Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: REBEL CENTURY 3.0 finds this mate in 15 to 17 moves. Easily!

Author: Robert Hyatt

Date: 15:10:21 11/23/00

Go up one level in this thread


On November 22, 2000 at 18:21:39, Timothy J. Frohlick wrote:

>[D]8/8/8/3B4/8/2K1N3/8/k7 w - - 0 1
>
>The Rebel Century 3.0 program knows that it must drive the black king to the
>corner which has the same-colored bishop color i.e. h1.  It does so easily in
>about three minutes of play time.  The variations are numerous.
>
>The one reason that I don't like tablebases is the fact that the damned computer
>will usually solve the problem the same way using the same stupid moves every
>time.  Goodbye Turing Test.
>
>Whip out your Rebel Century 3.0 program on the Rebel 11 disk and try this
>yourself.  You will be most entertained if you set the clock at 15 minutes total
>time and then try to outplay the RC3 program.  I do not think that any of you
>can avoid a mate in over 17 moves.
>
>Tim Frohlick


Doing what you suggest doesn't help the Turing test problem.  The program will
_still_ play the same moves each time...



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.