Author: Dann Corbit
Date: 11:32:23 01/03/06
Go up one level in this thread
On January 03, 2006 at 06:46:22, Uri Blass wrote: >On January 03, 2006 at 06:30:51, Gian-Carlo Pascutto wrote: > >>On January 02, 2006 at 18:29:31, Otello Gnaramori wrote: >> >>>On January 02, 2006 at 15:11:08, Vincent Diepeveen wrote: >>> >>>>In my eyes what rybka doing is very basic. Like fruit's eval + simple kingsafety >>>>+ very little more than that, but real well tuned. >>>> >>>>In my definition that is not much. That's real simple stuff. >>> >>> >>>I don't think it is so simple...I strongly suspect that Rybka has a special >>>evaluation fuction with a _lot_ of knowledge inside...the tuning isn't enough to >>>explain that brilliant performances. >>>I tend to agree with Uri. >> >>Wouldn't you have thought the same thing about Fruit if you had never seen the >>source? >> >>-- >>GCP > >I think that fruit also has a lot of knowledge inside of it relative to weaker >programs. > >Weaker programs may have bigger evaluation but bigger evaluation is not chess >knowledge. As Christophe Theron has aptly pointed out (on several occasions), better search also results in more knowledge.
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.