Author: Robert Hyatt
Date: 09:32:42 11/25/97
Go up one level in this thread
On November 25, 1997 at 01:31:12, Jouni Uski wrote: >On November 24, 1997 at 09:55:34, Robert Hyatt wrote: > >> >>The first cut thru crafty found #5 quickly. It turned out to be a bug. >>Here's why finding this quickly is bad: > >Mchess keeps Na5 at least next 5 iterations (to ply 15). You see it has >more knowledge than fastsearchers. I disagree on this position. Claiming this is a "draw" is a clear error because no program can either (a) search to the draw [> 100 plies at least] or (b) evaluate this as a draw, because it is too special-cased and moving something on the board by a square or two can change this critically. If Mchess has an eval of -3, join the club as that is what Crafty gets, and also what others seem to get... the tactical issue is to avoid losing outright, which most programs see Na5 does. Key issue: does mchess say 0.00 or -3.00 for the eval? If it is the former, it is correct and amazing (or buggy) while if it is the latter, this is not anything special, just following a tactical line of play...
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.