LOD Bias Adjustment - Found a way to change/adjust it (NVIDIA), setting not "broken"

Discussion in 'General Discussion' started by Spinelli, Mar 27, 2014.

  1. Robert Gödicke

    Robert Gödicke Registered

    Joined:
    Oct 5, 2010
    Messages:
    409
    Likes Received:
    11
    ^What Lazza said.

    I don't know exactly how Alt+M works, if it works exactly like clamp (only overriding all negative settings with 0.0 while leaving the positive ones intact) or if it really just sets every LOD bias to 0.0 regardless of it's initial setting. I read somewhere that it's the latter one which would make Alt+M act a little bit different compared to clamp, but since I highly doubt track makers will use positive values somewhere, Alt+M should work exactly like clamp in our case, yes.

    For your second problem Spinelli, a nice solution would be if ISI would implement Alt+M in a way where we could simply set our desired value for all materials (and make it permanent, of course) in the *.PLR or something. Right now, it's impossible.



    The easiest fix to get rid of all the negative values at once (and being able to adjust them manually afterwards) would be if nVidia actually fixed or revived the clamp setting in their drivers. Since rFactor 1 doesn't have this Alt+M functionality, this would be the only way to be able to treat all textures the same, no matter their very different initial LOD settings. I am trying to get some attention in their forums, but it's like fighting windmills over there when random users simply tell me that the screenshots are not affected by different LOD settings, but by AA being set on and off. :)

    So the issue is clearly visible to them, but they just don't want to understand that clamp would actually really make such a huge difference. So far no dev has answered in that topic. If you want to join me bumping this thread until we get an official response, feel free to do so, just leaving a reply like "Make our rFactor 1 beautiful again! Bring back clamp functionality!":

    https://forums.geforce.com/default/...clamp-negative-lod-bias/post/4153213/#4153213

    The more people will post there, the better the chance we will get a reply.
     
  2. TechAde

    TechAde Registered

    Joined:
    Oct 13, 2010
    Messages:
    606
    Likes Received:
    38
    No, I don't think so, as I understand it Alt-M disables all mip map lod bias entries in the materials. This is also hinted at by the message that appears when Alt-M is pressed. Clamp does what you describe, setting all negative bias materials to zero but not affecting anything with a positive bias (not that anyone ever uses a positive bias).

    It should be easy enough to test and confirm by someone who has the time & tools to do so.
     
  3. AK1504

    AK1504 Registered

    Joined:
    Feb 28, 2014
    Messages:
    110
    Likes Received:
    44

    Thx but i found the solution in Thread Display Options. Show pictures wasnt enabled.
     
  4. Lazza

    Lazza Registered

    Joined:
    Oct 5, 2010
    Messages:
    12,393
    Likes Received:
    6,609
    Lazy post on my part... 'remembered'/imagined seeing "Negative" in the message it shows and ran with it... *sigh*

    So is there something to be gained by having different bias on different materials?
     
  5. Synthbent

    Synthbent Registered

    Joined:
    Mar 31, 2014
    Messages:
    5
    Likes Received:
    0
    YOU ARE MY HERO. I have been playing around with every type of AA I could since I bought rF2 several moths ago. I had pretty much come to a conclusion that I either could have it run smooth and look like total hell, or run poorly and have it look good with SGSSAA on. I tried your first method, with 8x multisampling in Nvidia Inspector and it looks fantastic and runs at a smooth 60 fps with no drops whatsoever. Thank you so much!
     
  6. Joe Campana

    Joe Campana Registered

    Joined:
    Jan 19, 2012
    Messages:
    129
    Likes Received:
    2
    alt-M 'mip bias disabled' = forces all MIP LOD bias to 0.0. it is not a clamp.
     

Share This Page