Author: pavel
Date: 01:03:54 08/29/00
Go up one level in this thread
On August 28, 2000 at 21:19:15, Mogens Larsen wrote: >On August 28, 2000 at 20:59:08, Chessfun wrote: > >>No it would also be a which book is better match, unless the book is the same. > >That depends on what you want to test. Whether it's the difference between >native or winboard version directly, or the difference between packages, ie. WB >vs. CB. I find the latter more important, and so does Pavel and Roger I think, >because then you have two different approaches to engine+book+learning against >each other. This is more relevant from SSDF viewpoint IMO as other engines use >their own book if the opportunity exist. Otherwise all programs might as well >use the same book when tested by SSDF. But since it's your decision, I won't >interfere again, I promise :o). Just my opinion. > >Mogens. interesting..... I think we should first find out the purpose of the experiment. ( as mogens said) If I were to do the experiment my purpose would be to find out if the CB (or cb adapter) version plays at the same strength as the WB version. So I would use the auto232 adapter to play the WB crafty on one ocmputer and the CB crafty on other. OR we can use a single program (eg, YT version of sos) and play both versions of crafty (CB and WB) a reasonable abount of game against sos..... as for opening book. If it is under winboard why not use the downloadable version of crafty book?, in that case there is no one to blame. If it is under CB I guess it is reasonable to use same opening book (general.ctg??) for both the programs. as for time control, I would prefer 1hr per game. but again, (as mogens said), It is your experiment, its your time and your energy, you resources......and so on. you can do whatever you want. I am here just to give my opinion.... thanks Pavel :)))
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.