Just like The Matrix, but with no Keanu or fighting or bullet time or Carrie-Anne Moss.
November 10, 2003 10:02 AM   Subscribe

"He's not in this for the paycheck. He really takes the 'defender-of-humanity' thing seriously." Gary Kasparov faces another, still-tougher computer opponent, but this time in VR! "For the first time the man will meet the machine on its own turf, the virtual world," is the spin on this latest twist on the Kasparov-vs-Computer tradition. You can watch the match online starting tomorrow, but note that ESPN (!) will cover the entire match - "nearly 18 thrilling hours of live chess," Wired notes wryly.
posted by soyjoy (29 comments total)
 
Well I'm psyched and kind of sad to be so far from New Yawk. It's interesting; someone in one of these sites cites how Kasparov usually gets ahead early by exploiting holes in the machine's opening book - presumably by playing openings which have never been tried before, and creating a massive, unanticipated move-tree for the machine.

It will be awesome to see how this one turns out.
posted by kaibutsu at 10:27 AM on November 10, 2003


Nice. I recently viewed the "Kasparov vs Deep Blue" exhibit at the Smithsonian's Information Age exhibit, and was surprised to learn that, technically, Kasparov was not beaten; he resigned the final game in frustration short of a checkmate. Let's see if he's able to keep a cool head this time around.

Not that I'll be watching the Live Chess. Too boring. Just tell me who wins.
posted by brownpau at 10:36 AM on November 10, 2003


I think the VR 3-D board gimmick is kind of silly...unfortunately, the VR company is sponsoring the match, so it is here to stay!
posted by crunchburger at 10:39 AM on November 10, 2003


Resigning a chess game loses just as much as being mated does. And, he was roasted, which was why he resigned. Most GM games are not played out until checkmate.
posted by crunchburger at 10:41 AM on November 10, 2003


Virtual Chess? Man, nobody's on nobody's side...
posted by vorfeed at 10:42 AM on November 10, 2003


(Thanks for the clarification, crunchburger. I think you can tell from my own uninformed remarks how much chess I play.
posted by brownpau at 10:54 AM on November 10, 2003


Is this like Keanu Reeves in the Matrix? I think it's more like John Henry and the Steam Engine.

I wonder if Kasparov will beat the computer only to perish - as he pushes his queen into checkmate position - as a big throbbing blood vessel in his overheated cranium busts?

"He died heroically for all of humanity....


next contender........"

posted by troutfishing at 10:56 AM on November 10, 2003


Well, although the game is just as lost in terms of scoring, there is a difference between resigning and being checkmated - the possibility, however slim, that you might have been able to win or draw if you hadn't resigned.

BTW I remember following that match with a 486 and a chessboard in front of me. When he gave up that game I was flabbergasted - the situation looked perfectly even to me! Of course he could see hundreds of combinations I couldn't - but no matter, I still felt vindicated later when he got slammed for resigning.
posted by soyjoy at 11:01 AM on November 10, 2003


Oh, are we talking about different games? I was referring to the last game of the 1997 Deep Blue match. In that game Kasparov's position was perhaps not totally hopeless when he resigned, but it was lost (especially against a supercomputer).

Game 2, which your link refers to, is the notorius resigned-in-a-drawn-position game, and that went a little over my head too. As I recall, GK psyched himself out there, assuming that the thing would not have let him have an escape. So, he didn't look for it. The drawing lines were tricky but not really very hard compared to the kind of stuff he calculates routinely.
posted by crunchburger at 11:12 AM on November 10, 2003


> I want my 3-D glasses so I can watch Kasparov fight "inside
> the computer" against the evil Sylvester Stallone.

Next match after this one will be Kasparov vs. a nanobot and will take place in Kasparov's hypothalamus. Reserve your seats now.
posted by jfuller at 11:16 AM on November 10, 2003


Yeah, I didn't mean to imply my link was about the same game, crunchburger. But the "psyching himself out" thing is exactly what may be the margin of loss or victory in a human v. machine contest. Even if Kasparov can come up with viable openings that are completely off-book, the computer will relentlessly work out the best thread of moves in reaction to it, and the computer is never gonna say "this is hopeless!"
posted by soyjoy at 11:37 AM on November 10, 2003


Yeah, Battle Chess is cool, but it's no Archon. I hear that's all that Bobby Fischer will play these days.
posted by eatitlive at 12:19 PM on November 10, 2003


You can watch the match online starting tomorrow, but note that ESPN (!) will cover the entire match - "nearly 18 thrilling hours of live chess," Wired notes wryly.


I guess I did my part by e-mailing ESPN when they showed the last game of Kasparov - Deep Junior earlier this year. Hey, at least it's more exciting than another repeat of "The World's Strongest Man."
posted by gyc at 12:48 PM on November 10, 2003


I'll watch at least some of it. I love chess. Of course, I watched some of the scrabble (!) tournament yesterday and that was interesting despite the fact that I don't even like to play scrabble.
posted by callmejay at 12:57 PM on November 10, 2003


Not as many know this, but stavrosthewonderchicken is rated #2 and #4 respectively in the world in the sports of Yahtzee and Boggle, and in the summer of 1996 he went mano-a-Maco against a Classic IIe in a humanity vs. tiny-screened Applinity dustup for the title of "Yahtzee Champ of All Intelligent Creatures, Meat and/or Beige Division." At first, it seemed that all was lost for the Meat contender, since the Classic was getting three's by the bagfull, but eventually the game was ruled a "gimme" due to Stav catching the Classic using some cheapass diceroller it got off TUCOWS. Disappointing - but to have seen Stavros pitching-the-bones and whispering a cold, arrogant "Yahtzee, motherf**ker" to the snotty little brown box? To coin a phrase - priceless.
posted by UncleFes at 1:16 PM on November 10, 2003


You watch whatever you want, but I don't get it. Ten years from now computers will be able to beat any human at chess. Who cares exactly when this happens the first time? Did they hold races between human sprinters and early automobiles 130 years ago?
posted by Triplanetary at 1:21 PM on November 10, 2003


See, triplanetary, but it does matter... The programming aspect of it is a challenge to use your time to find move trees and narrow them down such that you win. It's about intelligent design at this point in history, though brute force plays a part in it. It is still a competition, because the game hasn't been mapped out in its entirety. I find it fascinating that the human methods of game play developed over the last hundred years are holding their own against the machines. It means that our non-brute force methods were/ are actually pretty goddamned good. Which suggests that we are probably more intelligent than we give ourselves credit for.

I suppose studying chess a bit helps the understanding of what the competition is over. Chess developed over something more like a network of dumb animals than over any kind of brute force method. It's a competition of methods, and it looks like what we had was pretty good.

And remember that the human vs. machine chess competition is really only a warm up for when human vs. machine Go starts getting off the ground. I think that's where the test of the future is going to lie...
posted by kaibutsu at 2:05 PM on November 10, 2003


I find it fascinating that the human methods of game play developed over the last hundred years are holding their own against the machines.

To be fair, kaibutsu, it is a game we designed, after all. If computers had created chess it would be a verrrrrry different game, one we might get four or five moves into before getting trounced.

That said, there are times when I'm playing that I go through a couple of combinations in my head and notice something unexpected developing out of them, deriving intrinsically from the way the pieces move and where they tend to land on their first couple moves, and I marvel all over again at the built-in geometry and proportionality of this game that was created out of many human minds' collaboration. This is usually, I guess, to avoid marveling at my own stupidity for not noticing the same things sooner.
posted by soyjoy at 2:42 PM on November 10, 2003


"To be fair, kaibutsu, it is a game we designed, after all. If computers had created chess it would be a verrrrrry different game, one we might get four or five moves into before getting trounced."

OK...the computers would be running on software designed by humans. Hard to get the ghost out of the machine in that sense.


Anyway, I lost interest as soon as I figured out this company is not using X3D. They are just hijacking the name. A lot of good people have put a lot of work into the X3D tech for them to dilute the name like this. Sad.
posted by john at 3:02 PM on November 10, 2003


If I remember correctly, questions arose during a previous match that Kasparov lost about whether the opponent (Deep Blue) software was static for the duration of the contest. The creation of plasticity for symbol manipulating processes yielding intelligent behavior would seem to be of some interest in these matters.
posted by fold_and_mutilate at 3:08 PM on November 10, 2003


I think the main thing of interest is seeing if the computer can truly play like a human. Previously, just a few short years ago, computers played world-class chess, but in a very un-humanlike way, and there were anti-computer strategies created to hone in on tendencies of computers towards certain kinds of moves. I think that in the last Kasparov computer match, the computer showed that sometimes it is able to calculate so deeply that they play very human-like moves, moves that humans make not because of calculating deeply like a machine, but as a result of seeing the board as a whole and based on intuition and feeling.
posted by gyc at 6:17 PM on November 10, 2003


Follow game 1 live here (flash), beginning 1PM EST . Or, go to WorldChessRating and click the 'follow live commentary' link, which opens a new browser window.
posted by crunchburger at 10:02 AM on November 11, 2003


"moves that humans make not because of calculating deeply like a machine, but as a result of seeing the board as a whole and based on intuition and feeling."

Probably too late to really pick nits and be heard, but from my understanding, moves are only products of intuition and feeling in speed chess. Rather, moves tend to be made based on near-philosophical understanding of the game make-up combined with an understanding of the fundamental logic of the game. Intuition does play a certain role, but I think the ideas we have about the game itself (such as quick development, how to go about getting center control, etc.) are what really make human chess human.
posted by kaibutsu at 10:35 AM on November 11, 2003


This is fascinating to watch - er, at least, check in on. Kasparov has just given up a pawn to gain attacking chances, but Fritz seems to be keeping the short-term advantage very viable. Two things strike me:

1) Man, I can't believe how much this "live" technology has improved since IBM tried to handle the first two matches and botched it in several ways.

2) Man, I can't believe I misspelled Kasparov's first name in the FPP! That's something I used to know. Apologies all round.
posted by soyjoy at 10:46 AM on November 11, 2003


You watch whatever you want, but I don't get it. Ten years from now computers will be able to beat any human at chess. Who cares exactly when this happens the first time? Did they hold races between human sprinters and early automobiles 130 years ago?

I think part of the reason might be that we were never the fastest animals around, or the strongest - we learned to compete not in absolute terms, but in a subclass, to be the fastest human, not the fastest entity. But as chess-players, or generally as reasoners, it had long been the case that the best human chess player=the best chess player, period. It will be interesting to see whether human chess playing continues competitively once the computers are completely unbeatable, the way marathon running continues despite the fact that the clunkiest yugo could beat any human.
posted by mdn at 11:18 AM on November 11, 2003


the computer is never gonna say "this is hopeless!"

Heh. Whoever said this hadn't seen today's game. Fritz was on the attack for the last dozen moves, first menacing and then checking Kasparov's king. Kasparov had a back-channel opportunity for mate once the pressure let up, and Fritz, even though "tactically" ahead, apparently threw up its hands and forced a draw by repetition. Fascinating. And, uh, enlightening.
posted by soyjoy at 1:39 PM on November 11, 2003


Well, Fritz is not tactically ahead - if it stops checking, it gets mated. Draw.

Actually they have something called the 'contempt factor' in computer chess which adjusts how the program will behave in situations like that. I'm not really sure how it works, but I like the name.
posted by crunchburger at 2:02 PM on November 11, 2003


Also: Computers don't get tired.
posted by soyjoy at 8:41 PM on November 13, 2003


...heh... but they also don't get long-term strategy. Fascinating stuff.
posted by soyjoy at 8:16 AM on November 17, 2003


« Older addictive java games   |   If we didn't have such a thing as an airplane... Newer »


This thread has been archived and is closed to new comments