A 1500 human has the capability to lose in ANY position, except perhaps with two bare kings.

Who knew that there could be a match more boring than Kramnik-Leko ?

Kramnik's simplifying strategy got him a winning position in Game One, so it wasn't unreasonable for him to try it again in Games Three and Five.

pundit--

Your displeasure might not be a good enough reason for Kramnik to abandon a style that's carried him succeesfully through the last three WCC matches.

Great players dominate the game, but the smartest players find a way to "break" the game.

Standing under the basket swatting away shots, George Mikan broke college basketball. Rockefeller broke the oil game. And by "neutralizing" 1. e4 and routinely stifling tactical geniuses Kasparov, Topalov, and DF10, Kramnik suggests the possibility of "breaking" the game of chess.

After Mikan, basketball adopted the goal-tending rule. After Rockefeller, Congress passed anti-trust laws. Maybe after Kramnik, chess will add a new opening position every year.

Playing the Botvinnik would be insane against the computer. I doubt Kramnik could make 30% in the Botvinnik against Anand, let alone Fritz. He did this when he was naive and the computer was not as strong. I think this play it safe strategy is best, when one can win 1/10 games or so against the beast. Let's face it.... and move on.

I think Kramnik is playing very well. Only one bad move in 5 games. Pretty good.

I think he has to stay with his strategy.

Humans lose to the computer because of the human's emotional attempt to crush the computer. A silly mistake that will loss for sure. Kramnik has to hope that he can find a closed position where his strategic understanding can find some moves that Fritz does not consider. Trying to knock heads in a tactical melee is suicide.

Kramnik has displayed a very deep understanding of chess. Bravo. He has shown he is deservedly the World Champion. Topalov would probably be lucky to go 1 point in 6 games against this program. However, Topalov's style is very effective against other humans but I dont think he would hold up against Fritz.

I am not putting Topalov down. I think he is a very good player but his style is not suitable for play against the computer. just my opinion.

Greg Koster is right-Kramnik has found a way to hold his own against a thing that can out-calculate him ten to one. If we dont like it, we'll just have to change the rules.

Actually, computers are so good at tactical evaluation that it is pointless playing competitive chess with one. Given that, Kramnik is doing great in not being chomped 5-0, the way Michael Adams was some time ago.Proves that brute force computation cannot steamroller positional play. Also, demonstrates how little we know of the workings of the human mind that a human, thinking not more that 2 or 3 moves ahead can hold fort against a computer that can think 18.

Actually, playing Botvinnik variation against the computers is not as crazy as it seems. Some guy on chesspro.ru forums, apparently at least a master, posted a game where he defeated Fritz in Botvinnik Variation.

1. d4 d5 2. c4 c6 3. Nc3 Nf6 4. Nf3 e6 5. Bg5 dxc4
6. e4 b5 7. e5 h6 8. Bh4 g5 9. Nxg5 hxg5 10. Bxg5 Nbd7 11. g3 Bb7 12. Bg2 Qb6 13. exf6 O-O-O 14. O-O c5 15.d5 b4 16. Rb1 Qa6 17. dxe6 Bxg2 18. e7 Bxf1 19. Qd5 Bh6 20. Bxh6 Bd3 21. Qa8+
Nb8 22. exd8=Q+ Rxd8 23. Re1 bxc3 24. Bf4 Qb6 25. bxc3 Bf5 26. f3 Bd7 27. Rd1 Qb7 28. Qxb8+ Qxb8 29. Bxb8 Kxb8 30. g4 Kc7 31. Kg2 Ba4 32. Rxd8 Kxd8 33. f4 Ke8 34. Kg3 Bb3 35. a3 Kf8 36. f5 Bc2 37. Kf4 Kg8 38. h4 a6 39. g5 Bd3 40. h5 Be2 41. h6 1-0

The idea is that in some crazy lines with nonstandard material (like 3 pawns vs. a minor piece), computers are often wrong in their evaluation. So that guy got this position after 32 moves: http://chesspro.ru/guestnew/upload/diags/776109.gif

The engine (black) thought it was a little better, but the reality was that white is winning.

Given that, Kramnik is doing great in not being chomped 5-0, the way Michael Adams was some time ago.

-- Posted by: naren at December 3, 2006 22:15

Adams did not have the enormous benefit of months of playing against the -exact- same engine that he would play in his match, like Kramnik did.

If Adams had been able to play hundreds of practice games against Hydra, as no doubt Kramnik has against his -copy- of DF, then I hardly think Adams would have lost his match 5-0. Adams would have realized that playing normal chess was a recipe for disaster, which he did not find out until the actual match.

Kramnik had this -enormous- advantage over Adams so it is hardly surprising that his results are much better.

The engine (black) thought it was a little better, but the reality was that white is winning.

-- Posted by: Russianbear at December 3, 2006 23:47

Yes, this is how you beat the computer. You play into positions that it misevaluates, as in the excellent diagram that you posted above.

Forget Kramnik. Petrosian and his exchange sacrifices would have given the computer fits.

And that's why I think Topalov may have a chance. Topalov also loves to sacrifice the exchange and the computer could easily misevaluate the resulting position with its material imbalance, like the Botvinnik variation that you gave above.

This match is the most boring match I have ever had the misfortune to witness. I'm not sure if I will even watch the final game...

Anyone else think Fritz will win by +2 at this point?
Regards

>>Your displeasure might not be a good enough reason for Kramnik to abandon a style that's carried him succeesfully through the last three WCC matches.<<

Sure. Then again, Kramnik's success in the last three WCC matches might not be a good enough reason for pundit to enjoy his chess playing style.

I will not avoid pointing out the obvious, such as that Kramnik's dominance of Kasparov was limited to two games of a single match, that he was more often than not at a disadvantage coming out of the opening against Topalov and won each time more due to the latter's late-mid and early-endgame stumbles, and finally that the only thing Kramnik has so far proven against DF10 is that it's still possible not to get blown out by a computer.

Here is a question for those more familiar with the match's rules, what's to stop Kramnik from repeating the moves of a previous game?

The first game was still Kramnik's best. Kramnik was superior in this game because he knew the computer wouldn't play Bxf3 which would end in a completely drawn position. Since the comp didn't play it Kramnik reached a serious advantage with playing Ne1 and exchanging the white bishops.
In the following games he didn't reach a position where the computer could have made bad positionell choices. The exchanges didn't help any party.

Hmm... maybe Kramnik play could be commended, ... but he only got one superior position in the first game, and was sweeting in the others.

Compare with Kasparov-Deeper Blue game 1, Kasparov-X3d Fritz game 3 (fine closed position maneuvers), or Kasparov-Deep Junior game 1 (fine demolition right after a novelty).

I'm not even sure the technical endings thingummy is a real such a brillant strategy... many moves, and many opportunities to slip or blunder for an human (like Kramnik faild to convert a winning position in game 1)... anyway Kramnik will always play Kramnik-style.

Kramnik played well again. He played for win but, alas, it just doesn't work anymore as in good old times when comps where blind to long term strategies.

17.h4 began a postional K-side attack and one could have assumed few years ago that Fritz would play some meaningless moves and "see" the danger when it would have been too late.

No such luck however, Fritz played b5 liquidating Kramnik's weak pawn and damaging its own pawn structure for the sake of getting counterplay on Q-side. You got to admire somebody taking the right way when at crossroads, when a difficult decision with many trade off is to be made.
Comps are really all round good nowadys.

Nice and logical anti-computer try from Kramnik but it just didn't work. I love the guy, he should have become an accountant.

Oh rubbish, zarghev; Kramnik was much better in game two.

Yuriy; the operators can change up to ten ply in the opening book of any variation played before. So they can stop him if they want.

Russianbear's game is all theory if I'm not mistaken: a game van Wely-Smeets (at least up to move 25/26 or so; not sure about the rest). And he's right that Fritz hasn't got a clue about it; I could beat it like this myself if I got lucky with its opening book.

I'm still very surprised Kramnik didn't play the Berlin; this is another opening about which computers have no clue. Maybe he thinks they will have cured that by beefing up its opening book there.

I wonder if the idiot who thinks Kramnik's been training against Deep Fritz for months actually noticed Elista. I'm guessing that took priority over training for an exhibition match like this.

I attended yesterday's game and have made another set of pictures and uploaded the best to Flickr:

http://www.flickr.com/photos/schwarti/sets/72157594403819156/

Have fun!

Andreas

I hope this is the last man vs. machine event - utterly uninteresting.

As Kramnik keeps repeating in his press conferences he's trained for about 2 weeks for this match. Compare that to the Fritz team who've had months to tune their opening book for Kramnik (doesn't make showing Fritz's screen in the opening seem so bad).

Also interesting... in the press conference after game 5 (http://chesspro.ru/_events/2006/fritz13.html - in Russian), Kramnik says he played 60 games with the computer at a 10-min time control. He won 2 and made "a lot" of draws, so, as he says, the way the match has gone here hasn't been a great suprise.

I don't understand the criticism. Anyone who tries a gung-ho approach against the computer's doomed. If you're going to go for heavy complications you need to have it all worked out at home (also a possibility if Kramnik had more time to prepare - or right up Kasparov's street). Topalov-like attacks - seizing the initiative, putting pressure on your opponents and hoping blunders go unnoticed - are suicidal against computers. Fun as it might be to watch ;)

Good game today, a very interesting example of open dynamic, yet ultimately positional gameplay. Looked like the opposite of Game Four with this time Kramnik having an advantage and Fritz neutralizing it and perhaps even ending up slightly better at one point (certainly with more opportunities for Vlad to err than in G4). I will leave it to stronger chess minds than mine to tell me how much trouble either party was in.

Interesting decision by computer to play for a draw. How does it make decisions like that? Does that mean it thought Kramnik was better or that it's own chances of winning were low? Does DF even have another way of proceeding after 31. Rh4?

Translation of some of the comments made by Kram here:

http://www.chesspro.ru/_events/2006/fritz13.html

Mathias says DF for the first time thought it was worse off coming out of the opening.
Kramnik disagrees: "What, was I better off? I don't understand why people say this. I never saw White as having an advantage today."
Q: Compared to the human matches, is this more difficult?
A: It takes greater concentration. That's obvious. Takes much more energy than when you play people, but that's ok, remember, I am a professional. But I feel fine. And I think I am playiing well. Except for the event in the second game.
Did you lose your best opportunitiy today?
A: Why no. I didn't see myself as having an advantage at all. It was sharp, approximately equal play and counterplay. I think it turned out to be a clean and quite good game.
Q: Even showing your best game, you can't beat the computer. Is it impossible?
A: Well, no... It's very hard to beat the computer, but it is possible. Maybe I will be able to do this in the last game. I will tell you a little secret. I have played with this program many games (60 or so) in rapid time control, 10 minute each. That's much harder for a man than longer time control. And I played with black every time. Well, I managed to win only twice out of the 60 games. Lots of draws. So I wouldn't be surprised if there are many draws here in this match. I knew that if I played well there would be many draws. If I play badly there will be few draws. *laughs*

Says that the most you can win against a computer with such a short match is one game. "Maximum." And he says he still has this chance.

Q: Is the time of anti-computer chess gone?
A: Yes, I try to play like I usually play against people. And the computer, btw, also plays like a regular strong GM. The time of all these special "anti-computer" tricks is gone.
Q: Do you regret being unable to fully prepare for this match?
A: Well, what is one to do when things are as they are. Of course I hope we will have another match like this and I will have a couple of months to prepare...But, since I only had two weeks, whether I regret this or not, it doesn't matter. I prepared in physical, psychological and chess terms as best I could, considering how much time I had.
Asked if he thinks computer progressed strongly since the Bahrain match.
Says he thinks is progressed considerably and this can be shown mathematically, but the play has been equal right now. Says he had chances to win in the first and second game, and you can't say that he had no chances. "I can't speak for others but I personally can battle a computer. It's too early to speak about the far future."
Asked about Topalov rematch, says he hasn't heard anything official from FIDE, so it's just talk for now.

An idea for those who think man-machine matches
have been killed by opening theory. Play all
games at odds of pawn and move. It will still
be recognizable chess and there will be lots more wins. Or if you think this will make the games too one-sided, just remove both queens knight pawns at the beginning. Humanity won't be at the
disadvantage that Fischer random confers.

> Here is a question for those more familiar with the match's rules, what's to stop Kramnik from repeating the moves of a previous game?

The fact that Fritz's moves have a little randomness on them. From the rules:

"The Deep Fritz Team is not required to disclose the exact hash table size for the match. It is understood that hash table size does not influence playing style but rather introduces a small element of non-determinism into the move selection process. The Deep Fritz Team has to notify the Arbiter of the Hash table size so that they can reproduce the programs calculations."

>An idea for those who think man-machine matches
have been killed by opening theory....>

They have not been killed by it but rather they are still happening precisely because forcing the comps to follow the theory allows humans to bypass the middle game and reach directly the endgame, reach simplified postions where they still have a chance to draw.

Well let's hope the programmers don't catch on and turn off the machine's opening book altogether then, eh, Ovidiu?

They know very well what is going on and why Kramnik demanded such thing. But they accept it
(and hope that no one notices or at least no one who counts) since it is part their agreement with Kramnik. It was necessary so as to have this match being held and seem "balanced", seem a "match".

It is not to test anything about computer-human chess, it is a big publicity show for Chessbase acted by their poster boy.

It is like a big, money making, flashy Hollywood movie : a lot of hype, a lot of patzers artwise watching it, and little, if anything, as art content.

For all of Ovidiu's numerous posts I still don't understand what his objection to the match rules is. That Kramnik is able to see what line Fritz will follow in a given opening? Sorry, but you could counter argue that otherwise Fritz has an advantage by having access to world's opening database knowledge. As for quality of the "art" of the match, it's relative. I enjoyed watching the games so far and so have, seemingly, a lot of the higher-rated, more chess analysis-oriented posters on here. I will take this match over Sofia anyday.

> I will take this match over Sofia anyday.

enjoy then Yuriy, taste is not a matter for arguments and it is not my intention to spoil
your fun, vox populi vox dei

I don't understand this opening book thing at all. I thought it is just for Kram to recognize when Fritz is out of the book and has to calculate itself.

Ovidiu,

"...it is not my intention to spoil your fun...."

For the past ten days you've been re-repeating the same points:
1) the match rules are tilted in Kramnik's favor 2) to give him and DF10 a roughly equal chance of success
3) Kramnik and Chessbase benefit from the match
4) Kramnik is employing a sensible strategy against the computer and
5) you are bored

If your point is not to spoil Yuriy's fun, then what exactly IS your point?

Mig identified your IP address as coming from Neptune. Did you by any chance move there from Bulgaria?

According to the opening book. In game one Fritz was out of the book after move seven by Kramnik. Then it began to calculate, but, surprisingly, it made moves that pulled it later in the book again. A move like c5 played by Fritz is a normal move that also a human being would play. Therefore, where was the difference in this position if it played with book or without book? The result would have been the same.
Today's computers do not make silly moves like h5 like they did twenty years ago, even without book.
KLramnik has an advantage, but not because he see's the book, his advantage is he can chose between moves that are sharp or solid. Fritz always plays the best move it discovers whether Kramnik is in time trouble or not.

Ovidiu,

Thanks for a classy reply. Don't worry, I understand that when the same subject comes up for discussion a person may have the same thought.

2 wins out of 60 against Fritz with only the black pieces in 10 minute games is actually somewhat impressive, and makes me kind of surprised that Kramnik hasn't found a way to win one of 5 games at a long time control, 3 of which were with white. Of course, the small sample size has a lot to do with that too.

I wonder if Kramnik thought it might also be helpful to prepare with white as well, since that's half the match.

Samer,

Don't think Vlad meant that was the extent of his preparation--just a little trick he employed as part of it.

I think Kramnik felt he was on the verge of winning game 2 when me make the fatal error.

I think Kramnik is playing very well. The little error of mate in one was simply a little tidbit to keep all the chess fans arguing and fighting to the death.

>I think Kramnik felt he was on the verge of winning game 2 when he made the fatal error.>

I think this too. He played Qe3 calmly, had 30 min left, and without mate he was clearly won after the Qs exchange. However with only 2 wins out of 60 games one should have been more suspicious.

Better, worse ... these assessments of positions have a different meaning for a computer than it does for a human. In each case its an estimate of the odds (probability) of finding a winning variation in the future (or conversely, being forced int a losing variation). The challenge for human's is that the odds have to be fairly strong for them to have a chance to win .. and it gets worse as computer's calculating powers increase. But those strong odds of the existence of a winning variation often bring with it similarly higher odds of the existence of a losing variation ... one which the computer will find! Kramnik is using the right strategy.

Yuriy,

Many thanks for the translation.

"For all of Ovidiu's numerous posts I still don't understand what his objection to the match rules is. That Kramnik is able to see what line Fritz will follow in a given opening?"

And I still wonder how Kramnik is supposed to be able to see this. ESP?

It will be interesting to see if the Fritz team will use this opportunity to play e4 instead of d4 for this final game. Kramnik needs to go for the win, so playing for a draw against an e4 from Fritz is probably not his plan if he has any pride. We might get the sharp tactical game we wanted to see from Fritz if they use this opportunity. I am hoping the Fritz team will play e4 to show A) they really want to win this match (they will either get the open game they want, or an easy draw and match win), B) to see how Kramnik responds to it (sharp Sicilian and I would respect him even if he loses again, Petroff or Berlin, and he is a wimp), C) a cut-throat game of tactics this match has been missing.

If however they want to be 'friendly' to Kramnik (or this is a big publicity stunt, or any other conspiracy theory), they can throw out another d4 opening that will help Kramnik get the type of game he wants and can more easily control. So for me there is some intrigue left in the match, the very first move of Game 6 being one of them.

Playing the Berlin against 1.e4 would seem to be the best way (or a good way) to play for a win, a sharp Sicilian a good way to lose. I certainly don't hope for the latter, it could get ugly. A nice Berlin would be welcome.

Today is going to be the most interesting game since I'm sure Kramnik will play for the win. Good luck, Vladdy.

Well I think Team Fritz can avoid the Berlin and get a sharp game if they play for a Scotch opening instead of Bb5. I think I remember Kasparov liking the Scotch game with White so from that I would think it leads to sharp play. I don't know enough about the Petroff or if its easy to get into sharp positions if Kramnik chooses 2..Nf6 or what can be done from there. All I know is that from recent Petroff games it seems like it gets down to a drawn endgame by around move 25 or so after a flurry of piece exchanges.

I just want to see something sharp and with lots of pieces and maybe queens around for a while. What interested me most was seeing a World Class player take on the newest Fritz in a tactical battle. So far I don't really think that has happened. Team Fritz has its best chance to make this happen IF they want to put Kramnik and their own beast to a test. If they play d4 and we get another Catalan or Slav or the like, I would see it as a real opportunity lost. How many Man-Machine matches are left in history, and how many of those games can the human be 'forced' to play an open game, going for the win? Not many I suspect. I'm hoping for e4 and some guts from Kramnik.

I'm sure Fritz will play 1.e4, as Kramnik's Petroff or Berlin are really designed as drawing defences for Black. Which means he'll probably go for a Shveshnikov Sicilian - that could be fun !

David

A lot of people in an optimist mood today
but I don't see Kramnik offering his fans anything else than usual. Let's hope however from CBase.

The Berlin may have been "designed as [a] drawing defence" against Kasparov, but it definitely offers plenty of imbalances and a strategic complexity where Kramnik should be far superior to the machine. Still very hard to win, needless to say, but it's got to be one of the best alternatives.

Kramnik seems to have prepared Petroff (a left over from his preparation against Topalov ?) and it won't be reasonable ( from some point of view) for CBase to try 1.e4, save that they want to play 2.Nc3-Nf6 3.f4 or something.

1.d4 and QGA again seems probable.

I too have a feeling Fritz will play d4, despite the fact it stood very well in game four and very badly in game two.

If it does go e4 then surely either a Berlin or a Sveshnikov. I would have thought the Berlin gave better chances than a 9 Nd5 Svesh (assuming the machine's not interested in the Pono-Kramnik stuff, and even if it is, perhaps.

I believe Kramnik will play for a draw as a damage limitation exercise and then claim he only lost the match because of a one-off blunder. I think Fritz will probably play 1.e4 as it did better with this. If so, I predict another Petroff as this is more watertight for a draw than the Berlin. I think it is most unlikely that Kramnik will play a Sicilian.

Najdorf Sicilian. Bravo Kramnik, however this turns out! Now if I can only stay awake for it all.

Well, how wrong can a forum be? Wonder when Big Vlad last played the Najdorf as Black? Stern will be happy, anyway.

The Petroff has worse statistics at elite level and far worse in Kramnik's personal practice than the Berlin, so I don't know what this 'more watertight for a draw' nonsense is.

Acirce, here is your answer from the rules of the match, as discussed in detail on this board:

"The computer will consult an opening book during the game. During the match, the opening book may not be modified, except that up to 10 ply of additional moves may be added in the opening variation of the game which has most recently been played (not counting adjournment sessions) and the weightings of specific moves may be modified so that the different variations, already present in the opening book, will be preferred by the program."

"As long as Deep Fritz is “in book”, that is playing moves from memory and not calculating variations, Mr. Kramnik sees the display of the Deep Fritz opening book. For the current board position he sees all moves, including all statistics (number of games, ELO performance, score) from grandmaster games and the move weighting of Deep Fritz. To this purpose, Mr. Kramnik uses his own computer screen showing the screen of the Deep Fritz machine with book display activated. "

Yes, but as discussed that doesn't give him any idea what the computer is going to play.

Kramnik playing Najdorf ? This is not a serious game.

why is the machine taking so long for the 9th move? Its out of book already?

seems that is out yes

If I know the probability of you playing each move as well as what you believe a best response is to each move that gives me a pretty good idea of what each move is going to be. Or is the idea the opposite, that Kramnik has access to the world opening knowledge in the form of DF opening database?

So Kramnick sees on his 8th move that Qc7 is not in Fritz's book. Kramnick is prepared, Fritz on his own.

Here is where Kramnick seeing the opening screen on Fritz is used to Kramnick's advantage - we hope! If Fritz "solves" this "novelty", we humans are in big trouble!

Hey 10.Re3 ??
It's a computer move !!

It looks bad

10.Re3 ! the solution.. hahaha ...now Kramnik is out his own book too

It wouldn't surprise me if Kramnik faces difficulties very early in this game.

Now I think 11...Kh8

...0-0

Oh My! Kramnick castles right into the jaws of defeat? This isn't supposed to be helpmate!

10.Re3...i guess Fritz is allowed to see Moroz opening book...

Re3->g3 is not a very impressive plan.

"If I know the probability of you playing each move as well as what you believe a best response is to each move that gives me a pretty good idea of what each move is going to be."

But Kramnik doesn't know what the probability of Fritz playing each move is and he doesn't know what Fritz believes is the best response to each move. That is exactly the point. I know about the rules, they have been as discussed in detail on this board.

Oh, and I see there's a Najdorf...just like I thought...

Don't you get the creepy feeling Fritz will announce a mate in 12, starting with 18. Rxg7 or 18. Bxe6

Kramnik's biggest advantage might be that he can always play a move that he knows takes Fritz out of book.

Fritz doesn't have a clue in this position - Kramnik should win this one based on White's lack of a real plan.

18...Ng8 seems a great move, intending Bf6
Fritz is playing stupid moves since Re3.
Kramnik has an advantage, but it's so difficult to convert it into something concrete...

Yes, he does--he has access to the stats and move weighings for the opening.

Yes, but only from his own side when it is his move.

Ng8 and then Nb1 -

A modern homage to Chigorin and Steinitz!

I think 20.c3 is good for Fritz

18..Ng8! 19.Nb1! where Kramnik loves to redeploy his peices can Fritz be far behind ? ...I suspect that this is an "all comps" game. Danailov will issue a press release on this soon.

Black seems to have equalized,
but white has no structural problems.
Should black try to create some
with, e.g., 20.. a4 and .. Bxb2
giving white two isolated pawns
(but a Qside passer)?

20.c3 g6 Kram is dominating
I hope he will go on this way until he will find something concrete

c4 square is good for the knight.
Maybe Fritz turned the tables....

Fritz is slightly better (after having played the opening as an 1600)...now is the point where Kramnik may begin to go downhill if he doesn't find a plan.

I predict that Black will double rooks on the b-file and
ultimately sac the exchange on b2.

How's that for going out on a limb?

Is Black about to resign here? Or am I just seeing ghosts?

If the former rather than the latter, then what does that imply about all the preceding "live commentary" on this thread?

Why should he be about to resign? Anyway, even if he is, all it would imply is that none of us has a clue about the game at this level, which would hardly be news to me as far as my own comprehension is concerned; don't know about the rest of you.

Good points. Now I see Kramnik wasn't pondering resignation on his move 26, but was figuring how to construct at least a temporary barricade (still rather leaky-looking to me).

Yeah, rdh, the thought did occur to me that if he found a good defense and my last remark got challenged for arrogance, I could just come back with: "Kramnik sees more than I do. What can I say?"

I don't understand Qb7. What was the point of that?

I don't know about resigning yet, but given the meta-situation (White with an advantage, complicated position, Kramnik short on time, opponent is a computer), I would not fancy Kramnik's chances to hold this. I hope he'll impress me though.

Very modern-looking, computerish Sicilian to my eyes (not surprisingly). The machine refuses to go all in and in the end just takes a pawn on the queenside. At least I assume that's it's idea of the end. I don't quite understand why Kramnik needed to leave that en prise - what was the need to put the queen on b7, I wonder.

Kramnik should not resign yet, although his position is clearly worse. A pawn down, weakened kingside pawn structure, lack of piece coordination...It seems to me that DF10 has the pawn and the compensation, as they say.

31.Qxc6 ?! comp like decision, 31.Qh4 would have kept the pressure and won faster..anyway, Kramnik won't save this.

Perhaps resigning is premature. But from my patzer-perspective it doesn't look too good when you're black and white has a dangerous queenside pawn majority, and on top of that is able to calculate a "quadzillion" moves per second.

Queens are exchanged. I think now Kramnik's only chance is to push the k-side pawns. The rook at f3 is awkward and could become a target.

Kramnik has at least escaped into an ending; while it should be lost I wouldn't say it's totally hopeless yet.

Really, Ovidiu? Strikes me as a very human decision (and Susan P, as well, I see).

Ba4 and b3! I love it. These things just play a different game to us, don't they?

Wonder where that a6 rook is going? If this was SomeHuman -v- Ulf Andersson, I'd be predicting ...Ra6xa4 followed by White gradually losing all his queenside pawns.

Ovidiu

How does this game fit into your Kramnik as posterboy its all a marketing ploy by chessbase polemic?

andy,

No need to ask. We'll be enlightened soon enough.

this last one doesn't fit, the other 5 do, look at the global picture

Fritz would have won in style this match, only now we realize what could have been.

Kramnik's team spent the entire night trying to find a way out of Ovidiu's box.

--if he plays Petroff or Berlin he's boring and cowardly
--if he plays Najdorf he's foolhardy and "it's not a serious game."
--if he loses he's bad
--if he draws he's boring
--if he wins, DF10 is bad or the game was fixed.

Ba4 is a giant pawn. I think Black could have reasonable drawing chances.

No doubt in reality the machine is beginning the consolidation process with its customary aplomb, but if you were I were playing one would say that White hasn't covered himself in glory over the last few moves. Black has the d-file; the a4 bishop doesn't look ideally placed, and can one even dream about a knight getting to d4 one day?

Oh well, according to Susan P 36...Bh6 ws a horrible move and the machine has simply been fluently increasing its advantage. Good to know I understand nothing. At least Kramnik doesn't either, though, so that's a relief.

The position is a desaster for Kramnik. I didn't understand his whole game. Had he ever a plan? I think Ba6 was a bad thing. The bishop was useless there.

If Kramnik figures out a way to escape from this, he should change his name to Houdini.

Kramnik went wrong with 24..Rb6? , after 25.e5! he was practically lost. It has been a 25 moves game, the rest has been Kramnik's agony.

Is it so bad? 42...f6 now and 43...e5; can I dream about a knight to d4 (after ...Kf7, bien sur, but can the machine stop Ne6-d4?)

Oh the humanity!
Would someone please put us out of Kramnik's misery?

acirce,

I don't see anything in the rules to suggest Kramnik is unable to look when it's Fritz's move. The statement on the subject is: "As long as Deep Fritz is “in book”, that is playing moves from memory and not calculating variations, Mr. Kramnik sees the display of the Deep Fritz opening book." Since Kramnik has access to the program, he also can determine these things beforehand, knowing for example that in a Sicilian Fritz will go a certain way, that in Berlin a certain position is very likely to come up, etc.

hmmm ok Kramnik has gone native has he? I was hoping for a more ingenious explanation but I guess I will take your advice and focus on the first 5 games - the contrived missed technical win followed by the theatrical false blunder and three dry pre-arranged hardly knockabout draws. I thought you might suggest this thumping [looks like 20 moves to a lost position]was a devilish attempt to throw us off the scent. You know, give us the impression of a competitive match rather than a collusive farce between the mighty chessbase and their expensive fall guy.

He can look all he likes while it's Fritz's move, but (i) the machine moves instantly while it's in book, and (ii) it doesn't help to know the move your opponent's about to play before he plays it (except timewise). The book only displays options one move ahead at a time.

He didn't have the thing in advance, so Yuriy's last point is wrong. The machine could have been primed to play anything at all. Chessbase have been bragging about hiring a top GM famed for his opening expertise to assist them with the book for the match, after all.

OK, if ....Kg7 was the best move evidently things are terrible.

andy, I will reply more on this game after is over, now let's see the finish of this one.

Kramnik used the opening book trick again in this game just as before but it was not enough.

Fritz did not have Qc7 in the opening book and Kramnik saw this an played it.
What he hoped for actually did happen, just as when it happens when you play Sicilian against an 1600 player : he plays as if against 1..e5 (no f4 or g4 and pawn assault etc.) and the result is that you get to play not a sharp game but a postional one with Black being better with his free control of the center.

However after the opening Fritz simply outplayed Kramnik, and in style. He played the middle game after 18. Bc1 as a +3000 and Kramnik went wrong with 24..Rb6 and was lost after 25.e5! (maybe even sooner he went wrong : the plan with 15..Rb8 and c5 was weaker then that with 15..d5)

What's to stop the A pawn from marching north unless black sacks the bishop?

What's to stop the A pawn from marching north unless black sacks the bishop?

Great game. The best of this match and a great game for Fritz (and for justice as it eventually prevailed against the backstage schemers).

Ovidiu,

Let's try to be consistent. DF10's creator and Kramnik are evil, backstage schemers. Thus the only just result would have been if both DF10 and Kramnik had lost.

So will we now have The Fritz Attack in the Sicilian, characterized by Re1 Re3 and Rg3?

Grandmasters everywhere booking up on the irrefutable Fritz Attack, and the Najdorf goes out of fashion?

No talljen, I have just posted a reply to Andy over what happened, After 8..Qc7 Fritz was out of the book ( Kramnik saw that it did not have it) and started computing and played a bad Sicilian as pazters always do, and what Kramnik hoped for.

But it was not enough, Qs where still on board, and Fritz played awesome from 18th on.

Ovidiu has taken my earlier observation made in real-time (that Black was dead after 25.e5!), and tried to make a shotgun wedding between it, and his personal theories about how Kramnik exploits his access to what is and isn't in the computer's opening "book".

The problem with doing that is, Ovidiu is also endorsing the naive comments of various others (such as "fogof") that White, once out of its book, played "like a 1600" into the early middlegame.

These people trashed Fritz's Re3 and subsequent maneuvering for a piece-attack against the Black King. However, that plan was, and is, perfectly reasonable, even normal.

In general, a Rook lift is hardly a novel or overly risky concept (although it can backfire if taken to an extreme -- see Radjabov-Anand, Dubai 2002 for a particularly instructive and spectacular example). Even in the Najforf sub-line with 6.Bc4, attacking the K-side with pieces instead of a pawn storm is actually widely played these days. For instance, in popular ...b5 lines, I think the e White plan of Qd1-f3-g3 followed by Bh6 (to force the f6-Knight back to e8) and then piece maneuvering, eventually displaced the formerly automatic f2-f4 followed by either f4-f5 or e4-e5.

Finally, those who contend White adopted a poor plan or no plan after Kramnik took it out of book with ...Qc7, have a heavy burden of proof to meet: Since their spokesman Ovidiu now concedes (belatedly it seems) that White was winning by move 25, where then did the tide turn? If the machine's attacking plan really was so foolish, how then did it reach a won position just 15 moves later? -- and against the widely acknowledged (even by Kramnik-trashers like Ovidiu) defensive wizard of our time?

But ovidu where will the black Najdorf player deviate from what Kramnick played, in what was a rarely examined, moribund Bc4 Najdorf?

Jacobs, your "priority" aside, I suggested that the plan with Rb8,c5 while undertandable was, seems to me, weaker than that with d5 direct (instead of Rb8).

Kramnik just did not know what to do in an unfamilar setup. Similar as the game with Topalav
in Elista after 10.f4 when he went wrong with "b5-b4" and was lost in few moves while the "c5" plan was the way to go, as Golubev wrote in the game analysis.

OK, how could Kramnik not have the book in advance when Ovidiu is able to say below that Qc7 is not in the book? Is the advantage simply limited to Kramnik being able to take the game out of theory (and knowing that he is doing so)?

Fritz showed you clearly how to beat Kramnik in a match: get the lead and make Kramnik play must-win games.

Kramnik looked totally out of it today. He never seemed to have a plan aside from trying to hold on.

In fact, he's spent the last four games of the match trying to hold inferior positions, three times drawing and losing badly today.

Yuriy, at this point I am myself fed up with this "opening book" topic but in short yes,

Kramnik sees what moves Fritz expects from him which is to say what lines it has stored in the opening book. Kramnik can choose to follow one of
them if he knows it and he knows that its end is an endgame or, as in this game, to play something which isn't there at all and thus force Fritz to start computing. After 8..Qc7 Fritz started computing, as those who followed on playchess.com could notice.

And since Fritz isn't capable to rediscover the main strategies of playing the Sicilian by mere computing it played some garbage, as Kramnik expected and, as I wrote, most <1600 player do whne faced with such situation. Kramnik expected postional advanatge, and was not far from it by move 15.

Ovidiu,

Such a scenario is not that different from playing a regular GM. You can either follow a previously known line or take it off the known theory and hope that your novelty gives you a better position.

Fritz may not be able to discover new "strategies" for playing Sicilian, but it certainly is capable of coming up with the best move to make in a position.

Yuriy, a GM is not bound to robotically follow a line if he happens to know it, as Fritz is, and you do not know when you play him OTB what his "book", what are the lines he has memorized, happens to be.

Had Kramnik not been able to see on screen that 8..Qc7 wasn't there he would have not tried it. What if it were ? Sozin-Najdorf gets pretty sharp after 9.f4 (f5) etc.

Kramnik played the opening intelligently again but he however just got confused in an advantageous but unfamiliar position and devised on spot what turned out to be a wrong plan.

That's my opinion, let's agree that we disagree and let it like that.

David Bronstein died today...What a player!

Although Kramnik has played the Sicilian before, it's about as conducive to his style of play as the Caro-Kann is(was)to Kasparov's. What it tells me is that he really wanted to win this last game, but wasn't fully prepared to play that defence at this particular juncture. The last 15-20 moves for Fritz reminded me of the way Fischer used to torture an opponent when he got an advantage [which was quite often], very reminiscent.

Sorry to hear about Bronstein, and yes he was an original.

notyetagm, when was Kramnik worse in game 5? That game was very balanced throughout.

Sorry, off topic but Susan Polgar reports on her blog that FIDE have proposed a new two-year cycle for the WCC, with a 128 player event (not a KO event) ending with a match to decide a challenger for the champion in the next year. It's not prefect, but it's the most sensible suggestion I've heard for a while.

http://susanpolgar.blogspot.com/2006/12/important-changes-with-world.html

looking for Kramnik's better plan, how about 13..e5 (instead of 13..a5 ) ?

1.e4 c5 2.Nf3 d6 3.d4 cxd4 4.Nxd4 Nf6 5.Nc3 a6 6.Bc4 e6 7.0-0 Be7 8.Bb3 Qc7 9.Re1 0-0 10.Re3 Nc6 11.Rg3 Kh8 12.Nxc6 bxc6 13.Qe2 a5

[13...e5 14.Bg5 (14.Be3 Ne8 15.Qh5 f5; 14.Rd3 a5 15.Qf3 Ba6 16.Rd1 Rab8 17.Bg5) 14...Ne8 15.Bxe7 Qxe7 16.Rd1 f5]

Kramnik has a natural anti-computer style against something like Fritzg which is a tactical monster. He probably felt obligated to play fighting chess in the last game to see if he can level the match.

Kramnik always produces when it matters - like the last game at Brissago when he beat Leko to level the match and kept the title.

But Fritz is no Leko. It doesn't get nervous. It's ruthless in executing :-(

I recall a game a few years ago when Judith Polgar played tactical game against a computer (I don't recall which one; Junior I think) and won.

Kramnik is a peace-loving guy. He doesn't like to get into blood curdling tactical lines. We need players like Polgar or Topalov to bring on the fire works :-) (even Kasparov!!)

>Kramnik is a peace-loving guy. He doesn't like to get into blood curdling tactical lines.>

yeah, he played Nadorf and only harmed himself

> He probably felt obligated to play fighting chess in the last game ..>

he had a press conference yesterday, Yuriy translated for us, where he got carried with wishful thinking :

"I can't speak for others but I personally can battle a computer."

Sure, I guess that he ended up believing what he was saying and look what happened.

I guess the sponsors of the match were worried that with this Kramnik strategy, they weren't able to show that DF 10 is a very strong engine. I have the impression that Kramnik played this opening just because they needed to show DF strengths; remember that the primary purpose of the match is to sell a product!! I congratulate Chessbase for their marketing strategy; being able to involve the World Champion in this is full of praise and Kramnik won't be dissapointed with some extra money...

Sounds weird, but this was a win-win situation for everybody...

"David Bronstein died today...What a player!"

Any chance you could post a link to your source? I can't find any news anywhere on this.

Because it just happened today, english sources are not yet available. But the information is 100% correct.

Note: I guess I consider pertinent to say I am not Ovidiu ... :) and I am fan of Kramnik; but I don't like at the idea that the world champion was permissive to use his name in such a idea.

I hope this match would end the idea to compare chess programs with humans; given the fact that in the short term free engines would be stronger than any human player, efforts could be directed to make the engines more useful for the training of an average player.

I was following the game in a chess server and I could realize the tendency of people using the engines to follow the game, so a lot of "+1.42 for white, -2.07, etc...", but how much the current engines can help those people to get a better understanding of the game? I believe a lot more should be done in that direction (I know something has been done, but there are plenty of things to do) rather than just making stronger machines who just squeeze the current knowledge on chess programming without any significative advances in that area...

Sadly, the news about Bronstein is true.

http://surov.livejournal.com/24856.html

Battling, in any language, does not mean tactical, open, Sicilian or Najdorf. A positional or a Berlin game is a battle as well. Whether Kramnik can battle a PC or not is for you to decide. But his quote does not by any way mean "I can outplay computer in dynamic open positions".

Sorry to hear about the death of David Bronstein, an extraordinary player.

I have a book from him about his matches with Botvinnik, as well as his famous "200 Open Games" (I dont have his book on Zurich 1953, which I know is a classic too) and I must say I always liked his fresh and improvisational approach to chess; reminiscent of the present days of blitz all over the world, part of his legacy... Rest in peace

>Note: I guess I consider pertinent to say I am not Ovidiu ... :) and I am fan of Kramnik; but I don't like at the idea that the world champion was permissive to use his name in such a idea.>

Sandor, if I were you who would you be ?

Kramnik got a beating and cashed half million dollars, not bad. Now Topalov will play DF-10 in Sofia and, if he wins, will declare himself WCC and if he loses DF-10 remanins "the man".

> but how much the current engines can help those people to get a better understanding of the game?>
how much did it help Kramnik ? this is not an issue, wait to see the next "move" from CBase :

they will create a test/examination institution ( simliar to Microsoft certificates) "eGM".
Why waist time in tournaments ? Study and train yourself at home with DF-10 and when you feel ready come and take the exam in Berlin and if you pass you will be declared "eGM" (CB) and given a certificate.

In time few such people will prove themselves also in OTB tournaments ( the exam will be serious enough) and this in turn will being respectability to those who have passed the exam but have not played much in tournaments..etc. you get the picture.

CBase (DF-10 in fact) owns now the WC title and Chessbase will take over FIDE too, and over the much coveted GM title.

NY Times is quick to report of Kramnik's looss:

http://www.nytimes.com/2006/12/05/crosswords/chess/05cnd-chess.html?hp&ex=1165381200&en=92caa45502331dee&ei=5094&partner=homepage

Sandorchess,

Get Bronstein's book on the 53 Candidate Matches. You can read it, re-read it, and then read it again and you will never be disappointed. One of the best chess books, ever.

While I completely agree with chesstraveler, I would like to add that Zürich/Neuhausen 1953 was a tournament, not matches.

It was against Bronstein that I played the first game of chess where I kept score: a simultaneous exhibition almost 40 years ago. In a way, it was he who got me started. I will never forget him.

One of the first tribues to Bronstein:

http://tinyurl.com/yjkquo

Charley is right, it was a tournament. I always think of it as a CM because the winner [Smyslov] earned the right to play Botvinnik in 54 for the WC.

Bronstein's death is sad news, but doesn't it seem poetic, or fitting, that Bronstein, such a wonderful player with a truly "human" style, died on the same day that a chess computer has probably surpassed human players once and for all?

The passing of David Bronstein is sad, indeed. Yet another one of the Giants from the Soviet School departs, and the chess world is lessor for his absence.

Interestingly, Bronstein was very much interested in Computer Chess, and how computers would influence the future of chess.

He wrote a fanciful little book in the early 1980s, entitled, appropriately enough "Chess in the Eighties". In it he describes some of his own experiences in playing against Chess computers. They were much weaker then, but still about Master strength.

I believe there was also a game (earlier, in the 1960s?) where he played the King's Gambit against the computer and announced something like a Mate in 10 moves. A really cute game.

Later, in the 1990s, he played in some of those Dutch "Human vs. Computer" events--the AEGON Computer Chess tournament.

Title: Chess in the eighties /
Uniform Title: Prekrasnyi i iarostnyi mir. English
Author(s): Bronshtein, David Ionovich, 1924-
Smolian, Georgii L’vovich.
Publication: Oxford ; New York : Pergamon Press,
Edition: 1st English ed.
Year: 1982
===========================================

Here are some comments by Bronstein, about the 2002 match between Kramnik and Deep Blue:
http://www.chessdate.com/?cd=articles&id=280

Page 28 of the California Chess Journal has an article on Bronstein vs. the Computers.

http://www.chessdryad.com/articles/ccj/pdf/jan_feb_03.pdf

At least, there are some decent collections of Bronstein's games that have been published.

He was a real chess genius, the likes of whom we shall not see again.

Why should Kramnik losing mean that computers have conquered mankind. Few facts:

- Kramnik didn't play well. Maybe he was tired and not well prepared so soon after the match with Toppy.

- This loss doesn't prove that similar or higher rated players like Topalov, Anand, Leko, Ivanchuk will also lose to F10 with better preparation.

- Kramnik may be work champion but that doesn't make him the strongest/top player - his rank is #3/4 and was rated even lower till recently. It is also not clear if F10 is the best computer program.

Let's not fall for the propaganda being spread by chessbase and commentrators paid by it.

Kapalik

The result is another nail in the coffin. I'm not sure that one can draw broad conclusions about the match result. Kramnik lost game two due to a shocking fluke, and he was outplayed in Game 6, when he essayed a defense that he never plays, and for which he lacks "the touch". He would have done better to play a Sveshnikov, which can be equally unbalanced, but is more strategic in nature.

I'd like to see Anand--or even Topalov-- have a crack at this Deep Fritz. Either of them would be better suited for matching up against Fritz.

Let's face it: if the "anti-computer" strategies, which seem to involve insipid openings, aren't bringing good results, then maybe it is time to revert to more maximalist strategies. As White, the Champion of the Humans ought to play for positional and strategic complications, and rely on prepared novelties to obtain a big edge. However, the resulting positions ought not be so sharp that the attack/initiative ends up being "va Banque".

Fritz might not be the strongest of the silicon beasts. It'd be interesting to see Top GMs take on Rybka, or Hydra. It would also be interesting to have an "Absolute" Computer Chess Championship between Fritz, Rybka, Hydra, and, say, Junior. Make it a Quadruple Round Robin, so that the event will be of sufficient length to be meaningful. It's clear that computer championships utilizing the Swiss System have a format that too much influences the final standings.

I would like to thank Kramnik and Chessbase on behalf of the young generation, for completely destroying chess as a profession. From this moment on, I will purchase no chess products, nor participate in chess tournaments. Anyone that tries to make a living playing this game will be battling windmills.

A quad RR with Fritz, Rybka, Hydra, and Junior would be silly. Fritz and Junior would battle each other for the last two places. A 100-game match between Hydra and Rybka--that would be telling.

It'd be interesting to see Top GMs take on Rybka..

GM Christiansen played a month or so ago a match with Rybka on WCN (rapid/25 min), it was 2-0 for Rybka.

>Why should Kramnik losing mean that computers have conquered mankind. ?>

of course, hope never dies, the mythological/eschatological return of Bobby Fischer will save the chesskind and the evil computers will be obliterated.Amen.

A lot of people refer to the "half a million dollars".

Hopefully for Kramnik, it is half a million euros, which is about $666,000.

looks like Bronstein passed away

See http://www.doggers-schaak.nl/?p=627&lp_lang_view=en for an interesting article on computers and the future of chess

Just wanted to say thanks to Andrey, dz and DOug for posting their reports on the passing of GM David Bronstein.

I've just posted this on Mig's ChessNinja Forum: "65th Square - David Bronstein"

http://www.chessninja.com/cgi-bin/ultimatebb.cgi?ubb=get_topic;f=9;t=001256;p=1#000007

It's now been officially reported on 'e3e5' -"06.12.05 David Bronstein (1924-2006)"

http://e3e5.com/petersburg/

A passage from my CN tribute post:

In 1970 David wrote: "A machine which can play chess with people is one of the most marvelous wonders of our 20th century!" and in 1978, he mused about the possibility of an electronic grandmaster. Bronstein, in fact dared not make a prediction, but world champions Botvinnik said, "yes, there will be." Smyslov said, "No", and Tal said, "I hope not."

From my friends over at ChessDryad (Kerry, Richard, Mark & Frisco)... Previously pointed out by DOug above (CCJ, Vol. 17, No. 1 Jan-Feb).

Regards, - Mal
Berkeley, CA

"...I would like to thank Kramnik and Chessbase on behalf of the young generation, for completely destroying chess as a profession. From this moment on, I will purchase no chess products, nor participate in chess tournaments. Anyone that tries to make a living playing this game will be battling windmills.
Posted by: Eo at December 6, 2006 00:54..."

What the hey? You won't buy a well-written, engaging, stimulating chess book ever again just because you didn't like the Kramnik-computer match?

That makes no sense. Play online- have fun, learn some stuff from good books, play in the park- impress the spectators around your park table.

Chess is life at its best!

I dont understand everyone's misplaced fascination with the strength of Rybka... It really isnt that strong.

I agree with Eo, in that Kramnik's loss marks an important milestone in the senility and eventual death of chess.

>That makes no sense. Play online- have fun, learn some stuff from good books, play in the park- impress the spectators around your park table.>

Andy, I think that the guy had a point. I remember my sister giving up piano after listening another girl playing way better the same piece she was exercising. "I will never be able to play like this!" my sister said and that was the end of her piano player career.

So, he has a point, it just doesn't make any sense to do something, or even live life, if others can easily do it or have it better than you.

Let's face it-- let's "lose with dignity"(TM) as Susan would have it or at least pretend so-- the future of chess is secure but it belongs to comps.
It is simply the logical next step of the evolution by natural selection that Darwin
explained to us.

"The future of chess is secure but it belongs to comps."

The Bonn match rules gave Kramnik a handicap to somewhat level the playing field. But would would be an artistically "fair" set of human-comp rules?

--DF10 carried an opening book and endgame tablebases into the match, so a human should be allowed to do likewise.
--DF10 could examine several thousand of the human's old games. So maybe six months before the match, DF10 could turn over to the human a few thousand of its games. Both sides could then improve their play in secret in the half-year leading up to the match.

A century or so ago it might have been fun to watch a man race against an automobile; but those days are long gone. The future of transportation belongs to cars, aircraft, and spacecraft but I still enjoy running and watching track meets.

how about DF gives pawn odds to kramnik?

Post-last game conference:
http://www.chesspro.ru/_events/2006/fritz15.html

Q: Were you surprised at all by Fritz-10 in this match?
A: When I got this program and began to prepare, there was an unpleasant surprise when I realized how strong it was. Of course two weeks is not enough to prepare for such an event. Team ChessBase prepared for about half a year, I suppose. I understood that this will be very difficult, but actually the games turned out better for me than I thought they would. The sixth game today is the only exception...I couldn't manage against a computer. But in reality we played about equal through the match. Of course, I am not satisfied with the score, but i don't know that anybody, in the conditions I was in, could have gotten a better result. And the games, I thought, were interesting and even not bad in quality.

Q: Vladimir you recently told us that you played 60 games against Fritz-10 with black. Did you play the Sicilian even once?
A: No, I didn't...I was actually not intending to play the Sicilian. But in the Petroff it's not easy to win, let's be frank, and I had to do something extraordinary, in one day to prepare an opening in which ten thousands of games have been played...So I went for broke. What could I do? It didn't succeed this time. I don't see particular difference in losing 3.5-2.5 and 4-2...Even today during the game there were a couple of moments when I could have simplified position and there would have probably been a draw, but I didn't want to do that.
Q: The Rook maneuver e1-e3-93, which some commentators have called "childlike" and some "amateur", did it surprise you?
A: Here I must say that the computer "guessed" correctly...Especially against a human opponent it is unpleasant to have a rook on g3 and constantly have a threat against your king...I think ChessBase intentionally prepared this for a human opponent. If against this computer another computer was playing, he would have no difficulty at all in constantly controlling his kingside. But it's very hard for a human: the play is on queenside but there are constantly some threats on kingside. Constantly there are threats of attack on kingside. I used a considerable amount of time to control the entire board and at some moment lost too much time and the computer developed such a great attack. After e4-e5 it was probably impossible to save the game.

Also when asked about the future said he would play such a match again. He said he will have more experience, and will try to find more time to prepare.

Matthias Wullenwebber (sorry, don't know correct spelling with Latin letters) head analytic of Chessbase says they don't anticipate such an event in the next few years. (private comment to Vasilyev, not part of press-conference as far as I can tell)

> The future of transportation belongs to cars, aircraft, and spacecraft but I still enjoy running and watching track meets.>

I do not enjoy such things but I see your point.
Well yes, we can still enjoy watching the fights between humans. It does not matter that they are patzers by DF-10 standards (24..Rb6? 25.e5! +-)
as long they are both of the similar strength and there is the excitement of the struggle and the uncertainty of the result.

>

Yeah, at least we can get candidates matches again with computers.. they don't ask for a million bucks to play... or fuss about the match conditions.. or going to the rest room

Thanks, Yuriy.

I enjoy seeing the battle of strongest chess-playing minds on the planet in a field of chess. I couldn't care less that there are computers who can do it better.

I have known for a long time that I will probably never be the strongest player on the mind and that there are hundreds of players who I will never be as good at. That hasn't stopped me from playing, following or enjoying chess and I don't see why it should anybody else.

If the death of chess does come through computers, it will probably come not through them outplaying us but through their use exhausting the study of theory of chess. The game will not be as much fun if every position has been studied, every opening explored, every line analyzed. It will make the game even more a matter of studying and less of thinking.

Yuriy, I appreciate your honesty, I for one I gave up playing chess seriously at 17 when it became apparent to me that I would not make it big, and I know may similar stories.

Nevertheless you miscalculate. Computers not only play better than GMs but they degrade the traditional respect given to strong chessplayers. They take away the charisma and mystique associated with being able to master such game and, conversely, make chess skill to appear trivial, not an worthy accomplishment.

You will see the effects of comps on chess not now but in a generation as less and less intelectually gifted young men will chose to study it in depth so to be able to battle their cellphones..

"I dont understand everyone's misplaced fascination with the strength of Rybka... It really isnt that strong."

Sure. After all, it's only the strongest chess-playing entity that ever existed on Earth, and would likely beat all World Champions that have existed.

On a unrelated note, now that Fritz has beaten Kramnik, maybe there will be a "post-reunification match" which would be "Fritz-10 vs Kasparov" :-)
(or maybe "Fritz-10 vs Topalov").

As for this defeat being a nail in the coffin of chess, remember that from now on, even more than before, *comments* of Grand Masters will be lose strongly of their value... along with it the prestige of Grand Mazters.

Comments like "chess is being played out", "boring game today", "Kramnik is a patzer" and "computers make chess appear trivial" degrade GMs and chess a lot more than existence of a great engine.

This is a case of perception creating reality.

Chess interest has been waning with advance of technology because of creation of more exciting, colorful forms of entertainment available to the youngmen. I doubt not being able to beat their computer is going to matter now, since most of us haven't been able to do so for over ten years.

> I doubt not being able to beat their computer is going to matter now, since most of us haven't been able to do so for over ten years.>

New York Times was quick to notice it in minutes after it finished. You may wonder way since most of the readers have little clue on playing chess.
But it still made a good news exactly because of that "social" (charisma etc.) dimsension of chess which I mentioned.

>Chess interest has been waning with advance of technology because of creation of more exciting, colorful forms of entertainment available to the youngmen.>

these are "games" and nobody has taken them seriously, nowhere close to how chess has always been regarded.(up to now at least)

There will never been a big news of NYT who won the championship on "NeedforSpeed 4", it would sound as a joke. Chess however is deeply embedded in the "social fabric" of the western world. It is a game too of course but it has a different cultural history and significance.

I am not sure that you see me point but I won't start another heated argument as that with the opening-book.
I do not intend to degrade chess, I still play myself and in tournaments, I am making a point which I think is first factual and only as a secondary problem unpleasant too.

Yuriy's point is well taken.

Chess should have received a great boost from the computer and internet revolutions. But chess was too busy committing institutional suicide in the critical years 1993-2006: alienation of big bucks sponsors, a split title, and the collapse of the WCC structure.

As a result the public cares as little about chess as they do about another once popular sport with similar leadership problems: how many people know (or care) who's the current heavyweight boxing champion?

It is painful to imagine how prosperous the chess world could be today if the old WCC structure had continued in place and if big bucks sponsors had been courted and welcomed rather than shunted aside and slimed.

"Fritz might not be the strongest of the silicon beasts. It'd be interesting to see Top GMs take on Rybka, or Hydra. It would also be interesting to have an "Absolute" Computer Chess Championship between Fritz, Rybka, Hydra, and, say, Junior. Make it a Quadruple Round Robin, so that the event will be of sufficient length to be meaningful."

You're forgetting that the top GMs don't play for free. It won't happen unless someone is willing to pay for it, and I'm not sure who would. Chessbase was able to find sponsorship, because they're the Microsoft of chess. That's why Fritz was in the match, and not Rybka.

I am not sure how many more of these big-money matches are left. They make sense only if the humans have a chance, and right now it's looking like they don't. As it is, Kramnik benefited from special rules he'd never have against a human opponent, and he lost anyway.

What's more, the computers are improving rapidly, but the best human isn't going to get much better than Kramnik. We are fast approaching the time when these matches will be pointless, because the human has virtually no chance.

I agree with Greg Koster that human-to-human chess will remain interesting. Track meets didn't go out of fashion, just because no human can outrun an automobile.

I notice that noone online wants to play against computers- only humans. Playing against computers is no fun. I only do it for training purposes - and I'm sure most would agree.

Kramnik just did it for the money, as would any rational person.

Furthermore, I will continue to enjoy playing through game collections of the most brilliant human chess games.

Can anyone tell me how many copies have sold of that book (if it even exists ) "The 100 best computer versus computer chess Games" ?

uhh.. three copies sold worldwide? I thought so.

"I agree with Greg Koster that human-to-human chess will remain interesting. Track meets didn't go out of fashion, just because no human can outrun an automobile."

People, you miss the point here (or, at least part of it)! Noone ever suggested that it's kind of the "essence" of humans to be able to run fastest in the world. So, that a car (or an elephant, jaguar, even alligator) can outrun us, is no "threat" to our specialness.
We do feel however, that at least part of what seperates humans from crocdiles is our ability to think, plan, imagine, dream etc, i.e. our cognitive abilities.
If chess, deemed by some to be a valid way of "measuring" those abilities, is no longer something we excel in, compared to machines, that changes part of how we have to look at ourselves.

Or so the argument goes.

I'm not saying that I support that view, just that the quoted argument is no way of coping with that difficulty.

>People, you miss the point here (or, at least part of it)! Noone ever suggested that it's kind of the "essence" of humans to be able to run fastest in the world.>

Thanks for this point Albrecht !

Al,

Nicely done.

Part of the chess' interest lies in exploring a new path. Chess might feel less interesting if we find ourselves merely re-treading ground that's already been thoroughly worked out by computers.

The "race to the poles" last century might not have been quite as exciting if androids had already been there and photographed the route and the goal.

>If chess, deemed by some to be a valid way of "measuring" those abilities, is no longer something we excel in, compared to machines, that changes part of how we have to look at ourselves.>

We will not give up (not yet prepared for this) to to our special status of "intelligent"-things as opposed to machines....

What I suggested was that because of the computer chess it would be chess itself which would be relegated to "not so intelligent" and its masters (GMs and good players in general) would lose their status in eye of the public.

There will be a shift (for the worse) in the perception of chess by the general public so as to accomodate this, now undoubtful, reality that machines can beat us anytime.

My calculator adds more error-free than I can, but I still find math puzzles fun, and enjoy working out math problems in my head (as much as I ever did).

My computer spell-check can beat any of those smarty-pants spelling bee winners, but the spelling bee contests still go on.

Computers can carry out long long series of inferences and carry out logic proofs better than me (and I studied philosophy), the essence of human reasoning, and we are bested! So it is all useless, we should just give it all up, don't even get out of bed, machines rule forever.

Right.

off-topic/

what happens with Mig ? his blog hasn't been updated, is he still in Bahamas ?

"what happens with Mig ? his blog hasn't been updated, is he still in Bahamas ?"

I know it's amazing to conceive that the guy might actually take a few days off from chess.

On future of chess:

Please give me an example of an intellectual discipline or activity the interest in which waned because of development of a superior computer ability.

Interest in chess of course is dropping, as it is in nearly all older forms of entertainment and pasttime.

On sponsorship in "computer era":

1. On casual level, chess did receive a boost. People have better opportunity to play each other, meet and discuss online.
2. I am unaware of chess having success securing sponsors before the split. And for that matter, there was no money in chess at that point. It probably became necessary when GMs started to have to spend more time on chess, training became expensive and chess became the method of income for everybody involved.
3. During the split, nothing has prevented computer/techonlogy companies from promoting an occasional spectacle; Intel, IBM, Deep Fritz, MSN, Hydra, etc. have put on computer vs man spectacles. They don't seem too troubled by our internal haggles. There is always money for freak show sponsorships that benefit computer companies.
4. There was only one serious offer in history of our sport of outside sponsorship for WCC and that is Intel. They didn't seem to care about the chess split, absence of established cycle. None of the GMs were particularly thrilled about what Intel wanted and when Kasparov bolted, the deal collapsed.
5. No sponsors rushed at the idea of sponsoring the end of the split or staging subsequent cycles now that it's over.
6. Current head of FIDE has a very poor to non-existent record of securing business sponsors for any chess events.

I see no cause relationship between the split and the lack of sponsorship.

>I know it's amazing to conceive that the guy might actually take a few days off from chess.>

not really only that now one would have expected him to comment on the end of this big match..according to chessbase the evnt got more attention in the world media than Elista ( which seems true, even yahoo had it as main news today)

It would be nice to see a computer match longer than 6 games.

Bronstein's death as reported by FIDE...

http://www.fide.com/news.asp?id=1190

Hey, it's not Mig's fault, blame the Redhead! See what happens when a person gets a life. Damn it Mig, get rid of her and get back to chess. ;)

Does DF10 replace kramnik at WC ?

or the Greyhead, I gather that Mig works for Kasparov so he may be busy with some duty

>Does DF10 replace kramnik at WC ?>

for all that it matters it seems so, unconfirmed rumor is that Topalov wants to play DF-10 in Sofia using an identical hardware in few months, the idea seems to be that from now on who beats DF-10 in match can declare himself champ.

Ovidiu:
I think many people here do not have a hope. It was only a short match. Just 6 games, and all the talk Chess is dead. I still hope that if there is a long match of 20-24 games, then it is possible that humans can win. Examine the games 1-5 and you will know that Kramnik was not outplayed by this silicon beast. In the last game and in a must win position, he lost. There are Super Computers today. But who made them if not humans? The DP10, Junior and other run of the mill programs are all the same. Remove tablebases, opening books and let them play. The theory of evolution is false and even the author himself confessed it on his deathbed.

Ovidiu,

What is your source for the fact that DF10 replaces Kramnik at WC? Or that from now on who beats DF-10 in match can declare himself champ? I understand that the rumor is unconfirmed but I would like to at least to know the reliability of the source spreading it.

Chess sponsorship would be a natural for any computer/internet firm. But during the computer/internet explosion of the past fifteen years, chess has never been ready with an attractive-to-sponsors, stable organization.

Shoe companies fork over millions of sponsorship dollars to associate their name with stable organizations like NCAA or NBA basketball.

There are "flakes" in those organization are a relatively minor part of the game. But any big-money sponsor would think twice about associating himself with a game featuring such influential "flakes" as Kirsan and Danailov.

The flakes

"It was only a short match. Just 6 games, and all the talk Chess is dead. I still hope that if there is a long match of 20-24 games, then it is possible that humans can win."

I can't conceive of a 20-24 game human-comp match, given that Kramnik vs Topalov was only 12 games. Obviously, for a match of that length, the cost goes way up. For this match, Kramnik was paid $500k for showing up, plus another $500k if he won. What would you have to pay him for 20-24 games, and who would pay it?

I also think the human's chance of winning goes not up, but way down, as match length increases. In a long match, fatigue begins to take its toll. When two humans are playing, at least they both suffer equally, but Fritz never gets tired. In a 20-24 game match, the human's blunder rate would start to go up.

Those are two very good reasons (cost & fatigue) why none of the human-comp matches have gone on that long, and none will.

Yuriy, I can't disclose my sources but they are close to the Topa-Danailov team. Don't take it too seriously, it is the kind of absurd joke that people make while half-thinking that it might be in fact an idea. While DF-10 can be bought on line organizing a match is way more trouble.

I would rather think of a Rybka(C)-DF10(C) match on heavy hardware and brodacasted on line.
This could be a great success as Rybka-talk is all the rage nowdays on chess servers and Rybka has lots of die hard fans.
Since Kasparaov or Tal appeared on the chess scene has't been such a passionate talk about a new "player" who might be the strongest etc.
Such a +3000 "clash of the titans" to find out if Rybka is that good could be a great match success if spiced with GM commentary etc.

"unconfirmed rumor is that Topalov wants to play DF-10 in Sofia using an identical hardware in few months..."

Topalov wants a rematch with Kramnik too. That doesn't mean it's happening.

Ovidiu,

What a surprise to Kramnik, to finish playing a computer match and THEN be told that it had been for the world championship.

To even hint that the ever-proper Bulgarians would condone such a thing is a vicious slander.

tjallen:

I very much agree with the spirit of your post, my own favourite example being that a pile of dead computers makes a much better hockey goalie than a person. As a graduate English student, however (though never a spelling bee participant), I can assure you that innumerable masses of my peers can spell MUCH better than your computer spell-check can.

It's a bit of an aside at this point, but upon reading this afternoon's comments relating the machine's dethronement of man in chess, to humankind's sense of "uniqueness" within the natural world (i.e. our ability to think, plan, create, etc.), I was reminded of some points made in "Of a Fire on the Moon," Norman Mailer's account of the 1969 Apollo mission, the first and only manned moon-landing.

Mailer wrote then that the moon itself would never be seen in the same way it had been for millenia. Once conquered (walked upon) by men, the emotional power it had held over us, its philosophical and religious significance as the nearest piece of the formerly unknowable "heavens," was forever shattered.

And from a different perspective on the moon landing, Mailer reported the despair of his contemporary and literary peer, the African-American novelist James Baldwin. Baldwin was depressed and angry, viewing the conquest of the moon as a triumph of the "white man's technology," a symbolic vindication of the extant American power structure, resource allocation, military-industrial complex, etc., that lay at the center of the space program.

Perhaps today's thinkers and philosophers will draw similarly grand implications from the computer's triumph in chess. (Note, however, that this is essentially a non-chess question, with little connection to the debates here about whether the match was fair, whether engines really "understand" the game or are just dumb calculating machines, and the like.)

Along similar lines, I'd note that Vassily Smyslov has written eloquently in a philosophic/religious vein, about human creativity versus computer brute force.

I don't have the link offhand, but I found online an interview with him from only a few years ago, where he equated the computer form of chess play with "the Devil."

>Vassily Smyslov has written eloquently in a philosophic/religious vein, about human creativity versus computer brute force.>

He is old and such mystico/religious rants are understandable.

Computers are still at the beginning and it is their brute force that they can rely upon for now just as the dinosaurs at the beginning of the biological evolution. But they are getting more subtle in thinking as fast pace. DF-10 has shown very good postion play in this match.

We just need get over this difficult moment and accept that the evolution goes on just as it has always gone.
We are creating in fact the future minds which one day will be all round better and replace us.

Thx for mentioning the interview, Jacobs.
It beats what Fischer usually produces.

Smyslov's interview, chess and Holy Ghost

http://www.gmsquare.com/interviews/smyslov.html

"The meaning of life is to let your soul soak in as much of the Holy Ghost as possible. God does not need the unprepared. He takes either those who are ready, or those who are without hope. God grants the others an opportunity to improve their karma"

Supposing that Fritz's freakish Rook and Bishop moves in this game
are correct, then there really is something distinctly different about this algorithm.

It's as though there is a hidden extra value term of "potential energy" in the evaluation of chess positions that no one knew about.

on the "chess death by comps" discussion:

altho a game, chess is not only about winning, it's about understanding how to win. and more generally, about understanding a small, tidy, man-made system. that's what a game is: understanding and (trying to) dominating its rules.
the interest in chess doesn't have to vanish because programs can win more times than humans. the games will be there to be seen and studied. form a patzer point of view (as i see any game today, played by comps or human masters) or from the expert point of view.

>.. chess is not only about winning, it's about understanding how to win..>

I agree with Lasker that chess is, above all, a struggle, a fight. And that humans have lost it.

It's "logical" or "natural" that if anybody would be interested in sponsoring chess events it would be the people whose product is the most chess-related, ie: makers of chess software. Computer/Technology companies whose product is less chess-related such as Dell or Sony, I am not sure they would sponsor chess events and the past backs up my assertion here.

Chess sponsorship largely falls in two categories: 1) chess software makers putting on a spectacle Man vs The Software Maker's Machine to promote release of their latest program
2) random benefactor, who for some reason wants to spend money on chess (Corus, Nahed Ojjeh, occasional third world dictator). I don't believe for a second these people make any money off chess.

It is easy to say that the flakes of chess world are the reason chess can't get sponsorship. This is especially easy because there have always been flakes in our world, whether they were world champions (Fischer) or FIDE presidents (Campomanes). It's easy to blame Illumzhinovs and Danailovs but I wonder if any potential sponsor staying away from chess right now even knows who they are? The truth of the matter is chess world is a flake in and of itself and must come with a bowed head and an opened hand to the corporate world as an intelligent, dignified and persistent beggar. Then and only then will we have more cigar companies and retired philantropists offering us money.

Ovidiu,

The reason we listen to Mig's rumors is because they often come true and we know him to be well-connected in the chess world.
The reason I choose to ignore your rumors is because you can't provide anything to back them up, you haven't delivered any scoops in the past and because this is exactly the kind of rumor somebody wanting to substantiate the whole "chess has been mortally wounded and its torch passed to computers" would start.

[Note: I am not talking about the idea that Topalov would want to play DF10. That's to be expected, though DF10 have made comments to the degree they don't want to play GM now (not until next software release, anyway) I am talking about the idea of Kramnik having been replaced by DF10 as world champion and that "whoever beats DF10 is the champion now".]

I hate to be blunt, because I think you are a good guy. But stuff like this is to be disseminated very carefully.

it can be a battle and a fight, but the fight is also in terms of understanding (and analysis and theories, etc.).

I don't know if anyone here will care, but the "understanding" issue (debated from time to time here, posed most recently by "kuanchainkein") is most important to those of us whose primary aim in following chess is to improve OUR OWN understanding of the game, and thereby raise our own standard of play.

It's odd to see that goal espoused here by just a small minority of readers/posters on what is, after all, a chess blog.

From the standpoint of the majority who define themselves as chess "fans", the paramount question is always who is "better," i.e., who wins or will win.

For those people -- even those who are strenuously arguing here against Ovidiu -- his (Ovidiu's) overall perspective is actually the most logical one (even if his expression of it isn't always the most cogent).

In my case, I said at the outset that I didn't care who would win this match. Ovidiu took that initial comment of mine as agreement with his proposition that Kramnik is a boring player and no one could care about anything Kramnik does; of course, I meant nothing of the sort.

Rather, for those of us who strive to better understand chess for ourselves, the sporting aspect of man-vs-machine matches (and of man vs man matches, for that matter) is of little concern. Clearly, becoming a better PLAYER than the human champion, does not in itself make the engine a better TEACHER. And the biggest market for the software companies is among people (including even some GMs, perhaps?) who wish to use software to deepen their own understanding of the game.

Engine-chess will always differ to some degree from human chess, because we have different strengths and weaknesses.

We humans can never ever duplicate the computer's strength, which lies in (potentially unlimited) brute-force calculation. Engines, on the other hand, can and are gradually assimilating more and more of humans' area of expertise: positional / strategic principles; properly weighting the trade-offs between various values such as pawn structure, king safety, etc.; and quantifying how each sort of material or positional advantage or disadvantage may hold greater or lesser importance under differing circumstances.

So Ovidiu was right to state that as programs keep getting stronger, they are also becoming more human-like (assimilating our strengths, but not our weaknesses), "more subtle in thinking."

Still, it is well to remember where this aspect of their improvement is coming from: Programmers (and the chess experts who advise them) are simply getting better at reducing the more subtle facets of HUMAN POSITIONAL KNOWLEDGE to formulas that can be turned into computer code.

"We are creating in fact human minds which one day will be all round better and replace us."

A very pretty thought. But it might be worth pondering which is the better metaphor for computers as human minds which one day will surpass and "replace us": is it Fritz/Shredder/Hydra/Junior, etc.;
or on the other hand, is it The Terminator / The Matrix / and the intellectual ancestor to those films - Frankenstein.

Nicely put, Jon Jacobs; and the last paragraph made me grin ear to ear.

"Engines, on the other hand, can and are gradually assimilating more and more of humans' area of expertise: positional / strategic principles; properly weighting the trade-offs between various values such as pawn structure, king safety, etc.; and quantifying how each sort of material or positional advantage or disadvantage may hold greater or lesser importance under differing circumstances."

That is true. It's only a matter of time a program with ALL possible strengths is created, since chess is finite. Or perhaps they'll brute-force exhaust the game.

Anyway, I suspect that one of the reasons Kasparov retired is because he saw that programs had become stronger than the top humans... a man with his ego would find that demoralizing.

->We just need get over this difficul