Author: Danniel Corbit
Date: 09:42:37 09/16/98
Go up one level in this thread
I think that generalizations about both machines and humans have an inherent danger. That is because every program and every person plays differently. Machines do make blunders! It is usually caused by a programming error or bad opening book or the like. Blunders in well tested programs are rare, of course. Also, some humans will blunder a lot, others hardly ever. I think *I* would have a very hard time to tell if I was playing a GM or a computer, because either way, I would know I was simply getting handed my hat. That a GM says he could not tell, is no doubt the truth. But that does not mean that a different GM could not tell, or that even an C player could not tell. Certainly, there are clear tendencies in computers and humans, but I don't think you can rely on them 100%. A related question is: "After 24 hours of analysis _after_ the match, could you determine if you played a computer or a human?" I think this one is more interesting. What would be the exact criteria to determine it? [Not some vague generalities, but an exact formulation of a method to determine]. I also believe that any formulation will never be 100% reliable, since computer programs do blunder, and ten different computers may choose a different move from the same starting position.
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.