Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Rybka code

Author: Chrilly Donninger

Date: 07:51:24 12/12/05

Go up one level in this thread


On December 12, 2005 at 09:25:42, Peter Eizenhammer wrote:

>
>>
> My main point was:
>>Its not MTD().
>>Chrilly
>
>I still don't get the point:
>Am I right to say that your basis thesis was:
>Fruit uses MTD, Rybka does not, so -> no clone.
>
>Well, Fruit does not use MTD, so what is left of your argument?
>Or do I miss something?
>
>Peter

Sorry. I should have written 2 different postings. I mixed up two messages in 2
posting.
Message 1 is: Rybka uses PVS (or Alpha-Beta) and not MTD(). This refers to
discussions I found in the archive, where Vasik Rajlich mentioned that he uses
in his programm MTD().

Message 2 is: I have looked in the code and found no direct similarieties with
Fruit. Both use PVS, but almost all programms (besides SOS) use PVS. Thats no
argument for a clone and of course no argument that is no clone.
According my inspections are the board-representation, the movegenerator, the
eval and also the search different. Its unavoidable that there are some general
techniques like PVS in common. Every programm builds on the known start of the
art.

Chrilly







This page took 0.04 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.