Komodo - Rybka in Danger?

[Arturo Ochoa]

Sorry forgot the link. Here it is:
http://rybkaforum.net/cgi-bin/rybkaforu ... ?tid=15105

Besides of being forced to read this forum, since I still work with one engine, The Baron, as a booker and I must be up-to-date of the new coming, I hope that Komodo's Authors keep some savviness and canniness.

Any "melange" with that boring, nasty, obsessive and excessive subject that frighten any human in normal mind status, would be another infamous momentous of the last months in this forum.

Welcome to the new real star.... (if it is true)

/A8A.

[Uri Blass]

<snipped>

The Elo increase depends on time limit (almost always, the faster the rate of play, the greater the increase due mostly to less draws).

I think that it is dependent on the change that you do

I read that one of the improvement is better move ordering and I think that I can expect it to help more in longer time control.

Uri

[Uri Blass]

I'm looking forward to taking a new exciting legit engine for a spin.
In your tests, what's the ELO increase vs. Doch?
Another question: do you explicitly try 'give it a playing style' or is it more about tuning eval/search, and whatever gets the highest elo you're happy with?

We try to make the engine play as much in a human style as possible, but currently strength trumps style. When it's a close call we try to make it evaluate more like a human grandmaster would.

If you try the engine to play as much in a human style then it may be good if you change the evaluation to evaluate branches and not leaves(I think that it is a good idea in many games and not only in chess).

The idea is that if I calculate some line and see that the position becomes better for me then it should increase the evaluation and not everything is static knowledge.


I will give an example.
When you calculate a line you can evaluate every node in the branch.


If you calculate in some line something like
1.a4 evaluation 0.20 for black
1...Rb8 evaluation 0.20 for black
2.a5 evaluation 0.10 for black
2...Ra8 evaluation 0.09 for black
3.b4 evaluation 0.00

Then the final evaluation should be better than 0.00 because white improved the position.

practically you have an array of number of static evaluations from white point of view.

eval[0]=-0.20(evaluation of the root)
eval[1]=-0.20(evaluation of the position 1 ply after the root)
eval[2[=-0.10(evaluation of the position 2 plies after the root)
eval[3]=-0.09(evaluation of the position 3 plies after the root)
eval[4]=0.00(evaluation of the position 4 plies after the root)

I think that you should have a function that translate this array to a number that is bigger than 0.00 for white.

Humans do not think in terms of exact numbers but they can think that some idea is good because they can improve the position based on their calculation.

Uri

[Leto]

two games between Komodo 64 vs Naum x64 1CPU, 40/4:

[Event "Komodo testing 1"]
[Site "Microsoft"]
[Date "2010.01.21"]
[Round "1.1"]
[White "Komodo64 1.0 JA"]
[Black "Naum 4 x64 1CPU"]
[Result "1/2-1/2"]
[ECO "A25"]
[Annotator "0.29;0.09"]
[PlyCount "221"]
[EventDate "2010.01.21"]
[EventType "tourn"]
[TimeControl "40/180:0/0:0/0"]

{Intel(R) Core(TM)2 Quad CPU Q9650 @ 3.00GHzW=21.8 ply; 2,367kN/s; Perfect
v8.32.ctg B=20.6 ply; 1,485kN/s; Perfect v8.32.ctg} 1. c4 {B/0 0} e5 {B/0 0}
2. g3 {B/0 0} Nc6 {B/0 0} 3. Bg2 {B/0 0} Nf6 {B/0 0} 4. Nc3 {B/0 0} Bc5 {B/0 0}
5. a3 {B/0 0} a6 {B/0 0} 6. b4 {B/0 0} Ba7 {B/0 0} 7. Bb2 {B/0 0} O-O {B/0 0}
8. e3 {B/0 0} d6 {B/0 0} 9. Nge2 {B/0 0} Re8 {B/0 0} 10. Qc2 {0.29/14 5} Bg4 {
(Ne7) 0.09/15 4} 11. h3 {(Nd5) 0.47/15 4} Bh5 {(Be6) 0.08/15 5} 12. d3 {(0-0)
0.42/16 3} Qd7 {(Bg6) 0.11/16 11} 13. O-O {(g4) 0.36/14 5} h6 {(Bg6) 0.16/15 12
} 14. g4 {(Rac1) 0.38/13 4} Bg6 {0.08/16 6} 15. Rac1 {(Ng3) 0.33/15 8} Bb6 {
(Rab8) 0.04/15 8} 16. Ng3 {0.47/15 8} Rab8 {(Rad8) 0.09/14 11} 17. Bf3 {(Nce4)
0.40/14 3} a5 {(Ne7) 0.00/16 10} 18. b5 {0.35/16 5} Nd8 {0.00/17 2} 19. Na4 {
0.46/16 6} Ne6 {0.07/16 3} 20. Rfd1 {(Nxb6) 0.40/16 20} Nc5 {0.03/16 10} 21.
Nxc5 {(Ne4) 0.40/17 4} Bxc5 {0.03/17 13} 22. Qd2 {0.45/17 7} Bb6 {(a4) 0.03/16
7} 23. Bc3 {(a4) 0.41/15 4} Bc5 {(a4) 0.19/15 5} 24. a4 {(Bxa5) 0.72/16 4} e4 {
(c6) 0.17/14 5} 25. dxe4 {(Be2) 0.49/16 4} Nxe4 {0.08/15 3} 26. Bxe4 {0.67/17 4
} Bxe4 {0.09/16 2} 27. Qb2 {(Nxe4) 0.54/17 7} f5 {(f6) 0.00/13 2} 28. gxf5 {0.
38/15 4} Bxf5 {0.00/14 3} 29. Bxg7 {(Nxf5) 0.32/16 3} Bxh3 {0.19/13 2} 30. Bxh6
{(Bd4) 0.30/16 3} Re6 {0.00/15 2} 31. Bf4 {0.30/16 3} Rg6 {0.00/16 3} 32. Rd5 {
0.28/16 12} Bf5 {(Qf7) 0.00/16 5} 33. Qd2 {(Kf1) 0.14/14 6} Ba3 {(Bb4) -0.18/
15 4} 34. Rd1 {(Ra1) 0.00/15 7} Be4 {-0.27/14 4} 35. Rh5 {0.20/14 3} Bf3 {-0.
14/15 6} 36. Rh4 {(Rh2) 0.13/15 3} Qg7 {-0.22/15 5} 37. Rb1 {(Qd4) 0.13/15 3}
Qf6 {(Be4) -0.26/15 7} 38. Qd3 {(Rh2) 0.22/15 3} Re8 {-0.28/15 3} 39. Bh6 {
(Kf1) 0.19/15 3} Bg4 {-0.26/14 4} 40. Kf1 {(Qd5+) 0.16/15 4} Bc5 {(Bb4) -0.33/
16 13} 41. Bf4 {0.16/16 4} Re6 {(Bf3) -0.33/16 4} 42. Rh2 {0.19/17 4} Bf3 {-0.
34/16 2} 43. Re1 {(Rh3) 0.14/16 3} Bb4 {-0.33/15 5} 44. Rc1 {(Rb1) 0.14/17 4}
b6 {(Qb2) -0.23/15 6} 45. Rb1 {0.20/16 7} Qg7 {(Re8) -0.24/15 6} 46. Qb3 {(Rc1)
0.14/16 3} Qf7 {(Qf6) -0.33/15 3} 47. Qb2 {(Qd3) 0.06/17 5} Rg7 {(Qf6) -0.48/
16 3} 48. Qc2 {0.00/18 3} Reg6 {(Qg6) -0.48/16 8} 49. Rc1 {-0.46/16 6} Qe6 {-0.
46/16 5} 50. Rh4 {-0.25/18 7} Ba3 {-0.41/16 5} 51. Re1 {-0.31/19 5} Rh7 {-0.54/
16 7} 52. Rxh7 {-0.27/14 0} Kxh7 {-0.54/6 0} 53. Qf5 {-0.27/17 3} Qxc4+ {-0.51/
16 2} 54. Kg1 {-0.35/16 5} Qe6 {(Bb2) -0.43/17 9} 55. Qxe6 {-0.45/18 5} Rxe6 {
-0.43/6 0} 56. Nf5 {-0.44/17 4} Bc5 {-0.51/17 4} 57. Kh2 {-0.33/17 3} Be4 {-0.
49/17 2} 58. Ng3 {(Nd4) -0.46/18 4} Bd5 {(Bc2) -0.46/16 4} 59. Ra1 {(Kh3) -0.
46/16 3} Kg6 {-0.49/17 5} 60. Kh3 {(Ne2) -0.47/18 4} Re8 {-0.56/16 3} 61. Kg4 {
-0.53/18 6} Rf8 {(Bb3) -0.57/16 2} 62. e4 {-0.47/17 5} Be6+ {-0.72/16 7} 63.
Kf3 {-0.66/20 4} Bxf2 {-0.56/16 2} 64. Rh1 {-0.65/20 4} Bc5 {(Bxg3) -0.50/16 6}
65. Rh6+ {-0.62/19 3} Kf7 {-0.45/18 2} 66. Rh7+ {-0.67/18 6} Ke8 {-0.47/19 6}
67. Rxc7 {-0.71/21 4} Rf7 {-0.60/19 10} 68. Rxf7 {-0.70/22 4} Bxf7 {-0.47/20 6}
69. e5 {-0.80/21 4} d5 {(Bb3) -0.43/20 6} 70. Nf5 {-0.60/19 8} Bh5+ {(Bg6) -0.
72/19 3} 71. Kg3 {-0.76/21 5} Bd1 {(Kf7) -0.72/20 3} 72. Be3 {-0.61/21 4} Bxe3
{-0.92/21 4} 73. Nxe3 {-0.65/19 1} Bb3 {-0.58/21 4} 74. Kf4 {(Kf3) -0.64/24 4}
Kd7 {-0.92/23 5} 75. Nf5 {-0.56/25 4} Bxa4 {-0.92/23 2} 76. Nd4 {-0.55/25 4}
Bd1 {-0.92/24 3} 77. Ke3 {-0.54/26 4} Bh5 {(Bg4) -0.92/25 5} 78. Kd3 {(Kd2) -0.
54/26 8} Bg6+ {(Ke7) -0.92/26 10} 79. Kc3 {-0.55/27 4} Be4 {(Be8) -0.92/25 5}
80. Kb3 {(Kb2) -1.00/27 9} Bb1 {(Bd3) -0.92/26 4} 81. Kb2 {(Kc3) -0.65/27 4}
Bd3 {(Bh7) -0.92/27 3} 82. Kc3 {-0.96/26 4} Bg6 {-0.92/26 4} 83. Kb3 {(Kd2) -1.
19/26 7} Be4 {(Ke7) -0.92/27 13} 84. Kc3 {(Kb2) -1.20/26 5} Ke7 {-0.92/26 3}
85. Kb3 {-1.22/24 3} Bg6 {-0.92/26 2} 86. Kc3 {-0.94/26 4} Be8 {(Bb1) -0.92/25
2} 87. Kb3 {(Nf5+) -1.20/24 3} Bh5 {(Bd7) -0.92/25 3} 88. Nf5+ {(Kc2) -1.11/23
3} Kd8 {(Kd7) -0.92/24 3} 89. Kc3 {-1.20/24 4} Bg6 {(Kd7) -0.92/24 2} 90. Nd4 {
-1.28/25 5} Be4 {(Ke7) -0.92/26 5} 91. Nc6+ {(Ne2) -1.19/25 3} Kc7 {-0.92/26 4}
92. Nd4 {-1.19/26 5} Bg2 {-0.92/26 4} 93. Nf5 {(Ne2) -1.23/25 4} Kd7 {(Kd8) -0.
92/26 4} 94. Nd4 {-1.37/27 11} Bf1 {-0.92/27 4} 95. Kb3 {(Kb2) -1.15/26 5} Bc4+
{(Ke7) -0.92/28 2} 96. Ka4 {(Kc3) -1.20/26 4} Bd3 {(Ke7) -0.92/28 2} 97. Ka3 {
(Kb3) -1.21/24 3} Ke7 {(Ke8) -0.92/28 3} 98. Kb3 {(Kb2) -1.25/26 3} Be4 {-0.92/
27 4} 99. Kc3 {-1.30/27 3} Kf7 {-0.92/26 2} 100. Kb3 {(Kd2) -1.63/22 4} Bg6 {
(Kg6) -0.92/27 4} 101. Kc3 {-1.21/21 4} Ke7 {-0.92/27 10} 102. Ne2 {-1.22/24 3}
Bb1 {(Bh5) -0.92/26 4} 103. Kd4 {(Nd4) -0.45/25 4} Ba2 {(a4) -1.23/23 2} 104.
Nc3 {-0.05/28 4} Bb3 {-1.23/26 3} 105. Nxd5+ {-0.05/30 5} Bxd5 {-1.23/29 3}
106. Kxd5 {-0.05/30 1} a4 {(Kd8) -1.23/30 3} 107. Kc4 {(Kd4) 0.00/38 3} Kd7 {
(Ke6) -1.23/33 4} 108. Kb4 {0.00/41 4} Ke6 {-1.12/40 3} 109. Kxa4 {0.00/86 4}
Kxe5 {0.00/0 1} 110. Kb3 {0.00/0 0} Kd5 {0.00/0 0} 111. Kb4 {0.00/0 0} 1/2-1/2


[Event "Komodo testing 1"]
[Site "Microsoft"]
[Date "2010.01.21"]
[Round "2.1"]
[White "Naum 4 x64 1CPU"]
[Black "Komodo64 1.0 JA"]
[Result "1/2-1/2"]
[ECO "B80"]
[Annotator "0.28;0.29"]
[PlyCount "341"]
[EventDate "2010.01.21"]
[EventType "tourn"]
[TimeControl "40/180:0/0:0/0"]

{Intel(R) Core(TM)2 Quad CPU Q9650 @ 3.00GHzW=24.8 ply; 1,668kN/s; Perfect
v8.32.ctg B=26.1 ply; 2,689kN/s; Perfect v8.32.ctg} 1. Nf3 {B/0 0} c5 {B/0 0}
2. e4 {B/0 0} Nc6 {B/0 0} 3. d4 {B/0 0} cxd4 {B/0 0} 4. Nxd4 {B/0 0} e6 {B/0 0}
5. Nc3 {B/0 0} Qc7 {B/0 0} 6. g3 {B/0 0} a6 {B/0 0} 7. Bg2 {B/0 0} Nf6 {B/0 0}
8. O-O {B/0 0} d6 {B/0 0} 9. Re1 {B/0 0} Rb8 {B/0 0} 10. Nxc6 {B/0 0} bxc6 {B/
0 0} 11. Qd3 {0.28/13 6} Be7 {0.29/15 6} 12. a3 {(e5) 0.19/13 6} O-O {-0.03/15
4} 13. Rb1 {(b3) 0.08/15 10} e5 {(a5) -0.11/16 6} 14. b4 {(b3) 0.06/15 6} h6 {
(Be6) -0.16/15 6} 15. Bd2 {(h3) 0.06/15 6} Rd8 {-0.18/15 7} 16. Qe2 {(Qc4) 0.
00/15 5} Be6 {-0.14/15 4} 17. Bf3 {0.00/16 9} Re8 {(Qb6) -0.21/16 15} 18. Kg2 {
(Qxa6) 0.00/15 6} Rbc8 {(Ra8) -0.29/14 3} 19. Rb2 {(Qxa6) 0.00/14 5} Qa7 {(Rb8)
-0.25/14 5} 20. Be3 {(Qe3) 0.00/13 5} Qb7 {-0.27/15 7} 21. Bd2 {0.00/13 6} Rb8
{(Qa7) -0.24/16 19} 22. Rbb1 {0.00/15 8} Qa7 {(Qb6) -0.19/14 3} 23. Qd3 {(Be3)
0.00/15 6} Red8 {-0.25/13 3} 24. Qe2 {0.00/14 8} Rd7 {(Qd4) -0.29/14 4} 25. Na4
{(Qe3) 0.02/15 8} Rdb7 {(Rdd8) -0.19/14 6} 26. Ra1 {(Nc3) 0.02/14 5} Rc7 {(Rd7)
-0.11/13 5} 27. c4 {(Rab1) 0.13/14 6} Rcc8 {(c5) -0.07/14 5} 28. c5 {(Rab1) 0.
12/14 7} dxc5 {-0.02/16 4} 29. Be3 {0.10/15 3} Nd7 {0.05/16 3} 30. Bg4 {0.16/
16 4} Bxg4 {0.16/16 6} 31. Qxg4 {0.16/6 0} Qc7 {(g6) 0.17/16 4} 32. Bxh6 {0.33/
16 7} g6 {0.00/17 3} 33. Red1 {0.33/17 4} Nf6 {0.09/17 3} 34. Qf3 {(Qh4) 0.35/
17 6} cxb4 {0.19/17 4} 35. Nc5 {0.36/17 10} Qa7 {(Bxc5) 0.47/17 4} 36. axb4 {
0.36/16 3} Rxb4 {0.42/18 4} 37. Nxa6 {0.37/17 5} Rd4 {0.52/18 4} 38. Rdc1 {0.
41/17 8} Qd7 {0.54/17 3} 39. Nc5 {0.41/17 6} Qg4 {0.59/18 9} 40. Ra7 {0.44/18 6
} Qxf3+ {0.57/16 3} 41. Kxf3 {0.44/6 0} Bxc5 {0.55/18 4} 42. Rxc5 {0.46/18 6}
Nxe4 {0.60/18 4} 43. Rxe5 {0.46/18 2} Nd6 {0.58/17 6} 44. Be3 {(Kg2) 0.44/17 4}
Rd3 {0.58/17 3} 45. Ke2 {0.43/16 2} Rb3 {0.66/18 7} 46. Ree7 {0.49/16 4} Rf8 {
0.68/17 5} 47. Red7 {(Bc5) 0.46/16 2} Nb5 {0.68/16 9} 48. Ra6 {0.51/17 7} Rc8 {
0.72/17 3} 49. h4 {(Kf3) 0.48/17 8} Kg7 {(Kf8) 0.60/15 3} 50. Kf1 {(Kf3) 0.65/
16 3} Kf6 {(c5) 0.91/16 4} 51. Bg5+ {0.74/16 3} Kg7 {0.99/18 4} 52. Rb6 {0.77/
18 4} Rb1+ {0.96/18 4} 53. Kg2 {0.78/19 2} Nc3 {1.11/20 4} 54. Rxb1 {0.77/19 3}
Nxb1 {1.02/17 0} 55. Bc1 {(Rd3) 0.78/19 3} Kg8 {1.14/18 3} 56. Rd3 {0.90/18 2}
Rb8 {1.19/19 4} 57. Be3 {0.89/19 4} Kg7 {1.13/20 18} 58. Bf4 {(Kf3) 0.91/19 8}
Rb5 {1.11/18 4} 59. Rd6 {0.90/19 5} Nc3 {1.13/18 3} 60. Rxc6 {0.88/19 5} Nd5 {
1.11/18 6} 61. Bd6 {0.95/19 7} f5 {(f6) 1.12/18 4} 62. Be5+ {0.98/18 4} Kf7 {
1.10/19 4} 63. Kh3 {(Kf3) 0.97/19 7} Rb6 {(Rb1) 1.05/19 5} 64. Rxb6 {(Rc1) 1.
03/20 2} Nxb6 {1.10/20 0} 65. Bc3 {(Bd4) 1.03/22 2} Ke6 {1.05/21 3} 66. Kg2 {
1.03/23 3} Nd5 {(Na4) 1.20/24 4} 67. Bd4 {1.03/23 4} Kd6 {(Ne7) 1.19/22 5} 68.
f3 {(Kf3) 1.03/24 7} Kc6 {(Ke6) 1.27/21 5} 69. Kf2 {1.03/22 3} Nb4 {(Kd6) 1.27/
22 3} 70. g4 {(Ke3) 1.16/21 4} fxg4 {1.28/24 4} 71. fxg4 {1.16/23 4} Kd5 {1.37/
24 3} 72. Bc3 {1.15/24 3} Nc6 {1.33/26 4} 73. Kf3 {(Ke3) 1.04/25 5} Ke6 {(Nd8)
1.14/25 3} 74. Kf4 {(Ke4) 1.15/25 4} Nd8 {(Kf7) 1.14/27 4} 75. Ke4 {(Bd2) 1.15/
26 5} Nf7 {(Kf7) 1.15/28 3} 76. Bb2 {(Bg7) 1.15/27 6} Kd6 {(Nd8) 1.16/28 5} 77.
Kf4 {(Bf6) 1.15/26 4} Kd5 {(Ke6) 1.14/28 5} 78. Bc3 {(Bg7) 1.15/27 6} Kd6 {
(Ke6) 1.14/27 6} 79. Ba5 {(Bd4) 1.15/26 4} Ke6 {1.19/26 4} 80. Bc7 {1.15/28 6}
Kd5 {(Kf6) 1.15/27 4} 81. Kf3 {(Bb6) 1.15/26 3} Kc6 {1.19/25 3} 82. Bb8 {(Bg3)
1.15/26 2} Kd5 {1.16/27 3} 83. Kf4 {(Bg3) 1.15/27 3} Ke6 {1.14/27 4} 84. Ba7 {
(Ke4) 1.15/27 2} Kd5 {(Nd8) 1.18/27 3} 85. Bf2 {(Be3) 1.15/26 3} Nd6 {(Ke6) 1.
14/25 8} 86. Bg1 {1.15/25 4} Ke6 {(Nf7) 1.16/26 3} 87. Bd4 {(Kg5) 1.15/25 2}
Nf7 {(Kf7) 1.15/28 4} 88. Ba1 {(Bg7) 1.15/27 3} Nd8 {1.15/29 4} 89. Bc3 {(Bb2)
1.15/27 3} Nf7 {(Kf7) 1.16/28 6} 90. Bd4 {1.15/28 5} Nd8 {1.15/27 3} 91. Ba1 {
(Bb2) 1.15/27 2} Nf7 {(Kf7) 1.14/27 4} 92. Bg7 {1.15/29 4} Nd8 {1.14/27 5} 93.
Ke4 {(Be5) 1.15/28 4} Nf7 {(Kf7) 1.26/24 3} 94. Bd4 {(Kd4) 1.15/27 2} Ke7 {
(Nd6+) 1.21/26 3} 95. Ba1 {(Kf3) 1.15/25 4} Ke6 {1.16/27 4} 96. Bc3 {(Bg7) 1.
15/27 3} Nd6+ {1.16/26 6} 97. Kf3 {(Ke3) 1.15/27 2} Kd5 {(Nb7) 1.20/23 4} 98.
Kf4 {(Bg7) 1.15/25 2} Ke6 {(Nf7) 1.22/26 8} 99. Be1 {(Kg5) 1.15/27 6} Nf7 {
(Nb5) 1.14/24 3} 100. Ba5 {(Bb4) 1.15/27 4} Ke7 {(Ne5) 1.15/26 4} 101. Ke4 {
(Bb4+) 1.15/27 5} Ke6 {1.12/27 6} 102. Bc7 {1.15/27 2} Nh6 {(Kf6) 1.05/25 4}
103. Kf3 {(Kf4) 1.15/27 2} Nf7 {1.12/27 4} 104. Bg3 {(Kf4) 1.15/28 4} Ke7 {
(Kd5) 1.12/26 4} 105. Bh2 {(Kf4) 1.15/28 4} Ke6 {(Kf6) 1.17/28 4} 106. Bc7 {1.
15/28 4} Kd5 {(Nh6) 1.15/26 4} 107. Ba5 {(Bf4) 1.15/28 7} Ne5+ {(Nd6) 1.17/26 4
} 108. Kf4 {1.15/28 6} Ke6 {1.19/26 4} 109. Bc7 {(Kg3) 1.15/28 6} Nf7 {(Nd3+)
1.16/28 4} 110. Bb8 {(Bb6) 1.15/28 3} Kd5 {(Kf6) 1.14/29 5} 111. Ba7 {(Ke3) 1.
04/28 13} Ke6 {(Ne5) 1.17/28 4} 112. Bf2 {(Bd4) 1.15/28 10} Nd6 {(Ne5) 1.16/26
5} 113. Bc5 {(Bd4) 1.15/27 3} Nc4 {1.18/28 4} 114. Bd4 {(Kg5) 1.15/26 4} Nd6 {
1.18/28 4} 115. Bb2 {(Kg5) 0.96/26 17} Nc4 {(Kf7) 1.17/27 4} 116. Bg7 {0.95/26
7} Nd6 {(Kf7) 1.17/28 4} 117. Kg5 {(Bd4) 0.92/25 8} Kf7 {1.16/29 4} 118. Bd4 {
(Be5) 0.86/26 5} Nb5 {(Ne4+) 1.15/27 5} 119. Be5 {(Bf2) 0.86/27 6} Ke6 {(Na3)
1.16/29 4} 120. Kf4 {(Bb8) 0.86/27 8} Na3 {1.15/27 6} 121. g5 {(Bc3) 0.86/28 4}
Nc4 {0.66/26 6} 122. Bd4 {0.86/27 2} Nd6 {0.67/30 5} 123. Bc3 {0.86/27 3} Nf5 {
0.71/27 3} 124. Be1 {0.86/27 2} Ng7 {(Ne7) 0.68/30 4} 125. Ke4 {0.85/26 3} Kd6
{(Nf5) 0.64/28 4} 126. Bf2 {0.85/26 5} Nh5 {(Ke6) 0.65/29 4} 127. Be3 {(Bd4) 0.
85/27 3} Ng3+ {(Ke6) 0.63/28 4} 128. Kf3 {0.85/27 6} Nh5 {(Nf5) 0.61/29 4} 129.
Bf4+ {0.85/29 5} Kd5 {0.63/30 3} 130. Bh2 {(Bc7) 0.85/29 3} Ng7 {0.46/31 11}
131. Bg3 {(Kg4) 0.85/29 4} Ne6 {(Nf5) 0.46/31 5} 132. Ke3 {(Kg4) 0.85/27 4} Nd4
{0.46/31 4} 133. Kf4 {0.85/28 3} Ne6+ {(Ne2+) 0.46/31 4} 134. Kf3 {(Kg4) 0.75/
28 4} Kd4 {(Nd4+) 0.46/31 6} 135. Bh2 {(Bf2+) 0.75/28 4} Kd5 {0.46/31 4} 136.
Ke3 {(Bb8) 0.75/28 3} Nd4 {0.46/31 3} 137. Kf4 {0.75/28 2} Ne6+ {0.46/29 5}
138. Kg4 {0.75/30 4} Kd4 {(Ke4) 0.60/28 19} 139. Bf4 {(Bd6) 0.75/26 2} Ng7 {
(Ke4) 0.46/30 3} 140. Bc7 {(Bg3) 0.75/28 4} Ne6 {(Ke4) 0.65/30 5} 141. Bb6+ {
(Bg3) 0.75/28 4} Ke4 {(Ke5) 0.47/30 3} 142. Bg1 {(Bf2) 0.75/29 3} Nf4 {(Nd8) 0.
46/32 4} 143. Bc5 {0.75/28 5} Ne6 {(Ng2) 0.46/31 4} 144. Bb6 {(Bd6) 0.75/29 2}
Nf4 {(Ng7) 0.46/33 4} 145. Bf2 {(Ba7) 0.75/29 3} Ne6 {0.46/32 4} 146. Bg3 {
(Be1) 0.75/30 5} Ng7 {(Kd5) 0.46/29 3} 147. Be1 {0.75/30 6} Nh5 {(Nf5) 0.00/31
3} 148. Bd2 {(Bb4) 0.75/31 6} Ng7 {0.00/33 3} 149. Bb4 {(Ba5) 0.75/31 5} Nh5 {
(Nf5) 0.00/35 6} 150. Bf8 {(Bc3) 0.75/31 3} Kd4 {(Nf4) 0.00/32 4} 151. Bd6 {
(Ba3) 0.75/32 5} Ke4 {0.00/36 5} 152. Ba3 {(Bf8) 0.75/32 4} Nf4 {(Ng7) 0.00/38
3} 153. Be7 {(Bd6) 0.75/30 6} Nh5 {(Ne6) 0.00/37 6} 154. Bc5 {(Bf6) 0.75/32 6}
Kd5 {(Ng7) 0.00/35 3} 155. Bf2 {(Bb4) 0.75/32 4} Ke4 {(Ng7) 0.00/37 6} 156. Be1
{(Bb6) 0.75/32 7} Ke5 {(Ng7) 0.00/37 5} 157. Bc3+ {(Bb4) 0.75/32 7} Ke4 {(Kd6)
0.00/36 3} 158. Bb4 {(Bb2) 0.75/32 7} Nf4 {(Ng7) 0.00/37 4} 159. Ba3 {(Bf8) 0.
75/31 11} Nh5 {(Ne6) 0.00/38 3} 160. Bd6 {(Bb4) 0.75/33 12} Kd5 {(Ng7) 0.00/37
5} 161. Bh2 {(Bb4) 0.75/32 3} Ke4 {(Ng7) 0.00/35 3} 162. Bc7 {(Bb8) 0.75/32 17}
Kd5 {(Ng7) 0.00/34 3} 163. Kf3 {0.75/29 3} Ng7 {0.00/34 6} 164. Bb8 {(Bg3) 0.
75/27 16} Kd4 {(Ne6) 0.00/34 6} 165. Bf4 {(Bh2) 0.75/28 6} Nf5 {0.00/33 3} 166.
Bg3 {(Kg4) 0.75/28 3} Ng7 {(Kd5) 0.00/35 5} 167. Bf2+ {(Bc7) 0.75/27 3} Ke5 {
(Kd5) 0.00/33 4} 168. Kg4 {(Ke3) 0.75/28 9} Kd5 {(Nf5) 0.00/36 3} 169. Be1 {
(Kf3) 0.75/30 3} Kd4 {(Ke4) 0.00/35 5} 170. Kf3 {(Ba5) 0.00/24 4} Nh5 {(Kd3) 0.
00/36 10} 171. Bb4 {0.00/60 0} 1/2-1/2

[lkaufman]

I taught Rybka 3 that two rooks on the seventh were worth more than two times one rook there, and it probably made R3 something like one Elo point stronger. We're not yet down to looking for single Elo point improvements in Komodo, so at this point Komodo only gets double credit for two rooks on the seventh, not more.

[lkaufman]

That name "pigs" for rooks on the 7th was around before Yasser was.

[lkaufman]

We have some hints that you are correct about move-ordering. If so, Komodo may turn out to be stronger than we think at serious time limits.

[benstoker]

Is this info about R3 that you provide here and elsewhere at all public knowledge?

I taught Rybka 3 that two rooks on the seventh were worth more than two times one rook there, and it probably made R3 something like one Elo point stronger. We're not yet down to looking for single Elo point improvements in Komodo, so at this point Komodo only gets double credit for two rooks on the seventh, not more.

[lkaufman]

Well, this stuff was made public by the clones, and anyway Vas had asked me to keep such general eval knowledge confidential for a year, and it's now been a year and a half. Still I won't reveal exact values of terms in R3 even if they are more or less public due to the clones (or "derivatives" if you prefer). I doubt that they would be of much use to a non-clone program anyway, as different programs require different values for terms.

[maxchgr]

Larry, I did not realize that you are working with other authors on their engines now. Are you no longer helping Vas with Rybka? Are you also helping on engines other than Komodo now? Please do explain to me why you have made that choice as I am rather curious.