Author: blass uri
Date: 10:27:57 06/20/99
Go up one level in this thread
On June 20, 1999 at 04:41:59, Marcus Kaestner wrote: >On June 19, 1999 at 18:22:40, Anatoli wrote: > >>Hi ! >>It will be very interesting if somebody runs a 10-20 game match between the >>World Champion Shredder and Hiarcs 7.32 on two fast mashines. I am still very >>surprised by Hiarcs poor performance. There was a big mess about it and its >>2600+ rating. Now everybody is after Ferret and Shredder and Hiarcs 7.32 is not >>a star anymore ?! I personally thing that a 7 round robin tournament is >>absolutely unsuitable for challenging the World best program. Programs must play >>with each other using White and Black colous and the tournament should be >>longer. >>Thanks > >this question is very easy to answer. >on PIII-500 this shredder 4 scores more than 70% against H7.32 in 30m blitz. >hiarcs needed 21 games for the first victory! >longer time controls are further better for shredder. What is the evidence that longer rime controls are further better for shredder? Here are some test positions based on some correspondence games(mainly of me) with the results of dark thought. I did not give position number 2 because I am not sure if the solution for this position is the only correct move. I am interested in the result of shredder. These position are hard for chess programs and dark thought results are very good(in most of the cases better than the commercials that I know). position number 1: r1bqr1k1/pp1p1ppp/5n2/1N2n3/2P1PB2/6P1/PP1Q1K1P/4RB1R b - - 0 1 (bm = d6) "DarkThought" locks onto d6 in iteration #12 after roughly 14 minutes while running on a 600MHz Alpha-21164a (see below). 00:13:55 12.02[ -27, -26] d6 Nxd6 Qb6+ Kg2 Bg4 Bxe5 Rxe5 b4 Qc6 Qd4 >=-24 (-0.19) #211168866 ----- position number 3: 7r/1k2P2P/4p3/1pp5/8/2pp4/1n3PP1/R4K1R w - - 0 1 (bm = g4) "DarkThought" locks onto g4 in iteration #13 after roughly 54 minutes while running on a 600MHz Alpha-21164a (see below). 00:53:34 13.04[ 177, 178] g4 c2 >=178! (1.39) #1054920952 ----- position number 4: r4rk1/2p3pp/p7/1p1pq3/8/2P2N2/PPQ2KPP/R1B5 w - - 0 1 (am = Kg1) "DarkThought" abandons Kg1 for good in iteration #13 after 37 seconds while running on a 600MHz Alpha-21164a (see below). 00:00:27 13.01[ -32, 48] Kg1 Rxf3 gxf3 Qe1+ Kg2 Re8 Kh3 Re5 Qd2 Qg1 =-31 (-0.24) #8198247 00:00:37 13.02[ -31, -30] Qd2 Rxf3+ gxf3 Qxh2+ Kf1 Qh1+ Kf2 c6 Qe3 Rf8 >=-30! (-0.23) #11425757 ... 00:08:36 14.02[ -6, -5] Qd3 Qe4 Qd4 Kh8 Kg1 Rae8 >=-5! (-0.04) #149371636 ----- position number 5: 3r1b1k/pp6/2n2q1p/2Bp1B1Q/5P2/1P4p1/P7/4R1K1 w - - 0 1 (bm = Re6) "DarkThought" locks onto Re6 in iteration #14 after roughly 37 minutes while running on a 600MHz Alpha-21164a (see below). 00:37:19 14.16[ -225, -224] Re6 Bxc5+ >=-176! (-1.38) #586903348 ----- 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.