|Nov-26-11|| ||kappertjes: Amazing game, Nf5, Qh6 very nice. From this tourney it seems that top computers have a lot more decisive games than top humans. Perhaps it is that computers can find and take smaller differences and push them to a win more ruthlessly. A matter of tipping scales by finding the smallest imbalance and not letting go?|
|Nov-26-11|| ||computer chess guy: Computers will play the sac 28. ♘hf5+ without necessarily seeing a win: the destruction of Blacks's kingside is enough compensation. After 32. ♖e1 though, Black is really struggling. Houdini likes 32. .. ♘ac7 instead of .. c4, but that doesn't seem to completely solve Black's problems.|
|Nov-27-11|| ||MethodMan: I think 22. Nxe5 maintains a draw.|
|Nov-27-11|| ||Al2009: Frankly speaking, I'm astonished whenever I see that a player (no matter he/she/it could be: computer, super GM, etc.) could really play a move like 12.Bd1 (?) which is one of the strategically worst move I've ever seen in opening phase, it's also obstructing Rooks' connection!|
After 12.Bd1, Bc4 13. Re1, Rfd8 (with idea Bh8 and e5) Black could get a clear positional advantage.
And so, I do prefer players using a strategic "human" logic, surely computers have no rival in calculating tactical variations, but after moves like 12.Bd1 (?), really I still do prefer humans!
|Nov-27-11|| ||OhioChessFan: 12. Bd1 is a very ugly move. I don't know why the engines like 23. Qc6 but there you go. 24...c4 would have been more consistent after the Queen move. 24...h5 was an amateurish move and quite surprising.|
|Nov-27-11|| ||SteinitzLives: 12 . . . . Bd1 is an example of the maxim: If you can't find a good square for a piece, don't move it", and "if you can't find a good square for any of your pieces, move the one that will hurt you the least".|
White is in a brief momentary move choice zugzwang, despite how normal the position looks. The outcome may offend purists, but Bd1 frees the white Q from defense of the c pawn, which supports the maxim that "Every piece should be given the right job".
Furthermore, Bd1 keeps the white Q from being overworked since it must defend the pontiff on h6.
|Nov-27-11|| ||Rob Lob Law: Hey, if 12. Bd1 works, why would you still prefer human play? Besides, even super GM's make these kinds of moves sometimes! Take for example Kramnik Carlsen from the Tal memorial this year.http://www.whychess.org/node/3020|
|Nov-27-11|| ||Penguincw: Black is up a piece but no way to hold a draw.|
|Nov-28-11|| ||Jamboree: I'm confused about the last move. White played the mysterious 41. Rf3 and black resigned. But why play that move in the first place? I thought the whole point was that white was going to play 41. Nd6+ Rxd6 42. h7 and suddenly the queening h pawn wins it for white. The tricky 42. ... Nf8?! fails to 43. Qxd6 Nxh7 44. Qg6+ winning the knight and the game.|
So, if the simple 41. Nd6+ won the game easily, I have two questions: Why did white instead play 41. Rf3; and then, why did black resign? I presume that 41. Rf3 also wins, but the win seems more obscure than the direct tactics of 41. Nd6+. Computers generally don't resign until they see an unavoidable mate or overwhelming material advantage, so what is the crushing continuation after 41. Rf3?
|Nov-28-11|| ||Jamboree: I think I'm beginning to see the answer to my own question. After my suggested 41. Nd6+ Rxd6 42. h7, black has the amazing 42. ... Rd1+ 43. Kh2 Qc1!! because then after 44. h8/Q+ Kd7! White has two queens and a rook on the board and yet no way to finish black off with a direct mate! The two connected black knights support each other and prevent many key checks, and white eventually will run out of checks unless he sacs one of the queens back for the two knights. Which means that unless white sacs back, eventually black will be able to play Ra1+, possibly chasing white's king into danger and/or skewering the white queen on h8.|
Looks like white will probably be able to win the endgame in this line anyway, but it probably had a more distant event horizon than the win resulting from 41. Rf3.
|Nov-28-11|| ||Check It Out: <SteinitzLives: [12.]Bd1 frees the white Q from defense of the c pawn>|
Interesting point. 17.Rc1 does the same, defending the c pawn to free the bishop so it can counter the penetration of black's LSB.
|Nov-28-11|| ||sevenseaman: After analysing this one I do not feel so averse to inhuman games, after all an extension of the human enterprise and endeavor.|
Do computers too have blind spots? I guess not, they will only be overwhelmed when something beyond their horizon comes along. I believe their brilliancies should consist of run-of-mill effective coping with threats & good play to create some of their own.
|Nov-28-11|| ||Penguincw: Congratz to Junior for a strong performance.|
|Nov-28-11|| ||knighterrant999: Why don't I care about computer chess? Is it just me?|
|Nov-28-11|| ||kevin86: The f-file should deliver mate.|
|Nov-28-11|| ||kingscrusher: <Al2009:> Bd1 looks to have strategic implications because it makes way for Ne2-g3 to reinforce the Dark square attack campaign around the opponents K. The K like pawn structure does represent a "slow moving target" and in that respect K-attacks can be seen as kind of strategic in nature. |
Strategically by black playing the b to the a6-f1 diagonal as the game played out, meant it was liable to exchange thus weakening black on the light squares.
It was actually a logical consequence of these two factors that Nf5+ later came to be a winning combinatory tactic.
Tiny dynamic details (and potential K-attack manoevers) can be the basis of a "strategic crush", not just moves which are like a slap in the face from a big wet fish IMHO.
|Nov-28-11|| ||Arcturus: Bit more dynamic than I expected !
When I see Junior I'm asking for his autograph!
|Nov-28-11|| ||Phony Benoni: <knighterrant999: Why don't I care about computer chess? Is it just me?>|
You're hardly alone. I've never liked computer games since Fidelity Prestige beat me at the 1982 US Open. But they are now performing an important function, exemplified by such "ugly" moves as <12.Bd1>.
Chess has been in danger of dying the Draw Death ever since draws were invented. It definitely seemed that way in the 1920s when Capablanca was ruling the roost and complaining that things had gotten too easy.
But first came the post-WWI Hypermoderns, then the dynamic style usually associated with the Soviet School of Chess. They brought new concepts, different ways of evaulating positions and planning. Things got interesting again.
I think computers are performing that same function today. True, many of the games seem like pointless byte shuffling. But they are also revealing more than was dreamed in our chess philsophy, and it's up to people to perceive and apply these lessons in their own games.
There may be some computer "ideas" that don't transfer well to human play, that simply aren't practicable in a context where human frailities play a part. But that's for us to decide.
|Nov-28-11|| ||King Death: The zenith of dull chess was in the Alekhine-Capablanca match that featured 32 games of 34 in different lines of the Orthodox Queen's Gambit Declined. We can all be thankful that Nimzovich started playing his defense a lot, then Bronstein and Geller came along and made hay with the King's Indian in the 1940s.|
|Nov-28-11|| ||sevenseaman: <12. Bd1 here and 20...Kh6 in Junior vs Woodpusher, 2011 are two extremes, the pragmatic and the absurd.
The latter move may be more a result of <Woodpusher> being in too exalted a company many streets ahead of itself.|
In general beauty/ugliness in chess moves is purely a human concept. The computer can only be expected to be objective from the efficiency aspect.
A dilemma? Its up to us; to resolve or decide to live with! Man shouldn't be happy only being derisive of his own creation.
|Nov-28-11|| ||RandomVisitor: After 14.Qe3 you just have to love the possible 14...Bg4!? - only a computer would play it. [15.hxg4? Nxg4]|
1: Junior (Computer) - The Baron (Computer), World Computer Chess Championship 2011
click for larger view
Analysis by Rybka 4.1 x64:
[-0.01] d=21 14...Bg4 15.Bxg7 Bxf3 16.Bxf6 Bxd1 17.Raxd1 exf6 18.Qd2 d5 19.h4 h5 20.g3 Kg7 21.Kg2 Qc7 22.exd5 Nxd5 23.Nxd5 cxd5 24.c3 Qc4 25.Qc2 Rc8 26.Kh3 f5 27.Kg2
|Nov-28-11|| ||Gypsy: A couple of interesting moves: <12.Bd1...> and <18.Be2...>. |
Grew to like the former nearly instantly, the latter took more time.
Would like to know what does Rybka (for instance) say about the alternatives like <12.Rac1...> and <18.Ne2...>.
|Nov-30-11|| ||nimh: Draw rate depends on two factors.
1) the average strength of all participants.
2) the difference in playing strength between the strongest and the weakest participant.
In Tal memorial the rating difference between Carlsen and Nepo was less than 100 points, with the draw rate of 78%. Since Woodpusher managed to lose all games, the recent WCCCT probably had the difference of at least a thousand rating points with the draw rate of 39%.
For a comparison, Rybka 4.1 against himself at 1 second per move on my i7 860 results in 54% draw rate. In the Unofficial World Blitz Correspondence Chess Championship on the Rybka forum the difference was ca 400 points and the draw rate was 68%.
<Perhaps it is that computers can find and take smaller differences and push them to a win more ruthlessly. A matter of tipping scales by finding the smallest imbalance and not letting go>
This explanation isn't correct because it takes into account only one side of the equation. Games are decided by the combined effort of two players.