Page 9 of 20
Posted: Tue Sep 04, 2007 8:16 pm
by SURFiNG
just a thing i don't ever understood...
<subsurface_scattering>
<scattering_coefficient_spectrum>
<rgb>
<rgb>11.873 13.293 14.589</rgb>
</rgb>
</scattering_coefficient_spectrum>
How to translate in BLENDIGO??? I mean i can't reach values like 11.8 on it!
Any suggestion?
Posted: Tue Sep 04, 2007 8:57 pm
by zuegs
set RGB color to 0.8138, 0.9112, 1.000 and the gain to 14.589
or set RGB color to 0.1187, 0.13293, 0.1458 and gain to 100
both solutions should give the same right output (except if blendigo uses gamma correction for sss....
) you will tell us
Posted: Tue Sep 04, 2007 9:00 pm
by SURFiNG
what do you mean by "scale to 100" ?
the gain value?
Posted: Tue Sep 04, 2007 9:19 pm
by zuegs
yes, sorry, scale=gain
[Edited/corrected post above]
Posted: Tue Sep 04, 2007 9:22 pm
by CoolColJ
yes
and check the scene file to make sure RGB gamma = 1 not 2.2, or the colour won't be the same
it probbaly sets it to 2.2 so, you will have to manually change it
Posted: Tue Sep 04, 2007 9:40 pm
by SURFiNG
damn i feel retard!!!!!
here is an example:
Code: Select all
<medium>
<name>milk_scattering_medium</name>
<ior>1.35</ior>
<cauchy_b_coeff>0.0</cauchy_b_coeff>
<precedence>10</precedence>
<absorption_coefficient_spectrum>
<rgb>
<rgb>0.001 0.003 0.013</rgb>
</rgb>
</absorption_coefficient_spectrum>
<subsurface_scattering>
<scattering_coefficient_spectrum>
<rgb>
<rgb>11.873 13.293 14.589</rgb>
</rgb>
</scattering_coefficient_spectrum>
<phase_function>
<uniform/>
</phase_function>
</subsurface_scattering>
</medium>
So in Blendigo will be:
Absorption: Gain: 1 values: 0.001 0.003 0.013
Scattering: Gain 100: values: 0.119 0.132 0.145
Then edit the mat file and change the RGB gamma to 1 (why is 2.2 default?)
Another thing i don't understand is:
Why the color doesn't change in blendigo visual colortab if i change the gain? Is it normal?
Bury me with spits
Posted: Tue Sep 04, 2007 11:03 pm
by CoolColJ
2.2 gamma is to match the processing of colours in Indigo with what you see in Blender and other program's colour sliders
But if your using SSS RGB values from scientfic papers gamma has to be set to 1. Some exporters allow you to change the gamma, a good idea IMO
RGB will only go from 0 to 1 within 3D programs. That's why you won't see any change, this multiplier is only for Indigo which works on a greater range with spectral stuff.
SPIT!
Posted: Tue Sep 04, 2007 11:26 pm
by Kram1032
You'd need a HDR-screen to get colours out of the 0-1 range
Posted: Wed Sep 05, 2007 12:15 pm
by SURFiNG
ok thank you mates!!!
I need that GAMMA switch on blendigo Smart otherwise i'll go postal!!!! Please
Posted: Wed Sep 05, 2007 4:20 pm
by Big Fan
ok someone help me with my confusion please
this gamma being 1.0 only applies to SSS if you want to use data from papers otherwise it continues to be 2.2, is that it?
all other gamma in indigo stays as 2.2 regardless?
and at the moment there is a bug in Indigo for SSS that renders for example milk orange, yes?
Surfing you want a button on the SSS UI to switch between the two gammas to make life easy?
I suppose I could do something about that for smartden if he is not around...
I dont think he likes my coding though
Posted: Wed Sep 05, 2007 5:16 pm
by CoolColJ
yeah pretty much just for RGB absorbtion and scattering
Posted: Wed Sep 05, 2007 9:56 pm
by zuegs
i think gamma should be 2.2 only for real colors (diffuse and albedo). Most others as absorbtion, SSS and other funny stuff should have a gamma of 1.0 as this values express some special physical values (as absorption per meter stuff - look units in indigo-manual
).
Posted: Thu Sep 06, 2007 3:06 am
by Whaat
zuegs wrote:i think gamma should be 2.2 only for real colors (diffuse and albedo). Most others as absorbtion, SSS and other funny stuff should have a gamma of 1.0 as this values express some special physical values (as absorption per meter stuff - look units in indigo-manual
).
I agree
Maybe the best compromise is to use gamma = 1.0 for all absorption and scattering values but then have a second color swatch displayed in the UI that shows the actual gamma corrected color value that will be exported to Indigo. I think Maxwell
has something like this.
This way, users can use actual real-world values from papers and it makes sharing materials much easier between the different exporters if everyone is using gamma=1.0. (Sharing IGM files starts to get pretty confusing if everyone is using different gamma values).
Posted: Thu Sep 06, 2007 4:29 am
by Kram1032
now, THAT's an interesting idea
so every value has to be corrected with gamma 0,45454545454545454545454545454545
(1/2.2)
Posted: Thu Sep 06, 2007 4:39 am
by Whaat
Kram1032 wrote:now, THAT's an interesting idea
so every value has to be corrected with gamma 0,45454545454545454545454545454545
(1/2.2)
Not really....Only the values that will be displayed on the second swatch. This only serves as visual feedback so you can see what Absorption and Scattering values are actually being exported. The gamma tag would be exported as 1.0 for absorption and scattering. Gamma tag would be 2.2 for all other situations.