Please report all spam threads, posts and suspicious members. We receive spam notifications and will take immediate action!
Results 1 to 6 of 6

Thread: Further NVIDIA optimizations for 3DMark03?




  1. #1
    Join Date
    Nov 2001
    Location
    Taipei, Taiwan
    Posts
    4,308

    Default

    Once upon a time, there was Quake/Quack—a benchmark "optimization" exposed by the simple renaming of a file. Surely such a simplistic method of benchmark detection wouldn't be used in graphics drivers again. And surely such a simple method wouldn't be laying under all the other layers of "optimizations" for 3DMark03 in NVIDIA's Detonator FX drivers.

    Or would it?

    Quake/Quack, meet 3DMark/3DMurk. Read on to find out what happened when we renamed "3DMark03.exe" and fired off some tests.


    http://www.tech-report.com/etc/2003q...3/index.x?pg=1
    Cameron "Mr.Tweak" Wilmot
    Managing Director
    Tweak Town Pty Ltd

  2. #2
    Join Date
    Dec 2002
    Location
    Isle of man
    Posts
    1,332

    Default

    So is this it?? can we finally say that nvidia are "cheating in 3dmark03??:confused:
    What the fcuk happened to my sig?

  3. #3
    Join Date
    Nov 2001
    Location
    New England Highlands, Australia
    Posts
    21,907

    Default

    Or is it as some ppl have been sayin' for a while and 3DMark 03 just isn't a proper benchtest? :?:

  4. #4
    Join Date
    Apr 2002
    Posts
    3,141

    Default

    "Did ATI "cheat" on 3DMark2003?

    Version 3.30 of 3DMark03 not only defeated the DetonatorFX driver's app detection mechanisms, but it also managed to uncover an ATI optimization that landed the company in a bit of hot water.

    ATI's optimization involves using a more optimized shader program in Game Test 4 for the water shader program, but according to ATI, this water shader renders "…the scene exactly as intended by Futuremark, in full-precision floating point. Our shaders are mathematically and functionally identical to Futuremark's, and there are no visual artifacts; we simply shuffle instructions to take advantage of our architecture….However, we recognize that these can be used by some people to call into question the legitimacy of benchmark results, and so we are removing them from our driver as soon as is physically possible. We expect them to be gone by the next release of CATALYST."

    The line between a bona fide optimization and a cheat is sometimes not well-defined, but in our estimation, ATI's optimization is suspect, and at the end of the day, it bought them all of a 9% increase in performance, or a barely mention-worthy 1.7fps.

    This small gain leads us to question why ATI would implement this optimization in the first place, since the risk of being perceived as cheating would far outweigh a less than 2fps performance increase. However, ATI has been forthcoming in its public statements about the nature of this optimization, and has stated that because of the potential perception of being considered a cheat (and probably because it didn't buy them much performance anyway), that this optimization will be pulled from the next Catalyst driver release." :)

    :cheers:

  5. #5
    Join Date
    Nov 2001
    Location
    Texas, USA
    Posts
    4,825

    Default

    Quote Originally Posted by The__tweaker
    "Did ATI "cheat" on 3DMark2003?

    Version 3.30 of 3DMark03 not only defeated the DetonatorFX driver's app detection mechanisms, but it also managed to uncover an ATI optimization that landed the company in a bit of hot water.

    ATI's optimization involves using a more optimized shader program in Game Test 4 for the water shader program, but according to ATI, this water shader renders "…the scene exactly as intended by Futuremark, in full-precision floating point. Our shaders are mathematically and functionally identical to Futuremark's, and there are no visual artifacts; we simply shuffle instructions to take advantage of our architecture….However, we recognize that these can be used by some people to call into question the legitimacy of benchmark results, and so we are removing them from our driver as soon as is physically possible. We expect them to be gone by the next release of CATALYST."

    The line between a bona fide optimization and a cheat is sometimes not well-defined, but in our estimation, ATI's optimization is suspect, and at the end of the day, it bought them all of a 9% increase in performance, or a barely mention-worthy 1.7fps.

    This small gain leads us to question why ATI would implement this optimization in the first place, since the risk of being perceived as cheating would far outweigh a less than 2fps performance increase. However, ATI has been forthcoming in its public statements about the nature of this optimization, and has stated that because of the potential perception of being considered a cheat (and probably because it didn't buy them much performance anyway), that this optimization will be pulled from the next Catalyst driver release." :)

    :cheers:
    The__tweaker,
    Anytime you use quotes from other sources, please include a reference to the source. Folks work hard on their content and they deserve the recognition if it is going to be used somewhere else.
    Old age and treachery will overcome youth and skill
    My Toys

  6. #6
    Join Date
    Apr 2002
    Posts
    3,141

    Default

    Sorry, it was found on Futuremark.

Thread Information

Users Browsing this Thread

There are currently 1 users browsing this thread. (0 members and 1 guests)

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •