Author: Uri Blass
Date: 10:20:15 12/04/05
Go up one level in this thread
On December 04, 2005 at 12:27:38, Bill Angelopoulos wrote: >If you do not believe me ask Kurt... I know that kurt says that the best personality in personality games is not the best personality against other programs. It does not mean that it is useless to test program against previous version. result against one opponent always may be misleading but if program lose against previous version then it means that it may perform worse also against other programs that have similiar behaviour to the previous version. I think that when you want to estimate playing strength games against previous version should be part of the test. Note also that in the case of toga and fruit no program is previous version of the other program or personality of the other program. They have the same source(fruit2.1) but they are different. Uri
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.