Author: Ernst A. Heinz
Date: 14:19:53 12/03/97
Go up one level in this thread
On December 03, 1997 at 16:45:36, Don Dailey wrote: > >Hi Ernst, > Hi Don, good to have you on an electronic computer chess forum at last! :-) >Do you give attacked squares partial credit or none at all? None at all as far as "mobility" is concerned -- "space" is a different animal, of course. But I do not like to score "space" ... >I read your article on Dark Thought. It sounds like you have >a great program. I mentioned early that I only just recently >heard of rotated bit boards. I heard about this from the article. Nice to hear that the article was of use to our competitors ... :-) >I'm interested in what evaluation heuristics people apply to the >queen. Currently we do nothing but I don't believe this is good >enough. We occasionally find completely useless squares for the >queen and this gets us into trouble. But mobility seems to have >some bad side effects too. Does anyone want to share some info >here? Yes, queen mobility seems to be no good at all -- I have discarded it long ago, too. "DarkThought" seems to do fine with scoring the queen in a very similar way as a rook (i.e., file/rank connection, placement on semi-open and open files, and so on). =Ernst=
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.