Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Yace (who else) shows mate in 17 after 2 minutes

Author: Uri Blass

Date: 04:31:40 01/17/02

Go up one level in this thread


On January 17, 2002 at 06:59:28, David Rasmussen wrote:

>On January 17, 2002 at 05:26:40, Uri Blass wrote:
>
>>On January 17, 2002 at 05:01:24, Jouni Uski wrote:
>>
>>>in Celeron 300. But I didn't check, if it keeps the solution. Pro programs
>>>(play) are helpless again...
>>>
>>>Jouni
>>
>
>Can we have a FEN of the position?
>
>>They have no chance because of null move pruning.
>>Chessmaster may have better chance because I believe that it does not use the
>>wrong null move algorithm(I define the wrong null move algorithm as null move
>>algorithm that can cause stupid blunders like not finding mate in 2).
>>
>>Yace is not using the wrong null move algorithm and I believe that the same is
>>for chessmaster.
>>
>
>What do you mean by "wrong null-move algorithm"? Which programs uses this, and
>what _is_ it? And what is the right algorithm?
>
>/David


The wrong algorithm is an algorithm that does not have zunzwang detection
Correct algorithm should always solve mate in 2 problems.

I believe that big part of the top programs can miss mate in 2 when the time is
not important

I also remember Deep Fritz for doing mistakes that were result of null move
pruning when time could not help(it was not mate in 2 but Deep Fritz missed
simple tactics).

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.