Author: Paul Jacobean Sacral
Date: 19:03:08 10/20/05
Go up one level in this thread
On October 20, 2005 at 11:46:34, Robert Hyatt wrote: >I first thought I had some sort of bug, but after playing back through the >position, this is less a bug and more just missing deep tactics. Would be >interesting to see what kinds of scores at different depths other programs see >starting at move 66... I got similar results like those already posted with Hiarcs 9 and Shredder 9, both preferring 65.Reb1 etc. Both engines agree on 67.Ke3 being better, keeping white's clear advantage as it seems, than 67.Rxc5 capturing a poisonous pawn, whereafter the evaluations drop, and on 69.Ra5 as a considerable alternative to 69.Ke3 (for instance 69...Qxa5 70.Qxf2), although the position remains being a minefield. Here is the "fresh" engine output after deleting learnfiles and hash: P4/3.0 GHz, 128 MB hash each. Crafty 20.1 - unknown, Blitz:5' [D]r5k1/6p1/3P1p2/1Rp1pP2/q1P1Q2P/2PK4/5rP1/2R5 w - - 0 1 Analysis by Hiarcs 9: 67.Rxc5 Qa2 68.Qe3 Rxg2 +- (3.38) Depth: 1/2 00:00:00 +- (2.34) Depth: 3/11 00:00:00 67.Qd5+ Kh7 68.Ke3 Qa2 69.Re1 Rxg2 +- (3.02) Depth: 3/11 00:00:00 +- (1.68) Depth: 5/18 00:00:00 17kN 67.Rxc5 Qa2 68.Qe1 Rd8 69.Rd5 Rxg2 +- (1.73) Depth: 5/18 00:00:00 20kN +- (2.04) Depth: 5/18 00:00:00 21kN 67.Ke3 Qa2 68.Qxa8+ Qxa8 69.Kxf2 Qe4 70.g3 Qd3 71.Rcb1 Qd2+ 72.Kf1 Qxd6 +- (2.08) Depth: 5/18 00:00:00 28kN +- (1.42) Depth: 9/28 00:00:04 859kN 67.Rxc5 Qa2 68.Qe1 Rd8 69.Ke3 Rf4 70.g3 Rxf5 71.Ke4 g6 72.Ke3 Rxd6 +- (1.43) Depth: 9/33 00:00:10 2258kN ² (0.70) Depth: 10/33 00:00:17 3931kN 67.Ke3 Rf4 68.Qd5+ Kh7 69.g3 Rxc4 70.Rb7 Qa3 71.Kd2 Qa2+ 72.Rc2 Rd4+ 73.Qxd4 cxd4 74.Rxa2 ² (0.70) Depth: 10/33 00:00:18 4038kN +- (1.41) Depth: 11/34 00:00:59 12358kN Crafty 20.1 - unknown, Blitz:5' [D]3r2k1/6p1/3P1p2/2R1pP2/2P4P/2PK4/q4rP1/2R1Q3 w - - 0 1 Analysis by Hiarcs 9: 69.Rd5 g6 +- (1.79) Depth: 1/14 00:00:00 ± (1.16) Depth: 4/14 00:00:00 69.Rc6 Rxg2 70.Ke3 Rg4 ± (1.29) Depth: 4/16 00:00:00 ² (0.55) Depth: 6/17 00:00:00 10kN 69.Ke3 Rf4 70.g3 Rxf5 71.Ke4 g6 ² (0.70) Depth: 6/18 00:00:00 18kN ² (0.68) Depth: 7/19 00:00:00 41kN 69.Rc6 Rxg2 70.Rb6 Rh2 71.Qe3 Rc8 72.Rb4 e4+ 73.Kxe4 ² (0.69) Depth: 7/19 00:00:00 52kN -+ (-1.87) Depth: 8/25 00:00:01 345kN 69.Ke3 Rf4 70.g3 Rxf5 71.Qe2 Qa7 72.Ke4 Qxc5 73.Kxf5 Qc8+ 74.Ke4 Rxd6 -+ (-1.86) Depth: 8/25 00:00:02 499kN = (-0.20) Depth: 8/25 00:00:02 570kN 69.Ra5 Rxd6+ 70.Ke3 Qb2 71.Rb1 Qc2 72.Rd5 Rxd5 73.cxd5 Rxg2 74.Rb8+ Kh7 75.Rb4 Qxf5 76.d6 = (-0.20) Depth: 8/26 00:00:03 888kN ³ (-0.66) Depth: 11/34 00:00:31 8008kN 69.Ke3 Rxd6 70.Qxf2 Rd3+ 71.Kxd3 Qxf2 72.Rd5 Qxf5+ 73.Kd2 Qf2+ 74.Kd3 Qxg2 75.Rd8+ Kh7 76.Rd1 Qg3+ 77.Kd2 Qh2+ 78.Kd3 ³ (-0.65) Depth: 11/34 00:00:36 9543kN ³ (-0.62) Depth: 11/34 00:00:58 14628kN Crafty 20.1 - unknown, Blitz:5' [D]r5k1/6p1/3P1p2/1Rp1pP2/q1P1Q2P/2PK4/5rP1/2R5 w - - 0 1 Analysis by Shredder 9: 67.Qd5+ Kh8 +- (2.63) Depth: 2/8 00:00:00 +- (2.63) Depth: 3/8 00:00:00 67.Ke3 Rf4 +- (1.82) Depth: 3/8 00:00:00 ± (1.03) Depth: 14/41 00:00:46 17572kN Crafty 20.1 - unknown, Blitz:5' [D]3r2k1/6p1/3P1p2/2R1pP2/2P4P/2PK4/q4rP1/2R1Q3 w - - 0 1 Analysis by Shredder 9 SE: 69.Rd5 Rf4 70.Rd1 Qxc4+ 71.Kc2 Rxf5 ³ (-0.48) Depth: 3/7 00:00:00 µ (-0.74) Depth: 3/13 00:00:00 69.Rc6 Rxg2 70.Rb6 Rg4 71.Rd1 Qxc4+ 72.Kc2 Re4 73.Qg3 Qa2+ 74.Kc1 Re2 µ (-0.73) Depth: 3/13 00:00:00 -+ (-1.71) Depth: 9/33 00:00:01 442kN 69.Ra5 Qxa5 70.Qxf2 Qa4 71.c5 -+ (-1.70) Depth: 9/33 00:00:02 902kN µ (-0.90) Depth: 9/33 00:00:02 1021kN 69.Ke3 Rf4 70.g3 Rxf5 71.Qe2 Qa3 72.Ke4 Qxc5 73.Kxf5 Qxd6 74.Kg4 Qd7+ 75.Kf3 Qf5+ 76.Kg2 Rd3 µ (-0.89) Depth: 9/33 00:00:02 1058kN -+ (-1.68) Depth: 10/37 00:00:05 2341kN 69.Rc6 Rxg2 70.Ke3 -+ (-1.67) Depth: 10/37 00:00:05 2401kN -+ (-1.67) Depth: 10/37 00:00:06 2464kN 69.Ra5 Qxa5 -+ (-1.66) Depth: 10/37 00:00:06 2585kN ³ (-0.56) Depth: 15/41 00:00:47 20123kN The FEN show move no. 1 each, but actually the analysis was always started from within the complete notation. Might matter for being reproduceable on other computers. Yours truly Paul J. Sacral
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.