Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Tablebases are not perfect

Author: Robert Hyatt

Date: 05:35:03 10/08/98

Go up one level in this thread


On October 08, 1998 at 01:53:13, blass uri wrote:

>
>On October 08, 1998 at 00:13:15, jonathan Baxter wrote:
>
>
>>This system will handle winning positions with mate lengths from 1 to
>>126.
>>Losing positions may be from "lost in 0" (checkmated) up to "lost in
>>125".
>>All draws are zero.
>
>This is a mistake because the 50 moves rule should be considered.
>There are positions of mate in N in KRB vs KR that are practically draw because
>of the 50 move rule and the program may do mistake in this evaluation.
>
>Uri


I'm not aware of any deep mates in KRB vs KR...  this is generally a quick
draw...  ditto for KRN vs KR.  The mates are pretty rare in this opening, and
in setting up a few random ones, I couldn't create a single won position unless
I had the losing side set up hanging the rook immediately...

I believe this is basically a draw, however...

things like KNN  vs KP certainly has deep mates beyond 50 however...



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.