Memory Stress Testing: Epox 9NDA3+

The memory stress test is very basic. It simply tests the ability of the Epox 9NDA3+ to operate at its officially supported memory frequency (400MHz DDR) at the lowest memory timings OCZ PC3200 Platinum Rev. 2 modules will support. All DIMMs used for stress testing were 512MB double-sided (or double-bank) memory. Memory was only tested in Dual-Channel mode 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. We ran our own Memory Bandwidth tests 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 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 2.6V.

Stable DDR400 Timings - 2 DIMMs
(2/4 DIMMs populated)
Clock Speed: 166MHz
CAS Latency: 2.0
RAS to CAS Delay: 2T
RAS Precharge: 10T*
Precharge Delay: 2T
Command Rate: 2T
*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. We ran our own Memory Bandwidth tests 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 run at a TRas setting of 10.

Tests with all four DIMM slots populated on the Epox board were something of a disappointment, however. Most 939 boards that we have tested require 2T Command Rate with 4 DIMMs in two dual channels. Not surprisingly, 2T was also required with 4 dimms on the 9NDA3+. However, this Epox is the only 939 board we have tested that absolutely forces DDR333 when four DIMMs are installed. Some other boards that we have tested also try to force DDR333 with 4 DIMMs, but this can normally be corrected to DDR400 by selecting that speed in BIOS. However, the Epox sets DDR333 with 4 DIMMs even if you specify DDR400. With no way to even try DDR400 with 4 DIMMs, we are left with the forced DDR333 setting with 4 dimms in our memory tests.

Epox can possibly correct this strange 4-DIMM memory behavior with a BIOS update. For now, if you absolutely need a 939 board that supports 4 DIMMs at DDR400, you will need to look at other 939 motherboards, since the Epox is the only 939 board that we have tested which cannot run at DDR400 with 4 DIMMs. Of course, we did not need to slow memory timings at all with 4 DIMMs, though we would hardly expect that with memory running at DDR333 instead of DDR400.

FSB Overclocking: Epox 9NDA3+ Performance Test Configuration: Epox 9NDA3+
Comments Locked

36 Comments

View All Comments

  • TrogdorJW - Monday, October 25, 2004 - link

    Wes (#23) - I'm not 100% positive, but I'm almost sure that UT2K3/UT2K4 are DX8.1 in terms of the 3D code, but they require DX9 in order to run. Sort of like how Doom 3 is an OGL game but requires DX9 for input/output functions. AFAIK, no Unreal engine game has shipped with DX9 features, but that will probably change soon.

    You could even make the argument that most of the UT engine is still DX7, with only a few DX8 additions. What does that mean? That pixel and vertex shaders are not absolutely necessary to get great graphics. Useful? Yes. Required? Nope. :)
  • AtaStrumf - Monday, October 25, 2004 - link

    I had 2 Epox nForce 2 Ultra 400 boards (got a new one after I RMA-ed the first one) and not one worked perfectly. That was reason enough for me to stay clear of Epox boards from that moment on. After reading tihs review, it seems that they still send out unfinished/not properly tested products, so I sugest to all to pick some other brand, because something is systematicly wrong at Epox. They are just not a good choice for the enthusiast anymore.

    ABIT seems to have improved a bit though, but I don't garantee anything.
  • NedFlanders - Monday, October 25, 2004 - link

    any new info on the FSB for the ep-9nda3+ when using 4 sticks? Epox's website has no owner's manual for this board on ANY of their worldwide websites. No BIOSs either.
  • Wesley Fink - Monday, October 25, 2004 - link

    #22 -
    My UT2004 package says "DirectX version 9.0b or higher required." The website says DirectX 8.1. If someone can provide a definitive answer and a link I will change the category.
  • CrystalBay - Monday, October 25, 2004 - link

    Hi Wes, I'm just curious why UT2k4 is considered a DX9 benchmark ?
  • Bonesdad - Monday, October 25, 2004 - link

    I think the review was quite good. I did not get the sense that this is a "positive" review or even a wholly "negative" review. I think you did an admirable job of giving readers your observations and letting them make the final purchase decision. Which is the goal of a good review.

    Based on current information about this board, I will skip it too. I'll stick with my 8RDA+ until the field is more populated with 939s.
  • NedFlanders - Monday, October 25, 2004 - link

    my Epox 9kda3+ is en route. The whole reason why i waited for 939 was to use all of my RAM sticks (4). Should I just return this thing un-opened to newegg or is there the potential that i can run 4 256 hyperX's at 400?

    Thanksdidilyanks
  • Gholam - Monday, October 25, 2004 - link

    Does Cool'n'Quiet work on this board with four DIMMs? Because 8KDA3+ has this annoying problem with Cool'n'Quiet not working with two or three DIMMs present...
  • ksherman - Monday, October 25, 2004 - link

    Thanks for changing the article title Wesley! Now its more representative of the review ;)
  • LocutusX - Monday, October 25, 2004 - link

    This board seems like "too little, too late". Still, I think AT has done a good job on this review as it has indicated where most of the problem areas lie so people can decide for themselves if its an important issue or not. As for me, I`d skip this one.

Log in

Don't have an account? Sign up now