Author: blass uri
Date: 03:17:39 04/27/00
Go up one level in this thread
On April 26, 2000 at 20:41:31, Flemming Rodler wrote: >Hi all, > >I was just wondering how people determine if a new version of the program they >are developing is better than the current. Are there other methods than just >testing it against a whole bunch of other chess engines? > >/Flemming There are other methods. If after every small change you do a lot of games to prove that the change is positive you cannot get a lot of progress and it is sometimes better practically to do changes without being at least 95% sure that you made a progress. There are cases when you do not need to do it to be convinced that you made a progress. For example if you found an idea to do the same thing faster(no change in evaluation or search rules) then you need only to test speed. If you find that the new version needs less time to do the same thing in many positions then you do do not need games to be practically sure that the new version is better. 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.