Author: Kurt Utzinger
Date: 09:25:01 04/23/05
Go up one level in this thread
On April 23, 2005 at 09:21:16, Christian Koch wrote:
>Hi @all,
>
>Spike 0.0 and Spike 0.9a are the same from the point of playing strength.
>
>To see an ELO difference from 72 points between 0.9 and 0.9a is a little bit
>dubious. You need much more games to draw such conclusions.
A further point: in my opinion a gauntlet with only
two games vs X engines is a very unreliable - if not
the most unreliable - method of testing a chess program
as one leaves the results more or less to chance. I'd
prefer 10 matches at 10 games to 50 matches at 2 games.
Kurt
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.