Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Some Philosophical questions on the limits of Computer chess

Author: Angrim

Date: 18:10:20 01/28/02

Go up one level in this thread


On January 26, 2002 at 19:36:03, Jesper Antonsson wrote:

>On January 25, 2002 at 17:25:56, Albert Silver wrote:
>>The way you put it, it sounds as if there are very few
>>non-losing moves (i.e. a narrow road to avoid losing against perfect play)
>>whereas I believe there are many many roads to a draw that even perfect play
>>from the other side would not easily avoid.
>
>Well, assume that chess is a draw and that in each position in the drawn part of
>the game tree, there is four moves that keep the draw. If you have to play
>randomly for, say, fifty moves before the game is settled, the likelyhood that
>you keep the draw against perfect play is (4/36)^50 = 2*10^-46%. Not very good
>chances, eh? But on the other hand, how easy is it for a super GM to always play
>one of the four moves for 50 moves? If the likelyhood of him choosing the wrong
>move is just 5% per move, he'll still lose 8% of the time. But I think chess is

Odds of holding the draw would be (0.95)^50=7.7% so he would lose 92.3%
No comment on the assumptions, just the math :)

Angrim

>harder than that, and that GMs would almost never be able to draw a perfect
>player, even as white.
>
>/Jesper



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.