Author: Dann Corbit
Date: 12:25:09 07/29/02
Go up one level in this thread
On July 29, 2002 at 15:14:07, Daniel Clausen wrote: >On July 29, 2002 at 15:12:12, Dann Corbit wrote: > >>What sort of comments? About statistical significance? > >Yup :) The conclusion reached "they are still close in strength" is sufficiently inexact as to render argument in either direction as fairly pointless. With 60 games, you can probably get an ELO error bar of a few hundred or so. The average figure of equality is also in agreement with the conclusion. If the original conclusion was "They are exactly of the same strength." then that would be something to quibble about. Within a couple hundred ELO, they are probably about the same strength. Both are very strong, BTW. With Pepito's new book, it appears to learn very effectively. Play a long match against Crafty or Yace or whatever strong opponent you like. You will observe the effects of learning as time goes by.
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.