[michaelplogue]

Could it be due to I was talking about fixing Nd's of provided material presets instead of Nd core which has been perceived faulty and working precisely perfect? I think, yes I meant this.

Ehh? Translation please ..... anyone?

[tom]

OK, Exit Tomish, Enter English.

Nd is working properly.
The thing needs to be fixed is not Nd of Maxwell.
The thing needs to be fixed is Nd value of some obsolete material presets.
Clear now?

[mverta]

do you have experience how the different ND´s affect the materials appearence ?

perhaps i´ll do a couple of test to investigate this...

I have a lot of experience, actually

Basically, for low-roughness objects, it sort-of "pushes" the reflection strength at the edges... great for auto paint and "wet" looks. Just do a test with Nd 3 and Nd 20 and you'll immediately see it; again, for low-roughness objects. As roughness increases, the effect becomes less evident, and by the time you're up to like roughness 70 it barely does anything. You can also try between 3 and 1 to see how rapidly the surface darkens into.. well, into the first post in this thread. Still, there may be a use for it in some circumstances between 2 - 3. And again, this is for SOLIDS, not for transmissives. With transparent objects, you're cool at 1.33 or 1.5 or whatever.

_Mike

[tom]

...about fresnel and Nd, you may want to peek:
http://www.maxwellrender.com/forum/view ... 38&start=0

[SJ]

For Maxwell being a renderer based on real-world physical principles I'd like to know how the Nd-thing relates to reality. Have solid objects really such differences in refractivity and does it affect the appearance similarly?
I prefer to really understand how a parameter behaves and why it works the way it does. Try and error is not really effective.

[tom]

I prefer to really understand how a parameter behaves and why it works the way it does. Try and error is not really effective.

Naturally, Nd is working perfectly same as Index of Refraction in other applications. Only difference is Nd=1.0 which is simulated correctly by Maxwell, we say "ghost".

[michaelplogue]

The thing needs to be fixed is Nd value of some obsolete material presets.
Clear now?

So, Just to be sure: The 'fix' will be that when we create a new material, the Nd default/preset will be a higher value - as opposed to a value one would expect to see in a dielectric material. Is this correct?

[SJ]

Naturally, Nd is working perfectly same as Index of Refraction in other applications. Only difference is Nd=1.0 which is simulated correctly by Maxwell, we say "ghost".

Ghost To me, the principle of refraction is very logical in context with transparent objects. But I don't get how "solid" objects can have refraction and why it affects their appearance so much...
Edit: I just read one of the threads linked above. It seems to explain the thing.

[tom]

But I don't get how "solid" objects can have refraction and why it affects their appearance so much...

No, in short, Nd controls both refraction and fresnel for transmissives but only fresnel for opaques, that's why.

[SJ]

But I don't get how "solid" objects can have refraction and why it affects their appearance so much...

No, in short, Nd controls both refraction and fresnel for transmissives but only fresnel for opaques, that's why.

aaah, that's a thing one can contemplate about. Thanks!

[michaelplogue]

But I don't get how "solid" objects can have refraction and why it affects their appearance so much...

This is what I'm trying to figure out as well.......

I'm tryng to come to grips as to why a solid material would have an index of refraction (Nd). I would have thought that one would use a coating or sss layer to simulate these sort of effects.

[tom]

So, Just to be sure: The 'fix' will be that when we create a new material, the Nd default/preset will be a higher value - as opposed to a value one would expect to see in a dielectric material. Is this correct?

No again, let me tell the whole story from the beginning:
- Months before when fresnel was not working properly, we created these material presets (I mean MXM files) with leaving default Nd as 1.0 which is incorrect but working with old system like Nd=max now.
- Later, fresnel approach is fixed and material presets become obsolete.
- Unfortunately we left these obsolete materials in installation package and this is where all the headache comes from.

[arch4d]

...about fresnel and Nd, you may want to peek:
http://www.maxwellrender.com/forum/view ... 38&start=0

so, if i get it right, by setting the ND values we set a fresnel curve as we can set it in C4D for example (there it is as a simple gradient map), right?
where the colors picked for reflection and reflection 90 set the maximum and minimum reflectance?

[mverta]

Yes, plus search the forum for .r2 files.

_Mike

[tom]

so, if i get it right, by setting the ND values we set a fresnel curve as we can set it in C4D for example (there it is as a simple gradient map), right?
where the colors picked for reflection and reflection 90 set the maximum and minimum reflectance?

It's strictly not a falloff control curve as assumed. The curve comes from fresnel and is not user editable. You may think it as a naturally forced gradient falloff (it's fresnel itself) and yes the picked colors for each end corresponds to ref 0 and ref 90.