Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Importance of L2 cache speed/size for diff programs (was:..Genius speed?

Author: fca

Date: 07:44:40 08/12/98

Go up one level in this thread


On August 12, 1998 at 06:59:14, Tom Kerrigan wrote:

>On August 12, 1998 at 04:48:56, fca wrote:
>
>>I tend to agree with Bob - as you recall, the parentage of this discussion was
>>my surprise that such a huge speed ratio for Junior existed between the P200MMX
>>and P2/300, which others later attributed to L2-cache size/MHz.  I disagreed.
>
>I didn't say L2 cache is the only reason it's faster.

Agreed.

>The P6 core is maybe 35-75% faster than the P5 core. I assume you knew this.

A safe assumption :-)

> I was just pointing out
>that with the faster core and the faster L2 cache speed, the performance
>difference you see is explainable.

You wrote:
"Aside from software differences, the Pentium MMX/200 has a 66MHz L2 cache
(possibly smaller than 512k) whereas the Pentium II/300 has a 512k 150MHz L2
cache. If a program really bangs on the L2 cache, it will go much faster on the
Pentium II."

But since the core (P2/300 vs P200MMX) is so much faster, the extra/faster cache
(even if accessed a lot) simply serves to alleviate what the faster core would
*otherwise* have made into a bottleneck.  L2-hit rates etc suggest in itself
this would not be able to increase the speed ratio above that the cores deliver.

So, I am still surprised at the 2.5x reported.  Aren't you?

Also, I am not sure why the F5'd Junior engine should run **20%** slower than in
native, on identical hardware.  What am I missing here?

Kind regards

fca



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.