Corus 2011

General discussion about computer chess...
BB+
Posts: 1484
Joined: Thu Jun 10, 2010 4:26 am

Re: Corus 2011

Post by BB+ » Thu Jan 20, 2011 3:06 pm

Nepomniachtchi-Shirov was all book-prep. Black gets a fortress when down Q+2p vs RN+p, with computers at least favouring White by +2 (and more). Very nice.

Grischuk has played a peculiar exchange sac against Wang Hao:
[White "Wang Hao"][Black "Grischuk"]
1. d4 Nf6 2. c4 g6 3. g3 Bg7 4. Bg2 O-O 5. Nc3 d6 6. Nf3 Nc6 7. O-O a6 8. b3 Rb8 9. Nd5 Bg4 10. Bb2 Nxd5 11. cxd5 Nb4 12. e4
f5 13. h3 fxe4 14. hxg4 exf3 15. Bxf3 Qe8 16. g5 Rxf3 17. Qxf3 Qf7 *This might be good for the draw, though I'm not yet convinced. EDIT: I should add that Grischuk is (as usual) almost an hour behind on the clock.

I am not expert enough to say much against Carlsen's play as White in the Ruy against L'Ami. It's one of the slow-moving games that computers don't always judge well. Anand seems to have blunted Smeets in the Breyer. Giri gave Vachier-Lagrave the Bishop pair in their Petrov, though it will not be easy to convert this. V-L also has an extra hour on the clock. Aronian accepted weakened pawns to exchange Queens -- Kramnik will likely torture this to no end, but my guess is Black should hold somewhat easily. Ponomariov seems to have got the best out of the opening, and has the Bishop pair, so Nakamura might be wary to open the position. But with a mobile centre, I think it's somewhat inevitable. This could be a good test for both players.

BB+
Posts: 1484
Joined: Thu Jun 10, 2010 4:26 am

Re: Corus 2011

Post by BB+ » Thu Jan 20, 2011 4:00 pm

Aronian has sacrificed not one but now two pawns, getting the Bishop pair and some temporary pressure until White gets fully situated with development. The computers preferred freeing the wRh1 via 19. Ke2, but VK played the slower move 19. Ne5 (Black gains a tempo in many lines via Rc6 to attack this Knight, and then invading on the c-file).
[White "Kramnik"]
[Black "Aronian"]
[Result "*"]
1. d4 d5 2. c4 c6 3. Nc3 Nf6 4. cxd5 cxd5 5. Bf4
5... Nc6 6. e3 Bf5 7. Rc1
7... e6 8. Qb3
8... Qb6
9. Qxb6 axb6
10. Bb5 Be7
11. f3 O-O
12. a3
12... Rfc8 13. Nge2 Na5
14. Na4 Rxc1+ 15. Nxc1 Nc4
16. Bxc4 dxc4 17. Nxb6 Ra6 18. Nxc4 Nd5
19. Ne5 *
Carlsen is down a pawn against L'Ami, but should not lose, due to piece activity and Black's weak pawns. I don't think a 2600+ player should lose this as Black without a genuine mishap, so MC will probably still be winless. The principal advantage in Smeets-Anand is that White has to make 20 moves in ~30 minutes in a position which could turn complicated. I think Anand should make it as difficult as possible by playing quickly himself. :) Nakamura and Ponomariov are still progressing slowly (well, at least Nakamura is), and at move 18 I think Black has good chances with the Bishop pair. Vachier-Lagrave versus Giri looks headed toward a draw, though White has some attacking ideas in the coffeehouse class, and Black is behind on the clock. Wang Hao still is up the exchange, and is trying to apply pressure to Black's King (likely eventually via the h-file, though not right away).

BB+
Posts: 1484
Joined: Thu Jun 10, 2010 4:26 am

Re: Corus 2011

Post by BB+ » Thu Jan 20, 2011 4:42 pm

I expect Carlsen-L'Ami to be draw agreed, perhaps after the time control [White has some threats to the Black King and the some assorted tricks, but Black should not be fooled]. Kramnik-Aronian also looks to be a matter of time, though it could be more lengthy (oops, just agreed at move 26 -- cutely, Black had just given up the exchange for 2 pawns, so is it mutual fear? :o ). Nakamura's position continues to deteoriate. I didn't mention it before, but his b5 push made the c4 pawn weak. He has survived worse, though. It seems as though Vachier-Lagrave pushed one too many pawns, and now Black (Giri) might have the chances on the queenside (the kingside now being locked [oops, now unlocked, and White looks better again, though I'd expect any heavy piece ending to be drawn]). Wang Hao is still the exchange up, and Grischuk needs to make like 13 moves in 2-3 minutes (with 30s/move increment).

Forgot Smeets-Anand. Not much happening there after the Queens were exchanged. White has a Bishop pair, Black has a strong Knight at d3 and 30 extra minutes. If Smeets reaches move 40, he should have a draw (and he might even have a minor edge, but I can't imagine he would be unwilling to take the draw).

BB+
Posts: 1484
Joined: Thu Jun 10, 2010 4:26 am

Re: Corus 2011

Post by BB+ » Thu Jan 20, 2011 4:57 pm

[White "Wang Hao"][Black "Grischuk"][Result "*"]
1. d4 Nf6 2. c4 g6 3. g3 Bg7 4. Bg2 O-O 5. Nc3 d6 6. Nf3 Nc6 7. O-O a6 8. b3 Rb8 9. Nd5 Bg4 10. Bb2 Nxd5 11. cxd5 Nb4 12. e4 f5 13. h3 fxe4 14. hxg4 exf3 15. Bxf3 Qe8 16. g5 Rxf3 17. Qxf3 Qf7 18. Qe2 Rf8 19. f4 Nxd5 20. Rf2 h6 21. Raf1 c6 22. g4 e6 23. Qe4 hxg5 24. fxg5 Qe8 25. Rxf8+ Bxf8 26. Re1 Kf7 27. Kg2 b5 28. a4 b4 29. Bc1 a5 30. Bf4 Qd7 31. Bg3 Be7 32. Rf1+ Kg7 *Find the shot? White did, though of course he can win in whatever manner.

In other news, L'Ami blundered at the time control (should just check with 39... Ra2+), and now Carlsen can perhaps imagine winning. Still unlikely IMO, though MC will certainly give a go. Vachier-Lagrave now has a minor edge (safer King) in a Queen ending against Giri, though without a blunder it should be drawn. Smeets made the time control in a double rook and opposite bishop ending. Unless he wants to push his luck (and the kingside pawns), it will be draw agreed.

EDIT: I might point out that the reason why L'Ami should play Ra2+ is that if White doesn't guard the g2-pawn, then Black can draw via hoovering up the White pawns (giving up a piece for the e-pawn if necessary).
Last edited by BB+ on Thu Jan 20, 2011 5:09 pm, edited 1 time in total.

Jeremy Bernstein
Site Admin
Posts: 1226
Joined: Wed Jun 09, 2010 7:49 am
Real Name: Jeremy Bernstein
Location: Berlin, Germany
Contact:

Re: Corus 2011

Post by Jeremy Bernstein » Thu Jan 20, 2011 5:08 pm

BB+ wrote:[White "Wang Hao"][Black "Grischuk"][Result "*"]
1. d4 Nf6 2. c4 g6 3. g3 Bg7 4. Bg2 O-O 5. Nc3 d6 6. Nf3 Nc6 7. O-O a6 8. b3 Rb8 9. Nd5 Bg4 10. Bb2 Nxd5 11. cxd5 Nb4 12. e4 f5 13. h3 fxe4 14. hxg4 exf3 15. Bxf3 Qe8 16. g5 Rxf3 17. Qxf3 Qf7 18. Qe2 Rf8 19. f4 Nxd5 20. Rf2 h6 21. Raf1 c6 22. g4 e6 23. Qe4 hxg5 24. fxg5 Qe8 25. Rxf8+ Bxf8 26. Re1 Kf7 27. Kg2 b5 28. a4 b4 29. Bc1 a5 30. Bf4 Qd7 31. Bg3 Be7 32. Rf1+ Kg7 *Find the shot? White did, though of course he can win in whatever manner.
Beautiful!

jb

BB+
Posts: 1484
Joined: Thu Jun 10, 2010 4:26 am

Re: Corus 2011

Post by BB+ » Thu Jan 20, 2011 5:40 pm

Ponomariov played it coolly, winning a pawn w/o giving up any structural gains, though he eventually found himself also in time pressure. At the time control, Nakamura still faces an uphill battle to draw, though there is much play left (OTOH, I can't imagine Ponomariov losing unless he really chokes -- White has no real plusses). Smeets thought for 30 minutes on move 41, and looks to play on. However, even if he wins a pawn somehow, I can't see victory coming. Carlsen has given it a big think, and I have to say that in practise against a lower-rated player, this has pretty good winning chances (especially with both sides having passed pawns, so that there is volatility). My impression, though, is that "best play" would lead to a draw (for instance, via a rook endgame where White is a pawn up with connected passers, but Black's has one of his own and it is far enough advanced to ensure the draw). It looks "finely balanced" from the human perspective.

BB+
Posts: 1484
Joined: Thu Jun 10, 2010 4:26 am

Re: Corus 2011

Post by BB+ » Thu Jan 20, 2011 6:04 pm

Anand gave up an exchange (the engines are happy with it, though I really don't see why it was necessary) to try to simplify to a draw, while Ponomariov just needs to avoid any mis-steps to get his win. EDIT: Ponomariov just agreed to a draw!! :!: Carlsen took a 2nd long think, and chose (at least at first) to play at Black's King. I think 42. h4 could have been superior, as Black has to respond with 42...g5, though maybe this line just leads to exchanges too rapidly. MC could always return to this if nothing pans out with 42. Nf8+.

Anand managed to lose RR vs RB (with 2 pawns, none passed, for each side) to Radjabov 3 years ago here, in a game where he could have claimed a 3-rep but did not. Here there are more pawns, and thus the Rooks are likely not so strong.

BB+
Posts: 1484
Joined: Thu Jun 10, 2010 4:26 am

Re: Corus 2011

Post by BB+ » Thu Jan 20, 2011 6:48 pm

Delicately balanced indeed in Carlsen-L'Ami. The live peanut gallery is almost laughable, being so knowledgeable with their engines (sometimes themselves erroneous). After a tricky sequence with Rd1, eventually L'Ami lunged forward the a-pawn (almost from zugzwang), and MC pushed the e-pawn a bit too soon it seems (he did well to hold it back previously, to restrain the Black Knight). I think engines can hold after 46. Nc5 in any event, but the draw becomes notably less hairy for Black after 46. e6. However, there are lines like 47. Ke2 Re1+ 48. Kd3 (after 46. e6 Nd6) -- is it obvious for Black to play g5 [and I'm only going on feel here -- others moves could draw, and this might even lose]? Perhaps without this resource, it again becomes dicey... I think both players are doing about as well as can be expected (in earlier ages, the game would have been adjourned, and even then play was often of a variable level). In any event, MC played 47. h4?!, and now Black has some winning chances too (or chances for White to blunder, I should say), thanks to the push of the a-pawn that was pooh-poohed so readily.

Anand seems to have looked to exploit Smeets in time trouble again (perhaps why he gave up to exchange -- he gets a dangerous passer, and White has to play precisely, at least for a few moves), but after the dust clears there should be no win for either side (the White Rook is just too strong in R vs B as a general principle, and Black will never get the passer(s) sufficiently fearful).

BB+
Posts: 1484
Joined: Thu Jun 10, 2010 4:26 am

Re: Corus 2011

Post by BB+ » Thu Jan 20, 2011 7:02 pm

Smeets just played the berzerk 57. g3? --- maybe he forgot that Black can play d2 after Rg2? Time trouble, but he still gets 30s per move... Anand has all the play for the win now, though maybe it's not yet a surety (with White's probable mind-set it could be, but two pawns up in Rook ending is never that sure).

Carlsen has played 50. e7, which I guess forces L'Ami to play some only-moves (50... Ne4+ 51. Kf3 Nxg5+ 52. Bxg5+ Rf1+, for instance -- note that White is mated after 53. Kg3? in this line, so must go to the e-file!) to get the draw [which I'm sure all the spectators saw all along :lol: ]. EDIT: Black played 50...a2 (loses the pawn, but deflects the white Rook off the 7th rank?!), and now faces a very uphill grind. Unlike much of this, I really don't see why Ne4+ wasn't played, as Black has to do it sooner or later in any event. MC only has 2 mins plus the increment to reach move 60, so again there could be more twists. Black gave up the g-pawn to get space for his King, and after the exchange of passers (a-pawn for e-pawn), this will be White's main trump.

BB+
Posts: 1484
Joined: Thu Jun 10, 2010 4:26 am

Re: Corus 2011

Post by BB+ » Thu Jan 20, 2011 7:21 pm

Smeets made a further blunder, and quickly resigned.
Carlsen has this against L'Ami, and it seems very difficult to tell whether White has a theoretical win or not (my guess is "no", Black is able to deal with the passer):
5N2/4b3/8/R5Pp/6k1/3KB1n1/6P1/5r2 w - - 2 57At least White now has no losing chances. :) [Furthermore, Black has many ways to go wrong, and it's all now "increment chess" after a long day's work]. [For instance, when Black can simply give up a minor piece for the g-pawn in some lines if the other minors come off, and draw R+N+g3 vs R+h4?].

In any event, Black made two quite poor moves at the end of time control (showing how little the margin for error really is), first failing to re-position the Knight, and then guarding the wrong rank (against the g-pawn) with the Rook. All in all, this is about what I expect when a 2800 player has such an endgame against one rated 150+ points lower (blunders on both sides, but the better player wins).

Post Reply