Author: Gabor Szots
Date: 04:07:36 01/03/06
Go up one level in this thread
On January 03, 2006 at 06:56:16, Thorsten Czub wrote: >the question is what the test suites measure. > >it is very often the case that test suite do not measure strength. > >e.g. programs solving test suites very good are not automatically good chess >programs in real games. I agree. For example, Rybka (default) solves 97 of Arasan5 and 17 of WM, which are under HIARCS 10 (default). But in practical play Rybka is better, it seems.
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.