Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: It makes no sense to test Shredder 5 at all ...

Author: Mike S.

Date: 07:45:26 05/30/01

Go up one level in this thread


On May 30, 2001 at 02:37:03, Jouni Uski wrote:

>... because Deep Shredder is 1) faster even with single CPU as 99,99% of PC
>are and will be(!) 2) better in all test suites and 3) plays better than S5
>e.g. in my own test Deep Shredder - Shredder 5 21,5-18,5.

Could you post some examples where Deep Shredder 5 is (significantly) better on
single CPU, than Shredder 5?

I have tried this comparison with 8 test positions, where Shredder 5 had needed
mid-range solving times. I 5 of these, DS's result was within ±10%, one was
better, two were worse. Do you have such data (equal/better/worse)?

I'd like to decide if Shredder 5, or Deep Shredder 5 to use on single CPU... In
general, if no important other improvements have been added, I would assume that
the SMP version has a small overhead or something like this (additional, but
useless code for single CPU operation) which results in a small disadvantage.

Maybe SMP experts can acknowledge this, or tell that it's nonsense :o) ?

Thanks,
M.Scheidl



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.