Author: Don Dailey
Date: 07:32:20 07/14/98
Go up one level in this thread
On July 13, 1998 at 10:23:15, Shaun Graham wrote: >P.S. > > Of course Fritz could never play by simply telling moves secretly to a player, >because it's CHEATING! However, for a good scientific investigation, the removal >of the player vs computer bias would need to be eliminated. > >This happens to call to mind the famous John VonNeuman incident at a US chess >tornament a few years ago, where an individual "REPORTEDLY" recieving computer >information was defeating very strong players including i believe I.Ms (this >story has i'm sure undergone many evolutions, so i do not hold as gospel, So >i'm not sure about I.Ms but indeed very strong players ) I think this would be a great scientific experiment. I would love to know how much this affects results, I'll bet it makes a lot of difference in favor of the humans. But for Fritz to get it's GM norms, it would be unfair to hide it's identity (I know you're not saying this, I'm just making the point.) Every GM has a style that is well known and it's part of the game. So we cannot ever include this as part of the formula for calculating Fritz strength, instead it becomes one of Fritz (and any other program) liability and this is only fair. - Don
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.