Author: Daniel Shawul
Date: 04:34:54 03/18/05
Go up one level in this thread
On March 18, 2005 at 07:29:23, Daniel Shawul wrote: >On March 17, 2005 at 16:46:57, Ross Boyd wrote: > >>On March 17, 2005 at 15:55:10, Dann Corbit wrote: >> >>>Not surprising. >>> >>>Fruit does not have a pawn hash table, which is the key thing to solving this >>>thing fast. >> >>Can't agree with you Dann. >> >>The key to solving it fast is a properly working transposition table with a good >>replacement scheme. > Some strong commercail engines like DeepSjeng can't solve it too. > This position is a bad one for tuning a replacement scheme. >daniel > > Adding pawn hashing made only a very small difference to >>TRACE 1.28. I suspect a little glitch in Fruit. >> >>For proof, here's TRACE 1.27's output... its just as fast as 1.33B and there's >>no pawn hashing. >> >>FEN: 8/k7/3p4/p2P1p2/P2P1P2/8/8/K7 w - - 0 1 >> >>TRACE 1.27: >> 2 00:00.030 40 40 +0.70 Ka2 Kb6 >> 2 00:00.030 41 41 +0.80 Kb2 Kb6 >> 2 00:00.030 42 42 +0.80 Kb2 Kb6 >> 3 00:00.030 75 75 +1.00 Kb2 Kb6 Kc3 >> 4 00:00.030 157 157 +1.00 Kb2 Kb6 Kc3 Kc7 >> 5 00:00.060 289 289 +1.10 Kb2 Kb6 Kc3 Kc7 Kc4 >> 6 00:00.060 486 486 +1.00 Kb2 Kb6 Kc3 Kc7 Kc4 Kd7 >> 7 00:00.060 794 794 +1.00 Kb2 Kb6 Kc3 Kc7 Kc4 Kd7 Kd3 >> 8 00:00.070 1.279 1.279 +1.10 Kb2 Kb6 Kc3 Kc7 Kc4 Kb6 Kd3 Kc7 >> 9 00:00.070 1.846 1.846 +1.10 Kb2 Kb6 Kc3 Kc7 Kc4 Kb6 Kd3 Kc7 Ke3 >> 10 00:00.080 5.321 532.100 +1.00 Kb2 Kb8 Kc3 Kb7 Kc4 >> 11 00:00.090 6.764 676.400 +1.00 Kb2 Kb8 Kc3 Kb7 Kc4 Kb6 Kd3 Kc7 Ke3 >>Kd7 Kd3 >> 12 00:00.090 8.335 833.500 +1.00 Kb2 Kb8 Kc3 Kb7 Kc4 Kb6 Kd3 Kc7 Kc4 >>Kb6 >> 13 00:00.100 9.965 498.250 +1.00 Kb2 Kb8 Kc3 Kb7 Kc4 Kb6 Kd3 Kc7 Ke3 >>Kd7 Kf3 Ke7 Ke3 >> 14 00:00.100 13.624 681.200 +1.00 Kb2 Kb8 Kc3 Kb7 Kc4 Kb6 Kc3 Kb7 >> 15 00:00.110 17.070 426.750 +1.00 Kb2 Kb8 Kc3 Kb7 Kc4 Kb6 Kd3 Kc7 Ke3 >>Kd7 Kf3 Ke7 Ke2 Kd7 Kd3 >> 16 00:00.110 24.673 493.460 +0.91 Kb2 Kb8 Kc3 Kb7 Kd3 >> 17 00:00.130 29.554 369.425 +1.00 Kb2 Kb8 Kc3 Kb7 Kd3 Kc7 Ke3 Kd7 Kf3 >>Ke7 Ke2 Kd8 Kd2 Kd7 Kc3 Ke7 Kc4 >> 18 00:00.150 42.911 306.507 +1.00 Kb2 Ka8 >> 19 00:00.170 49.482 309.262 +1.00 Kb2 Ka8 Kc3 Kb7 Kd3 Kc7 Ke3 Kd7 Kf3 >>Ke7 Ke2 Kd8 Kd1 Kd7 Kd2 Kc8 Kc2 Kb8 Kc3 Kb7 >> 20 00:00.180 55.357 307.538 +1.00 Kb2 Ka8 Kc3 Kb7 Kd3 Kc7 Ke3 Kd7 Kf3 >>Ke7 Ke2 Kd8 Kd1 Kd7 Kd2 Kc8 Kc2 Kb8 Kc3 Kb7 >> 21 00:00.200 61.474 307.370 +1.00 Kb2 Ka8 Kc3 Kb7 Kd3 Kc7 Ke3 Kd7 Kf3 >>Ke7 Ke2 Kd8 Kd1 Kd7 Kd2 Kc8 Kc2 Kb8 Kc3 Kb7 >> 22 00:00.230 75.746 329.330 +1.00 Kb2 Ka8 Kc3 Kb7 Kd3 Kc7 Ke3 Kd7 Kf3 >>Ke7 Ke2 Kd8 Kd1 Kd7 Kd2 Kc8 Kc2 Kb8 Kb3 Kc8 >> 23 00:00.290 107.566 384.164 +1.00 Kb2 Ka8 Kc3 Kb7 Kd3 Kc7 Ke3 Kd7 Kf3 >>Ke7 Ke2 Kd8 Kd1 Kd7 Ke1 Kd8 Ke2 Kd7 Ke3 Kd8 >> 23 00:00.310 119.141 384.325 +1.01 Kb1 >> 23 00:00.350 134.987 397.020 +3.01 Kb1 >> 23 00:00.380 156.072 410.715 +3.52 Kb1 Kb7 Kc1 Kb8 Kc2 Kb7 Kc3 >> 24 00:00.411 168.266 420.665 +3.59 Kb1 Kb7 Kc1 Kb8 Kc2 Kb7 Kc3 Ka6 Kd2 >>Kb7 Ke2 Kc8 Kf2 Kd7 Kg3 Ke7 Kh4 Kf6 Kh5 Ke7 >> 25 00:00.461 182.779 406.175 +3.59 Kb1 Kb7 Kc1 Kb8 Kc2 Kb7 Kc3 Ka6 Kd2 >>Kb7 Ke2 Kc8 Kf2 Kd7 Kg3 Ke7 Kh4 Kf6 Kh5 Ke7 >> 26 00:01.671 207.891 314.986 +3.59 Kb1 Kb7 Kc1 Kb8 Kc2 Kb7 Kc3 Ka6 Kd2 >>Kb7 Ke2 Kc8 Kf2 Kd7 Kg3 Ke7 Kh4 Kf6 Kh5 Ke7 >> 27 00:01.811 242.456 303.070 +3.59 Kb1 Kb7 Kc1 Kb8 Kc2 Kb7 Kc3 Ka6 Kd2 >>Kb7 Ke2 Kc8 Kf2 Kd7 Kg3 Ke7 Kh4 Kf6 Kh5 Ke7 >> 28 00:01.132 300.777 273.433 +3.59 Kb1 Kb7 Kc1 Kb8 Kc2 Kb7 Kc3 Ka6 Kd2 >>Kb7 Ke2 Kc8 Kf2 Kd7 Kg3 Ke7 Kh4 Kf6 Kh5 Kf7 >> 29 00:01.432 424.764 299.129 +3.76 Kb1 Kb7 Kc1 Kb8 Kc2 Kb7 Kc3 Ka6 Kd2 >>Kb7 Ke2 Kc8 Kf2 Kd7 Kg3 Ke7 Kh4 Kf6 Kh5 Kf7 >> 30 00:02.712 522.858 305.764 +3.76 Kb1 Kb7 Kc1 Kb8 Kc2 Kb7 Kc3 Ka6 Kd2 >>Kb7 Ke2 Kc8 Kf2 Kd7 Kg3 Ke7 Kh4 Kf6 Kh5 Kf7 >> 31 00:02.594 766.436 297.068 +3.86 Kb1 Kb7 Kc1 Kb8 Kc2 Kb7 Kc3 Ka6 Kd2 >>Kb7 Ke2 Kc8 Kf2 Kd7 Kg3 Ke7 Kh4 Kf
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.