Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: tablebases questions to nalimov

Author: guy haworth

Date: 04:21:30 12/22/99

Go up one level in this thread


I believe Eugene generates P-endgame EGTBs 'as a whole' and not 'a slice at a
time', each slice being one pawn-formation.

Peter Karrer has pioneered the 'slice at a time' approach by modifying Eugene's
code ... and this indicates that this is a good way to go for the future.

The experience may affect both the way EGTBs are generated and the way they are
stored in files before compression.

But for now, you do the whole endgame at one go.  There will be 3.8x as many
positions as for a pawnless endgame.

Rgds, Guy



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.