Author: Dann Corbit
Date: 12:21:29 12/11/00
Go up one level in this thread
On December 11, 2000 at 14:40:12, Dieter Buerssner wrote: >On December 11, 2000 at 10:03:19, Jon Dart wrote: > >>I've been building a small test suite for a while based on positions >>from Arasan games. Recently I took a closer look at this, weeded out >>very easy and very hard positions, and a few where I wasn't sure there >>was a clear right answer. Then I added some more positions from recent >>games. So now I have a condensed version with 70 positions, which follows. > >Thanks, for posting the interesting test suite. > >>Arasan is now getting about 30 right at 60 seconds/move on a PIII/733 >>with 16MB hash. Crafty 17.6 got 48 right on the same machine. > >Compared to this, I get a surprisingly well result with my current version >of Yace. 52 right. AMD K6-2 475, 30 MB hash. Of course, it is possible, that >some solutions will be "lost" at longer search time. > >>3r3k/6p1/7p/1N6/2pP4/p3PbP1/R4P2/5K2 b - - bm Rd8; id "arasan2.11"; > >Do you mean "bm Rb8" here? Surely it must be, as a glance at the digram will reveal: [D]3r3k/6p1/7p/1N6/2pP4/p3PbP1/R4P2/5K2 b - - acd 22; acn 660178447; acs 1000; bm Rd8; ce 93; id "arasan2.11"; pv Rb8 Nxa3 Ra8 Ra1 Ra5 d5 c3 Nc2 Rxa1+ Nxa1 Bxd5 Ke2 Bc4+ Kd1 Bd3 Nb3 Kg8 Nc5 Bb1 Kc1 c2 e4 Kf7 e5;
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.