Houdini beats Rybka4 with 57 – 43 %

Discussion of computer chess matches and engine tournaments.

Moderators: hgm, Rebel, chrisw

Albert Silver
Posts: 3019
Joined: Wed Mar 08, 2006 9:57 pm
Location: Rio de Janeiro, Brazil

Re: Houdini beats Rybka4 - AT LTC - with 57 – 43 %

Post by Albert Silver »

beram wrote:

Code: Select all

LTC games
T8100 (6,29 Fritzmark), LTC - 20m/40+10m/20+(10m+12s) Nunn2 – 
yet after 36 games of 50 played
                            
1   Houdini 1.01 w32 2_CPU   +10/=22/-4 58,33 %    21/36
2   Deep Rybka 4 w32         +4/=22/-10 41,67 %    15/36

Regarding the special Rybka 4 time control settings for short time control. This doen't count for my used LTC (20m/40+10m/20+(10m+12s))

I did earlier also test Fire 1.31 at this LTC and here Deep Rybka wins:

Code: Select all

T8100 (6,29 Fritzmark) LTC 20m/40+10m/20+(10m+12s)  50 games Bram privat suite 1.2
                      
1   Deep Rybka 4 w32  +15/=28/-7 58.00%   29.0/50
2   Fire 1.3 w32      +7/=28/-15 42.00%   21.0/50
No, I completely agree, they do not play better at repeating time controls. They may even play worse.

For 40 moves in 20 mins, I would suggest Normal Move = 100, and Max Move =150 as tested by the CCRL. I honestly don't know how it will act for the 3rd time control though.
"Tactics are the bricks and sticks that make up a game, but positional play is the architectural blueprint."
Albert Silver
Posts: 3019
Joined: Wed Mar 08, 2006 9:57 pm
Location: Rio de Janeiro, Brazil

Re: Houdini beats Rybka4 with 57 – 43 %

Post by Albert Silver »

Gino Figlio wrote:Deep Rybka 4 SSE42 x64 TC3100150 vs. Houdini x64 POPCNT_4CPU
2 CPU each, ponder off TC 1/0
Quad i7-920 4.0 GHz
GUI Aquarium 4.0.5
Narrowbook, played each opening twice

117-113, + 6 elo
I should warn you, that at 1+0, 100/150 is not the best by any margin. Try instead 72/115 with a TC Buffer at 1 (instead of the default 3).
"Tactics are the bricks and sticks that make up a game, but positional play is the architectural blueprint."
Gino Figlio
Posts: 95
Joined: Sun Jan 10, 2010 6:10 am
Location: Lamar, Colorado, USA

Re: Houdini beats Rybka4 with 57 – 43 %

Post by Gino Figlio »

Thanks, I will try those settings. I replied to one of your posts in the rybka forum but it was moderated so I deleted it.
It is fun to try to find different settings, maybe some of the other engines need to learn how allowing the user to customize options can be a plus.
kingliveson

Re: Houdini beats Rybka4 with 57 – 43 %

Post by kingliveson »

Albert Silver wrote:
Gino Figlio wrote:Deep Rybka 4 SSE42 x64 TC3100150 vs. Houdini x64 POPCNT_4CPU
2 CPU each, ponder off TC 1/0
Quad i7-920 4.0 GHz
GUI Aquarium 4.0.5
Narrowbook, played each opening twice

117-113, + 6 elo
I should warn you, that at 1+0, 100/150 is not the best by any margin. Try instead 72/115 with a TC Buffer at 1 (instead of the default 3).
Setting TC Buffer to 1 might will cause Rybka to lose on time on occasion.

[Event "Fire vs Rybka RND P02 1+0"]
[Site "W0111D0001"]
[Date "2010.06.05"]
[Round "4"]
[White "Deep Rybka 4 SSE42 x64"]
[Black "Fire 1.31 x64"]
[Result "0-1"]
[ECO "A00"]
[Annotator "-0.29;-0.32"]
[PlyCount "294"]
[TimeControl "60"]

{AMD Phenom(tm) II X4 940 Processor W=11.0 plies; 214kN/s B=15.2 plies; 5,
527kN/s} 1. d3 e5 2. c3 d5 3. d4 Nc6 4. dxe5 Nxe5 {Both last book move} 5. Nf3
{-0.29/12 1} Nxf3+ {-0.32/16 1 (Nc6)} 6. exf3 {-0.27/14 1} Nf6 {-0.32/16 0} 7.
Bd3 {-0.31/13 1 (Be3)} Bd6 {-0.35/17 1} 8. O-O {-0.32/13 1} O-O {-0.36/17 1} 9.
Be3 {-0.33/13 1 (Re1)} c5 {-0.35/16 1 (h6)} 10. Re1 {-0.30/13 1} Be6 {-0.31/15
1 (h6)} 11. Nd2 {-0.31/12 1} h6 {-0.37/14 0 (Re8)} 12. f4 {-0.29/12 2 (Qa4)}
Qb6 {-0.32/14 1 (Ng4)} 13. b4 {-0.18/13 1 (f5)} Qc7 {-0.46/16 1} 14. bxc5 {-0.
27/13 2} Bxf4 {-0.42/15 1} 15. Bxf4 {-0.29/13 1} Qxf4 {-0.42/14 0} 16. Rb1 {-0.
29/13 2} Qc7 {-0.50/16 1} 17. Nb3 {-0.28/12 1 (Re3)} b6 {-0.32/16 8 (Rfe8)} 18.
cxb6 {-0.25/13 1} axb6 {-0.32/14 0} 19. Qc2 {-0.25/13 1 (Qd2)} Ng4 {-0.42/16 2
(Ra3)} 20. g3 {-0.09/12 1} Rfc8 {-0.27/16 1} 21. Rbc1 {-0.09/12 1} Bd7 {-0.27/
15 0 (Ne5)} 22. Nd4 {0.00/13 1 (Bh7+)} Qc5 {-0.23/15 2 (Nf6)} 23. Qb2 {-0.03/
12 3 (f3)} Nf6 {-0.15/14 1 (Re8)} 24. Rc2 {-0.01/13 2} Ra3 {-0.21/15 1 (Re8)}
25. Nb3 {0.00/14 3} Qd6 {-0.17/16 1} 26. Qc1 {-0.10/12 1 (Nd4)} Rca8 {-0.44/14
1} 27. Qd2 {-0.22/11 1} Ba4 {-0.41/14 0} 28. Rb2 {-0.11/11 1} h5 {-0.32/14 2
(Qc6)} 29. f3 {0.06/10 1 (Bb1)} Rc8 {-0.23/13 1 (Nd7)} 30. Re3 {0.00/11 1} Bxb3
{-0.25/14 0 (Qc7)} 31. Rxb3 {0.00/12 1 (axb3)} Rca8 {-0.26/14 0} 32. Bb1 {0.03/
12 1 (Rxa3)} Qc5 {-0.18/14 1 (R3a4)} 33. Kf2 {0.00/12 1 (Kg2)} g6 {-0.21/15 2
(R8a4)} 34. Kg2 {-0.02/11 2 (Qd4)} Kg7 {-0.20/14 0} 35. Re5 {-0.02/10 0} h4 {
-0.26/14 1 (Qc4)} 36. Re3 {-0.07/10 1} hxg3 {-0.29/14 1 (Rc8)} 37. hxg3 {-0.05/
11 0} Rc8 {-0.30/14 0 (R3a4)} 38. Rxa3 {-0.19/13 2} Qxa3 {-0.30/12 0} 39. Kh3 {
-0.19/13 1 (c4)} Rh8+ {-0.51/14 1 (Rc4)} 40. Kg2 {-0.05/6 0} Qf8 {-0.65/15 1
(Rc8)} 41. Re5 {-0.28/10 1 (Qe1)} Qg8 {-0.98/14 0 (Rh7)} 42. Kg1 {-0.85/10 0
(Qd4)} Qh7 {-1.09/13 0} 43. Qg2 {-0.85/11 0} Qh6 {-1.09/12 0 (Rc8)} 44. Re1 {
-0.52/10 0} Rc8 {-1.19/14 0} 45. Qf2 {-0.58/10 0} Rxc3 {-1.19/12 0 (Qh5)} 46.
Kg2 {-0.79/10 0} b5 {-1.19/14 0} 47. Qd4 {-0.79/12 1} Rc1 {-1.19/15 0} 48. Qf2
{-1.02/11 0} d4 {-1.22/15 0} 49. Bd3 {-1.04/13 0} Rxe1 {-1.27/16 0} 50. Qxe1 {
-1.04/9 0} Qg5 {-1.20/16 0} 51. Qb4 {-1.05/13 0} Qe3 {-1.20/16 0} 52. Qxb5 {-0.
95/12 0} Qd2+ {-1.20/14 0} 53. Kf1 {-0.97/12 0} Qxa2 {-1.20/16 0} 54. Qe5 {-1.
03/12 0} Qa1+ {-1.20/16 0} 55. Ke2 {-1.03/15 0 (Kf2)} Qb2+ {-1.16/18 0} 56. Ke1
{-1.03/16 0 (Kf1)} Qg2 {-1.21/18 0 (Qc1+)} 57. Be2 {-0.94/14 0} d3 {-1.10/18 1
(Qg1+)} 58. Bxd3 {-0.94/14 0} Qxf3 {-1.10/17 0} 59. Be2 {-0.92/14 0} Qg2 {-1.
10/18 0 (Qe4)} 60. Qc3 {-0.93/12 0} Qg1+ {-1.11/18 0} 61. Bf1 {-0.96/14 0} Qb6
{-1.11/18 0 (g5)} 62. Be2 {-0.94/12 0 (Bg2)} Kh7 {-1.11/17 0 (Kg8)} 63. Qe5 {
-0.97/13 0 (Bf3)} Qg1+ {-1.12/15 0} 64. Kd2 {-0.97/13 0} Qf2 {-1.16/16 0} 65.
Kd3 {-1.00/12 0} Qb6 {-1.13/16 0 (Kg8)} 66. Bf3 {-1.04/12 0} Nd7 {-1.19/15 0}
67. Qd5 {-1.04/12 0 (Qd4)} Qf6 {-1.26/14 0} 68. Ke3 {-1.09/13 1 (Qxd7)} Qc3+ {
-1.21/16 1} 69. Kf2 {-1.09/13 0} Ne5 {-1.21/17 0} 70. Be2 {-1.06/12 0} Qb2 {-1.
21/17 1 (Qa1)} 71. Kf1 {-1.06/12 0 (Qd6)} Kg7 {-1.24/15 0 (Kg8)} 72. Qc5 {-1.
11/13 0 (Qe4)} Kg8 {-1.21/16 0 (Nd7)} 73. Qc8+ {-1.06/14 0 (Qe3)} Kh7 {-1.19/
18 0} 74. Qc5 {-1.09/14 0} Kg7 {-1.20/18 0 (Qb1+)} 75. Qd5 {-1.06/14 0 (Qe3)}
Qc1+ {-1.21/18 0 (Qb1+)} 76. Kf2 {-1.06/14 0} Qc3 {-1.21/18 1 (Qc7)} 77. Qe4 {
-1.18/13 0 (Qd6)} Qb2 {-1.24/16 0 (Qc5+)} 78. Ke3 {-1.18/12 0 (Kf1)} Qc1+ {-1.
23/15 0} 79. Kf2 {-1.17/14 0} Qc5+ {-1.20/17 0} 80. Qe3 {-1.17/14 0} Qd5 {-1.
20/16 0 (Qxe3+)} 81. Qc3 {-1.12/13 0 (Kg1)} g5 {-1.21/15 0 (f5)} 82. Qb4 {-1.
08/11 0 (Kg1)} Kg6 {-1.18/15 0} 83. Qb6+ {-1.06/11 0 (Ke3)} f6 {-1.19/16 0} 84.
Qb8 {-1.06/11 0} f5 {-1.19/15 0 (Kg7)} 85. Qe8+ {-1.08/11 0} Nf7 {-1.19/15 0}
86. Qg8+ {-1.08/11 0 (Bf3)} Kf6 {-1.19/15 0} 87. Bh5 {-1.08/13 0 (Qc8)} Qd4+ {
-1.21/16 0} 88. Kf1 {-1.08/14 0} Qa1+ {-1.21/16 0 (Qd3+)} 89. Kg2 {-1.08/16 0
(Kf2)} Qb2+ {-1.21/16 0} 90. Kg1 {-1.08/15 0 (Kf1)} Qb1+ {-1.20/17 0 (Qc1+)}
91. Kh2 {-1.08/16 0 (Kg2)} Qc2+ {-1.02/17 0} 92. Kg1 {-1.08/16 0} Qc5+ {-1.02/
17 0} 93. Kg2 {-1.08/16 0} Qc6+ {-0.99/17 0 (Qc2+)} 94. Kh2 {-0.96/14 0 (Kg1)}
Qe6 {-1.02/16 0 (Qc2+)} 95. Qg6+ {-0.82/12 0} Ke7 {-1.02/14 0} 96. Qg7 {-0.82/
0 0} Qf6 {-1.04/16 0 (g4)} 97. Qg8 {-0.77/10 0 (Qh7)} Nd6 {-0.83/14 1 (Qb2+)}
98. Bf3 {-0.77/9 0} Qb2+ {-0.83/14 0} 99. Kh3 {-0.81/10 0} Nf7 {-0.82/15 0}
100. Bh5 {-0.82/10 0} g4+ {-0.82/14 0} 101. Bxg4 {-0.82/10 0} Kf6 {-0.82/14 0
(Qc1)} 102. Qf8 {-0.79/8 0} fxg4+ {-0.82/14 0} 103. Kxg4 {-0.79/10 0} Qe2+ {-0.
82/15 0 (Qb7)} 104. Kh3 {-0.78/12 0 (Kf4)} Qf1+ {-0.82/15 0 (Qf3)} 105. Kh2 {
-0.78/12 0} Qf2+ {-0.82/15 0} 106. Kh3 {-0.78/12 0} Qf3 {-0.82/16 0 (Qe3)} 107.
Kh2 {-0.78/12 0} Qh5+ {-0.82/16 0 (Qe4)} 108. Kg2 {-0.78/12 0} Qd5+ {-0.82/16 0
} 109. Kf2 {-0.78/6 0} Qf5+ {-0.82/16 0} 110. Kg2 {-0.78/3 0 (Ke3)} Qe4+ {-0.
82/16 0 (Qc2+)} 111. Kh2 {-0.78/13 0 (Kf2)} Qc2+ {-0.82/17 0 (Qh7+)} 112. Kg1 {
-0.78/7 0 (Kh3)} Qc1+ {-0.82/16 0 (Qc3)} 113. Kh2 {-0.78/13 0} Qd2+ {-0.78/17
1 (Qb2+)} 114. Kh3 {-0.78/13 0 (Kg1)} Qd3 {-0.78/16 0 (Qe3)} 115. Kh2 {-0.78/
12 0 (Kg2)} Qh7+ {-0.74/15 0} 116. Kg2 {-0.78/9 0} Qh5 {-0.74/15 0 (Qf5)} 117.
Kf2 {-0.78/10 0} Qh2+ {-0.74/14 0 (Qg4)} 118. Kf3 {-0.78/8 0} Qh1+ {-0.71/14 0}
119. Kf2 {-0.78/10 0} Qd5 {-0.71/14 0 (Qe4)} 120. g4 {-0.77/6 0} Qd4+ {-0.71/
14 0 (Qd2+)} 121. Kf3 {-0.77/9 0} Qg1 {-0.79/15 0 (Qd3+)} 122. Ke2 {-0.77/8 0
(Ke4)} Qxg4+ {-0.79/14 0} 123. Kd3 {-0.77/9 0} Qf5+ {-0.79/14 0 (Qf3+)} 124.
Kc3 {-0.77/7 0} Qf3+ {-0.79/14 0 (Qd5)} 125. Kb4 {-0.77/6 0 (Kd4)} Qe4+ {-0.79/
13 0 (Qd3)} 126. Kb3 {-0.77/8 0 (Kc3)} Qd3+ {-0.79/12 0 (Qe3+)} 127. Kb4 {-0.
77/9 0 (Ka4)} Qd4+ {-0.79/13 0 (Qd2+)} 128. Kb3 {-0.77/10 0 (Ka3)} Qd5+ {-0.79/
14 0 (Qd1+)} 129. Kc3 {-0.77/8 0} Qc6+ {-0.79/14 0 (Qe5+)} 130. Kb4 {-0.77/7 0
(Kd4)} Qb6+ {-0.79/14 0 (Qb7+)} 131. Kc3 {-0.77/9 0} Qe3+ {-0.79/15 0} 132. Kb4
{-0.77/8 0 (Kc4)} Qd2+ {-0.79/14 0} 133. Kc4 {-0.77/6 0 (Kb3)} Qb2 {-0.79/15 0
(Qe3)} 134. Kd3 {-0.77/4 0} Qb3+ {-0.79/15 0} 135. Kd4 {-0.77/6 0} Qg3 {-0.79/
14 0} 136. Kc4 {-0.77/4 0} Qe5 {-0.79/14 0 (Qf4+)} 137. Kd3 {-0.77/5 0 (Kb3)}
Qb2 {-0.79/14 0 (Qb5+)} 138. Kc4 {-0.77/5 0} Qc2+ {-0.79/15 0 (Qe2+)} 139. Kb4
{-0.77/8 0 (Kd4)} Qb1+ {-0.79/15 0 (Qd3)} 140. Ka5 {-0.77/2 0 (Kc3)} Qf5+ {-0.
72/15 0 (Qa2+)} 141. Kb4 {-0.77/3 0} Qg4+ {-0.72/15 0 (Qd3)} 142. Kb3 {-0.77/2
0 (Kc3)} Qe6+ {-0.79/16 0 (Qd1+)} 143. Kb4 {-0.77/0 0 (Kc3)} Qe1+ {-0.79/15 0}
144. Kc4 {-0.77/10 0 (Kb3)} Qd2 {-0.79/15 0 (Qa1)} 145. Kb3 {-0.77/1 0} Qd1+ {
-0.72/14 0 (Qe3+)} 146. Kc3 {-0.77/3 0 (Kb4)} Qc1+ {-0.72/15 0 (Qd5)} 147. Kb4
{-0.77/10 0} Qa1 {-0.65/15 0 (Qf4+) Time} 0-1
beram
Posts: 1187
Joined: Wed Jan 06, 2010 3:11 pm

Re: Houdini beats Rybka4 with 57 – 43 %

Post by beram »

Nunn2 Match at LTC has ended:
second half Rybka4 got back a little, but still a win for Houdini
I take my hat of mr Houdart, thank you Robert for this great engine.

Code: Select all

T8100, 20m/40+10m/20+(10m+12s)  0
                            
1   Houdini 1.01 w32 2_CPU  +11/=33/-6 55.00%   27.5/50
2   Deep Rybka 4 w32        +6/=33/-11 45.00%   22.5/50
kind regards Bram