iXBT Labs - Computer Hardware in Detail






Brief Account of Testing
Modern 3D Accelerators in DOOM III


  1. Introduction
  2. Video cards' features
  3. Testbed configurations, benchmarks
  4. Test results: Performance
  5. Test results: Quality
  6. Conclusions

I will not blow the trumpet and shout that the long-expected event has come true. We shall publish a separate article about DOOM III later, where the game and its features will be reviewed and analyzed. So those of you who have opened this article to learn something about the game may relax and have some coffee before plunging into the Martian battle or before uninstalling the game and complaining in forums that the game did not come up to your expectations of a miracle...

I'm a tester above all, and for me this game is a new benchmark to analyze the performance of video cards. Even if in the long run the majority of gamers declare it the worst sequel from id Software, the performance of video cards in this game will still remain very important because the DOOM III engine will be used in other games to be released in future.

It goes without saying that first of all I tested the latest generation accelerators, many people want to know their disposition in this game. These are ATI RADEON X800 Series and NVIDIA GeForce 6800 Series. We have published a lot of articles about these cards, so I will not dwell on the features of these GPU. If you wish, you can look through the list of the past articles on the subject.

Theoretical materials and reviews of video cards, which concern functional properties of the GPU ATI RADEON X800 (R420) and NVIDIA GeForce 6800 (NV40)

I have added two more video cards to the existing five video cards of the latest generation: to estimate the advantage over the past colossi from the previous generation. And as GeForce 6800 still has no match in the ATI camp, it will be interesting to compare this low-end product from the new series with the past idols of hardcore gamers.

By the way, I want to note the buzz circulating in Internet about the forthcoming release of ATI RADEON X800 XT (not PE), which will become the competitor for GeForce 6800 GT, and X800 PRO will take one step down and will compete with GeForce 6800. That's why I compared 6800 not only with the past kings of 3D graphics but with X800 PRO as well.

So, it's high time to introduce our contenders.

Video Cards

NVIDIA GeForce 6800 Ultra
Leadtek WinFast A400 Ultra TDH, 256MB, 400/1100 MHz, 16x1

NVIDIA GeForce 6800 GT
Leadtek WinFast A400 GT TDH, 256MB, 350/1000 MHz, 16x1

NVIDIA GeForce 6800
MSI NX6800-TD128, 128MB, 325/700 MHz, 12x1

NVIDIA GeForce FX5950 Ultra
MSI FX5950 Ultra-VTD256, 256MB, 475/950 MHz, 4x2

ATI RADEON X800 XT Platinum Edition
Gigabyte GV-RX80X256V RADEON X800 XT Platinum Edition, 256MB, 520/1120 MHz, 16x1

HIS Excalibur X800 Pro IceQ II Limited Edition 256MB, 475/900 MHz, 12x1

ASUS A9800XT RADEON 9800 XT Limited Edition, 256MB, 412/730 MHz, 8x1

I suppose there is no point in commenting on these cards, all of them were tested in our lab.

Installation and Drivers

Testbed configurations:

  • Athlon 64 3200+ based computer
    • AMD Athlon 64 3200+ (L2=1024K) CPU
    • ASUS K8V SE Deluxe mainboard based on VIA K8T800
    • 1 GB DDR SDRAM PC3200
    • Seagate Barracuda 7200.7 80GB SATA HDD

  • Operating system – Windows XP SP1; DirectX 9.0c
  • Monitors: ViewSonic P810 (21") and Mitsubishi Diamond Pro 2070sb (21").
  • ATI drivers v6.476 (CATALYST 4.9beta); NVIDIA drivers v61.77.

VSync is disabled.

Note that we used a beta version of CATALYST 4.9, which contains ICD OpenGL (alpha version) rewritten from scratch. The driver is still in the development stage, so the current ATI test results can be considered preliminary.

Without this driver the ATI test results would have been worse by 20-40%, but let's not put the cart before the horse and see the performance.

Test results: performance comparison

We used the following test applications:

  • DOOM III (id Software/Activision) – OpenGL, test settings are specified in the tests, demo SAM3

If you want to get the demo-benchmarks, which we use, contact me at my e-mail.

Concerning the benchmark, we did not use Demo1 installed with the game but recorded our own demo instead - SAM3 (timedemo sam3 console command is used for the test) This is done to avoid possible optimizations for the existing game benchmark.

Important information: The maximum FPS in the game is limited to 60 frames per second, but when we run timedemo this limitation is lifted. There is no point in measuring the minimum FPS, because even the most powerful 3D accelerators may often slump to 15-20 FPS, so it's not the issue.

AF8x modes on the diagrams mean Anisotropic 8x Quality; AA+AF - Anisotropic 16x Quality (FX5950 - 8x Quality) + Anti-Aliasing 4x Quality

Ultra High Quality Mode

In this mode texture precompression is disabled and anisotropy 8x is activated by default.

High Quality Mode

In this mode texture precompression is enabled and anisotropy 8x is activated by default.

Summary table of performance differences

Test results: quality comparison

One could say right away, that in 90% cases you will not see any difference at all, because DOOM III is mostly roaming in dark scary nooks, which are sometimes lighted by lamps and hell fire, which is quite enough to get a complete idea of the devil's lair in our understanding.

That's why I will provide examples, where you can at least make out some details. Note the lack of obvious glitches to be seen, there are just some rendering differences between ATI and NVIDIA.

GeForce 6800
Example 1, Ultra High Quality

Example 1, High Quality

Example 2, Ultra High Quality

Example 2, High Quality

Example 3, Ultra High Quality

Example 3, High Quality

Example 4, Ultra High Quality

Example 4, High Quality

Example 5, Ultra High Quality

Example 5, High Quality

Example 6, Ultra High Quality

Example 6, High Quality

As we can see, there is almost no difference, except for the grating in the first example. This is probably due to different implementations of filtering (optimization), or for some other reason.


Thus, about the speed on the whole:

  • NVIDIA GeForce 6800 versus ATI RADEON 9800 XT – in some modes the gain is doubled, only in AA+AF the advantage in high resolutions decreases due to the CPU bus bandwidth bottleneck in 6800.
  • NVIDIA GeForce 6800 versus NVIDIA GeForce FX 5950 Ultra – the same picture, but the gain is a little smaller. But on the whole it changes nothing.
  • NVIDIA GeForce 6800 versus ATI RADEON X800 PRO – the CPU bus bandwidth bottleneck is obviously the only reason of defeat in AA+AF at 1600x1200, in all other aspects the first card is at advantage. It's interesting to note that X800 PRO features an inexplicable breakaway from X800 XT (up to 1.5 times!), which obviously cannot be explained by the difference in architecture. Perhaps it's the problem of driver v4.9, which will be improved before its release.
  • NVIDIA GeForce 6800 GT versus ATI RADEON X800 PRO – no need to comment, if 6800 outscores its opponents then GT will do it a fortiori. The advantage reaches 1.5 times!
  • NVIDIA GeForce 6800 Ultra versus ATI RADEON X800 XT – the victory is NVIDIA's, the advantage is up to 27%.
  • NVIDIA GeForce FX 5950 Ultra versus ATI RADEON 9800 XT – a little advantage, but still it's an advantage. The parity can be seen only in a couple of modes.

We have already mentioned the quality, on the whole there is no difference. So we can certify that at this stage NVIDIA GeForce 6800 (the entire series!) is much more attractive for DOOM III than the ATI X800 series. Of course, the driver will improve, but even if we assume that ATI would manage to level the performance of its top X800 XT with the rival 6800 Ultra, in all other cases it's difficult even to count on that, the advantage of NVIDIA is too great.

Such poor results of ATI and brilliant ones of NVIDIA are easy to explain. We all remember from the release of DOOM III Alpha (its leakage, actually) that the salt of the game is in active shadow usage implemented via the stencil buffer. NVIDIA has been assuring us since the times of GeForce FX that Z operations and stencil operations are accelerated twofold in comparison to pixel operations. So with its 16 pipelines 6800 Ultra processes 32 Z and stencil results at a cycle. This is not the case in ATI cards.

Then, you should not forget about UltraShadow, which optimizes the shadow processing operations of video cards based on NVIDIA chips (truncating unnecessary work in scene parts). Though it's difficult to say how effectively it all works in this game. And the last stronghold of NVIDIA success (but not the least) is shader optimization (right up to replacing them with more effective and faster ones). Of course we mean the useful optimization without the quality loss.

What concerns weaker video cards, we shall test their performance in different modes later. It goes without saying that the High Quality and Ultra High modes are out of the question. So you know, you cannot expect the high quality graphics here. DOOM III in all its glory is still for the most powerful and expensive accelerators. And don't forget about the demands to the CPU, system bus, and memory. Yes, DOOM III can almost completely drain the upgrade purse of a devoted gamer.

More games based on the DOOM III engine are coming, but the nearest release is of the long-awaited Half-Life2, so we are looking forward to new discoveries.

Andrey Vorobiev (anvakams@ixbt.com)

August 19, 2004

Write a comment below. No registration needed!

Article navigation:

blog comments powered by Disqus

  Most Popular Reviews More    RSS  

AMD Phenom II X4 955, Phenom II X4 960T, Phenom II X6 1075T, and Intel Pentium G2120, Core i3-3220, Core i5-3330 Processors

Comparing old, cheap solutions from AMD with new, budget offerings from Intel.
February 1, 2013 · Processor Roundups

Inno3D GeForce GTX 670 iChill, Inno3D GeForce GTX 660 Ti Graphics Cards

A couple of mid-range adapters with original cooling systems.
January 30, 2013 · Video cards: NVIDIA GPUs

Creative Sound Blaster X-Fi Surround 5.1

An external X-Fi solution in tests.
September 9, 2008 · Sound Cards

AMD FX-8350 Processor

The first worthwhile Piledriver CPU.
September 11, 2012 · Processors: AMD

Consumed Power, Energy Consumption: Ivy Bridge vs. Sandy Bridge

Trying out the new method.
September 18, 2012 · Processors: Intel
  Latest Reviews More    RSS  

i3DSpeed, September 2013

Retested all graphics cards with the new drivers.
Oct 18, 2013 · 3Digests

i3DSpeed, August 2013

Added new benchmarks: BioShock Infinite and Metro: Last Light.
Sep 06, 2013 · 3Digests

i3DSpeed, July 2013

Added the test results of NVIDIA GeForce GTX 760 and AMD Radeon HD 7730.
Aug 05, 2013 · 3Digests

Gainward GeForce GTX 650 Ti BOOST 2GB Golden Sample Graphics Card

An excellent hybrid of GeForce GTX 650 Ti and GeForce GTX 660.
Jun 24, 2013 · Video cards: NVIDIA GPUs

i3DSpeed, May 2013

Added the test results of NVIDIA GeForce GTX 770/780.
Jun 03, 2013 · 3Digests
  Latest News More    RSS  

Platform  ·  Video  ·  Multimedia  ·  Mobile  ·  Other  ||  About us & Privacy policy  ·  Twitter  ·  Facebook

Copyright © Byrds Research & Publishing, Ltd., 1997–2011. All rights reserved.