Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: not using nullmove?

Author: Tim Foden

Date: 04:00:53 02/18/04

Go up one level in this thread


On February 18, 2004 at 04:24:54, Dann Corbit wrote:

>On February 18, 2004 at 03:36:14, Tim Foden wrote:
>
>>On February 17, 2004 at 20:55:13, Dann Corbit wrote:
>>
>>>Here's a dumb idea:
>>>
>>>Write a program to scan a Nalimov database, but throw away everything except
>>>won/lost/drawn/broken (needs 2 bits per reflected board position to store the
>>>outcome state).
>>>
>>>Then write a table.
>>>
>>>For up to the 4 man tables, it should be really tiny and fit into ram without
>>>any fuss.
>>>
>>>Seems like one single program could write a recognizer for anything [for which a
>>>Nalimov or Edwards or Thompson EGTB exists].
>>
>>This is almost exactly the idea that is well documented in Heinz's book!  :)
>>
>>He gets all the required 3+4 piece tables to fit in just over 15 MB.
>
>I read his book, but I don't remember his suggestion of canibalization of
>existing tb files to create bitbases.  I do remember the notion of extreme
>compression of bitbases.

He generated his smaller bitbases from the Edwards ones.  He had to do some work
to generate correct KPKP ones though.

The indexing schemes are talked about extensively in chapter 5.  In chapter 6 he
talks about "Knowlegeable Endgame Tablebases", which use knowledge, including
"bitbases" to encode tablebase knowledge that can be held in memory.

Ch 6, p 100, para 3, "Interestingly enough, we derived the necessary a-priori
knowledge from Edwards' distance-to-mate tablebases [67] and their result
summaries as publicly available on the Internet.  Hence, Edwards' standard
tablebases provided the ombiscient feedback which not only enabled but actually
sparked the development of our knowledgeable endgame databaes in the first
place."

Ch 6, p 112, para 2, "We created the knowledgeable endgame databases of
DarkThought with the help of a double-checking tool that automatically converts
Edwards' tablebase files into our knowledgeable format."

Cheers, Tim.



This page took 0.02 seconds to execute

Last modified: Thu, 07 Jul 11 08:48:38 -0700

Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.