Author: stuart taylor
Date: 10:24:39 05/04/03
I would propose that any program which gets 50% or over against all oncomers after 50 games, is not just a "little" better than the others, but considerably better. Atleast 40 elo. The only chance for a runner up to be less than 40 elo behind is if it only loses to the top one (and at a very small margin) but scores quite similar against all others, like the top one does. It will take alot of work to overcome a program which has consistent all-round success, according to results, even after relatively few samples. Anyway, that's been my line of thinking ever since I've been posting here. Sounds familiar? S.Taylor
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.