Volumetric caustics position rules

[Reaversword]

I have a volumetric caustic scene working (classical participating media volume + crystal sphere + light casting photons).

The fact is, I haven't very clear rules of Vray volumetric caustics. In understand it not always behaves as in real life it should be.

I'm using volumetric caustics, mainly for get a clue of whats really happening with my caustics, more like a "photons viewer" like other thing..


Although my real purpose is a scene lighting where light becomes from a jacuzzi (full of water), creating a caustic lightning all over the scene, so with this purpose I've been trying to understand how Vray caustic system works. So I make some tests, maybe its me who is wrong, so I would like to share this with the forum.


This is first test. Two crystal balls and a reflector (half donut), inside a fog volume. Light is OUTSIDE volume:


Second test. Exactly the same, but this time, with ligth INSIDE volume. No caustics anywhere. Why?. Its that physically correct?


Third test.
Blueball and Light, out of Volume, Green ball and reflector inside it. Caustics doesn't go through green ball. Is this physically correct?. I have done another test (not published), turning green ball in a blue ball (for the blue beam light), still doesn't go through the ball. The same with two balls with no color (fog vraymtl color to white), no light beam go-throught.


Fourth test. The same as the third, but with light INSIDE blue ball, and this two still OUT of the volume. Green ball and reflector inside volume. Should not be "blue" light exiting from inside blue ball?



Fifth one: Light INSIDE blue ball, and both INSIDE volume. Green ball and reflector, outside


I need another post to continue (forum limitations).

[Reaversword]

Sixth test: Light OUTSIDE blue ball, both INSIDE volume. Green ball and reflector outside volume



Scene to play:
VvolCaustics.zip


How could we see light converging clearly inside crystal balls?

Would be possible to see light beam from light source to the final receiver object going through all refractors/reflectors (every one with its own ior) without any "visual cut" in the beam?

Vray volumetric caustics respects TIR? (Total internal reflection, it reflects completely all photons travelling throught the water if they are received in the water-air frontier with an angle close to be at glance?)

Which would be the way to see a light beam be casted from a luminarie inside the water (Jacuzzi), travelling throught water, exiting at the air in a foggy day and finally, being received by a wall (or reflected by a metal piece till light beam meets a diffuse/light-absorbent material)?

I'm made a mess.

[Svetlozar Draganov]

Hello Reaversword,

Thank you very much for the interesting post - it is a tricky one , but I'll try to explain you why is that happening and what must be done in order to achieve expected result.

Case 1: Everything is ok - the result is expected.

Case 2: The problem here is that the light is placed directly inside the fog and Vray doesn't know where the fog starts and where ends and as a result it doesn't shoot correct rays for caustics. The same thing applies if you have a camera inside of the fog - and the result is missing fog. In such situations when Light or camera are placed inside of fog-objects you have add an object with inverted normals with the same fog-material applied and place the camera or light inside that objects. By doing this Vray will be able to determine start and end point of the fog and the calculation from now on will be accurate


Case 3: Caustics after the green ball are missing because the intensity of the light passed through the blue ball is very very low and furthermore Cutoff Threshold is ON by default which clamps light with low intensity. If you set that cutoff to 0 and if you increase light intensity or caustic multiplier you will be able to see how the light is passing through 2,3 or more objects - in my example I even copied the blue ball after the green ball and the caustics are visible through all of them.

[Svetlozar Draganov]

Case 4: Same as Case 1 - but this time the issue is not in the fog objects but in the blue-ball-object. Again Vray doesn't have input and output point of that objects, and the rays are not fired properly, but if you add the to ball a little thickness the issue is solved.


Case 5 - Here we have a combination of Case 2 and Case 4 - so we already know how to fix them separately Just add an object with inverted normals and with the same fog-material wrapping around the light and create some thickness to the blue ball.



Here are all Maya files:: 342_Caustics.zip

[Reaversword]

Beauty. It's wonderfull see all working and understand how and why.

Thank you very much Svetlozar, this is really interesting. Now I have some funny hours waiting me with volumetric caustics. Everything perfectly explained.

I has no idea that with camera this issues happens too!. Good to kwow!.


By the way, I know I insist perhaps more than I should, but, one more time, remember add this important info to the documentation page of volumetric caustics (or in the example page).

Thank you one more time Svetlozar!.