Fall 2003 Video Card Roundup Part I - ATI's Radeon 9800 XT
by Anand Lal Shimpi & Derek Wilson on October 1, 2003 3:02 AM EST- Posted in
- GPUs
AquaMark 3
Despite what some people would like everyone to think, Aquamark3 is really a test of how people developing software now envision DirectX 9 pixel and vertex shaders will be used in the future. The situation is very reminiscent of the first Sony PlayStation: the first games that used the technology were limited by the hardware until developers really learned to work with the hardware rather than on the hardware. As time progressed, we went from what were essentially ports of 16bit console games to amazingly complex and beautiful games like Gran Turismo 2. The same thing will happen with shader technology, and no amount of guessing and throwing functions at a gpu will tell you how its performance will really be in the future. Essentially, my advice is that any piece of software that claims it is a valid predictor of future performance should be taken lightly. We based our decision to include Aquamark3 on its popularity in the community. Aquamark3 is a cool piece of software, with some pretty neat tests, and a high score in any benchmark can still earn bragging rights in the forums. The only Aquamark3 test we ran was the publicly available 1024x768 4xAF noAA in order to maximize the usefulness of these numbers to the community. Our drivers were set to allow application control of AF and AA.
We can see almost a pairing off of the cards in direct competition with one another from each camp. ATI pulls ahead by an insignificant margin in the case of the top cards, but the 5600 Ultra falls way behind in this test. Image quality appears to have improved for NVIDIA in this benchmark over what has been reported of previous drivers, and the NV38 handled the massive overdraw portion of the test the smoothest of all the cards. We will be taking a much closer look at image quality very soon, but until then, it looks like ATI and NVIDIA have equal footing in the Aquamark3 arena and we are left to find more useful information about their differences elsewhere. We would also like to point out that the 9700 Pro held its own in this test inching out the standard 9800.
Facebook
Twitter
RSS
263 Comments
View All Comments
Anonymous User - Friday, October 3, 2003 - link
FSAA does work in Halo you need to add two lines to the config.txt file to enable it. FSAA is working fine in Halo now.Anonymous User - Friday, October 3, 2003 - link
Where are the DX9 benchmarks?What is going on at Anandtech? Why all the Dx9 titles?
Old cards can do dx8 well I want to see how dx9 titles run. Aquamark is mostly dx8.
You for some reason are using buggy Nvidia drivers for this test why?
Something is fishy here. I smell a sellout.
Anonymous User - Friday, October 3, 2003 - link
Anonymous User - Friday, October 3, 2003 - link
Good article and nice new testing sweet. But look into adding SOE's Planetside to the mix that game eats anything less then a 5600 for lunch running at no more then 20 fps. my heavily oced 5600 (350/550) never gets over 70 or so.Anonymous User - Friday, October 3, 2003 - link
#206everything that was able to run aa/af was run in aa/af ... how can you complain about that?
There is exactly one (sucky) dx9 game out that they didn't test: TRAOD ...
meh
Anonymous User - Friday, October 3, 2003 - link
I'd like to see Nascar Racing 2003 tested, rather than F1Challenge. Since F1C is CPU limited, it makes the results rather useless for GPU testing.As #159 notes, starting from the back of a full-field AI race will definitely show what your hardware is capable of doing. But the AI calculations may eat up a lot of CPU cycles. (FWIW, NR2003 is multithreaded and MP-aware, so this scenario might make for a good CPU/system test.)
However, one could create a _replay_ of a full-field race. The replay is then repeatable on any system. And, although I haven't tested this, I imagine the replay might be more GPU-intensive since there's less real-time AI and physics processing happening.
OTOH, both games have DX8.x graphics engines AFAIK.
Anonymous User - Friday, October 3, 2003 - link
Almost all the games were cpu-limited.Relatively few used AA/AF, which is even more important with a slow cpu, given that you have videocard power to burn. Another failure.
Few of the games were DX9. Is this some sort of sop for Nvidia?
All-in-all a very annoying and disappointing non-review.
rms
Anonymous User - Friday, October 3, 2003 - link
Still using Flash for benchmarks.. again? Come on, cut that out.Anonymous User - Friday, October 3, 2003 - link
i dont even play games at 1024x768 cause i have an nvidia and it does suck!Anonymous User - Friday, October 3, 2003 - link
Pete --The ATI 9600 Pro would not run Homeworld 2 at all ... Oops on leaving that out of the write up, but that's a good catch on your part.
NWN problems are known, but didn't exist until introduced by the Cats released *after* NWN was on the shelves (so says Bioware iirc).
But we will touch on this in the next article.
The 9600 Pro will be addressed when we do our budget card section of the roundup ...
J Derek Wilson
(Wading through 180 posts as I work on the next set of benchies and IQ tests)