Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: REBEL 10.0 vs CM5500

Author: Timothy J. Frohlick

Date: 14:44:35 11/21/98

Go up one level in this thread


On November 21, 1998 at 10:40:29, James T. Walker wrote:

>Hello,
>This is game 11 of my 12 game match.  CM5500 now leads 6-5.  I've been bored
>with this match from the start but I'm beginning to develope an appreciation
>for "positional" chess.  I prefer the fireworks of Fritz/Junior but this game
>was interesting to watch the score climb for CM5500.  I dislike the sac of the
>Bishop by CM5500 in the ending but this is becoming "Normal" for programs.
>If it leads to a quicker win, then by all means, but I saw no evidence that
>winning without the Bishop was faster than winning with it.  Still an enjoyable
>game.
>
>CM5500 AMD K6-2 350 MHZ 64 MEG HASH
>REBEL 10.0 PII 333 MHZ 100 MEG HASH
>
>
>[Event "12 GAME MATCH @ 40/90"]
>[Site "FL"]
>[Date "11.21.1998"]
>[Round "11"]
>[White "CM5500"]
>[Black "REBEL 10.0"]
>[Result "1-0"]
>[WhiteElo "2550"]
>[BlackElo "2535"]
>[ECO "A35"]
>
>1. c4 c5 2. Nc3 Nc6 3. Nf3 Nd4 4. e3 Nxf3+ 5. Qxf3 g6 6. d4 Bg7 7. dxc5
>Bxc3+ 8. bxc3 Qa5 9. Rb1 Qxa2 10. Bd3 f5 11. e4 fxe4 12. Qxe4 Nf6 13. Qd4
>O-O 14. O-O Qa5 15. Re1 Rf7 16. Bg5 Qd8 17. Be2 Qf8 18. Bf3 e6 19. Bxb7
>Bxb7 20. Rxb7 Qc8 21. Rb5 a6 22. Rb6 Ne8 23. Bf4 a5 24. Reb1 Nc7 25. Qe4
>Na6 26. Bd6 a4 27. Ra1 Kg7 28. Rxa4 h6 29. Qb7 Qxb7 30. Rxb7 Kf6 31. Rb6
>Nc7 32. Rxa8 Nxa8 33. Ra6 g5 34. Rxa8 Kf5 35. Kf1 h5 36. Rc8 h4 37. Rg8
>Rh7 38. h3 Kf6 39. Ke2 Kf7 40. Rxg5 Kf6 41. Rg4 Kf7 42. Rf4+ Kg6 43. Kf3
>Kg5 44. Ke4 Kg6 45. Be5 Re7 46. Rg4+ Kf7 47. Rxh4 Kg6 48. Rg4+ Kf7 49.
>h4 Ke8 50. Rg8+ Kf7 51. Rg7+ Kf8 52. h5 Rxg7 53. Bxg7+ Kxg7 54. g4 Kf6
>55. h6 Kg6 56. g5 Kh7 57. f4 Kg6 58. f5+ exf5+ 59. Kf4 Kf7 60. Kxf5 Kg8
>61. Kf6 Kh8 62. g6 Kg8 63. h7+ Kh8 64. Kf7 d6 65. g7+ Kxh7 66. g8=Q+ Kh6
>67. Qg6# *
>Jim Walker
In reply:

What do you think would happen if you ran Rebel 10 on the K6-350Mhz vs CM5500 on
the P II 333Mhz?  Do you think that a 30% increase in Rebels' play would occur?
You should try it and then average the results between the two series of games.
You should also switch the hash table levels so that Rebel 10 gets 60Mb and
CM5500 gets 100Mb.  It is obvious that these are both excellent programs.  When
we find out the optimal settings for Rebel 10 you could repeat the games on
identical computers.

Respectfully,

Tim Frohlick



This page took 0 seconds to execute

Last modified: Thu, 15 Apr 21 08:11:13 -0700

Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.