Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Maybe you're right but maybe you're not!

Author: Russell Reagan

Date: 13:32:04 04/07/03

Go up one level in this thread


On April 07, 2003 at 16:19:26, Albert Bertilsson wrote:

>Hi Russell!
>
>I've thought about that, and look up table is faster than the PopCount16 that I
>have. This is when testing is done is repetetive loops but in the real program
>I'm worried about cache misses. For the PopCount I prefer then non look up
>version, still I wonder if anybody knows of a way to speed it up.

If you're worried about cache misses, use a 256 element table. I can't imagine
that table not being in the cache on any modern processor when it is used very
frequently.

>For the IndexOfAnyOne function I have nothing even close to look up tables in
>performance, but if anyone can think of something...

The only thing I can think of is bsf on the PIII, which I believe is 1 or 2
cycles for 32-bit values.



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.