Page 1 of 2
Multiple BSDF roughness
Posted: Tue Aug 03, 2010 10:32 pm
by David Solito
Can someone explain me what is the interest to have for example in the same layer :
a bsdf with a roughness of 30 with a weight of 100
a other bsdf with a roughness of 10 with a weight of 100
rather than only one bsdf with a roughness of 20 (30+10/2) with a weight of 100.
I can understand that a material can have multiple roughness controlled with a weightmap, but with no map how can one roughness value or the other influence the material?
Re: Multiple BSDF roughness
Posted: Tue Aug 03, 2010 11:30 pm
by zdeno
non material i homogenus
because on deep nucleus level there are some hills and vallyes or crystalic structure
and one photon of light bounce from polished hill and other bounce in deep hole (losts some energy and change some colors.
mapping two BSDFs with different roughness will give you the same results If you tile maps to eg. 9999 ( as with just weight 100)
just run few tests its the best way to get it. sadly manual is not even 2% detailed as it should be.
Re: Multiple BSDF roughness
Posted: Tue Aug 03, 2010 11:33 pm
by Half Life
The roughness of each BSDF is represented in the final material (if somewhat less intense) -- one BSDF would only have that specific roughness.
Interestingly you would likely not have much roughness 20 in your first scenario so the two materials would not match at all.
Best,
Jason.
Re: Multiple BSDF roughness
Posted: Tue Aug 03, 2010 11:39 pm
by David Solito
Half Life wrote:The roughness of each BSDF is represented in the final material (if somewhat less intense) -- one BSDF would only have that specific roughness.
Interestingly you would likely not have much roughness 20 in your first scenario so the two materials would not match at all.
Best,
Jason.
Can your explanation be more detailled?
Re: Multiple BSDF roughness
Posted: Tue Aug 03, 2010 11:49 pm
by Half Life
30% mixed at 1/2
10% mixed at 1/2
+_______________
= half at 30% and half at 10% -- not 20%.
Try it and you'll see -- the real math is beyond me, I wouldn't understand if it was explained to me... but the practical application is obvious with a few tests.
Best,
Jason.
Re: Multiple BSDF roughness
Posted: Wed Aug 04, 2010 12:08 am
by David Solito
ok I tested. the 20% roughness material is less reflective than 30% @ 1/2 + 10% @ 1/2.
But why?

Re: Multiple BSDF roughness
Posted: Wed Aug 04, 2010 12:11 am
by Half Life
I've heard soup and cake batter analogies -- but I think you'll need Tom or one of the other masters here for the indepth scoop on why.
Best,
Jason.
Re: Multiple BSDF roughness
Posted: Wed Aug 04, 2010 12:31 am
by David Solito
Thanks 1/2 Life.

So I wait that a maxwell guru explain it...

Re: Multiple BSDF roughness
Posted: Wed Aug 04, 2010 1:26 am
by zdeno
this thread could be funny I subscribed already
don't get me wrong but if You don't catch this it could be though nut to crack.
Re: Multiple BSDF roughness
Posted: Fri Aug 06, 2010 12:07 pm
by Tok_Tok
It's pretty simple acutally. It is more clear when you stretch the settings a lot more.
So let's take the first BSDF and give it a roughness of 60 and another that will have a roughness of 20. Now you will have a material that has a very rough
overall look but the second BSDF will still give it a shiny specular spot. If you would take one BSDF and give that a roughness of 40 (inbetween 60 and 20) you would not
get the same effect. Essentially when you use more BSDF's you can control the specularity much more of your material, giving it a very hard highlight but still have that rough look
over the rest of the material.
Now, the thing you are probebly crack your mind on is that the weight on both layers are 100, so couldn't be that you see both the layers.. Maxwell automatically devides the layers
into 2. So 2 layers of 100 will be 50/50. But it also works this way if you use 80/25. (Maxwell does the math for this one

)
Hopefully it's more clear now

Re: Multiple BSDF roughness
Posted: Fri Aug 06, 2010 7:36 pm
by David Solito
Hello Tok_Tok,
Thanks for your comments. Concerning the layer weight division, I knew that but the question stay the same. If you take the result maxwell does for the addition of the 2 weight, the roughness interpretation stay different if you put that result in only one BSDF. Why is it so? Aditive mode is IMO just what it says, an addition and in this case with a maximum value of 100. What i would to know is how does it work ? When a material need more BSDF for roughness control? How many layers...
Re: Multiple BSDF roughness
Posted: Fri Aug 06, 2010 9:01 pm
by zdeno
studdy any carpaint ... there is sharp 0 roughness hard specular reflection and "metalic" 40-50 roughness glossy. It is designed for purposes like this.
when You set just 25 roughness there will be dramatic difference .
Re: Multiple BSDF roughness
Posted: Fri Aug 06, 2010 9:06 pm
by David Solito
Ooops! sorry for the confusion. It's about normal blending mode and not aditive. Aditive option is really ok for me to use (for glossy plastics, car paints...)
Re: Multiple BSDF roughness
Posted: Fri Aug 06, 2010 10:36 pm
by tom
David, additive or normal doesn't matter. Make 2 BSDFs weighted at 50/50; one of them roughness 90 and the other roughness 0. You suggest it would look like 1 BSDF @ roughness 45, right? But, it doesn't... So, tell us why.

Re: Multiple BSDF roughness
Posted: Sat Aug 07, 2010 12:05 am
by David Solito
Fact1: 100/100 or 50/50 is the same
Fact2: Playing with the extreme value (0-90) is a good thing to see the difference.
Fact3: Additive / Normal blending is just for layers (a little mistake in my question

)
Fact4: Effectively 90 and 0 BSDF roughness values give not the same aspect as 45
Fact5: The aspect is impossible to achieve with one BSDF (it can be rough and in the same time glossy)
Fact6: With 2 BSDF, You keep the 0 reflectance color of the 90 roughness BSDF but you can see the 0 and 90 reflectance of the 0 roughness BSDF.
Fact6: I begin to understand the utility of this option, but I can't explain to myself or imagine mentaly how the 2 BSDF are merged
