Computer Chess Club Archives


Search

Terms

Messages

Subject: ply and time to search: fruit and rybka

Author: Joseph Ciarrochi

Date: 12:54:35 01/06/06

Go up one level in this thread


Interesting discussion folks, even if it was prompted my my mistaken "ply
tournmenet." :)

It makes sense that rybka uses deeper ply. that would at least explain the
ubsurd imbalance in these results.

But one more thing is wierd to me. I run rybka versus fruit in arena using fixed
ply of 3. I do this not for the result but because arena times each move and the
total game and i thought maybe it could give us some info about what is going
on.
 Well, it looks like fruit and rybka use the same amount of time. I expected
that rybka would take a bit more time, given it is probably doing deeper search.

I wonder if the arena clock is inaccurate at these low levels? Or are there
other interpretations


On January 06, 2006 at 07:19:47, Joseph Ciarrochi wrote:

>If you run tournments limiting ply to 3, Rybka absolutely kills fruit and
>fritz9.
>Rybka sure forms a good first impression.:)
>
>
>
>
>
>                                                 1             2            3
>1   Rybka 1.0 Beta 32-bit    ****       94.0 - 7.0    94.5 - 6.5
>189.5/203
>2   Fritz 9                7.0 - 94.0    ***          58.5 - 42.5
>77.0/220
>3   Fruit 2.2              6.5 - 94.5   42.5 - 58.5        **
>55.5/221
>
>
>
>These numbers seem so imbalanced. I wonder if there is not some sort of bug in
>the fritz gui or something.
>
>best
>Joseph



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.