Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: The Rybka Flamewar & question for Vasik

Author: Kirill Kryukov

Date: 00:45:08 02/17/06

Go up one level in this thread


Hi Vincent,

On February 16, 2006 at 21:05:54, Vincent Diepeveen wrote:

>Moves say nothing. Add in fruit a small kingsafety and it looks less like Fruit
>than any other program.

That would be an interesting experiment indeed.


>That's the point Chrilly and others are making. Their point is they can read
>assembly and statements done in this forum do not reflect what he sees posted
>there.
>
>Enrico makes the point there is also a discrepancy in nodes a second. At slow
>hardware it gets initially the same nps first 5 ply like it gets at hardware
>factor 20 faster; at bigger depths then at the slow hardware it gets retuned
>then the output of nps and also at the faster hardware it gets retuned.
>
>Again an inconsistency between the real truth and shown truth in the display.
>
>Why all this show and fanfare?
>
>Vincent

I believe there is slight misunderstanding because of lacking explanatino of
correlation tables. I will try to add detailed explanation soon.

The good point of those correlation stats is that they show objective comparison
of engine playing style, derived simply from their games. So it does not depend
on disassembling and any subjective observations, but only on the evaulation and
moves in the games.

Best,
Kirill



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.