Author: Bela Andrew Evans
Date: 17:38:11 06/18/98
Go up one level in this thread
Human-computer testing is great -- only problem is finding humans. These programs are better than 99.1 to 99.9% of all human chess players. I personally have a USCF rating of 2189 right now, which makes me better than approximately 90% of all US tournament chess players -- and I bet that most of the top programs running of today's fast PCs would crush me game after game. So if Fritz 5 beats me 10-0, Rebel 9-1, Hiarcs 9.5-.5 -- any ratings these programs get from crushing me won't mean a whole lot. But hey, if you can find a bunch of 2300 and up players willing to play 10 games against the computer programs, you'll have some interesting and significant data. Any CCC members care to volunteer for this experiment? Bela Evans
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.