Mirror reflecting light (like laser or sun bouncing of mirror)

[Morne]

I'm trying to get a highly reflective object (mirror) black diffuse white reflect to bounce the sunlight. I'm getting nowhere fast. Do I have to use caustics for this?

[jacksc02]

Yeah enable caustics

[Morne]

I tried that, but it only shows the dot of the sun, not the actual light coming from all directions onto the mirror. So what is happening is that my sun is acting as a laser beam casting a dot, instead of one huge stream of light.
Attached is a test scene of what I'm talking about (max 2011 vray2 beta 1 or 2)

sun study solar pipe.zip

[Morne]

Increasing the sun's "size multiplier" to 200 seems to be getting there almost. But this can't be the correct way of doing it as this just makes the dot bigger. As this is for a sun study I need to be able to get a correct result.

I've also increased the sun's caustic subdivs to 4000 and put the gloabl multiplier on 17000. Search distance on 100mm Max photons on 60. I've now made the walls and ceiling medium grey (180,180,180)
This is the result so far.



How can I get it smoother and more "correct"?
Everything is perpendicular to each other so I would also expect the light to go all the way to the left wall, and not stop short of the wall

[AJ Jefferies]

Vray Sun has the additional parameter for the 'photon emit radius' - so you need to make sure this encompasses any & all objects you wish to receive/generate caustics.

I'll have a look at your file & see what I can get out of it

[AJ Jefferies]

Ok, I messed around with some of your scene settings (sorry about that) but hopefully it sheds some light (oh!) on the matter

http://moonjam.com/CG/Caustics.zip

[Morne]

Thanks. The "photon emit radius" is the one I missed. It worked great on the little test scene above.

Now I'm wokring on my actual scene and I'm getting some strange things happening.
I have a room lit only from a long chrome pipe in the ceiling. I'm doing a sun study animation for this scenario but getting some weird things as you can see from the attached images.

This is taken from a single frame where the sun just just creeps in the pipe:

1. With Caustics OFF, you can see the bit of direct sun creeping in. But the scene hardly gets lit up at all.


2. With Caustics ON, the scene is lit nicely but the direct sunlight area is gone and there is some weird larger shadow in that area.


3. Caustics On again, but this time the sun sits much lower as now it is mid afternoon. Caustics doesn't look right. Or is it?


Any ideas?

[jujubee]

You could also try a Vraylight set to skylight portal above the hole.

[AJ Jefferies]

Interesting... I'll have a play with a similar scene at lunchtime & see what happens.

[Morne]

You could also try a Vraylight set to skylight portal above the hole.

How would I then link this to the daylight system? (It needs to be an accurate sun study animation) To put it in perspective the hole is 2500mm wide and the pipe is couple of stories high.

Interesting... I'll have a play with a similar scene at lunchtime & see what happens.

Thanks I'm looking forward to your results.

I'm running a 100 frame test animation to see the effect of the sun moving. We're having a fat debate in the office whether or not the image no3 above is correct or not? Why would there be a "blank spot" in the middle of the light?

On a side note, is the caustic calculation multithreaded and can it be interpolated between animation frames?

[vlado]

I'm running a 100 frame test animation to see the effect of the sun moving. We're having a fat debate in the office whether or not the image no3 above is correct or not? Why would there be a "blank spot" in the middle of the light?

The result is correct and agrees quite well with the result from other render engines. The "blank spot" appears because of the cylindrical shape (in the real world, you will see it if you look at the caustics ring inside a glass).

On a side note, is the caustic calculation multithreaded and can it be interpolated between animation frames?

It is multithreaded, yes. It is not interpolated, but caustics are somewhat more consistent in animation as they start from the light, rather than the camera.

Best regards,
Vlado

[Morne]

Thanks for the reply Vlado. It is very good news that the result is correct (in 3D terms).
For the design however it is bad news as we didn't want the light to spread out like that with the sun at an angle. Maybe we'll put in some angled mirrors in the tube
(Boss doesn't want a diffuser at the bottom)