Author: Christos Gitsis
Date: 13:34:54 01/25/06
Go up one level in this thread
On January 25, 2006 at 13:03:36, Christopher Conkie wrote: >What would you say is the fairest method of testing an opening book? Is there a >way to bench mark a book? How about playing really long matches Program_A versus Program_A, Program_B versus Program_B, etc., with one instance of each program using one opening book, and the other instance using another? I guess that if the same opening book were used for both instances, the results of the matches would ultimately approach 50%-50%. If the results are different than that, it could possibly mean that one opening book is stronger than the other. Of course this would require many long matches, which probably nobody is interested in running. Regards, Christos.
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.