Links |
|
Two leading chess programmers (REBEL programmer Ed Schröder and CHESS-TIGER programmer Christophe Théron) have joined forces. Playing Strength [Event "Simultaneous game against Lautier"] [Site "Clichy"] [Date "1999.10.09"] [White "Rebel Tiger"] [Black "GM Joël Lautier"] [Result "1-0"] 1. e4 c5 2. Nf3 Nc6 3. Bb5 d6 4. O-O Bd7 5. Re1 Nf6 6. c3 a6 7. Bxc6 Bxc6 8. d4 cxd4 9. cxd4 Bxe4 10. Nc3 Bxf3 11. Qxf3 e6 12. Qxb7 Qc8 13. Qf3 Be7 14. d5 e5 15. Bg5 O-O 16. Ne4 Nxd5 17. Nxd6 Qe6 18. Rad1 Bxg5 19. Rxd5 Bf6 20. b3 Rad8 21. Qd1 Qd7 22. Qd3 Qc6 23. Rd1 Rd7 24. Ne4 Rxd5 25. Qxd5 Qxd5 26. Rxd5 Be7 27. Rxe5 Rc8 28. Kf1 Bb4 29. Rd5 h6 30. Rd4 Rc1+ 31. Ke2 a5 32. Rc4 Ra1 33. Rc2 f5 34. Nd2 Kf7 35. Nf3 Kf6 36. Kd3 Rf1 37. Ke3 Ke6 38. Nd4+ Ke5 39. f4+ Kd5 40. Nxf5 Bc5+ 41. Kd3 Rxf4 42. Nxg7 Rd4+ 43. Kc3 Rf4 44. Rd2+ Ke5 45. Re2+ Kd5 46. Ne6 Bb4+ 47. Kd3 Rf1 48. Nc7+ Kd6 49. Nb5+ Kc6 50. Nc3 Rc1 51. Rc2 Rh1 52. g3 Rf1 53. Kd4 Bc5+ 54. Ke5 Bd6+ 55. Ke6 Ba3 56. Ne4+ Kb6 57. Nd6 Re1+ 58. Kd7 1-0 Results I have so far at 60 moves in 30 minutes: Tiger - Nimzo 732 15.5 - 8.5 Tiger - Fritz 532 9.5 - 6.5 Tiger - Fritz 6 14.5 - 5.5 Tiger - Junior 5 8.0 - 8.0 Tiger - Hiarcs 732 10.5 - 5.5 Tiger scored 63%, which is roughly 100 Elo above the average of the other 5 programs. That's a huge difference. We'll see how this continues.Enrique's games can be downloaded HERE A NEW CHALLENGE Since the worldchampionships in Paderborn ended I can tell you that I am very positive about the cooperation between Rebel and Chess Tiger. The reputation of Rebel is well known, but I also met with Christophe Theron and in my opinion he is a very talented programmer. I am sure that Chess Tiger has a high potential, the more because this program is already having a very high playing strength. For me this means that - besides my work for Ed and the Rebel openingbook - I have found a new challenge: Help Christophe and Chess Tiger to become a very strong chess program. In Paderborn Chess Tiger was playing with my openingbook and in the future I will also work on optimising the book for the Tiger. Somehow I feel very confident about the future, because with Rebel and Chess Tiger I am able to work for two great chess programs. For Chess Tiger I want to share my chess knowledge to improve the playing strength of this chessprogram, and I am also happy that my future improved books will be a part of the Tiger. With these joint forces I am sure that we will be able to improve a lot. First goal is to participate in the open Dutch championships, later this year. Of course I cannot wait this event coming closer and closer! This will be the first test after the WC in Paderborn. All in all I think this cooperation (between Christophe, Ed and me) can lead to newer, better and stronger engines. Of course we'll have to wait what the future brings us, but as far as I am concerned, it is going to be a bright one! Jeroen Noomen Thoralf Karlsson chairman of the SSDF (November '99) THE SSDF RATING LIST 1999-11-26 69727 games played by 202 computers Rating + - Games Won Oppo ------ --- --- ----- --- ---- 1 Chess Tiger 12.0 DOS 128MB K6-2 450 MHz 2696 44 -40 317 72% 2533 2 Fritz 5.32 128MB K6-2 450 MHz 2671 45 -41 297 72% 2506 3 Nimzo 7.32 128MB K6-2 450 MHz 2663 37 -35 409 69% 2526 4 Nimzo 99 128MB K6-2 450 MHz 2644 52 -48 214 67% 2520 5 Hiarcs 7.32 128MB K6-2 450 MHz 2636 42 -39 320 67% 2509 6 Junior 5.0 128MB K6-2 450 MHz 2619 54 -50 190 65% 2508The good result for Chess Tiger 11.8 on P90 made us very curious about what Christophe Therons latest program could achieve on faster hardware. Now we know! After 317 tournament games Chess Tiger 12.0 DOS K6-2 450 MHz has taken the first place on the rating list! It's rating is 2696, which is 25 points more than Fritz 5.32 has on the same hardware! Very impressing! As can be seen SSDF has played with a DOS-version of Chess Tiger. Will the final commercial Windows-version Rebel-Tiger be slower and therefore weaker? No, surprisingly the opposite seems to be the case! Windows does slow the program down with 2-3%, but since the Windows compiler is better than the DOS-compiler, the net effect is a speed advantage for the Windows version. Our preliminary tests with a beta-version indicates that the Windows version is about 8-10% faster at three minutes per move. Related Links |