Author: Ed Schröder
Date: 09:50:51 11/08/04
Go up one level in this thread
On November 08, 2004 at 11:49:55, ThatsIt wrote: >Hi Ed ! > >>There is an even more easy way. Run List vs List or Ruffian vs Ruffian, in other >>words let engines play against itself. As we are playing with reserved colors >>each 2-game cycle should produce indentical games. >> > > O.K. the assumptions is that ruffian vs ruffian will > produce identical results. > Fine, but if i want to find out which engine is better, > engineXX or engineYY, e.g. in activchess this method > don't help me very much ;-) > > And when i try to find out this with the "normal" > method e.g. tournament- or engineduelmode under cb-guis, > i will get first an 75-85, next time (running under the > complete same conditions) an 80-80 and at least an 85-75, > what should i think about ? > Hopefully this effect will disappear with higher timecontrols ?! My second experiment seems to indicate that. But I am as confused as you. Ed
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.