news
news archive
SEND NEWS!

articles
FORUMS!
links
contact

PC Database

Folding Team

SETI@HOME Team
RC5 Team
Genome Team

OCAU chat

SlotA Showdown - Page 4

Jump to page:

Artificial Benchmark:
As explained on the previous page, I have set the CPU's to 115MHz FSB, for a core speed of 805MHz, and set the L2 ratio on the Athlons to 1/3. This will demonstrate how much effect the smaller but faster L2 cache on the Thunderbirds can have:

Sandra - and yes, I am particularly proud of the nausea-inducing colour scheme of this graph :) ..:

The Thunderbird here shows an 8% increase in FPU memory speed. This drops to a 3.5% difference in CPU memory speed.. the rest stay identical, which is to be expected. In Quake3:

The "fastest" setting shows just under 4% difference.. High Quality is limited by the fillrate of the GeForce SDR again.

3DMark2000:

In 640x480x16, we see a nearly 7% performance lead by the Thunderbird. Again, hardly spectacular, but it's clear there's no performance hit from the smaller cache. Obviously, the higher speed outweighs the smaller L2 size. This is good news, because on-die cache is cheaper than off-die. To conclude with regards to performance, we can see that, in the same way that Intel's move from 512Kb of slower off-die cache to 256Kb of full-speed on-die cache (between the Katmai and Coppermine Pentium III's), AMD's "Thunderbird" update to the Athlon line has made them marginally faster at the same core speed.

Finally, I wanted to detail my experiences getting the Thunderbird working on the KA7, a KX133-chipset motherboard..

NEXT PAGE - KA7 <-> Thunderbird notes and Conclusions..

Major Sponsors:

All content copyright 1999-2002 James Rolfe. All rights reserved. No reproduction allowed without written permission.
Interested in advertising on OCAU? Contact us for info.