programmable clipping

Discussion in 'Rendering Technology and APIs' started by 2008 IQ is unacceptable, Sep 16, 2008.

  1. Davros

    Legend

    Joined:
    Jun 7, 2004
    Messages:
    17,900
    Likes Received:
    5,342
    please explain how it looks like it uses a w buffer
     
  2. Tokelil

    Regular

    Joined:
    Mar 27, 2002
    Messages:
    329
    Likes Received:
    0
    Location:
    Denmark
    What makes you think it looks like w-buffer is used? The "only" difference between the two shots seems to be texture filtering to me, which has little to do with the depth buffer.
     
    #42 Tokelil, Sep 23, 2008
    Last edited by a moderator: Sep 23, 2008
  3. Johnny_Physics

    Newcomer

    Joined:
    Sep 12, 2003
    Messages:
    205
    Likes Received:
    3
    Location:
    Norway
    Tokelil: Exactly! Now look at the textures between Shadowman's puny draw distance, declared as great, compared to the Giants-shot enormous draw distance, declared as bad and containing z-artifacts or whatever.

    I'm starting to think that these "z-buffer artifacts" is nothing more than normal use of mip-maps and texture filtering, and iq's ultimate w-buffered dreamboat would involve no fog, point filtering and no mip-maps.
    Like a high resolution Quake 1 in software with mip-level 0 or whatever that command was. Using an extreme negative lod bias(like some do) will achieve a similar effect. It's like rubbing sand in your eyes, but some people see it as "high quality" and there is(or used to be) specially "tweaked" drivers to enforce that on a driver level.
     
  4. 2008 IQ is unacceptable

    Banned

    Joined:
    Sep 13, 2008
    Messages:
    262
    Likes Received:
    0
    The whole game does because of the huge draw distances and openness of the environments.

    @tokelil:

    I never said that the depth buffer had anything to do with the filtering. Try to refrain from putting words in my mouth, but I'm not mad at you. lol.
     
  5. 2008 IQ is unacceptable

    Banned

    Joined:
    Sep 13, 2008
    Messages:
    262
    Likes Received:
    0
    I've always preferred positive lod bias if I couldn't have neutral.

    I can't stand the shimmering and over-sharpness of LOD's below zero.
     
  6. Davros

    Legend

    Joined:
    Jun 7, 2004
    Messages:
    17,900
    Likes Received:
    5,342
    im talking about the picture in question IQ
     
  7. Andrew Lauritzen

    Andrew Lauritzen Moderator
    Moderator Veteran

    Joined:
    May 21, 2004
    Messages:
    2,633
    Likes Received:
    1,258
    Location:
    British Columbia, Canada
    I think you're really mistaking bad LOD for depth buffer precision problems. The latter is a total non-issue nowadays, while doing good LOD is still hard. So while I can accept whining about LOD (although doing some research and coming up with a better scheme would be more productive!), absolutely none of the modern IQ problems with games have anything to do with depth buffer precision...
     
  8. 2008 IQ is unacceptable

    Banned

    Joined:
    Sep 13, 2008
    Messages:
    262
    Likes Received:
    0
    I never said that.

    The 2 are totally unrelated; the depth buffer has to with depth precision, LOD has to do with sharpening or blurring textures.
     
  9. Johnny_Physics

    Newcomer

    Joined:
    Sep 12, 2003
    Messages:
    205
    Likes Received:
    3
    Location:
    Norway
    No, he's talking about the kind of LOD that deals with changing the complexity of the geometry depending on their distance from the virtual camera/viewport, not textures.
     
  10. Davros

    Legend

    Joined:
    Jun 7, 2004
    Messages:
    17,900
    Likes Received:
    5,342
    lod also has to do with polygons
    an object far away will be drawn will fewer polys than one close up
    it would be a huge waste to draw an object with 10,000 polys if it only occupies a dozen pixels on screen

    ps: i finally got around the timelimit in dxtweaker and selected NONPOW2CONDITIONAL but the problem still remains with crimson skies
    (ive only tried the demo cos it was quicker to download than find my cd and had to run in win98 compatability mode could that be why dxtweaker doesnt work ?)
     
  11. kyetech

    Regular

    Joined:
    Sep 10, 2004
    Messages:
    532
    Likes Received:
    0
    I cringed all the way through this thread.
     
  12. MDolenc

    Regular

    Joined:
    May 26, 2002
    Messages:
    696
    Likes Received:
    446
    Location:
    Slovenia
    I don't think you have provided a single screenshot where there would be any artefacts caused by depth buffer precision.
     
  13. Andrew Lauritzen

    Andrew Lauritzen Moderator
    Moderator Veteran

    Joined:
    May 21, 2004
    Messages:
    2,633
    Likes Received:
    1,258
    Location:
    British Columbia, Canada
    This tells me that you are indeed muddling the concepts in your mind. I suggest you go do some reading on geometry LOD, progressive meshes, etc. If you have time, even go look at some micropolygon renderers like REYES.
     
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...