Author: Vasik Rajlich
Date: 08:14:22 12/02/04
Go up one level in this thread
On December 02, 2004 at 05:06:50, Uri Blass wrote:
>On December 02, 2004 at 04:41:58, Vasik Rajlich wrote:
>
>>On December 01, 2004 at 16:27:20, Peter Berger wrote:
>>
>>>aka round 2 ;)
>>>
>>>The complicated and exciting Hertneck variation of the French Winawer is
>>>certainly very tough to understand for the computers.
>>>
>>>Movei out of book at move 15 chose 15. O-O, which is extremely untopical. Only
>>>15. h4 can be the correct and logical move here, but I guess very few programs
>>>will choose this move without book.
>>>
>>>movei00_8_295 - Junior 8.0.0.2, Computer chess game 2004
>>>[D]r1b1qrk1/pp4pp/2n1p1p1/3pP1N1/2pP2Q1/P1P1B3/2P2PPP/R3K2R w KQ - 0 1
>>>
>>>Junior maybe manouvred a bit carelessly in the opening, 20. ..Qf5 should be ?! .
>>>
>>>A nice move by movei was 22. g4:
>>>
>>>movei00_8_295 - Junior 8.0.0.2, Computer chess game 2004
>>>[D]r3r1k1/p2bn1p1/1p2p1pp/3pPq2/2pP1P1Q/P1P1B3/1RPN2PP/5RK1 w - - 0 1
>>>
>>>I won't comment on the following manouvres, some of them a bit painful to watch,
>>>but not decisive at all -let's move on to the critical point:
>>>
>>>movei00_8_295 - ba2axf, Computer chess game 2004
>>>[D]2q2r2/p3nrpk/1p2p1pp/3pP3/b1pP1PPN/P1P1B3/1RP3QP/3R3K w - - 0 1
>>>
>>>37. g5 ? - ouch.
>>>
>>>The first exchange sac is very typical and probably easy for humans (and
>>>computers?) 38. ..Rxf4
>>>
>>>[D]movei00_8_295 - Junior 8.0.0.2, Computer chess game 2004
>>>2q2r2/p3nrpk/1p2p1p1/3pP1Pp/b1pP1P1N/P1P1B3/1RP3QP/1R5K b - - 0 1
>>>
>>>But the second one is just great! 41. ...Rxh4!
>>>
>>>movei00_8_295 - Junior 8.0.0.2, Computer chess game 2004
>>>[D]5q2/p3n1pk/1p2p1p1/3pP1Pp/b1pP1r1N/P1P3QP/1RP5/1R5K b - - 0 1
>>>
>>>The position after black's 46th move - that's the dream of any black Caro-Kann
>>>player with 3. ...g6 come true :
>>>
>>>movei00_8_295 - Junior 8.0.0.2, Computer chess game 2004
>>>[D]8/p5pk/1p2p1p1/3pPn1p/b1pP1q2/P1P4P/1RP5/1R2Q2K w - - 0 1
>>>
>>>Junior took the lead by 2-0 in the match with this great game.
>>>
>>>Junior 8.0.0.2 , P223 MMX, 16 MB Hash, 3-4 men tablebases
>>>Movei , Athlon 64 3200+(2.0GHz), 64 MB Hash
>>>
>>>Game in 2 hours, 10 seconds increment/move
>>>
>>>[Event "Computer chess game"]
>>>[Site "XP3200"]
>>>[Date "2004.12.01"]
>>>[Round "2"]
>>>[White "movei00_8_295"]
>>>[Black "Junior 8.0.0.2"]
>>>[Result "0-1"]
>>>[TimeControl "7200+10"]
>>>
>>>1. e4 e6 2. d4 d5 3. Nc3 Bb4 4. e5 c5 5. a3 Bxc3+ 6. bxc3 Ne7 7. Qg4 O-O 8.
>>>Bd3 Nbc6 9. Qh5 Ng6 10. Nf3 Qc7 11. Be3 c4 12. Bxg6 fxg6 13. Qg4 Qf7 14.
>>>Ng5 Qe8 15. O-O h6 16. Nf3 Qf7 17. Rab1 b6 18. Nd2 Bd7 19. f4 Ne7 20. Rb2
>>>Qf5 21. Qh4 Rfe8 22. g4 Qf8 23. Ra1 Ba4 24. Kh1 Rad8 25. Qg3 Nc6 26. Nf3
>>>Kh7 27. Qh3 Rd7 28. Nh4 Rf7 29. Rba2 Qe7 30. Rg1 Nd8 31. Bc1 Qd7 32. Qg2
>>>Nc6 33. Be3 Ne7 34. Rb2 Rd8 35. Rgb1 Rdf8 36. Rd1 Qc8 37. g5 h5 38. Rdb1
>>>Rxf4 39. Bxf4 Rxf4 40. Qg3 Qf8 41. h3 Rxh4 42. Qxh4 Qf3+ 43. Kh2 Nf5 44.
>>>Qe1 Qf4+ 45. Kg1 Qxg5+ 46. Kh1 Qf4 47. Kg1 Qf3 48. Kh2 h4 49. Rc1 g5 50.
>>>Rbb1 Ne3 51. Qd2 Qg3+ 52. Kh1 Qxh3+ 53. Kg1 Qg4+ 54. Kh2 Qg3+ 55. Kh1 h3
>>>56. Re1 Qf3+ 57. Kg1 Ng4 58. Re2 Qg3+ 59. Kh1 Be8 60. Rf1 Bg6 61. Rf8 Be4+
>>>62. Rxe4 dxe4 63. Qe2 g6
>>>{Movei->white resign} 0-1
>>
>>Nice report.
>>
>>Also worth pointing out is white's strange use of rooks - doubled on the b-file,
>>rather than somewhere on the f&g files. It's hard to get an engine to put its
>>rooks on the right closed files.
>>
>>Vas
>
>What is the definition of closed files?
>By movei's definition closed files are files that are not open and
>open files are files when there is no friendly pawn that prevent the rook to
>go forward(there may be pieces or pawns of the opponent that prevent it to go
>forward but if you remove all pieces except friendly pawns the rook can go to
>the 8th rank).
>
>By movei's definition the b file is an open file because there are no friendly
>pawns in that file.
>
>Movei has a bonus of 0.1 pawns for rook on open file and panelty of 0.1 for
>closed file and I am not sure if that bonus and panelty is productive.
>
>It was productive in the positions that I tested at the time that I implemented
>it but I did not play enough games and it is possible that the positions were
>misleading and in different positions I could get different results.
>
>Uri
Well, it's not an easy problem.
One thing I noticed quite some time ago from watching Rybka play is that when a
file is totally open (ie. no pawns for either side), it is very rarely a
strategic mistake to move a rook to it. This doesn't quite apply to a half-open
file, which is much more likely to be useless.
Of course this is just a cheap observation - better to "do it right" and try to
evaluate where the rook really has prospects based on the position.
Vas
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.