Author: Thorsten Czub
Date: 07:18:26 01/10/06
Go up one level in this thread
On January 10, 2006 at 09:59:12, Arnold Gove wrote: >I don't understand. You mean that when other programs lose they do know what is >happening? > >AG others lose for other reasons. e.g. Genius cannot evaluate exchange lines in the main line. he can only count the material (in the end position) when you have exchanges in the main line, but genius does not know if the end position is lost or won, from the point of view of a chess position. he has only a static exchange evaluator. and therefore very often when genius loses in chess games, you will see that it evaluates its own position positive, you have few exchanges in the main line, the forced exchanges happen on board. and instantly genius sees: oh - i am lost ! I guess the problem with fritz is as system immanent as the asymmetric search and the static exchange evaluator is system immanent in genius. those things will never change and thats the reason those programs get beaten. if richard would have ever changed this. maybe rewrite genius new from the scratch, he could have had a chance to make progress. but he failed to do so. and was overtaken. the programs have strength and weaknesses in very different situations. depending on how the programs worked. e.g. Chess System Tal was good in bringing a position into the winning point for itself, but weak in winning the won situations. this had to do with the way it was designed.
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.