Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Program settings (Re: Programmers Should Take A Cue From Chessmaster)

Author: Albert Bertilsson

Date: 09:38:17 07/04/03

Go up one level in this thread


On July 04, 2003 at 09:05:12, Ryan B. wrote:

>I would drop the piece tables for the queen and rook.  For the rook open file,
>semi open file bonuses along with a small mobility bonus should work just fine
>and be more accurate for good positioning of the rook.  For the queen, mobility
>and king proximity work better than piece table bonuses for me.  Naturally both
>should get 7th rank bonus.  For the rest of the pieces Sjeng and Beowulf have
>excellent examples of good piece square values.  For knight outpost I would
>suggest something like this.
>{
>      0,   0,   0,   0,   0,   0,   0,   0,
>      0,   0,   0,   0,   0,   0,   0,   0,
>      0,   0,   0,   0,   0,   0,   0,   0,
>      0,   0,   1,   5,   5,   1,   0,   0,
>      0,   0,   5,   8,   8,   5,   0,   0,
>      0,   2,   7, 12, 12,   7,   2,   0,
>      0,   0,   2,   4,   4,   2,   0,   0,
>      0,   0,   0,   0,   0,   0,   0,   0
>};

Hi Ryan!

Currently I have no code special code for rooks, so I'll stick with the
piece-square-table until I do. The same goes for mobility for the queen and
stuff like 7-th rank bonus.

What do you mean by king proximity for the queen? Should I keep the queen close
to the own king or the opponent king? Or should that be depending on the overall
situation?

I'll try that knight-outpost-table when I add code for knight outposts.

/Regards Albert



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.