Memory Stress Testing

The SLI version of the Gigabyte nForce4 handled memory exactly as we found on the K8NXP-9, which is very good indeed. We did not see any issues with memory on this board that have been reported as problems on the older nForce3 Ultra Gigabyte boards.

The memory stress tests measure the ability of the K8NXP-SLI 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.

By 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.

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.

Overclocking: Gigabyte K8NXP-SLI Test Setup
Comments Locked

58 Comments

View All Comments

  • bob661 - Thursday, November 25, 2004 - link

    Actually, there were quite a few people that asked for resolutions higher than 16x12. I think for SLI benches, high res testing is probably the best. People that didn't read the initial article would get the wrong impression of SLI.
  • Omega215D - Thursday, November 25, 2004 - link

    what i'd like to know is when are they going to start phasing out those useless serial ports and maybe the LPT printer ports? I figure if your going to spend money on performance parts might as well get USB printers and game controllers.

    Also, any idea on when USB will be fully bootable that way PS2 ports can be phased out? Just wanna save on IRQs and make room for more USB's or Firewires so I don't have to rely on slot taking expansions.
  • Wesley Fink - Thursday, November 25, 2004 - link

    #2 - Read on in the review. I did test at 1280 x1024 also because 1024x768 was limiting in some benches - but not as much as you might expect. The 1024x768 was provided mostly to allow comparison to past motherboard reviews.

    I believe most will run 1280x1024 which is doable with most 19" flat panels. The cheapest flat panel I know of that runs 1600x1200 resolution is the 20.1" Dell at about $750. That's 16x12 entry level.

    Most any decent 19" CRT can support 1600x1200 as you stated, but have you ever tried to play a game at 1600x1200 on a 19" CRT. I tried it just to see for this review and it was pretty ugly. However 16x12 was OK on the 22" Diamondtron, though I prefer 1280x1024 on the 19" flat panels for most gaming.

    Anand got criticized for gearing everything in his SLI launch review to 1600x1200, so it figures I would get criticized for 1280x1024 which was suggested in the comments on his review. Guess we will never please everybody.
  • AnnoyedGrunt - Thursday, November 25, 2004 - link

    Well, this was a Mobo review, so that's one reason why it would make sense to run in resolutions that were CPU limited instead of video card limited. I doubt there would be much of a difference in SLI performance between the ASUS and Gigabyte solutions (although it certainly wouldn't hurt to have more data).

    -D'oh!
  • Decoder - Thursday, November 25, 2004 - link

    #2 . I agree with you 100%. I run my Dell 2001FP at 1600x1200 and i will going SLI so that i can play games at 1600x1200 only.
  • Decoder - Thursday, November 25, 2004 - link

  • cnq - Thursday, November 25, 2004 - link

    "Since 1600x1200 normally requires a 20" or larger flat panel monitor, we did not report 1600x1200 results, since most readers will not run at that resolution."

    Wesley, with all due respect, there is not a single reader who is waiting to buy SLI who says to himself, "Boy I'd really like to see how fast SLI does at middling resolutions. Yeah, that's why I'm splurging thousands for an SLI rig. 1024x768 all the way baby!"

    *ALL* SLI buyers will be running EVERYTHING at 16x12 or even 20x15, since current single card solutions can keep up at 10x7 and 12x10. No one would be dumb enough to pair up a couple of 68GT's or 68U's and then limit themselves to resolutions that a single 66GT could handle.

    People who buy SLI are going to have money to spend, so don't worry, their monitors WILL be able to handle 16x12. An ordinary $250 19" CRT can do that, much less the kind of monitors that SLI fanatics will have.

    Any SLI prospective purchasers out there planning to run their games at 1024x768? Didn't think so.

    C'mon Wesley, really. You're doing quite a disservice benchmarking SLI at resolutions that aren't even graphics card limited in the first place.
  • HiroProt - Wednesday, November 24, 2004 - link

    Great review, Wesley!

    Looks like Gigabyte has done it this time around. The question is: when can we buy these boards? I assume that the non-SLI version will be available first, since it already hit 1.0.

    Do you have any more availability info, Wesley?

Log in

Don't have an account? Sign up now