Computer Chess Club Archives


Search

Terms

Messages

Subject: An experiment ...

Author: GuyHaworth

Date: 01:57:20 01/22/04

Go up one level in this thread



Rafael Andrist ran an experiment on the classic 'Browne-BELLE' KQKR maxDTC
positions using my Reference Endgame Player model.

Browne has exhibited an apparent competence of '20' in terms of the model, so we
pitched an REP of 'competence 20' as KQ against the EGT as KR.

This was described in our ACG-10 (Graz) paper which Rafael presented.  It was a
follow-up to an exposition of the original idea at Maastricht (2002) and in the
ICGA_J (2003).


If you are interested in how well a chess-engine would do without the EGT, have
it play the KRPP side against the EGT as KR.

An 'EC' DTC EGT of KRPPKR is not available yet but Eugene (and I think Marc with
FEG) have created DTM EGTs.

g



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.