[eric nixon]

Maybe this helps explain things better, for the most common situation regarding the cup geometry;

create a crazy mesh on the glass object if it was an image of liquid being poured

Yes, but that wouldnt be very hard for most 3D apps, its just very hard to do an animated version.

[hatts]

Glad you could use it photomg1

Eric in the end you're absolutely right; having a reflectionless glass layer at that liquid boundary is the last step to make the most out of this setup. The thing is, I haven't yet produced a render that shows visible "doubled" reflections at the top of the liquid, nor do the doubled reflections seem to add drastically to each other. I think the gap, if small enough, perhaps removes the necessity to give the glass layer no specular.

I remember from the other thread that the topic of frosted glass came up. It sounds like you're using poly selections. That is probably the most flexible solution. Below is my solution using mapped roughness instead.

[eric nixon]

Well my (badly explained) example on the first page was a rough glass of sorts with a little sss. I think it renders faster with just one reflection, thats the main thing.

Ofcourse the camera might not even be viewing the top surface but maxwell will study it for sure, so best to make it as correct as possible.

[bograt]

I thought I would re-open this thread instead of starting a new one.
I am working on a glass candle model but as the wax is rougher than the glass I am getting glossy reflections on the surface of the wax.
Does anyone have a solution for this? Is it just a matter of applying the reflective properties of the wax to a duplicate of the glass material and apply this material to the membrane polygons?
Cheers

Jules

[tom]

Maybe Tom can confirm this.

I agree the top without reflection is more correct than ever. Because, a liquid surface with a lower Nd would be less reflective than a typical glass topcoat. So, it's essential and not only needed for rough glass. Anyway....Thanks, Eric! The method is accurately impressive...

[seghier]

is the diagram useful if we use metal or sss plastic material with water or milk ?

[bograt]

Thanks tom (and Eric). It does seem like the most realistic option but the last time I tried it also seemed to remove the reflection from the liquid material.
I have had other multi/sub-object glitches in max before so it could be down to that but I will look into it. Thanks

edit: It's working well now, not really sure what was wrong before. Thanks

[eric nixon]

Its worth mentioning, that you dont need to worry about the normal directions on the various geometries, ofcourse its always nice to have correct normals for the bump maps, but as far as refraction is concerned it doesnt matter, the rays are just counting there way in and out thru the layers of the objects. They remember what objects they,ve passed through - 'the object stack'.

I noticed some mis-info about this small detail on the forum.

Perhaps removing that issue makes it easier to model in a certain app, and someone on the internets has an easier day

[Bubbaloo]

Its worth mentioning, that you dont need to worry about the normal directions on the various geometries, ofcourse its always nice to have correct normals for the bump maps, but as far as refraction is concerned it doesnt matter, the rays are just counting there way in and out thru the layers of the objects. They remember what objects they,ve passed through - 'the object stack'.

I noticed some mis-info about this small detail on the forum.

Perhaps removing that issue makes it easier to model in a certain app, and someone on the internets has an easier day

True, Maxwell treats all faces as double sided, regardless of normal direction.