Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: K+P vs. K

Author: Don Dailey

Date: 17:02:05 01/20/98

Go up one level in this thread


On January 20, 1998 at 18:31:59, Stuart Cracraft wrote:

>
>I know tablebases exist. But is it worth it to
>have a K+P vs. K algorithm in one's static eval
>to recognize and score such positions?
>
>--Stuart

Hi Stuart,

There are a large number of tablebases that exist.  But in fact
only 3 or 4 cover 95% or more of the positions you are likely
to enter and misplay.  This one may very well be the single most
important one to have, not for playing the ending correctly which
is easily accomplished with a couple of rules, but for knowing
how to trade down from more complex endings.   You definitely
want this one.

I did my own K&P vs K database before the tablebases existed. It
only takes 24K I think and is a single bit per position.  It  is
faster than doing a regular eval because it is a single table
lookup (with a little conversion of symetries.)

- Don



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.