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:
A new bus speed
Heat and Hertz
Test Configurations
Cachemem, Sandra
Everest and Speed Derating
Sysnark2004
Internet ContentCreation
Office Productivity
WorldBench5 -1
WorldBench5 -2
3D Rendering
3ds max
Cinebench2003
Futuremarks
Gaming A-Z
DOOM3 FarCry
Conclusions and Paradoxes

Give Us Some Feedback to Help Us Improve our Reviews

 Intel Pentium4 3.46 Extreme Edition / 925XE chipset
(Review by MS November 16)
Intel P4 560+ At:

Memory Performance

ARGUABLY, WE were running a bit out of spec here since we set in both DDR2 platforms the latencies to 3:3:3:8 (CL:tRCD:tRP and tRAS) at 533 MHz data rate whereas most specs call for either 4:4:4:12 or even 5:4:4:13 latency settings. To make a long story short, relaxing the timings to 4:4:4:12 gave us on average an approximately 4-5% performance hit and there is really no point in going there anyway since there are enough memory modules out there capable of sustaining the lower latencies. On the other hand, using higher latency modules made things downright ugly performance wise.


Cachemem 2.65

Memory access latencies for the 925XE at 1066 MHz host bus frequency (solid bars) compared to the 925X running at 800 MHz (transparent bars). in both cases, the memory frequency was 533 MHz data rate at 3:3:3:8 [t] latencies. At higher stride length and consequently an increased number of page misses we see a gain of almost 10% in data access for the new 925XE chipset running in pseudo-synchronous mode as opposed to the original 925X version that needed to translate the data from a double width 533 MHz memory bus (effective 1066 MHz) to an 800 MHz interface. Keep in mind that the CPUs used here are both ExtremeEditions with a low latency pipeline.

Intel P4 Northwood 2.4
(hard to find)

Memory Bandwidth

As long as we were running SiSoft Sandra, the system behaved as expected ... sort of, at least.

Sandra 2004 SP2

There is clearly an improvement of the 925XE chipset over the older 925X chipset. However, as we alluded to earlier, there are also differences between the different CPUs, i.e. Northwood / Gallatin vs. Prescott. Not only that, but there is also a difference between the 560 and the 550 processor.

Basically the same picture as above and we might have missed something, had we left it at that.

next page:     => Memory Performance Part II =>

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