Marvellous Morozevich
30 December 2006 11:42 AM CET | By Peter Doggers | Filed under: Reports | Tags: Blunder, Endgame
It was Marvellous Morozevich again, this week in Pamplona. He won the not very weak tournament with the splendid score of 6 out of 7 and noted a Rybkayan tpr of 2951. What a shame he won’t be there, in Wijk aan Zee, in two weeks from now.
It’s tempting to say that Morozevich’s score should have been 6,5 out of 7, because he didn’t win a Q-R ending. But in the last round his opponent Korneev shouldn’t have lost.
Morozevich-Jakovenko
Pamplona (5), 2006
A position to remember. The rook on h2 makes it quite difficult for White to win this deceptively easy position. 111. Qe5! wins most quickly: 111… Kg1 (111… Rg2+ 112. Kh3 +-; 111… Kg2 112. Qd4 Kf1 113. Kg3 Rg2+ 114. Kf3 +-) 112. Kg3 Rg2+ 113. Kh3 Kf1 114. Qa1+ Kf2 115. Qb2+ +-. Moro went wrong with 111. Kf3?? when 111… Rf2+! was an instant draw.
While replaying the game, watch Jakovenko’s almost inhuman defending. 79…Rg8! (leads to mate in 17) and 83…Rg4! (mate in 24) are relatively the best moves in the position and the Russian finds them.
In the last round the badly playing ‘open tournament tiger’ Korneev, who for a change participated in a strong closed group, had to finish all that was left in the lethal goblet:
Morozevich-Korneev
Pamplona (7), 2006
31…Qxd5?? (31… Rxe6! 32. dxe6 Qxe6 33. Qf3 Qxb3 34. Rxg7 is hard to judge but certainly not worse for Black) 32. Rxa6+! Kb8 33. Qf4+ Rd6 34. Rxb7+! Kxb7 35. Rxd6 Qxb3 36. Qf7+ 1-0
We will see Alexei Shirov at Corus. He scored a tpr of 2755 and only drew one game. The prettiest move of the tournament was from his hand:
Shirov-Illescas
Pamplona (7), 2006
22. d5! (Much nicer than 22. Qe2 Qxd4 23. Rad1, which is also very good for White) 22… Rxd1 23. Raxd1 e3 24. Rfe1 b5 25. dxe6 bxa4 26. Rd7 Qb6 27. Rxe3! (stronger than 27. Rc7+ Qxc7 28. Bxc7 Kxc7 29. Rxe3 Kd6) 27… Qxb2+?! Better is 27… Qb5 when 28. Bg3! is most accurate (certainly not 28. e7? Nxe7 29. Rdxe7 Bxe7 30. Rxe7 Qd5+ 31. Kg3 Qf3+ 32. Kh4 Qxf4 33. Re8+ Kd7 34. f8Q Qxh2+ 35. Kxg4 Qg2+ 36. Kf4 Qf2+ draw) 28… Qxb2+ 29. Kh1 +-) 28. Kg3 g5 29. Bxg5 Nd4 (29… Qh8 30. Bf4 Qh3+ 31. Kf2 Qh4+ 32. Kg2 +-) 30. Rd8+! Kb7 31. Rxd4 Qxd4 32. e7 Qh8 33. e8Q Qh3+ 34. Kf4 Bd6+ 35. Kf5 1-0
Final standings:
1 MOROZEVICH Alexander 2747 RUS * ½ 1 1 ½ 1 1 1 6 2 JAKOVENKO Dmitry 2671 RUS ½ * ½ ½ 1 1 ½ 1 5 3 SHIROV Alexei 2720 ESP 0 ½ * 1 0 1 1 1 4½ 4 BAUER Christian 2585 FRA 0 ½ 0 * 1 ½ 1 1 4 5 WOJTASZEK Radoslaw 2630 POL ½ 0 1 0 * ½ ½ ½ 3 6 ILLESCAS CORDOBA Miguel 2620 ESP 0 0 0 ½ ½ * 1 ½ 2½ 7 LAZNICKA Viktor 2596 CZE 0 ½ 0 0 ½ 0 * 1 2 8 KORNEEV Oleg 2657 RUS 0 0 0 0 ½ ½ 0 * 1
Funny crazy rook. My Rybka-engine doesn’t see the draw even after an hour thinking. It may of course be my weak engine (Pentium III 1GHz) but surely Rybka should have recognized the mechanism instantly anyway? Is there anyone whose engine does see the draw?
Indeed funny crazy rook. My Rybka engine has no problem to find the rook move. Used rybka 2.2 en 2.1c on a AMD thunderbird 1.3 GHz 128 Mb
and it plays it instantly
Instantly? Toch handig..tablebases.
Waarom doet Moro niet mee in Wijk aan Zee? Ik denk omdat hij (relatief) faalt in de allersterkste toernooien, denk aan de Tal Memorial onlangs waar hij laatste werd. Daarentegen liggen de iets minder sterke toernooien zoals Biel en Pamplona hem uistekend en haalt hij vaak enorm uit.
@Heer. Bezemer,
Helaas, ook zonder tablebases. Anders had ik niet gepost.
Sorry, even without tablesbases, otherwise I would not have posted.
It does find it, but only as it’s the only way not to get mated, evaluation is still +5. It can only know it’s draw with tablebases.. [Running on a Athlon X2 3800]
@grploeg. Obviously I didn’t mean that Rybka finds the *move*, but the right *idea* and especially the correct *evaluation* (0.00). Otherwise I would not have posted
The draw is fifty moves in the future, too far away -still- for the programs to assess the situation properly (since programs can “only” calculate, not “understand” positions).
Well OK but in principe the mechanism should be rather easy to put in a rule. I imagine something like: ‘Look for the position of the king - is it in a stalemate position? Then find pieces that can, if taken, achieve this stalemate.’ Tim Krabbe has formulated a rule for ‘crazy rooks’ that could easily be put into a formal rule I think. But I guess that building such a rule into the software costs more time than possible (half-)points… By the way, I am told Fritz9 finds the mechanism and correct evaluation almost instantly!
My Fritz 9 doesn’t. It gives the game away via: 1. Kf3, Tf2+ 2. Ke3, Te2+ 3. Kd3, Td2+ 4. Kc3, Tc2+ 5. Kb3, Tg2 6. Df3 Kh2.
I believe this position is winnable for White.
[...] exactly one year ago, right under the daily puzzle. Funny enough, today one year ago the header was “Marvellous Morozevich”, because the Russian had just won the Pamplona tournament with 6 out of 7. One year later he wins [...]