and successfully hacked into howie and ohmega, but.. > hmmm.. I still never seen the BASIC.XXX@ or whatever. > > Is it a part of the game to discover where this thing is hiding > from me (if so, I'll keep trying) or is it another bug in my UM? It's definitely not a bug. /Alexandre -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060723/a32cb4a8/attachment.html From plakal at gmail.com Sun Jul 23 16:34:12 2006 From: plakal at gmail.com (Manoj Plakal) Date: Sun Jul 23 16:34:14 2006 Subject: [icfp-discuss] basic In-Reply-To: References: <200607232128.12373.bartoschek@gmx.de> <44C3CEF2.2090001@cs.cmu.edu> Message-ID: <5230c2820607231334u52313086ie7d6292e57da3514@mail.gmail.com> read the comments in the original basic program ... On 7/23/06, Andrey V. Stolyarov wrote: > > > Hmmm, > > all the other puzzles are at least visible in the game, but I'm > totally unable to understand what to do with BASIC in order to > turn it into a puzzle. I've tried to run several programs with > that /bin/qbasic, tried to drop it into some troubles (but it > successfully shows respective error messages), needless to say > I've run the hack.bas program (after removig all the crap nearby > its end) and successfully hacked into howie and ohmega, but.. > hmmm.. I still never seen the BASIC.XXX@ or whatever. > > Is it a part of the game to discover where this thing is hiding > from me (if so, I'll keep trying) or is it another bug in my UM? > > > > -- > Croco > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > From kurt.rinnert at cern.ch Sun Jul 23 16:35:27 2006 From: kurt.rinnert at cern.ch (Kurt Rinnert) Date: Sun Jul 23 16:35:33 2006 Subject: [icfp-discuss] basic In-Reply-To: References: <200607232128.12373.bartoschek@gmx.de> <44C3CEF2.2090001@cs.cmu.edu> Message-ID: <20060723203527.GB10442@vulcan> Oh come on! Use the source! On Mon, Jul 24, 2006 at 12:19:53AM +0400, Andrey V. Stolyarov wrote: > > > Hmmm, > > all the other puzzles are at least visible in the game, but I'm > totally unable to understand what to do with BASIC in order to > turn it into a puzzle. I've tried to run several programs with > that /bin/qbasic, tried to drop it into some troubles (but it > successfully shows respective error messages), needless to say > I've run the hack.bas program (after removig all the crap nearby > its end) and successfully hacked into howie and ohmega, but.. > hmmm.. I still never seen the BASIC.XXX@ or whatever. > > Is it a part of the game to discover where this thing is hiding > from me (if so, I'll keep trying) or is it another bug in my UM? > > > > -- > Croco > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss -- Kurt Rinnert, University of Liverpool - Department of Physics From drl at cs.cmu.edu Sun Jul 23 16:37:10 2006 From: drl at cs.cmu.edu (Dan Licata) Date: Sun Jul 23 16:37:45 2006 Subject: [icfp-discuss] Day 3! Message-ID: <20060723203710.GA26581@cs.cmu.edu> Now that we're in the third day of the contest, we have a few announcements for you: - Congratulations to our current leaders, Team Smartass, kuma-, and PLOP. We are happy to report that over 300 teams have now completed the UM, many of them in the past 12 hours. We're very impressed with everyone's hard work! - We've posted a couple of new transcription errors in the balance problem (user id 'yang') to the FAQ. http://www.icfpcontest.org/rulesfaq.shtml - Remember that the scoreboard will not be updated after 4:00 AM (EDT). - We will make an announcement later today about how to submit code and commentary for the Judges' prize. The Judges' prize is awarded at our discretion for the coolest, most interesting bit of programming---for example, clever unexpected solutions to puzzles, an alarmingly fast UM implementation, etc. -Dan From geepokey at yahoo.com Sun Jul 23 17:26:13 2006 From: geepokey at yahoo.com (Gee Pokey) Date: Sun Jul 23 17:26:16 2006 Subject: [icfp-discuss] Who's got the fastest sandmark? In-Reply-To: <1307-89755@sneakemail.com> Message-ID: <20060723212613.92447.qmail@web37212.mail.mud.yahoo.com> On my AMD64 3800+ (2.4GHZ) - 2:28.37 Minutes C:\icfp06>\cygwin\bin\time icfp_j8.exe > run.txt 0.01user 0.00system 2:28.37elapsed 0%CPU (0avgtext+0avgdata 97024maxresident)k Simple C implementation with calloc/malloc/free. That Celeron number looks pretty good in comparison. Why do you think your Opteron 244 slower than the Celeron ? -Gumby of Canivsar ----- Original Message ---- From: vbzoli To: icfpcontest-discuss@lists.andrew.cmu.edu Sent: Sunday, July 23, 2006 11:34:50 AM Subject: Re: [icfp-discuss] Who's got the fastest sandmark? My results... on a 1.5GHz Celeron M laptop: real 1m19.109s user 1m15.105s sys 0m0.340s on an Opteron 244 box: real 1m41.522s user 1m38.178s sys 0m0.704s Simple C implementation with calloc/malloc/free. Circular buffer used for free array indices. Compiled with gcc -O6. And linux, of course. :o) -- vbzoli _______________________________________________ icfpcontest-discuss mailing list icfpcontest-discuss@lists.andrew.cmu.edu https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss From phallien at yahoo.com Sun Jul 23 17:25:30 2006 From: phallien at yahoo.com (P. Hallien) Date: Sun Jul 23 17:29:35 2006 Subject: [icfp-discuss] RE: UMIX Password Cracking Message-ID: <20060723212530.50332.qmail@web35514.mail.mud.yahoo.com> [ Warning, Spoilers May Follow ] Please help me maintain my sanity! I'm at my wits-end on this. :) Here's my patch to the hack.bas routine to append two digit decimal numbers DD to the dictionary words. As far as I can tell from adding PRINT statements, it's doing the right thing, but fails to find a valid password for either 'gardener' or 'bbarker'. Am I missing something blindingly obvious, or is access to the two above accounts gained in a different way? [ I've found passwords for all other accounts (ohmega, howie->yang,hmonk) ] Thanks! P. Hallien ... DXL REM DXLV DIM j AS INTEGER DL DIM k AS INTEGER DLV i = I DLX j = XLVIII DLXV k = XLVIII DLXX IF CHECKPASS(username, words(i) + CHR(j) + CHR(k)) THEN GOTO DCXXXV DLXXV k = k + I DLXXX IF k > LVII THEN GOTO DXC DLXXXV GOTO DLXX DXC j = j + I DXCV IF j > LVII THEN GOTO DCV DC GOTO DLXV DCV PRINT "." DCX i = i + I DCXV IF i > pwdcount THEN GOTO DCXXV DCXX GOTO DLX DCXXV PRINT "No Match for User: " + username + CHR(X) DCXXX END DCXXXV PRINT "Match Found! for User: " + username + CHR(X) DCXL PRINT "Password: " + words(i) + CHR(j) + CHR(k) + CHR(X) DCXLV END ============= RE: Tom Murphy > Only a few users have insecure passwords. You should try to work on the other puzzles in order to unlock more accounts... Samuel Burri > Plus you have to append decimal digits eg. asciicodes 48 through 57 __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com From drl at cs.cmu.edu Sun Jul 23 17:33:28 2006 From: drl at cs.cmu.edu (Dan Licata) Date: Sun Jul 23 17:34:04 2006 Subject: [icfp-discuss] RE: UMIX Password Cracking In-Reply-To: <20060723212530.50332.qmail@web35514.mail.mud.yahoo.com> References: <20060723212530.50332.qmail@web35514.mail.mud.yahoo.com> Message-ID: <20060723213328.GC26581@cs.cmu.edu> First of all, please do not post problem solutions to this discussion list. However, to answer your question, not all accounts can be hacked by even the extended cracker. -Dan On Jul23, P. Hallien wrote: > [ Warning, Spoilers May Follow ] > > > Please help me maintain my sanity! I'm at my wits-end on this. :) > > > Here's my patch to the hack.bas routine to append two digit decimal numbers DD to the dictionary > words. > > As far as I can tell from adding PRINT statements, it's doing the right thing, but fails to find a > valid password for either 'gardener' or 'bbarker'. > > Am I missing something blindingly obvious, or is access to the two above accounts gained in a > different way? > > > Thanks! > P. Hallien > > ============= > > RE: > > Tom Murphy > Only a few users have insecure passwords. You should try to work on the > other puzzles in order to unlock more accounts... > > Samuel Burri > Plus you have to append decimal digits eg. asciicodes 48 through 57 > > > > __________________________________________________ > Do You Yahoo!? > Tired of spam? Yahoo! Mail has the best spam protection around > http://mail.yahoo.com > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > From papierfalter at googlemail.com Sun Jul 23 17:41:34 2006 From: papierfalter at googlemail.com (Wolfram Fischer) Date: Sun Jul 23 17:41:12 2006 Subject: [icfp-discuss] guest problems Message-ID: <44C3ED0E.6020402@googlemail.com> I've still problems logging in as guest. I'm always asked for a password, regardless of me providing a '\n' (== 0xA) to the um. So, my question is: What is valid input for the um? For example, is 'g' 'u' 'e' 's' 't' '-1' valid (-1==all bits high) or must I add 0xA somewhere in the sequence or am I totally wrong??? Thank you very much in advance and with best regards, Wolf From windenntw at gmail.com Sun Jul 23 17:49:09 2006 From: windenntw at gmail.com (Antonio Vargas) Date: Sun Jul 23 17:49:13 2006 Subject: [icfp-discuss] guest problems In-Reply-To: <44C3ED0E.6020402@googlemail.com> References: <44C3ED0E.6020402@googlemail.com> Message-ID: <69304d110607231449x30fbce26jc781c71e051d6165@mail.gmail.com> Please note that -1 == EOF, which the UMIX uses later on for logout for example. So, what you need to enter is g u e s t ENTER, at least that's what I type here and it works. ps. which language are you using? ps2. it did make a diference for me to explicitely use unbuffered io... On 7/23/06, Wolfram Fischer wrote: > I've still problems logging in as guest. > I'm always asked for a password, regardless of me providing a '\n' (== > 0xA) to the um. > > So, my question is: What is valid input for the um? > > For example, is 'g' 'u' 'e' 's' 't' '-1' valid (-1==all bits high) or > must I add 0xA somewhere in the sequence or am I totally wrong??? > > Thank you very much in advance and with best regards, > Wolf > > > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > -- Greetz, Antonio Vargas aka winden of network http://network.amigascne.org/ windNOenSPAMntw@gmail.com thesameasabove@amigascne.org Every day, every year you have to work you have to study you have to scene. From scsibug at imap.cc Sun Jul 23 17:49:19 2006 From: scsibug at imap.cc (Greg Heartsfield) Date: Sun Jul 23 17:49:30 2006 Subject: [icfp-discuss] guest problems In-Reply-To: <44C3ED0E.6020402@googlemail.com> References: <44C3ED0E.6020402@googlemail.com> Message-ID: 'g' 'u' 'e' 's' 't' '-1' Is not valid input. You must provide "guest\n" (0xA), and nothing else. Programmatically, you should be able to just read chars from stdin, and pass those along unchanged. You don't even need to implement the -1 end-of-input special case in order to get login working. Thanks, Greg Heartsfield On Jul 23, 2006, at 4:41 PM, Wolfram Fischer wrote: > I've still problems logging in as guest. > I'm always asked for a password, regardless of me providing a '\n' (== > 0xA) to the um. > > So, my question is: What is valid input for the um? > > For example, is 'g' 'u' 'e' 's' 't' '-1' valid (-1==all bits high) or > must I add 0xA somewhere in the sequence or am I totally wrong??? > > Thank you very much in advance and with best regards, > Wolf > > > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss -------------- next part -------------- A non-text attachment was scrubbed... Name: PGP.sig Type: application/pgp-signature Size: 186 bytes Desc: This is a digitally signed message part Url : http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060723/8ec18657/PGP.bin From papierfalter at googlemail.com Sun Jul 23 18:10:35 2006 From: papierfalter at googlemail.com (Wolfram Fischer) Date: Sun Jul 23 18:11:20 2006 Subject: [icfp-discuss] guest problems In-Reply-To: <69304d110607231449x30fbce26jc781c71e051d6165@mail.gmail.com> References: <44C3ED0E.6020402@googlemail.com> <69304d110607231449x30fbce26jc781c71e051d6165@mail.gmail.com> Message-ID: <44C3F3DB.6080204@googlemail.com> Thanks to everyone for the hints! I made the mistake and thought (without veryfieng) '\n'=='0xA', which is wrong. Now I'm a big step further - my um segfaults ^^. I'm using ANSI C '89 this year (last year Java, the year before Delphi/C -- next year I'll try to finally use a language with some kind of lambda-calculus ;-) How many bytes must the initial dump have? (is 15924059 Bytes the correct length?) Best regards, Wolf Antonio Vargas schrieb: > Please note that -1 == EOF, which the UMIX uses later on for logout > for example. So, what you need to enter is g u e s t ENTER, at least > that's what I type here and it works. > > ps. which language are you using? > > ps2. it did make a diference for me to explicitely use unbuffered io... > > > On 7/23/06, Wolfram Fischer wrote: >> I've still problems logging in as guest. >> I'm always asked for a password, regardless of me providing a '\n' (== >> 0xA) to the um. >> >> So, my question is: What is valid input for the um? >> >> For example, is 'g' 'u' 'e' 's' 't' '-1' valid (-1==all bits high) or >> must I add 0xA somewhere in the sequence or am I totally wrong??? >> >> Thank you very much in advance and with best regards, >> Wolf >> >> >> _______________________________________________ >> icfpcontest-discuss mailing list >> icfpcontest-discuss@lists.andrew.cmu.edu >> https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss >> > > From micha-1 at fantasymail.de Sun Jul 23 18:20:51 2006 From: micha-1 at fantasymail.de (Michael Wohlwend) Date: Sun Jul 23 18:14:43 2006 Subject: [icfp-discuss] guest problems In-Reply-To: <44C3F3DB.6080204@googlemail.com> References: <44C3ED0E.6020402@googlemail.com> <69304d110607231449x30fbce26jc781c71e051d6165@mail.gmail.com> <44C3F3DB.6080204@googlemail.com> Message-ID: <200607240020.52266.micha-1@fantasymail.de> Am Montag, 24. Juli 2006 00:10 schrieb Wolfram Fischer: > Thanks to everyone for the hints! > I'm using ANSI C '89 this year (last year Java, the year before Delphi/C > -- next year I'll try to finally use a language with some kind of > lambda-calculus ;-) OCaml :-) > How many bytes must the initial dump have? (is 15924059 Bytes the > correct length?) I have 15923865 byte. cheers, Michael From the.michael.e at gmail.com Sun Jul 23 18:57:51 2006 From: the.michael.e at gmail.com (Michael E.) Date: Sun Jul 23 18:57:53 2006 Subject: [icfp-discuss] guest problems In-Reply-To: <200607240020.52266.micha-1@fantasymail.de> References: <44C3ED0E.6020402@googlemail.com> <69304d110607231449x30fbce26jc781c71e051d6165@mail.gmail.com> <44C3F3DB.6080204@googlemail.com> <200607240020.52266.micha-1@fantasymail.de> Message-ID: <83d675d20607231557u36ceecc6xadb314427afef657@mail.gmail.com> > > How many bytes must the initial dump have? (is 15924059 Bytes the > > correct length?) > I have 15923865 byte. 15.901.920 byte. br, Michael Etscheid From jonathan.roewen at gmail.com Sun Jul 23 17:46:48 2006 From: jonathan.roewen at gmail.com (Jonathan Roewen) Date: Sun Jul 23 19:00:31 2006 Subject: [icfp-discuss] guest problems In-Reply-To: <44C3ED0E.6020402@googlemail.com> Message-ID: <37106724187-BeMail@beos> > I've still problems logging in as guest. > I'm always asked for a password, regardless of me providing a '\n' (= > = > 0xA) to the um. > > So, my question is: What is valid input for the um? The valid input is 'g' 'u' 'e' 's' 't' '\n' From shawnyar217 at yahoo.com Sun Jul 23 19:18:11 2006 From: shawnyar217 at yahoo.com (Shawn Yarbrough) Date: Sun Jul 23 19:18:13 2006 Subject: [icfp-discuss] Who's got the fastest sandmark? (fast C++) In-Reply-To: <1307-89755@sneakemail.com> Message-ID: <20060723231811.33326.qmail@web53414.mail.yahoo.com> What's your SuperPI[2^20] time? On my Athlon XP 2700+ with 1 GB memory my SuperPI[2^20] is 49 seconds. My UM implemention's SANDmark finishes in: real 2m18.729s user 2m13.576s sys 0m0.872s I think my UM speed rating then is: 139/49 = 2.831 This is on Ubuntu Linux written in C++ (GNU g++) with portable, clean OO code organized neatly into classes and functions. Shawn vbzoli wrote: My results... on a 1.5GHz Celeron M laptop: real 1m19.109s user 1m15.105s sys 0m0.340s on an Opteron 244 box: real 1m41.522s user 1m38.178s sys 0m0.704s Simple C implementation with calloc/malloc/free. Circular buffer used for free array indices. Compiled with gcc -O6. And linux, of course. :o) -- vbzoli --------------------------------- Talk is cheap. Use Yahoo! Messenger to make PC-to-Phone calls. Great rates starting at 1?/min. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060723/3a40b4b7/attachment.html From tom7 at cs.cmu.edu Sun Jul 23 19:27:29 2006 From: tom7 at cs.cmu.edu (Tom Murphy) Date: Sun Jul 23 19:27:31 2006 Subject: [icfp-discuss] guest problems In-Reply-To: <83d675d20607231557u36ceecc6xadb314427afef657@mail.gmail.com> Message-ID: 15901920 is the size of the codex volume id 7 output. With this you will experience a few minor bugs and your scores from the gardener user account will not be counted--you should download the new codex (volume id 8), whose output has size 15923864. Tom > > > How many bytes must the initial dump have? (is 15924059 Bytes the > > > correct length?) > > I have 15923865 byte. > 15.901.920 byte. > > > br, > > Michael Etscheid > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > > [ NEW! : http://tom7.org/ ] [ OLD! : http://fonts.tom7.com/ ] From druffer at worldnet.att.net Sun Jul 23 20:57:36 2006 From: druffer at worldnet.att.net (Dennis Ruffer) Date: Sun Jul 23 20:57:40 2006 Subject: [icfp-discuss] Day 3! In-Reply-To: <20060723203710.GA26581@cs.cmu.edu> References: <20060723203710.GA26581@cs.cmu.edu> Message-ID: As this year's contest winds down and I have yet to get to the password prompt, I have to ask a basic question. Is it possible that a specific decryption key would cause a umix image to be generated that would require more memory than I can get? This breaks down to 2 questions. 1) Does the umix memory requirements vary, based upon decryption key? 2) What is the upper limits on how much memory it will request? I now have my UM running with virtual RAM mapped to a 2 Gb disk file. This is extremely slow but I am hoping it does something, other than crashing, before 4am. However, in case it doesn't, I wanted to get these questions registered before the contest closes. I have even tried my decryption key on code from another author. He's successfully gathering points, but my umix generated by his code still crashes. DaR From kurt.rinnert at cern.ch Sun Jul 23 21:11:09 2006 From: kurt.rinnert at cern.ch (Kurt Rinnert) Date: Sun Jul 23 21:11:17 2006 Subject: [icfp-discuss] Day 3! In-Reply-To: References: <20060723203710.GA26581@cs.cmu.edu> Message-ID: <20060724011109.GC10442@vulcan> As a test, why not try to register again with a different team name etc. and check whether the new key also causes problems? However, I doubt that a specific key causes memory exhaustion, considering the very good overall design of the whole thing. Still not impossible, though. Maybe your point gathering friend simply has more RAM... - Kurt On Sun, Jul 23, 2006 at 08:57:36PM -0400, Dennis Ruffer wrote: > As this year's contest winds down and I have yet to get to the > password prompt, I have to ask a basic question. > > Is it possible that a specific decryption key would cause a umix > image to be generated that would require more memory than I can get? > > This breaks down to 2 questions. > > 1) Does the umix memory requirements vary, based upon decryption key? > 2) What is the upper limits on how much memory it will request? > > I now have my UM running with virtual RAM mapped to a 2 Gb disk > file. This is extremely slow but I am hoping it does something, > other than crashing, before 4am. However, in case it doesn't, I > wanted to get these questions registered before the contest closes. > > I have even tried my decryption key on code from another author. > He's successfully gathering points, but my umix generated by his code > still crashes. > > DaR > > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss -- Kurt Rinnert, University of Liverpool - Department of Physics From rvandam00 at gmail.com Sun Jul 23 21:16:12 2006 From: rvandam00 at gmail.com (Robert Van Dam) Date: Sun Jul 23 21:16:19 2006 Subject: [icfp-discuss] DIV error In-Reply-To: <20060723181649.18542.qmail@web50713.mail.yahoo.com> References: <20060723175614.58713.qmail@web50715.mail.yahoo.com> <20060723181649.18542.qmail@web50713.mail.yahoo.com> Message-ID: <1815f00607231816v7081c683ha5c516ebf37f2fb5@mail.gmail.com> Did you ever get that DIV error fixed? We've reimplemented in C since our previous attempt in a higher level language couldn't handle all the memory thrashing and the sandmark gives a DIV error that I can't trace down. The codex tells me wrong key which I encountered before and discovered a math error on my part but after having checked all the math routines, everything looks clean. On 7/23/06, Andy wrote: > Its surely fun and a great contest organised by you guys....thats the reason I participated...but > too bad I was not able to get it running atleast by y'day...but I guess would continue to see what > unravels....:) > > > > > --- Andy wrote: > > > Thanks Tom. > > Is it worth to continue now that many have already submitted and in lead or the contest is of > > the > > nature where there is still a chance for the latecomers ? Just wondering :( > > > > Andy > > > > > > > > > > --- Tom Murphy wrote: > > > > > > > > Also, make sure that you are doing unsigned division. > > > At this point it should be possible to single-step through the > > > instructions and check that they are doing the correct thing... > > > > > > Tom > > > > > > > Possibly truncating/rounding issue? > > > > > > > > --David > > > > > > > > On 7/23/06, Andy wrote: > > > > > Hi All, > > > > > > > > > > I completed coding the UM and tried running it and am getting the message "DIV error". I > > > checked > > > > > to make sure that there is div by zero or negaivtive value conditions ...anybody had this > > > problem > > > > > ? > > > > > > > > > > Thanks, > > > > > Andy > > > > > > > > > > > [ NEW! : http://tom7.org/ ] > > > [ OLD! : http://fonts.tom7.com/ ] > > > > > > _______________________________________________ > > > icfpcontest-discuss mailing list > > > icfpcontest-discuss@lists.andrew.cmu.edu > > > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > > > > > > > > > __________________________________________________ > > Do You Yahoo!? > > Tired of spam? Yahoo! Mail has the best spam protection around > > http://mail.yahoo.com > > _______________________________________________ > > icfpcontest-discuss mailing list > > icfpcontest-discuss@lists.andrew.cmu.edu > > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > > > > > __________________________________________________ > Do You Yahoo!? > Tired of spam? Yahoo! Mail has the best spam protection around > http://mail.yahoo.com > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > From drl at cs.cmu.edu Sun Jul 23 21:16:50 2006 From: drl at cs.cmu.edu (Dan Licata) Date: Sun Jul 23 21:16:52 2006 Subject: [icfp-discuss] Day 3! In-Reply-To: References: <20060723203710.GA26581@cs.cmu.edu> Message-ID: <20060724011650.GA27422@cs.cmu.edu> On Jul23, Dennis Ruffer wrote: > As this year's contest winds down and I have yet to get to the > password prompt, I have to ask a basic question. > > Is it possible that a specific decryption key would cause a umix > image to be generated that would require more memory than I can get? > > This breaks down to 2 questions. > > 1) Does the umix memory requirements vary, based upon decryption key? Definitely not. It executes the same instructions no matter what. The values of registers and memory might be different, but there is no control dependency. > 2) What is the upper limits on how much memory it will request? SANDmark allocates a total of 1597336168 bytes, and there is at most 1227412 live at any time. > I have even tried my decryption key on code from another author. > He's successfully gathering points, but my umix generated by his code > still crashes. Your friend probably has a bug. By the way, teams aren't supposed to collaborate. -Dan From druffer at worldnet.att.net Sun Jul 23 21:21:51 2006 From: druffer at worldnet.att.net (Dennis Ruffer) Date: Sun Jul 23 21:21:57 2006 Subject: [icfp-discuss] Day 3! In-Reply-To: <20060724011109.GC10442@vulcan> References: <20060723203710.GA26581@cs.cmu.edu> <20060724011109.GC10442@vulcan> Message-ID: Both of us are running OSX, which I believe has a 2 Gb limit per thread, but I also tried it on a system with 8 Gb installed. Getting a new key is something I hadn't thought of, but my other run will tell me for sure how much it needs. Thanks for the suggestion though. DaR On Jul 23, 2006, at 9:11 PM, Kurt Rinnert wrote: > As a test, why not try to register again with a different team name > etc. > and check whether the new key also causes problems? > > However, I doubt that a specific key causes memory exhaustion, > considering the very good overall design of the whole thing. > Still not impossible, though. > > Maybe your point gathering friend simply has more RAM... > > - Kurt > > On Sun, Jul 23, 2006 at 08:57:36PM -0400, Dennis Ruffer wrote: >> As this year's contest winds down and I have yet to get to the >> password prompt, I have to ask a basic question. >> >> Is it possible that a specific decryption key would cause a umix >> image to be generated that would require more memory than I can get? >> >> This breaks down to 2 questions. >> >> 1) Does the umix memory requirements vary, based upon decryption key? >> 2) What is the upper limits on how much memory it will request? >> >> I now have my UM running with virtual RAM mapped to a 2 Gb disk >> file. This is extremely slow but I am hoping it does something, >> other than crashing, before 4am. However, in case it doesn't, I >> wanted to get these questions registered before the contest closes. >> >> I have even tried my decryption key on code from another author. >> He's successfully gathering points, but my umix generated by his code >> still crashes. >> >> DaR >> >> _______________________________________________ >> icfpcontest-discuss mailing list >> icfpcontest-discuss@lists.andrew.cmu.edu >> https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > > -- > Kurt Rinnert, University of Liverpool - Department of Physics > > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss From dufresnep at fastmail.fm Sun Jul 23 21:23:04 2006 From: dufresnep at fastmail.fm (Paul Dufresne) Date: Sun Jul 23 21:23:04 2006 Subject: [icfp-discuss] DIV error In-Reply-To: <1815f00607231816v7081c683ha5c516ebf37f2fb5@mail.gmail.com> References: <20060723175614.58713.qmail@web50715.mail.yahoo.com> <20060723181649.18542.qmail@web50713.mail.yahoo.com> <1815f00607231816v7081c683ha5c516ebf37f2fb5@mail.gmail.com> Message-ID: <1153704184.15046.266690498@webmail.messagingengine.com> On Sun, 23 Jul 2006 19:16:12 -0600, "Robert Van Dam" said: > Did you ever get that DIV error fixed? Hey, I did not even saw it, I did not have time to try your version yet. So no, I did not fixed it yet. :-) -- http://www.fastmail.fm - A fast, anti-spam email service. From aguilar.james at gmail.com Sun Jul 23 21:25:16 2006 From: aguilar.james at gmail.com (James Aguilar) Date: Sun Jul 23 21:25:19 2006 Subject: [icfp-discuss] Wires Message-ID: All, Has anyone managed to get more than 30 points on the first wires problem? My best result so far is 1673 bytes. Has anyone got a shorter implementation than that? -- James -- [?] James Aguilar [@] 18100 NE 95th St. #RR3088 / Redmond, WA 98052 [#] 314 494 0450 [!] In the wind and the rain, my darling, say goodbye -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060723/71e4498c/attachment.html From druffer at worldnet.att.net Sun Jul 23 21:26:00 2006 From: druffer at worldnet.att.net (Dennis Ruffer) Date: Sun Jul 23 21:26:04 2006 Subject: [icfp-discuss] Day 3! In-Reply-To: <20060724011650.GA27422@cs.cmu.edu> References: <20060723203710.GA26581@cs.cmu.edu> <20060724011650.GA27422@cs.cmu.edu> Message-ID: By any stretch, there is no collaboration. Only a sanity test. Hmm, it's still a mystery then. DaR On Jul 23, 2006, at 9:16 PM, Dan Licata wrote: > On Jul23, Dennis Ruffer wrote: >> As this year's contest winds down and I have yet to get to the >> password prompt, I have to ask a basic question. >> >> Is it possible that a specific decryption key would cause a umix >> image to be generated that would require more memory than I can get? >> >> This breaks down to 2 questions. >> >> 1) Does the umix memory requirements vary, based upon decryption key? > > Definitely not. It executes the same instructions no matter what. > The > values of registers and memory might be different, but there is no > control dependency. > >> 2) What is the upper limits on how much memory it will request? > > SANDmark allocates a total of 1597336168 bytes, and there is at most > 1227412 live at any time. > >> I have even tried my decryption key on code from another author. >> He's successfully gathering points, but my umix generated by his code >> still crashes. > > Your friend probably has a bug. > > By the way, teams aren't supposed to collaborate. > > -Dan From triska at gmx.at Sun Jul 23 21:35:46 2006 From: triska at gmx.at (Markus Triska) Date: Sun Jul 23 21:30:44 2006 Subject: [icfp-discuss] Wires In-Reply-To: References: Message-ID: <17604.9202.633156.290376@localhost.localdomain> James Aguilar writes: > > Has anyone managed to get more than 30 points on the first wires problem? > My best result so far is 1673 bytes. Has anyone got a shorter > implementation than that? We also have 30 points, 3696b. I doubt that file size matters -- more likely #boxes... Markus Triska. From aguilar.james at gmail.com Sun Jul 23 21:45:42 2006 From: aguilar.james at gmail.com (James Aguilar) Date: Sun Jul 23 21:45:45 2006 Subject: [icfp-discuss] Wires In-Reply-To: <17604.9202.633156.290376@localhost.localdomain> References: <17604.9202.633156.290376@localhost.localdomain> Message-ID: On 7/23/06, Markus Triska wrote: > > > We also have 30 points, 3696b. I doubt that file size matters -- more > likely #boxes... verify said that area is the criteria, which meshes with what I've seen. I got it up to 31 by being tricky. However, there is no more trickiness I can do now. Any further improvements would have to be straight up huge revisions. I'm moving on to rev. :) -- James -- [?] James Aguilar [@] 18100 NE 95th St. #RR3088 / Redmond, WA 98052 [#] 314 494 0450 [!] In the wind and the rain, my darling, say goodbye -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060723/8dbbaf2b/attachment.html From drl at cs.cmu.edu Sun Jul 23 21:52:35 2006 From: drl at cs.cmu.edu (Dan Licata) Date: Sun Jul 23 21:52:55 2006 Subject: [icfp-discuss] Wires In-Reply-To: <17604.9202.633156.290376@localhost.localdomain> References: <17604.9202.633156.290376@localhost.localdomain> Message-ID: <20060724015235.GA27497@cs.cmu.edu> The criteria is area, which is defined to be the length of the longest line times the number of lines. -Dan On Jul24, Markus Triska wrote: > James Aguilar writes: > > > > Has anyone managed to get more than 30 points on the first wires problem? > > My best result so far is 1673 bytes. Has anyone got a shorter > > implementation than that? > > We also have 30 points, 3696b. I doubt that file size matters -- more > likely #boxes... > > Markus Triska. > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > From aguilar.james at gmail.com Sun Jul 23 21:57:33 2006 From: aguilar.james at gmail.com (James Aguilar) Date: Sun Jul 23 21:57:37 2006 Subject: [icfp-discuss] Wires In-Reply-To: <20060724015235.GA27497@cs.cmu.edu> References: <17604.9202.633156.290376@localhost.localdomain> <20060724015235.GA27497@cs.cmu.edu> Message-ID: On 7/23/06, Dan Licata wrote: > > The criteria is area, which is defined to be the length of the longest > line times the number of lines. That's an interesting* definition of area. When you say "line" I assume you're talking about wires, not borders of modules? *non-intuitive -- [?] James Aguilar [@] 18100 NE 95th St. #RR3088 / Redmond, WA 98052 [#] 314 494 0450 [!] In the wind and the rain, my darling, say goodbye -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060723/6a5a71f4/attachment-0001.html From aguilar.james at gmail.com Sun Jul 23 21:59:47 2006 From: aguilar.james at gmail.com (James Aguilar) Date: Sun Jul 23 21:59:50 2006 Subject: [icfp-discuss] Wires In-Reply-To: References: <17604.9202.633156.290376@localhost.localdomain> <20060724015235.GA27497@cs.cmu.edu> Message-ID: Ooooh, you mean line as in line of the file. OK, that's fine then. On 7/23/06, James Aguilar wrote: > > On 7/23/06, Dan Licata wrote: > > > The criteria is area, which is defined to be the length of the longest > > line times the number of lines. > > > That's an interesting* definition of area. When you say "line" I assume > you're talking about wires, not borders of modules? > > *non-intuitive > -- > [?] James Aguilar > [@] 18100 NE 95th St. #RR3088 / Redmond, WA 98052 > [#] 314 494 0450 > [!] In the wind and the rain, my darling, say goodbye > -- [?] James Aguilar [@] 18100 NE 95th St. #RR3088 / Redmond, WA 98052 [#] 314 494 0450 [!] In the wind and the rain, my darling, say goodbye -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.andrew.cmu.edu/pipermail/icfpcontest-discuss/attachments/20060723/2984891e/attachment.html From drl at cs.cmu.edu Sun Jul 23 22:04:02 2006 From: drl at cs.cmu.edu (Dan Licata) Date: Sun Jul 23 22:04:12 2006 Subject: [icfp-discuss] Wires In-Reply-To: References: <17604.9202.633156.290376@localhost.localdomain> <20060724015235.GA27497@cs.cmu.edu> Message-ID: <20060724020402.GA27541@cs.cmu.edu> Yes, that's what I meant. Line in the file. -Dan On Jul23, James Aguilar wrote: > Ooooh, you mean line as in line of the file. OK, that's fine then. > > On 7/23/06, James Aguilar wrote: > > > >On 7/23/06, Dan Licata wrote: > > > >> The criteria is area, which is defined to be the length of the longest > >> line times the number of lines. > > > > > >That's an interesting* definition of area. When you say "line" I assume > >you're talking about wires, not borders of modules? > > > >*non-intuitive > > > > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss From kurt.rinnert at cern.ch Sun Jul 23 22:17:00 2006 From: kurt.rinnert at cern.ch (Kurt Rinnert) Date: Sun Jul 23 22:17:02 2006 Subject: [icfp-discuss] Day 3! In-Reply-To: References: <20060723203710.GA26581@cs.cmu.edu> <20060724011109.GC10442@vulcan> Message-ID: <20060724021700.GE10442@vulcan> I never observed more than ~450MB with my implemenation. (I did not bother to tweak the C++/STL memory managament). In case you are interested in another test case I can send you the source (after the contest is over). Considering Dan's reply you probably both have a bug. Then again, who has not? At least you know... On Sun, Jul 23, 2006 at 09:21:51PM -0400, Dennis Ruffer wrote: > Both of us are running OSX, which I believe has a 2 Gb limit per > thread, but I also tried it on a system with 8 Gb installed. > > Getting a new key is something I hadn't thought of, but my other run > will tell me for sure how much it needs. > > Thanks for the suggestion though. > > DaR > > On Jul 23, 2006, at 9:11 PM, Kurt Rinnert wrote: > > >As a test, why not try to register again with a different team name > >etc. > >and check whether the new key also causes problems? > > > >However, I doubt that a specific key causes memory exhaustion, > >considering the very good overall design of the whole thing. > >Still not impossible, though. > > > >Maybe your point gathering friend simply has more RAM... > > > > - Kurt > > -- Kurt Rinnert, University of Liverpool - Department of Physics From jonathan.roewen at gmail.com Sun Jul 23 23:18:53 2006 From: jonathan.roewen at gmail.com (Jonathan Roewen) Date: Sun Jul 23 23:19:30 2006 Subject: [icfp-discuss] Adventure memory usage Message-ID: <57041012163-BeMail@beos> Is it just me, or does adventure require a crazy amount of memory to run? I don't know if it's a bug in the UM (everything else appears to run correctly), or it's just seriously whack ;-) I get an out of memory exception with only 256MB physical memory available. Jonathan From flyfree at gmail.com Sun Jul 23 23:26:46 2006 From: flyfree at gmail.com (Zhenzhong Xu) Date: Sun Jul 23 23:26:48 2006 Subject: [icfp-discuss] Adventure memory usage In-Reply-To: <57041012163-BeMail@beos> References: <57041012163-BeMail@beos> Message-ID: <7b2bf4900607232026s562bf7e6t29ac97174b1c9ce9@mail.gmail.com> i implemented some sort of "garbage collector" in my UM, but still sometimes the UM will use up to 400 Meg of memory. Z. On 7/23/06, Jonathan Roewen wrote: > Is it just me, or does adventure require a crazy amount of memory to > run? I don't know if it's a bug in the UM (everything else appears to > run correctly), or it's just seriously whack ;-) > > I get an out of memory exception with only 256MB physical memory > available. > > Jonathan > _______________________________________________ > icfpcontest-discuss mailing list > icfpcontest-discuss@lists.andrew.cmu.edu > https://lists.andrew.cmu.edu/mailman/listinfo/icfpcontest-discuss > From kevin.watkins at gmail.com Sun Jul 23 23:47:45 2006 From: kevin.watkins at gmail.com (Kevin Watkins) Date: Sun Jul 23 23:47:48 2006 Subject: [icfp-discuss] Adventure memory usage Message-ID: <6b2ccb110607232047i624aacb4lcc6a89c3a9114b4d@mail.gmail.com> The adventure game allocates many arrays in the course of play, and occasionally abandons many of them at once. For example, after starting the game and using the "look" command a few dozen times in a room with a pile of junk, I hit a high-water mark of 7667982 arrays containing a total of 35105947 platters, or about 140 Mbytes. >: look Junk Room You are in a room with a pile of junk. A hallway leads south. [ ... ] >: state=interrupt r0=00009cb4 r1=00009cb4 r2=00000004 r3=001f1407 r4=00009caa r5=00005740 r6=00000000 r7=0000008f *00009cb4: b0000003 in r3 sim> s Instruction counts: cmovnz 6.83% 83245368 ld 17.03% 207702324 st 13.96% 170207679 add 2.26% 27616728 mul 0.24% 2965120 div 0.16% 1972360 nand 5.75% 70112449 halt 0.00% 0 alloc 1.82% 22206673 free 1.80% 21936407 out 0.00% 35506 in 0.00% 135 jump 7.57% 92273991 movi 42.57% 519152706 trap14 0.00% 0 trap15 0.00% 0 total 100.00% 1219427446 Allocation counts: total-words 95216810 current-words 5083819 max-words 35105947 total-arrays 22206674 current-arrays 270267 max-arrays 7667982 dup-arrays 0 sim> From iano at quirkster.com Sun Jul 23 23:42:41 2006 From: iano at quirkster.com (Ian Osgood) Date: Mon Jul 24 00:30:06 2006 Subject: [icfp-discuss] Which teams are small? Message-ID: Any indication on which of the top teams are in the running for first place? Grandeus