Author: Uri Blass
Date: 04:57:57 01/02/02
Go up one level in this thread
On January 02, 2002 at 05:44:15, Gian-Carlo Pascutto wrote: >On January 02, 2002 at 01:47:26, Timothy J. Frohlick wrote: > > >>It seems that finally the score is nearly a point higher after over ten fold the >>number of moves searched even though Rd1-d8 was considered early on. > >The trick is that the threat after Rdd8 is hidden. White will >mate but needs several quiet moves to set up the mate. This will >kill most overselective searchers. > >[D]1R1R4/k3rp1p/1p6/pP4p1/1P6/4qPB1/6PK/8 b - - 0 35 > >How long before you engine sees it's already over? > >-- >GCP My weak engine is not selective(it only uses check extension and alpha beta with no null move pruning) but even at depth 9 after some minutes and more than 20M nodes it cannot see that it is already over. main line at depthes 6-9 Re5 2.Ra8+ Kb7 3.Rdb8+ Kc7 4.Re8 f6 5.Ra7+ Kd6 6.Bxe5 fxe5 7.bxa5 bxa5 8.Rxh7 score 1.2 for white that does not mean that the game is over and the line leads to a draw. only at depth 10 after 128124097 nodes it can start to see problems and the score for Re5 is -1.85 when the main line begins with 1...Re5 2.Rbc8 Kb7 3.Rc6 f5 4.Rd7+ Kb8 5.Rdc7 Rb5 6.Re7+? Re5? Bxe5+? Ka8 unfortunately the line 6.Rc8+ Kb7 7.R6c7+ Ka6 8.Ra8# is 11 plies for it because 11 plies were done when the king is not in check and it cannot see it. I told it only to tell me the time when it changes it's mind so it did not tell me the exact time but I know the speed of it is usually 100-200 knodes per second(except endgames) so it means that it needed more than 10 minutes to finish iteration 10. Commercial programs are clearly better Deep Fritz has no problem to see that black is losing in a few seconds and losing means more than 5 pawns advantage for white and not 1.85 pawns for white. New position 1R1R4/k3rp1p/1p6/pP4p1/1P6/4qPB1/6PK/8 b - - 0 1 Analysis by Deep Fritz: 1...Qxf3-- ± (1.25) Depth: 1/3 00:00:00 1...Qxf3-- 2.gxf3 +- (4.72) Depth: 1/4 00:00:00 1...a4! ² (0.41) Depth: 1/4 00:00:00 1...a4! 2.Ra8+ Kb7 -+ (-1.56) Depth: 1/5 00:00:00 1...g4! -+ (-1.59) Depth: 1/10 00:00:00 1...g4! 2.bxa5 -+ (-1.81) Depth: 1/10 00:00:00 1...Qc3! -+ (-2.09) Depth: 1/11 00:00:00 1...axb4! -+ (-2.63) Depth: 1/11 00:00:00 1...axb4! 2.Ra8+ Kb7 -+ (-3.03) Depth: 1/11 00:00:00 1...axb4-- +- (#2) Depth: 2/5 00:00:00 1...axb4-- +- (#2) Depth: 2/6 00:00:00 1...Re5! +- (4.25) Depth: 2/8 00:00:00 1...Re5! 2.Ra8+ Kb7 3.Rdb8+ Kc7 ³ (-0.47) Depth: 2/12 00:00:00 1...Re5 2.Ra8+ Kb7 3.Rab8+ Kc7 4.bxa5 ³ (-0.38) Depth: 3/15 00:00:00 1kN 1...Re5-- = (0.00) Depth: 4/13 00:00:00 1kN 1...Re5-- = (0.00) Depth: 4/15 00:00:00 3kN 1...Re5-- ² (0.66) Depth: 5/17 00:00:00 5kN 1...Re5-- 2.Ra8+ Kb7 3.Rdb8+ Kc7 4.Re8 f6 5.Bxe5+ fxe5 ± (0.78) Depth: 5/17 00:00:00 9kN 1...Re5 2.Ra8+ Kb7 3.Rdb8+ Kc7 4.Re8 f6 5.Bxe5+ fxe5 ± (0.78) Depth: 6/18 00:00:00 19kN 1...Re5 2.Ra8+ Kb7 3.Rdb8+ Kc7 4.Re8 f6 5.Bxe5+ fxe5 6.bxa5 ± (0.78) Depth: 7/23 00:00:00 50kN 1...Re5 2.Ra8+ Kb7 3.Rdb8+ Kc7 4.Re8 f6 5.Bxe5+ fxe5 6.Ra7+ Kd6 7.Rd8+ ± (0.87) Depth: 8/24 00:00:00 164kN 1...Re5 2.Ra8+ Kb7 3.Rdb8+ Kc7 4.Re8 f6 5.Bxe5+ fxe5 6.Ra7+ Kd6 7.Rd8+ ± (0.87) Depth: 9/27 00:00:00 416kN 1...Re5 2.Ra8+ Kb7 3.Rdb8+ Kc7 4.Re8 f6 5.Bxe5+ fxe5 6.Ra7+ Kd6 7.Rd8+ ± (0.87) Depth: 10/28 00:00:01 1027kN 1...Re5 2.Ra8+ Kb7 3.Rdb8+ Kc7 4.Re8 f6 5.Bxe5+ fxe5 6.Ra7+ Kd6 7.Rd8+ ± (0.91) Depth: 11/30 00:00:03 2532kN 1...Re5-- ± (1.22) Depth: 12/32 00:00:07 6008kN 1...Re5-- 2.Rbc8 Kb7 3.Rc6 Qf4 4.Bxf4 gxf4 5.Rd7+ Kb8 6.Rxb6+ Kc8 7.Rxf7 +- (5.66) Depth: 12/35 00:00:09 7687kN 1...Qe5! +- (5.62) Depth: 12/35 00:00:11 9269kN 1...Qe5-- +- (5.94) Depth: 13/27 00:00:13 10495kN 1...Qe5-- 2.Bxe5 Rxe5 3.bxa5 bxa5 4.b6+ Ka6 5.Rd7 Kb5 6.b7 Re6 7.Ra8 +- (8.09) Depth: 13/32 00:00:15 12160kN 1...Re5! +- (8.06) Depth: 13/32 00:00:17 13606kN 1...Re5! 2.Rbc8 Kb7 3.Rc6 Qf4 4.Bxf4 gxf4 5.Rd7+ Kb8 6.Rxb6+ Kc8 7.Rxf7 +- (5.37) Depth: 13/36 00:00:24 19830kN 1...Re5-- +- (5.69) Depth: 14/27 00:00:28 22501kN 1...Re5-- +- (5.69) Depth: 14/37 00:00:44 36835kN 1...Re5 2.Rbc8 Kb7 3.Rc6 Qf4 4.Bxf4 gxf4 5.Rd7+ Kb8 6.Rdc7 Re8 7.bxa5 +- (5.84) Depth: 15/40 00:01:28 73167kN 1...Re5-- +- (6.16) Depth: 16/32 00:01:51 90752kN 1...Re5-- +- (6.16) Depth: 16/41 00:03:54 195008kN (Blass, Tel-aviv 02.01.2002) Uri
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.