interesting Komodo-Stockfish eval difference

Discussion of anything and everything relating to chess playing software and machines.

Moderators: hgm, Rebel, chrisw

jdart
Posts: 4367
Joined: Fri Mar 10, 2006 5:23 am
Location: http://www.arasanchess.org

interesting Komodo-Stockfish eval difference

Post by jdart »

From one of Stefan Pohl's LTC games. Stockfish 9 seems really reluctant to play Komodo's move exf6 here (allowing Bxh2+):

[D] 2kr2r1/pbqn1p1p/1p2pn2/3pP1p1/P1pP1b2/2P2Q2/1PBN1NPP/R1B2RK1 w - - 0 15

But Komodo 11.01 gives this move a + score even with a relatively shallow search.

Komodo wound up with 2 Knights vs Rook and 2 pawns, and won the game:

[pgn]
[Event "long"]
[Site "?"]
[Date "2018.02.26"]
[Round "30.3"]
[White "Komodo 11.2.2 64-bit"]
[Black "Stockfish 190218 64 BMI2"]
[Result "1-0"]
[ECO "D00"]
[PlyCount "177"]
[EventDate "2018.02.23"]
[EventType "tourn"]
[Source "Doe"]

1. d4 d5 2. e3 Nf6 3. Nd2 e6 4. Bd3 Nbd7 5. f4 b6 6. Nh3 Bb7 7. Nf2 c5 8. c3
Bd6 9. Qf3 Qc7 10. O-O O-O-O 11. a4 c4 12. Bc2 Rhg8 13. e4 Bxf4 14. e5 g5 15.
exf6 Bxh2+ 16. Kh1 h5 17. a5 bxa5 18. Rb1 g4 19. Qe2 Qg3 20. Re1 Qh4 21. Nf1
Bc7+ 22. Kg1 Qxf6 23. b3 cxb3 24. Rxb3 Nb6 25. Bd1 a4 26. Ra3 g3 27. Nh3 e5 28.
Be3 Qc6 29. Qa2 Rg4 30. Bxa4 Nxa4 31. Rxa4 a5 32. Rxa5 Bxa5 33. Qxa5 f6 34. Rb1
Qa6 35. Qc5+ Qc6 36. Qe7 h4 37. Nd2 Re8 38. Qf7 Rgg8 39. Qh7 Rh8 40. Qc2 exd4
41. Bxd4 Qd7 42. Qd3 Qd6 43. Nf1 Qa6 44. Qg6 Rhg8 45. Qh7 Rh8 46. Qc2 Qd6 47.
Qg6 Rhg8 48. Qh5 Rh8 49. Qg4+ Qe6 50. Qg7 Qe7 51. Qxe7 Rxe7 52. Bxf6 Rhe8 53.
Bxe7 Rxe7 54. Nf4 Rc7 55. Ne3 Rxc3 56. Nf5 Ba6 57. Nxh4 Bd3 58. Nxd3 Rxd3 59.
Nf5 Kd7 60. Kf1 Ke6 61. Rb3 Rd1+ 62. Ke2 Ra1 63. Nxg3 Ke5 64. Kf3 Rd1 65. Re3+
Kf6 66. Nh5+ Kf7 67. Ke2 Rd4 68. Rf3+ Ke6 69. Rf4 Rc4 70. Rxc4 dxc4 71. g4 Ke5
72. Ke3 Ke6 73. Ke4 c3 74. Ng3 c2 75. Ne2 Kf6 76. Kf4 Ke7 77. Kf5 Kf7 78. g5
Kg7 79. g6 Kf8 80. Kf6 Ke8 81. g7 Kd7 82. Nc1 Kc6 83. Ke5 Kd7 84. g8=Q Kc7 85.
Qg2 Kb6 86. Kd6 Ka6 87. Kc6 Ka5 88. Kc5 Ka4 89. Qa8# 1-0
[/pgn]
Dann Corbit
Posts: 12542
Joined: Wed Mar 08, 2006 8:57 pm
Location: Redmond, WA USA

Re: interesting Komodo-Stockfish eval difference

Post by Dann Corbit »

Komodo is really good at material imbalance. Since Larry wrote the inaugural paper on the subject we should not be surprised.
Taking ideas is not a vice, it is a virtue. We have another word for this. It is called learning.
But sharing ideas is an even greater virtue. We have another word for this. It is called teaching.
User avatar
Guenther
Posts: 4610
Joined: Wed Oct 01, 2008 6:33 am
Location: Regensburg, Germany
Full name: Guenther Simon

Re: interesting Komodo-Stockfish eval difference

Post by Guenther »

[D]2kr2r1/pbbn1p2/5P2/p2pp2p/2pP2pq/2P5/1PB1QNP1/1RB1RNK1 w - - 0 23

Are the evals available too? IMHO 22...e5! should at least hold for Black and it seems White was a bit lucky here.
https://rwbc-chess.de

trollwatch:
Talkchess nowadays is a joke - it is full of trolls/idiots/people stuck in the pleistocene > 80% of the posts fall into this category...
jdart
Posts: 4367
Joined: Fri Mar 10, 2006 5:23 am
Location: http://www.arasanchess.org

Re: interesting Komodo-Stockfish eval difference

Post by jdart »

Game with evals:

[pgn]
[Event "long"]
[Site "?"]
[Date "2018.02.26"]
[Round "30.3"]
[White "Komodo 11.2.2 64-bit"]
[Black "Stockfish 190218 64 BMI2"]
[Result "1-0"]
[ECO "D00"]
[Annotator "0.15;-0.77"]
[PlyCount "177"]
[EventDate "2018.02.23"]
[EventType "tourn"]
[Source "Doe"]
[TimeControl "1200+5"]

{Int@l(R) Core(TM) i7-6700HQ CPU î 2.60GHz 2591 MHz W=26.2 plies; 7.218kN/s;
SALC_V5_hc_10m.ctg B=29.7 plies; 7.322kN/s; SALC_V5_hc_10m.ctg} 1. d4 {0.00/0 0
} d5 {0.00/0 0} 2. e3 {0.00/0 0} Nf6 {0.00/0 0} 3. Nd2 {0.00/0 0} e6 {0.00/0 0}
4. Bd3 {0.00/0 0} Nbd7 {0.00/0 0} 5. f4 {0.00/0 0} b6 {0.00/0 0} 6. Nh3 {
0.00/0 0} Bb7 {0.00/0 0} 7. Nf2 {0.00/0 0} c5 {0.00/0 0} 8. c3 {0.00/0 0} Bd6 {
0.00/0 0} 9. Qf3 {0.00/0 0} Qc7 {0.00/0 0} 10. O-O {0.00/0 0} O-O-O {0.00/0 0}
11. a4 {0.15/27 114} c4 {-0.77/29 123 (Thg8)} 12. Bc2 {0.31/26 26} Rhg8 {
-0.82/26 13 (a5)} 13. e4 {0.60/27 22} Bxf4 {-0.76/28 36 (dxe4)} 14. e5 {
0.49/26 21} g5 {-0.83/27 14} 15. exf6 {0.43/27 35 (Sg4)} Bxh2+ {-1.40/25 11}
16. Kh1 {0.43/5 0} h5 {-1.42/28 37} 17. a5 {0.48/28 68 (De3)} bxa5 {-1.14/30 87
} 18. Rb1 {0.18/25 53 (De2)} g4 {-0.60/26 32 (La6)} 19. Qe2 {0.13/25 14} Qg3 {
0.00/30 129 (La6)} 20. Re1 {0.46/26 36 (Td1)} Qh4 {-0.07/27 12} 21. Nf1 {
0.52/25 12} Bc7+ {-0.12/30 69 (Ld6+)} 22. Kg1 {0.42/24 16} Qxf6 {-0.10/30 26}
23. b3 {0.46/24 13} cxb3 {-0.08/32 13} 24. Rxb3 {0.51/24 14} Nb6 {0.00/32 21}
25. Bd1 {0.41/25 62 (Ld3)} a4 {0.00/28 60 (Td6)} 26. Ra3 {0.53/24 20 (Tb1)} g3
{-0.24/24 6 (Td6)} 27. Nh3 {0.46/21 13} e5 {0.31/27 50} 28. Be3 {0.65/26 92
(dxe5)} Qc6 {0.02/24 16 (Kb8)} 29. Qa2 {0.75/24 14} Rg4 {0.34/29 143 (f6)} 30.
Bxa4 {0.72/25 41} Nxa4 {0.35/29 40} 31. Rxa4 {0.81/27 19} a5 {0.23/29 69} 32.
Rxa5 {0.91/27 37} Bxa5 {0.35/30 18} 33. Qxa5 {0.85/26 12} f6 {0.66/27 28} 34.
Rb1 {0.89/24 13} Qa6 {0.41/29 18 (Td6)} 35. Qc5+ {1.32/27 77 (Db4)} Qc6 {
0.22/27 6} 36. Qe7 {1.28/25 21 (Da5)} h4 {0.69/26 21 (Tgg8)} 37. Nd2 {1.67/26
14 (Df7)} Re8 {0.84/28 35} 38. Qf7 {1.61/26 15} Rgg8 {0.87/28 15} 39. Qh7 {
1.61/28 44 (Sf3)} Rh8 {1.05/27 23} 40. Qc2 {1.74/24 10 (Df5+)} exd4 {1.43/30
60 (Te7)} 41. Bxd4 {1.85/25 15} Qd7 {1.84/30 50} 42. Qd3 {1.76/23 11 (Sf3)} Qd6
{2.05/29 39} 43. Nf1 {1.70/27 43 (c4)} Qa6 {1.58/24 15} 44. Qg6 {1.71/27 38
(Dd1)} Rhg8 {1.64/29 13} 45. Qh7 {1.65/26 21 (Dc2)} Rh8 {1.73/29 5} 46. Qc2 {
1.40/28 70 (Dg6)} Qd6 {1.63/29 21 (Te2)} 47. Qg6 {1.73/26 18 (Dd3)} Rhg8 {
1.66/24 4} 48. Qh5 {1.83/25 7} Rh8 {1.77/30 21 (Dd7)} 49. Qg4+ {1.97/28 9} Qe6
{1.84/31 3} 50. Qg7 {1.69/29 18} Qe7 {2.50/32 20} 51. Qxe7 {1.93/28 6} Rxe7 {
2.62/29 6} 52. Bxf6 {1.96/30 7} Rhe8 {2.56/32 22} 53. Bxe7 {2.01/29 6} Rxe7 {
2.76/30 5} 54. Nf4 {2.15/28 12} Rc7 {2.80/30 12 (Te4)} 55. Ne3 {2.21/29 9} Rxc3
{2.84/25 2} 56. Nf5 {2.24/29 7} Ba6 {2.86/28 6} 57. Nxh4 {2.68/30 37} Bd3 {
2.80/28 2 (Lc4)} 58. Nxd3 {2.94/31 10 (Sxd5)} Rxd3 {3.54/30 5} 59. Nf5 {
3.56/31 31} Kd7 {3.88/33 11} 60. Kf1 {3.57/33 9 (Te1)} Ke6 {3.95/33 5 (Tc3)}
61. Rb3 {3.62/29 6} Rd1+ {4.03/37 5} 62. Ke2 {3.62/5 0} Ra1 {4.18/30 7 (Tg1)}
63. Nxg3 {3.62/30 6} Ke5 {4.37/30 5 (Tg1)} 64. Kf3 {3.78/32 28 (Kf2)} Rd1 {
4.39/31 2 (Kd6)} 65. Re3+ {4.05/25 5 (Kf2)} Kf6 {5.08/34 9} 66. Nh5+ {4.81/32
10} Kf7 {5.76/29 6} 67. Ke2 {5.22/36 18 (Sf4)} Rd4 {6.22/33 5 (Tg1)} 68. Rf3+ {
17.72/27 4 (Kf3)} Ke6 {132.61/34 5} 69. Rf4 {18.86/29 39} Rc4 {#22/48 6} 70.
Rxc4 {#34/18 1} dxc4 {#21/49 3} 71. g4 {#19/1 0} Ke5 {#18/0 0} 72. Ke3 {#18/0 0
} Ke6 {#17/0 0} 73. Ke4 {#17/0 0} c3 {#16/1 0} 74. Ng3 {#16/0 0} c2 {#15/0 0}
75. Ne2 {#15/1 0} Kf6 {#14/0 0} 76. Kf4 {#14/1 0} Ke7 {#13/0 0} 77. Kf5 {
#13/1 0} Kf7 {#12/0 0} 78. g5 {#12/1 0} Kg7 {#11/0 0} 79. g6 {#11/1 0} Kf8 {
#10/0 0} 80. Kf6 {#10/1 0} Ke8 {#9/1 0} 81. g7 {#9/0 0} Kd7 {#8/0 0} 82. Nc1 {
#8/0 0} Kc6 {#7/0 0} 83. Ke5 {#7/0 0} Kd7 {#6/0 0} 84. g8=Q {#6/1 0} Kc7 {
#5/1 0} 85. Qg2 {#5/0 0} Kb6 {#4/0 0} 86. Kd6 {#4/1 0} Ka6 {#3/0 0} 87. Kc6 {
#3/1 0} Ka5 {#2/0 0} 88. Kc5 {#2/0 0} Ka4 {#1/0 0} 89. Qa8# {#1/1 0} 1-0
[/pgn]
User avatar
pohl4711
Posts: 2441
Joined: Sat Sep 03, 2011 7:25 am
Location: Berlin, Germany
Full name: Stefan Pohl

Re: interesting Komodo-Stockfish eval difference

Post by pohl4711 »

My advice: Use my SALC books and opening-sets for engine-play and you will get many, many of these interesting, spectacular games. And less boring draws.

http://www.sp-cc.de/salc-openings.htm


Stefan (SPCC)