[Aldaryn]

Whoa.. this is big.

[rivoli]

well yes, thanks for sharing your valuable knowledge on this matter fack.

i think your name is mispelled though.

[Mihai]

You're going to love this thread then

http://www.maxwellrender.com/forum/viewtopic.php?t=2906

It's called "conclusion", 20 or so pages worth of conclusions. Just read what Thomas An. posted and there is also a photo comparison on page 13 I think.

[Thomas An.]

You're going to love this thread then

http://www.maxwellrender.com/forum/viewtopic.php?t=2906

It's called "conclusion", 20 or so pages worth of conclusions. Just read what Thomas An. posted and there is also a photo comparison on page 13 I think.

You don't need to read all 20 pages.

All answers and comparisons are now summed in the first page.

Keep in mind that with the beta 1.2.1 and the bug in dielectrics this will look like its not working (but it is).
http://www.maxwellrender.com/forum/viewtopic.php?t=4275

[noseman]

I was just browsing through this thread and after drinking some wine, I started thinking!
I may have a possible solution but I'm too drunk to think it over.
Maybe someone has already suggested it, so please be kind to me

here goes nothing:



Which case of this post (http://www.maxwellrender.com/forum/viewtopic.php?t=2906) are we talking about here?

I am rendering a test scene at the moment and I'm using thick glass to get an amplified view of the final effect. I will let it run until tomorrow morning. I will post the result then!
Any speculations, bets e.t.c

[Axe]

According to Oscar:
http://www.maxwellrender.com/forum/viewtopic.php?t=4275

Dialetrics intersecting each other won't work right now. I believe this was tried previously in this thread:

http://www.maxwellrender.com/forum/viewtopic.php?t=2906

[Thomas An.]

Hi noseman,

Your method will not produce correct refractions in the liquid, because a beam coming from the side will behave as:
EnterGlass1-->EnterLiquid-->ExitLiquid-->EnterGlass2-->ExitGlass2-->ExitGlass1

You can double-check by looking at the propagation diagram:
http://www.maxwellrender.com/forum/viewtopic.php?t=2906

Besides. I don't think you can do this test now, bcause there is an issue with dielectrics inside (or partially inside) dielectrics with the beta 1.2.1.

[tom]

hi noseman, don't worry brother, we tested every possible situation

[noseman]

just for the sake of it, here is the test:

[Thomas An.]

noseman,

This is nice but not too usefull. We don't know if the refraction is correct.
Use the experimental scene file from the link, so that we can compare with the actual photo.

[noseman]

I can't seem to find the link Thomas. Can you tell me where it is?

Thanks!

[Thomas An.]

I can't seem to find the link Thomas. Can you tell me where it is?

Thanks!

noseman,

Here is the link:
http://www.maxwellrender.com/forum/view ... cene#27837
http://mywebpages.comcast.net/ThomasAna ... d_Glass.7z

You can modify the glass in that file to your technique and then we can see if it matches the photo. Also, tell us what ND values you are using.

[noseman]

will do, thanks!

[Micha]

Thomas,

Thank you for the tutorials. Do somebody have images that compare both methods? Must be the glass splitted allways?

[Frances]

I wonder if you all are splitting hairs when you should just be splitting surfaces.

Mihai, I'm not sure how you got coplanar dielectric surfaces to render without being screwed up looking. When I tried it...well look:



In life, the liquid is, for all intents and purposes, indistinguishable from the adjacent side of the vessel in which it resides. In order for the surfaces to appear as one, there can only be one surface or the renderer becomes confused. As advanced as MW is, we are still a couple of generations away from being able to simulate this condition with a simple mechanical duplication of a real world situation (one solid residing in another).

Here is the evil-no-good-lyin'-cheatin' split version (pardon the reflection noise):



(11hr 24min/21 samples reached)