Gigabyte 7NNXP (nForce2 Ultra 400): Gigabyte Goes nForce
by Evan Lieb on July 5, 2003 10:53 PM EST- Posted in
- Motherboards
Gigabyte 7NNXP: Stress Testing
We performed stress tests on the 7NNXP in several different areas and configurations, including:
1. Chipset and motherboard stress testing was conducted by running the FSB at 222MHz
2. Memory stress testing was conducted by running RAM at 400MHz with two DIMM slots filled and at 400MHz with all four DIMM slots filled in dual DDR mode at the lowest timings possible.
Front Side Bus Stress Test Results:
As standard practice, we ran a large load of stress tests and benchmarks to ensure the 7NNXP was absolutely stable at each overclocked FSB speed. These stress tests included Prime95 torture tests, which were run in the background for a total of 24 hours.
We proceeded to run several other tasks, such as data compression, various DX8 games, and light apps like Word and Excel with Prime95 running in the background. Finally, we reran our entire benchmark suite, which includes Sysmark 2002, Quake3 Arena, Unreal Tournament 2003, SPECviewperf 7.0, Jedi Knight 2 and XMPEG. In the end, 222MHz FSB was the highest overclock we were able to achieve with the 7NNXP without encountering any reliability issues.
Memory Stress Test Results:
This memory stress test examines the motherboard’s ability to operate at its officially supported memory frequency (400MHz DDR), and at the lowest supported memory timings that our Corsair TwinX LL modules support:
| Stable Dual DDR400 Timings (2/4 banks populated) |
|
| Clock Speed: | 200MHz |
| Timing Mode: | N/A |
| CAS Latency: | 2.0 |
| Bank Interleave: | N/A |
| RAS to CAS Delay: | 2T |
| RAS Precharge: | 4T |
| Precharge Delay: | 2T |
| Command Rate: | N/A |
It’s not surprising to see that the 7NNXP achieved such low memory timings with just two memory modules spec’ed at DDR400 running in dual channel DDR400 mode. It is standard for any P4 or Athlon XP motherboard to achieve these aggressive performance settings, though only nForce2 and nForce2 Ultra 400 motherboards can achieve a RAS Precharge as low as 4T. However, we suggest you run your memory at CAS 2-2-2-5 instead, as we discovered some time ago that CAS 2-2-2-4 is a bit slower than CAS 2-2-2-5. Of course, the difference will never be noticeable in real world usage.
The following memory stress test is obviously a bit more strenuous on the memory subsystem than most memory stress tests, as it tests for stability when a desktop user installs four DIMMs running in dual DDR400 mode at the most aggressive memory timings attainable in the BIOS:
| Stable Dual DDR400 Timings (4/4 banks populated) |
|
| Clock Speed: | 200MHz |
| Timing Mode: | N/A |
| CAS Latency: | 2.0 |
| Bank Interleave: | N/A |
| RAS to CAS Delay: | 3T |
| RAS Precharge: | 6T |
| Precharge Delay: | 2T |
| Command Rate: | N/A |
It is not unusual to see such lax timings for such a strenuous memory subsystem test such as this. However, we have seen better and so in this case, the 7NNXP does not perform as well as most other modern-day motherboards. Again, the real world performance difference is negligible, so we urge you not to flip out ove small variances in memory timings as seen in the two charts above.
We tested all these memory timings using several stress tests and general applications to guarantee stability. We initiated the tests by running Prime95 torture tests; a grand total of 24 hours of Prime95 was successfully run at the timings listed in the above charts. We also ran Sciencemark (memory tests only) and Super Pi. All three stress tests could not make the 7NNXP fail at the timings listed in the above charts.
Facebook
Twitter
RSS
29 Comments
View All Comments
Anonymous User - Tuesday, July 8, 2003 - link
Yes, uhm, can we go back to GIF? I hate Flash. Look, I'm on dialup, and I'm lagged a LOT by big downloads, but I still prefer GIFs to Flash (which seems to load a tad faster). My friends all agree, and most of them are dialuppers too. Please, at least put up a poll or something so you can see how we all really feel about this stupid Flash stuff.NovaPolice - Monday, July 7, 2003 - link
It looked very good on paper but it hated every brand of ddr I had on hand. I went through a couple gigs of sticks before deciding to send it back.Anonymous User - Monday, July 7, 2003 - link
I have tested both this board, and the MSI K7N2Delta. The MSI board was FAR MORE stable, and a bit (7%) faster. I wonder why in this test the Gigabyte board is compared with an EPOX via board: Just to let it outperform, which is otherwise not possible?Anonymous User - Monday, July 7, 2003 - link
They say that CPU voltage is adjustable only up to 1.85v and that chipset voltage is adjustable by 5, 7.5 or 10%. I believe this to be wrong however. What the reviewer believes to be chipset voltage is actually CPU voltage also!! Check out the review over at Digital Daily and they show how voltage up to 2.035v is possible using a combination of these 2 seperate CPU voltage controls! Quite a large difference and large overlook by the reviewer.WooDaddy - Monday, July 7, 2003 - link
Evan, are you or anyone at Anandtech going to discuss the differences between the Ultra 400 and the standard nForce 2 chipsets? Maybe a general review with reference boards? Is there a big performance difference or limiting factor for future upgrades?Anonymous User - Monday, July 7, 2003 - link
The review does not include information on the presence of Heat Sink Mounting Holes. Suggestion, add info on mounting holes or lack of to all motherboard reveiws.Anonymous User - Monday, July 7, 2003 - link
It appears that the performance test results in the Gigabyte 7NNXP review are presented in Macromedia Flash Player 6.This is unfortunate, since I am among the many web users who dislike the Flash format, and resent the numerous pop-up "ads" requesting that version 6 be installed.
I regret that I will be unable to view these results, to find out what additional complexities required them to be displayed in Flash, rather than a standard, multi-platform/browser form such as GIFs. And I hope that this is not an indication that future results will be Flash-formatted, since I am a frequent AnandTech reader and will miss whatever informative content is Flashed.
Thanks.
Anonymous User - Sunday, July 6, 2003 - link
The Anandtech review claims Gigabyte replied in 33 hours..I Wonder who in Gigabyte Should I contact ,Or what e-mail address i should mail to,
To get a Reply in ANY time frame.
CrystalBay - Sunday, July 6, 2003 - link
Nice evaluation...However perhaps the six phase power DMS prolongs the overall life of these boards, rather than inherently more stable overclocking...Replacing a board thirteen months, two or three years down the line (If lucky) is common place...I like to see more board makers produce the bare bones 400's like Solteks NV400-64L...Why because these boards can be thrown away once a year or so...