Maxwell Render | Official public forum

I read about birefringence in my old book of physic (double refraction) ; i tried yesterday to create calcite material
"The measure of birefringence (δ) [delta] is the difference between the indices of refraction of the two rays. δ = ne − no"


left : test of calcite material , right : glass material

For calcite :
calcite CaCO3
no=1.658
ne=1.486
δ= −0.172

The calcite material created with two bsdf's
The first with nd = no
The second with nd = ne
That give birefringence effect but i am not sure if maxwell give the right effect
Because there exist two types of birefringence (double refraction) : uniaxial negative and uniaxial positive
In this case we need to apply the law : δ = ne − no to get the right effect.
The question : if maxwell can give the right effect ; how?
Or maxwell need birefringence in material editor ?
----------------
Realistic rendering of birefringency in uniaxial crystals :
http://www.cg.tuwien.ac.at/research/pub ... -paper.pdf

Birefringency-Based Realtime rendering of translucent material :
http://www.ijfcc.org/papers/150-K00012.pdf

http://physics.info/refraction/

ohhhhh, this is nice. I've been mulling over this effect in my head lately for my bottle renders but I never knew exactly how to approach it. when looking at real whiskey bottles i've noticed they all have this sort of "blurry" or "greasy" look to them, very similar to that calcite material.

just to clarify, that image isn't something you rendered in maxwell, it is something rendered in that paper?

even though it's not physically correct way to do it, could you emulate this using two bsdf in a material with one of the bdsf having a flat normal map to create the illusion of those offset refractions? I remember a thread a few years ago where somebody created a glass material that looked like there was a layer of glass inside of a layer of glass (or something like that) by using a normal map to hap one layer look offset.

many materials have the same effect if you see in the last link
with normal map i didn't tried that .
the image rendered in maxwell ; i tried to increase the effect by increasing the nd of the second bsdf ( nd=2.5) but that not real.
calcite :
bsdf : nd=1.658
bsdf : nd=1.486
δ < 0
----------------
ice :
bsdf : nd=1.309
bsdf : nd=1.313
δ > 0

nd in maxwell don't accept negative value.
if that will be possible ; the mixing of the two bsdfs give exact value of : δ
bsdf : nd= -1.658
bsdf : nd= 1.486

Seghier, I think what you render is pretty correct because, when we assume crystals as isotropoic entities, it should be as simple as mixing 2 BSDF with different Nd as you've already done. The paper is talking about view dependant anisotropic crystals which is not possible to render with current Maxwell.

thank you
δ : negative or positive
how mixing of two bsdfs solve the equation : δ = ne − no
i hope see solution in the future maxwell

Actually, you are right about it. Because, a real birefringent material is view dependant and your example calcite should keep its main refraction intact while its second refraction moving when you rotate the object on the paper. This can't be simulated with the current material system.

in the pdf attached they explain how they can render this material
i read also about metamaterials with negative nd ; it's great if maxwell can reder these materials

We can consider improvement in this are but, maybe not so soon as it's obviously not something frequently requested.

requested for realism
i hope see this feature before v4

gotcha, it sounded like you weren't able to get the effect in maxwell but you definitely were. I'll have to give this a try as soon as i have time. thanks for experimenting with this!

Tom confirm that actual version of maxwell can't render this materials ; but we can get the good result
you can try 2 or 3 bsdfs with different nd's and different colors

in the left : triple refraction / in the right : double refraction

seghier wrote:requested for realism
i hope see this feature before v4

speaking about realism i really would prefer to get working caustics first... after almost 10 years

numerobis wrote:

seghier wrote:requested for realism
i hope see this feature before v4

speaking about realism i really would prefer to get working caustics first... after almost 10 years

but caustics work !