reflected / refracted caustics

[super gnu]

hi... caustics problem... is there any way to differentiate between reflected and refracted caustics in vray? it would be great if there were seperate multipliers for each, but since there arent, what material properties control the strength of the two types?

i have a swimming pool, with vray lights set into the side of the pool.

the caustics reflected from the underside of the water surface onto the sides of the pool look great, but very little light escapes through the surface and up the walls around the pool... to get the caustics outside the pool bright enough, i have to completely bleach out those below the surface.

i have the water surface set to not cast shadows, it is a standard material with vray maps since it gives me more control over the opacity settings.
i tried a vray mat, but i have the same problem....

i have a fresnel falloff on the opacity since it makes the ripples in the water look lovely, but adjusting the opacity doesnt seem to help anyway.

ill post an image if necessary but i guess the problem is pretty self explanatory....

[super gnu]

somebody must know something! surely....

[losbellos]

try with color mapping 3 or so HSV exponential.(dark multiplier)
All multiplier (caustics, light on 1) looks okay.

[super gnu]

colour mapping? that will colour correct the whole image! surely there must be a way to control reflected and refracted caustics independantly of each other??

this is gonna be part of a huge scene, and i dont want to have to set up the whole scene so that i can colour correct it just for one bit of caustics...

[losbellos]

your welcome, good luck!

[super gnu]

sorry...pretty rude of me... thanks for your help, im just a bit stressed with a deadline approaching and no one seems to be able to answer my question!

do appreciate it

[losbellos]

show me an image of how far you got. And color mapping is a little different then color correction. You can do things with color mapping, with color correction never. Its basicly a dark and light multiplier.
You could benefit from it quite much...

[losbellos]

late this evening I'll show an example...

[super gnu]

http://www.smoothe.com/new/robin

here it is.... its a very rough test, just to get materials and lighting... it will be used in a big project, landscaped beach club with thousands of plants, people, buildings... (people diving into the water etc.)

as u can see the caustics practically stop at the surface (there are some getting through but not a lot...) i appreciate that i can boost them up using the colour mapping, ( basically colour correction on vray's internal 64 bit image as far as i know) but surely this means that anything else in the image with the same brightness range will be boosted by the same amount?

id LOVE to be proved wrong, cozs it would sort out my problems... buty what im really after is: which material properties determine the strength of the reflected caustics and which control strength of the refracted ones?

you surely must be able to adjust them relative to each other rather than just brightening up the darker ones?

anyway, thanks for your help losbellos, youre the only one to attempt it, and i was rude to u

[DanielBrew]

Super Gnu

Just a thought - is the water a surface with no thickness or is it a solid filling the pool? The reason I ask is that I had a similar problem with some curtains. I modelled them by extruding an open spline. They wouldn't let light through until I gave them a thickness.

[super gnu]

it is indeed a surface with no thickness... hmmm.. let me try giving it some thickness, or maybe make it double sided... hmm

[DanielBrew]

I think giving it a thicknes would work better, that or making the water a volume which fills the pool as real water would. If you choose the 'fill the pool' option I would make the water bigger than the pool so that the sides of the water and the sides of the pool are not coincidental. See all the posts re. wine glasses filled with liquid.

[losbellos]

in my case even without (sorry I missed this) color mapping does work well. If u use lights below the water surface It that makes troube, you wont see the caustics. Lower maybe the multiplier. By the way your image looks cool to me. Here is my try.(without thinking too much on design

[J_Bug]

Do your lights in the pool have any falloff or attenuation on them?

--Jon

[losbellos]

I just checked on the net (google:"pool night") I just didnt find any image, where caustics is noticably visible where the lights are underwater.