Author: blass uri
Date: 23:31:27 01/21/99
Go up one level in this thread
On January 21, 1999 at 23:20:54, Steven Juchnowski wrote: >I have played Junior5 against other programs using >separate PC's. > >Since the moves are entered manually, there have >been occasions where Junior's opponent has lost >on time according to Junior's clocks. Some of these >games have in fact ended in victory for the opponent. > >What I have noticed is that Junior's learning is based >on the lost in time result rather than the final result. > >This has led to Junior replaying a previously lost game. > >Fritz5 does not appear to have this problem as Fritz will >always prompt you for the scoreline upon exiting the game, >irrespective if the game was lost on time or not. I think that it is a mistake to learn only by result. This is a stupid way of learning because if your opponent lost and missed a win then it can repeat the same game without missing the win. I believe that Amir Ban is not responsible for the learning of Junior5. I know that he believes that learning is not important and there is simple things against learners(some randomness and big book) Chesstiger in enrique's tournament has not big book and has no learner but Fritz tried without success to repeat the same win again and again because of some randomness. Uri
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.