Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Chess program improvement project (copy at Winboard::Programming)

Author: Stuart Cracraft

Date: 22:53:01 03/06/06

Go up one level in this thread


On March 06, 2006 at 22:49:11, Dann Corbit wrote:

>Unstoppable passed pawn.  What does your program see as far as pawn score here?
>[D] 8/7p/5k2/5p2/p1p2P2/Pr1pPK2/1P1R3P/8 b - - bm Rxb2

It went to 5000 with a 10x millipawn convert from centipawn that I had
missed in the original cloning of 4.7's eval.

That's from a factor which is the square of the rank to up to 23x the
value based on square status in front times 10 overall.

There is a connected passer credit of 500, regardless of rank as well,
not 4.7 but GNU, so the total comes to > 5000 so the connected passers
eventually are worth a rook which is canonical wisdom. I don't know if
it is good for a program to have such a high credit overall though.

Stuart



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.