in my match i expected even more draws as most of FB appears to be R3 code(most not all).yanquis1972 wrote:i dont see why that indicates anything is wrong. if i am right that 8CPU scaling isn't good for FB then it would be = to somewhat weaker than rybka. it came out to something like +60 elo (a couple more games were run and rybka won one of them) in my tests.
right now i am running FB vs IHv73 @ 4+2 and there are 16 draws in 18 games. this is using a truncated 5 move book. however these two engines are very close to identical imo.
Firebird 1.0 beta vs Rybka 3, 15+10
Moderators: hgm, Rebel, chrisw
-
- Posts: 305
- Joined: Sat Jan 02, 2010 11:31 pm
- Location: bristol,uk
Re: Firebird 1.0 beta vs Rybka 3, 15+10
-
- Posts: 1187
- Joined: Wed Jan 06, 2010 3:11 pm
Re: Firebird 1.0 beta vs Rybka 3, 15+10
And yet at longer time control an even better performance:
Compaq 6710b, Blitz 25m, Nunn2 Test Match T8300 2100 Mhz, 64 Mb Hash
1 FireBird 1.0 beta w32 +16/=28/-6 60.00% 30.0/50
2 Rybka 3 32-bit +6/=28/-16 40.00% 20.0/50
I am convinced, we have a new leader in these chain of Ip/Rob/Ig/Iv/ namely Firebird with sustained strong play and yet multi pv and other profi futures. Unbelievable how strong this far from perfect engine yet plays, after years of Rybkadomination. Now Vas wil have to improve for 100 ELO otherwise no one buys a Rybka 4. For us who want the strongest possible engine these are sensational times.
greets Bram
Compaq 6710b, Blitz 25m, Nunn2 Test Match T8300 2100 Mhz, 64 Mb Hash
1 FireBird 1.0 beta w32 +16/=28/-6 60.00% 30.0/50
2 Rybka 3 32-bit +6/=28/-16 40.00% 20.0/50
I am convinced, we have a new leader in these chain of Ip/Rob/Ig/Iv/ namely Firebird with sustained strong play and yet multi pv and other profi futures. Unbelievable how strong this far from perfect engine yet plays, after years of Rybkadomination. Now Vas wil have to improve for 100 ELO otherwise no one buys a Rybka 4. For us who want the strongest possible engine these are sensational times.
greets Bram
-
- Posts: 235
- Joined: Fri Mar 10, 2006 9:19 am
- Location: Quezon City, Philippines
Re: Firebird 1.0 beta vs Rybka 3, 15+10
Before, Rybka was only fighting against an upstart. Now she has a whole family of upstarts to deal with. Daddy Ippo, sons Ivanhoe and Robbo, and now a pet Firebird.
But Vas never ceased to amaze us with previous Rybkas. I have a feeling R4 will be another revelation that will silence everything. We are in an exciting period of computer chess
But Vas never ceased to amaze us with previous Rybkas. I have a feeling R4 will be another revelation that will silence everything. We are in an exciting period of computer chess
-
- Posts: 1738
- Joined: Sun Dec 13, 2009 6:09 pm
Re: Firebird 1.0 beta vs Rybka 3, 15+10
I agree with this. This puts some pressure on Rybka's creator to improve it. Imagine there were no RL/Igor/Ivan/FB, it would be the same boring tournaments won by R3 by a large margin. Now R3 has competition or is it backwards?Eraserheads wrote:We are in an exciting period of computer chess
-
- Posts: 305
- Joined: Sat Jan 02, 2010 11:31 pm
- Location: bristol,uk
Re: Firebird 1.0 beta vs Rybka 3, 15+10
i disagree with you guys.all these comic engines are unstable on big hardware.here is an example with FB on 8 cpu.as soon as it is losing it crashes.this has happened on more than one occasion.Carlos777 wrote:I agree with this. This puts some pressure on Rybka's creator to improve it. Imagine there were no RL/Igor/Ivan/FB, it would be the same boring tournaments won by R3 by a large margin. Now R3 has competition or is it backwards?Eraserheads wrote:We are in an exciting period of computer chess
here is an example of FB crashing when it is about to lose game.
[Event "0000-7160E6B413, 20'/40+20'/40+20'"]
[Site "0000"]
[Date "2010.01.19"]
[Round "3"]
[White "FireBird 1.0 beta x64"]
[Black "Rybka 3"]
[Result "0-1"]
[ECO "B83"]
[Annotator "0.03;0.03"]
[PlyCount "166"]
[TimeControl "40/1200:40/1200:1200"]
{Intel(R) Core(TM)2 Extreme CPU X9775 @ 3.20GHz 3591 MHz W=20.7 plies; 10,
207kN/s; HS-Masterbook.ctg B=17.0 plies; 355kN/s; 3 TBAs; HS-Masterbook.ctg}
1. e4 {B 0} c5 {B 0} 2. Nf3 {B 0} e6 {B 0} 3. d4 {B 0} cxd4 {B 0} 4. Nxd4 {B 0}
Nf6 {B 0} 5. Nc3 {B 0} d6 {B 0} 6. Be2 {B 0} Be7 {B 0} 7. O-O {B 0} a6 {B 0} 8.
a4 {B 0} O-O {B 0} 9. Be3 {B 0} Nc6 {B 0} 10. Nb3 {B 0} b6 {B 0 Both last book
move} 11. Nd4 {0.03/22 74} Bb7 {0.03/18 123} 12. Nxc6 {0.04/22 34} Bxc6 {0.03/
16 0} 13. Qd4 {0.01/22 68} Qc7 {0.02/17 25} 14. f3 {0.02/21 78} Nd7 {0.00/16 50
} 15. Rfd1 {0.00/22 56} Rfd8 {0.01/17 167} 16. Bf2 {0.00/22 40 (Qb4)} Nc5 {-0.
01/17 16} 17. Qe3 {0.00/21 50 (Bg3)} Bb7 {0.00/17 31 (Bf6)} 18. Bg3 {0.08/21 54
} h6 {0.00/18 47 (Rac8)} 19. h4 {0.07/21 64} Rdc8 {0.00/19 240 (Re8)} 20. Kh1 {
0.05/21 126 (Qd2)} Bc6 {0.00/17 21} 21. Qf2 {0.09/20 46} Be8 {-0.04/17 54 (Nb7)
} 22. Bf4 {0.11/20 28 (Qd4)} Rd8 {-0.07/17 18 (Kh8)} 23. Qg3 {0.03/22 31} Kh8 {
-0.09/17 15} 24. Rd2 {0.05/21 11} Nb7 {-0.05/17 63} 25. Be3 {0.00/22 28} Bf6 {
0.00/17 13} 26. Qf2 {0.00/23 24} Nc5 {0.00/18 15} 27. Bd4 {0.05/22 37} Be5 {0.
00/18 37} 28. Kg1 {0.00/21 127 (g3)} Kg8 {-0.06/14 16} 29. Rdd1 {0.00/20 11
(f4)} Qe7 {0.00/16 31 (Rab8)} 30. g3 {0.00/21 16} Qc7 {0.00/17 17 (Bc6)} 31.
Kg2 {0.12/20 19} Bxd4 {0.00/17 37} 32. Rxd4 {0.12/21 28} Bc6 {0.01/17 11} 33.
Rd2 {0.09/21 22} Nd7 {0.01/18 26} 34. Rad1 {0.11/21 20} Nc5 {0.04/18 25} 35. b3
{0.11/21 13} Nb7 {0.04/18 19 (Rd7)} 36. Qd4 {0.11/21 18} Nc5 {0.05/17 8} 37.
Qe3 {0.11/21 6} Rd7 {0.04/16 16 (Rab8)} 38. Ra1 {0.14/21 17 (Bc4)} Rdd8 {0.04/
18 14} 39. b4 {0.10/22 33 (Re1)} Nb7 {0.05/18 21} 40. Rb1 {0.09/22 16} Rdc8 {
0.04/17 19} 41. Rb3 {0.13/22 69} Rd8 {0.04/17 31} 42. Bd3 {0.13/23 57 (Qf2)}
Be8 {0.01/18 38 (Rdb8)} 43. Bf1 {0.13/23 48} Bc6 {0.07/18 41} 44. Be2 {0.13/22
19 (Nb1)} Rdc8 {0.04/19 40} 45. Bc4 {0.13/22 41 (Qg1)} Rd8 {0.00/19 31} 46. Bf1
{0.13/22 37 (Be2)} Rdc8 {0.04/18 30} 47. Bd3 {0.13/22 18} Be8 {0.00/18 64 (Rf8)
} 48. Nd1 {0.21/22 29 (Ne2)} Rcb8 {0.06/19 33} 49. Rc3 {0.24/23 13} Qd8 {0.12/
19 42} 50. Nb2 {0.15/24 46} a5 {0.15/19 43} 51. b5 {0.15/24 34 (Rb3)} Nc5 {0.
02/18 27} 52. Bc4 {0.22/24 112 (Be2)} Rb7 {0.07/18 20} 53. Bb3 {0.14/24 101}
Rc8 {0.05/18 25} 54. Nd3 {0.18/23 36 (Rd1)} Nxb3 {0.00/19 16} 55. Rxb3 {0.25/
23 44 (Rxc8)} f6 {0.01/17 14} 56. e5 {0.20/24 33 (Nb2)} Bf7 {0.08/16 26 (fxe5)}
57. Rb1 {0.39/22 19} Qe7 {0.10/18 21} 58. exd6 {0.31/24 43} Qxd6 {0.08/19 17}
59. Nf2 {0.32/24 6} Qb8 {0.22/19 100 (Qc5)} 60. Rbd1 {0.64/22 25} Rbc7 {0.15/
19 150} 61. Qe2 {0.26/24 189} Rc5 {0.15/19 18 (e5)} 62. Ne4 {0.23/22 16 (c4)}
Re5 {0.00/17 19} 63. f4 {0.29/22 7} Rd5 {0.00/18 17 (Bh5)} 64. c4 {0.35/22 9}
Rxd2 {0.00/17 0} 65. Rxd2 {0.32/23 12} Rd8 {0.00/19 37} 66. c5 {0.20/23 63}
bxc5 {0.00/19 29} 67. Nxc5 {0.04/23 38} e5 {0.00/19 19} 68. f5 {0.19/22 4} Rd5
{0.00/18 11 (Rxd2)} 69. Rc2 {0.23/19 5 (Rxd5)} Qb6 {0.00/17 53 (Rd4)} 70. Nb3 {
0.23/18 4 (Ne6)} Qd6 {-1.05/15 23 (Rd8)} 71. h5 {0.00/16 4 (Kh2)} Kh7 {-1.48/
15 9 (Qb4)} 72. Qc4 {-0.55/16 2} Qa3 {-2.75/15 44 (Qf8)} 73. Qg4 {-2.14/16 5}
Rd8 {-3.15/17 43} 74. Rd2 {-2.35/16 4} Rf8 {-2.75/14 0} 75. Nxa5 {-2.63/17 4}
Qc3 {-2.75/14 0} 76. Rd7 {-2.43/16 2} Qxa5 {-3.35/15 30} 77. Kh2 {-2.71/16 1
(Qd1)} Qe1 {-3.93/15 19 (Qc3)} 78. Qf3 {-2.78/13 1} e4 {-4.13/13 0 (Qb4)} 79.
Qg2 {-3.23/13 1} Bxh5 {-4.42/16 14 (Qb4)} 80. Qg1 {-3.32/9 0} Qe2+ {-4.22/13 0}
81. Qg2 {-3.16/6 0} Qc4 {-4.22/13 0} 82. g4 {-6.84/22 600} Bf7 {-4.22/12 0} 83.
b6 {-6.80/21 76 (Re7)} Qxa4 {-5.42/13 22 Time} 0-1