ABOUT BENCH FORUMS PODCAST LOGIN REGISTER
PC Components
Smartphones & tablets
Systems
ENTERPRISE & IT
GUIDES

FIRST LOOK - nForce4: Gigabyte K8NXP-9

 by Wesley Fink on November 12, 2004 8:00 PM EST
49 Comments | Add A Comment
49 Comments

Overclocking and Memory Stress Testing

With the limited ranges of CPU clock speeds available in the pre-release Gigabyte BIOS, it was not possible to test overclocking thoroughly. We were able to reduce multipliers and easily reach 250FSB, which was the top speed available in the pre-release BIOS. However, until we receive the updated BIOS, we cannot determine the full overclocking capabilities of the K8NXP-9.

Since there have been many reports of issues with memory on the earlier Gigabyte nForce3 Ultra, more effort went into testing memory performance on the K8NXP-9. Our memory stress test measures the ability of the K8NXP-9 to operate at its officially supported memory frequency (400MHz DDR), at the lowest memory timings that OCZ PC3200 Platinum Rev. 2 modules will support. All DIMMs used for stress testing were 512MB double-sided (or double-bank) memory. To make sure that memory performed properly in Dual-Channel mode, memory was only tested using either one dual-channel (2 DIMMs) or 2 dual-channels (4 DIMMs).

Stable DDR400 Timings - One Dual-Channel
(2/4 DIMMs populated)
Clock Speed: 200MHz
CAS Latency: 2.0
RAS to CAS Delay: 2T
RAS Precharge: 10T*
Precharge Delay: 2T
Command Rate: 1T
*Several memory tests have shown that memory performs fastest on the nVidia nForce chipsets at a TRas (RAS Precharge) settings in the 9 to 13 range. Memory Bandwidth tests were run with memtest86 with TRas settings from 5 to 15 at a wide range of different memory speeds. The best bandwidth was consistently at 9 to 11 at every speed, with TRas 10 always in the best range at every speed. The performance improvement at TRas 10 was only 2% to 4% over TRas 5 and 6 depending on the speed, but the performance advantage was consistent across all tests. All benchmarks were therefore run at a TRas setting of 10.

Using two DIMMs in Dual-Channel 128-bit mode, the memory performed in all benchmarks at the fastest 2-2-2-10 timings, at default voltage. As a further test, we ran every pair of PC3200 DIMMs that we had in the lab at their fastest timings, at DDR400 in DC mode. We had no problem with any RAM that we tested. It appears that the problems with Dual-Channel mode and memory compatibility, which were reported on the earlier Gigabyte nF3 Ultra board, have been improved greatly or eliminated on this nForce4 version.

Stable DDR400 Timings - 4 DIMMs
(4/4 DIMMs populated)
Clock Speed: 200MHz
CAS Latency: 2.0
RAS to CAS Delay: 2T
RAS Precharge: 10T*
Precharge Delay: 2T
Command Rate: 2T

Tests with all four DIMM slots populated on the Gigabyte required a 2T Command Rate with 4 DIMMs in two dual channels. This is the pattern seen on other top-performing Socket 939 boards. However, an "Auto" setting for Command Rate would not boot, and we had to force a 2T setting in BIOS with 4 DIMMs. Once 2T was set in BIOS, there was no problem running 4 DS DIMMs at the same aggressive 2-2-2-10 timings we had used with 2 DIMMs.

Basic Features Test Setup
PRINT THIS ARTICLE
Comments Locked

49 Comments

View All Comments

  • arswihart - Saturday, November 13, 2004 - link

    quote by Wesley Fink:
    "#6 - Full performance comparisons of nForce3 Ultra and nForce4 were run at nF4 launch at http://www.anandtech.com/cpuchipsets/showdoc.aspx?... Performance of nF3 and nF4 is basically the same - the only real difference is PCIe instead of AGP."

    but you are showing here that the production board from Gigabyte is a little different than the reference nf4 boards. I think it would be helpful to at least include one good nf3-250gb board in some future nf4 round-up or review, for comparison's sake, as i think its more practical at this point to compare nf3 to nf4, rather than reference nf4 to production nf4.

    Thanks for the review though, and I'm also interested in the price of these nf4 boards. I've seen somewhere quoting these boards on average at like $180

  • Whizzmo - Saturday, November 13, 2004 - link

    Possible Typo:
    Page 2, below the second mobo pic, the following:

    Four ports are 3Gb/s ports provided by the nForce3 chip, and

    should probably be:

    Four ports are 3Gb/s ports provided by the nForce4 chip, and


    Danke :)

  • johnsonx - Saturday, November 13, 2004 - link

    oh, ok... 1.5Gbps and 3Gbps signalling rates, which translate down to 150MB/s and 300MB/s data rates, respectively. The SATA uses 8b/10b encoding, so 10 bits of signalling are need for each 8 bits of data.

    Anyway, nevermind.

  • johnsonx - Saturday, November 13, 2004 - link

    ok, it's late, so I may be tired and crazy....

    but what's all this about 1.5Gb SATA and 3Gb SATA? I thought standard SATA is 150MB/s (SATA-150), while the new SATA 2.0 spec runs at 300MB/s. Even converting those speeds to Gigabits per second, you get 1.2Gbs and 2.4Gbps.

  • Jalf - Saturday, November 13, 2004 - link

    Would be nice to see it compared to ATI's A64 board. That looked like a pretty good performer as well

  • stelleg151 - Saturday, November 13, 2004 - link

    Any clues as to when we will be able to get our hands on one?

  • RyanVM - Saturday, November 13, 2004 - link

    Too bad the secondary SATA controller isn't on the PCIe bus.

  • xtknight - Saturday, November 13, 2004 - link

    wow, very interesting. looks like the gigabyte mobo is a winner. by the way, doom 3 belongs under OpenGL benchmarks.

  • PorBleemo - Friday, November 12, 2004 - link

    So much for that "Fatal1ty". :P

  • ProviaFan - Friday, November 12, 2004 - link

    Page 2 "Four ports are 3Gb/s ports provided by the nForce3 chip"

    Oops?
Copyright © 2024. All rights reserved.
BENCH
TOPICS
FOLLOW
ABOUT

Log in

Don't have an account? Sign up now