Author: Drexel,Michael
Date: 04:22:03 06/03/03
Go up one level in this thread
On June 03, 2003 at 06:54:14, Joachim Rang wrote: >On June 03, 2003 at 06:47:31, Drexel,Michael wrote: > >>You cant match two programs and state that the program which wins >>the match is stronger. >>Is that so hard to understand? >>Smar Think is clearly weaker according to ratinglists which are based on much >>more games than 20. > >actually it is a good poitn to start. If a program wins an enginematch with >sufficient games (20 are not enough) against another program, there is a high >possibility, that this program is stronger. To verify or to falsify this >assumption, one has to play against other opponents. Yes, but there are too many exceptions > >I assume, that in 80 % the program which won the enginematch is stronger against >other opponents too. I think you have to match both programs at least against 3 recent versions of the commercial programs and 3 of the strongest amateur programs before such a statement is maybe justified. Number of games per match >= 40 (for 6 opponents) Michael > >I doubt that SmarThink will win against ruffian if sufficient games are played.
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.