Author: Barry Culp
Date: 03:36:41 01/21/06
Go up one level in this thread
On January 21, 2006 at 02:33:09, Kurt Utzinger wrote: >On January 20, 2006 at 21:55:54, robert flesher wrote: > >>My questions are regarding my previous match between Rybka beta and Hiarcs 10 >>modified. In this match I set both engine to have maximum learning within the >>gui under the engine match. So for clarification under book options I set >>"Variety of play" to the far left side, "Influence of learn value", and Learning >>strength" both to the far right side. Also, I checked tournement book, thus >>getting what I believed to be the best book settings for learning. However, many >>complained and stated unfair. So I plan to start over the 1000 game match, but >>before I do so can all the people who think they have a better method to please >>enlighten me (sarcasm aside!!). I want to avoid book repetition, although I do >>not like preset posistions. I like the full game from opening to endgame. I do >>not see what was wrong with my previous setting so any help would be welcomed. >>Btw I will of course update to the newer Rybka beta 11/12, cheers > > No test is unfair as long as all engines play under > the exact same conditions in my opinion. > Best regards > Kurt Using different books (Hiarcs10.ctg for Hiarcs and Junior9.ctg for Rybka) means that you are testing engines and books. I would expect Hiarcs to perform very well in these conditions. If you want to test just a set of Hiarcs engine parameters, then the book setup should be changed. In any case, in a 200 or so game match, if you end up with 1 out of every 6 games with the exact same opening and the exact same colors, I think it would be reasonable to question the test conditions.
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.