Author: Robert Hyatt
Date: 17:11:46 07/25/98
Go up one level in this thread
On July 25, 1998 at 19:00:53, Mark Gaddis wrote: >Does anyone know of a set of chess positions which can be used to check both the >tactical and positional prowess of a chess program? It might be possible to get >an approximate rating for a program by analysing the percentage of correct >solutions determined and the time it took to arrive at these solutions. This >would save a tremendous amount of time versus playing many actual games against >a rated opponent. There is no such animal, unfortunately. The problem is we have no way to "initially rate" the programs so that known programs could be normalized against some known suite results to produce a decent formula for computing the rating. Programs are all different, so there's nothing to say that a *new* program's results on a test suite would mean anything at all relative to a chess rating. We keep hoping for such a magic bullet, but none exist. Everyone runs win at chess, a known suite that is pretty easy. I don't know how everyone else does, but I've reported my results here: on my 4 processor machine I get 299/300 at one minute per move. I have also posted 30 second and 15 seconds results in the past, but don't have them handy...
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.