NVIDIA claims top DX9 GPU marketshare spot.

Discussion in 'Graphics and Semiconductor Industry' started by Jerky, Aug 1, 2003.

  1. RussSchultz

    RussSchultz Professional Malcontent
    Veteran

    Joined:
    Feb 6, 2002
    Messages:
    2,855
    Likes Received:
    55
    Location:
    HTTP 404
    Well, when I said "truly acceptable" I meant 60+ FPS, or 180+ if you're a quakehead, particularly for twitch type games. I doubt that even the 9800 pro could do that for every pixel being shaded with PS2.0. (Assuming the shader was something difficult enough to require PS2.0)

    But yes, I see your point about MSAA. Thank you for the clarification.
     
  2. RussSchultz

    RussSchultz Professional Malcontent
    Veteran

    Joined:
    Feb 6, 2002
    Messages:
    2,855
    Likes Received:
    55
    Location:
    HTTP 404
    Demalion: unless you can stop with the "inability to understand basic english" slights, you can count this discussion as over.

    But to continue it one more time: here you are again, insisting I refuse to believe anything else possible but my own conclusion.

    Do I need to quote to you AGAIN where I explicitely state that I don't agree with this idea that my conclusion is the only acceptable one and any other is completely wrong?

    Why should I even both to discuss my reasoning if you continue to assert that I have no reasoning; that I don't accept anybody elses conclusion; and that I assert that all other conclusions are wrong? ESPECIALLY SINCE I'M ACTIVELY AND EXPLICITELY SAYING THE EXACT OPPOSITE!

    If you could just back down and accept that my assertion of my conclusion does not preclude your assertion of your conclusion, perhaps we could address the meat of our differing conclusions.

    Or is it more fun to you to go round and round?
     
  3. Joe DeFuria

    Legend

    Joined:
    Feb 6, 2002
    Messages:
    5,994
    Likes Received:
    70
    I think Russ just likes to hear himself argue. ;)
     
  4. digitalwanderer

    digitalwanderer Dangerously Mirthful
    Legend

    Joined:
    Feb 19, 2002
    Messages:
    17,130
    Likes Received:
    1,660
    Location:
    Winfield, IN USA
    I sure hope not! I got in the Russ/demalion pool late and got one of the high post numbers before one of ya goes absolutely psycho on the other. ;)
     
  5. Myrmecophagavir

    Newcomer

    Joined:
    Dec 28, 2002
    Messages:
    136
    Likes Received:
    0
    Location:
    Oxford, UK
    I think they both do; trouble is, no-one else does.
     
  6. RussSchultz

    RussSchultz Professional Malcontent
    Veteran

    Joined:
    Feb 6, 2002
    Messages:
    2,855
    Likes Received:
    55
    Location:
    HTTP 404
    I'm just bored at work. Figure its here arguing over semantics, or fark arguing over politics.
     
  7. andypski

    Regular

    Joined:
    May 20, 2002
    Messages:
    584
    Likes Received:
    28
    Location:
    Santa Clara
    Fair enough - certainly you are right if every pixel is shaded with very long PS2.0 shaders, however a mix of short and long ones should give very acceptable performance, depending on resolution.

    Assuming an execution rate of only 1 instruction per clock per shading pipe (very much the worst possible case) then a simple calculation shows that a 9800 pro could run a 64 instruction PS2.0 shader at 1024x768 resolution at 60 fps with an overdraw of 1.

    Code:
    1024*768 = 786432 pixels
    
    * 60 fps = 47185920 pixels per second
    
    380MHz * 8 pipes = 3040 MPixels/second
    
    / 64 instructions = 47500000 Pixels per second
    
    47500000 / 47185920 P = 1.007 (overdraw for 60 fps) 
    Normally shader performance should be higher than this, and in addition early and hierarchical Z rejects can speed up the overall effective shading speed immensely, so even with the longest possible shader (under PS2.0 limitations) you could probably support an overdraw significantly greater than 1 in this case.

    Reducing the length of an average shader down to 32 instructions (still easily long enough to require PS2.0) will give more headroom. Considering that the most complex shader model available before PS2.0 had a maximum instruction count of 16 ALU operations you can see that even averaging 32 instruction shaders would be a huge leap forward in complexity.

    Admittedly with shaders this complex you won't be playing at huge frame rates and incredibly high resolutions, but I think the performance available in our current parts is well targeted to enable extensive use of complex shaders (compared to previous generations) right from the start.
     
  8. RussSchultz

    RussSchultz Professional Malcontent
    Veteran

    Joined:
    Feb 6, 2002
    Messages:
    2,855
    Likes Received:
    55
    Location:
    HTTP 404
    So it would have approximately the same performance as a VoodooI, but look much nicer? (A voodoo I had 45Mpix/s, didn't it?)
     
  9. andypski

    Regular

    Joined:
    May 20, 2002
    Messages:
    584
    Likes Received:
    28
    Location:
    Santa Clara
    Not a bad analogy to the baseline performance, although that is worst case (compared to the Voodoo1's best case). I would expect the overall performance would typically be significantly higher than that (I would have thought it would generally be more like a Voodoo 2).

    Maybe I should hack up Quake 1 to run with 64 instruction pixel shaders and see what performance I get ;)
     
  10. digitalwanderer

    digitalwanderer Dangerously Mirthful
    Legend

    Joined:
    Feb 19, 2002
    Messages:
    17,130
    Likes Received:
    1,660
    Location:
    Winfield, IN USA
    Excuse me, but are you guys seriously suggesting that the 5200 performs at near V2 levels?!?!

    I think I'm confused.... :?:
     
  11. RussSchultz

    RussSchultz Professional Malcontent
    Veteran

    Joined:
    Feb 6, 2002
    Messages:
    2,855
    Likes Received:
    55
    Location:
    HTTP 404
    Do, please! It'd be a great demonstrator and tidy little benchmark program.
     
  12. RussSchultz

    RussSchultz Professional Malcontent
    Veteran

    Joined:
    Feb 6, 2002
    Messages:
    2,855
    Likes Received:
    55
    Location:
    HTTP 404
    No, the 9800 does. When running only PS2.0 shaders of 64 instructions.

    The 5200 would likely run at about 1/8th of a V2 or worse under the same conditions.
     
  13. andypski

    Regular

    Joined:
    May 20, 2002
    Messages:
    584
    Likes Received:
    28
    Location:
    Santa Clara
    I think you are. :lol:
     
  14. demalion

    Veteran

    Joined:
    Feb 7, 2002
    Messages:
    2,024
    Likes Received:
    1
    Location:
    CT
    Well, I don't just say "you are displaying an inability to process basic English", I demonstrate where you are failing to do so.

    Showing my demonstration to be in error will work towards your requirement for me to stop saying you are wrong. So will moving on to a different topic of conversation.
    Holding a conversation with me and continuing to require me to explain the problem with your statements, and simply dictating to me that I'm not allowed to find that problem, will not.

    If you think there is something wrong with those choices outside of your personal criteria of feeling uncomfortable when not getting what you want, you are free to make a case for it.
    Until then, those are your options.

    Here is the fallacy you propose:

    You said, after our conversation progressed, that properly could include performance.
    I said we were making progress and you were admitting to the error in your prior statement.
    You are maintaining that you were not wrong in your prior statement, while circumventing any and all reasons I propose why the latter statement contradicts it, by insisting that you aren't now saying what I proposed was wrong with your prior statement. Of course you aren't, that was why I said we were making progress from your original statement.

    Again: *I'm not saying you're refusing to believe anything else about properly now, I'm saying your original statement denied that properly could include performance by agreeing with someone who proposed that it could not, and by attacking me when I said "properly can include performance" to that person.
    Your responses also propose that "You weren't disagreeing with saying properly could include performance" in your reply to me, and you ignored my question about what you thought I was saying when you...disagreed with me.

    * What I'm saying you refuse to accept is that your modified statement indicates something different than what your initial statement did. I.e., that there was a problem with your original statement.
    * I'm also saying your original statement's presented commentary doesn't make sense, and I've explained why in my first reply to it.

    You are addressing my replies on the premise that I'm telling you that what you are proposing right now is that properly cannot include performance, when this entire branch of discussion started when I was saying what you were now proposing was different than what you had proposed. The discussion of why I think you had proposed such a thing is nowhere addressed by you in your pursuit of this nonsensical address.
    This is something I explained earlier as well...take a look of some of the explanations you've been skipping. :-?

    Now. Had. :-? Why should I think you make sense when you propose that I'm arguing what you are doing now when it was drawing a distinction between what you now said and what you had that prompted this discussion?!

    "Backing down" comes before addressing the "meat of our differing conclusions"?!
    Also, get the "conclusions" straight...it might also help if you didn't just keep skipping over my explanations and restating things that ignore what I just tried to explain to you. :shock:
     
  15. Dave Baumann

    Dave Baumann Gamerscore Wh...
    Moderator Legend

    Joined:
    Jan 29, 2002
    Messages:
    14,079
    Likes Received:
    648
    Location:
    O Canada!
    Ahhhh, I lament for my database space.
     
  16. Malo

    Malo Yak Mechanicum
    Legend Veteran Subscriber

    Joined:
    Feb 9, 2002
    Messages:
    6,952
    Likes Received:
    3,032
    Location:
    Pennsylvania
    LOL Russ and demalion taking up 10% of total database eh?
     
  17. demalion

    Veteran

    Joined:
    Feb 7, 2002
    Messages:
    2,024
    Likes Received:
    1
    Location:
    CT
    I'd offer to pay for the storage I'm taking up, but I doubt I could afford it. :(
    :p
     
  18. RussSchultz

    RussSchultz Professional Malcontent
    Veteran

    Joined:
    Feb 6, 2002
    Messages:
    2,855
    Likes Received:
    55
    Location:
    HTTP 404
    If that is the bone of your contention, you have seriously wasted a lot of my time and yours.

    Again, in plain english for you:

    Me stating an opposing conclusion with my reasoning does not equate to me denying that any other conclusion is possible.

    Consider the following:
    Somebody else: "JoeSmith would never drive drunk."
    You: "I think he was drunk and drove off the side of the cliff after dosing off. Look, there's a broken beer bottle in the car"
    Me: "I never knew Joe to drink and drive. There's a dead deer here on the side of the road, I think its safe to assume he lost control after hitting the animal"

    Do you see me denying that it is possible that Joe, our hapless driver, was driving drunk? No. I'm simply asserting my view on the situation.

    If you don't understand that, I don't know what I can quote or illustrate to help you. And, we'll just call this conversation over.
     
  19. digitalwanderer

    digitalwanderer Dangerously Mirthful
    Legend

    Joined:
    Feb 19, 2002
    Messages:
    17,130
    Likes Received:
    1,660
    Location:
    Winfield, IN USA
    Is Joe ok after the accident or not? :|
     
  20. RussSchultz

    RussSchultz Professional Malcontent
    Veteran

    Joined:
    Feb 6, 2002
    Messages:
    2,855
    Likes Received:
    55
    Location:
    HTTP 404
    Immaterial. The semantics are much more important. ;)
     
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...