3DMark05 scores courtesy the Inq

Discussion in '3D Hardware, Software & Output Devices' started by trinibwoy, Sep 28, 2004.

Thread Status:
Not open for further replies.
  1. Reverend

    Banned

    Joined:
    Jan 31, 2002
    Messages:
    3,266
    Likes Received:
    24
    Ich, if games use depth shadow mapping, I think DST is a good idea. This isn't about NVIDIA influencing anyone -- the fact is that DST is not a bad thing.

    The argument here is whether game developers would code their game whereby DST is an option or if it is hardcoded and therefore penalize non-NV hw. The smart thing is detection; if NV card exists, use DST; if not, default shouldn't be DST. It breaks things. Like this HardOCP review : http://www.hardocp.com/article.html?art=NjI4LDM= , which is exactly what "DST enabled by default" means. And I'm talking games here, not a benchmark.
     
  2. Bjorn

    Veteran

    Joined:
    Feb 6, 2002
    Messages:
    1,775
    Likes Received:
    1
    Location:
    Luleå, Sweden
    If you're supposed to emulate the behaviour of future games then the guidelines should be to try and use the same features as they will. Otherwise call it something other then game tests.
     
  3. trinibwoy

    trinibwoy Meh
    Legend

    Joined:
    Mar 17, 2004
    Messages:
    10,430
    Likes Received:
    433
    Location:
    New York
    I don't fully get it. Are people having a problem with

    1. DST is not part of the DX spec
    2. DST is there but 3dc isn't
    3. DST causes visual difference - is it better/worse?
    4. DST is on by default


    I see no problem with 1 and 4 if it wasn't for 3. I understand the problem with 2. But what is wrong with defaulting to DST/3dc - wouldn't non-capable cards just default to what they can do? Or does the setting break something on non-capable cards?
     
  4. ERP

    ERP Moderator
    Moderator Veteran

    Joined:
    Feb 11, 2002
    Messages:
    3,669
    Likes Received:
    49
    Location:
    Redmond, WA
    I thought PCF/DST would be a hot topic when they announced the use of Shadow maps...

    I think the problem is that they haven't really clarified what the benchmark is supposed to show.

    If it is supposed to give an indicator of upcoming game performance, then they should choose to use the features they believe will be used in future titles. Which probably does include NVidias depth textures.

    If it's supposed to be a pretty shader benchmark, then clearly they should run the same shader on all hardware.

    At the end of the day benchmarks are benchmarks they demonstrate how fast your hardware combination will run that benchmark and give you some marginal ability to project performance.

    IMO these "complex" benchmark tests are somewhat less usefull than simple ones, there are too many variables to determine why hardware X is faster than hardware Y.
     
  5. trinibwoy

    trinibwoy Meh
    Legend

    Joined:
    Mar 17, 2004
    Messages:
    10,430
    Likes Received:
    433
    Location:
    New York
    Thought this was funny. From ATI via Hexus.net :

     
  6. Ichneumon

    Regular

    Joined:
    Feb 3, 2002
    Messages:
    414
    Likes Received:
    1
    I see what you're getting at now Rev. From a games standpoint it needs to be "off by defult" and "on when card support is detected". Though no matter how you set it, somebody looses... either performance on new cards supporting the technique, or potential for errors on new cards not supporting it. Someone will bitch either way.

    I still think FM should have the default bench path not use vendor specific optimizations. There can be an optimized path which can be run to see differences. Maybe their business model makes a truly impartial graphics benchmark not feasible with all the competing interests i guess... but I just think it invalidates a large portion of the usefulness of the bench.
     
  7. Reverend

    Banned

    Joined:
    Jan 31, 2002
    Messages:
    3,266
    Likes Received:
    24
    Taking advantage of IHV/HW-specific features is a problem for a benchmark. Especially when they are enabled by default.

    3DMark05 is a benchmark, not a game. How many times have I said that?
     
  8. Fodder

    Fodder Stealth Nerd
    Veteran

    Joined:
    Jul 12, 2003
    Messages:
    1,112
    Likes Received:
    9
    Location:
    Sunny Melbourne
    Right now SM3.0 and 3Dc are both IHV-specific, but both are open standards that can be adopted by anyone. Is SM3.0 more valid for inclusion in a default run because it's part of an official DirectX standard?

    (Don't mind me, I'm just trying to pinpoint exactly where people stand on this)
     
  9. Bjorn

    Veteran

    Joined:
    Feb 6, 2002
    Messages:
    1,775
    Likes Received:
    1
    Location:
    Luleå, Sweden
    The problem is, what would the benchmark tell us about future games if it didn't use the features that these games probably will ? And we buy these cards to play games aren't we ?

    Seems like a "damned if you do, damned if you don't" thing.
     
  10. trinibwoy

    trinibwoy Meh
    Legend

    Joined:
    Mar 17, 2004
    Messages:
    10,430
    Likes Received:
    433
    Location:
    New York
    I understand this Rev but what makes a benchmark fair? If I were to bring out a benchmark that was heavily vertex limited without regard for the actual vertex load in future games wouldn't that be cow-towing to ATI's superior vertex processing abilities? It's not a perfect analogy but it has the same end result.
     
  11. trinibwoy

    trinibwoy Meh
    Legend

    Joined:
    Mar 17, 2004
    Messages:
    10,430
    Likes Received:
    433
    Location:
    New York
    This is the gist of the question in my last post :)
     
  12. jvd

    jvd
    Banned

    Joined:
    Feb 13, 2002
    Messages:
    12,724
    Likes Received:
    8
    Location:
    new jersey
    There are a ton of upcoming games including half life 2 that will use 3Dc . So why isn't it in ? Last time i saw the list there were about 8 or 9 games coming out with 3dc support in the future and this list was out before the x800s were announced.

    So why does one feature get in and not the other ?

    I can under stand sm3.0 since its part of dx 9 . But 3dc and this other feature are not part of dx 9. At the very least 3Dc is an open standard (dunno about nvidia's shadow thing) so it has the option of being supported by everyone.


    But once again , both will be in upcoming games , why does one get in and the other doesn't ?
     
  13. digitalwanderer

    digitalwanderer Dangerously Mirthful
    Legend

    Joined:
    Feb 19, 2002
    Messages:
    17,261
    Likes Received:
    1,778
    Location:
    Winfield, IN USA
    It i$ a de$ign deci$ion that i$ ba$ed on a number of influencing factor$.... ;)
     
  14. trinibwoy

    trinibwoy Meh
    Legend

    Joined:
    Mar 17, 2004
    Messages:
    10,430
    Likes Received:
    433
    Location:
    New York
    That is the true point of contention - not simply that DST is included but that 3dc was not. And like digi said it's probably all about the benjamins.
     
  15. jvd

    jvd
    Banned

    Joined:
    Feb 13, 2002
    Messages:
    12,724
    Likes Received:
    8
    Location:
    new jersey
    right , i was just answering your question of why its such a big deal.

    dst gives 10=15% increase if u look at reverands numbers on a 6800gt.

    if that was not included it would be loosing pretty badly to a x800pro.

    But what if 3Dc was added that say gave ati cards a 10-15% improvement .

    Would that not just be a kick in the face to nvidia ?
     
  16. trinibwoy

    trinibwoy Meh
    Legend

    Joined:
    Mar 17, 2004
    Messages:
    10,430
    Likes Received:
    433
    Location:
    New York
    No it would be a boost in 3dmark05 for x800 owners. BTW which upcoming games are going to be using 3dc?
     
  17. digitalwanderer

    digitalwanderer Dangerously Mirthful
    Legend

    Joined:
    Feb 19, 2002
    Messages:
    17,261
    Likes Received:
    1,778
    Location:
    Winfield, IN USA
    Mebbe FM fudges the games weighting on the scores to keep parity 'tween the two companies? :|
     
  18. DSN2K

    Newcomer

    Joined:
    Oct 4, 2003
    Messages:
    146
    Likes Received:
    3
  19. StarGazer

    Newcomer

    Joined:
    Apr 13, 2004
    Messages:
    48
    Likes Received:
    0
    Serious Sam 2 and HL2 for sure.
     
  20. jvd

    jvd
    Banned

    Joined:
    Feb 13, 2002
    Messages:
    12,724
    Likes Received:
    8
    Location:
    new jersey
    I'm talking about if dst wasn't in there but 3Dc was . It would be a kick in the face to nvidia users and it would be bitched about till 3dmark 2007 .

    i will see if i can find the list and i will post it if i can
     
Loading...
Thread Status:
Not open for further replies.

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...