Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: WCCC 2002: Round 2 - WARP vs. DIEP /Including testposition(s)

Author: Uri Blass

Date: 16:17:05 07/07/02

Go up one level in this thread


On July 07, 2002 at 17:18:02, Uri Blass wrote:

>On July 07, 2002 at 15:34:59, Peter Berger wrote:
>
>>On July 07, 2002 at 08:38:32, Eduard Nemeth wrote:
>>
>>[Event "WCCC"]
>>[Site "Maastricht"]
>>[Date "2002.07.07"]
>>[Round "2"]
>>[White "WARP"]
>>[Black "Diep"]
>>[Result "1/2-1/2"]
>>[ECO "A33"]
>>[PlyCount "116"]
>>[EventDate "2002.??.??"]
>>
>>1. c4 c5 2. Nc3 Nf6 3. Nf3 Nc6 4. d4 cxd4 5. Nxd4 e6 6. g3 Qb6 7. Nb3 Ne5 8. e4
>>Bb4 9. Qe2 d6 10. f4 Nc6 11. Be3 Bxc3+ 12. bxc3 Qc7 13. Bg2 b6 14. O-O O-O 15.
>>g4 Bb7 16. g5 Nd7 17. Rf3 Ne7 18. Rh3 Rfc8 19. Qh5 Nf8 20. Nd2 Neg6 21. Rf1 Qc6
>>22. Rf2 Qd7 23. f5 exf5 24. exf5 Bxg2 25. Kxg2 Ne5 26. Bd4 Re8 27. Kh1 Rac8 28.
>>f6 Ng4 29. Rg2 Re1+ 30. Bg1 Ne5 31. Qh4 Rd1 32. fxg7 Kxg7 33. Ne4 Qf5 34. Qh6+
>>Kh8 35. Nxd6 Rxd6 36. Qxd6 Ne6 37. Rhg3 Rd8 38. Qe7 Rd1 39. Qe8+ Kg7 40. Qa4
>>Qb1 41. Qb3 Qc1 42. c5 Nf4 43. Rc2 Rxg1+ 44. Rxg1 Qe3 45. Qb1 Qf3+ 46. Rgg2
>>Nxg2 47. Rxg2 Nd3 48. Qg1 bxc5 49. h4 Nf4 50. Qf2 Qxg2+ 51. Qxg2 Nxg2 52. Kxg2
>>Kg6 53. Kf3 Kf5 54. c4 a5 55. a4 Ke5 56. Ke3 Kf5 57. Kf3 Ke5 58. Ke3 Kf5
>>1/2-1/2
>>
>>I think your analysis misses a few interesting turning points in this game - at
>>least one of them should be good for a testposition.
>>
>>Diep played on a Dual AMD K7 in this game, so no really big hardware although
>>quite OK of course :).
>>
>>The position after opening and in early middlegame was unbalanced.
>>
>>At move 31 the position is about equal with probably a little advantage for
>>Diep.
>>
>>Then Diep commited a horrible blunder 31...Rd1?? which should have lost the
>>game:
>>
>>WARP - Diep
>>[D]2r2nk1/p2q1ppp/1p1p1P2/4n1P1/2P4Q/2P4R/P2N2RP/4r1BK b - - 0 1 am Rd1
>>
>>This one could be difficult for some other engines, too.
>
>Here is the end of the analysis of movei at depth 11
>
>11 -26 83673 127553954 a7a5 h4e1 d7h3 g1d4 f8d7 e1e4 h3h5 g2g1 a5a4 f6g7 g8g7
>d2f3
>11 -25 101403 153850144 e5g6
>11 -12 104045 158135756 e5g6 h4e1 d7h3 g1d4 g7f6 g5f6 h3d7 e1f2 f8e6 f2f5 d7a4
>f5b1
>11 -11 110510 168071972 c8d8
>11 -8 117392 178523348 c8d8 f6g7 g8g7 h4e1 d7h3 g1d4 h3e6 e1e4 g7g8 e4b7 e5c6
>b7c6 e6e1 g2g1 e1d2
>11 -7 145975 222148125 g8h8
>11 -6 153489 233612692 g8h8 h4e1 d7h3 f6g7 h8g7 g1d4 h3e6 d2e4 f8d7 e4f6 d7c5
>f6e4 c5d3
>11 -6 153489 233612692 g8h8 h4e1 d7h3 f6g7 h8g7 g1d4 h3e6 d2e4 f8d7 e4f6 d7c5
>f6e4 c5d3
>
>
>>
>>You mentioned the next one, the awful 38. Qe8+?? which in my opinion already
>>spoils the game to a draw ( although this would be very difficult to prove of
>>course)
>>
>>WARP - Diep
>>[D]3r3k/p4p1p/1p1Qn3/4nqP1/2P5/2P3R1/P5RP/6BK w - - 0 1 am Qe8+
>>
>>This is probably an engine-specific problem, other engines probably have no
>>problems here. Easiest looks Qf6+ with a sure win.
>>
>>Without diagram another maybe interesting one:
>>
>>WARP - Diep
>>8/p4pkp/1p2n3/4n1P1/Q1P5/2P3R1/P5RP/1q1r2BK w - - 0 1 am Qb3
>>
>>Qb3 ?! isn't really a bad move but it looks quite senseless. If white wants to
>>play c5 why not do it right now ? After Qb3 the queen is at a worse position for
>>c5.
>
>Movei considers c5 as the best move
>
>9 77 2087 3239103 a4a7 e6f4 a7b7 f4g2 b7g2 d1e1 g2a8 e5c4 a8g2
>9 78 3038 4800466 c4c5
>9 151 4927 7680207 c4c5 e6c5 a4f4 d1e1 f4f6 g7g8 f6d8 g8g7 d8e7 b1e4 e7f6 g7g8
>f6d8 g8g7
>9 151 6350 10033388 c4c5 e6c5 a4f4 d1e1 f4f6 g7g8 f6d8 g8g7 d8e7 b1e4 e7f6 g7g8
>f6d8 g8g7
>10 144 7717 12179728 c4c5 e6c5 a4f4 d1e1 f4f6 g7g8 g5g6 f7g6 f6d8 g8g7 d8e7 e5f7
>e7a7 g7f8
>10 144 13781 21733749 c4c5 e6c5 a4f4 d1e1 f4f6 g7g8 g5g6 f7g6 f6d8 g8g7 d8e7
>e5f7 e7a7 g7f8
>
>
>>
>>
>>But then there was also a rather brilliant move that might be a good
>>testposition IMHO:
>>
>>WARP - Diep
>>[D]8/p4pkp/1p2n3/2P1n1P1/8/1QP3R1/P5RP/2qr2BK b - - 0 1 bm Nf4
>>
>>While the obvious 42. .. Nxc5 ?! might have been sufficient to hold after a long
>>and dangerous fight, 42...Nf4 !! is clearly the best move, pretty, surprising,
>>witty and a clear draw although very deep.
>>The mainline to get this with drawscore should be something like 20 ply. Can
>>your engine find this one, too ?
>
>Movei00_72f says that it is only 10 plies to find the move and not 20 plies.
>finding a draw score may be a different story
>
>10 -125 5262 8544649 e6c5 b3b4 g7g8 b4b5 g8f8 b5e2 d1e1 e2h5 c5e4 h5h6 f8g8
>10 -124 9464 15267487 b6c5
>10 -122 13314 21308383 b6c5 b3b5 e6f4 b5c5 d1d5 c5a7 f4g2 h1g2 c1c2 g1f2 e5d3
>a7b6 c2a2
>10 -121 15280 24563000 e6f4
>10 -81 17115 27474407 e6f4 g2c2 c1a1 c3c4 a1d4 b3e3 b6c5 e3d4 c5d4 c4c5
>10 -81 18411 29649175 e6f4 g2c2 c1a1 c3c4 a1d4 b3e3 b6c5 e3d4 c5d4 c4c5
>11 -87 22910 36681099 e6f4 g2c2 c1a1 b3b2 a1b2 c2b2 b6c5 h2h4 g7g6 b2b7 f4e2
>g3g2 e2c3 b7a7
>
>Uri

The real problem is finding Rxg1+ and not finding Nf4

Movei needs depth 12 and more than 20 minutes to find it.

11 -79 27842 45833908 c1a1 b3b2 a1b2 c2b2 b6c5 h2h3 g7g6 h1h2 f4h5 g1c5 h5g3
h2g3 g6g5 c5a7
12 -84 39470 65881027 c1a1 b3b2 a1b2 c2b2 b6c5 b2f2 e5d3 f2f3 g7g6 g3g4 g6f5
g4g2 c5c4
12 -83 81790 134888247 e5d3
12 -75 109220 178474497 e5d3 c3c4 c1a1 h2h3 f4h3 b3c3 a1c3 c2c3 d3f2 h1h2 h3g1
g3g1 d1g1 h2g1 f2e4 c3a3 b6c5 a3a7 e4g5
12 -74 145026 234770439 d1g1
12 -39 169803 276443079 d1g1 g3g1 c1e3 b3b1 e3e4 c2g2 e5d3 b1f1 g7g8 h2h3 f4h3
h1h2 h3g1 h2g1 e4e3 g1h1 b6c5
12 -39 172812 281897501 d1g1 g3g1 c1e3 b3b1 e3e4 c2g2 e5d3 b1f1 g7g8 h2h3 f4h3
h1h2 h3g1 h2g1 e4e3 g1h1 b6c5

Uri



This page took 0.11 seconds to execute

Last modified: Thu, 07 Jul 11 08:48:38 -0700

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