TC 120+30 in addition ponder
Hardware used: http://pasteboard.co/crZfUwOU1.png
Next Komodo vs GM match - 2 pawns
Moderator: Ras
-
- Posts: 2822
- Joined: Fri Sep 25, 2015 9:38 pm
- Location: Sortland, Norway
-
- Posts: 2822
- Joined: Fri Sep 25, 2015 9:38 pm
- Location: Sortland, Norway
Re: Next Komodo vs GM match - 2 pawns
Exchange Odds
-----------------
[pgn][Event "Auto 232 Game"]
[Site "https://lichess.org/1aHXRXcp"]
[Date "2016.07.17"]
[White "Stockfish_x64_modern (3200)"]
[Black "H4 Pro | Contempt 0 (3200)"]
[Result "1/2-1/2"]
[WhiteElo "?"]
[BlackElo "?"]
[PlyCount "301"]
[Variant "From Position"]
[TimeControl "-"]
[ECO "?"]
[Opening "?"]
[Termination "Normal"]
[FEN "rnbqk1nr/pppppppp/8/8/8/8/PPPPPPPP/1NBQKBNR w Kkq - 0 11"]
[SetUp "1"]
[Annotator "lichess.org"]
11. e4 c5 12. Nc3 Nc6 13. Bb5 Nd4 14. Nge2 Nxb5 15. Nxb5 d6 16. c4 a6 17. Nbc3 Nf6 18. O-O b5 19. d3 O-O 20. Bg5 Ng4 21. b3 h6 22. Bc1 Nf6 23. Ng3 e6 24. Be3 Qa5 25. Qc1 Ng4 26. Bd2 Qd8 27. Nce2 Qe7 28. Qc2 Ne5 29. f4 Nd7 30. f5 Nf6 31. Qb2 bxc4 32. dxc4 Bb7 33. Qc1 Kh7 34. Rf4 Rfe8 35. Nc3 Rac8 36. h3 Bc6 37. Kh2 Nd7 38. Rf2 Rg8 39. Qd1 Rcd8 40. Qg4 Rde8 41. Qe2 Bb7 42. Qh5 Nf6 43. Qf3 Rgf8 44. Qf4 Bc6 45. Be3 Kh8 46. Kg1 Nh7 47. Bd2 Rg8 48. Qg4 Ref8 49. Qh5 Qd8 50. Bf4 Ba8 51. Qd1 Qf6 52. Qe1 e5 53. Bd2 Rc8 54. Rf3 Qd8 55. Nf1 Nf6 56. g4 Nh7 57. h4 Bc6 58. Rh3 a5 59. Qe2 a4 60. bxa4 Ra8 61. g5 Bxa4 62. Nh2 Bc6 63. Rd3 Qe7 64. Ng4 h5 65. Ne3 g6 66. Ned5 Qb7 67. Nf6 Rgd8 68. Be1 Ra3 69. Qd2 Qe7 70. Nfd5 Qa7 71. Nf6 Qb8 72. Kh2 Kg7 73. Qf2 Ra7 74. Rf3 Rf8 75. a4 Rb7 76. a5 Nxf6 77. gxf6+ Kxf6 78. fxg6+ Kg7 79. Bd2 f6 80. a6 Ra7 81. Qe2 Qd8 82. Nd5 Bxd5 83. cxd5 Rh8 84. Qf1 Qe7 85. Rf5 Raa8 86. Be3 Rab8 87. Rf2 Rbc8 88. Rf5 Rcf8 89. Rf2 Qc7 90. Rb2? { (0.00 → -1.18) Mistake. Best move was Rg3. } (90. Rg3 Rh8 91. Qf5 Bd7 92. Qf1 Bc6 93. Qf5 Bd7 94. Qf1 Bc6 95. Qf5) 90... Rb8 91. Rf2 Qe7?! { (-1.34 → -0.40) Inaccuracy. Best move was Rxa6. } (91... Rxa6 92. Rf5 Rh8 93. Rxh5 Kxg6 94. Nd5 Kf7 95. Nxf6 Ra2 96. Nd5 Rxd2 97. Qxd2 Rxh5 98. Qf2+) 92. Rf5 Ra8 93. Rf2 Rhf8 94. Qe2 Qd7 95. Rf5 Rh8 96. Qf3 Raf8 97. Qf1 Qa4 98. Bg5 Qc2+ 99. Kg1 fxg5 100. Rf7+ Rxf7 101. Qxf7+ Kh6 102. g7 Qd1+ 103. Kf2 Qd4+ 104. Kf1 Qd3+ 105. Kf2 Qd2+ 106. Kf1 Qc1+ 107. Kf2 Ra8 108. g8=N+ Rxg8 109. Qxg8 Qf4+ 110. Kg1 Qg4+ 111. Kf1 Qd1+ 112. Kf2 Qc2+ 113. Kg1 Qb1+ 114. Kf2 Qb2+ 115. Ke1 Qb4+ 116. Kf2 Qd4+ 117. Kf1 Qd3+ 118. Kf2 Qd2+ 119. Kf1 Qf4+ 120. Kg1 Qe3+ 121. Kf1 Qh3+ 122. Ke2 Qg4+ 123. Kf1 Qf3+ 124. Kg1 Qd1+ 125. Kf2 Qd4+ 126. Kf1 Qa1+ 127. Kf2 Qa2+ 128. Ke3 Qa3+ 129. Kf2 Qa2+ 130. Ke1 Qb1+ 131. Kf2 Qc2+ 132. Kf3 g4+ 133. Ke3 Qb3+ 134. Kf2 Qf3+ 135. Ke1 Qxe4+ 136. Kd2 Qf4+ 137. Kd1 Qf3+ 138. Kd2 Qg2+ 139. Kc1 Qf1+ 140. Kc2 Qc4+ 141. Kd2 Qxa6 142. Qe6+ Kg7 143. Qe7+ Kg6 144. Qe6+ Kh7 145. Qd7+ Kh6 146. Qe6+ Kg7 147. Qe7+ Kh8 148. Qd8+ Kh7 149. Qe7+ Kg8 150. Qe8+ Kg7 151. Qe7+ { Draw } 1/2-1/2[/pgn]
https://en.lichess.org/1aHXRXcp#20
-----------------
[pgn][Event "Auto 232 Game"]
[Site "https://lichess.org/1aHXRXcp"]
[Date "2016.07.17"]
[White "Stockfish_x64_modern (3200)"]
[Black "H4 Pro | Contempt 0 (3200)"]
[Result "1/2-1/2"]
[WhiteElo "?"]
[BlackElo "?"]
[PlyCount "301"]
[Variant "From Position"]
[TimeControl "-"]
[ECO "?"]
[Opening "?"]
[Termination "Normal"]
[FEN "rnbqk1nr/pppppppp/8/8/8/8/PPPPPPPP/1NBQKBNR w Kkq - 0 11"]
[SetUp "1"]
[Annotator "lichess.org"]
11. e4 c5 12. Nc3 Nc6 13. Bb5 Nd4 14. Nge2 Nxb5 15. Nxb5 d6 16. c4 a6 17. Nbc3 Nf6 18. O-O b5 19. d3 O-O 20. Bg5 Ng4 21. b3 h6 22. Bc1 Nf6 23. Ng3 e6 24. Be3 Qa5 25. Qc1 Ng4 26. Bd2 Qd8 27. Nce2 Qe7 28. Qc2 Ne5 29. f4 Nd7 30. f5 Nf6 31. Qb2 bxc4 32. dxc4 Bb7 33. Qc1 Kh7 34. Rf4 Rfe8 35. Nc3 Rac8 36. h3 Bc6 37. Kh2 Nd7 38. Rf2 Rg8 39. Qd1 Rcd8 40. Qg4 Rde8 41. Qe2 Bb7 42. Qh5 Nf6 43. Qf3 Rgf8 44. Qf4 Bc6 45. Be3 Kh8 46. Kg1 Nh7 47. Bd2 Rg8 48. Qg4 Ref8 49. Qh5 Qd8 50. Bf4 Ba8 51. Qd1 Qf6 52. Qe1 e5 53. Bd2 Rc8 54. Rf3 Qd8 55. Nf1 Nf6 56. g4 Nh7 57. h4 Bc6 58. Rh3 a5 59. Qe2 a4 60. bxa4 Ra8 61. g5 Bxa4 62. Nh2 Bc6 63. Rd3 Qe7 64. Ng4 h5 65. Ne3 g6 66. Ned5 Qb7 67. Nf6 Rgd8 68. Be1 Ra3 69. Qd2 Qe7 70. Nfd5 Qa7 71. Nf6 Qb8 72. Kh2 Kg7 73. Qf2 Ra7 74. Rf3 Rf8 75. a4 Rb7 76. a5 Nxf6 77. gxf6+ Kxf6 78. fxg6+ Kg7 79. Bd2 f6 80. a6 Ra7 81. Qe2 Qd8 82. Nd5 Bxd5 83. cxd5 Rh8 84. Qf1 Qe7 85. Rf5 Raa8 86. Be3 Rab8 87. Rf2 Rbc8 88. Rf5 Rcf8 89. Rf2 Qc7 90. Rb2? { (0.00 → -1.18) Mistake. Best move was Rg3. } (90. Rg3 Rh8 91. Qf5 Bd7 92. Qf1 Bc6 93. Qf5 Bd7 94. Qf1 Bc6 95. Qf5) 90... Rb8 91. Rf2 Qe7?! { (-1.34 → -0.40) Inaccuracy. Best move was Rxa6. } (91... Rxa6 92. Rf5 Rh8 93. Rxh5 Kxg6 94. Nd5 Kf7 95. Nxf6 Ra2 96. Nd5 Rxd2 97. Qxd2 Rxh5 98. Qf2+) 92. Rf5 Ra8 93. Rf2 Rhf8 94. Qe2 Qd7 95. Rf5 Rh8 96. Qf3 Raf8 97. Qf1 Qa4 98. Bg5 Qc2+ 99. Kg1 fxg5 100. Rf7+ Rxf7 101. Qxf7+ Kh6 102. g7 Qd1+ 103. Kf2 Qd4+ 104. Kf1 Qd3+ 105. Kf2 Qd2+ 106. Kf1 Qc1+ 107. Kf2 Ra8 108. g8=N+ Rxg8 109. Qxg8 Qf4+ 110. Kg1 Qg4+ 111. Kf1 Qd1+ 112. Kf2 Qc2+ 113. Kg1 Qb1+ 114. Kf2 Qb2+ 115. Ke1 Qb4+ 116. Kf2 Qd4+ 117. Kf1 Qd3+ 118. Kf2 Qd2+ 119. Kf1 Qf4+ 120. Kg1 Qe3+ 121. Kf1 Qh3+ 122. Ke2 Qg4+ 123. Kf1 Qf3+ 124. Kg1 Qd1+ 125. Kf2 Qd4+ 126. Kf1 Qa1+ 127. Kf2 Qa2+ 128. Ke3 Qa3+ 129. Kf2 Qa2+ 130. Ke1 Qb1+ 131. Kf2 Qc2+ 132. Kf3 g4+ 133. Ke3 Qb3+ 134. Kf2 Qf3+ 135. Ke1 Qxe4+ 136. Kd2 Qf4+ 137. Kd1 Qf3+ 138. Kd2 Qg2+ 139. Kc1 Qf1+ 140. Kc2 Qc4+ 141. Kd2 Qxa6 142. Qe6+ Kg7 143. Qe7+ Kg6 144. Qe6+ Kh7 145. Qd7+ Kh6 146. Qe6+ Kg7 147. Qe7+ Kh8 148. Qd8+ Kh7 149. Qe7+ Kg8 150. Qe8+ Kg7 151. Qe7+ { Draw } 1/2-1/2[/pgn]
https://en.lichess.org/1aHXRXcp#20
-
- Posts: 34
- Joined: Sat Jul 02, 2016 1:47 pm
Re: Next Komodo vs GM match - 2 pawns
The problem is that komodo plays very drawish and is forcing each simplification of the complexity of a position in order to receive a forced draw or even a 0.01 evaluation increase in odd games.
This is why odd games with komodo (and stockfish) are counterproductive to demonstrate the world the achievments of todays chess programming.
The old vehicle "comtempt factor" found in the early 1970s is absolute not the solution for 2016s chess programms to deal with odd games.
In all odd games i observed the defensive drawish playing style of komodo.
The climax was: In one of the last odd games the GM was the guy who avoided the draw that komodo wanted to force too early. The GM was the guy how wanted to make the game more interesting.
For the future, odd games are boring with komodo until the programmers make deep changes in komodo to prepare komodo for odd games against much weaker opponents.
This is why odd games with komodo (and stockfish) are counterproductive to demonstrate the world the achievments of todays chess programming.
The old vehicle "comtempt factor" found in the early 1970s is absolute not the solution for 2016s chess programms to deal with odd games.
In all odd games i observed the defensive drawish playing style of komodo.
The climax was: In one of the last odd games the GM was the guy who avoided the draw that komodo wanted to force too early. The GM was the guy how wanted to make the game more interesting.
For the future, odd games are boring with komodo until the programmers make deep changes in komodo to prepare komodo for odd games against much weaker opponents.
-
- Posts: 6259
- Joined: Sun Jan 10, 2010 6:15 am
- Location: Maryland USA
- Full name: Larry Kaufman
Re: Next Komodo vs GM match - 2 pawns
I use (rating diff)/15 as a guide for setting Contempt in a normal game. So about 50 would be right in this match without handicap. I raise it a bit for handicap games; I used 60. For today's games I'll raise it further to 75. I think that play suffers too much to use values like 150 when the handicap is only rated around this level. Maybe half the handicap value is reasonable.Laskos wrote:Do you think Komodo should pro-actively seek draws? Several moves before the sacrifice, Komodo was effectively one pawn down in all reasonable variations, having positional advantage, with an eval of -0.70. The position was winnable for Komodo, if GM continued with his inaccuracies. With Contempt set at 150, Komodo doesn't go for the draw-sacrifice even when given direct opportunity.lkaufman wrote:First game was a draw; Komodo played the famous Lasker two-bishop sacrifice to force perpetual check still down two pawns.
We could make a change to avoid early rep draws without changing other play. Perhaps we will in the future.
Komodo rules!
-
- Posts: 6259
- Joined: Sun Jan 10, 2010 6:15 am
- Location: Maryland USA
- Full name: Larry Kaufman
Re: Next Komodo vs GM match - 2 pawns
Komodo won game 3 when Erenburg blundered a piece. Much of his advantage was gone, but he was still better but very short of time (down to 2 min plus increment 15". So score 2-1 in favor of Komodo. Last game in about 40 min.
Komodo rules!
-
- Posts: 1346
- Joined: Sat Apr 19, 2014 1:47 pm
Re: Next Komodo vs GM match - 2 pawns
If last game is a draw, I win my bet twice in a row !
-
- Posts: 6259
- Joined: Sun Jan 10, 2010 6:15 am
- Location: Maryland USA
- Full name: Larry Kaufman
Re: Next Komodo vs GM match - 2 pawns
Sorry, Komodo won the final game to take the match 3 to 1. Good day for Komodo, since we also won the World Rapid Championships with 14.5 out of 15.JJJ wrote:If last game is a draw, I win my bet twice in a row !
Komodo rules!
-
- Posts: 1346
- Joined: Sat Apr 19, 2014 1:47 pm
Re: Next Komodo vs GM match - 2 pawns
Komodo did 2 wins with 2 pawn handicap ? Wooo... But at least I was very close !
Congrat Komodo, I was sure improving 50 elo since last time he play 2 pawn handicap would have make a great difference.
Congrat Komodo, I was sure improving 50 elo since last time he play 2 pawn handicap would have make a great difference.
-
- Posts: 6259
- Joined: Sun Jan 10, 2010 6:15 am
- Location: Maryland USA
- Full name: Larry Kaufman
Re: Next Komodo vs GM match - 2 pawns
JJJ wrote:Komodo did 2 wins with 2 pawn handicap ? Wooo... But at least I was very close !
Congrat Komodo, I was sure improving 50 elo since last time he play 2 pawn handicap would have make a great difference.
Here is game 4. I seem to have done something wrong in saving game 3, might have to get the score from chess.com.
Scoring 3 out of 4 giving two non-edge pawns to a nearly 2600 FIDE GM seems unbelievable to me. I never imagined such a thing back when Rybka was giving one pawn (rotating thru all 8) to similarly rated GMs. Erenburg suggested that in future matches we should give both material and time handicaps to GMs. Maybe the standard 90' + 30" for GM vs. 15' + 5" for Komodo, plus f7 handicap. He thinks that with double the time GMs will do much better with these handicaps. Perhaps we'll try this at some point.
Event "Rapid 50m+35s-45m+15s"]
[Site "Potomac"]
[Date "2016.07.17"]
[Round "4"]
[White "Komodo 1668.00 64-bit"]
[Black "Erenburg, Sergey"]
[Result "1-0"]
[Annotator "Kaufman,Larry"]
[SetUp "1"]
[FEN "rnbqkbnr/pppppppp/8/8/8/8/PPPP2PP/RNBQKBNR w KQkq - 0 1"]
[PlyCount "75"]
[TimeControl "2700+15"]
{12288MB, Book2016.ctg, 24core} 1. Nf3 {0} d5 {17} 2. d4 {0} Bg4 {17} 3. c4 {0}
e6 {121} 4. Be3 {106} c6 {222} 5. Nc3 {193} Nd7 {60} 6. cxd5 {122} cxd5 {176}
7. Bd3 {0} Ngf6 {43} 8. O-O {92} Bd6 {145} 9. Qb3 {0} Bxf3 {141} (9... Qb8) 10.
Rxf3 {60} Qb6 {664} (10... Ng4 11. Bf4 {7} Qb8 (11... Bxf4 {10} 12. Rxf4 {1}
Qg5 13. Raf1 Ne3 14. R1f2 Nxg2 15. R4f3) 12. Bxd6 Qxd6 13. Rg3 h5) (10... Qb8)
11. Nb5 {71} O-O {127} 12. Nxd6 {109} Qxd6 {16} 13. Rc1 {167} Qb6 {288} 14. Qc2
{0} Ne4 {128} 15. Bxe4 {0} dxe4 {15} 16. Qxe4 {49} Rac8 {48} 17. Rd1 {45} f5 {
120} (17... f5 {8}) 18. Qb1 {0} Qc6 {51} 19. d5 {56} exd5 {39} 20. Rxf5 {18}
Qe6 {40} 21. Rxf8+ {165} Rxf8 {25} 22. Qd3 {23} Nf6 {361} 23. Bxa7 {73} b6 {55}
24. Qd4 {3} Rc8 {125} (24... Qe2 {10} 25. Bxb6 Rc8 {4} 26. Qd3 {12} Qxb2) 25.
Qxb6 {42} Qe2 {60} (25... Qd7 {18} 26. Qf2 {3} Ne4 {3} 27. Qd4 {0} Rc2 {1} 28.
h3 {3} Nf6 {9} 29. a3 {4} Qb5 {0}) 26. Rf1 {49} Rc2 {19} 27. Qd8+ {53} Kf7 {42}
28. Qxd5+ {31} Kg6 {48} 29. Qf5+ {0} Kf7 {20} 30. Qh3 {43} Qe4 {124} (30... Qe6
) 31. Qb3+ {36} Kg6 {25} 32. Bf2 {15} h6 {102} 33. Re1 {80} Qf5 {44} 34. Bd4 {
22} Nd5 {52} 35. Qg3+ {42} Qg5 {90} 36. Qd3+ {0} Qf5 {25} 37. Qxf5+ {15} Kxf5 {
21} 38. Re5+ {32} 1-0
Komodo rules!
-
- Posts: 10948
- Joined: Wed Jul 26, 2006 10:21 pm
- Full name: Kai Laskos
Re: Next Komodo vs GM match - 2 pawns
Would it be possible to have a match in the same conditions as this one, but against GM Petr Neuman?lkaufman wrote:JJJ wrote:Komodo did 2 wins with 2 pawn handicap ? Wooo... But at least I was very close !
Congrat Komodo, I was sure improving 50 elo since last time he play 2 pawn handicap would have make a great difference.
Here is game 4. I seem to have done something wrong in saving game 3, might have to get the score from chess.com.
Scoring 3 out of 4 giving two non-edge pawns to a nearly 2600 FIDE GM seems unbelievable to me. I never imagined such a thing back when Rybka was giving one pawn (rotating thru all 8) to similarly rated GMs. Erenburg suggested that in future matches we should give both material and time handicaps to GMs. Maybe the standard 90' + 30" for GM vs. 15' + 5" for Komodo, plus f7 handicap. He thinks that with double the time GMs will do much better with these handicaps. Perhaps we'll try this at some point.
Here the performance of Komodo was amazing. Handicaps were roughly 600-650 ELO points (as derived by us awhile ago), 3-1 is 200 points, so a performance of Komodo in these conditions of around 3400 FIDE. Previously, we had a combined performance of Komodo in these handicap matches of 3250 FIDE.