Author: Thorsten Czub
Date: 07:32:29 03/08/02
Go up one level in this thread
On March 08, 2002 at 06:42:15, Jorge Pichard wrote: >After 12 games I decided to stop the match since Shredder 6 was winning by a big >margin. >Shredder 6 (2715) - Shredder Paderborn (2715) [C95] >Athlon 1.2 Ghz, Blitz:60' My Place (1), 06.03.2002 And what do you think does this tell us ?? That Shredder6 is stronger than Shredder6 Paderborn ? How do you know ? Because you let them play against each other?? And you really believe it would say anything about which program is better ? I will tell you something. let shredder6 play versus shredder6. if you get 50 % you have measured that the programs are similar. if you get any other program, and you get a result far from 50%, you will measure that this program is different. but is says you almost nothing if this program is weaker. maybe if you play many games it loses all, than it is weaker. maybe if you play many games and it wins all. than it is stronger. but all within the range of not similar (draw), not winning (mate against) and not losing (mate) is NOT KNOWN. there is a position. And it is maybe draw. maybe mate. or maybe mate against. so it is 1/2-1/2, 1-0, or 0-1. but anything that is not those 3 stages, is very difficult to say. when e.g. a program gets 40% against shredder6, it can be that this program is stronger than all other chess programs, and shredder6 is an angst gegner. so : how do you want to measure chess engines by let them play version X versus X+1.
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.