Navigate:

Advice
Beginners
BIOS Guide
CPUs
Links
Mainboards
Memory
Network
Storage
Video/Sound Cards

Contact
Forum
SiteMap
Sponsors
WebNews
Home
. .

Prices:

Mainboards

ABIT
ASUS
Chaintech
Shuttle
Soyo
Tyan

CPU
Intel
P4 2.4C-800
P4 2.6C-800
P4 2.8C-800
P4 3.0-800
P4 3.2-800

AMD
AthlonXP
XP 1700+
XP 2000+
XP 2400+
XP 2500+
XP 2700+
XP 3000+
XP 3200+

Athlon64
Athlon64 3200+
Athlon64 FX-51

Opteron
Opteron 240
Opteron 242
Opteron 244
Opteron 246

Memory

Corsair
Crucial
Kingston
Mushkin
OCZ

Search Prices:


























































































































LOSTCIRCUITS

SHORTCUTS:
The Multiplier Legacy
Validating the New, Supporting the Old
Apples to Apples and Clocks to Clocks
System Disclosure, SiSoft Sandra Memory
PCMark2002, A Different Perspective
Benchmarks that won't: CodeCreatures, Comanche4
SPEC ViewPERF and diminishing returns
Fraggin the Quake3 Arena Scores
Expendable and the Chipset Frequency, 3DMark2001SE
3DMark2001SE: Behind the Looking Glass
Profiling Performance, Will Barton Challenge Hammer?
AMD Processor Steals
 333 MHz FSB for the Athlon   
The Unkept Promises
(Review by MS, September 2, 2002)
PCMark 2002

PCMark2002 is a perfect application for testing the different speed settings since the CPU and HDD scores measured in parallel to the bandwidth provide an internal control for the accuracy of the benchmark as well as the validity of some interpretations that we had earlier, i.e. no real changes in CPU and/ or IDE performance (the latter being dependent among other things on the PCI bus frequency) whereas the memory performance should vary. Unlike the situation in SiSoft Sandra, PCMark2002 includes much more complicated tasks than just scrolling down the column addresses and prefetching the bits from adjacent locations. Therefore, we do not expect the bus saturation that evens out the PC2100 and PC2700 scores courtesy of the FSB bottleneck.


The deck is stacked differently in PCMark than in SiSoft Sandra: increasing the memory bus frequency from 266 to 333 MHz data rate yields an 8% bandwidth increase that is topped by an additional 9.5% increase when the FSB is increased to 333 MHz as well.

It is really important here to stress the fact that only a minority of application uses streaming data transfer mode, the rest are a mixture of random and "interrupted" in page access, where interrupted means that cache accesses may negate in page accesses and / or eventually lead to page expiration.

next page:    => 3D Applications =>

Click here! All advice and educational articles on LostCircuits are free, but if you feel you can, please make a small donation to us!
Thank you!

General disclaimer: This page only reflects the author's personal opinion and assumes no responsibility whatsoever regarding any of the contents or any damages that may occur explicitly or implicitly from reading the contents of this site. All names and trademarks mentioned in this review are the exclusive property of the respective parent companies.
All contents of this site are protected by international copyright laws. Reproduction of the contents even in parts is not allowed except after written permission by the author and referral to this site.
Copyright 2002 - 2008 LostCircuits