Someone explain this to me. I'm currently learning Substance designer and in the tutorial they say that the shader has a hardcoded 4% F0(Fresnel 0° angle) for dialectrics, since most of them basically have this or a very close value anyway, but Arnold and other offline renderers use float numbers instead of percent to calculate fresnel, is 1.52 IOR equal to 4%? And how can I see the grayscale value?
you should just use real life IOR values
http://forums.cgsociety.org/archive/index.php?t-513458.html
>>562335
How can I do that in substance? How does Rubber 1.5191 IOR translate to % in Substance?
>>562336
i don't think it has anything to do with the fresnel values. this is related to the shader and not the textures.
substance have proprietary shader which you cannot change. however you can change the IOR value in a render engine like arnold.
i mean, you should. i can do it in cycles then im sure you can do it in arnold
>>562342
It does, the 4% reflectivity on F0 is hardcoded since most dielectric materials use values from 2% to 5%, 4% is a nice middle-ground value, you can change it in case you're working with materials that are the exception, or if you really want to be as physically accurate as possible, but it needs an extra step. What I'm trying to understand is, when importing Substance materials to Arnold or other renderers, how do I convert the percentage value to IOR units? Is 4% = to 1.4? and 9% = to 1.9? What about non-metallic materials that have huge values? I mean, there are like 2 or 3 in the universe but how would you go about doing that? Would you just make them metal since at that point there's no diference?
>>562345
well i personally tweak my IOR in regular numerical values and not in like you just showed so i don't know what to tell you.
>In case you ever need it. The conversion from IOR to F0 is:
((1.0 - IOR) / (1.0 + IOR))^2
just copied it from here, im not a math guy so i don't know what the hell that means
feel free to dig in
https://blender.stackexchange.com/questions/73009/how-to-make-a-node-group-to-calculate-fresnel-via-f0-values
pic related is how my shader looks like
>>562348
and this is how my PBR shader looks like, but i didn't make it myself.
i think this is the F0 convertion
>>562349
Just use Disney shader and forget about this shit. Spend time on something more important.
>>562364
Unrelated but this movie doesn't look too good on screenshots.
>>562364
people are going crazy over PBR and rightfully so. its easy fast and looks good.
but you don't need a clusterfuck of shaders in most cases, just good art
>>562371
That's one of the advantages of PBR, you don't need a clusterfuck of shaders, just one standard/uber shader.