Subscribe via RSS Feed

Chess Fraud

[ 29 ] January 22, 2011 |

Cheating’s not just for cyclists:

An announcement on Friday on the French Chess Federation’s Web site accuses three of its own players of cheating during last year’s Chess Olympiad in Khanty-Mansiysk, Russia.

The players accused of cheating are grandmasters Sébastien Feller, France’s fifth-ranked player, Arnaud Hauchard, No. 16 in the country, and Cyril Marzolo, an international master, ranked No. 46.

The announcement does not provide any details about what it is they are said to have done, but it is extraordinary for a national federation to accuse its own players in this way. Usually, accusations of cheating involve getting the help of chess computers during games as computers have surpassed people as chess players. Such aid is illegal under the rules.

France is the No. 3 ranked country in chess, and Feller, 19, is one of its most promising young talents.

He was actually the only one who played on the national team during the Olympiad, so perhaps the federation was accusing the other two of helping him. He started out with two wins, two losses and a draw in his first five games before winning three games and drawing his last. His record is on the Olympic Web site.

The allegations of computer cheating strike me as rather more severe than similar allegations in other sports. Steroids and other performance enhancing drugs simply increase the ability of the elite athlete to perform; if every athlete were held to the same standard, competition would presumably (allowing that different bodies react differently to certain training methods) remain the same. As the most advanced chess computers now exceed the capabilities of the best players, using them to cheat (as opposed to training) actually replaces the individual human contribution with something else. This strikes me as problematic.

Share with Sociable

Comments (29)

Trackback URL | Comments RSS Feed

  1. John says:

    Isn’t it pretty damned depressing that computers are better at chess than people now? At that point, high level chess starts to seem kind of pointless.

    • Meh, that was pretty inevitable wasn’t it? As computing technology improved, it was basically a given that computers would surpass humans at a game like chess, with a limited number of possible moves and the ability to quickly and completely chart every possible move ahead of time is huge.

      • hv says:

        It is quite possible to make games that are very similar to chess that challenge computers.

      • Ed Marshall says:

        When I was about six (in 1981), I remember I had a book full of things like “will cars fly in the future”, “when will there be robots”, ,etc.. futurology.

        I remember that “will computers be able to beat humans at chess” was more or less no. I wish I still had the book, but I believe the argument was that a grandmaster was playing five moves ahead. They showed that using brute force to plot out the possible moves in that way it would take some horrible amount of time at the processor speeds of the day. I don’t think they put it as “brute force”, but the idea that you *had* to plot out every possible move was just assumed. It was absolutely not gonna happen for hundreds and hundreds of years, if ever.

        • klk says:

          Funny – not long after that I remember reading someone (and my surely faulty memory wants it to be pianist Murray Perahia, but take that for the very little it’s worth) saying that “computer beats human at chess” was analogous to “automobile beats human at speed,” so I think there was that thinking around in the 80s, too.

  2. [...] This post was mentioned on Twitter by Robert Farley, LG&M. LG&M said: Chess Fraud http://is.gd/28aMON [...]

  3. Surely the solution is to implant computers into the heads of chess players to improve their capacity to calculate alternative moves. In the name of the Omnissiah, brethren. /cogwheel gesture

  4. LosGatosCA says:

    Problematic seems very diplomatic. Machines replacing humans doesn’t invalidate competition for either-as long as Jimmy Johnson doesn’t try to run a marathon in his car.

  5. gmack says:

    Yes, instead of having humans participate in weightlifting competitions, I think we should just bring in forklifts.

    • Ian says:

      A scene from the 1904 Olympic Marathon

      “Frederick Lorz stopped running because of exhaustion after nine miles (14.5 km). His manager gave him a lift in his car for the next eleven miles (17.7 km), but after the car broke down, Lorz continued on foot back to the Olympic stadium, where he broke the finishing tape and was greeted as the winner of the race.”

  6. Jeremy says:

    So in the future, the chess equivalent of peeing in a cup will be shaving their head to check for computer implant surgery scars?

  7. Rick Massimo says:

    This is both more and less worrying than you’re making it out to be.

    On one hand, practically speaking it just isn’t possible for a complete chess know-nothing to be computer-assisted all the way to the grandmaster or beyond. Chess tournaments are at least slightly more organized than that.

    Then again, in high-level chess games (and I think I’m just good enough to know how good high-level chess is, not good enough to play it) (and I’m probably being presumptuous even to say that) there are only a few major junctures; the rest of carrying a strategy out cleanly. And at those junctures, the possibilities are pretty well apparent to any serious chess player. A grandmaster wouldn’t need someone spelling out every move; a hand signal once or twice in the course of a game indicating something as simple as “go for it” or “head for the endgame” would be a huge advantage.

  8. asdfsdf says:

    How does one catch somebody cheating with computer aid?

    Also, not only are computers better than people at chess, computers are better than people at Starcraft as well.

    • Murc says:

      No, they are not.

      Talk to me when a non-cheating AI takes a series off actual high-level Starcraft pros instead of some no-names from the ICCUP ladder. That experiment, while impressive, was the equivalent of beating that guy who plays Chess down at the park.

      Take a game off Jaedong or Flash and we’ll talk.

      • hv says:

        Murc, I agree that that link doesn’t prove the claim. However, I do think that computers are very likely to prove better than humans at Starcraft. I think there are tiny but real advantages in multi-tasking, micromanagement, focused fire, exploring, optimizing resource flow, etc. Also, please consider how hard it was for chess developers to entice world champions to face their computers; let’s give the Starcraft AI devs a bit of patience.

        Let me put it back to you:
        Have IBM fund it and we’ll talk.

    • Rick Massimo says:

      Well, you find someone playing way above their level and you watch them for frequent trips to the bathroom, to stretch their legs, etc. And/or, in a recent case, you see a guy playing way above his level while wearing a sweater in July and an earpiece that he claims is a hearing aid but you Google the brand name and it’s a radio receiver.

      More challengingly, you look for someone who’s already really good but seems to be getting the occasional hand signals from a manager/friend/compatriot who keeps leaving the tournament hall. That’s tough because they’re already really good and, like I said above, it doesn’t take much. One or two hand signals a game would give a strong player a prohibitive edge.

  9. Some Guy says:

    On the plus side, now we can remake War Games, but with the game being Chess instead of Tic Tac Toe, since theoretically, computers could be so aware of the massive, but limited, number of moves each other could make, resulting in perpetual stalemates.
    It will stare Shia Leboufe, Justin Beiber, and 19 Year Old Starlette de Jour #XXXX. And it will be terrible.

    • hv says:

      computers > humans != chess is solved

      Also, maybe when chess is solved, we might learn that White wins by virtue of moving first. So the War Games remake may have the computer decide to launch.

  10. Jamie says:

    This is a straight cheating claim. Either they did, or they didn’t.

    Computers are now better at chess than humans. This is a fact. It is a consequence of the prop ties of the game. To the extent that some people would ike that to mean something, well, you can place your hope oin Go. For a while, at least. Fact is, games involving combinatorial forward strategy are doomed. LGM tie in: what does this do to military strategy? What has it done already?

    There is already a useful l, and interesting, variant of chess that is played with computer assistance. It may be the first game presaged by science fiction – Vernor Vinge presaged it in _the Peace wars_. Go look at “Advanced Chess” on Wikipedia.

    I still say, so what? I love chess. I used to play a lot, this is not the Singularity. Yu have to solve the human to get there, and i think we’re comfortably far from that.

    • hv says:

      Fact is, games involving combinatorial forward strategy are doomed.

      Some game theorists think it is possible to create “combinatorial forward strategy” games that will scale to any computational brute force capacity.

  11. Jamie says:

    Damn, my apologies. I’m on an iPad, and typing does suck on it. Not sure if the answer is fuck a buncha,or get a pwirerless keyboad, but if the difficulties of typing ‘fuck a buncha’ arer any indication, I’ll be keeping my laptop,

  12. rea says:

    This sort of thing has long been a problem in grandmaster tournaments–the only novelties are (1) the involvement of computers, and (2) that anything was done about it.

    For a notorious example see the game Fischer-Kovacevic, Zagreb 1970, in which two Soviet players, the ex-world champions Smyslov and Petrosian, gave the young Yugoslavian player the winning moves during the game. The whole chess world knew it happened; nothing was done.

  13. hv says:

    Computers have changed a lot about chess. Killed correspondence and adjournments. But created cheating!

    Cheating in chess is surprisingly recent (at top levels). Until* affordable computer engines surpassed human performance, it wasn’t really possible to “wire” the correct moves to a player. From whom would you get the moves? That player would be entered in the tournament more profitably. Moreover, for long periods, chess has been dominated by relatively few personalities each era. No one besides top GMs was beating Capablanca or Alekhine or Fischer or Kasparov or plenty of others in their primes, no matter who sent moves to their opponents; so there was a clear ceiling to these shenanigans. No one was going to highjack anything important. It might have been possible, pre-computers, to help a player to play up a hundred rating points; but, as Rick Massimo suggests above, you had to start with talented players and you still aren’t beating top flight opponents.

    Cheating really only occurred in the kind of event like rea describes above… a nearby grandmaster slips a move to the opponent of a rival.

    ====

    * technically, this was slightly before when computers could beat humans, because computers were substantially out-performing humans in some (known) areas of chess a bit earlier

    • Rick Massimo says:

      Computers have changed a lot about chess. Killed correspondence and adjournments. But created cheating!

      It’s a shame about correspondence chess (though I still play in no-engines-allowed tournaments, where we rely on the honor system, the fact that there are no money prizes and that if we were using engines we’d all be way higher-rated), but I think killing adjournments was a good thing. Players were getting help from countryment, coaches and other players who had an interest in the outcome of the game.

      The other thing computers did was tip the strategy/tactics balance. All sorts of ugly-looking plans turned out to be playable, and even those of us who can’t remember reams of computer-generated opening analysis remember the general principle that active pieces are not just more valuable than a pretty-looking pawn formation, but WAY more valuable.

Leave a Reply

You must be logged in to post a comment.