Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Research idea? re: weight optimization

Author: Sune Fischer

Date: 09:06:31 12/29/01

Go up one level in this thread


On December 29, 2001 at 11:47:52, Rafael Andrist wrote:

>>BTW, I have done this and the results are crap. I don't think there's a good way
>>to translate move frequency into location desirability, i.e., GMs play Nf3 a
>>heck of a lot more than Ne5, but f3 is not a better location for a knight than
>>e5 is...
>
>I did it with very good results for the opening phase, I can show you some
>statistics if your interested. Are you sure your code worked correctly? Anyway I
>have thrown all piece square stuff out and evaluate otherwise.

He could be right, the move Ne5 is usually so strong that the opponent will not
allow it, so really great moves may not be very frequent moves.

-S.



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.