Rybka 4 Beta joke?

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

Moderator: Ras

Edward German

Rybka 4 Beta joke?

Post by Edward German »

Now a Rybka 4 Beta 11.

This is just a trick of Rybka 4, right? That can not be normal! What is your opinion?

All games in this row:

[Event "5m + 0s, rated"]
[Site "Engine Room"]
[Date "2010.05.20"]
[Round "?"]
[White "Highendman, Rybka 4 Beta 11 SSE"]
[Black "Strangler_M, Rybka 3"]
[Result "0-1"]
[ECO "B92"]
[WhiteElo "2627"]
[BlackElo "2647"]
[PlyCount "152"]
[EventDate "2010.05.20"]
[EventType "blitz"]

1. e4 {B/0 0} c5 {B/0 0} 2. Nf3 {B/0 0} d6 {B/0 0} 3. d4 {B/0 0} cxd4 {B/0 0}
4. Nxd4 {B/0 0} Nf6 {B/0 0} 5. Nc3 {B/0 0} a6 {B/0 0} 6. Be2 {B/0 0} e5 {B/0 0}
7. Nb3 {B/0 0} Be7 {B/0 0} 8. O-O {B/0 0} O-O {B/0 0} 9. Be3 {B/0 0} Be6 {B/0 0
} 10. Qd2 {B/0 0} Nbd7 {B/0 0} 11. a4 {B/0 0} Rc8 {B/0 0} 12. a5 {B/0 0} Qc7 {
B/0 0} 13. Rfd1 {B/0 0} Rfe8 {B/0 0} 14. h3 {0.59/18 45} h6 {B/0 0} 15. Bf3 {
0.59/17 5} Qc6 {-0.14/15 23} 16. Kh2 {(Ëa1-a4) 0.59/17 0} Red8 {-0.23/15 6} 17.
Qe2 {(Êðh2-g1) 0.60/16 8} Nf8 {-0.23/16 8} 18. Nc1 {(Êðh2-g1) 0.62/17 14} N8h7
{-0.23/15 5} 19. h4 {(Êc1-d3) 0.63/18 2} h5 {-0.30/15 4} 20. Bb6 {0.52/19 9}
Re8 {-0.24/16 0} 21. g3 {0.54/18 4} Ng4+ {-0.24/15 34} 22. Kg1 {
(Êðh2-g2) 0.53/21 0} Nhf6 {-0.25/16 6} 23. Nd3 {0.53/19 9} Bd8 {-0.17/16 0} 24.
Bxd8 {0.53/18 4} Rexd8 {-0.17/17 0} 25. Ra3 {(Ñf3-g2) 0.52/16 9} Qc7 {
-0.33/14 4} 26. Ra4 {0.47/15 8} Bc4 {-0.33/14 0} 27. Qe1 {(Ëd1-a1) 0.47/16 3}
Re8 {-0.41/13 13} 28. Nb4 {(Ñf3xg4) 0.48/17 9} Be6 {-0.33/13 3} 29. Rd2 {
(Êb4-d3) 0.48/17 31} Qd7 {-0.30/15 18} 30. Qa1 {(Ëa4-a1) 0.48/15 5} Rc7 {
-0.43/13 3} 31. Bh1 {(Êb4-d3) 0.49/15 10} Qc8 {-0.42/13 3} 32. Nd3 {0.47/15 8}
Qb8 {-0.42/15 0} 33. Bf3 {(Ôa1-a3) 0.49/16 13} Rec8 {-0.41/14 3} 34. Kg2 {
(Ëa4-b4) 0.44/15 4} Rc6 {-0.41/14 5} 35. Rb4 {(Ôa1-e1) 0.44/15 0} Qc7 {
-0.40/12 4} 36. Nc1 {(Ôa1-a4) 0.43/14 4} Rc5 {-0.65/13 6} 37. Ra4 {0.31/17 0}
Bd7 {-0.61/14 3} 38. Ra3 {0.26/17 0} Rc4 {-0.56/14 5} 39. Nb3 {
(Êc1-d3) 0.31/16 0} Be6 {-0.61/11 2} 40. Ra4 {(Ôa1-e1) 0.40/16 4} Rc6 {
-0.44/12 4} 41. Nc1 {(Ëa4-b4) 0.40/15 1} Bc4 {-0.43/12 3} 42. Qa3 {
(Ëa4-b4) 0.41/16 0} Qb8 {-0.43/12 3} 43. Nd3 {0.41/16 0} Be6 {-0.43/14 4} 44.
Kg1 {(Ôa3-a1) 0.41/16 3} Kh7 {-0.41/13 2} 45. Qa1 {(Êðg1-g2) 0.42/14 6} Kh8 {
-0.43/13 9} 46. Rb4 {(Ôa1-e1) 0.42/14 5} Qc7 {-0.41/13 2} 47. Qa3 {0.42/15 5}
Kg8 {-0.40/14 2} 48. Kg2 {(Ôa3-a4) 0.41/16 0} Rd8 {-0.40/13 3} 49. Nc1 {
(Êðg2-g1) 0.41/16 0} Bc4 {-0.41/13 2} 50. Nb3 {(Êðg2-g1) 0.41/15 3} Nd7 {
-0.41/12 2} 51. Qa1 {(Ñf3-d1) 0.41/14 2} Be6 {-0.42/13 2} 52. Rd3 {
(Ôa1-a3) 0.39/14 7} Rc8 {-0.43/12 3} 53. Rd1 {0.40/16 0} Ndf6 {-0.46/13 4} 54.
Qa3 {0.40/15 0} Rxc3 {-0.86/13 14} 55. bxc3 {-0.12/15 0} Qxc3 {-1.14/14 12} 56.
Rxb7 {(Ëb4-b6) -1.02/15 0} Nxf2 {-1.33/13 1} 57. Kxf2 {-1.11/16 0} Qxc2+ {
-1.33/14 10} 58. Rd2 {-1.11/18 0} Nxe4+ {-1.37/15 6} 59. Bxe4 {-1.11/18 0} Qxe4
{-1.37/15 8} 60. Nc5 {-1.15/18 0} Rxc5 {-1.34/16 4} 61. Rb8+ {
(Ëb7-b6) -1.21/18 0} Bc8 {-1.22/12 2} 62. Qf3 {-1.06/16 4} Qg6 {-1.36/14 0} 63.
Qd3 {-1.20/14 4} Qxd3 {-1.35/15 0} 64. Rxd3 {-1.35/15 0} d5 {-1.36/17 1} 65.
Rxc8+ {-1.47/17 1} Rxc8 {-1.44/19 0} 66. Rxd5 {-1.47/19 0} Re8 {-1.44/19 2} 67.
Ke3 {-1.59/21 0} Re6 {-1.53/18 0} 68. Ke4 {-1.59/21 2} Kh7 {-1.63/18 0} 69. Rd8
{(Ëd5-d3) -1.72/18 4} Rg6 {-1.67/17 1} 70. Rd3 {-2.08/18 3} Rc6 {-1.67/20 0}
71. Kd5 {(Ëd3-d7) -2.08/18 2} Rc1 {-1.71/18 0} 72. Kxe5 {-2.30/19 1} Rc5+ {
-1.88/18 0} 73. Rd5 {(Êðe5-d6) -2.30/19 0} f6+ {-2.72/19 1} 74. Ke4 {
(Êðe5-d4) -10.32/20 1} Rxd5 {-4.99/20 2} 75. Kxd5 {-10.32/22 0} g5 {-4.99/21 5}
76. Kd4 {-14.39/22 0} Kg7 {
Highendman resigns (Lag: Av=0.27s, max=3.7s) -5.07/19 3} 0-1

[Event "5m + 0s, rated"]
[Site "Engine Room"]
[Date "2010.05.20"]
[Round "?"]
[White "KERCH, Rybka 3"]
[Black "Highendman, Rybka 4 Beta 11 SSE"]
[Result "1-0"]
[ECO "D38"]
[WhiteElo "2643"]
[BlackElo "2611"]
[PlyCount "225"]
[EventDate "2010.05.20"]
[EventType "blitz"]

1. d4 {B/0 0} Nf6 {B/0 0} 2. c4 {B/0 0} e6 {B/0 0} 3. Nf3 {B/0 0} Bb4+ {B/0 0}
4. Nc3 {B/0 0} d5 {B/0 0} 5. Bg5 {B/0 0} Nbd7 {B/0 0} 6. cxd5 {B/0 0} exd5 {
B/0 0} 7. e3 {B/0 0} c5 {B/0 0} 8. Bd3 {B/0 0} Qa5 {B/0 0} 9. Qc2 {B/0 0} c4 {
B/0 0} 10. Bf5 {B/0 0} O-O {B/0 0} 11. O-O {B/0 0} Re8 {B/0 0} 12. Nd2 {B/0 0}
g6 {B/0 0} 13. Bxd7 {B/0 0} Nxd7 {B/0 0} 14. e4 {B/0 0} Bxc3 {0.25/17 13} 15.
Qxc3 {B/0 0} Qxc3 {0.25/19 4} 16. bxc3 {B/0 0} dxe4 {0.25/19 7} 17. Nxc4 {
0.32/15 5} b6 {0.14/17 0} 18. Ne3 {0.39/17 12} Bb7 {(Ña6) 0.27/18 0} 19. c4 {
0.51/16 7} Kg7 {(f6) 0.38/17 2} 20. a4 {0.66/16 6} h6 {(a5) 0.47/18 40} 21. Bf4
{0.67/16 3} a5 {0.57/18 0} 22. Rab1 {0.73/17 26} Ba6 {0.67/18 15} 23. Rfc1 {
0.67/17 0} Rac8 {(Ëec8) 0.67/16 7} 24. h4 {0.86/14 5} Red8 {0.73/17 2} 25. h5 {
0.86/15 2} Rf8 {0.76/15 3} 26. Rc2 {0.86/15 2} f5 {(Ëfe8) 0.79/15 7} 27. hxg6 {
1.27/14 2} Kxg6 {0.79/19 0} 28. g3 {1.26/15 3} Rf7 {(Ëfd8) 0.79/19 1} 29. Kg2 {
1.20/16 6} h5 {0.79/19 19} 30. Rbb2 {1.20/18 0} Rh7 {0.79/18 6} 31. Kh3 {
1.20/18 0} Rf7 {(Ëch8) 0.79/19 22} 32. Rc1 {1.20/17 5} Rc6 {(Ëg7) 1.09/18 6}
33. Kh4 {1.20/16 9} Rc8 {(Ëf8) 1.23/18 0} 34. Rcc2 {1.20/16 3} Rff8 {
(Ëg7) 0.85/15 4} 35. Rb1 {1.20/15 2} Rf7 {(Ëh8) 1.13/17 7} 36. Rc3 {1.20/16 2}
Rff8 {(Ëh7) 1.13/16 11} 37. Kh3 {1.20/15 2} Rh8 {1.11/17 9} 38. Rcc1 {1.20/17 0
} Rc6 {(Ëh7) 1.13/18 11} 39. Rc2 {1.20/16 4} Ra8 {(Ëcc8) 1.13/18 3} 40. Rbc1 {
1.19/14 5} Rf8 {(Ëa7) 1.06/15 7} 41. Rc3 {1.20/14 4} Rf7 {(Ëh8) 1.07/18 18} 42.
Rb1 {1.18/16 4} Rc8 {1.06/17 1} 43. d5 {1.18/17 12} Rf6 {(Ëh7) 1.06/16 4} 44.
Kg2 {1.18/15 2} Rff8 {0.87/17 4} 45. Rcc1 {1.18/16 0} Kf6 {(Ëfe8) 0.87/16 1}
46. Rb2 {1.18/15 5} Rfd8 {(Kðg6) 0.92/15 3} 47. Rb3 {1.18/13 3} Rg8 {0.92/17 3}
48. Kh3 {1.18/15 0} Rg6 {0.92/17 4} 49. Rb2 {1.18/15 1} Rc5 {(Ëgg8) 0.92/16 0}
50. Kh2 {1.18/15 5} Rg8 {0.92/17 3} 51. Rcc2 {1.18/15 3} Ra8 {(Kðg6) 0.92/17 2}
52. Rb1 {1.18/13 3} Kg6 {(Ëh8) 0.92/18 0} 53. Kh3 {1.18/14 2} Rh8 {
(Ëcc8) 0.92/18 0} 54. Rh1 {1.18/14 2} Rcc8 {(Ëd8) 0.92/18 2} 55. Ra1 {1.18/14 2
} Nc5 {(Ëh7) 0.92/18 0} 56. Rb1 {1.18/14 1} Nd7 {0.92/17 1} 57. Kh4 {1.18/16 2}
Rhf8 {(Ëhg8) 0.92/17 0} 58. Rbb2 {1.18/16 5} Rfe8 {(Ëf6) 1.01/16 2} 59. Rc1 {
1.18/15 1} Rc5 {(Ëf8) 0.92/16 1} 60. Bd6 {1.18/15 10} Rcc8 {1.11/20 0} 61. Kh3
{1.18/16 1} Kf6 {(Ëh8) 0.80/16 1} 62. Bf4 {1.18/16 3} Rc5 {(Kðg6) 0.89/15 1}
63. Rbb1 {1.18/14 3} Rg8 {(Kðg6) 0.90/16 2} 64. Bd6 {1.18/14 2} Rcc8 {0.80/14 0
} 65. Rc2 {1.18/15 2} Kg6 {(Ëge8) 0.89/17 0} 66. Kh4 {1.18/14 6} Kf6 {
(Ëh8) 0.90/18 0} 67. Bf4 {1.18/14 1} Rg4+ {(Ëg6) 0.90/16 0} 68. Kh3 {1.18/15 1}
Rg6 {(Ëgg8) 1.02/16 0} 69. Rcc1 {1.18/14 2} Rc5 {(Ëgg8) 0.82/13 1} 70. Kh4 {
1.18/13 1} Rg7 {(Ëg8) 0.82/15 1} 71. Rc2 {1.18/14 2} Rg8 {0.82/15 1} 72. Kh3 {
1.18/15 2} Re8 {(Ëcc8) 0.82/14 1} 73. Rbb2 {1.18/14 2} Rcc8 {(Ëh8) 0.81/17 1}
74. Rc3 {1.18/14 8} Rg8 {(Ëc5) 0.89/18 0} 75. Rb1 {1.18/14 1} Rc5 {
(Kðg6) 0.82/17 1} 76. Kh4 {1.18/14 2} Rg6 {(Ëcc8) 0.89/17 0} 77. Rb2 {1.18/14 3
} Rc8 {0.89/16 0} 78. Rcc2 {0.98/14 4} Rc5 {(Ëg7) 0.82/13 1} 79. Rc1 {1.18/14 2
} Rc8 {(Ëg8) 0.89/16 0} 80. Rcb1 {1.18/13 1} Kf7 {0.89/17 0} 81. Rc2 {1.18/14 1
} Kf6 {0.81/14 1} 82. Kh3 {1.18/15 10} Rc5 {(Ëgg8) 0.81/18 0} 83. Bd6 {
1.05/12 1} Rc8 {0.81/18 0} 84. Ba3 {1.13/13 2} Rgg8 {0.81/16 0} 85. Kh4 {
1.13/12 8} Rg4+ {0.00/14 0} 86. Kh3 {0.00/16 1} Rgg8 {0.00/21 0} 87. Rd1 {
0.95/12 1} Kg6 {0.64/12 1} 88. Kh4 {0.89/12 1} Ne5 {0.68/11 1} 89. Rb2 {
0.72/12 0} Nxc4 {(Ëb8) 1.07/12 1} 90. Nxc4 {1.00/12 1} Bxc4 {0.96/16 0} 91.
Rxb6+ {1.06/13 0} Kh7 {1.01/16 0} 92. Bb2 {1.08/14 2} Be2 {(f4) 0.86/13 1} 93.
Rb7+ {1.18/13 1} Kg6 {1.12/13 0} 94. Re1 {1.18/14 0} Bg4 {(Ñf3) 1.09/14 0} 95.
Bd4 {1.35/12 1} Rb8 {1.21/13 0} 96. Ra7 {1.42/13 1} Rgc8 {(Ëgf8) 1.41/12 1} 97.
Be5 {1.46/12 1} Rb3 {1.41/15 0} 98. Rxa5 {1.55/12 1} Rc2 {(Ëb6) 1.84/12 1} 99.
Ra6+ {1.88/12 1} Kf7 {1.84/13 0} 100. Rh6 {1.98/12 1} Rc5 {2.15/10 1} 101. Rh7+
{2.16/13 0} Kg6 {2.34/12 0} 102. Rg7+ {2.53/12 3} Kh6 {3.03/14 0} 103. Rd7 {
3.33/12 4} Bf3 {(Kðg6) 3.44/14 0} 104. Rd6+ {6.23/10 4} Kh7 {2.44/5 0} 105. Kg5
{7.53/10 2} Kg8 {6.49/13 0} 106. Kg6 {7.48/9 0} Rc8 {(Ëxd5) 6.49/12 1} 107. Rc1
{5.11/9 0} Rbb8 {3.55/11 0} 108. Rxc8+ {14.17/11 0} Rxc8 {13.11/13 0} 109. Rd7
{16.83/11 0} h4 {15.75/13 0} 110. Rh7 {21.99/11 0} Kf8 {15.75/12 0} 111. d6 {
21.97/11 0} Bh5+ {#7/15 0} 112. Kxh5 {21.97/12 3} Rd8 {(Ëe8) #6/17 0} 113. Rh8+
{Highendman,Rybka 4 Beta 11 SSE4 resigns (Lag: Av=0.25s, max=1.5s) 21.97/8 0}
1-0

[Event "5m + 0s, rated"]
[Site "Engine Room"]
[Date "2010.05.20"]
[Round "?"]
[White "Yet Another Fool, Rybka 3"]
[Black "Highendman, Rybka 4 Beta 11 SSE"]
[Result "1-0"]
[ECO "E15"]
[WhiteElo "2613"]
[BlackElo "2596"]
[PlyCount "134"]
[EventDate "2010.05.20"]
[EventType "blitz"]

1. d4 {B/0 0} Nf6 {B/0 0} 2. c4 {B/0 0} e6 {B/0 0} 3. Nf3 {B/0 0} b6 {B/0 0} 4.
g3 {B/0 0} Ba6 {B/0 0} 5. b3 {B/0 0} Bb4+ {B/0 0} 6. Bd2 {B/0 0} Be7 {B/0 0} 7.
Bg2 {B/0 0} c6 {B/0 0} 8. Bc3 {B/0 0} d5 {B/0 0} 9. Ne5 {B/0 0} Nfd7 {B/0 0}
10. Nxd7 {B/0 0} Nxd7 {B/0 0} 11. Nd2 {B/0 0} O-O {B/0 0} 12. O-O {B/0 0} Rc8 {
B/0 0} 13. e4 {B/0 0} c5 {B/0 0} 14. exd5 {0.30/14 7} exd5 {B/0 0} 15. dxc5 {
0.27/15 14} dxc4 {0.44/14 6} 16. Nxc4 {0.23/15 11} Bxc5 {(Tc8xc5) 0.34/16 0}
17. Re1 {0.51/14 9} Rc7 {(La6-b5) 0.72/15 4} 18. Ne3 {0.58/14 4} Bxe3 {
0.74/16 0} 19. Rxe3 {0.52/15 4} h6 {(Sd7-c5) 0.67/15 2} 20. Qd4 {0.60/13 6} Nf6
{0.81/15 4} 21. Qf4 {0.56/16 1} Bb7 {0.93/17 5} 22. Be5 {0.72/17 1} Bxg2 {
(Tc7-d7) 0.89/19 40} 23. Bxc7 {0.73/15 4} Qa8 {1.22/18 0} 24. Bd6 {0.66/16 5}
Nd5 {1.00/18 0} 25. Qe5 {0.96/17 16} Rc8 {(Tf8-d8) 1.01/18 0} 26. Rd3 {
1.00/17 5} Re8 {(Sd5-f6) 1.15/16 1} 27. Qd4 {1.06/15 3} Be4 {1.22/17 0} 28. Rd2
{1.02/17 6} Qc8 {1.23/16 5} 29. Re1 {1.02/16 1} Bf3 {1.22/17 4} 30. Rxe8+ {
0.99/17 1} Qxe8 {1.22/19 0} 31. Qe5 {0.99/18 4} Qc8 {1.10/18 0} 32. Qe1 {
0.99/18 5} b5 {(a7-a5) 0.88/14 15} 33. b4 {1.09/14 3} Qh3 {1.30/16 4} 34. Qf1 {
1.08/17 1} Qe6 {1.50/18 23} 35. Bb8 {1.23/17 0} a6 {1.39/15 9} 36. h3 {
1.23/16 0} f5 {(De6-f6) 1.34/15 10} 37. a3 {1.35/14 2} f4 {1.72/14 10} 38. Kh2
{1.46/16 0} Qf5 {(De6-e4) 1.91/16 5} 39. g4 {1.77/15 3} Qe4 {2.71/15 7} 40. Qc1
{1.83/15 1} Kf7 {(Kg8-f8) 2.48/15 7} 41. Bd6 {2.34/15 7} Kf6 {
(Lf3-g2) 2.51/16 2} 42. Qc2 {2.78/14 2} Qxc2 {(g7-g5) 3.05/17 5} 43. Rxc2 {
2.83/16 1} h5 {3.20/18 2} 44. gxh5 {2.87/17 0} Bxh5 {3.31/19 3} 45. Rc6 {
3.18/19 3} Kf5 {3.36/20 4} 46. Rxa6 {3.20/19 1} Be2 {3.36/18 2} 47. Rc6 {
3.36/19 4} Bf1 {(g7-g6) 3.77/18 14} 48. Kg1 {3.39/19 4} Be2 {3.91/19 0} 49. Rc8
{3.52/18 5} Bd3 {(f4-f3) 3.84/18 5} 50. Re8 {3.72/17 3} Nc3 {(Ld3-c4) 3.90/19 2
} 51. Rf8+ {3.85/16 1} Ke4 {3.90/19 0} 52. Rxf4+ {4.05/17 3} Kd5 {3.90/21 0}
53. Rf3 {4.14/19 3} Ne2+ {4.26/20 5} 54. Kh2 {4.14/19 0} Be4 {
(Ld3-b1) 4.26/21 3} 55. Re3 {4.24/18 0} Nd4 {4.32/22 0} 56. Bc5 {4.24/20 1} Ne6
{4.32/21 0} 57. Kg3 {4.24/20 3} g6 {(Le4-b1) 4.42/21 0} 58. f3 {4.32/20 4} Bb1
{(Le4-c2) 4.60/21 0} 59. Re1 {4.35/20 2} Bd3 {(Lb1-f5) 4.78/22 9} 60. Rc1 {
4.47/19 2} Ng5 {(Ld3-f5) 4.78/22 0} 61. Be7 {4.62/17 2} Ne6 {4.99/20 5} 62. Bf6
{4.83/20 0} Nf8 {5.03/20 5} 63. Kf4 {4.94/19 0} Nd7 {5.03/19 4} 64. Bb2 {
5.00/20 0} Nb6 {5.41/19 4} 65. Rc5+ {5.02/19 3} Kd6 {5.41/20 0} 66. Be5+ {
5.02/19 18} Ke7 {5.78/21 0} 67. Rc7+ {5.12/18 2} Ke8 {
Highendman,Rybka 4 Beta 11 SSE4 resigns (Lag: Av=0.16s, max=0.4s) 5.78/18 6}
1-0

I don't understand it. What is wrong here?

ED.
jpqy
Posts: 556
Joined: Thu Apr 24, 2008 9:31 am
Location: Belgium

Re: Rybka 4 Beta joke?

Post by jpqy »

Hmm..already Beta 11 ..they just try things out,to see how the engine react on some settings and use the best one for the final release.
Loosing these games now says nothing about the strenght..the surprice at the end will be even more!

We will see...
yanquis1972
Posts: 1766
Joined: Wed Jun 03, 2009 12:14 am

Re: Rybka 4 Beta joke?

Post by yanquis1972 »

yes, there's probably tons of toying around with settings...by accounts i've read R4 is about 50 elo with a much different evaluation. if the eval itself added the elo that would be one thing but it appears this is not the case...it's just markedly different (less 'solid', more speculative), yet the search is presumably faster -- depending on what vas chooses to do, of course.
lkaufman
Posts: 6284
Joined: Sun Jan 10, 2010 6:15 am
Location: Maryland USA
Full name: Larry Kaufman

Re: Rybka 4 Beta joke?

Post by lkaufman »

I would estimate that about half of the predicted +50 Elo is from faster and/or better search and the other half from better eval. The eval is not so different from R3 in general, just in certain areas (king safety and endgames mostly).
User avatar
Dr.Wael Deeb
Posts: 9773
Joined: Wed Mar 08, 2006 8:44 pm
Location: Amman,Jordan

Re: Rybka 4 Beta joke?

Post by Dr.Wael Deeb »

lkaufman wrote:I would estimate that about half of the predicted +50 Elo is from faster and/or better search and the other half from better eval. The eval is not so different from R3 in general, just in certain areas (king safety and endgames mostly).
Rybka 4=a personality of Rybka 3 with minor tweaks and +40-60 Elo at best....
Diagnosing the upcoming release regards,
Dr.D
_No one can hit as hard as life.But it ain’t about how hard you can hit.It’s about how hard you can get hit and keep moving forward.How much you can take and keep moving forward….
User avatar
Rubinus
Posts: 1213
Joined: Thu Jan 18, 2007 4:05 pm
Location: Prague
Full name: Pavel Háse

Re: Rybka 4 Beta joke?

Post by Rubinus »

SSE is version for old machines, 32bit.

My guess: Rybka 4 improved cca 50-80 ELO, than 3.
Edward German

Re: Rybka 4 Beta joke?

Post by Edward German »

jpqy wrote:Hmm..already Beta 11 ..they just try things out,to see how the engine react on some settings and use the best one for the final release.
Loosing these games now says nothing about the strenght..the surprice at the end will be even more!

We will see...
Yes, Rybka-Team is testing with different parameters, I think!

See as example this game:

[Event "16m + 0s, rated"]
[Site "Engine Room"]
[Date "2010.05.20"]
[Round "?"]
[White "Future Breeze, Rybka 3 DynamicBree"]
[Black "Highendman, Rybka 4 Beta 11 SSE"]
[Result "*"]
[ECO "D30"]
[WhiteElo "2436"]
[BlackElo "2590"]
[PlyCount "44"]
[EventDate "2010.05.20"]
[EventType "rapid"]

1. Nf3 {0} Nf6 {B/0 0} 2. d4 {0} e6 {B/0 0} 3. c4 {0} d5 {B/0 0} 4. Bg5 {0}
dxc4 {B/0 0} 5. e4 {0} Kd7 {0.55/15 62} 6. Nc3 {28} c6 {0.71/12 19} 7. e5 {23}
h6 {2.09/17 19} 8. Bh4 {0} g5 {2.13/17 8} 9. exf6 {0} gxh4 {2.01/18 12} 10.
Ne5+ {0} Kc7 {2.15/17 8} 11. Nxf7 {0} Qxf6 {2.15/18 12} 12. Nxh8 {0} Qg7 {
1.65/13 15} 13. Qf3 {21} Qxh8 {2.46/17 21} 14. Bxc4 {0} Kb6 {2.46/13 25} 15.
O-O {23} Be7 {3.32/14 39} 16. Rfe1 {21} Qf8 {4.27/16 51} 17. Bxe6 {0} Na6 {
4.52/16 13} 18. Qg4 {25} Bb4 {3.71/13 55} 19. a3 {0} Bd6 {5.14/16 36} 20. Bxc8
{0} Qxc8 {4.46/14 10} 21. Qxh4 {24} Qf5 {5.65/13 68} 22. Qxh6 {14} Bf4 {
5.73/14 1} *

Position after move 5:

Future Breeze,R - Highendman,R, 16m + 0s, rated rapid 2010
[d]rnbq1b1r/pppk1ppp/4pn2/6B1/2pPP3/5N2/PP3PPP/RN1QKB1R w KQ - 0 1

Rybka 4 Beta 11 played here 5...Kd7?? :roll:

Very funny this King on d7. Or not? 8-)

ED.
jpqy
Posts: 556
Joined: Thu Apr 24, 2008 9:31 am
Location: Belgium

Re: Rybka 4 Beta joke?

Post by jpqy »

Hi Ed,if you see this kind of test opening..it's like the engine is already much stronger and then try to see if it can even win from a bad start position,or at least make a draw...

Why else you do this..you want to see how the engine react on these games!

JP.
beram
Posts: 1187
Joined: Wed Jan 06, 2010 3:11 pm

Re: Rybka 4 Beta joke?

Post by beram »

lkaufman wrote:I would estimate that about half of the predicted +50 Elo is from faster and/or better search and the other half from better eval. The eval is not so different from R3 in general, just in certain areas (king safety and endgames mostly).
Dear Larry, your mentioned 50 ELO gain of Rybka4, is this based on your testgames or are they just a guess ?

Kind regards Bram
Jim Walker
Posts: 436
Joined: Thu Mar 09, 2006 2:31 am

Re: Rybka 4 Beta joke?

Post by Jim Walker »

Only 50 Elo gain after 2 years? Shame on Vas. I would guess he is holding out the best for his "Super Rybka" on line.