Author: Tord Romstad
Date: 02:23:50 12/04/05
Go up one level in this thread
On December 02, 2005 at 19:25:15, Mig Greengard wrote: >Of the seriously wrong evaluations I've collected most have to do with pawn >majorities with decisive endgame implications. Humans may lose such positions to >computers anyway of course, but analysis and instruction are very different. >Beating humans so easily because of the tactical superiority has masked the >evaluation deficiencies that are obvious when you use them to analyze human >games. Hi Mig, Actually, most of us are much more interested in such evaluation deficiencies than in forced mates. If you have a big collection of positions which programs tend to misevaluate badly, I (and many others, I am sure) would be very grateful if you could post them. Tord
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.