Author: Andrew Wagner
Date: 07:14:51 05/15/04
Hi all. The LeChess group on ICC has a fascinating correspondence game going on
with IM Radoslaw Jedynak. He's agreed to let use analyze with computers, so
don't think I'm trying to cheat or anything. First off, here's the whole game so
far:
[Event "?"]
[Site "?"]
[Date "2004.03.21"]
[Round "?"]
[White "LeChess"]
[Black "Moonspell"]
[Result "*"]
[ICCResult "Game adjourned by mutual agreement"]
[Opening "French: Winawer, advance, 5.a3"]
[ECO "C17"]
[NIC "FR.10"]
[Time "08:05:09"]
1. e4 e6 2. d4 d5 3. Nc3 Bb4 4. e5 c5 5. a3 Ba5 6. Qg4 Ne7 7. dxc5 Bxc3+ 8.
bxc3 Qa5 9. Bd2 Ng6 10. Nf3 Nd7 11. Bd3 Nxc5 12. O-O Ne4 13. Bxe4 h5 14.
Qxg6 fxg6 15. Bxg6+ Kd8 16. c4 Qa4 17. cxd5 exd5 18. Bf7 Bg4 19. Bxd5 Kc7
20. Rab1 b6 {Game adjourned by mutual agreement} *
Position after 13...h5:
[d]r1b1k2r/pp3pp1/4p1n1/q2pP2p/4B1Q1/P1P2N2/2PB1PPP/R4RK1 w kq h6
Now here, Qxg6 appears to be the only real reasonable move. I should also note
that there was a game, Ponomariov-Borovikov, where white tried 13.Be3 h5 Qh3 and
got a lousy position, though he wound up winning in the end. I'll put that game
at the end of this post. So, does your program find Qxg6? Or suggest something
else? Crafty likes 14.c4 after 15 minutes on my little Celeron 2.4. I can't say
with certainty that it's wrong, but all the strong players I've talked to have
said that Qxg6 is best. Here's Crafty's analysis:
White(2): setboard r1b1k2r/pp3pp1/4p1n1/q2pP2p/4B1Q1/P1P2N2/2PB1PPP/R4RK1 w kq h
6 0 14
White(1): go
clearing hash tables
crafty is behind 2.0 on time, reducing by 1/6.
time surplus 0.00 time limit 15:00 (15:00)
depth time score variation (1)
7-> 0.15 0.00 1. Qh3 dxe4 2. Ng5 Qxe5 3. Rfe1 O-O
4. Qxh5
8 0.22 -0.22 1. Qh3 dxe4 2. Ng5 Bd7 3. Rfe1 Nxe5
4. Qe3 O-O-O 5. Qxe4
8-> 0.30 -0.22 1. Qh3 dxe4 2. Ng5 Bd7 3. Rfe1 Nxe5
4. Qe3 O-O-O 5. Qxe4
9 0.47 -0.09 1. Qh3 dxe4 2. Ng5 Bd7 3. Qe3 O-O 4.
Qxe4 Qxe5 5. Qxb7
9-> 0.66 -0.09 1. Qh3 dxe4 2. Ng5 Bd7 3. Qe3 O-O 4.
Qxe4 Qxe5 5. Qxb7
10 1.49 -0.46 1. Qh3 dxe4 2. c4 Qc7 3. Ng5 Bd7 4.
Bc3 Bc6 5. Rad1 Nxe5
10 3.01 -0.26 1. c4 Qa6 2. Qxg6 fxg6 3. Bxg6+ Ke7
4. Bb4+ Kd7 5. cxd5 exd5 6. Bf5+ Kd8
7. Bd3
10-> 3.17 -0.26 1. c4 Qa6 2. Qxg6 fxg6 3. Bxg6+ Ke7
4. Bb4+ Kd7 5. cxd5 exd5 6. Bf5+ Kd8
7. Bd3
11 4.93 -0.08 1. c4 hxg4 2. Bxa5 dxe4 3. Nd2 Nxe5
4. Bc3 Rh5 5. Nxe4 Kf8 6. Nd6
11-> 5.68 -0.08 1. c4 hxg4 2. Bxa5 dxe4 3. Nd2 Nxe5
4. Bc3 Rh5 5. Nxe4 Kf8 6. Nd6
12 7.52 -0.06 1. c4 hxg4 2. Bxa5 dxe4 3. Nd2 b6 4.
Bc3 Bb7 5. Rfe1 O-O 6. Nxe4 Rad8
12-> 9.68 -0.06 1. c4 hxg4 2. Bxa5 dxe4 3. Nd2 b6 4.
Bc3 Bb7 5. Rfe1 O-O 6. Nxe4 Rad8
13 14.56 0.05 1. c4 hxg4 2. Bxa5 dxe4 3. Nd2 b6 4.
Bc3 Bb7 5. Rfe1 O-O-O 6. Nxe4 Bxe4
7. Rxe4
13-> 18.81 0.05 1. c4 hxg4 2. Bxa5 dxe4 3. Nd2 b6 4.
Bc3 Bb7 5. Rfe1 O-O-O 6. Nxe4 Bxe4
7. Rxe4
14 28.58 -0.05 1. c4 hxg4 2. Bxa5 dxe4 3. Nd2 Nxe5
4. Bc3 e3 5. fxe3 Rh5 6. Rf4 Kf8 7.
Raf1 f6
14-> 50.02 -0.05 1. c4 hxg4 2. Bxa5 dxe4 3. Nd2 Nxe5
4. Bc3 e3 5. fxe3 Rh5 6. Rf4 Kf8 7.
Raf1 f6
15 1:19 0.10 1. c4 hxg4 2. Bxa5 dxe4 3. Nd2 Nxe5
4. Bc3 e3 5. fxe3 Rh5 6. Rf4 Kf8 7.
Raf1 f6 8. Bd4
15-> 1:54 0.10 1. c4 hxg4 2. Bxa5 dxe4 3. Nd2 Nxe5
4. Bc3 e3 5. fxe3 Rh5 6. Rf4 Kf8 7.
Raf1 f6 8. Bd4
16 2:55 0.01 1. c4 hxg4 2. Bxa5 dxe4 3. Nd2 O-O
4. Bc3 Bd7 5. Nxe4 Bc6 6. Rae1 Nh4
7. Re2 Rad8 8. c5 Nf5
16-> 4:59 0.01 1. c4 hxg4 2. Bxa5 dxe4 3. Nd2 O-O
4. Bc3 Bd7 5. Nxe4 Bc6 6. Rae1 Nh4
7. Re2 Rad8 8. c5 Nf5
17 8:11 0.06 1. c4 hxg4 2. Bxa5 dxe4 3. Nd2 O-O
4. Bc7 Bd7 5. Rfe1 Bc6 6. a4 Rac8 7.
Bd6 Rfd8 8. Nxe4 Bxe4 9. Rxe4
time=15:00 cpu=97% mat=3 n=711647084 fh=91% nps=790k
ext-> chk=22147163 cap=1360802 pp=796529 1rep=1381084 mate=93317
predicted=0 nodes=711647084 evals=222241549
endgame tablebase-> probes done=0 successful=0
SMP-> split=0 stop=0 data=0/32 cpu=14:39 elap=15:00
White(1): c4
time used: 15:00
Now fast-forward to after 20.b6. Our intended move is Rb4 here, giving the
following position: [d]r6r/p1k3p1/1p6/3BP2p/qR4b1/P4N2/2PB1PPP/5RK1 b - -
After another 15-minute search, Crafty gives this position +1.5 for white, and
finds what seems to be black's best response (21...Qd7), but it insists on white
taking the exchange right away:
White(1): setboard r6r/p1k3p1/1p6/3BP2p/qR4b1/P4N2/2PB1PPP/5RK1 b - - 1 21
Black(1): st 900
search time set to 900.00.
Black(1): go
clearing hash tables
time surplus 0.00 time limit 15:00 (15:00)
depth time score variation (1)
7 0.20 1.37 1. ... Qxa3 2. Nd4 a6 3. Rc4+ Kb8 4.
Bxa8 Kxa8 5. e6
7-> 0.26 1.37 1. ... Qxa3 2. Nd4 a6 3. Rc4+ Kb8 4.
Bxa8 Kxa8 5. e6
8 0.33 -- 1. ... Qxa3
8 0.72 2.78 1. ... Qxa3 2. Rc4+ Kd7 3. Bc6+ Ke6
4. Ng5+ Kxe5 5. f4+ Kd6 6. Bxa8 Rxa8
7. Bb4+ Kd5 8. Bxa3 Kxc4
8 0.91 1.80 1. ... Qd7 2. Bxa8 Rxa8 3. Nd4 Rd8
4. f3 Be6 5. Rd1
8-> 1.18 1.80 1. ... Qd7 2. Bxa8 Rxa8 3. Nd4 Rd8
4. f3 Be6 5. Rd1
9 1.49 1.62 1. ... Qd7 2. Bxa8 Rxa8 3. Nd4 Rf8
4. h3 Be6 5. Nb5+ Kb8 6. Nd6
9-> 1.97 1.62 1. ... Qd7 2. Bxa8 Rxa8 3. Nd4 Rf8
4. h3 Be6 5. Nb5+ Kb8 6. Nd6
10 2.81 1.48 1. ... Qd7 2. Bxa8 Rxa8 3. Bf4 Qe7
4. Ng5 a5 5. Ra4 Rf8 6. Re1
10-> 3.46 1.48 1. ... Qd7 2. Bxa8 Rxa8 3. Bf4 Qe7
4. Ng5 a5 5. Ra4 Rf8 6. Re1
11 5.30 1.38 1. ... Qd7 2. Bxa8 Rxa8 3. Bf4 Qe7
4. Nd4 g5 5. Nb5+ Kc6 6. Bg3 h4 7.
Nd4+ Kb7
11-> 6.97 1.38 1. ... Qd7 2. Bxa8 Rxa8 3. Bf4 Qe7
4. Nd4 g5 5. Nb5+ Kc6 6. Bg3 h4 7.
Nd4+ Kb7
12 11.51 1.38 1. ... Qd7 2. Bxa8 Rxa8 3. Bf4 Qe7
4. Ng5 Rf8 5. f3 Be6 6. Re1 Rf5 7.
Nh3
12-> 16.42 1.38 1. ... Qd7 2. Bxa8 Rxa8 3. Bf4 Qe7
4. Ng5 Rf8 5. f3 Be6 6. Re1 Rf5 7.
Nh3
13 28.98 1.19 1. ... Qd7 2. Bxa8 Rxa8 3. Bf4 Qe7
4. Nd4 g5 5. Be3 Qxe5 6. Rb5 Qf6 7.
h3 a6 8. Rxg5
13-> 41.21 1.19 1. ... Qd7 2. Bxa8 Rxa8 3. Bf4 Qe7
4. Nd4 g5 5. Be3 Qxe5 6. Rb5 Qf6 7.
h3 a6 8. Rxg5
14 1:13 1.36 1. ... Qd7 2. Bxa8 Rxa8 3. Bf4 Qe7
4. Ng5 Rf8 5. g3 Rd8 6. e6+ Kb7 7.
Re1 Rd1 8. Rxd1 Bxd1
14-> 1:46 1.36 1. ... Qd7 2. Bxa8 Rxa8 3. Bf4 Qe7
4. Ng5 Rf8 5. g3 Rd8 6. e6+ Kb7 7.
Re1 Rd1 8. Rxd1 Bxd1
15 3:06 1.28 1. ... Qd7 2. Bxa8 Rxa8 3. Bf4 Kb7
4. Nd2 Rc8 5. Nc4 Ka8 6. Ne3 Rf8 7.
f3 Be6 8. Rfb1 Qf7
15-> 4:43 1.28 1. ... Qd7 2. Bxa8 Rxa8 3. Bf4 Kb7
4. Nd2 Rc8 5. Nc4 Ka8 6. Ne3 Rf8 7.
f3 Be6 8. Rfb1 Qf7
16 14:47 1.53 1. ... Qd7 2. Bxa8 Rxa8 3. Nd4 a5 4.
Rc4+ Kb7 5. Be3 Qd5 6. Rc6 Rc8 7. Rg6
Qxe5 8. h3 Bf5 9. Nxf5 Qxf5 10. Rxb6+
time=15:00 cpu=99% mat=-1 n=796512617 fh=90% nps=884k
ext-> chk=26705110 cap=1046560 pp=1802519 1rep=1460535 mate=108069
predicted=0 nodes=796512617 evals=301574707
endgame tablebase-> probes done=0 successful=0
SMP-> split=0 stop=0 data=0/32 cpu=14:54 elap=15:00
Black(1): Qd7
time used: 15:00
Those I've studied the position with seem to think that taking the exchange is
bad, that white should keep increasing the pressure instead. So, what do you all
make of this wild, crazy game? What do your engines think of these two
positions, and what do you think? Here's the Pono game I referred to earlier,
which we think we've improved on:
[Event "Governor's Cup"]
[Site "Kramatorsk"]
[Date "2001.10.06"]
[Round "8"]
[White "Ponomariov, Ruslan"]
[Black "Borovikov, Vladislav"]
[Result "1-0"]
[WhiteElo "2684"]
[BlackElo "2593"]
[Opening "French: Winawer, advance, 5.a3"]
[ECO "C17"]
[NIC "FR.10"]
1. e4 e6 2. d4 d5 3. Nc3 Bb4 4. e5 c5 5. a3 Ba5 6. Qg4 Ne7 7. dxc5 Bxc3+ 8.
bxc3 Qa5 9. Bd2 Ng6 10. Nf3 Nd7 11. Bd3 Nxc5 12. O-O Ne4 13. Be3 h5 14. Qh3
Qxc3 15. Nd4 Bd7 16. Rab1 Nxe5 17. Bxe4 dxe4 18. Qg3 Ng4 19. Rb3 Qa5 20.
Rxb7 Qe5 21. Qxe5 Nxe5 22. Rfb1 O-O 23. h3 Rfd8 24. Rc7 Rdc8 25. Rbb7 Rxc7
26. Rxc7 a6 27. Bf4 Ng6 28. Bd6 e5 29. Ne2 Be6 30. Ng3 Rc8 31. Nxe4 Rxc7 32.
Bxc7 Bf5 33. f3 f6 34. c4 Ne7 35. Kf2 Kf8 36. Ke3 Ke8 37. f4 exf4+ 38. Bxf4
Bxe4 39. Kxe4 Kd7 40. Bd2 Ke6 41. Kd4 Nf5+ 42. Kc5 Nd6 43. Bf4 Ne4+ 44. Kb6
g5 45. Bh2 Nd2 46. c5 Nc4+ 47. Kb7 Kd5 48. c6 Na5+ 49. Kxa6 Nxc6 50. Kb6 f5
51. a4 h4 52. Bg1 g4 53. Be3 Ne5 54. Bg5 g3 55. Kb5 Nc4 56. Kb4 Nd6 57. a5
Kc6 58. Bxh4 f4 59. Bg5 f3 60. gxf3 g2 61. Be3 Nf5 62. Bf2 Nh4 63. f4 Nf3
64. f5 g1=Q 65. Bxg1 Nxg1 66. f6 Kd7 67. f7 Ke7 68. a6 {White wins} 1-0
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.