[Thomas An.]

Hi everyone,

In view from this earlier discussion, It appears that SSS was not fully functional in the previous beta 1.2.0:
http://www.maxwellrender.com/forum/view ... highlight=

In the current beta 1.2.1 the behavior of the absoption/scattering is now showing in the following chart:

• The chart bellow is a grid of wedges grouped in six
• Each wedge has a varied thickness with beams running behind it at the 1x, 2x, 4x, 6x and 8x thickness position.
• Each groupling of six wedges is backlit by an emitter B110W
• The chart also tests the effects of color lightness

[tom]

wonderful test!!!!!!!!!!!!!!

[spekoun]

That is great test...

[Axe]

Let me see if I get this right:

Lower scattering values (as long as it's above zero) result in more light coming straight through, thus a brighter object.

Absorption values set to max (1) will not absorb all light, but will allow a large percentage to transmit through.

Thickness (and object base colour) seems to be the predominate factor in controlling to what degree the object on the opposite side is visible.

Excellent test Thomas! I've saved it with my Maxwell documentation files.

[Becco_UK]

Thank you for taking the time to produce this. Very useful.

[Hervé]

this is really great Thomas... could not dream better ! Thanks a lot !

[Micha]

Wow, Thank you very much for this helpful chart.

[Aldaryn]

Thanks for this detailed chart! Im sure you've saved a lot experimenting time for all of us!

[rivoli]

this is just great thomas, thanks for the really useful chart.

[Maximus3D]

Great work Thomas that one should be a given for the pdf manual

/ Max

[x_site]

:: wow thaks so much ::

[abgrafx3d]

Very nice work Thomas and great info

Thanks alot!

[oz42]

Thanks Thomas, you finished what I started

[Thomas An.]

Thank you all !


One thing that puzzles me about these results is that the scattering parameter serves to darken the object in much the same way as the absorption parameter. Both parameters seem to have the same effect.

But... if that is the case... why have two parameters ?
(unless the SSS implementation is not 100% complete, even in this 1.2.1 release ?)

Also, if the absorption is 1.0, then I would expect an opeque object.

[Mihai]

Awesome test Thomas! Especially the fact you used varying thickness.

It's still a mystery what scattering (and even absorption) do, according to the manual with scattering at 1 you should get brighter objects..:

"Scattering coef
Defines the amount of light to be scattered back from sub-surface. When increased, object will become more translucent."

But looking at your pics, when the absorption is between 0.2-0.6, then increasing the scattering either darkens the object or doesn't change appearance.

Then if you look at the 0.8 abs one, the 0.2 sc one is darker, then gets brighter, brightest with 0.6 sc, then starts getting darker again.....

What is also strange is how the blurryness remains the same for different settings, so obviously we need also a density setting.