The Daily Update: Dortmund, Biel, and Rybka-Stockfish
Wednesday, July 21, 2010 at 9:27PM The three games in Dortmund today were a study in contrast. Two of the three games were uninspired draws, while the last game was a weird and wacky blunderfest. If a weird virus were to erase Leko - Ponomariov and Le Quang Liem - Kramnik from the world's computers, no one would ever notice, so I won't say anything more about them. Mamedyarov - Naiditsch is a game White probably wishes a virus would wipe out, but alas, you'll have your chance to see it first. After some weird transpositions and a Naiditsch bailout, White obtained a very pleasant edge pretty much for free. Mamedyarov was almost always better and was even winning once or twice, but he repeatedly let Naiditsch off the hook. (Time trouble was almost surely a big factor in this.) Still, he was always at least equal until his very last move, which was an outright blunder giving up the exchange for nothing. Disgusted, he resigned without even waiting to see whether Naiditsch might have suffered the same hallucination he did.
That game, with my notes, can be replayed here. As a service to insomniacs, I also give the other two games, but without annotations.
Standings After Round 6:
1. Ponomariov 4
2. Le Quang Liem (!) 3.5
3-4. Kramnik, Mamedyarov 3
5. Naiditsch 2.5
6. Leko 2
In Biel, So beat Rodshtein and Caruana beat Negi (quickly), in both cases with Black. So is thus in clear first with 2.5/3, half a point ahead of Caruana and Tomashevsky.
Finally, the Rybka 4 - Stockfish 1.8 match is almost over. Assuming Stockfish converts its currently winning position, the score will be 28-19 in favor of Rybka 4 with one game to go.
Reader Comments (8)
Strange but true: There was no real time trouble for Mamedyarov - final clock times (still given at the Chessdom live site http://livechess.chessdom.com) were 15 minutes for Mamedyarov vs. 1:13 for Naiditsch. What apparently had happened, already for a few moves before: Mamedyarov was blitzing along in Naiditsch's time trouble.
Is Mamedyarov becoming the new Ivanchuk? According to the tournament homepage, after yesterday's game he immediately ran away in anger - no info on the fate of concrete pillars in the vicinity. And his sudden resignation in the previous round againt Ponomariov was also "vintage Chucky", at least even his opponent was surprised ("I didn't understand why he resigned in the end. Sure, I am better, but one can still fight" - also translated from German at the tournament homepage).
BTW, I think Le Quang Liem - Kramnik wasn't quite as boring as you make it, but that's a matter of taste. It may have been a narrow escape for Kramnik and got most of the attention at other sites, making Mamedyarov-Naiditsch just sort of a footnote. I fully agree that Leko-Ponomariov was genuinely boring ... .
> BTW, I think Le Quang Liem - Kramnik wasn't quite as boring as you make it
Agreed. Kramnik had to deal with an interesting novelty in his Catalan. White changing the pawn structure with 17.Rac1 to create attacking chances. Kramnik suffered some time disadvantage and probably discomfort before bailing out with his queen sac. Shipov gives some interesing lines, for instance 24.Qe4 Bxd6 25.Rc6 Qa7 26.exd6!
full translation of Shipov's comments here: http://www.chessintranslation.com/2010/07/shipovs-live-commentary-on-le-quang-liem-kramnik/
Nice to see that Martin has listened to your feedback on starting positions for the computer matches. Stockfish-Houdini looks a more interesting prospect as a result. It'll be interesting if Stockfish manages to beat Houdini, after the latter did significantly better against Rybka. Presuming there's not something odd going on with the style of each, or some odd bias in the change in starting positions, it'd lend some credence to Kenneth's point that 48 (or 32) games isn't a large enough sample to give a clear indication of strength. [[also assuming that Houdini 1.03a is a genuine improvement over 1.02 - presuming the 'a' is for 'alpha', that might not be an entirely safe assumption]]
Hi Dennis,
I wanted to comment on your analysis after 32.Bxg5 etc. In your line after 40.g4 I think it is better for black to not give the check Qd3+ and should instead take with the knight one move earlier. The idea is that black's chances are slim if he has to exchange queens on e3, but and it's easier to avoid it if the queen is on d4 (and can go to a1 after Qg5). Sorry I don't have the actual move numbers.
I agree that my comments about the Kramnik game were a bit hyperbolic and that he did have to sweat a little. But I also think the only reason it got much attention was because Kramnik was involved, not because of it was intrinsically fascinating. (But perhaps I'm wrong.)
Serdal: Assuming you mean 44...Nxe5 (instead of 44...Qd3+ - why not just open a second browser window or tab so you can see the move numbers?), then it looks like White is doing well with 45.Qg5 Qa1 46.Rc1 Qa3+ 47.Qe3 Qd6 48.Qf4 - not winning here, either, but I think his chances are better than in the 44...Qd3+ line.
Hi Dennis,
you correctly 'guessed' what I meant - sorry about that. I just thought that in the 44...Nxe5 line that there could follow 45.Qg5 Qa1 46. Rc1 Qa3+ 47.Qe3 (as you give), but my idea was that Nc4 could be ok for black, because a queen trade doesn't seem so harmful as long white doesn't get an e pawn to undermine the c4 outpost for black's knight. Maybe white does have decent winning chances, I don't know, but it looks to me like black's best bet.
Thanks
Serdal,
White wins, with a little but not too much work. After the natural moves 47.Qe3 Nc4 48.Qxa3 Nxa3 49.Rc7 Kg7 50.Ra7 Nc4 51.f4 Kf6 White needs an idea, and one that works is to bring his king around on the kingside. After 52.Kh4 Kg6 53.Ra8 Black is in a sort of zugzwang.
(1) 53...Kg7 54.Kg5 Nb6 55.Ra7! Nxa4 56.f5 exf5 57.gxf5 wins.
(2) 53...Kh6 54.f5 d4 55.Rd8 Ne5 56.Rd6 Kg7 57.fxe6 fxe6 58.Rxe6 Nc4 59.Kg5 Kf7 60.Re4 Nd2 61.Kf5! wins.
(3) 53...Kf6 54.Kh5 d4 55.Ra7! d3 56.Rd7 and now another parting of the ways:
(3a) 56...Nb2 57.Kh6 e5 58.Rd6+ Ke7 59.fxe5 Nxa4 60.Rxd3 will win - White will bring his king to the queenside if necessary to separate the knight from the a-pawn.
(3b) 56...d2 57.g5+ Kg7 (57...Kf5 58.Rxf7+ Ke4 59.Rd7 Kxf4 60.g6 wins) 58.g6 Nb6 59.Rxd2 fxg6+ 60.Kg5 Kf7 61.Rd4 Nd5 62.Rc4 e5 (62...Kg7 63.Rc5 is easy) 63.fxe5 Ke6 64.Rc5 Kxe5 65.Rxa5 Kd6 66.Rb5 Kc6 67.Rb1 and wins.
Thanks Dennis,
I was far too optimistic for black's side, you're right of course. I hadn't thought white could break through so easily on the kingside. The solidity of black's position is just 'optical'. Thanks for your analysis, I don't want to bother you any more.
Keep up the good work, Serdal