Author: Dann Corbit
Date: 02:01:26 12/13/03
Go up one level in this thread
On December 13, 2003 at 04:31:15, Uri Blass wrote: >On December 12, 2003 at 20:43:25, Dann Corbit wrote: > >>In solving WAC at 5 seconds per position on a 950 MHz AMD Athlon machine, Crafty >>solves 294 out of 300. That is pretty run of the mill as far as top engines go. > >No > >Knightdreamer is not a top engine and it solved 297/300 in 5 seconds in slower >hardware(733 mhz PIII). And shredder 6 solves 285/300 on similar hardware. >see http://www3.tripnet.se/~owemelin/johan/KnightDreamer.html >and look in version history. > >I do not have chessmaster9000 but I believe that it is slightly better than >knightdreamer in solving test suites. > >I also think that some of the problems in Wac are solved for positional >reasons(for example Rb4). That one is solved by blockade detection. >personally I do not like 0.00 scores for positions that one side has material >advantage and you only suspect that it is a draw. It *is* a draw until a way to break through the wall is found that is better than simply losing material. >The question is if detect draw does not detect draw for non drawn position. >If it means that in most of the cases it is draw but there are mistakes then it >may be better to use something between 0.00 and 1 pawn advantage for the better >side based on the static evaluation without detect draw. > >0.00 is justified only if you are sure that the position is a draw or both sides >have the same chances. > >Note that even if you detect other moves as +1 for black crafty can solve it >because it detected Rb4 as +1.18 for black at depth 14. Crafty detects the potential draw for the same reason that you would see it. Crafty breaks the dam for the same reason that you would break it.
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.