Author: Stephen A. Boak
Date: 18:47:33 11/19/99
Go up one level in this thread
All my above discussion makes me wonder if we are really dealing with a bell-shaped curve or not--I tend to doubt it. But let me SEE, :), Standard Error of Estimate is thought to be a Bell-shaped curve! This might work after all. Hmmm. Well, I do have some linear extrapolation techniques that might be able to take 90Mhz, 200Mhz, 450Mhz, etc, results and 'predict' (please don't shoot me just yet!) the nominal ratings of various programs on one 'standard' Mhz. It wouldn't be perfect, mind you, and would be highly controversial I'm sure, and might spur a debate of 500 postings, nearly rivaling.... And there are non-linear methods I like to devise--whatever works, I always say, at least in mathematics. More thoughts later. I will be tied up for about a week or so. Working on knot theory now. --Steve :)
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.