Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Just learning capability?

Author: Dann Corbit

Date: 21:05:26 06/12/00

Go up one level in this thread


On June 12, 2000 at 20:15:38, mike schoonover wrote:

>this is just my humble opinion.
>instead of programs having opening books,let them have opening programs.
>sub routines if you will.they would have an under standing of the openings,
>postionenal,and tactical.
>i know from a programing piont of view this must be a nightmare,
>but,consider the middle game sub routines.
>difficult,yes,impossible,no.
>
>my point is a human can't page through his opening library when playing a game.
>a computer can with deadly accuracy.
>i beliave the same to be true with ebtb's.

Since we are interested in fairness, I subit the following:

0.  The human players cannot use any opening they have memorized.  After, that
is unfair use of stored knowledge.  If someone is allowed to use a known
opening, it will be only under the conditions that they have never played a game
of chess or read a chess book -- but only been given a sheet of paper holding
the rules of the game.  Of course, after reading it, it must be set aside (no
cheating).

1.  The human can only think for 1e9/1e21 = one trillionth as long, since the
human brain can perform 1e21 computations per second [Morozevitch's book on AI]
while a good computer can barely muster one billion calculations per second.

Having these two minor restrictions set in place, we can have a truly fair chess
game and everyone will be happy.



This page took 0.01 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.