Author: Pete Galati
Date: 21:31:48 03/20/00
Go up one level in this thread
On March 21, 2000 at 00:05:16, Vincent Vega wrote: >I predict that the conventional wisdom that doubling the time per move increases >chess program's Elo rating (in a pool of chess programs) by a constant value >will be proven wrong. I believe that as time per move increases there will be >more and more draws between programs so that Elo rating increase won’t be as >high as it was for slower computers or faster time controls. I hope to run a >small experiment to test this when I have some free time. There probably are more draws, hard to say, I never checked. Frank has plenty of tournaments available for download so that you'd have plenty examples to look through. Probably as big a factor would be that programs will vary in relative strength to eachother, depending on the time control used, that is some will excell in Blitz, but others will excell in long time controls. Pete
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.