Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: WCCC Hardware Uniformity

Author: GuyHaworth

Date: 16:37:44 07/07/04

Go up one level in this thread



The most recent attempts by the ICGA to organise 'uniform platform' events have
not attracted great support from either authors or sponsors.

Whatever machine was chosen, the subtleties of hardware-architecture would
favour one engine or another.

Experiments in self-play with different depths of 'basic search' on the two
sides indicate decreasing advantage from the extra ply as the 'core' search
depth increases.  These experiments have only gone to 12 ply, and the search
depths being achieved in WCCC_2004 are typically 15, 16 and more ply.

My understanding is that typical fan-out is effectively 3 or so, so an extra ply
requires more than 2x the hardware power.  Given this, I do not think that the
extra power confers much more ELO, but of course, one still wishes to have the
quickest machine under one's software.


The ICGA WCCC is not a test of the 'best search engine' but of the whole 'chess
playing system' consisting of hw, sw, opening book, EGTs, author and operator.

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.