Author: James T. Walker
Date: 05:29:52 10/11/99
Go up one level in this thread
On October 11, 1999 at 02:31:07, Harald Faber wrote: >On October 10, 1999 at 13:05:08, James T. Walker wrote: > >>>I wouldn't even expect a 50% score both programs playing with the same settings. >> >>Hello Harald, >>I once played a match of Crafty/tablebases vs Crafty/no tablebases. The >>tablebases I had were only a few 5 man(No KRPKR) so the tablebases were >>essentially useless. The score was 201.5 to 198.5. (The tablebases Crafty won) >>So I think it is reasonable to expect a program to play even against itself. >>Jim Walker > >Don't forget you didn't play the program vs itself, you played Crafty+TB vs >Crafty-TB! The point is it was a similiar experiment. The experiment was to see the difference between having TB and not having TB using the same program as a reference. Also as I mentioned, I only had a few TB's and their effect was minimal. That is why the score was almost identical. Again, I say that playing a program against itself, it is reasonable to expect it to play very even. That being the case then any changes such as PB/NoPB should account for the difference in score. Same with TB/NoTB. This is a pretty simple concept. Jim Walker
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.