Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: a relatively easy mate in 24 to solve

Author: Uri Blass

Date: 13:49:04 12/20/01

Go up one level in this thread


On December 20, 2001 at 16:42:29, Uri Blass wrote:

>On December 20, 2001 at 16:00:09, Dieter Buerssner wrote:
>
>>On December 20, 2001 at 10:45:11, Uri Blass wrote:
>>
>>[Much snipped]
>>>I think that this time I can say mate in at most 18
>>>in my next move even if my opponent declines the condition and
>>>the only thing that I need to be careful is not to do
>>>a mistake in writing the moves and not to be dead.
>>
>>It seems, that you trusted the scores of a certain program for the pawn endgame.
>>What would happen, if there is was an horrible bug in that program, and the
>>scores were wrong. Will the author be dead?
>>
>>Cheers,
>>Dieter "still alive" Bürßner
>
>I also checked with another program before sacrificing the rook.
>It did not see the mate but it saw a decisive score for black.
>
>The worst thing that I can imagine if yace is wrong is that I do a mate in more
>moves because I also can see clearly the win for black and the only thing that I
>cannot see without computer help is the distance to mate.
>
>Uri

Note that yace changed it's mind from mate in X to mate in Y when Y>X in part of
the cases after learning.

I believe that in all of these cases there is a mate in at most X moves and my
experience is that yace later see again a shorter mate than mate in Y.

Uri



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.