Unreal Tournament 2003 IQ no AA/AF

We did the same old benchmark with UT2003 we always do. Here are the screenshots we took.

NVIDIA 45.23

NVIDIA 52.14

ATI Catalyst 3.7

Tron 2.0 Performance 4xAA/8xAF Unreal Tournament 2003 Performance no AA/AF
Comments Locked

117 Comments

View All Comments

  • Anonymous User - Saturday, October 11, 2003 - link

    Great job with the review. I'm so happy to finally see benchmarks in more than just FPS's. I'm always curious to see what kind of benefits can be had by upping my video card in RTS games for example. Take a rest, your brain must be fried from all that benching.

    One question. I have an LCD monitor and I can't get Generals:Zero Hour to run at 1280x1024. How did you manage to get that resolution for your benches since it is not offered in the game menu?
  • Anonymous User - Friday, October 10, 2003 - link

    Anadtech is a total liar. The 52.14 quality of picture sucks and so does my 5900 card sucks. Its garbage like it was before and the 52.14 drivers for sure are not helping it become better. As for the pictures i dont know how he dares even to say there is no difference in quality it for sure is a big problem the quality with all the games and programs i tried out.

  • Anonymous User - Friday, October 10, 2003 - link

    The TRYTH about the det. 52.14 driver:

    A) OpenGL filtering appears to be just fine.

    B) D3D filtering suffers from the following "optimizations"

    Application Mode:
    1) True trilinear is never utilized at all...on any texture stages. It's now all "pseudotrilinear"

    Control Panel Mode
    1) Same pseudotrilear as above
    2) Proper Aniso level selection is only applied to texture stage 0. No matter what the aniso level selected (2x-8x), only 2x is applied to stages 1-7.
  • Anonymous User - Thursday, October 9, 2003 - link

    I'd like to know what happened to all the CPU scaling analysis? I hope there's a part 3 which includes these same cards.
  • Anonymous User - Thursday, October 9, 2003 - link

    Good to see you are reworking the Flight Sim 9 BM. Your achieved FR in part 1 were so much higher than what real simmers are getting. Many people would be happy to get a reliable 25 FPS out of the game. Once there we can worry about IQ. So you need to push the texture sliders etc to put a real load on the system in this game. A continuing argument with this game is the importance of CPU/memory relative to graphics card in achieving acceptable frame rates. Anything in your testing that could shed light on this would be invaluable.

    scott s.
    .
  • Pete - Thursday, October 9, 2003 - link

    As long as I'm bothering you, I'd like to request Halo numbers with AF for your next review/roundup. AF really spruces up IQ, IMO, and it's a shame (almost pointless) to buy $500 cards and not run at the highest IQ possible. I'd also appreciate comparison pics with AF, as well. Thanks!
  • Pete - Thursday, October 9, 2003 - link

    Whoops, that was the URL of the inline pic, which works. This is the (broken) link to the large pic: http://images.anandtech.com/reviews/video/roundups...
  • Pete - Thursday, October 9, 2003 - link

    Derek, the Halo Det45.23 large pic link doesn't work: http://images.anandtech.com/reviews/video/roundups...
  • Anonymous User - Thursday, October 9, 2003 - link

    Might I suggest you remove the Tech part of your
    Sitename ? there's not much tech anymore
    Anyway this is not about Nv vs Ati, it is Nv vs DX9, It still states on NVidia's site, the FX series are DX9 card which they are not!!
    Inform the people as it should & not as your Nv-Paymaster is telling you!!
  • Anonymous User - Thursday, October 9, 2003 - link

    "We still urge our readers not to buy a card until the game they want to play shows up on the street."

    What if we want to play DNF? Should I just wait till it's done?

    Sounds like you're in nVidia's pocket more like it. Most people like games, not just one... so upgrading has immediate effects for most.

Log in

Don't have an account? Sign up now