Author: Uri Blass
Date: 15:41:08 12/08/03
Go up one level in this thread
On December 08, 2003 at 18:30:57, Steven Edwards wrote: >I've been testing Symbolic's low level search on the BWTC test suite. Althouh >the program can solve almost all of the 1,001 positions in a few seconds or less >each, there are a few holdouts. > >Specifically: > >(BWTC.0057 BWTC.0134 BWTC.0317 BWTC.0544 BWTC.0548 BWTC.0570 BWTC.0624 >BWTC.0640) > >resist solution even after up to 600 seconds wall time each on an 800 MHz PPC >G4. > >Here are the first two missed problems. Note that in both cases the key move >was selected, but then rejected for a better scoring move. Can these be >considered valid best move results? Yes In the second position movei likes Nf6 after deeper search because of even better score(+7.50 at depth 10 instead of +6.50 for Nxh7). 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.