|
Advice Beginners BIOS Guide CPUs Links Mainboards Memory Network Storage Video/Sound Cards Contact Forum SiteMap Sponsors WebNews Home
|
. | . |
|
CPU Intel P4 840 D P4 820 D P4 630 P4 640 P4 650 P4 660 P4 670 AMD Athlon64 3500+ 3700+ 3800+ 4000+ X2-3800+ X2-4200+ X2-4400+ X2-4600+ X2-4800+ 1-Way Opteron Opteron 144 Opteron 146 Opteron 148 Opteron 150 Opteron 152 2-Way Opteron Opteron 240 Opteron 242 Opteron 244 Opteron 246 Opteron 248 Opteron 250 Opteron 252 2-Way Dual Core Opteron Opteron 270 Opteron 275 nVidia GF 7800GT GF 6800GT GF 6600GT ATI R X850 XT PE R X850 XT R X800 XT PE R X800 XT R X800 XL Memory Corsair Crucial Kingston Mushkin OCZ |
LOSTCIRCUITS |
|
| DDR-II Roundup The State of the Art | |
|
(Review by John Cook, May 19, 2005) |
|
256 MB OCZ Low Latency DDR (I) Starting at: |
Test Setup
| Motherboard: | Asus P5AD2-E Premium - Bios 1004 Official |
| CPU: | Intel 550-LGA775 3.4GHz |
| Power Supply : | Zippy HP2-6460P 460W |
| Case: | Silverstone TJ06 |
| Cooling: | AlphaCool Nexxos Bold with Single 120mm fan/radiator |
| OS: | Windows XP Professional w/SP2 |
This review could almost be considered phase 3 after two earlier drafts. While I still consider there to be limitations of this present write-up, I cannot justify extending the amount of time I have either put into or put off this review. In a perfect world I would not have encountered the problems I had in setting up a good DDR2 testing platform, however life is rarely so compliant, and again Murphy and his corresponding law were quick to strike in this case.
To elaborate, the motherboard I had originally chosen for the DDR2 roundup was the Abit AA8XE board based on the then new 925XE chipset. Officially supporting 1066fsb, I thought this should allow headroom for overclocking my 3.4GHz P4 by using the well known available 14x multiplier on the 3.4 and up P4 CPU’s with bios selection. While the board functioned fine at 266FSB or 1066 when factoring in the quad bus speed, there was basically no headroom afforded thereafter because of the PLL issue described in detail in an earlier report. While Asus had rectified this issue with a bios update to allow a higher bus VTT (termination voltage), Abit has yet to see fit to include this on the more budget/affordable AA8XE and chosen to only implement this fix on the much more expensive AA8XE Fatality board. I somewhat chuckle that I call the AA8XE a budget/affordable board given it’s $165 price point, however considering the $240+ alternatives in the Asus P5AD2-E and Fatality boards, the $165 tag seems relatively affordable. So with a board that couldn’t clock past 267MHz without the bus PLL jumping around and invalidating any data, I was left to source another option.
An Asus P5AD2-Premium board was obtained, based on the Intel 925X chipset which had shown itself to be a stable performer in many reviews. Unfortunately, above 270MHz front side bus, I would lose my network connection via the onbard NIC, and the PCI Express frequency made running an Nvidia card impossible. Actually, the Nvidia card didn’t like to be in the system past 245MHz bus or so, and therefore I had to use a Radeon X700Pro 256mb PCI Express card. This allowed higher bus speeds to be obtained, but again as the bus speed was raised, certain onboard features started to become fragile and I felt this invalidated the board as solid testing platform.
Hence the ASUS P5AD2-E Premium as the final board I gathered, and the one eventually used for testing. While more stable and not suffering the issues described above. I was limited to a bus speed of 285MHz of total 24/7 tortured stability. Anything past this point and I experienced anomalies such as random crashes, and or failure to post. While I have definitely seen many people on various forums and sites get much higher front side bus speeds, I can only describe to the readers that this is my experience, and that my CPU is not at present the limiting factor as I have taken the processor up towards 4.2GHz at the 17x multiplier and lower front side bus speeds. Therefore, I am left with a platform that I can test with relative assurance that data collected up to 285MHz is valid, but here clarified to the reader that this is a current limitation of testing and some modules tested may in fact test higher, and in all probability will do so.
|
OCZ 2 x 512 MB PC2-4200 DDR2 RAM |
next page: => The Modules: Corsair's Lineup =>
If you enjoyed reading this article and found it useful, please consider making a small donation to LostCircuits.