Whatever happened with Futuremark and nVidia with 53.03?

Discussion in 'Graphics and Semiconductor Industry' started by digitalwanderer, Jan 5, 2004.

  1. digitalwanderer

    digitalwanderer Dangerously Mirthful
    Legend

    Joined:
    Feb 19, 2002
    Messages:
    17,219
    Likes Received:
    1,738
    Location:
    Winfield, IN USA
    Did the FM time limit expire? Are they supposed to be doing SOMETHING now about it, or am I jumping the gun?
     
  2. bloodbob

    bloodbob Trollipop
    Veteran

    Joined:
    May 23, 2003
    Messages:
    1,630
    Likes Received:
    27
    Location:
    Australia
    53.03 shouldn't be approved because well the 52.16 shouldn't be improved and I believe if anything 53.03 have more cheats then the 52.16. So I don't see why you are asking why hasn't the newer drivers been approved.
     
  3. Tim Murray

    Tim Murray the Windom Earle of mobile SOCs
    Veteran

    Joined:
    May 25, 2003
    Messages:
    3,278
    Likes Received:
    66
    Location:
    Mountain View, CA
    52.16 was approved for 340 because 340 broke the cheats.
     
  4. bloodbob

    bloodbob Trollipop
    Veteran

    Joined:
    May 23, 2003
    Messages:
    1,630
    Likes Received:
    27
    Location:
    Australia
    http://www.futuremark.com/community/drivers/?approved

    You just keep believing that barron.
     
  5. Geo

    Geo Mostly Harmless
    Legend

    Joined:
    Apr 22, 2002
    Messages:
    9,116
    Likes Received:
    213
    Location:
    Uffda-land
    Doesn't the readme on the 53's specifically state that they reenabled the. . .uh. . .compiler. . .that the 340 build "broke"?
     
  6. Deathlike2

    Regular

    Joined:
    Aug 17, 2003
    Messages:
    542
    Likes Received:
    5
    Yep. :lol:
     
  7. Tim Murray

    Tim Murray the Windom Earle of mobile SOCs
    Veteran

    Joined:
    May 25, 2003
    Messages:
    3,278
    Likes Received:
    66
    Location:
    Mountain View, CA
    340 broke the cheats that impacted the actual score, didn't it? as for why anyone would cheat in the theoretical tests, that's beyond me... but it worries me that FM couldn't prevent NV from detecting it.

    and didn't Stealthy bring this up a while ago? damn him for not keeping me up to date. damn him. ;)
     
  8. Deathlike2

    Regular

    Joined:
    Aug 17, 2003
    Messages:
    542
    Likes Received:
    5
    If you read some of those 3DMark03 threads here (the big ones), I believe Patrick (the Futuremark developer) stating something along the lines of one of NVidia's cheats was not blocked (the PS 2.0 one). The patch was rushed to be posted... and here we are. Then you have Futuremark's website with its "scribbled in" notes. :wink:
     
  9. bloodbob

    bloodbob Trollipop
    Veteran

    Joined:
    May 23, 2003
    Messages:
    1,630
    Likes Received:
    27
    Location:
    Australia
    Yeah I believe the problems were discovered after they released the patch and approved the drivers instead of revoking the drivers which would lead to all sorts of problems they decided to add stuff to the notes saying its only validate to bench between nvidia cards.

    If they revoked the drivers they would have shown that there process is flawed and doesn't work as well as leaving nvidia with no drivers meaning 3dmark could not be used with nvidia cards using the 340 patch ( you can still use the 330 patch as long as you have never attempted to install 340 and seen the updated EULA ) so either futuremark would have had to release a new patch or have nvidia @$#@ off as well as OEMs.
     
  10. ByteMe

    Banned

    Joined:
    May 27, 2003
    Messages:
    136
    Likes Received:
    7
    Is it me or did futuremark take a bad situation and make it worse?
     
  11. StealthHawk

    Regular

    Joined:
    May 27, 2003
    Messages:
    459
    Likes Received:
    0
    Location:
    I exist
    Yep. The answer was that Futuremark didn't want to spent the extra time it would take to examine and defeat cheats in the Pixel Shader2.0 test since it doesn't affect the final score. The instead decided to release a patch that would fix all the tests contributing to the final score, and they caught some other things too I believe- such as Ragdoll.
     
  12. cthellis42

    cthellis42 Hoopy Frood
    Legend

    Joined:
    Jun 15, 2003
    Messages:
    5,890
    Likes Received:
    33
    Location:
    Out of my gourd
    The only answer at this point is to remind people what IS kosher while they hash out relations with nVidia, or go all out and take them on directly. At this point I'm voting they take them on directly, because I'm just sick and tired of the whole thing, with FutureMark being the only whipping boy.

    The ONLY thing I can think of otherwise is that they may work up a "final patch" which makes the program unable to be manipulated in that fashion--say by writing random, unique shaders of mathmatical equivalence but different structure before each test, so there is NO way to predict the order and prepare shortcuts. Despire the waffling, I could accept a reprogramming of 3DMark that doesn't seek to counter illicit optimizations, but makes it so that they CANNOT take place again.

    Still, the silence is rather deafening. If that option is not being pursued, I'd rather we get EVERYTHING in the light at this point--full disclosure instead of "duelling PR statements" once in a blue moon.

    The only other thing I can think is that they're holding off until it's closer to the NV40's launch, just to make them suffer a bit more. :p
     
  13. Quitch

    Veteran

    Joined:
    Jun 11, 2003
    Messages:
    1,521
    Likes Received:
    4
    Location:
    UK
    Thus they ended up in a situation where they had nVidia style guidelines. They said one thing, but the implementation of said guideline was rather... different.
     
  14. Hanners

    Regular

    Joined:
    Jul 12, 2002
    Messages:
    816
    Likes Received:
    57
    Location:
    England
    Well, they've added Catalyst 3.10 to the approved driver list, but not the ForceWare 53.03 - I think that should tell you everything you need to know. ;)
     
  15. cthellis42

    cthellis42 Hoopy Frood
    Legend

    Joined:
    Jun 15, 2003
    Messages:
    5,890
    Likes Received:
    33
    Location:
    Out of my gourd
    They're abiding by their guidelines, they're just not shouting out about the 53.03's... which wasn't a part of said guildlines. :p People can make their own stink about it if they want, since if the 53.03's haven't been accepted it's because (shock of shocks!) the illicit optimizations are back in.

    I don't believe they ever stated they would be making press releases for any IHV whose latest driver does not qualify and thumb their noses at them, did they? Problem is if they say almost anything, the PR wars we've seen shooting back and forth before will happen again, and if it happens again at this point it's probably a "for good"-type situation. They may still be trying to salvage the nVidia situation behind the scenes so it doesn't turn ugly and impact other areas.

    I'd like to see the gloves come off, but then I have nothing riding on the situation. :p
     
  16. pino

    Newcomer

    Joined:
    Dec 16, 2003
    Messages:
    109
    Likes Received:
    0
    Location:
    Naples, Italy
    what ?


    I really don't get it, Futuremark states that 52.16 are approved but then they say that 52.16 have specific optimiziation (cheats?).
    How the hell can they approve them ?

    They should change their driver guidelines or remove the 52.16 from approved drivers list and rename their benchmark 3dAti2003.
     
  17. Quitch

    Veteran

    Joined:
    Jun 11, 2003
    Messages:
    1,521
    Likes Received:
    4
    Location:
    UK
    52.16 had optimisations for the PS 2.0 test which FutureMark knew of and didn't do anything about, they then approved these drivers. Regardless of the reasons for this, it is a violation of their own guidelines. It is also irrelevant how big a violation this is, because once you've shown that you will violate the guidelines...
     
  18. andypski

    Regular

    Joined:
    May 20, 2002
    Messages:
    584
    Likes Received:
    28
    Location:
    Santa Clara
    Now why should they do that when there are drivers from SiS and Matrox that have been approved as well?
     
  19. digitalwanderer

    digitalwanderer Dangerously Mirthful
    Legend

    Joined:
    Feb 19, 2002
    Messages:
    17,219
    Likes Received:
    1,738
    Location:
    Winfield, IN USA
    No, no it doesn't. I wanna hear about how they're going to pull nVidia into a dark alley and work 'em over with a bloody f-ing rubber hose! :evil:

    You and me both, but didn't they say something about "enforcing" their guidelines? Do you mean that their idea of "enforcement" is to not put the official drivers on the officially approved list?

    Has anyone seen a review with the 53.03 drivers benched in 3dm2k3 yet? (I can't recall)

    That's the way it's looking to me too. I remember there was supposed to be some form of grace period to allow the IHV & FM to work out the differences of what is/isn't a cheat, but we haven't heard ANYTHING out of them and it's pretty apparent that something is up with the 53.03 set.

    Worm, you lurking? :|
     
  20. Hanners

    Regular

    Joined:
    Jul 12, 2002
    Messages:
    816
    Likes Received:
    57
    Location:
    England
    They probably should, but again, the fact that they haven't tells you everything you need to know about how FutureMark are going to play this.
     
Loading...

Share This Page

  • About Us

    Beyond3D has been around for over a decade and prides itself on being the best place on the web for in-depth, technically-driven discussion and analysis of 3D graphics hardware. If you love pixels and transistors, you've come to the right place!

    Beyond3D is proudly published by GPU Tools Ltd.
Loading...