First Kramnik vs DeepFritz, In Progress

An anonymous reader writes "Reigning world chess champion Vladimir Kramnik played the first match in a series of eight against the world's strongest chess computer. 'After the game Vladimir Kramnik said that he was never worried about losing the typical Berlin endgame that arose in his first game against Deep Fritz. The World Champion is the master of this line and Fritz was unable to take advantage of the white pieces.' There is live coverage of the event at the main website." We've mentioned this match a few times before.

Comments

No longer in progress: Draw

[Phouk]

"Kramnik was never worried about losing..." out of context is a bit misleading: Kramnik didn't win either, it was just a draw.

(For those who don't read the articles... ;) )

Re:No longer in progress: Draw

[SashaM]

Except that DeepFritz had white in this game, and when playing black at this level, your primary objective is not to lose, winning being secondary.

Re:No longer in progress: Draw

[Mr. PJR]

There's a saying among chess grandmasters--"Draw with black, win with white."

Anyway, I have a copy of Deep Fritz--the same program Kramnik is playing against. It's a comercially available chess program, designed to run on multiple processor computers. If you have a spare 90USD lying around, pick up a copy. It's a brutally strong program. You can buy it here: http://icdchess.com/cgi-bin/store.cgi (I have no affiliation with ICDchess, other than as a satisfied customer)

Re:No longer in progress: Draw

[mako]

There's a saying among chess grandmasters--"Draw with black, win with white."

Would someone please elaborate on this? I have seen this in other comments and do not understand the reasoning.

Re:No longer in progress: Draw

[Anonymous Coward]

White makes the first move.

At a high enough level, to win with black (second move) is a major coup.

Re:No longer in progress: Draw

[Mr. PJR]

Would someone please elaborate on this? I have seen this in other comments and do not understand the reasoning

When playing the black pieces you have a disadvantage because black always moves second. Basically, the player of the white pieces has a one-tempo advantage and can to some extent determine the shape of the game (by choosing which variation of the opening will be played). Strong chess players can take advantage of the first move advantage when playing white by forcing black to defend or play an opening which the black player isn't strong at.

With these disadvantages, black's behavior is usually to hold on and try to draw. Then to try and win when he has the white pieces (because the players alternate colors).

Re:No longer in progress: Draw

[Order]

Anyway, I have a copy of Deep Fritz--the same program Kramnik is playing against. It's a comercially available chess program, designed to run on multiple processor computers. If you have a spare 90USD lying around, pick up a copy. It's a brutally strong program. You can buy it here: http://icdchess.com/cgi-bin/store.cgi (I have no affiliation with ICDchess, other than as a satisfied customer)

I'm curious - why did you buy it? How is it fun to play against a machine that's so strong you can't possibly win?

Re:No longer in progress: Draw

[xintegerx]

Maybe because owning Deep Fritz is like having a Grandmaster caged up in your own home, one who tirelessly makes near-"perfect" moves?

Wouldn't it also be a great tool for analysis and to check what the best move, in a game you have played previously, would have been?

Wouldn't it improve your game?

People would buy Deep Fritz would use it for training. I've heard this about playing chess:

1/3 of your competition must be the same level as you are (makes for an even, fun game.)

1/3 of the competition should be stronger than you (to challenge yourself to advance and learn--why not learn from DeepFritz?)

1/3 of the practice competition should be weaker than you are (a match where you kick butt and raise your self-esteem after losing to higher competition [DeepFritz])

And, it doesn't take a genius to figure any of the above out (from your sig ;)

Re:No longer in progress: Draw

[haystor]

First, Deep Fritz is the name of the multi-processor version. For only $45 you can get the single processor version. No point in paying double the price to get your butt kicked.

There are a ton of features that make playing against Fritz wonderful. You can set the rating strength that it should play at. Of course, if you aren't an above average player Fritz will still be too strong most likely.

As others have mentioned, analysis with something like this is great. I feed my games to Fritz and look over them again. It really can point out when and how I went wrong.

But, my absolute favorite feature that Fritz has is something caled sparring mode. In sparring mode, Fritz will intentionally make mistakes at times. The difference between this and the average program is that an average program will just make a random mistake sometime. In sparring mode you first have to put Fritz under pressure and into a tactically active situation and only then will it make a little blunder (which you still have to catch). By being rewarded this way, I was able to learn what positions were putting pressure on my opponent and how to take advantage of mistakes. This is a great feature because its training in the middle game, which is much more difficult to capture in notes and books.

Re:No longer in progress: Draw

[desideria]

Here's four of the reasons I own it.

1. To analyze your own games more quickly with the engine, which is great for having new insight to your own games and evolution as a player.

2. Faster thinking times. A stronger engine is, well, faster... especially with SMP.

3. To play chess - Deep Fritz plays at a variety of levels. most of which are far easier than what Kramnik is playing.

4. Integration with a program called Chessbase, which is as you may have guessed, Database application for chess games.

Hope this helps!
- Cath

Re:No longer in progress: Draw

[invenustus]

How is it fun to play against a machine that's so strong you can't possibly win?

That raises another point, actually. It hasn't been proven that the "perfect chess game" doesn't exist. It's possible that someone could publish a book that consisted of an opening move for white, and the response to every situation that black could create from there, which would lead to mate for white every time.

If that happens, the whole game of chess is going to become pointless, because anyone with that book - or enough of it memorized - is going to be unbeatable. Sure, you could throw the book away, but as you played, you'd know that what you were doing could be flat out wrong - no better than not blocking your tic-tac-toe opponent when he/she has 2 in a row.

Re:No longer in progress: Draw

[dxkelly]

"For only $45 you can get the single processor version. No point in paying double the price to get your butt kicked."

No point in paying anything. For many players I think, there is plenty of challenge in free chess programs available online.

Re:No longer in progress: Draw

[b0r0din]

You couldn't write that book or memorize everything in it, because the possible moves that black can make are way above a million after say, four or five rounds. Good luck memorizing 10^28 possible responses (and no, I don't know how many combinations there are, suffice to say it's very high)

Now a computer might be able to eventually play the perfect chess game, having pre-computed all possible responses for every move. But for now computers aren't smart enough. Which is why humans can still beat (or at least draw) even the best computers.

If that does ever happen, if computers become too powerful, then champs of the world might just move on to a new game, like Go, which few AI play well.

Re:No longer in progress: Draw

[ponxx]

> That raises another point, actually. It hasn't been proven that the "perfect chess game" doesn't exist.

Indeed it can be proven that a "perfect chess game" does exist. In fact it is quite obvious that it does. Chess is a finite problem (due to rules about repeating a position 3 times being considered a draw). The question really is whether the "perfect game" results in a win for either color or in a draw.

Re:No longer in progress: Draw

[Anonymous Coward]

Actually White's average advantage is *half* of a tempo (according to Hans Berliner):

1. e4 ---> White +1 tempo
1... e5 -> equal ±0
2. Nf3---> White +1 tempo
2... Nc6-> equal ±0
(etc.)
= average +0.5 tempos for White, not +1

(This is usually *not* given in common chess books, but makes sense IMO.)

Regards,
Mike Scheidl

(sorry that I skipped the registration process; I don't usually post here.)

Not much of a contest...

[Anonymous Coward]

For those that are interested, the verdict among the chess world is that the computer is going to be exposed as a joke in this match. There are certain positions in which computers are very strong (tactical positions -- where each player has many choices over the next few moves and there are dramatic consequences), but there are equally many where they are not (positions in which long term planning is necessary and individual moves seem purposeless). Kramnik is not just strong -- he knows how to steer the game. The first game he had black and was thus trying to draw. So of course he immediately turned the game into a slow, boring game in which the computer's power was useless. Kramnik has shown previously that his anti-computer play is top notch, and you can look for it to win the match for him without problem.

Re:Not much of a contest...

[s1234d]

This is all very well, but the problem is that he must "steer the game" without making many tactical mistakes. All of today's strong players would beat these computers if they could be sure of not making any tactical errors during the game and have the luxury of just worrying about strategy.

Re:Not much of a contest...

[shimmin]

The quip is that the perfect chessplayer opens like a book, plays like a genius, and closes like a machine.

The question that I'd like answered (and the Fritz team probably won't tell because telling would compromise their machine's strategies) is how deep is Fritz's opening book, and to what extent is it weighted to play into enormously complicated positions where the human is more likely to screw up and the machine's inherent stupidity is less of a handicap (more of the lines are plausible, so the machine's difficulty in distinguishing between plausible and implausible lines is less important than its ability to quickly look through a broad game tree.

Re:Not much of a contest...

[Anonymous Coward]

This a ridiculous comment. Computers play perfectly when it comes to tactics. No human ever does. Humans are prone to mental lapses and errors. Even the best players. If Kramnik makes a single tactical error, the game is over. That is how serious this is. Now this game is certainly more favorable than the match Kasparov played against Deep Blue, but it is by no means a sure thing. In fact as I write this Kramnik is said to be biting his nails as the 2nd match progresses. Considering that Kramnik is often called "the ice man" for his absolute calm under pressure, I would say the computer is not playing like a "joke," you momo. Come back when you've got a little bit of chess knowledge, so you don't look retarded when you post.

Re:Not much of a contest...

[njdj]

Computers play perfectly when it comes to tactics.

It's interesting that people have been saying this since about 1960 (when it was obviously nowhere near true). It still is not true today. It can't be, because to be absolutely certain that a move is tactically the best, you'd have to analyse to the end of the game. At some point in tactical analysis, a player (human or machine) has to make the judgement that a possible position several moves ahead is "quiescent", i.e. it's unnecessary to carry the tactical analysis further.

Finally

[Anonymous Coward]

Whoa! I FINALY did it. Hi, my name is Earl. I am a Junior at the Cage de Folles Technical School in Topequanga Mississippi. I like using Linex and have 1,378 MP3 files on my b0xen. I beleive in essenshial liburties and that the MPAA and RIAA are evul fronst for Macro$hit!!!I live in my paRENTs basemant. It is ok, I got DSL hookup and the floor is dry exsept for March and April when the silage baffles are cleared. I have 3 parakeets named Eric and Dick and Bruce. They live in the same cage and sometimes fight. They fight by trying to climb higher than each other in the cage and shitting on the birds beelow. I cant let them free of their cage cause they dont no how to survive outside. The cage is there hole world. My Uncle Robert shares the basement with me. He is a good guy and sometimes we wrestle.

That box is certainly smaller than...

[weird mehgny]

...Deep Blue

Re:That box is certainly smaller than...

[stefanlasiewski]

Gee whiz, Deep Blue looks like it was created during the age of Deep Throat!

In fact, the angle of the Deep Blue picture reminds me of the video cover!

Re:That box is certainly smaller than...

[Hater's Leaving, The]

1400kg, and 256 processors (Power 2 superduper chip, or some bloody stoopid name like that). That's 5.5kg per processor. How heavy's your PC?

I'd say it was a fairly modern and compact design, personally.

THL.

Summary could have been better....

[billstewart]

Saying "After the game, [the human] was never worried about losing..." does sound rather like he won, and makes it clear that the author knew the results by the time he wrote it....

Kasparov lost...

[Ulumuri]

Because of unfair playing conditions, and also because he didn't have access to DB before the match. Therefore, matches between Kramnik and Fritz will take place every other day, be adjourned after 60 moves, and Fritz will not be reprogrammed between matches.

Currently, opinion is siding with Kramnik. GMs Nigel Short and Raymond Keene predict a Kramnik win.

The game went as follows:

Deep Fritz(2807) - Kramnik,V [C67]
Brains in Bahrain Man-Machine Match. Manama (1), 04.10.2002
1.e4 e5 2.Nf3 Nc6 3.Bb5 Nf6 4.0-0 Nxe4 5.d4 Nd6 6.Bxc6 dxc6 7.dxe5 Nf5 8.Qxd8+ Kxd8 9.Nc3 h6 10.b3 Ke8 11.Bb2 Be7 12.Rad1 a5 13.a4 h5 14.Ne2 Be6 15.c4 Rd8 16.h3 b6 17.Nfd4 Nxd4 18.Nxd4 c5 19.Nxe6 fxe6 20.Rxd8+ Kxd8 21.Bc1 Kc8 22.Rd1 Rd8 23.Rxd8+ Kxd8 24.g4 g6 25.h4 hxg4 26.Bg5 Bxg5 27.hxg5 Ke8 28.Kg2 ½-½

Re:Kasparov lost...

[GigsVT]

Because of unfair playing conditions, and also because he didn't have access to DB before the match.

Why would this matter? You are either able to beat arbitrary opponent "n" or you aren't. Sounds like a bunch of whining to me.

Re:Kasparov lost...

[Anonymous Coward]

Shut the fuck up, VT. I've heard enough of your almost incessant mass produced stupid messages.

GigsVT is gay.

-The GigsVT Troll

Re:Kasparov lost...

[damiam]

Deep Blue was fed all of Kasparov's previous games, so it knew exactly what to expect and how to optimize its strategy. Kasparov had never seen a game played by Deep Blue. It's common among grandmasters to review that past games of your opponent to look for their weeknesses. Kasparov didn't have this chance, which put him at a disadvantage.

Re:Kasparov lost...

[Anonymous Coward]

The GigsVT Troll is gay!

-The GigsVT Meta-Troll

Re:Kasparov lost...

[jasonditz]

Kasparov beat himself.

I'm sorry because he's a great player and all, but all this crap about how unfair it was is pathetic.

Look over those matches. Deepblue didn't play spectacularly, Kasparov just played miserable. He used openings that he has never used in tournament play and just generally slopped his way through obscure theoretical lines.

Hell, I could've beaten him if he'd played like that.

Re:Kasparov lost...

[Anonymous Coward]

You're obviously just GigsVT attempting to reclaim his heterosexuality.

Leave my presence, peon.

-The GigsVT Troll

p.s: GigsVT is gay. Almost forgot to say it!

Re:Kasparov lost...

[cheezehead]

Currently, opinion is siding with Kramnik. GMs Nigel Short and Raymond Keene predict a Kramnik win.

That does not mean anything. I think they honestly believe that, but their judgement might be clouded by emotions. Nobody likes to be confronted with the fact that a machine is better at their job then they are.

When Kasparov lost his crucial game against Deep Blue, he openly stated that he thought the Deep Blue team was cheating, because the decisive move could never have been made by a computer (it was too good a move).

Kramnik may have outsmarted DeepFritz with his Berlin Wall, but that only buys him a draw (!). If they should ever play anything else, he better make sure he does not slip up , because I suspect that DeepFritz will capitalize on the slightest mistake.

Re:Kasparov lost...

[Ninja Programmer]

Deep Blue was simple a souped up version of Deep Thought II. Kasparov could look through all those games if he liked (he even played a previous match with it, which he won.)

None of the Deep Blue team is strong enough to program a specific strategy in particular to play against Kasparov, except US GM Joel Benjamin. And Joel knows nothing about computer chess programs. He came up with the opening book for Deep Blue and played a few training games to tune its positional coefficients, but there is little else he could do for that team. They could not feed it a specific anti-Kasparov strategy; they had to rely solely on technological advantages.

Kasparov also played in a style that he's never played in the past. It was bizarre, closed, and very passive. Completely opposite to how Kasparov plays normally (he is typically a very dynamic, tactical, and aggressive player.) Even given this choice of strategy, there are many other players in the world who can do that better that he. In any event, Kasparov's choice of playing strategy nullified any specific anti-Kasparov strategy they might have created.

Deep Blue was a very expensive technical experiment by IBM that played 6 games in its life, then was quickly dismantled and never heard from again, except in IBM's advertisements. Them trying to sell their computers based on its ability to beat Kasparov is no different from Intel trying to tell your that the Pentium 4 will improve internet speeds.

Re:Kasparov lost...

[Hater's Leaving, The]

Wasn't he on non-prescription medication for ?migraines? at the time?

Didn't actually _pick up the wrong piece_ (i.e. a mechanical error) during one game?

So yes, he was below par.

However, _not_ letting GK have access to DB's prior games was an asymmetric condition, which breaks the typical protocol for such competitions, and if you force me to use the word 'unfair' for that condition than yes, it was unfair.

THL.

Re:Kasparov lost...

[haystor]

Kasparov played some tactical games against DB, going straight against the computers strengths.

Kramnik is playing with no such intention of winning with panache. He is fully expected by those GM's to grind out a victory. Anti-computer play is particularly suited to Kramnik'style and he' not likely to change it.

Fritz is not suited to an anti-Kramnik style and Fritz has no hope of changing it.

Re:Kasparov lost...

[jasonditz]

Yeah, but lets say I, who have no professional tournament experience, was going to play Mr. Kasparov. Now, I've got probably 10-15 books which feature most of his significant games. He probably doesn't have access to ANY of my games. But if I went out and beat him cold in a tournament, would it be any less impressive? If I did would everyone be arguing I should have been forced to forget all those games before the match to "keep it fair"? Besides which, that database of prior games was more useful as a psych-out tool than an actual game tool. Nowhere in DB's list of games did he ever run into GK opening with the Caro-Kann or any of other openings he used (at least never against a competant opponent). He's always been loyal to a couple of openings... that's why his games are always so boring to read.

Re:Kasparov lost...

[Grab]

So why do tennis players and football teams watch replays of how their opponents play? If you know your opponent has a weakness in a particular area, you aim in on that area and use it. If you know your opponent is particularly strong in one area, you stay the hell away from that area.

The DB team had logs of all Kasparov's matches, and so could program DB to apply extra weighting to those lines which would succeed best against Kasparov. Kasparov went in totally blind.

Grab.

Kramnick will win it

[dh003i]

Kramnick is a master of defense, an immovable object, so to speak; he proved that by beating Kasparov, an irresistable force.

Kramnick will play the defense and wait for his opportunity -- for the critical mistake -- to take the win. And, unlike this score-calculating computer, once Kramnick has won one game, he won't bother taking any risks; he'll just play solid defense every match, aiming for the draw; whereas the computer would foolishly (if it wins) try to win each successive game.

Also, if I recall correctly, this isn't a strictly timed match; its not a 5 minute game. Don't expect a computer to ever win a blitz match, because computer's just don't have the insight to play well in those circumstances, which is where human innovation shows through.

Re:Kramnick will win it

[Mr. PJR]

Don't expect a computer to ever win a blitz match, because computer's just don't have the insight to play well in those circumstances, which is where human innovation shows through.

I have to utterly disagree with this statement. Blitz games are quick games, such as a time limit of 5 minutes per side per game Blitz games are where computers are strongest, where their tactical ability, coolness under pressure, and lack of obvious mistakes shine through.

In quicker games, even weaker chess programs can anihilate strong human grandmasters. It's the longer games where humans are able to hold their own.

Re:Kramnick will win it

[Tablizer]

Also, if I recall correctly, this isn't a strictly timed match; its not a 5 minute game. Don't expect a computer to ever win a blitz match, because computer's just don't have the insight to play well in those circumstances, which is where human innovation shows through.

I heard it described as humans being better at pattern recognition than computers, not necessarily "innovation". The computer is good at matching exact patterns, but not approximations, and even when it can it does not know how to process/handle the differences very well.

Re:Kramnick will win it

[skeptikos]

> Don't expect a computer to ever win a blitz
> match, because computer's just don't have the
> insight to play well in those circumstances,
> which is where human innovation shows through.

That was strange. I would expect computers to become almost undefeatable at blitz matches in the future (10 years maybe). My reasoning is as follows: Current computers can play at the level of the best grandmasters in slow games. I am not a chess player but I guess that the ratio between the times you have to think in "slow"/"fast" games could be about 100 or something of that magnitude. If you consider Moore's law+better clustering+algorithm improvements it is concievable that computers in the future will make moves that are as strong as those made today in "slow" games BUT at blitz match rate. A performace improvement of a factor of 100 is not an impossible jump. I still remember my 486 computer running at 33Mhz and now a P4 can almost do 3Ghz (I know Mhz!=performace don't flame me).

It is hard for me to believe that in about 10 years blitz match players will make moves as strong as current best grandmaster's when they have lots of time to think!. QED :)

Re:Kramnick will win it

[dh003i]

I completely disagree. Fast games require insight and innovation (and fastly-changing strategies), something which computers simply can't do. Computers play by trying to figure out all the possible moves for so many steps ahead.

Re:Kramnick will win it

[haystor]

Please check out www.chessclub.com
You can log on as a guest for 7 days I believe and watch rated games between players (which will include GM vs computer games).

Grandmasters regularly play computer programs in 5 minute games. The best computers online have ratings higher than any grandmaster.

The simple fact is that computers are so fast that they can do enough calculations to look ahead up to 5 or more moves in a matter of seconds. This means they can play "perfect" chess where perfect is defined by the standing in the next 5 moves. Grandmasters may be able to do this but they can't do it in 3 seconds.

Re:Kramnick will win it

[jasonditz]

Indeed, that's why the average PC chess program has been getting progressively stronger. There are iterative improvements in the algorithms, of course, but mostly its just more Instructions per Second. In the 6502 days, even the best written computer chess programs topped out at 1300 or so. By the 386 they were playing at candidate master levels, and now they can hold their own with about anybody. I remember writing a chess program in Fortran IV in college and wondering why it was so weak on the CP/M server in the lab (I almost always beat it in 20 moves or less) and yet when I ran the same code on a 386 it gave me such a tough match.

Re:Kramnick will win it

[Anonymous Coward]

About your last point: it IS a strictly timed match. 120 minutes for all your moves, an additional 60 being added at move 40. Not only that, but your comment about blitz matches is absolutely and utterly incorrect. In blitz matches computers are already EASILY stronger than any human alive. Check out the Internet Chess Club (http://www.chessclub.com/). Top GMs play there, as do computers. Take a look at the ratings. A P4 2 ghz with 512 megs of RAM running crafty on linux is stronger than all but maybe 100 (if that!) human beings in the world.

The reason for this is that in blitz games, the computer can fail to see long term positional implications because the human will ALWAYS fail to see a tactic. Even in a 5 0 game (five minutes to make all your moves) the computer is ALWAYS looking 5+ moves ahead -- all but the best humans simply can't match that.

The long timered game (like the games of this match!) are humanity's last bastion of resistance against total computer dominance. However, as I said in my post above ("not much of a contest..."), they do not have the long term planning to win here. Given time, the best humans can see all the tactics the machine does. No amount of time will allow a computer to form a strategic plan.

Re:Kramnick will win it

[cheezehead]

I would expect computers to become almost undefeatable at blitz matches in the future (10 years maybe).

I'm too lazy to look it up, but I think computers already are almost invincible in blitz games. I think the reason is twofold:

1. Computer programs have very deep opening books in their database. So do most chess players, but I doubt that a human can beat a computer at simple lookups.

2. Blitz games are very 'tactical'. Computer programs are relatively good at this type of play. To put it another way: subtle stategical nuances in a position usually don't become very obvious unless you think deep (i.e. a long time). Humans don't have this time in a blitz game. Also, consider the way Kramnik got his draw: with a type of game that was very strategic, and not very tactical.

Re:Kramnick will win it

[cheezehead]

Fast games require insight and innovation (and fastly-changing strategies), something which computers simply can't do.

No. Fast games require not making mistakes like giving away a piece. Computer programs (even $40 PC programs) have the annoying habit of not making such mistakes. Humans do, especially under time pressure.

Re:Kramnick will win it

[Anonymous Coward]

Just take for example the (fairly) recent events which made Short look foolish. He played an online opponent hinted to be Fischer. Short believed that his opponent to be the great enigmatic former world champion who was known for his brilliancies, yet it was pointed out that NO human player could have devestated Short in such a manner. The opponent would move his king forward several squares, move it back, then proceed to completely destroy Short. Now Nigel Short is by no means a weak grand master, challenging Kasparov in the mid 90s for the world championship. The point is that only a computer can give away several moves, give away castling advantage, and still win so convincingly against a grand master. Expert commentary said that the games at that speed could not possibly be played so well by a human, and that not even Kasparov who is probably the best lightening player alive (refer to vegas lightening champs) could have played in such a fashion.

refer to http://www.kasparovchess.com for old articals on this

Re:Kramnick will win it

[sciire]

Wishy Anand is the "fast chess" master... Kasparov is a turtle on this kind of games.

Poor DeepFritz...

[rherbert]

He can only manage a draw in the first game, then we steal all his CPU time by Slashdotting him!

1st game, not match.

[mondoterrifico]

I know this is just being picky, but they are having only one match, and have currnetly played 1 game withing this match. Cheers

Re:1st game, not match.

[mondoterrifico]

Damnit, i pulled a Taco there with that misspelling. Sorry.

Berlin endgame?

[Johnso]

He's good at the Berlin endgame, huh?

Too bad the Russians couldn't say the same in World War 2...

Re:Berlin endgame?

[21mhz]

He's good at the Berlin endgame, huh?

Too bad the Russians couldn't say the same in World War 2...

Why, they got to set a flag over Reichstag and all. That was actually a draw, too.

What happened...

[Espectr0]

...with Deep Blue? I would like to see a match between 2 top-playing chess computers for a change

Re:What happened...

[WiKKeSH]

Though this doesnt mean that Deep Fritz has ever played Deep Blue, I got this info from Deep Fritz's player stats http://www.brainsinbahrain.com/about/more_stats.ht ml:

1995: Won the world computer chess championship in Hong Kong, ahead of Deep Blue

Re:What happened...

[Anonymous Coward]

What happened with Deep Blue is that it was dismantled. Torn apart and parts used for other things or thrown on the scrap heap, that is.

Which means no, you won't ever see Deep Blue play against Deep Fritz until someone comes up with a working time machine.

First Post was...

[certron]

1.e4 followed by e5

then
2. Nf3 Nc6

Some notes on chess notation:
http://chess.about.com/library/ble21brd .htm
(it is common to omit the pawn designation, it seems)

Some opening moves (which was this one?):
http://chess.about.com/library/ble50ndx.ht m?PM=ss1 3_chess

The whole match:
1.e4 e5 2.Nf3 Nc6 3.Bb5 Nf6 4.0-0 Nxe4 5.d4 Nd6 6.Bxc6 dxc6 7.dxe5 Nf5 8.Qxd8+ Kxd8 9.Nc3 h6 10.b3 Ke8 11.Bb2 Be7 12.Rad1 a5 13.a4 h5 14.Ne2 Be6 15.c4 Rd8 16.h3 b6 17.Nfd4 Nxd4 18.Nxd4 c5 19.Nxe6 fxe6 20.Rxd8+ Kxd8 21.Bc1 Kc8 22.Rd1 Rd8 23.Rxd8+ Kxd8 24.g4 g6 25.h4 hxg4 26.Bg5 Bxg5 27.hxg5 Ke8 28.Kg2 ½-½

Hey, I learned something from the above links.

btw, where does 'DeepFrtiz' the name come from? The team flag looks to be Germany, but where did the name come from? They are using an 8-CPU Compaq machine, also. (Good thing chess opening moves are public domain... otherwise the US Fritz would be making sure they didn't fall into the hands of free citizens! erm. or something like that...)

Re:First Post was...

[mondoterrifico]

The term Deep refers to the fact that it is able to take advantage of multiple cpus. There is an alternate program simply called Fritz that cannot take advantage of smp.

Re:First Post was...

[s1234d]

The "Fritz" part was apparently a marketing decision. "Deep" seems to just refer to a multiprocessor machine these days.

Re:First Post was...

[martyn s]

I think "Fritz" means it's fully DRM compliant.

Re:First Post was...

[Hater's Leaving, The]

Fritz post?

Alexander Fritz was a German chess master in the 1800s. I'm guessing Fritz is named after him.

THL.

Re:First Post was...

[21mhz]

Good thing chess opening moves are public domain... otherwise the US Fritz would be making sure they didn't fall into the hands of free citizens!

Are you talking about Shallow Fritz?

Gratuitous Star Trek ref

[T-Kir]

...aiming for the draw; whereas the computer would foolishly (if it wins) try to win each successive...

Kind of reminds me of an ep of TNG, where Data lost a game against a Grand Master (when his strategy was to win), but won when he went for a draw (and the Grand Master gave up out of frustration).

Size doesn't matter...

[phorm]

In the PC world, anyways.
Many computers that took up whole offices/floors/buildings are beaten by a $90 graphing calculator nowadays.

One of the local business has a really year old computer that manages some critical software. They can't take it offline because the processes it handle are extremely important, and there is no software to attend to them nowadays (though I'm sure it wouldn't be difficult to attend). The thing takes up half a room, has virtually no circuit pathways (all wires inside), and all of it's displays are etched with a nasty case of several years' burn-in.

10 years from now, Kramnik may be expending all of his brainpower beating Chess 2012 on a game-boy equivilent...

Live coverage.....

[jimson]

An in related news, live play-by-play coverage of paint drying!

John Henry was still a steel-driving man, though,

[janeil]

And Kramnik is still the current champ, win or lose to Deep Fritz. I'm always surprised when people make a big deal out of a human chess player, even the champ, losing to a computer program. The program thinks for basically millions of man-years for each move, moves the pieces around on the board to see what happens, and feels no fatigue or pressure. Of course the programs will eventually beat any and all humans, big deal! My Dodge Caravan can run down the fastest sprinter, too, so what? Are the Olympics now somehow pointless?

I give credit to Kramnik for taking Fritz on, I hope he gets big $$$ at least. It'll make big news if he loses, and not much if he wins, so it's hard for him to come out much ahead except for a payoff.

As for human dignity (see the web site) I can't imagine how we lose or gain any, geez, the machines don't even gloat. And, we can still unplug the machines. Seems like the human programmer of Fritz keeps the human dignity balance covered.

Re:anyone got any weed?

[Anonymous Coward]

dude... is it as dry there as it is here? wtf is going on?

Deep Fritz?

[Anonymous Coward]

Is that like Deep Blue with DRM?

Skip the hype

[teetam]

This is NOT a man Vs. machine matchup in any way. This, like all the previous ones, is between the human chess player and the human computer scientist.

Just because a computer can multiply two large numbers faster and more accurately than I do, does not prove that it is mathematically superior.

Re:Skip the hype

[Anonymous Coward]

I agree. I'm tired of all this man vs. machine nonsense. It's programers and hardware engineers vs. a chess player. Software and hardware designed and written by humans, vs the human player.

I've met a lot of people that think that computers can actually think, they base their opinions on the media reports of man vs. machine

*sigh*

Once we have AI (if ever) then it'll be a touch different...

Time trial? Time stats?

[phorm]

Was there a time-limit on moves in this game? I would be interested in seeing how the two opponents worked under the stress of time-limitations. Stats that showed not only the moves but also the delay between decisions would be also be cool.

One thing about the computer, it won't get distracted. Humans are affected by their environment as well, so things such as changes in temperature etc could affect the outcome

Of course, this is an honest match of brain against brain. But maybe if the chess pieces were drawn to the shape of playboy models Fritz would get the advantage next time :-)

Computers are still limited to the logic of their creators. Humans are illogical, so therein lies the advantage. - phorm

Re:Time trial? Time stats?

[MalleusEBHC]

Humans are affected by their environment as well, so things such as changes in temperature etc could affect the outcome

Changes in temperature won't affect the outcome? Let's take away the heatsink/fan and see does. Man: 1, Egg cooker formerly known as a mutliprocessor chess computer: 0.

Re:Time trial? Time stats?

[phorm]

Of course. I actually thought of this, but figured that as long as the machine has a good fan and heatsink running, then raising the room temperate by 10-15 degrees probably wouldn't affect it so much as a human (humidity is another story though).
It's a notable point though, but in this case I was more pointing out standard environmental distractions, removing the fan/heatsink from the computer would be more along the lines of human error in the case of the design team.

Computers don't need to take pee breaks - phorm

Re:Time trial? Time stats?

[egomaniac]

Changes in temperature won't affect the outcome? Let's take away the heatsink/fan and see does. Man: 1, Egg cooker formerly known as a mutliprocessor chess computer: 0.

The heatsink and fan are integral parts of the computer, necessary for its survival. How well would the human do without his lungs?

An intact computer can generally withstand much higher temperatures than a human.

Details

[ehiris]

Does anybody know what type of processors Deep Fritz is using?

Is the software low level assembly programming or does it run on top of an OS?

Re:Details

[Anonymous Coward]

Are you retarded? Do you think they'd actually program something so complex in assembly?

Re:Details

[imsabbel]

Yes, they do. Deep Fritz is 100% X86 Asm. Thats the reason why they used an 8*Xeon 900 and not a sun E12000 or anything like that. I would give you a link but your attitude is anoying. Try google yourself.

Re:John Henry was still a steel-driving man, thoug

[StillAnonymous]

"Are the Olympics now somehow pointless?"

The olympics have been pointless for many, many years now, but not for that reason. It's all about money, marketing, product placement, and hype. Lately, the IOC has been the main driving force for the debacle that is the olympics.

Whether it's issues with drug tests, corrupt judges, payola being passed around regarding location decisions, the olympics are a disgrace and yes, pointless.

I'd say that this event is really just a challenge for AI programmers. When you program a game that's designed to compete against human players, it's always fun to improve it and make it stronger. Who better to have it compete against than the best of the best so you can find the weaknesses in the code?

Trivial

[heikkile]

Chess is a trivial game - computers beat most of humans most of the time, and even on the top level it is a very close call. Given the speed computers advance, it is only a matter of time before no human can beat a computer in chess.

Machines have beaten man in many trivial games (tic-tac-toe. 100m sprint, weather prediction, etc). They have also failed in several "obviously easy" challenges (speech interfaces, AI, ...)

Before they play GO, I will not worry about my job.

Re:Trivial

[Anonymous Coward]

GO wouldn't bother. I wouldn't worry until they can master the "easy" games... like charades or something.

Re:Trivial

[Chicane-UK]

Hm.. well you have to bear in mind that on challenges such as basic games and 'the 100m sprint' there only needs to exist a certain number of rules for a machine to compete / win. But when you talk about things like AI and speech interfaces.. well thats a whole new ballgame. Speech took mankind thousands of years to develop - and even now I would say we dont use it as effectively as we could.

And the less said about AI the better.. I dont think you can really apply rules and reasons to intelligence!

Re:Trivial

[sampson]

"given the speed computers advance, it is only a matter of time before no human can beat a computer in chess...Before they play GO, I will not worry about my job"

given the speed computers advance, it is only a matter of time before no human can beat a computer in go. Traditional methods can't be applied (currently) to go because of memory/computation restraints, but it's only a matter of time. you'd better be fixing up your resume.

Re:Trivial

[tempmpi]

Successfully applying traditional methods to Go would require much more than only 10, 100 or 1000x the processing power currenly available. It is like saying: "Oh cracking 2048 bit rsa keys isn't a problem, just wait until computers get fast enough." Just throwing more cpu power at it isn't going to work. Better algorithms are needed.

I started playing Go 2 month ago, and I can beat the computer in an even game most of the time. (igowin or gnugo 3.2) There are some computer go programms out there that are slightly better but not much better. Playing Go is very different from chess. Many moves are choosen by intuition and it thinking about the next move is often a shape recognition problem. Computers suck at shape recognition and humans are quite good at it. I don't think I will see a go playing software that can beat a human professional go player in my lifetime. Pro go players can beat go software with a 20 stones handicap.

Re:Trivial

[Moridineas]

You kinda touch on it in your msg, but just to clarify, it's not that Go has some magical core that makes it harder a game than Chess. The problem is more or less what's called in the AI world the branching factor. In Chess I think the average branching factor is something like 30. That means that I have an average of 30 moves, then my opponent has an average of 30 moves he could possibly reply with, etc. Go has an IMMENSE branching factor.

If you look at Chess as a game, a human must learn how to use certain pieces effectively, with limitations on possible moves, etc, within an 8x8 board. A computer doesn't need to concern itself with strategies of certain pieces--it can just compute 20 moves ahead of all the possible moves and pick the best line. With Go, even if you're playing on a newb board size of 9x9 or so, the branching factor can still be like 80. It IS purely a matter of computational powers. If computers could brute force Go as much as they've brute forced chess, then no human could beat the computer at Go. I give Go no more than 20 years before it's "cracked" too (on non-super computing equipment).

You do make a point in that playing Chess and Go are for humans very different games--Go is largely about shapes and patterns. Chess is much more about strategic use of certain pieces.

Re:Trivial

[tunah]

Before they play GO, I will not worry about my job.

And before they read slashdot, I will not worry about mine.

Re:Trivial

[Anonymous Coward]

tic-tac-toe, that game on a 3x3 grid played with X's and O's and you try to get 3 in a row.

Isn't it proven that when any experienced players play the best result is a draw? How do machines win at a game that is inheriently unwinnable?

100m sprint, this is compariable to my civic in a F1 race. There exists an inherent power/drag difference. Are you going to limit the power and drag of the machine, if so whats the point. Or do you propose a nonmobile machine can win?

Weather Prediction != trivial, and besides I seem to predict the weather as well as the National Weather Sevice (read: machine).

And your "obviously easy" examples are neither. And their main limitation of these examples is the programmer trying to express these ideas in a manner in which the machine operates (see below).

Computers currently excel in one major thing, crunching numbers. Computers can be good at chess because it can be expresses (by the programmers) numericaly. Simeraly finding insanly large prime numbes, finding the 10^500000th digit of pi, or breaking public key encryption can all be done better (faster) by computers, because they involve numbers.

Re:Trivial

[PissedOffGuy]

chess is not "trivial" just because computers are better than humans at it. when the computers and their future pattern-matching algorithms etc. drop your precious GO too, youll still play it just like millions of people play chess.

go will just be the last to fall, after backgammon, checkers, connect4, chess, etc. after go, theres always music composition, poetry writing, etc. that the machines will beat us at.

AI is always a moving target for people isnt it?

Re:Trivial

[RzUpAnmsCwrds]

They do.

Re:Trivial

[2short]

"tic-tac-toe, that game on a 3x3 grid played with X's and O's and you try to get 3 in a row. Isn't it proven that when any experienced players play the best result is a draw?"

Yes. Furthermore you can define "experienced" as "not a jibbering idiot". My 2 year old can play perfect tic-tac-toe.

Re:Trivial

[Anonymous Coward]

machines have never beaten man yet. You mean a programmer & some hardware has challenged a human player.

Re:anyone got any weed?

[Anonymous Coward]

buy some shitty Mexican brickweed and make cookies...cheap, and it gets you so fahked up. Just saute the weed in butter for like 45 mins, then bake 'em normally (Duncan hines choco chip cookie mix works best).

That shit gets you high!

I dont' have time now,

[3-State Bit]

but one thing I rememer hearing much about karpov, back when kasparov was beaten, was that he, though not world champion, would have made a more interesting match against computerland, because of the fact that he focuses less on tactics (trying to out-think the computer by looking at combinations into more moves ahead) and more on abstract, pattern-based (such as in go) strategy, at which computers suck. Kasparov proved (insofar as you believe playing conditions were fair) that computers can out tacticate people, but perhaps a person whose style leans more toward abstract strategizing ("I want to keep this column open, because I feel it will be very important later" versus "I want to force the computer to lose that pawn, because I think I can pull off a combination in 43 moves")

i do need to go, but here are some things for children of this post to do:
o Look up some original reference (I saw many, many) that talked about how Kasparov's playing style is perhaps less suited to showcasing humanity's superiority to computers than Karpov's was)
o Look up whether Kramnik most resembles Karpov's or Kasparov's style.

One last thing.
Is it still true that in Go, computers play with a 14-move advantage and still lose to people who aren't even world-champion? Go is a game in which, because at each point in the game, it is unclear what groups of stones are alive and what are dead, pattern-based thinking is much more important. Would Karpov (and perhaps Kramnik) have made a better Go player than chess player?

When I come back, I'll add more to the thread, to anyone who wishes to discuss it.

Re:I dont' have time now,

[Scarblac]

i do need to go, but here are some things for children of this post to do:
I'm not a child of your post, but I'm a drunk chess player.

Look up some refernece that talked about how Kasparov's playing style is perhaps less suited to showcasing humanity's superioty to computers than Karpovs's was.
Kasparov is lethal when he has the initiative. He wants complex, tactical, attacking positions, and he's better at them than any human. Unfortunately, these positions tend to depend on calculation, which is what computers shine at. Karpov, like Kramnik, is more about prophylaxis, which is preventing any active options the opponent may have.

Although it must be said that at top level, all these players have a universal style. You can't become the world top player with a purely positional or a purely tactical style. Give Karpov a position that calls for a tactical solution, he's likely to play it. Put Kasparov in a quiet, strategical position, he'll usually know exactly what to do.

The differences show, mostly, in the choice of openings. They like different setups. Karpov choses the Caro-Kann (1.e4 c6) vs 1.e4, which is a very positional, defensive opening. Kasparov goes for the throat with the sharpest lines of the Sicilian (1.e4 c5). [if you're not a serious chess player, please believe me, that one square further makes a huge difference].

Kramnik plays the Berlin (1.e4 e5 2.Nf3 Nc6 3.Bb5 Nf6). An opening that gives White a positional advantage - just not enough of an advantage to win. It typically leads to an endgame that's better for White, though, in the hands of grandmasters, not yet winning. And he knows it well. There's no way a computer will understand all the subtleties in these quiet positions, Fritz isn't going to beat him here.

On the other hand, Kasparov actually lost to Judit Polgar, the world's highest rated woman, in the recent Russia vs Rest of the World match. Kasparov had a huge plus score vs Polgar beforehand, but he was tired, thought he could get an easy draw in that line just like Kramnik does. But he couldn't (a report of the match, including comments on the Polgar-Kasparov game, is at Chess Cafe). He just doesn't have the feeling for defending those worse, yet not yet losing, passive positions.

So the difference in style in small, but it's certainly there. And Kramnik's is much better against computers.

Re:I dont' have time now,

[legLess]

Blockquothe the poster:

Is it still true that in Go, computers play with a 14-move advantage and still lose to people who aren't even world-champion?

Oh, yes. Computers go programs are not serious opposition for anyone other than a weak to mid-level amateur. Here's a quick run-down of the go handicap system, for those not in the know: for each point of rank, or strength, difference in the players one stone of the weaker player's is placed on the board in a a predetermined position. Rank goes from 50 kyu (can't spell "go") through 1 kyu to 1 dan, then to 9 dan, then 1 dan to 9 dan professional. Thus a 5 kyu would give a 15 kyu a 10-stone handicap, and in theory, a 9 dan professional could give 68 (20% of the board) stones to someone who'd never played the game - and still win.

So when the poster says "14-move advantage" he means "14-stone handicap," which is huge. It's worse than that, though. A couple years ago, a dan-level player (a woman, not that it matters) beat the current computer go champion after giving it 27 stones. I can't find a bloody link right now, so you'll have to take my word for it.

Go is a game in which, because at each point in the game, it is unclear what groups of stones are alive and what are dead, pattern-based thinking is much more important.

Go is all about pattern recognition. The game is huge - easily the most complex game that people have created (where "life" is not defined as a game :). The board is 19x19 - 361 places to play - and all the stones have equal value. It's not possible for a computer to look 1% as far ahead in go as in chess.

Would Karpov (and perhaps Kramnik) have made a better Go player than chess player?

Who can say? They're very, very different games. I've played go for years, and every now and then I play chess with my brother. It feels very cramped, legalized, and formal. Go flows like a river.

Re:I dont' have time now,

[pangu]

a dan-level player (a woman, not that it matters)

If it doesn't matter why do you bother to mention it?

Re:I dont' have time now,

[Anonymous Coward]

Maybe for more information when you go to verify it, asshat.

Re:I dont' have time now,

[pangu]

Maybe for more information when you go to verify it

In that case the poster should have just provided the person's name.

Re:I dont' have time now,

[legLess]

If I knew her name I would have Googled for it. The AC was right - I gave all the information I had so perhaps someone else would have better luck finding a link.

Re:I dont' have time now,

[Shimrod]

The ranking system you describe is not entirely correct, as professional ranks are closer together than amateur ranks.

An amateur that has just learned the rules of the game is usually ranked at 40th-30th kyu. This then ranges up to 1st kyu. From there is continues with amateur 1st dan trough amateur 7th dan. Professionals range from 1st dan to 9th dan, with a 1st dan professional being of roughly equal strength to an amateur 7th dan and the strongest professional 9th dans being equivalent to a fictional 10th dan amateur rank.

Usually, if a player gets more than 9 stones in advance, he is instead given 10 points for any stone above the 9th. The maximum possible number of points the opponent can make in such a game is 369 (361 points on the board, plus capturing the first 9 handicap stones, minus his own move), so the maximum possible difference in strength for which this system still works is a 45 grade difference. Anything higher than that and black can claim a default win because he has that many points ahead. The weakest possible player is therefore usually ranked at 35 kyu (35 kyu grades + 10 dan grades makes 45).

Info on Fritz

[thelexx]

Here's decent description.

From looking at an ad for Deep Fritz 7, it runs under Windows and uses up to eight processors. No idea what language it is written in.

Re:Info on Fritz

[hayne]

According to the discussion on rec.games.chess.computer, the calculation engine of earlier versions of Fritz were written in assembler. I would imagine the same is true of this version. The GUI was written using Visual C++.

Re:Info on Fritz

[Wdi]

The core routines of Deep Fritz are written in assembler.

Newsflash! Deep Fritz 'seriously damaged'.

[wackybrit]

I just mirrored a news story from the BBC that states Deep Fritz has been destroyed in a possible terrorist attack on the conference. For some reason the BBC removed the story minutes after it was published.

Re:Newsflash! Deep Fritz 'seriously damaged'.

[Anonymous Coward]

Not bad, except that it's clearly a rip from
http://news.bbc.co.uk/1/hi/technology/2300267.st m

The about links in the sidebar are too old. :) Oh, and the comment in the HTML telling me that it was saved from that URL was helpful too.

Re:Newsflash! Deep Fritz 'seriously damaged'.

[$carab]

Well, that and the fact that an 8-processor computer running 90 dollar software isnt going to cost almost 2 million pounds.

Unless Microsoft changed their licensing again...

Re:Newsflash! Deep Fritz 'seriously damaged'.

[frankthechicken]

You mean the reporters name Thissisa Hoeks didn't give anything away? Though your investigative skills impressed me nonetheless.

Machine wars!

[charlie763]

Has anyone ever thought about matching some of these machines against eachother?

Re:Machine wars!

[buswolley]

Duh, they do it all the time. Nothing more to say about that

it's a DRAW

[Anonymous Coward]

Deep Fritz (2807) - Kramnik,V [C67]
Brains in Bahrain Man-Machine Match. Manama (1), 04.10.2002

1.e4 e5 2.Nf3 Nc6 3.Bb5 Nf6 4.0-0 Nxe4 5.d4 Nd6 6.Bxc6 dxc6 7.dxe5 Nf5 8.Qxd8+ Kxd8 9.Nc3 h6 10.b3 Ke8 11.Bb2 Be7 12.Rad1 a5 13.a4 h5 14.Ne2 Be6 15.c4 Rd8 16.h3 b6 17.Nfd4 Nxd4 18.Nxd4 c5 19.Nxe6 fxe6 20.Rxd8+ Kxd8 21.Bc1 Kc8 22.Rd1 Rd8 23.Rxd8+ Kxd8 24.g4 g6 25.h4 hxg4 26.Bg5 Bxg5 27.hxg5 Ke8 28.Kg2 ½-½

Gah!

[CyberDruid]

I am totally serious when I say that slashdot needs a "-1 wrong" or "-1 factual error".

• True - Kramnik is a staunch defender.
• Computers are not big on taking risks to begin with (they hardly ever sacrifice material for instance) and they don't really "play for a win", but if the operators wanted it to play more drawish, that would not be a problem, provided that they are allowed to adjust some positional parameters.
• A 5 min game would be extremely difficult for Kramnik. Quick games are basically just about calculating tactics, since the deeper aspects become hidden behind both sides poor play. A human excels in stuff like planning and sometimes logical reasoning, which both takes some time to do. It is a well known fact that computers don't improve their play much when given longer time (programmers will recognize this problem as "the exponential wall").

On a side note: In this game Kramnik drew easily because he could do some logical reasoning that no computer has ever done. He understood that in the final position, the computer could manoeuver around as much as it damn well pleased, there were simply no legal moves that could ever threaten anything. A computer will have great difficulty understanding this, since the calculation of variations will not show this simple visual fact.

In my opinion Deep Fritz will never beat Kramnik in a Berlin Defence. The team could try to deviate earlier, perhaps by closing the position with 4.d3, but this will also be easy play for Kramnik. They could also skip the Ruy Lopez altogether and play 3.Bc4 (Italian) or 2.f4 (King's gambit) instead, but these moves are not so common among the extreme elite. Kramnik would probably equalize comfortably against these moves. IMHO the team should try either switching to 1.d4 (at least for one game, to see where it leads) or just try to head for equal but tactically complicated positions after the King's gambit or the Italian, mentioned above. Playing 1.c4 or 1.Nf3 would probably be unwise. Kramnik knows these waters extremely well and could probably easily steer the game to a dull and totally safe position.

My money is on Kramnik, he will probably not lose a single game.

Re:Gah!

[Ninja Programmer]

1. Kramnik is more than just a "staunch defender". He is the strongest all around chess player with the possible exception of Garry Kasparov. His greatest strength is his incredible depth of understanding chess from a positional point of view. That's how he beat Kasparov.
   
2. You are not up on the latest in computer chess results. The top programs have varying degrees of positional understanding which allows them to make some kinds of sacrifices. Deep Junior is especially renowned for its ability to make pure sacrifices on positional critieria.
   
3. Indeed in speed chess, computers reign supreme. But why not cite the actual results? A few years ago, Kasparov, Anand, and others were invited to play a few exhibition blitz games against a previous version of Fritz. They played a few dozen games, with Fritz winning by an extreme margin (something like 10-2 or so.)
   

(Deep) Fritz is a closed source program. It is impossible to say what its capable of in certain endgames. These programs have over a decade of programming in them, and they all have interesting strengths and weaknesses. While your analysis about its understanding of this endgame may be correct, I don't think your reasoning is necessarily well founded. Some ICCA (International Computer Chess Association) articles have shown methods for encoding algorithms for many endgames, such as pawn endgames in almost completely formulaic ways.

Of course Kramnik has been playing training matches against this precise version of Fritz for quite some time now. I would suggest that we *know* Fritz won't beat Kramnik in the Berlin Defence simply by virtue of him using it. The rules for this match stipulate that the Fritz team may not deviate its programming or openings while the match is going on. Thus if Fritz wants to change strategy, then its up to the program, not the programmers. It better have some kind of random opening generator, or opening learning system, otherwise a most embarassing thing will happen -- Kramnik will draw all his games with black with precisely the same moves.

Before I make a prediction, I want to see how Kramnik handles white. Of course, Kramnik is well known for his solid play, but if he can't make headway as white, he may feel he has to try harder, and screw up in the process. We shall see. But obviously drawing with black, and fairly easily shows that Kramnik is doing well so far, though perhaps not as well as we/he might have hoped.

Re:Gah!

[Frodo2002]

Here here. Computers started kicking GMs' butts at blitz years back. -1 plain wrong.

How many people can beat the computer?

[Discoteck]

Finially a /. post about chess@!! I was wondering when you geeks would get around to it. :)

Can anyone recommend some good chess strategy books? I found this link but it does not give a very good indication of what book would be better than others. I guess it would have to depend on what I want to read about since it is a game of strategy.

So I was interested in how many readers are able to beat the computer when playing say the ChessMaster 7000 - 9000 series? I was interested in buying a chess game that teaches you tactics and strategy. I had heard good things about the ChessMaster series. Are there better titles out there? I think for what they offer it is really good. You can look at most of the famous past chess games to see how the professionals think about the game, well I guess if you could understand them I guess you would be wasting time with the game.

I used to play Kunfuchess online alot until I was forced to connect on a dialup modem. It is a pretty addictive version of chess; anyone who likes chess and hasn't tried it, should.

While surfing for links for the loyal /. readers I came across a couple that might prove useful.

http://www.wolffchess.com/php/home.php3

Once you register, you can improve your chess with hundreds of Web-based exercises, specifically designed to complement my book, The Complete Idiot's Guide to Chess.

http://www.chessclub.com/

Come join us! Register a name, install our easy-to-use software, and then use that to connect to our playing site. You can try it for free! With over 25,000 paying members from all over the world, Chessclub.com is the longest running and most vibrant chess community on the internet. You can play games and get a rating, watch grandmasters play while discussing the game, take lessons, play in tournaments, play in simultaneous exhibitions, try chess variants like bughouse, crazyhouse and atomic, play chess programs of all levels, and much more.

Of course there are always the game sites that offer chess onlne. It is one of the more popluar classical games that are available by most any site. Here are some that I found.

http://games.yahoo.com/

http://www.pogo.com/

http://www.station.sony.com/

http://www.playsite.com/

http://www.gamespyarcade.com/

and the list keeps on going... I know that I forgot a couple but if you want to play online these links will be more than sufficent to get you going.

Re:How many people can beat the computer?

[primus_sucks]

Ok, I'll probably get flamed for this, but I like the series of chess books from MSPress - Play Winning Chess, Winning Chess Strategies, Winning Chess Tactics, Winning Chess Endings.

Re:How many people can beat the computer?

[mpsmps]

I agree, those are excellent. /.'ers can at least feel good that Seirawan complains in the Introduction to his outstanding "Winning Chess Brilliancies" that Microsoft limited him to only 2/3 of the games he wanted to put in. Also, Microsoft has let most of this excellent series go out of print, which is a travesty.

Re:How many people can beat the computer?

[Anonymous Coward]

Try the books by Jeremy Silman. I've found them to be pretty good.

Re:How many people can beat the computer?

[Discoteck]

series of chess books from MSPress

Is that Microsoft Press? What are they doing printing books about chess. Thanks for getting back to me. My Dad had showed some interest in becoming better at chess because I have beaten him everytime that we have played.

Re:How many people can beat the computer?

[primus_sucks]

Is that Microsoft Press?

Yes.

Re:How many people can beat the computer?

[haystor]

http://www.ex.ac.uk/~dregis/DR/coaching.html

Check out the above. That guy helped me learn the game a lot. Oh, and listen to what he says about not studying things that are beyond your ability (reading its ok, just don't study what you aren't ready for yet.)

Don't fall into the trap of studying chess and not playing it. I heard someone suggest that every beginner should learn the rules, the opening principles and basic tactics and then should play 200 games before picking up a book. I'd have to say that's good advice, the temptation is to read book after book when your best improvement would come from playing.

Re:How many people can beat the computer?

[Anonymous Coward]

I had a very good chess book that I had a long-time ago ... and lost it.

The book was titled "The Middle Game" by Dr. Max Euwe.

It is probably the best chess book that I ever had and through it, won several of my chess matches in chess inter-school competitions.

I was able to *understand* the position and how to deal with it ... if only I can get another copy of it.

Re:How many people can beat the computer?

[Scarblac]

Can anyone suggest some good chess strategy books?

"How to reassess your chess" by Jeremy Silman is probably the most-recommended chess strategy book, but it's not for beginners, more for somewhat advanced club players. A cheap, all-round good book to start with may be "The Mammoth Book of Chess", by Burgess and Nunn. Go to Amazon for reviews by people and sample pages, they're good for that sort of thing.

Of course there are always the game sites the offer chess online.

The best for Slashdot geeks should be FICS, at http://www.freechess.org, with its command line interface and geeky audience (usually 400+ players online). The best Linux client to play there is eboard.

Incidentally, SCID is a *great* GPL'ed chess database, originally for Linux but also ported to Windows, that makes Chessbase obsolete as far as I'm concerned.

Hope this helps.

Re:John Henry was still a steel-driving man, thoug

[Ulumuri]

Actually, Kramnik is not the World Champion currently. That honour rests with Ruslan Ponomariov, 18-years old, from Ukraine. (You can see his webpage here). Ponomariov became the sixteenth World Champion on January 23, 2002 after beating countryman Vasilly Ivanchuk.

Man vs Machine: A More Just Form of Competition

[CreamsicleSeventeen]

Kramnik should write a chess program that can play against DeepFritz. DeepFritz should give birth to a chess Grand Master.

computers better at blitz

[hayne]

dh003i wrote:
its not a 5 minute game. Don't expect a computer to ever win a blitz match, because computer's just don't have the insight to play well in those circumstances, which is where human innovation shows through.

Actually, the exact opposite is true. Computers regularly beat even the best human players at fast time controls (blitz) since the humans are much more prone to making mistakes when they don't have time to think a lot. This is not merely my opinion - I think you will find few people who are familair with computer chess who would think otherwise. For example, here's what Robert Hyatt (author of Cray Blitz & Crafty) said in 1999( rec.games.chess.computer )

So today, game/30 is no longer safe and the computers are probably better there. Game/60 is also becoming more difficult for the humans, although I think they can do pretty well at this time control. But at 40/2, where there are no "blitz" time controls at the end, the GM players begin to show exactly why micro chess programs are not yet GM-level players at this time control. They simply know "too much" it seems, and they are able to exploit weaknesses they see while the programs are usually quite oblivious to what is going on...

Your opinion...

[BTWR]

Do you think that if a human player beats this computer it will erase Deep Blue's victory from people's memory? I think not. No one seemed to care that the grandmasters could beat the pc's for years. They only cared when the humans lost to the silicon.

Re:Kasparov lost... - I'm conflicting here

[403Forbidden]

and Fritz will not be reprogrammed between matches.

On one hand, the human brain can be reprogrammed to suit the competitor, but on the other hand a brain recodes itself in a manner of speaking, a computer doesn't.

Perhaps they should make it so it can self-compile changed code? Would be hard, but a definate advantage.

Fair enough

[Rui del-Negro]

matches between Kramnik and Fritz will take place every other day, be adjourned after 60 moves, and Fritz will not be reprogrammed between matches.

Sounds fair. As long as Kramnik isn't reprogrammed either. ;-)

RMN
~~~

Boring? Oh please...

[403Forbidden]

if you wanna see boring look no further than the pitch drop experiment! http://www.physics.uq.edu.au/pitchdrop/pitchdrop.s html

I wrote the world's greatest chess program

[Rui del-Negro]

You may not be aware of this, because They tried to cover it up, but I once wrote the world's most powerful chess program.

My approach was simple: to compute every possible move in every possible game, and come up with the perfect sequence. It took 14 years to do, on a 700-CPU supercluster, but finally we 'solved' chess. The database was huge. The program was unbeatable.

Unfortunately it was also rather boring. The human would make the first move and, invariably, the computer would spend 4 hours sorting through the database and finally declared:

Checkmate in 14705 moves. I win.

RMN
~~~

Re:I wrote the world's greatest chess program

[Anonymous Coward]

Thats great that you did that.

However I do have a question, since there are only 20 possible first moves in chess for a side (2 moves for each 8 pawns, 2 moves for each 2 knights) why not just pre-compute the 14705 or so moves for each of the possible first moves and declare you've won in a O(1) table lookup instead of consulting your database for 4 hours.

Liar Liar your pants are on fire

Re:I wrote the world's greatest chess program

[Anonymous Coward]

Except that the longest possible chess game
is 5948 moves.

Re:I wrote the world's greatest chess program

[Rui del-Negro]

Because that would be cheating. ;-)

BTW, I didn't really do that. I don't even particularly like chess. I can never remember how the little horse-shaped ones move.

RMN
~~~

Now...

[ymgve]

...imagine a Beowulf cluster of DeepFri...

AUGH! Stop beating me!

All your pawns are belong to...

NO! Not the baseball bat!

How long til somebody mods DeepFritz to run Lin...

Ack! Ack! ugh...you win...

Kind of pointless...

[Eric Damron]

As computers become faster and programmers improve chess algorithms, we will reach a point where no human can beat the top computer player.

There are two types of play:

Tactical play which is fairly straight forward and usually involves an immediate advantage. The thought process might go something like: If I take his pawn on e4 he can take it back with his rook but then my knight can fork his rook and queen. If he sees the fork and doesn't take back I'll be a pawn up.

Strategic (or positional) play which is far less straight forward and usually involves a more subtle advantage. The thought process might go something like: If I trade my knight for his bishop we'll have bishops of opposite colors and with this pawn structure I'll be able to draw even though he is a pawn up...

Computer's are very good at tactical play. Basically they look at every legal move and legal reply. The faster the computer is the farther ahead it can look. Then it's a simple matter to count pieces. The number of possible positions rises exponentially with each move so computer speed becomes a limiting factor. With various branch pruning algorithms we still need to make a computer go a lot faster just to allow the computer to think ahead one extra ply for a given time period.

Positional play is a lot harder to program. For one thing, most of us suck at it so we couldn't tell a program how to do it. But even so there are master chess players who can program and others who have been willing to work with programmers and the result is that the algorithms for positional chess are maturing. It will only be a matter of time until the advances in positional chess algorithms and the brute force of very fast computers make it impossible for humans to beat computer programs.

But don't feel bad. For a human to compete against a computer in some kind of 'best in the world' contest is a little like letting a hydraulic jack enter a weight lifting contest.

Re:Kind of pointless...

[Anonymous Coward]

It will only be a matter of time until the advances in positional chess algorithms and the brute force of very fast computers make it impossible for humans to beat computer programs.

Actually, the branching factor in a chess game is so huge that a small increase in computer speed (say, doubling the processor speed) doesn't make much of a difference. You're better off trying to improve your algorithme, like the creators of Deep Fritz did. Deep Blue could search 200 Million moves per second, and Fritz can only search 2 millon, but (at least according to Kramnik) Deep Fritz is a better computer.
Besides, the speed of computers isn't increasing as much as it used to. With the lousy sales on the latest Pentium, do you think Intel is going to want to come out with another one twice as fast? That's going to be a problem for those who rely on the expected increase in brute power of computers to beat the next chess champion.

Re:Kind of pointless...

[Anonymous Coward]

There was this one book called blondie84 or something like that but the book explained the thing you were talking about. The book described that a neural network that had nodes about various types of spacial info understood how to use position to beat their opponents. This book was written for checkers but i'm sure u geeks can adapt it for chess.

Will Chess becomeTic-Tac-Toe?

[SuperMario666]

In Tic-Tac-Toe, once the players become sufficiently skilled (usually around age 5 or 6), no one ever wins.

I wonder at what point Chess matches between AI opponents reach this same impasse?

Re:Will Chess becomeTic-Tac-Toe?

[frankthechicken]

I wonder at what point Chess matches between AI opponents reach this same impasse?

After about half a litre of vodka,or in computer terms, about Windows ME.

Re:Will Chess becomeTic-Tac-Toe?

[buswolley]

I believe it has been proven, theoreticly, that white wins in the *perfect* game. Proven using complex mathematics.. Its not a map to the perfect game, but that being first means you win.

Re:Will Chess becomeTic-Tac-Toe?

[Frodo2002]

I find this hard to believe. Do you have a link, some evidence?

Re:Will Chess becomeTic-Tac-Toe?

[buswolley]

I am sorry. I just remember reading about it *somewhere*. Ill try to do some research into it. I know unsupported claims don't make it far at /.

well maybe some do.

geeks need to brush up on their history

[Anonymous Coward]

troll my ass - that was pretty clever!

Yeah, yeah.

[Anonymous Coward]

Ivan will show up stinking vodka drunk and there goes your fine theory. And anyway, he aint playing against gnuchess running on Linux, you dork.

Wrong

[Anonymous Coward]

You were doing very well up until you gave Go no more than 20 years before it is cracked.

If you use current strategies and scale according to Moore's law, then Go on a full sized board is literally safe for centuries. This is easy to verify. On a 19x19 board the branching factor is several hundred and naive evaluations only show up many dozens of moves later. Consider a mere branching factor of 100 with advantages being recognized a dozen moves later. That takes 10**24 which is roughly 2**80 evaluations. With Moore's law you improve by a factor of 2**80 in about 120 years. To brute-force several dozen moves forwards you will need literally centuries.

That is assuming that Moore's law lasts that long. Which classical computing can't without breaking physical laws. (Quantum computers could do it, in theory. But there are considerable issues there.)

Until a completely different strategy can be identified, top human players will continue to have nothing to worry about from computers in Go for the forseeable future.

Re:Wrong

[Moridineas]

Good point--but my feeling is that within the next 20 years SOMETHING big will change in terms of computing power. Maybe it will be massively parallel computers, maybe quantum as you mention, maybe something totally unforeseen. Also I'm predicting AI techniques FOR the game of Go will increase in their effectiveness. Just my guess though, it's really up in the air.

Re:Wrong

[cheezehead]

Until a completely different strategy can be identified, top human players will continue to have nothing to worry about from computers in Go for the forseeable future.

Hmmm. That's what they said 20 years ago about chess. Note that chess programs do not calculate every possible move, they weed out branches as early as possible. The evaluation algorithm is what counts, as well as the fact that computers have become much, much faster in the past 20 years.

What I find telling about this match, is that the DeepFritz team was disappointed with the draw! Not so long ago, a draw against the world champion was regarded as a huge victory!

Also, consider that chess is a more popular game than Go, and that a lot of research has been done with regard to chess algorithms. If the same amount of effort were to be invested in development of Go algorithms, then I'm not so sure that a computer could not beat humans at Go in the forseeable future.

Re:Wrong

[Anonymous Coward]

Until a completely different strategy can be identified, top human players will continue to have nothing to worry about from computers in Go for the forseeable future.

Hmmm. That's what they said 20 years ago about chess. Note that chess programs do not calculate every possible move, they weed out branches as early as possible. The evaluation algorithm is what counts, as well as the fact that computers have become much, much faster in the past 20 years.

Not by my understanding of the history. In the 1970's it was first learned that every additional ply that the computer searched was worth about 200 rating points. That fact combined with Moore's law made it inevitable that computers would one day beat any human. And in fact made reasonably accurate order of magnitude estimates plausible.

Program improvements have brought that victory forward by a few years, but they have not altered the basic fact. Increased CPU means deeper searches, means improved strength. A program from 1982 on good modern hardware would be a grandmaster player. A program from today on good 1982 hardware would not.

The problem with Go is that if there is a similar linear relationship between search depth and strength, Moore's law is so slow that the human advantage will last centuries rather than mere decades. And much of our chess research doesn't help much because in chess it is enough to have a bad evaluation and focus on increasing the depth of the search.

Re:Wrong

[Anders1]

No it's not. There's no limit to the number of moves in a game of chess; you can spend all eternity moving pieces back and forth if you like.

Actually, the 50-move rule allows you to put an upper bound on the number of moves in a game. In particular, there can't be more than 96 pawn moves and 30 captures, and pawn moves/captures can be at most 50 moves apart, meaning that a chess game can't last more than 6350 moves. This could probably be improved somewhat.

I know why!

[Drakkar]

I know why computers have problems with the last 2 examples you gave. But first, I will tell why no.1 no.2 and no.3 are easier to predict. They have been invented by us. Tic/Tac/Toe is a game we invented. We decided the rules! Same thing for the 100m sprint. We have decided long ago how to count time; in our scale. And weather, well comp's are not perfect yet !

But for languages and A.I., we have not, as humans, found the secret of the human body. We are not "Masters" of our bodies yet. God knows the solution! So in that sense, we only programmed the computers with the knowledge we know. Someone will always be able to fool them, by saying something that has never been said before.

My opinion

thanks for ignoring me qjkx

[Anonymous Coward]

I'd just like to thank you for ignoring me first. Anyways, chess is pointless next to Go, and if you don't play Go you can't call yourself any type of geek. As for your question, the answer is simple--when the minimax tree is solved.

chess is the Microsoft of board games qjkx

[Anonymous Coward]

Everybody know about it, but it is vastly inferior. Now, Go, on the other hand...

Thanks for ignoring me, everyone.

to quote a great Canadian band

[SomeGuyFromCA]

Paraphrased, "Even a computer that plays chess doesn't have the brains to run from a fire."

(see the lyrics to "Kasparov vs. Deep Blue" off Moxy Früvous' album Live Noise.)

Re:to quote a great Canadian band

[protonman]

Even a computer that plays chess doesn't have the
LEGS to run from a fire.

Re:to quote a great Canadian band

[Anonymous Coward]

Actually, no. Read the lyrics. Their point is, even if the computer can beat the human at chess, that shouldn't threaten the human because the human's still smarter, as proven by its ability to approach attractive women, run from danger, etc...

Flash

[Britz]

Those suckers use Flash 6 to show the live match. Only Win32 and Mac Flash Clients are available at Macromedia. *nix only goes up to Version 5x

Let's boycot them!

Re:Flash

[WetCat]

You could have easily see the match by logging as a guest (or may be by buying membership if you want) to www.chessclub.com
by using xboard. There is announcement on how to use it.

Re:Kasparov lost... - I'm conflicting here

[targo]

Perhaps they should make it so it can self-compile changed code? Would be hard, but a definate advantage.

Most chess programs are able to learn from their mistakes in a primitive way. They store earlier games and if they lost a game in a certain situation, they'll try a different variation next time.

PGN of the game

[Anonymous Coward]

For those interested in replaying it in Crafty or their Chess program, here is a PGN of the game:

[Event "Brains in Bahrain Man-Machine Match"]
[Round "1"]
[Site "Manama"]
[Date "4.10.??"]
[White "Fritz, Deep"]
[Black "Kramnik, V."]
[Result "1/2-1/2"]

1. e4 e5 2. Nf3 Nc6 3. Bb5 Nf6 4. O-O Nxe4 5. d4 Nd6 6. Bxc6 dxc6 7. dxe5 Nf5
8. Qxd8+ Kxd8 9. Nc3 h6 10. b3 Ke8 11. Bb2 Be7 12. Rad1 a5 13. a4 h5 14. Ne2
Be6 15. c4 Rd8 16. h3 b6 17. Nfd4 Nxd4 18. Nxd4 c5 19. Nxe6 fxe6 20. Rxd8+ Kxd8
21. Bc1 Kc8 22. Rd1 Rd8 23. Rxd8+ Kxd8 24. g4 g6 25. h4 hxg4 26. Bg5 Bxg5 27.
hxg5 Ke8 28. Kg2 1/2-1/2

Go cannot be solved by brute force alone.

[Anonymous Coward]

Go cannot be solved by brute force alone, so it really doesn't matter how strong the computers are that are trying to play Go. Until you get a decent AI computer, you will never see a great Go program.
Here is a recent nytimes article which explains it nicely.

In an Ancient Game, Computing's Future
By KATIE HAFNER
Early in the film "A Beautiful Mind," the mathematician John Nash is seen sitting in a Princeton courtyard, hunched over a playing board covered with small black and white pieces that look like pebbles. He was playing Go, an ancient Asian game. Frustration at losing that game inspired the real Mr. Nash to pursue the mathematics of game theory, research for which he eventually won a Nobel Prize.

In recent years, computer experts, particularly those specializing in artificial intelligence, have felt the same fascination -- and frustration.

Programming other board games has been a relative snap. Even chess has succumbed to the power of the processor. Five years ago, a chess-playing computer called Deep Blue not only beat but thoroughly humbled Garry Kasparov, the world champion at the time. That is because chess, while highly complex, can be reduced to a matter of brute force computation.

Go is different. Deceptively easy to learn, either for a computer or a human, it is a game of such depth and complexity that it can take years for a person to become a strong player. To date, no computer has been able to achieve a skill level beyond that of the casual player.

The game is played on a board divided into a grid of 19 horizontal and 19 vertical lines. Black and white pieces called stones are placed one at a time on the grid's intersections. The object is to acquire and defend territory by surrounding it with stones.

Programmers working on Go see it as more accurate than chess in reflecting the ineffable ways in which the human mind works. The challenge of programming a computer to mimic that process goes to the core of artificial intelligence, which involves the study of learning and decision-making, strategic thinking, knowledge representation, pattern recognition and, perhaps most intriguingly, intuition.

"A good Go player could make a move and other players say, `Yes, that's a good move,' but they can't explain to you why it's a good move, or how they even know it's a good move," said Dr. John McCarthy, a professor emeritus at Stanford University and a pioneer in artificial intelligence.

Dr. Danny Hillis, a computer designer and chairman of the technology company Applied Minds, said that the depth of Go made it ripe for the kind of scientific progress that comes from studying one example in great detail. "We want the equivalent of a fruit fly to study," Dr. Hillis said. "Chess was the fruit fly for studying logic. Go may be the fruit fly for studying intuition."

Along with intuition, pattern recognition is a large part of the game. While computers are good at crunching numbers, people are naturally good at matching patterns. Humans can recognize an acquaintance at a glance, even from the back. "Every Go book is filled with advice on patterns of different kinds," Dr. McCarthy said.

Dr. Daniel Bump, a mathematics professor at Stanford, works on a program called GNU Go in his spare time. "You can very quickly look at a chess game and see if there's some major issue," he said. But to make a decision in Go, he said, players must learn to combine their pattern-matching abilities with the logic and knowledge they have accrued in years of playing.

"If you watch really strong players," Dr. Bump said, "some seem to make fairly mundane moves, but at the end of the game they're ahead. Others do spectacular things."

One measure of the challenge the game poses is the performance of Go computer programs. The last five years have yielded incremental improvements but no breakthroughs, said David Fotland, a programmer and chip designer in San Jose, Calif., who created and sells The Many Faces of Go, one of the few commercial Go programs.

Mr. Fotland's program was the winner of a tournament last weekend in Edmonton, Alberta, that pitted 14 Go-playing programs -- including several from Japan -- against one another. But even The Many Faces of Go is weak enough that most strong players could beat it handily.

Part of the challenge has to do with processing speed. The typical chess program can evaluate about 300,000 positions per second, and Deep Blue was able to evaluate some 200 million positions per second. By midgame, most Go programs can evaluate only a couple of dozen positions each second, said Anders Kierulf, who wrote a program called SmartGo.

In the course of a chess game, a player has an average of 25 to 35 moves available. In Go, on the other hand, a player can choose from an average of 240 moves. A Go-playing computer would take about 30,000 years to look as far ahead as Deep Blue can with chess in three seconds, said Michael Reiss, a computer scientist in London.

If processing power were all there was to it, the solution would be simply a matter of time, since computers are growing ever faster. But the obstacles go much deeper. Not only do Go programs have trouble evaluating positions quickly, they have trouble evaluating them correctly.

Nonetheless, the allure of computer Go increases as the difficulties it poses encourage programmers to advance basic work in artificial intelligence. Graduate students produce dissertations on the topic, and a handful of researchers around the world devote much or all of their attention to it.

The game attracts people from all fields. For example, Chen Zhixing, a retired chemistry professor in Guangzhou, China, wrote a program called Handtalk, which dominated the computer Go field for several years. Dr. Bump, 50, whose field is number theory, has been playing Go for 35 years and taught himself the C programming language four years ago so he could write Go software. Mr. Fotland, 44, the creator of The Many Faces of Go has been working on computer Go for 20 years and is chief technology officer at Ubicom, a small semiconductor company in Silicon Valley.

All are very strong Go players, and it takes a strong Go player to write even a weak Go program. Mr. Fotland, for instance, said he had written programs for checkers, Othello and chess. The algorithms are all very similar, and it is not difficult to write a reasonably strong program, he said. Each of the games took him a year or two to finish. "But when I started on Go," he said, "there was no end to it."

Mr. Fotland said that his Go programming was especially weak when he was a beginning player. "A lot of the stuff I wrote was just plain wrong because I didn't understand the game well enough," he said.

Even when skill develops, however, translating it into a program is not an obvious task. "There's a certain stream of consciousness when you're looking at positions," Dr. Bump said. "You might look at 10 variations, but you don't really know what's going on in the back of your mind. Even a strong player doesn't know how his mind works when he looks at a position."

"We think we have the basics of what we do as humans down pat," Dr. Bump said. "We get up in the morning and make breakfast, but if you tried to program a computer to do that, you'd quickly find that what's simple to you is incredibly difficult for a computer."

The same is true for Go. "When you're deciding what variations to consider, your subconscious mind is pruning," he said. "It's hard to say how much is going on in your mind to accomplish this pruning, but in a position on the board where I'd look at 10 variations, the computer has to look at thousands, maybe a million positions to come to the same conclusions, or to wrong conclusions."

Dr. Reiss, who is the author of Go4++, a previous champion that placed second in last weekend's playoff, agrees with Dr. Bump. Dr. Reiss, who is an expert in neural networks, compares a human being's ability to recognize a strong or weak position in Go with the ability to distinguish between an image of a chair and one of a bicycle. Both tasks, he said, are hugely difficult for a computer.

For that reason, Mr. Fotland said, "writing a strong Go program will teach us more about making computers think like people than writing a strong chess program."

Dr. Reiss, who works on Go full time, said he would not think of devoting his time to any other problem. "It's a fundamentally interesting problem, but also it's just the right level of difficulty," he said. "If it was too easy it would have been solved already. If it was fantastically difficult, people might give up in frustration."

"I think in the long run the only way to write a strong Go program is to have it learn from its own mistakes, which is classic A.I., and no one knows how to do that yet," Mr. Fotland said. A few programs have some learning capabilities built into them.

Mr. Fotland's program, for instance, refers to a database of games played by strong players in deciding its moves, and Dr. Reiss's program employs a learning scheme for deciding which moves are interesting to look at.

Dr. Reiss said he had come up with an idea for a new Go program that would learn by analyzing professional games. But to pursue his idea would require too much work, he said, depriving him of time to continue making updates to his current program.

It seems unlikely that a computer will be programmed to drub a strong human player any time soon, Dr. Reiss said. "But it's possible to make an interesting amount of progress, and the problem stays interesting," he said. "I imagine it will be a juicy problem that people talk about for many decades to come."

and of course

[Anonymous Coward]

it'll also depend on how much this weeny will wine before each mach about "fareness" with chess theirs at least 4 different highly obscure rules to the game.

ok, thats some funny shit.

[Anonymous Coward]

mod this man up, and give him a hot dog!

Go

[heikkile]

The problem is more or less what's called in the AI world the branching factor

Yes, the branching factor makes a huge difference. Another, equally important difference is the cost of the evaluation function. In chess it only takes a few CPU cycles to see if the position is mate, or to count who has more material. There is no way to do this in go. Even the end of the game is non-trivial to recognize, and even then it is hard to say who won.

I do not know of any go program that does much of global reading. I know GnuGo does none at all, it "only" evaluates the position once (which includes lots of local reading), and uses some heuristics to propose moves and to estimate their effect. Then it chooses the best. This it can do in a matter of a few seconds.

I believe this sort of approach can be extended quite far, and take good advantage of increasingly powerful computers. But I doubt it will ever be sufficient to beat a professional player.

Perfect chess game some ways off...

[Goonie]

Searching the state space of chess isn't likely to happen in the near future. I gather estimates are around 10^43, which if I've done the arithmetic correctly is around 2^140. You ain't going to be searching that for years yet.

However, it's not necessary for computers to do that to beat world champions. They're getting better all the time with heuristic searches, thanks very much.

Re:Perfect chess game some ways off...

[invenustus]

I gather estimates are around 10^43

Does that number represet the total possible board states? Because if the goal is to create the perfect game, a lot of those states might never be reached.

Re:Perfect chess game some ways off...

[Goonie]

I think that was for legal games (with a game a complete sequence of moves leading to mate or a drawn position, not board states.

It might be possible to reduce that state space quite a bit with some intelligent pruning, but even so I doubt it could ever be explored with a deterministic machine.

Wrong

[Rui del-Negro]

No it's not. There's no limit to the number of moves in a game of chess; you can spend all eternity moving pieces back and forth if you like.

RMN
~~~

Mod parent up!!!

[RockyJSquirel]

This is NOT a man Vs. machine matchup in any way. This, like all the previous ones, is between the human chess player and the human computer scientist.

It's an interesting position... What IDIOT modded it down as redundant!!!!

Anyway, so your position is that until a program can LEARN chess on its own then we're just playing against the programmers, one off.

Rocky J. Squirrel

Re:Mod parent up!!!

[Anonymous Coward]

Well; there is a difference. There are two possible approaches to chess; one based on raw calculations and another based on experience and knowledge (the GM knows that move x doesn't work while the computer calculates that every result from move x is disadvantageous)

A computer scientist would probably be beaten easily by a professionnal chess player. However, give this computer scientist the computing strength of a super-computer, and you've got a usually fair match.

Mod parent down!!

[RockyJSquirel]

cough cough (hacking up blood).

All your jokes are...

Rocky Da Squirel

Deep Fritz's little brother

[Anonymous Coward]

Forget Deep Fritz, try this JavaScript CPU opponent (lil' Fritz?)

http://www.neilpearce.co.uk/chess/index.html

8 processors

[vlad_petric]

The most interesting fact, IMHO, is that Deep Fritz runs on only 8 processors! A draw with the World Chess Champion is thus a major achievement, as, while it probably still uses a brute force approach (minimax is after all brute force), it's nowhere near Deep Blue in terms of computing power. It is definitely revolutionary in that respect.

The Raven.

Second game.

[BitterOak]

Here is the second game, Fritz playing black this time, and losing. Some were puzzled as to why Fritz resigned at this point. I'm not expert enough to see why either.

1.d4 d5 2.c4 dxc4 3.Nf3 Nf6 4.e3 e6 5.Bxc4 c5 6.0-0 a6 7.dxc5 Qxd1 8.Rxd1 Bxc5 9.Kf1 b5 10.Be2 Bb7 11.Nbd2 Nbd7 12.Nb3 Bf8 13.a4 b4 14.Nfd2 Bd5 15.f3 Bd6 16.g3 e5 17.e4 Be6 18.Nc4 Bc7 19.Be3 a5 20.Nc5 Nxc5 21.Bxc5 Nd7 22.Nd6+ Kf8 23.Bf2 Bxd6 24.Rxd6 Ke7 25.Rad1 Rhc8 26.Bb5 Nc5 27.Bc6 Bc4+ 28.Ke1 Nd3+ 29.R1xd3 Bxd3 30.Bc5 Bc4 31.Rd4+ Kf6 32.Rxc4 Rxc6 33.Be7+ Kxe7 34.Rxc6 Kd7 35.Rc5 f6 36.Kd2 Kd6 37.Rd5+ Kc6 38.Kd3 g6 39.Kc4 g5 40.h3 h6 41.h4 gxh4 42.gxh4 Ra7 43.h5 Ra8 44.Rc5+ Kb6 45.Rb5+ Kc6 46.Rd5 Kc7 47.Kb5 b3 48.Rd3 Ra7 49.Rxb3 Rb7+ 50.Kc4 Ra7 51.Rb5 Ra8 52.Kd5 Ra6 53.Rc5+ Kd7 54.b3 Rd6+ 55.Kc4 Rd4+ 56.Kc3 Rd1 57.Rd5+ 1-0

Re:Second game.

[sciire]

Its a clear Passed Pawn End Game to V. Kramnik. I HATE when players dont finish their games.

Re:Second game.

[tlovie]

The game could unfold something like this:
after 57. ... Rxd5 58. exd5 Kd6
White has b4! forcing another passed pawn the queenside. Perhaps play would continue 59. b4 axb4+ 60. Kxb4 Kxd5
Now black cannot ignore the a-pawn since it will queen before black can push his e and f pawns through. Whereas white will simply let black kill the a-pawn and ensure that he queens the h-pawn by moving his king back to the kingside.

Still wrong

[Rui del-Negro]

That's a tournament / match rule. It's not a game rule.

RMN
~~~

Obviously extremely difficult to articulate...

[Kjella]

Tic-tac-toe rules? No problem.
Chess rules? No problem.
Laws of physics (100m sprint)? No problem.
Laws of statistics (weather forecast)? No problem.

Speech rules?
Intelligence rules (Short of simple logic analysis)?

The problem isn't the computer, but rather our inability to formulate and program the rules or adaptive algorithms that our brain work by, but that we take for granted and have no concious access to.

Your body knows how to break down an apple into essential chemical components and convert them to muscle movement (like your heart) and other things. But you can't sit down at a computer and tell it how.

Kjella

Tic-Tac-Toe solved

[gnalle]

I believe that you misunderstood her. He was referrring to tic tac toe which is easily solved by hand

Re:Tic-Tac-Toe solved

[Frodo2002]

Duh, I solved Tic Tac Toe when I was just a wee bairn. I was replying to the dude who thinks that mathematicians have "solved" chess. (which by the way I am pretty sure they have not.)

One more factual error

[JustAnotherReader]

Reigning world chess champion Vladimir Kramnik

The current (i.e. "Reigning") world chess champion is in fact Ruslan Ponomariov, not Kramnik.

Re:One more factual error

[sciire]

Ruslan Ponomariov is the FIDE World Champion NOT The World Champion, The World Champion was defined in the 2000 year and in its final was Garry Kasparov against Vladimir Kramnik. The Classic Chess World Champion is V. Kramnik.
(R. Ponomariov is the first FIDE World Champion who loose against a Woman, a week ago...) Lets go Girls! LMAO

Kramnik Wins Game 3

[Anonymous Coward]

For the third consecutive game, Kramnik takes the computer into a technical endgame. Score 2.5-0.5

[Event "Brains in Bahrain"]
[Site "Bahrain UAE"]
[Date "2002.10.08"]
[Round "3"]
[White "DEEP FRITZ"]
[Black "Kramnik,V"]
[Result "0-1"]
[BlackElo "2807"]
[EventDate "2002.10.04"]
[ECO "C45"]

1. e4 e5 2. Nf3 Nc6 3. d4 exd4 4. Nxd4 Bc5 5. Nxc6 Qf6 6. Qd2 dxc6 7. Nc3
Ne7 8. Qf4 Be6 9. Qxf6 gxf6 10. Na4 Bb4+ 11. c3 Bd6 12. Be3 b6 13. f4 O-O-O
14. Kf2 c5 15. c4 Nc6 16. Nc3 f5 17. e5 Bf8 18. b3 Nb4 19. a3 Nc2 20. Rc1
Nxe3 21. Kxe3 Bg7 22. Nd5 c6 23. Nf6 Bxf6 24. exf6 Rhe8 25. Kf3 Rd2 26. h3
Bd7 27. g3 Re6 28. Rb1 Rxf6 29. Be2 Re6 30. Rhe1 Kc7 31. Bf1 b5 32. Rec1
Kb6 33. b4 cxb4 34. axb4 Re4 35. Rd1 Rxd1 36. Rxd1 Be6 37. Bd3 Rd4 38. Be2
Rxd1 39. c5+ Kb7 40. Bxd1 a5 41. bxa5 Ka6 42. Ke3 Kxa5 43. Kd4 b4 44. g4
fxg4 45. hxg4 b3 46. Kc3 Ka4 47. Kb2 f6 48. Bf3 Kb5 49. g5 f5 50. Kc3 Kxc5
51. Be2 0-1

Kramnik won third game now... with blacks

[sciire]

Deep Fritz - Kramnik,V (2807) [C45] Brains in Bahrain Man-Machine Match. Manama (3), 08.10.2002 1.e4 e5 2.Nf3 Nc6 3.d4 exd4 4.Nxd4 Bc5 5.Nxc6 Qf6 6.Qd2 dxc6 7.Nc3 Ne7 8.Qf4 Be6 9.Qxf6 gxf6 10.Na4 Bb4+ 11.c3 Bd6 12.Be3 b6 13.f4 0-0-0 14.Kf2 c5 15.c4 Nc6 16.Nc3 f5 17.e5 Bf8 18.b3 Nb4 19.a3 Nc2 20.Rc1 Nxe3 21.Kxe3 Bg7 22.Nd5 c6 23.Nf6 Bxf6 24.exf6 Rhe8 25.Kf3 Rd2 26.h3 Bd7 27.g3 Re6 28.Rb1 Rxf6 29.Be2 Re6 30.Rhe1 Kc7 31.Bf1 b5 32.Rec1 Kb6 33.b4 cxb4 34.axb4 Re4 35.Rd1 Rxd1 36.Rxd1 Be6 37.Bd3 Rd4 38.Be2 Rxd1 39.c5+ Kb7 40.Bxd1 a5 41.bxa5 Ka6 42.Ke3 Kxa5 43.Kd4 b4 44.g4 fxg4 45.hxg4 b3 46.Kc3 Ka4 47.Kb2 f6 48.Bf3 Kb5 49.g5 f5 50.Kc3 Kxc5 0-1

Game 4 is a draw

[Pr3d4t0r]

Kramnik,V (2807) - Deep Fritz [D34]
Brains in Bahrain Man-Machine Match. Manama (4), 10.10.2002

1.d4 d5 2.c4 e6 3.Nf3 c5 4.cxd5 exd5 5.g3 Nc6 6.Bg2 Nf6 7.0-0 Be7 8.Nc3 0-0 9.Bg5 cxd4 10.Nxd4 h6 11.Bf4 Bg4 12.h3 Be6 13.Rc1 Re8 14.Nxe6 fxe6 15.e4 d4 16.e5 dxc3 17.exf6 Bxf6 18.bxc3 Qxd1 19.Rfxd1 Rad8 20.Be3 Rxd1+ 21.Rxd1 Bxc3 22.Rd7 Rb8 23.Bxc6 bxc6 24.Rxa7 Rb2 25.Ra6 Bd2 26.Rxc6 Bxe3 27.fxe3 Kf7 28.a4 Ra2 29.Rc4 Kf6 30.Kf1 g5 31.h4 h5 32.hxg5+ Kxg5 33.Ke1 e5 34.Kf1 Kf5 35.Rh4 Kg5 36.Re4 Kf5 37.Rh4 Kg5 38.Kg1 Kg6 39.g4 hxg4 40.Rxg4+ Kf5 41.Rc4 ½-½

Round 5: Kramnik lost

[Anonymous Coward]

[Event "Brains in Bahrain Man-Machine Match"]
[Site "Manama, Bahrain"]
[Date "2002.10.13"]
[Round "5"]
[White "Deep Fritz"]
[Black "Kramnik, Vladimir"]
[Result "1-0"]
[ECO "A00"]
[EventDate "2002.10.13"]
[SourceDate "2002.10.13"]

1.d4 Nf6 2.c4 e6 3.Nf3 d5 4.Nc3 Be7 5.Bg5 h6 6.Bh4 0-0 7.e3 Ne4 8.Bxe7 Qxe7 9.cxd5 Nxc3 10.bxc3 exd5 11.Qb3 Rd8 12.c4 dxc4 13.Bxc4 Nc6 14.Be2 b6 15.0-0 Bb7 16.Rfc1 Rac8 17.Qa4 Na5 18.Rc3 c5 19.Rac1 cxd4 20.Nxd4 Rxc3 21.Rxc3 Rc8 22.Rxc8+ Bxc8 23.h3 g6 24.Bf3 Bd7 25.Qc2 Qc5 26.Qe4 Qc1+ 27.Kh2 Qc7+ 28.g3 Nc4 29.Be2 Ne5 30.Bb5 Bxb5 31.Nxb5 Qc5 32.Nxa7 Qa5 33.Kg2 Qxa2 34.Nc8 Qc4?? 35.Ne7+ 1-0

Fritz evens the match at 3-3

[Pr3d4t0r]

Kramnik,V (2807) - Deep Fritz [A00] Brains in Bahrain Man-Machine Match Manama, Bahrain (6), 15.10.2002 1.d4 Nf6 2.c4 e6 3.Nf3 b6 4.g3 Ba6 5.b3 Bb4+ 6.Bd2 Be7 7.Bg2 c6 8.Bc3 d5 9.Ne5 Nfd7 10.Nxd7 Nxd7 11.Nd2 0-0 12.0-0 Rc8 13.a4 Bf6 14.e4 c5 15.exd5 cxd4 16.Bb4 Re8 17.Ne4 exd5 18.Nd6 dxc4 19.Nxf7!? Kxf7 20.Bd5+ Kg6 21.Qg4+ Bg5 22.Be4+ Rxe4 23.Qxe4+ Kh6 24.h4 Bf6 25.Bd2+ g5 26.hxg5+ Bxg5 27.Qh4+ [27.Qe6+ Nf6 28.f4 Bh4!!] 27...Kg6 28.Qe4+ Kg7 29.Bxg5 Qxg5 30.Rfe1 cxb3 31.Qxd4+ Nf6 32.a5 Qd5 33.Qxd5 Nxd5 34.axb6 axb6 0-1 [35.Rxa6? b2]