Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!


Forgot your password?
Puzzle Games (Games) Entertainment Games

Chess - 2070 CPUs vs 1 GM 248

jvarsoke writes "ChessBrain.net broke the world's record for 'largest number of distributed computers used to play a single game' by holding a chess match between Danish GM Peter Heine Nielsen and the equivalent of SETI@home (which similarly, has some people looking for a Mate). 2070 CPU's from 56 countries aided Black by running the chess program Beowulf, including a couple of University clusters. Their supernode ran Linux, and MySQL. The game was relayed by FICS. Results can be viewed here(1) and here(2)."
This discussion has been archived. No new comments can be posted.

Chess - 2070 CPUs vs 1 GM

Comments Filter:
  • by odeee ( 741339 ) on Tuesday February 03, 2004 @03:38AM (#8167155)
    I'd love to see a Beowolf cluster of those... Oh damn... it is =:-)
  • by Gogl ( 125883 ) on Tuesday February 03, 2004 @03:43AM (#8167176) Journal
    Or in case it gets Slashdotted or something, I may as well note who actually won the game (although I do think that is something that should have been noted in the submission itself but oh well).

    Our World Record attempt is now complete. We had serious technical difficulties early in the game, but managed to resolve them! The result of the game was a draw.
    • PS (Score:5, Informative)

      by Gogl ( 125883 ) on Tuesday February 03, 2004 @03:48AM (#8167200) Journal
      It was a draw by repetition. The human grandmaster had a position advantage and was able to force a draw that way despite being down a significant amount of material.
    • ... you can always tell the abstracts that were written by the authors - because they read as a sales pitch for the article and don't include the experimental results - usually to the annoyance of the reader. Descriptive abstracts, which are a summary of the article and include the results, are often written by others (even a "professional abstractor/interer" at times.)

      Dave Barry's blog is a good example of the former, Fark.om of the latter. In this story the editor should have added the results in the abs
  • by filtur ( 724994 ) on Tuesday February 03, 2004 @03:43AM (#8167177) Homepage
    Sure Chess it great, but can it find me a date?
  • by doomy ( 7461 ) on Tuesday February 03, 2004 @03:43AM (#8167178) Homepage Journal
    Nielsen,P - ChessBrain [E94]
    Guinness record attempt, 30.01.2004
    1.d4 g6 2.c4 Bg7 3.e4 d6 4.Nc3 Nf6 5.Nf3 0-0 6.Be2 e5 7.0-0 a5 8.Re1 exd4 9.Nxd4 Bd7 10.Bg5 Nc6 11.Nxc6 Bxc6 12.f3 Qd7 13.Qd2 Rfe8 14.Rac1 h5 15.Kh1 Nh7 16.Bh6 Bxh6 17.Qxh6 Re5 18.Nd5 Rae8 19.Qd2 b6 20.Bd3 Qd8 21.Rf1 Nf6 22.b3 Bb7 23.Qc2 Nd7 24.f4 R5e6 25.e5 c6
  • What's the point? (Score:3, Interesting)

    by syrion ( 744778 ) on Tuesday February 03, 2004 @03:44AM (#8167181)
    The problem with this is that it seems to assume that chess is a difficult problem. It isn't. Modern chess algorithms are really simple search-and- prune systems, relying on the computer's immense number-crunching ability to overcome the more heuristic human mind. Unfortunately, this isn't very interesting. What's the point? We know that computers can search faster than a human. See: Google. All these projects (DeepBlue, Fritz, this) accomplish is trivializing the game of chess, which is rather sad. Now, I'll be really annoyed when Go programs start improving to a 'decent amateur' level...
    • Re:What's the point? (Score:2, Interesting)

      by suchire ( 638146 )
      Mostly, yes, I'd have to agree with you, but there are also complicated mechanisms for searching. Designing a good algorithm isn't a simple matter. There are all sorts of problems that come with the basic minimax tree with alpha-beta pruning (the horizon effect being the most obvious of them all). If you think about it, there are actually quite a few different algorithms for chess game-tree searching, negascout and mtd(f) being the most popular. If you really want to see how "simple" chess programs are,
    • by vidarh ( 309115 ) <vidar@hokstad.com> on Tuesday February 03, 2004 @08:50AM (#8167920) Homepage Journal
      Any turn based board game are "really simple search-and-prune" systems. The problem is how to minimize the time taken by the search, and how to decide what a to prune, and how to decide which move to take.

      Brute forcing a chess game tree based on basic alpha beta minimax for instance is no way to play well against an experienced human player - first of all you won't get many moves ahead, and a good player that know how the computer work can easily set up a trap that will make the board look good X moves ahead, to make the computer to do stupid moves they can't easily reverse later.

      Second you face the problem of definining and weighting what a "good position" is. What is a good position depends on the strategy of the opponent.

      Most modern chess programs will augment the basic search and prune with a lot of heuristics to guide the search and weighting of choices, exactly for that reason. They also often contain massive databases of games, sequences of moves etc., to hunt for known strategies that humans might try to recycle against it.

      Chess isn't "simple". Chess is a game where it's easy to beat beginners, possible to beat intermediate players on modest hardware, and possible to face grand masters if you have lots of time and access to millions of dollars worth of hardware, and you can still expect to be surprised every now and again.

      It makes it interesting, because you have a good foundation to research algorithm improvements on, and because a good algorithm will be more and more useful as hardware costs come down, but it certainly doesn't invalidate the need for better algorithms.

      It's also interesting because better algorithms might help us appreciate how humans approach the problem, and as such benefit AI research.

    • As others have put in their replies, this isn't perfectly correct. There is a lot of research and advance to the field of computation to be found by producing an economical computer that can compete with the best human. Consider, for example, that it takes specialized equipment (to the best of my knowledge) to play against a human grand master, but wouldn't you recognize it as a breakthrough if some refinement of the time and space requirements could place that same level of computer opponent on your desk
      • Solving the best way to parallelize tough problems is a very important area. This was partially an attempt to address that area. (And it ran into bottleneck trouble!)
    • It would be more interesting if they could find a nice heuristic to make a Pentium IV beat a human, playing at around the same speed as the human.
  • by kamapuaa ( 555446 ) on Tuesday February 03, 2004 @03:46AM (#8167190) Homepage
    which similarly, has some people looking for a Mate

    May I suggest, that neither the SETI@Home, nor Chessbrain.net, is the best place where one can find a Mate.

    • May I suggest, that neither the SETI@Home, nor Chessbrain.net, is the best place where one can find a Mate.

      Oh, I don't know; I thought the same thing about Slashdot articles, but stranger things [aftenposten.no] have happened... [georgyforgov.com]
  • by vchoy ( 134429 ) on Tuesday February 03, 2004 @03:46AM (#8167192)
    To give credit to Danish GM Peter Heine Nielsen, I would have to say if there were only 2069 CPUs then he might of just won... :P (J/KING)

    More interestingly, would the ChessBrain.net team would of won with more CPUs?

    • "might of"

      "would of"

      Make the hurting stop!

      The sad part is you correctly said "would have" earlier in the post.

      Yeah yeah, evolving language. Some adaptations should be thrown in the chlorinated pool!

      I'm not usually a grammar nazi. But hey, chess is neat. Those fancy chess playing computers are going to take over the world some day, yessirree!
        Grammar Nazis... I hate these guys.
      • of course he wrote "would have" earlier in the post... he wouldn't form a contraction there: "I would've to say." It's less a matter of evolving language and more a matter of improper orthography...
      • Yeah yeah, evolving language. Some adaptations should be thrown in the chlorinated pool!

        Indeed! "might of" and "would of" are simply incorrect!

        As an example, take "would've" as the contraction in speaking "would have". The sound of the "'ve" bit is misheard as "of". Sure, sometimes it sounds like your saying "of", but it is NOT "of". It never has been. Some people are more familiar with the spoken form of English than the written - which is the reason for this error.

        To be on topic, my computer was one of

  • Results (Score:5, Insightful)

    by Stalyx ( 633692 ) on Tuesday February 03, 2004 @03:47AM (#8167195)
    "The game lasted several hours before resulting in a draw. Chess Grandmaster Peter Nielsen commented that he had set several traps for ChessBrain which computers normally fall for... but was surprised that ChessBrain refused them! "

    So what does this tell us? Nothing really, however it would be interesting how the computer will perform in a 5 match series.

    Although I still think the GM would win handily.

  • by Crypto Gnome ( 651401 ) on Tuesday February 03, 2004 @03:47AM (#8167198) Homepage Journal
    It's only a large aggregation, not really a cluster in that sense.

    Anyway apparently it worked! (ie not a cluster in that sense either)

    If it WAS implemented on the clustering technology we-all-know-and-love as Beowulf, would that make it a Beowulf-Squared?

    And, of course, we have to ask the (obvious) question(s)
  • Intangibles... (Score:5, Interesting)

    by John Seminal ( 698722 ) on Tuesday February 03, 2004 @04:03AM (#8167251) Journal
    Computers playing chess is not the same thing as two people playing the game.

    With two people, there are some elements that can not be programmed into a chess game. I remember in high school playing chess, there was a differance between playing a math academy team and a school best known for its basketball program. Expectations were different, the pressure was different. I remember the pressure of the state finals. There is the look the other person has, almost like poker. Can I bluff this person? Can I trick this person? What about the clock, can I manipulate that to cause an emotion in the other person.

    Maybe Spock can play a PC and have no differance in quality of play. But I prefer humans.

    • Re:Intangibles... (Score:2, Interesting)

      by Seraphim_72 ( 622457 )

      Reminds me of the kid who was a year older than me who was in the Chess Club. Big guy, joined the Marines right out of highschool, played on the football team etc. Anyway, when he would go into a match he would pull out his chair about 5 feet or so - really far. He would then sit down in it, bend at the waist, grab the table and pull it over to him with the board and pieces jumping all over making a huge racket. It invariably ended up with him sitting at the table fiddling with his pieces while some shimp o
  • Comp. vs. Comp. (Score:4, Interesting)

    by John Hurliman ( 152784 ) on Tuesday February 03, 2004 @04:08AM (#8167264) Homepage
    I want to see this cluster take on IBM's system!
  • by schm00 ( 639953 ) on Tuesday February 03, 2004 @04:12AM (#8167271)
    Has anyone ever written a system by which a large number of average chess players could collaborate to play a single game? The individuals could vote for the best move, and the majority would rule. Would a group like this be able to beat a high ranking player?
    • Offhand, I would think not. Tests with monkeys have shown that intelligence is not cumulative. Ten half-power monkeys just can't equal five regular monkeys no matter what, to put it simply. Assuming that each player acts intelligently, i.e. non-randomly, there is about epsilon chance of them winning. Where epsilon is the chance that one of those players does act randomly... and randomly picks the best move... enough times to win. 0.02EU
    • by sciencewhiz ( 448595 ) on Tuesday February 03, 2004 @04:20AM (#8167294)
      There are many systems like this. Chessworld.net is one, and they just challenged chessbrain to a match. You can see a full list of chessworld.net's ongoing games here: http://chessworld.net/chessclubs/event_show_chessw orld_summary_rowgames.asp
    • This was done, in Kasparov v World.

      It was done on the Zone.


    • No. Popular != best. 100 average people just gets you an average mob. If the average person is an expert then perhaps it works, but the average person isn't an expert.

      Many average eyes only make obvious bugs shallow. You need skilled eyes.

      A chess grandmaster aided by a bunch of high powered chess computers and programmes, might be able to beat the world number 1. The grandmaster provides strategy, and tells the computers which paths to look into. The computers provide search depth and protection against s
      • The grandmaster provides strategy, and tells the computers which paths to look into. The computers provide search depth and protection against stupid mistakes.

        This would be more interesting than a plain vote, which as you note would just aggregate mediocrity.

        Rather than a grandmaster or other real expert, for which there are limited pools of, perhaps you could use a voting system with the computer(s) providing the choices, and the users voting on the choices the computer made? While this might be le
        • The best way to benefit from the masses+computers? Given that "best" in this case is easily defined, you can use the "survival of the fittest" approach.

          Get the masses to play each other with computers providing assistance (depending on what sort of player you want). Top winner plays grandmaster/number 1 in the world. Genetic algorithm if you may.

          Many average eyes/people = good at avoiding obvious mistakes. Computers = very good at avoiding obvious mistakes.

          Computers so far aren't as good as the top human
          • Get the masses to play each other with computers providing assistance (depending on what sort of player you want). Top winner plays grandmaster/number 1 in the world. Genetic algorithm if you may.

            Maybe the trick is to get the masses to play the opposite side, kind of a devil's advocate -- have the machine provide it's suggestions to the humans who then evaluate its options by playing against them. This feedback could be used to asses its own N+1 strategies and perhaps dismiss those that were blocked more
    • Would a group like this be able to beat a high ranking player?

      I seriously doubt that the group would win. Some of the moves suggested by individuals in the group would likely be the best choice. But more votes would probably come in for another move - one which doesn't hold up as well.

      Some time back, I saw an average or slightly above average player play "everyone at the event" by allowing anyone who wanted to make one move in the game. Many people felt this put him at a disadvantage. But it actual

    • As others have pointed out, voting would be unlikely to work. I think that real collaboriation could, but the game would be impractical. The problem is that voting doesn't get you any benefit, and destroy any hope of a coherent strategy. Put together a proper group of people and let them discuss and educate eachother on the proposed moves and you might get somewhere, but coordination would be hell.
    • A friend of mine wrote something like this as part of his distributed computing course. Also know another group who did a UT 2003 mod which kind of worked like chess in real time. Pretty cool.
    • Has anyone ever written a system by which a large number of average chess players could collaborate to play a single game?

      Several years ago, there was a "Kasparov Against the World" event, wherein people could log into the net and vote for moved suggested by 5 grandmasters. This failed miserably (i.e. people sought to fix the voting, thus ruining the experiment.)

      OTOH, WorldChessNetwork does a "Grandmaster v. Everyone" event once a week or so, where all logged in players play against a GM by discussing,
  • by Doc Ruby ( 173196 ) on Tuesday February 03, 2004 @04:34AM (#8167323) Homepage Journal
    It's really "10 trillion neurons" vs. "2070 CPUs", but the neurons are about 40Hz, while the CPUs are in the GHz class. My bets are on the homegrown favorite, the MPP integrated analog processor with the "intuitive" OS. Although v2 of the digital SW will benefit from the digirally-distributed analog MPP network of metaprogramming, and might come out on top in round 2.

    "Chess is for computers" - Usenet 1997
      • "...will benefit from the digirally-distributed analog MPP network..."
      Ahh, digiral, must be some new Japanese technology! What will they think of next?
  • 2070 CPUs (Score:5, Funny)

    by Duncan3 ( 10537 ) on Tuesday February 03, 2004 @05:53AM (#8167513) Homepage
    1 CPU to beat the GM.
    +2069 CPU's so it could get on Slashdot.

    There are very few humans on the planet that can beat even one computer. That's been true for how many years now? Neither beating a GM or 2070 CPU's is impressive anymore.

    Someone go built a robot that can shovel snow, now THAT would be useful.
  • deep blue was one machine... here the GM can claim to have beaten thousands of machines all working against him..
    • Deep Blue was, indeed, a single computer but it had 516 processors that could assess 50 billion chess moves every three minutes. And that was 1997 technology! This machine was an RS6000 (RISC technology) that weighed 2,800 lbs and stood over 6' high. Perhaps in today's technology the equivalent would fit in a much smaller package.

      Happy Trails,


      • I don't have the book in front of me, so I can't verify it right now, but in "Behind Deep Blue," Feng-Hsiung Hsu, who built those chess processors, claims that you could fit a Deep Blue-class machine into something the size of a PocketPC today.
  • > and the equivalent of SETI@home (which similarly, has some people looking for a Mate).

    I had always wondered why people ran SETI@home; now I know: they have given up on mating fellow humans (Is their self esteem that low? Has obesity gotten that bad in America?) and are looking to find love with aliens, once we decrypt the personal ads they have been sending us via interstellar radio.

    (I think ambiguous appositives like these are a good reason to switch to Lojban [lojban.org])
  • by Knx ( 743893 ) on Tuesday February 03, 2004 @09:35AM (#8168119) Homepage
    There are approximately 35 moves per position in Chess (average value). Thus, the branching factor of the search tree is ~35 with a simple min-max search. Assuming that the program is always picking the best move to search first -- which is obviously not systematically the case -- alpha-beta pruning allows us to get a branching factor equal to approximately the square root of 35, that is: close to 6.

    Assuming that 2070 CPU are able to do the calculations 2070 times faster than 1 CPU -- which, again, is not the case -- it appears that the resulting supernode is able to 'see' up to 4 or 5 half-moves deeper than a single CPU in the same amount of time:

    6^4 < 2070 < 6^5

    It doesn't seem to be *that* useful. For most strategical positions, thinking 5 half-moves deeper just doesn't make any difference. Game 3 [x3dchess.com] of 'Kasparov vs X3D Fritz' is a good example: I'd be willing to bet that 2070 X3D Fritz playing together would have lost the game the same way, since the serious troubles caused by the pawns diagonal are still far beyond the resulting analysis depth. (Well... At least, I think so. I'm not a Chess expert!)

    Anyway, this is quite an interesting project. I hope to see it grow up in the future.

    -- Arnauld
    • There is a difference between "participated" and "worked flat out on the problem".

      In this case they had some serious bottleneck issues and at least the machines I had involved spent most of the time idle, throughout the game I probably got only about five moves per CPU, total.

  • ...Waffle Iron still hasn't taken the time to learn chess strategy; gets soundly beaten once again by a cluster of one Z80 running the chess cartridge on a 1992 vintage Gameboy.
  • Chessbrain is kind of a cool hack, and I would respect that, if they weren't filthy spammers. Here is a typical Chessbrain spam [tartarus.org]. Notice the spam body image is hosted off of chessbrain.net. (Filthy [gnu.org], filthy [lysator.liu.se], incompetant [redhat.com], spammers [wearlab.de].

  • but why does every chess player have funny hair?
  • My first contact with GNU was GNU Chess on Win3.1, mind you.

    It was a pretty good player (better than the other chess programs I had), but it was so desperately unstable, it'd crash at random times.

    Sheesh, some things never change.

God help those who do not help themselves. -- Wilson Mizner