Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: SSDF(Gromit 3.11.9 - Rebel Century 4)A1200, 2-2

Author: Robert Hyatt

Date: 12:00:08 08/26/02

Go up one level in this thread


On August 26, 2002 at 11:19:23, Peter Berger wrote:

>On August 26, 2002 at 11:03:48, Robert Hyatt wrote:
>
>>>'Tablebases'
>>>
>>>--
>>>GCP
>>
>>
>>Terrible argument.  Why?  If I build my own tablebases, the data will be
>>_identical_ to the tablebases you build, in terms of content.
>
>Will they ? Bruce Moreland posted that his tablebases are aware of the 50 moves
>rule for example.


Not that I know of.  Bruce did distance to mate, and every distance to mate
tablebase suffers from the 50-move-rule problem.

But even if someone _did_ include that, anybody else that does the same will
have _exactly_ the same "content".  Tables are finite and precise in what they
contain.  No variability from one to the next.  Perhaps different organization
and different index schemes.  But when you probe one you get _exactly_ the same
result as when you probe someone else's.  I used both Nalimov and Edwards for
quite a while, with no problems.

Books are nothing like that of course.

There the content is definitely unique.



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.