Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: How well did the CCC guess the new SSDF results?

Author: Robert Hyatt

Date: 19:20:01 01/29/99

Go up one level in this thread


On January 29, 1999 at 08:42:58, James T. Walker wrote:

>It seems to me this is about as good a way to test your program for improvements
>as any.  Since GM/IM players are not laying around waiting for a program to play
>against what do you suggest?  You run Crafty on ICC for thousands of hours and
>it plays against other computers so you can look for problems and make
>improvements.  I see very little difference except most of your games are blitz
>and probably not as good as playing at longer time controls for testing.
>Jim Walker

less than 10% of my games are against computers.  over 30% are against GM's,
another 25% against IMs.  But the main point is I tune against a _wide_ variety
of opponents, rather than specifically tuning for 3-4.  IE different goals,
since I don't care where I end up in the SSDF since sales volume isn't an issue.
For those that it is, they have to worry about where they end up for sales
reasons...



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.