Forum Groups
  All forums
    Help & Feedback
      Work in progress
      Finished Art
      Non-Max related

Featured Threads
  inspiration alert!!!
(36 replies)
  Indespensible MaxScripts, Plugins and 3rd Party Tools
(37 replies)
  The allmighty FREE Resources Thread !
(17 replies)
  spam alert!!!
(4886 replies)
  Maxforums member photo gallery index
(114 replies)
  Maxforums Member Tutorials
(89 replies)
  three cheers to maxforums...
(240 replies)
  101 Things you didnt know in Max...
(198 replies)
  A Face tutorial from MDB101 :D
(95 replies) Members Gallery
(516 replies)
(637 replies)
  Dub's Maxscript Tutorial Index
(119 replies)

Maxunderground news unavailable

Compositing, bit depth and blending modes.
show user profile  Boing
I'm trying to improve my technical grasp of rendering and am learning to render in passes and then composite together, i'm also getting my head around LWF and bit depths but i need a few answers to questions if anyone can help.

1) After reading and trying a number of combinations of passes and blending modes the most effective has been this

VraySpecular (Linear dodge add)
VrayReflection (Linear dodge add)
VrayRefration (Linear dodge add)
VrayLightiing (Linear dodge add)

All rendered with a vray gamma of 1 and saved in 32 bit tiff format, then composited.

Does this match with the norm? i've seen suggestions for multiplying Raw GI with Diffuse GI then grouping and adding Raw Lighting with Diffuse Lighting etc but it always left me with pale edge outlines on my objects, the above doesn't.

2) I've noticed sometimes a clean render in my frame buffer (vray gamma = 1.0) when saved out as a 32 bit tiff will have lots of little pink pixels scattered around which aren't present if saved as 8/16 bit or as a 32 bit exr. If i save as 32 bit tiff and change to 16 bit in photoshop the spots remain but become white. Can anyone explain this or give me a solution? I can post images/files if needed.

3) There is a render pass called VrayUnclampedColour, it seems more contrasty than the RGB pass which i presume therfore, is clamped but i'm curious, i thought the whole point of LWF was that you never clamped colours? surley that defeats the purpose of the exercise?

any help is appreciated, cheers.

read 1138 times
2/4/2011 4:11:58 PM (last edit: 2/4/2011 4:11:58 PM)
show user profile  Bolteon
1. the reason why the "mult rawGI by diffuse"... technique works is because it's done in nuke. photoshop doesn't handle things properly. that's why your getting discrepancies. you can get it close in ps but to do it right you need a proper linear compositing tool.

2. do post, that shouldn't be happening... not that you need full 32bit most of the time; 16 is more than enough in 99% of applications.

3. the unclamped pass (though i've never found a use for it personally) is just a render element of what the image would look like if you have no level of exposure on in your vray physical cam.

-Marko Mandaric

read 1122 times
2/4/2011 11:07:43 PM (last edit: 2/4/2011 11:07:43 PM)
show user profile  Boing
Cheers Bolteon, i thought the edge artifacts were a problem with my methodology.

I don't have exposure ticked on my vray cam but the unclamped pass is still more contrasty than the rgb pass. It's no big deal but i was curious about the name as it implies the rgb is clamped in some way.

I've uploaded those tiffs if you would have a minute to check them out, here's the links.

16 bit tiff (no dots)

32 bit tiff (dots)

32 bit exr (no dots)

they have all been saved from the same frame buffer, ie they aren't 3 different renders. Rendered with a vray gamma of 1.0 and saved likewise (the 16 bit requires gamma corection)

Let me know what you think.
read 1102 times
2/5/2011 3:21:52 PM (last edit: 2/5/2011 3:21:52 PM)
show user profile  Bolteon
tiffs the issue.

always save as exr's.

-Marko Mandaric

read 1095 times
2/6/2011 3:30:16 AM (last edit: 2/6/2011 3:30:16 AM)
show user profile  Boing
Well i could do that but i'd like to understand whats happening, have you seen this before? do you know specifically what the problem is? if it was in a lower bit depth i would have just put it down to an inability to display the pixel values but in a 32 bit file it's difficult to understand.
read 1072 times
2/7/2011 1:23:37 PM (last edit: 2/7/2011 1:23:37 PM)
#Maxforums IRC
Open chat window