[Thomas An.]

Note, none of these images are correct:

Yes, but which one is identical to the proposed design ?

(I know the first one is not)

[Mihai]

Ok, I see. I suppose all the rays that hit the liquid first would contribute to the incorrect result and making the image look so different.

Thanks a lot for these experiments A++

At least for now we can have accurate results for stills.

[Mihai]

To the proposed design? Which one?

[Thomas An.]

Can you sent me a 3ds or obj file of this scene (so that I can genrate your glass in Rhino and try it here ?)

Also, let me know what are the positions of your camera and target (so as to get the exact same view angle too)

[tom]

Thank you a lot Thomas!
Great effort there......
And I'm still curious about an engine improvement for this.

[Thomas An.]

To the proposed design? Which one?

The newest one, in the previous page (here)

[Mihai]

Oh, none of my images are like your last one (with the glass surrounding the liquid). I'm just curious though why in my image with the water slightly inset from the glass produces such a big difference in refraction.

I'll make those obj's in a min and post them here.

[Thomas An.]

...This makes me think having double sided materials is not a good solution since that would mean Maxwell would need too methods to calculate absorption etc....

I agree with you here. Double sided materials might convolute the paradigm. There must be a better (more elegant) solution.

[Mihai]

Can you import fbx?

[Thomas An.]

Oh, none of my images are like your last one (with the glass surrounding the liquid). I'm just curious though why in my image with the water slightly inset from the glass produces such a big difference in refraction.

I'll make those obj's in a min and post them here.

Easy... because if you do it that way the glass and the liquid faces are "looking" at each other and they are reflecting each other. So when a ray travels from the outer glass wall and then through the inner glass wall, it becomes microscopically trapped into a multitude of reflections in that tight space.

This does not happen at all with the new design.

[Mihai]

Ah right.....do you still want my glass then? I'm going to remodel it with your method and make another test.

[Thomas An.]

Can you import fbx?

I am afraid not.

But I can import:
3dm, 3ds, iges, step, dwg, dxf, lwo, ai, raw, stl, obj, ply, stl, gft, gts, a medium Sprite, french fries, a double whoapper with cheese, and coleslaw.

-

[Thomas An.]

Ah right.....do you still want my glass then? I'm going to remodel it with your method and make another test.

Ok, then if you do it then I don't need to. But I am curious of the result

[Thomas An.]

Thank you a lot Thomas!
Great effort there......
And I'm still curious about an engine improvement for this.

Thanks.

It seems (given the new explanations) Maxwell is doing things more correctly than I previously thought.

Given how thorough Maxwell already is, I suspect any improvement will be in the form of a trivial tweek in the paradigm somewhere. Also, the "tolerance" idea seems interesting (in a pair of facing surfaces A and B, exclude surface A from reflections in a space less than epsilon .... or something to this effect.)

Or maybe yet a more elegant solution than only Oscar can figure properly.

[Kabe]

Oh, none of my images are like your last one (with the glass surrounding the liquid). I'm just curious though why in my image with the water slightly inset from the glass produces such a big difference in refraction.

This is for a very simple reason: If a ray leaves a dense medium like glass into air, then it might be bend back into the medium, which is called total internal reflection [1]
This is responsible for a number of optical effects like the ones I've shown here for an old project of mine:
http://www.bsmooth.de/BSolutions/#TIR

So, when you have a small air gap between glass and liquid, then the effect is not neglible at all, but very prominent!

So, why does it work then with Thomas' setup? That's because the NDs are very close to each other, so TIR will happen only on very small spots, and most of these show TIR in nature, too.

The only drawback of the approach of Thomas An. is that it's a bit harder to model, but optically such a setup is the cleanest way to render it - probably this is true for most available renderers.

The approach I use normally is to make the liquid bigger than the glas - which works very well with Cinema, but not with M~R. well, a good reason to think of an elegant solution for the TA setup.

Kabe

[1]A note to our english native speakers: In the US it's often named internal refraction, which is plainly wrong - according to US phyicists who know