Author: Chris Carson
Date: 12:58:33 05/07/02
Go up one level in this thread
On May 07, 2002 at 13:26:57, Christophe Theron wrote:
>On May 07, 2002 at 00:07:33, K. Burcham wrote:
>
>>
>>Gambit Tiger 2.0
>>2x1533 amd mp
>>1 gig ram
>>192 megs hash
>>Tiger.ctg book
>>default settings
>>
>>[Event "ICC 90 30"]
>>[Site "Internet Chess Club"]
>>[Date "2002.05.06"]
>>[Round "2"]
>>[White "xracer"]
>>[Black "Petrovich"]
>>[Result "0-1"]
>>[ICCResult "White resigns"]
>>[WhiteElo "2580"]
>>[BlackElo "2529"]
>>[Opening "French: Tarrasch, Botvinnik variation"]
>>[ECO "C05"]
>>[NIC "FR.17"]
>>[Time "20:59:43"]
>>[TimeControl "5400+30"]
>>
>>1. e4 e6 2. d4 d5 3. Nd2 Nf6 4. e5 Nfd7 5. Bd3 c5 6. c3 b6 7. Nh3 Ba6 8.
>>Bxa6 Nxa6 9. O-O Nc7 10. Qg4 Nb8 11. Nf3 h6 12. Nf4 g6 13. Be3 Nc6 14. Rfc1
>>c4 15. b3 b5 16. Rc2 Be7 17. Rb1 Rb8 18. Rd1 Kd7 19. Ne1 Qg8 20. Ra1 Qh7 21.
>>Qe2 Rhg8 22. b4 a5 23. a3 Ra8 24. Rca2 Ra7 25. f3 Rga8 26. Nh3 Ra6 27. Nf2
>>R8a7 28. Qb2 Rb7 29. Qd2 Rba7 30. Qe2 Qh8 31. Nc2 axb4 32. Nxb4 Ra4 33. Nxc6
>>Kxc6 34. Bc1 Qf8 35. Ng4 h5 36. Nf6 Bxa3 37. Bxa3 Rxa3 38. Qc2 Qa8 39. Rxa3
>>Rxa3 40. Rb1 Qa5 41. Rc1 Ra2 42. Qb1 Qa3 43. h3 Rb2 44. Qa1 b4 45. cxb4 Qxa1
>>46. Rxa1 c3 47. Rc1 Nb5 48. Kf1 Rxb4 49. Ke2 Rxd4 50. Ke3 Rb4 51. Kd3 Rb2
>>52. Rc2 Rb1 53. Rxc3+ Nxc3 54. Kxc3 Rg1 55. g4 Rg3 56. gxh5 Rxf3+ 57. Kd4
>>Rxf6 58. exf6 gxh5 59. Ke3 e5 60. Kd3 Kd7 61. Ke2 Ke6 62. Kf3 Kxf6 63. Kg3
>>Kg5 64. Kf3 Kh4 65. Kg2 f5 {White resigns} 0-1
>>
>>
>>Petrovich(GM) tells you: :)
>>/tell petrovich good game
>>Petrovich(GM) tells you: yes
>>.i think you played very high level game with no errors
>>(told Petrovich)
>>Petrovich(GM) tells you: yes
>>Petrovich(GM) tells you: you played without plan
>>Petrovich(GM) tells you: Ra1-b1-d1-a1
>>Petrovich(GM) tells you: Ne1?
>>Petrovich(GM) tells you: only active plan was h2-h4
>>Petrovich(GM) tells you: to break kingside]
>>.which program does not know
>>(told Petrovich)
>>Petrovich(GM) tells you: otherwise Black has comfortable play
>>Petrovich(GM) tells you: b3-b4?
>>Petrovich(GM) tells you: you missed knight transfer into c2
>>Petrovich(GM) tells you: to protect Ra1 and to hold a-file
>>Petrovich(GM) tells you: understand me?
>>.yes
>>Petrovich(GM) tells you: when shall we play 3rd game?
>>.thursday? ok with you
>>(told Petrovich)
>>Petrovich(GM) tells you: okay
>>
>>
>>this was a completely different game compared to the first game.
>>with Petrovich knowing the french very well, he closed the position.
>>once the a-file was open, eval dropped.
>>I see two problems here,
>>1. this opening against a strong player with the lines that are in this book
>>2. once the position was closed, the program made no improvements in its
>>position with waisted moves. this gave the GM time to set up on the a-file with
>>both rooks and queen. once set up, the attack started and the game was over.
>>maybe this was the type of game that anti-human should have been on. not sure.
>>kburcham
>
>
>Yes, maybe antihuman would have helped.
>
>I think the best use of antihuman is to set it on DURING the game if the
>operator notices that the opponent tries to close the position.
>
>Naturally that requires an external human intervention, so I'm not sure it is
>fair.
>
>Actually I would have to build this detection logic in the program itself. There
>is already one, but it is aimed towards very obviously closed games. It would
>not detect a more subtle buildup.
>
>
>
> Christophe
Thanks Christope, The match is tied at 1-1, I am hopeful that GTiger can
improve on its score, the games are very good/entertaining. Keep up the great
work on the detection code.
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.