NV40 & The Test

Coming off of the extremely poor showing of their 5000 series and its associated Detonator drivers, the NV40 design represents an interesting design from NVIDIA born amid the bitter taste of driver scandals and overall inferior performance. NV40 not only brought the introduction of Shader Model 3.0, further blurring the lines between a CPU and a GPU with the addition of more high-level programming abilities, but also reintroduced the world to SLI when the NV40 went to PCI-Express. Overall the NV40 is a very different design than anything preceding it, giving NVIDIA a blank slate from which to work out performance improvements.

It should be noted that our driver selection policy is a little different for NVIDIA than it was with ATI, due to NVIDIA's previously inconsistent official driver release schedule, which did not see official drivers released often enough to fill our intended 2-month gap between driver revisions. Because of this, some of the drivers used in this test are NVIDIA driver releases that were not directly released to the public by NVIDIA itself (e.g. releases by OEMs who received multiple NVIDIA builds). However given the simply enormous number of such drivers, we used only Windows Hardware Quality Labs (WHQL) certified drivers, which means these are drivers NVIDIA was confident enough to release in a final form and submit to testing to Microsoft. Overall these drivers still stick to an approximate 2-month gap, making the results as comparable as possible with our ATI Catalyst results.

What hasn't changed is our primary game list, though we'll be adding a few additional titles with more limiting testing (see page 11). As with the X800 Pro, we'll be conducting extensive testing with the following games:
  • X2: The Threat
  • Doom 3
  • Half-Life 2
  • Far Cry
  • Battlefield 2
  • 3DMark05
  • D3DAFTester
We have also set aside several games that are console ports, for a comparison on how this influences performance changes(more on that later). Those games are:
  • Final Fantasy XI, Benchmark 2
  • Need For Speed Underground 2
  • The Chronicles of Riddick
  • Serious Sam 2
Our benchmarking setup for this series also remains unchanged, and is once again the following:

Benchmarking Testbed
Processor: AMD Athlon 64 3400+(S754)
Motherboard: Abit KV8-MAX3
Memory: 2GB DDR400 RAM 2:2:2
Hard Drive: 120GB Maxtor DiamondMax Plus 9
Power Supply: Antec TruePower 430W

All tests were done at 1280x1024 unless otherwise noted.

Index D3DAFTester
Comments Locked


View All Comments

  • Schugy - Thursday, May 11, 2006 - link

    nVIDIA has done quite a good job with linux drivers. It's simple to run this script and start your favourite game afterwards. nVIDIA's reward will be that I`ll buy a GF7800GT for AGP when it comes out. I wonder why the last anandtech linux story is more than 10 months old.
  • KickusAssius - Sunday, May 14, 2006 - link

    I have owned the 9000, 9500 PRO and 9700 PRO and I just hate ATI's drivers. I had problems with at least half of the games I played. Gradually they fixed certain problems, but other problems were never resolved even after contacting ATI directly via email and I was not the only one.

    Prior to this, I owned a Geforce 256 and Geforce 4 MX and Nvidia's drivers were nothing spectacular, but they always worked.

    Now, I have owned the 6800GT, and the 7800GTX and the drivers have been excellent for a long time now. The only problems I ever had were in CSS, but Nvidia fixed that problem relatively quickly. Also, I have never had a system crash as a result of Nvidia drivers, but several times this happened with my 9700 PRO, (still loved the card though especially when drivers got better). I think that now both sides have excellent drivers, though ATI's control panel is simply annoying.
  • johnsonx - Thursday, May 11, 2006 - link

    The image for Doom3 with the 60.72 driver can't be right; in the text you say there are major rendering issues with th 60.72 driver, but both images (normal and 'moused-over') look fine to me; the image presented as the 60.72 image looks identical to the 61.76 image presented below it.

  • ozzimark - Thursday, May 11, 2006 - link

    the pictures are too small for me to really even see anything
  • johnsonx - Thursday, May 11, 2006 - link

    oh, nevermind, i see it now
  • JarredWalton - Thursday, May 11, 2006 - link

    The rendering error isn't "major" -- that's why we kept the 60.72 results in the graph. You could certainly play the game with the 60.72 drivers and never realized there was a "problem". For anyone else trying to find the anomaly, look for the vertical lines right in the center of the screenshot.
  • synque - Thursday, May 11, 2006 - link

    I think the conclusion the article draws is completely meaningless. Nothing can be said about NVIDIA because they either failed to improve the drivers, or the drivers were close to "optimal" from the start.

    One could speculate that the ATI drivers weren't optimal because they could be improved. But even that'd be guessing, because the driver team most likely optimizes for popular games after they are released (which could lead to special optimizations).

    So I know exactly as much as I knew before reading the article. Weird.
  • z3R0C00L - Thursday, May 11, 2006 - link

    Umm not true,

    You forget that the x8x0 line are essentially built on an improved R300 design. They're not an all new part. This means ATi have gradually worked to improve there drivers. They were optimal to begin with since they're essentially the same driver SDK that reaches to the time of the 9700 Pro.

    I have a feeling that if a 9800 Pro were tested you'd see the same improvements in non video card bottlenecked situations.

    This proves that the Catalyst driver team is superior to the Forceware driver team. But most of us have known this since Catalyst 3.0. Heck even Microsoft has stated this as well as 3rd party driver analysts. ATi's card have less issues then nVIDIA cards and carry with them less issues not fixed from previous releases. There's a simple way of checking this.. read the Driver Release notes from both ATi and nVIDIA.. you'll notice FAR more unresolved issues with nVIDIA drivers then ATi drivers. Many of them major issues.

    One thing nVIDIA is better at, and this is a fact, is supporting older hardware (GeForce2,3,4). ATi's 8500 support is lackluster at best. I can't remember the last time a driver release focused on fixing issues that still exist since Catalyst 3.0 on the 8500 series.
  • Redofrac - Thursday, May 11, 2006 - link

    Since Catalyst 3.0? I'm going to have to disagree with that.
    From firsthand experience owning a 9700 Pro with the earlier catalysts, I have to say that they were crap. Having to try multiple releases of the driver to find one that was actually stable isn't quite what I'd use for the mark of a good driver team. Every once in a while they'd manage to get out a stable release, and I'd stick with that one until the next, lest random games start crashing on me or glitching.

    I can't speak for ATI's current drivers, which I'll assume are much better (dealing with drivers for a 9700 somewhat turned me away from ATI) but I find it a bit hard to say that 3.0 drivers were stable with a straight face.
  • LoneWolf15 - Friday, May 12, 2006 - link

    And I owned a Radeon 9700(non-pro) and had no problems at all; in fact, I'd daresay it's the best card I've ever owned in terms of performance/stability/longevity. But, one experience does not equal all.

    I'm not debating your experience, but what if you had a borderline power supply at the time, for example? That could easily cause some issues.

    Neither of us is a representative example of how the Radeon 95xx/97xx cards worked by ourselves. By and large though, the enthusiast community had very few issues with this series of cards.

Log in

Don't have an account? Sign up now