[Thomas An.]

edit: You mean when the ray enters from the other side, Maxwell still knows it entered the surface, but the normal angle is negative?

Here is a rule of thumb

1. If a ray hits the front of a surface then the deflection is calculated by the ND of this present as well as the previous front surface (if no previous exists, then use air)

2. If a ray hits the back of a surface then the deflection is calculated by the ND of the previous two front surfaces (the present ND is ignored).

[Mihai]

Aah, I was getting confused by the use of single sided membranes What is happening then when the ray hits the convex membrane is it thinks it has exited the previous material. Then ofcourse having B there will make a difference. So it's like Kabe says, Maxwell knows when a ray hit something, even when it hits it from the back side. That's what I wanted to understand. 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.....one the way it's doing it now, by knowing how long it travelled inside the medium before it exited, the other (if you use a double sided mat), would simply be a setting we choose for that material.

But having these two methods is confusing I think. IMO it's best to keep it as it is, and just give us a solution for when we need two dielectric surfaces touching.

[Thomas An.]

Based on the knowledge from this experiment, I believe I now have the correct solution to the liquid+glass problem.
It is rendering right now

Stay tuned !

[Thomas An.]

[Mihai]

Naaah, I don't believe you

Not without a detailed illustrated explanation.......................................

[tom]

oh shit

[daros]

oops.

[tokiop]

Thomas, congrats for your testing..

your image is very interesting, waiting to know more about it! Some of Maxwell's alpha testers are really doing a great job and make aviable features that are not already here and push to the Next Limit too..!

[tom]

Thomas, won't you tell us how?

[Thomas An.]

Thomas, won't you tell us how?

Hold on. I got to get some sleep too... from time to time
(I know, I know.... bad habits...)

[Kabe]

Hold on. I got to get some sleep too... from time to time
(I know, I know.... bad habits...)

Hey, do you know that *not* knowing how you did this could cost us our sleep, too?

Just kidding, you have 24 hours

Kabe

[Mihai]

Are you by any chance doubling one of the materials and slightly offsetting it? The glass or the liquid?

[Thomas An.]

In view of the findings from the first page:

• Intersection of ray R1 at point A evaluates as step1 Snell's law (in the propagation diagram). This is an incident membrane intersection and Maxwell treats this as an air-to-liquid interface
• Intersection of ray R1 at point B evaluates as step 2 Snell's law. This is also an incident membrane intersection and Maxwell (correctly) treats it as a glass-to liquid interface
• Intersection of ray R1 at point X1 evaluates similarly to step 5 Snell's law (of the propagation diagram). This is a non-incident intersection and the ND is irrelevant. Maxwell (correctly) uses the previous NDb and NDa in the stack to treat this as a liquid-to-glass transition
• Intersection of ray R1 at point X2 evaluates similarly to step 6 Snell's law (of the propagation diagram). This is a non-incident intersection and the ND is irrelevant. Maxwell (correctly) uses the previous NDa and NDo in the stack to treat this as a glass-to-air transition

Similarly the behavior of ray R2 is evaluates to air-->liquid**-->glass-->air (please see diagram for the "key" explanation of liquid**)

I believe this diagram will give the most correct liquid+glass result.

[Mihai]

So the reason for the glass to completely enclose the liquid is so the caustics are calculated correctly?

Because I remember the tests I did and the most correct looking one was were the glass was cut away where it met the liquid. Although in that one there was just a liquid "top", so a ray coming from above that entered then exited the liquid at the bottom, didn't know it was then entering glass at the bottom...

But if this small distance doesn't matter so much for the nd, why did my images look so different?

Note, none of these images are correct:

[Thomas An.]

So the reason for the glass to completely enclose the liquid is so the caustics are calculated correctly?

Because I remember the tests I did and the most correct looking one was were the glass was cut away where it met the liquid. Although in that one there was just a liquid "top", so a ray coming from above that entered then exited the liquid at the bottom, didn't know it was then entering glass at the bottom...

But if this small distance doesn't matter so much for the nd, why did my images look so different?

Well, theoretically it passed all the logic tests. Now if there is certain situation that a visual result is looks off ... then I don't know

The test you did with just liquid on top (which was also the design that I advocated last time) was an unbalanced design because a ray entering from above will meet only three surfaces instead of four... and this of course creates unusual results (such as with caustics or strange reflections when seen from the bottom).