Author: Dusan Dobes
Date: 00:49:02 02/10/99
Go up one level in this thread
On February 09, 1999 at 12:44:44, Robert A. Basham wrote: >On February 09, 1999 at 11:51:43, Dusan Dobes wrote: > >>On February 08, 1999 at 16:04:11, Robert A. Basham wrote: >> >>>I am wondering if any chess programs can find this move -- R x d5+ >>> >>>(from this possition). So far, none of my programs can find it! >>>Thanks for trying! >>> >>>regards, >>>Robert >> >>This is Phalanx XXI on a pentium/150, 9MB hash table, opening book off. >>Rxd5+ is found in 0.2 s. >> >>r1bk3r/pp2n2p/2n2N2/1R1p3Q/2p4P/P1p5/2qBBPP1/4K2R w K - 0 1 Rb5xd5 >> 3 -18 63 7956 Rb5xd5 Ne7xd5 Qh5xd5 Kd8-e7 Qd5-c5 Ke7-f7 >> Be2-h5 Kf7-e6 Bd2-g5 Qc2-b1 Bh5-d1 >> 3 -> 0:00.75 9806 0 turns >> 4 42 84 10917 Rb5xd5 !! >> 4 76 303 44839 Rb5xd5 Ne7xd5 Qh5xd5 Kd8-e7 Qd5-c5 Ke7xf6 >> Qc5-g5 Kf6-f7 Be2-h5 Kf7-e6 Bh5-g4 Ke6-f7 >> Bg4-h5 >> 4 -> 0:03.12 46104 0 turns >> 5 75 664 99650 Rb5xd5 Ne7xd5 Qh5xd5 Kd8-e7 Qd5-c5 Ke7-f7 >> Be2-h5 >> 5 -> 0:08.01 127234 0 turns >> 6 80 1640 261359 Rb5xd5 Ne7xd5 Qh5xd5 Kd8-e7 Qd5-c5 Ke7-f7 >> Be2-h5 Kf7xf6 Qc5-d6 Bc8-e6 Bd2-g5 Kf6-g7 >> Bg5-h6 Kg7-f6 >> 6 -> 0:22.61 365129 0 turns >> 7 80 4743 744487 Rb5xd5 Ne7xd5 Qh5xd5 Kd8-e7 Qd5-c5 Ke7-f7 >> Be2-h5 Kf7xf6 Qc5-d6 Bc8-e6 Bd2-g5 Kf6-g7 >> Bg5-h6 Kg7-f6 Bh6-g5 >> 7 -> 1:16.28 1223068 0 turns >> 8 80 19504 3035751 Rb5xd5 Ne7xd5 Qh5xd5 Kd8-e7 Qd5-c5 Ke7xf6 >> Qc5-g5 >> 8 -> 4:41.55 4377025 0 turns >> 9 137 58595 8829483 Rb5xd5 Ne7xd5 Qh5xd5 Kd8-e7 Qd5-c5 Ke7xf6 >> Qc5-g5 Kf6-f7 Be2-h5 Kf7-e6 Bh5-g4 Ke6-f7 >> Qg5-d5 Kf7-f8 Bd2-h6 Kf8-e8 Bg4-h5 Qc2-g6 >> Bh5xg6 Ph7xg6 Bh6-g7 Pc3-c2 Qd5xc4 >> 9 -> 14:25.86 13320376 0 turns >>Depth=9, Value=137, Time=865.86, Last turn=0.20, Nodes=13320376, N/s=15383 >>PV = Rb5xd5 Ne7xd5 Qh5xd5 Kd8-e7 Qd5-c5 Ke7xf6 Qc5-g5 Kf6-f7 >> Be2-h5 Kf7-e6 Bh5-g4 Ke6-f7 Qg5-d5 Kf7-f8 Bd2-h6 Kf8-e8 >> Bg4-h5 Qc2-g6 Bh5xg6 Ph7xg6 Bh6-g7 Pc3-c2 Qd5xc4 >>[ Solution correct 100.00%, 2/2 ] >> >>Dusan Dobes >------------------------ >Your test results are excellent..! Thanks! That is what I was looking for! >I will print this out. They are about the same as what Crafty shows... > >Regards, Robert This is a typical third repetition draw that is forced by one side, until depth=8; it turns to be won at depth=9. Phalanx has speculative code that tries to detect the forcing side. The evaluation is not just "draw" in this case but "at least draw", that means white can avoid repetition somewhere in the forced line and black can't - and black king's position is open. So, i add positive evaluation to the draw score. I was inspired by Botvinnik's comments to his game, where Botvinnik sacrificed minor piece to reach an unclear position where draw was guaranteed for him. I forgot who/when played the game, i only remember the Botvinnik's idea. Dusan
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.