Komodo Dragon Vs. GM Perelshteyn Where is the Link ?

[Chessqueen]

I replayed this game from GM Perelshteyn Vs Komodo Dragon using the same Opening up to move 7 the 2nd game was using Contempt = 125 T/C 2'+2"


[pgn][Event "knight odds match"]
[Site "Chess.com"]
[Date "2021.11.06"]
[Round "2"]
[White "Dragon-2.5-64bit-avx2"]
[Black "Gaviota-win32-0.84"]
[Result "1-0"]
[BlackElo "2500"]
[WhiteElo "2500"]
[TimeControl "120+2"]
[SetUp "1"]
[FEN "rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/R1BQKBNR w KQkq - 0 1"]
[Termination "normal"]
[PlyCount "137"]
[WhiteType "program"]
[BlackType "program"]

1. f4 {[%clk 0:15:10]} d5 {[%clk 0:15:08.5]} 2. Nf3 {[%clk 0:15:19.9]} g6
{[%clk 0:15:14.7]} 3. e3 {[%clk 0:15:29.8]} Bg7 {[%clk 0:15:23.4]} 4. Be2
{[%clk 0:15:39.7]} c5 {[%clk 0:15:26.3]} 5. O-O {[%clk 0:15:45.5]} Nc6
{[%clk 0:15:34.5]} 6. Qe1 {[%clk 0:15:48.4]} Bg4 {[%clk 0:15:35.7]} 7. d3
Nf6 8. Bd1 O-O 9. e4 dxe4 10. dxe4 e5 11. f5 gxf5 12. Bg5 f4 13. Kh1 h6 14.
Bh4 Qb6 15. c3 Be6 16. b3 Ng4 17. Rg1 Ne3 18. g3 Kh7 19. gxf4 exf4 20. Bg5
Nxd1 21. Rxd1 Rg8 22. Qh4 Qa5 23. Bxf4 Qxc3 24. Rg3 Qf6 25. Bg5 Qb2 26.
Rdg1 Rad8 27. Be3 Qf6 28. Rxg7+ Qxg7 29. Ng5+ Kh8 30. Nxe6 fxe6 31. Rxg7
Kxg7 32. Qxh6+ Kf7 33. Bg5 Rd6 34. Qf6+ Ke8 35. e5 Rd1+ 36. Kg2 Rd7 37.
Qxe6+ Ne7 38. h4 Rg6 39. Qg4 a6 40. Qh5 Rc7 41. Kh3 Rc6 42. Bf6 Re6 43. Qf3
b6 44. h5 Rh6 45. Kh4 b5 46. Kg5 Rh7 47. Qd3 Rf7 48. h6 c4 49. bxc4 bxc4
50. Qxc4 Rc6 51. Qe4 Rc1 52. h7 Rg1+ 53. Kh6 Rg6+ 54. Qxg6 Nxg6 55. Kxg6
Rxh7 56. Kxh7 Kd7 57. a4 a5 58. Kg6 Ke6 59. Kg7 Kd5 60. Kf7 Kc4 61. e6 Kb3
62. e7 Kxa4 63. e8=Q+ Kb4 64. Qc6 a4 65. Ke6 a3 66. Kd5 a2 67. Qb6+ Ka4 68.
Bc3 a1=Q 69. Qb4# 1-0[/pgn]

[Chessqueen]

I replayed this game from GM Perelshteyn Vs Komodo Dragon using the same Opening up to move 7 the 2nd game was using Contempt = 65 T/C 2'+2"


[pgn][Event "knight odds match"]
[Site "Chess.com"]
[Date "2021.11.06"]
[Round "4"]
[White "PlayKomodo"]
[Black "Gaviota-win32-0.84"]
[Result "1/2 1/2"]
[BlackElo "2500"]
[WhiteElo "2500"]
[TimeControl "120+2"]
[SetUp "1"]
[FEN "rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKB1R w KQkq - 0 1"]
[Termination "normal"]
[PlyCount "338"]
[WhiteType "program"]
[BlackType "program"]

1. c4 {[%clk 0:15:10]} e5 {[%clk 0:15:06.4]} 2. Nc3 {[%clk 0:15:19.9]} Nc6
{[%clk 0:15:09.4]} 3. g3 {[%clk 0:15:29.8]} g6 {[%clk 0:15:11.1]} 4. Bg2
{[%clk 0:15:33.2]} Bg7 {[%clk 0:15:19.9]} 5. d3 {[%clk 0:15:36]} Nge7
{[%clk 0:15:24.1]} 6. Rb1 {[%clk 0:15:40.4]} a5 {[%clk 0:15:30.5]} 7. a3
{[%clk 0:15:50.1]} O-O {[%clk 0:15:24.5]} 8. e3 d6 9. h4 Be6 10. h5 h6 11.
hxg6 fxg6 12. b4 e4 13. Nxe4 axb4 14. axb4 Ra2 15. O-O b5 16. Qb3 Ra4 17.
Bd2 Kh7 18. Qc2 Ra3 19. cxb5 Ra2 20. Qd1 Ne5 21. Bc3 Nd5 22. Bxe5 Bxe5 23.
Qc1 Bg7 24. d4 Qd7 25. Qc6 Qxc6 26. bxc6 Rb8 27. Rfc1 Rb5 28. Bf1 Rxb4 29.
Rxb4 Nxb4 30. Nc3 Ra5 31. Rb1 Nxc6 32. Nb5 Bf5 33. Rc1 Nb4 34. Nxc7 Bf8 35.
Ne8 Bc2 36. Nc7 Bf5 37. Ne8 Bc2 38. Nc7 Ra2 39. Bc4 Rb2 40. Nd5 Nxd5 41.
Bxd5 h5 42. Kf1 Be7 43. Ra1 Kg7 44. Ra7 Kf6 45. Kg1 Rb1+ 46. Kh2 g5 47. Ra5
Rb2 48. Kg1 Bg6 49. Bf3 h4 50. gxh4 gxh4 51. Kg2 Bf5 52. Bd5 Bg4 53. Be4
Rb8 54. Kh2 Bf8 55. Kg1 Rc8 56. Kh2 Rb8 57. Kg1 Rc8 58. Rb5 Bd7 59. Rh5 h3
60. Kh2 Rc3 61. Rh8 Bg7 62. Rd8 Ke7 63. Ra8 Bc8 64. Ra7+ Kf6 65. Bd5 Rc2
66. Kg3 Rc1 67. Rf7+ Kg6 68. Ra7 Bf5 69. Ra8 Bc8 70. Ra7 Bf8 71. Rf7 Bh6
72. Re7 Bf5 73. Re8 Rg1+ 74. Kh2 Rf1 75. Kg3 Rg1+ 76. Kh2 Rf1 77. Kg3 Kf6
78. Rh8 Bg7 79. Rd8 Ke7 80. Rb8 Rg1+ 81. Kh2 Rd1 82. Rb7+ Kf6 83. Rb8 Rd3
84. Kg3 Kg6 85. Rg8 Rc3 86. Bf3 Kf6 87. Bd5 Bd7 88. Kh2 Rc2 89. Kg3 Rc3 90.
Kh2 Rc2 91. Kg3 Rc1 92. Ra8 Bc8 93. Ra7 Bh6 94. Rf7+ Kg6 95. Re7 Bf5 96.
Re8 Kf6 97. Rd8 Ke7 98. Ra8 Rg1+ 99. Kh2 Rf1 100. Ra2 Kf6 101. Kg3 Rg1+
102. Kh2 Rf1 103. Kg3 Rg1+ 104. Kh2 Rd1 105. Kg3 Re1 106. Ra6 Rg1+ 107. Kh2
Rf1 108. Ra2 Be6 109. Bxe6 Kxe6 110. Kxh3 Rg1 111. Ra5 Rg5 112. Ra6 Kd5
113. e4+ Kxe4 114. Rxd6 Bg7 115. Rd7 Bxd4 116. Rf7 Be5 117. f3+ Ke3 118.
Kh4 Bf4 119. Ra7 Kxf3 120. Ra3+ Ke2 121. Rb3 Rg1 122. Ra3 Be3 123. Ra6 Rg7
124. Rd6 Bf4 125. Rd5 Rg1 126. Kh5 Ke3 127. Rf5 Rg7 128. Rd5 Ke4 129. Ra5
Rg2 130. Ra3 Kf5 131. Kh4 Rh2+ 132. Rh3 Rb2 133. Rf3 Rb6 134. Kh3 Rh6+ 135.
Kg2 Rh2+ 136. Kg1 Rh8 137. Rb3 Ke4 138. Kf1 Rh1+ 139. Ke2 Rh2+ 140. Ke1 Bd6
141. Kf1 Be5 142. Ra3 Bf4 143. Ke1 Bd6 144. Ra8 Bc5 145. Rd8 Bd4 146. Rf8
Kd3 147. Rf3+ Be3 148. Kf1 Ke4 149. Rg3 Rf2+ 150. Kg1 Ra2+ 151. Kf1 Kd3
152. Rh3 Rf2+ 153. Kg1 Rd2+ 154. Kh1 Ke4 155. Rh4+ Bf4 156. Rh3 Rb2 157.
Kg1 Be3+ 158. Kf1 Bd4 159. Ke1 Bf2+ 160. Kf1 Be3 161. Ke1 Bd4 162. Rh7 Kd3
163. Rh3+ Be3 164. Kf1 Rc2 165. Rg3 Rd2 166. Rh3 Rb2 167. Rg3 Rc2 168. Rh3
Rc7 169. Kg2 Rc4 {50 moves rule} *[/pgn]

[Chessqueen]

I replayed this game from GM Perelshteyn Vs Komodo Dragon using the same Opening up to move 7 the 2nd game was using Contempt = 65 T/C 2'+2"

[pgn][Event "?"]
[Site "Chess.com"]
[Date "2021.11.06"]
[Round "4"]
[White "PlayKomodo"]
[Black "Gaviota-win32-0.84"]
[Result "1/2 1/2"]
[BlackElo "2500"]
[WhiteElo "2500"]
[TimeControl "120+2"]
[SetUp "1"]
[FEN "rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKB1R w KQkq - 0 1"]
[Termination "adjudication"]
[PlyCount "205"]
[WhiteType "program"]
[BlackType "program"]

1. c4 {[%clk 0:15:10]} e5 {[%clk 0:15:06.4]} 2. Nc3 {[%clk 0:15:19.9]} Nc6
{[%clk 0:15:09.4]} 3. g3 {[%clk 0:15:29.8]} g6 {[%clk 0:15:11.1]} 4. Bg2
{[%clk 0:15:33.2]} Bg7 {[%clk 0:15:19.9]} 5. d3 {[%clk 0:15:36]} Nge7
{[%clk 0:15:24.1]} 6. Rb1 {[%clk 0:15:40.4]} a5 {[%clk 0:15:30.5]} 7. a3
O-O 8. b4 axb4 9. axb4 e4 10. Nxe4 d5 11. cxd5 Nxd5 12. O-O Nd4 13. Bb2 Bg4
14. Bxd4 Bxd4 15. Qd2 Ra3 16. Qc1 Rc3 17. Qd2 Ra3 18. Qc1 Rc3 19. Qd2 Rc6
20. b5 Re6 21. e3 Bb6 22. Ra1 Rfe8 23. Ra4 f5 24. Ng5 Rd6 25. Bxd5+ Rxd5
26. Qa2 Kg7 27. Ra8 Qxa8 28. Qxd5 Re7 29. h3 Be2 30. Re1 Kf6 31. Qc4 Bxd3
32. Qh4 Qe8 33. Nxh7+ Kg7 34. Ng5 Kg8 35. Qb4 Bxb5 36. Qb3+ Kf8 37. Qb2 Kg8
38. Qb3+ Kf8 39. Qc3 Re5 40. Rd1 Bc6 41. Kh2 Be4 42. Nh7+ Kg7 43. Ng5 Kf8
44. Nh7+ Kg7 45. Ng5 Qb5 46. h4 Bc6 47. Kh3 Kg8 48. Qa1 Be8 49. Qa2+ Kf8
50. Qa1 Kg8 51. Qa2+ Kf8 52. Qa1 Qa5 53. Qb2 Bxe3 54. fxe3 Qc5 55. Rc1 Qd6
56. Qxb7 Rxe3 57. Qxc7 Rxg3+ 58. Kh2 Rd3+ 59. Qxd6+ Rxd6 60. Kg3 Bd7 61.
Kf4 Rd4+ 62. Ke5 Rxh4 63. Kf6 Ra4 64. Nh7+ Kg8 65. Kxg6 Be8+ 66. Kxf5 Kxh7
67. Ke6 Ra6+ 68. Ke7 Bc6 69. Re1 Kh6 70. Rg1 Kh5 71. Kd6 Bg2+ 72. Ke5 Ra2
73. Re1 Kg4 74. Rb1 Kf3 75. Rb8 Ke3 76. Rb3+ Kd2 77. Rg3 Ra5+ 78. Kd4 Ba8
79. Rd3+ Ke1 80. Ke3 Bd5 81. Rc3 Kd1 82. Kd3 Ra4 83. Rc8 Be4+ 84. Ke3 Bb1
85. Rd8+ Kc1 86. Rd6 Re4+ 87. Kf2 Kc2 88. Rd7 Kc3 89. Rd8 Bc2 90. Rd6 Rh4
91. Kf3 Bb1 92. Rc6+ Kd4 93. Rb6 Be4+ 94. Kg3 Rh1 95. Rb4+ Ke5 96. Rb3 Rg1+
97. Kf2 Rg2+ 98. Kf1 Ra2 99. Re3 Kd4 100. Rg3 Bd5 101. Ke1 Bg2 102. Rb3 Be4
103. Rg3 {User Adjudication} *[/pgn]

[Chessqueen]

I replayed this game from GM Perelshteyn Vs Komodo Dragon using the same Opening up to move 7 the 2nd game was using Contempt = 65 and this game with Contempt = 125 T/C 2'+2"

Note: We should take the score of GM Perelshteyn with a grain of salt, since in only 6 games it could have gone either way, and if the amount of games and choice of Openings instead of 3 with the b1 and 3 with the g1 Knight missing WAS 3 games with the b1 missing and 4 games with the g1 missing total of 7 games, Komodo Dragon could have been the winner of the match


[pgn][Event "knight odds match"]
[Site "Chess.com"]
[Date "2021.11.06"]
[Round "4"]
[White "PlayKomodo"]
[Black "Gaviota-win32-0.84"]
[Result "1/2 1/2"]
[BlackElo "2500"]
[WhiteElo "2500"]
[TimeControl "120+2"]
[SetUp "1"]
[FEN "rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKB1R w KQkq - 0 1"]
[Termination "normal"]
[PlyCount "338"]
[WhiteType "program"]
[BlackType "program"]

1. c4 {[%clk 0:15:10]} e5 {[%clk 0:15:06.4]} 2. Nc3 {[%clk 0:15:19.9]} Nc6
{[%clk 0:15:09.4]} 3. g3 {[%clk 0:15:29.8]} g6 {[%clk 0:15:11.1]} 4. Bg2
{[%clk 0:15:33.2]} Bg7 {[%clk 0:15:19.9]} 5. d3 {[%clk 0:15:36]} Nge7
{[%clk 0:15:24.1]} 6. Rb1 {[%clk 0:15:40.4]} a5 {[%clk 0:15:30.5]} 7. a3
{[%clk 0:15:50.1]} O-O {[%clk 0:15:24.5]} 8. e3 d6 9. h4 Be6 10. h5 h6 11.
hxg6 fxg6 12. b4 e4 13. Nxe4 axb4 14. axb4 Ra2 15. O-O b5 16. Qb3 Ra4 17.
Bd2 Kh7 18. Qc2 Ra3 19. cxb5 Ra2 20. Qd1 Ne5 21. Bc3 Nd5 22. Bxe5 Bxe5 23.
Qc1 Bg7 24. d4 Qd7 25. Qc6 Qxc6 26. bxc6 Rb8 27. Rfc1 Rb5 28. Bf1 Rxb4 29.
Rxb4 Nxb4 30. Nc3 Ra5 31. Rb1 Nxc6 32. Nb5 Bf5 33. Rc1 Nb4 34. Nxc7 Bf8 35.
Ne8 Bc2 36. Nc7 Bf5 37. Ne8 Bc2 38. Nc7 Ra2 39. Bc4 Rb2 40. Nd5 Nxd5 41.
Bxd5 h5 42. Kf1 Be7 43. Ra1 Kg7 44. Ra7 Kf6 45. Kg1 Rb1+ 46. Kh2 g5 47. Ra5
Rb2 48. Kg1 Bg6 49. Bf3 h4 50. gxh4 gxh4 51. Kg2 Bf5 52. Bd5 Bg4 53. Be4
Rb8 54. Kh2 Bf8 55. Kg1 Rc8 56. Kh2 Rb8 57. Kg1 Rc8 58. Rb5 Bd7 59. Rh5 h3
60. Kh2 Rc3 61. Rh8 Bg7 62. Rd8 Ke7 63. Ra8 Bc8 64. Ra7+ Kf6 65. Bd5 Rc2
66. Kg3 Rc1 67. Rf7+ Kg6 68. Ra7 Bf5 69. Ra8 Bc8 70. Ra7 Bf8 71. Rf7 Bh6
72. Re7 Bf5 73. Re8 Rg1+ 74. Kh2 Rf1 75. Kg3 Rg1+ 76. Kh2 Rf1 77. Kg3 Kf6
78. Rh8 Bg7 79. Rd8 Ke7 80. Rb8 Rg1+ 81. Kh2 Rd1 82. Rb7+ Kf6 83. Rb8 Rd3
84. Kg3 Kg6 85. Rg8 Rc3 86. Bf3 Kf6 87. Bd5 Bd7 88. Kh2 Rc2 89. Kg3 Rc3 90.
Kh2 Rc2 91. Kg3 Rc1 92. Ra8 Bc8 93. Ra7 Bh6 94. Rf7+ Kg6 95. Re7 Bf5 96.
Re8 Kf6 97. Rd8 Ke7 98. Ra8 Rg1+ 99. Kh2 Rf1 100. Ra2 Kf6 101. Kg3 Rg1+
102. Kh2 Rf1 103. Kg3 Rg1+ 104. Kh2 Rd1 105. Kg3 Re1 106. Ra6 Rg1+ 107. Kh2
Rf1 108. Ra2 Be6 109. Bxe6 Kxe6 110. Kxh3 Rg1 111. Ra5 Rg5 112. Ra6 Kd5
113. e4+ Kxe4 114. Rxd6 Bg7 115. Rd7 Bxd4 116. Rf7 Be5 117. f3+ Ke3 118.
Kh4 Bf4 119. Ra7 Kxf3 120. Ra3+ Ke2 121. Rb3 Rg1 122. Ra3 Be3 123. Ra6 Rg7
124. Rd6 Bf4 125. Rd5 Rg1 126. Kh5 Ke3 127. Rf5 Rg7 128. Rd5 Ke4 129. Ra5
Rg2 130. Ra3 Kf5 131. Kh4 Rh2+ 132. Rh3 Rb2 133. Rf3 Rb6 134. Kh3 Rh6+ 135.
Kg2 Rh2+ 136. Kg1 Rh8 137. Rb3 Ke4 138. Kf1 Rh1+ 139. Ke2 Rh2+ 140. Ke1 Bd6
141. Kf1 Be5 142. Ra3 Bf4 143. Ke1 Bd6 144. Ra8 Bc5 145. Rd8 Bd4 146. Rf8
Kd3 147. Rf3+ Be3 148. Kf1 Ke4 149. Rg3 Rf2+ 150. Kg1 Ra2+ 151. Kf1 Kd3
152. Rh3 Rf2+ 153. Kg1 Rd2+ 154. Kh1 Ke4 155. Rh4+ Bf4 156. Rh3 Rb2 157.
Kg1 Be3+ 158. Kf1 Bd4 159. Ke1 Bf2+ 160. Kf1 Be3 161. Ke1 Bd4 162. Rh7 Kd3
163. Rh3+ Be3 164. Kf1 Rc2 165. Rg3 Rd2 166. Rh3 Rb2 167. Rg3 Rc2 168. Rh3
Rc7 169. Kg2 Rc4 {50 moves rule} *[/pgn]

[lkaufman]

I replayed this game from GM Perelshteyn Vs Komodo Dragon using the same Opening up to move 7 the 2nd game was using Contempt = 65 and this game with Contempt = 125 T/C 2'+2"

Note: We should take the score of GM Perelshteyn with a grain of salt, since in only 6 games it could have gone either way, and if the amount of games and choice of Openings instead of 3 with the b1 and 3 with the g1 Knight missing WAS 3 games with the b1 missing and 4 games with the g1 missing total of 7 games, Komodo Dragon could have been the winner of the match


[pgn][Event "knight odds match"]
[Site "Chess.com"]
[Date "2021.11.06"]
[Round "4"]
[White "PlayKomodo"]
[Black "Gaviota-win32-0.84"]
[Result "1/2 1/2"]
[BlackElo "2500"]
[WhiteElo "2500"]
[TimeControl "120+2"]
[SetUp "1"]
[FEN "rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKB1R w KQkq - 0 1"]
[Termination "normal"]
[PlyCount "338"]
[WhiteType "program"]
[BlackType "program"]

1. c4 {[%clk 0:15:10]} e5 {[%clk 0:15:06.4]} 2. Nc3 {[%clk 0:15:19.9]} Nc6
{[%clk 0:15:09.4]} 3. g3 {[%clk 0:15:29.8]} g6 {[%clk 0:15:11.1]} 4. Bg2
{[%clk 0:15:33.2]} Bg7 {[%clk 0:15:19.9]} 5. d3 {[%clk 0:15:36]} Nge7
{[%clk 0:15:24.1]} 6. Rb1 {[%clk 0:15:40.4]} a5 {[%clk 0:15:30.5]} 7. a3
{[%clk 0:15:50.1]} O-O {[%clk 0:15:24.5]} 8. e3 d6 9. h4 Be6 10. h5 h6 11.
hxg6 fxg6 12. b4 e4 13. Nxe4 axb4 14. axb4 Ra2 15. O-O b5 16. Qb3 Ra4 17.
Bd2 Kh7 18. Qc2 Ra3 19. cxb5 Ra2 20. Qd1 Ne5 21. Bc3 Nd5 22. Bxe5 Bxe5 23.
Qc1 Bg7 24. d4 Qd7 25. Qc6 Qxc6 26. bxc6 Rb8 27. Rfc1 Rb5 28. Bf1 Rxb4 29.
Rxb4 Nxb4 30. Nc3 Ra5 31. Rb1 Nxc6 32. Nb5 Bf5 33. Rc1 Nb4 34. Nxc7 Bf8 35.
Ne8 Bc2 36. Nc7 Bf5 37. Ne8 Bc2 38. Nc7 Ra2 39. Bc4 Rb2 40. Nd5 Nxd5 41.
Bxd5 h5 42. Kf1 Be7 43. Ra1 Kg7 44. Ra7 Kf6 45. Kg1 Rb1+ 46. Kh2 g5 47. Ra5
Rb2 48. Kg1 Bg6 49. Bf3 h4 50. gxh4 gxh4 51. Kg2 Bf5 52. Bd5 Bg4 53. Be4
Rb8 54. Kh2 Bf8 55. Kg1 Rc8 56. Kh2 Rb8 57. Kg1 Rc8 58. Rb5 Bd7 59. Rh5 h3
60. Kh2 Rc3 61. Rh8 Bg7 62. Rd8 Ke7 63. Ra8 Bc8 64. Ra7+ Kf6 65. Bd5 Rc2
66. Kg3 Rc1 67. Rf7+ Kg6 68. Ra7 Bf5 69. Ra8 Bc8 70. Ra7 Bf8 71. Rf7 Bh6
72. Re7 Bf5 73. Re8 Rg1+ 74. Kh2 Rf1 75. Kg3 Rg1+ 76. Kh2 Rf1 77. Kg3 Kf6
78. Rh8 Bg7 79. Rd8 Ke7 80. Rb8 Rg1+ 81. Kh2 Rd1 82. Rb7+ Kf6 83. Rb8 Rd3
84. Kg3 Kg6 85. Rg8 Rc3 86. Bf3 Kf6 87. Bd5 Bd7 88. Kh2 Rc2 89. Kg3 Rc3 90.
Kh2 Rc2 91. Kg3 Rc1 92. Ra8 Bc8 93. Ra7 Bh6 94. Rf7+ Kg6 95. Re7 Bf5 96.
Re8 Kf6 97. Rd8 Ke7 98. Ra8 Rg1+ 99. Kh2 Rf1 100. Ra2 Kf6 101. Kg3 Rg1+
102. Kh2 Rf1 103. Kg3 Rg1+ 104. Kh2 Rd1 105. Kg3 Re1 106. Ra6 Rg1+ 107. Kh2
Rf1 108. Ra2 Be6 109. Bxe6 Kxe6 110. Kxh3 Rg1 111. Ra5 Rg5 112. Ra6 Kd5
113. e4+ Kxe4 114. Rxd6 Bg7 115. Rd7 Bxd4 116. Rf7 Be5 117. f3+ Ke3 118.
Kh4 Bf4 119. Ra7 Kxf3 120. Ra3+ Ke2 121. Rb3 Rg1 122. Ra3 Be3 123. Ra6 Rg7
124. Rd6 Bf4 125. Rd5 Rg1 126. Kh5 Ke3 127. Rf5 Rg7 128. Rd5 Ke4 129. Ra5
Rg2 130. Ra3 Kf5 131. Kh4 Rh2+ 132. Rh3 Rb2 133. Rf3 Rb6 134. Kh3 Rh6+ 135.
Kg2 Rh2+ 136. Kg1 Rh8 137. Rb3 Ke4 138. Kf1 Rh1+ 139. Ke2 Rh2+ 140. Ke1 Bd6
141. Kf1 Be5 142. Ra3 Bf4 143. Ke1 Bd6 144. Ra8 Bc5 145. Rd8 Bd4 146. Rf8
Kd3 147. Rf3+ Be3 148. Kf1 Ke4 149. Rg3 Rf2+ 150. Kg1 Ra2+ 151. Kf1 Kd3
152. Rh3 Rf2+ 153. Kg1 Rd2+ 154. Kh1 Ke4 155. Rh4+ Bf4 156. Rh3 Rb2 157.
Kg1 Be3+ 158. Kf1 Bd4 159. Ke1 Bf2+ 160. Kf1 Be3 161. Ke1 Bd4 162. Rh7 Kd3
163. Rh3+ Be3 164. Kf1 Rc2 165. Rg3 Rd2 166. Rh3 Rb2 167. Rg3 Rc2 168. Rh3
Rc7 169. Kg2 Rc4 {50 moves rule} *[/pgn]

I confirmed your contention that the 125 value for Contempt was too high against Gaviota. I ran 1000 games each at 2' + 1" on one thread with latest Dragon giving knight odds to Gaviota0.86 64 bit, the first version that seems able to come out ahead of Dragon under these conditions, using four different Contempt values. With 125 I got -26.5 elo, with 100 I got -24 elo, with 80 I got -16 elo, and with 65 I got -20 elo. So Contempt 80 looks best, although margin of error is still large even with 1000 game samples. Of course results might be very different using 16 or 32 threads, but I suspect that thread count won't have a big effect on the proper Contempt setting. Maybe with lower Contempt, one more version update, opening book enhancements, and possibly choosing g1 odds more often than b1 odds, we might have equal chances vs. 2500 GM, but I suspect we'll need some new idea as well.

[Chessqueen]

I replayed this game from GM Perelshteyn Vs Komodo Dragon using the same Opening up to move 7 the 2nd game was using Contempt = 65 and this game with Contempt = 125 T/C 2'+2"

Note: We should take the score of GM Perelshteyn with a grain of salt, since in only 6 games it could have gone either way, and if the amount of games and choice of Openings instead of 3 with the b1 and 3 with the g1 Knight missing WAS 3 games with the b1 missing and 4 games with the g1 missing total of 7 games, Komodo Dragon could have been the winner of the match


[pgn][Event "knight odds match"]
[Site "Chess.com"]
[Date "2021.11.06"]
[Round "4"]
[White "PlayKomodo"]
[Black "Gaviota-win32-0.84"]
[Result "1/2 1/2"]
[BlackElo "2500"]
[WhiteElo "2500"]
[TimeControl "120+2"]
[SetUp "1"]
[FEN "rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKB1R w KQkq - 0 1"]
[Termination "normal"]
[PlyCount "338"]
[WhiteType "program"]
[BlackType "program"]

1. c4 {[%clk 0:15:10]} e5 {[%clk 0:15:06.4]} 2. Nc3 {[%clk 0:15:19.9]} Nc6
{[%clk 0:15:09.4]} 3. g3 {[%clk 0:15:29.8]} g6 {[%clk 0:15:11.1]} 4. Bg2
{[%clk 0:15:33.2]} Bg7 {[%clk 0:15:19.9]} 5. d3 {[%clk 0:15:36]} Nge7
{[%clk 0:15:24.1]} 6. Rb1 {[%clk 0:15:40.4]} a5 {[%clk 0:15:30.5]} 7. a3
{[%clk 0:15:50.1]} O-O {[%clk 0:15:24.5]} 8. e3 d6 9. h4 Be6 10. h5 h6 11.
hxg6 fxg6 12. b4 e4 13. Nxe4 axb4 14. axb4 Ra2 15. O-O b5 16. Qb3 Ra4 17.
Bd2 Kh7 18. Qc2 Ra3 19. cxb5 Ra2 20. Qd1 Ne5 21. Bc3 Nd5 22. Bxe5 Bxe5 23.
Qc1 Bg7 24. d4 Qd7 25. Qc6 Qxc6 26. bxc6 Rb8 27. Rfc1 Rb5 28. Bf1 Rxb4 29.
Rxb4 Nxb4 30. Nc3 Ra5 31. Rb1 Nxc6 32. Nb5 Bf5 33. Rc1 Nb4 34. Nxc7 Bf8 35.
Ne8 Bc2 36. Nc7 Bf5 37. Ne8 Bc2 38. Nc7 Ra2 39. Bc4 Rb2 40. Nd5 Nxd5 41.
Bxd5 h5 42. Kf1 Be7 43. Ra1 Kg7 44. Ra7 Kf6 45. Kg1 Rb1+ 46. Kh2 g5 47. Ra5
Rb2 48. Kg1 Bg6 49. Bf3 h4 50. gxh4 gxh4 51. Kg2 Bf5 52. Bd5 Bg4 53. Be4
Rb8 54. Kh2 Bf8 55. Kg1 Rc8 56. Kh2 Rb8 57. Kg1 Rc8 58. Rb5 Bd7 59. Rh5 h3
60. Kh2 Rc3 61. Rh8 Bg7 62. Rd8 Ke7 63. Ra8 Bc8 64. Ra7+ Kf6 65. Bd5 Rc2
66. Kg3 Rc1 67. Rf7+ Kg6 68. Ra7 Bf5 69. Ra8 Bc8 70. Ra7 Bf8 71. Rf7 Bh6
72. Re7 Bf5 73. Re8 Rg1+ 74. Kh2 Rf1 75. Kg3 Rg1+ 76. Kh2 Rf1 77. Kg3 Kf6
78. Rh8 Bg7 79. Rd8 Ke7 80. Rb8 Rg1+ 81. Kh2 Rd1 82. Rb7+ Kf6 83. Rb8 Rd3
84. Kg3 Kg6 85. Rg8 Rc3 86. Bf3 Kf6 87. Bd5 Bd7 88. Kh2 Rc2 89. Kg3 Rc3 90.
Kh2 Rc2 91. Kg3 Rc1 92. Ra8 Bc8 93. Ra7 Bh6 94. Rf7+ Kg6 95. Re7 Bf5 96.
Re8 Kf6 97. Rd8 Ke7 98. Ra8 Rg1+ 99. Kh2 Rf1 100. Ra2 Kf6 101. Kg3 Rg1+
102. Kh2 Rf1 103. Kg3 Rg1+ 104. Kh2 Rd1 105. Kg3 Re1 106. Ra6 Rg1+ 107. Kh2
Rf1 108. Ra2 Be6 109. Bxe6 Kxe6 110. Kxh3 Rg1 111. Ra5 Rg5 112. Ra6 Kd5
113. e4+ Kxe4 114. Rxd6 Bg7 115. Rd7 Bxd4 116. Rf7 Be5 117. f3+ Ke3 118.
Kh4 Bf4 119. Ra7 Kxf3 120. Ra3+ Ke2 121. Rb3 Rg1 122. Ra3 Be3 123. Ra6 Rg7
124. Rd6 Bf4 125. Rd5 Rg1 126. Kh5 Ke3 127. Rf5 Rg7 128. Rd5 Ke4 129. Ra5
Rg2 130. Ra3 Kf5 131. Kh4 Rh2+ 132. Rh3 Rb2 133. Rf3 Rb6 134. Kh3 Rh6+ 135.
Kg2 Rh2+ 136. Kg1 Rh8 137. Rb3 Ke4 138. Kf1 Rh1+ 139. Ke2 Rh2+ 140. Ke1 Bd6
141. Kf1 Be5 142. Ra3 Bf4 143. Ke1 Bd6 144. Ra8 Bc5 145. Rd8 Bd4 146. Rf8
Kd3 147. Rf3+ Be3 148. Kf1 Ke4 149. Rg3 Rf2+ 150. Kg1 Ra2+ 151. Kf1 Kd3
152. Rh3 Rf2+ 153. Kg1 Rd2+ 154. Kh1 Ke4 155. Rh4+ Bf4 156. Rh3 Rb2 157.
Kg1 Be3+ 158. Kf1 Bd4 159. Ke1 Bf2+ 160. Kf1 Be3 161. Ke1 Bd4 162. Rh7 Kd3
163. Rh3+ Be3 164. Kf1 Rc2 165. Rg3 Rd2 166. Rh3 Rb2 167. Rg3 Rc2 168. Rh3
Rc7 169. Kg2 Rc4 {50 moves rule} *[/pgn]

I confirmed your contention that the 125 value for Contempt was too high against Gaviota. I ran 1000 games each at 2' + 1" on one thread with latest Dragon giving knight odds to Gaviota0.86 64 bit, the first version that seems able to come out ahead of Dragon under these conditions, using four different Contempt values. With 125 I got -26.5 elo, with 100 I got -24 elo, with 80 I got -16 elo, and with 65 I got -20 elo. So Contempt 80 looks best, although margin of error is still large even with 1000 game samples. Of course results might be very different using 16 or 32 threads, but I suspect that thread count won't have a big effect on the proper Contempt setting. Maybe with lower Contempt, one more version update, opening book enhancements, and possibly choosing g1 odds more often than b1 odds, we might have equal chances vs. 2500 GM, but I suspect we'll need some new idea as well.

It never occurred to me that the best Contempt was in between 65 and 125, since you run 1000 games with each contempt 80 is a logical winner which is not to high and not to low

[Chessqueen]

I took this position which was from one of the game that GM Perelshteyn played vs Komodo Dragon after move 7 replaced it with Gaviota and set the contempt = 80 for this game and this is the result

[pgn][Event "Computer chess game"]
[Site "MININT-UB2PIMJ"]
[Date "2021.11.11"]
[Round "?"]
[White "Dragon-2.5-64bit-avx2"]
[Black "Gaviota-0.86-win64"]
[Result "0-1"]
[BlackElo "2500"]
[Time "23:43:14"]
[WhiteElo "2500"]
[TimeControl "120+2"]
[SetUp "1"]
[FEN "r2qk2r/pp2ppbp/2n2np1/2pp4/5Pb1/3PPN2/PPP1B1PP/R1B1QRK1 w kq - 1 1"]
[Termination "adjudication"]
[PlyCount "120"]
[WhiteType "program"]
[BlackType "program"]

1. Bd1 O-O 2. c3 Bf5 3. Be2 Re8 4. d4 Qb6 5. a4 c4 6. Qd1 Be4 7. h3 Na5 8.
g4 Nb3 9. Ra3 Bf8 10. Nd2 e6 11. Rxb3 cxb3 12. Nxb3 h6 13. Nd2 Qc7 14. Bb5
Red8 15. Nxe4 Nxe4 16. Kg2 Be7 17. Bd2 Kh7 18. Be1 Qb6 19. Qe2 a6 20. Bd3
f5 21. h4 fxg4 22. Qxg4 Qxb2+ 23. Kh1 Qb6 24. a5 h5 25. Qh3 Qc6 26. Qg2 Rg8
27. c4 Kh6 28. Rg1 Kh7 29. Qc2 Rab8 30. Kg2 Nd6 31. c5 Nf5 32. Kh2 Bf6 33.
Qe2 Qa4 34. Rg2 Bd8 35. Bf2 Qxa5 36. Qf3 Rf8 37. Kh3 Be7 38. e4 dxe4 39.
Bxe4 Qa2 40. Rg1 Rfd8 41. Bd3 a5 42. Qg2 Rg8 43. Qf3 Rgd8 44. Bb1 Qb2 45.
Be4 Bf6 46. Qg2 Qc3+ 47. Kh2 Rg8 48. Rd1 Bxh4 49. Bg1 Bg3+ 50. Kh3 Bxf4+
51. Bd3 Be3 52. Bxe3 Nxe3 53. Qe2 Nf5 54. Qe1 Qxe1 55. Rxe1 Nxd4 56. Be4
Rbc8 57. Rc1 Nb3 58. Re1 Nxc5 59. Bc2 b5 60. Rb1 b4 {User Adjudication} 0-1[/pgn]

[Chessqueen]

I took this position which was from one of the game that GM Perelshteyn played vs Komodo Dragon after move 7 replaced it with Gaviota and set the contempt = 85 for this game and this is the result, I will have to replay all the other games that Human Gms played vs Komodo Dragon at Knight Odds and set up the position after the 7 move and test it with contempt = 80 and contempt = 85 at T/C 2'+2" and see which contempt perform better

[pgn][Event "Computer chess game"]
[Site "MININT-UB2PIMJ"]
[Date "2021.11.11"]
[Round "?"]
[White "Dragon-2.5-64bit-avx2"]
[Black "Gaviota-0.86-win64"]
[Result "1-0"]
[BlackElo "2500"]
[Time "23:24:37"]
[WhiteElo "2500"]
[TimeControl "120+2"]
[SetUp "1"]
[FEN "r2qk2r/pp2ppbp/2n2np1/2pp4/5Pb1/3PPN2/PPP1B1PP/R1B1QRK1 w kq - 1 1"]
[Termination "normal"]
[PlyCount "71"]
[WhiteType "program"]
[BlackType "program"]

1. h3 Bd7 2. Bd1 Be6 3. Kh1 O-O 4. e4 dxe4 5. dxe4 Bc4 6. Rg1 e5 7. f5 gxf5
8. exf5 Re8 9. Bg5 Ba6 10. Nh2 h6 11. Bh4 Qb6 12. Ng4 Nxg4 13. Bxg4 Nd4 14.
Qg3 e4 15. Rae1 Be5 16. Qf2 Qxb2 17. Qd2 Bg7 18. f6 Bf8 19. Qf4 Qxc2 20.
Re3 Kh8 21. a4 Qxa4 22. Bg5 Bd6 23. Qxd6 hxg5 24. Qd5 Rg8 25. h4 Bc4 26.
Rh3 Rg6 27. hxg5+ Kg8 28. Qe5 Ne2 29. Bxe2 Be6 30. Rh4 Rh6 31. Rxh6 Bh3 32.
Qh2 Qa3 33. Rxh3 Qxh3 34. Qxh3 Rc8 35. Qh6 b5 36. Qg7# 1-0[/pgn]

[Chessqueen]

I took this position which was from one of the game that GM Perelshteyn played vs Komodo Dragon after move 7 replaced it with Gaviota and set the contempt = 85 for this game and this is the result, I will have to replay all the other games that Human Gms played vs Komodo Dragon at Knight Odds and set up the position after the 7 move and test it with contempt = 80 and contempt = 85 at T/C 2'+2" and see which contempt perform better. For some reason I believe that Contempt = 85 performs better than 80 against Gaviota-0.86-win64; by at least 2 Elo. I already have done several testing using contempt = 80 and 85 against most of the games that Komodo played against Humans GMs at Knight Odds; but of course we need to play at least 1000 games with these 2 contempt settings.

[pgn][Event "Computer chess game"]
[Site "MININT-UB2PIMJ"]
[Date "2021.11.11"]
[Round "?"]
[White "Dragon-2.5-64bit-avx2"]
[Black "Gaviota-0.86-win64"]
[Result "1-0"]
[BlackElo "2500"]
[Time "23:24:37"]
[WhiteElo "2500"]
[TimeControl "120+2"]
[SetUp "1"]
[FEN "r2qk2r/pp2ppbp/2n2np1/2pp4/5Pb1/3PPN2/PPP1B1PP/R1B1QRK1 w kq - 1 1"]
[Termination "normal"]
[PlyCount "71"]
[WhiteType "program"]
[BlackType "program"]

1. h3 Bd7 2. Bd1 Be6 3. Kh1 O-O 4. e4 dxe4 5. dxe4 Bc4 6. Rg1 e5 7. f5 gxf5
8. exf5 Re8 9. Bg5 Ba6 10. Nh2 h6 11. Bh4 Qb6 12. Ng4 Nxg4 13. Bxg4 Nd4 14.
Qg3 e4 15. Rae1 Be5 16. Qf2 Qxb2 17. Qd2 Bg7 18. f6 Bf8 19. Qf4 Qxc2 20.
Re3 Kh8 21. a4 Qxa4 22. Bg5 Bd6 23. Qxd6 hxg5 24. Qd5 Rg8 25. h4 Bc4 26.
Rh3 Rg6 27. hxg5+ Kg8 28. Qe5 Ne2 29. Bxe2 Be6 30. Rh4 Rh6 31. Rxh6 Bh3 32.
Qh2 Qa3 33. Rxh3 Qxh3 34. Qxh3 Rc8 35. Qh6 b5 36. Qg7# 1-0[/pgn]

[Chessqueen]

I already finished testing 1000 games with each contempt = 80 and 85 against Gaviota-0.86-win64, and contempt 85 is only slightly better by 2%