I was fiddling around with media, trying to figure it out (who isn't)
and I ran into an interesting issue. It would seem that all light
passing through a turbulent media are warped. When I had a room with
reflective walls, the reflection of everything on the far side of the
media object (in this case, a fireball) is warped and crazy looking. Is
there any way to introduce turbulence to a media object without messing
up reflections? I know you can see through the media object without
distortion, do the reflection rays follow a different series of
modifiers? This isn't a particularly critical issue to me, mostly a
point of curriosity.

Josh Sullivan

Post a reply to this message

Interesting effect, but I couldn't reproduce this.
I used 2 planes: one checkered, the other reflective; the camera looking at
the reflective plane. When a sphere with turbulent media (spherical mapping)
was added, the reflection of the checkered wall throug this media was not
warped at all. I tried all media types (scattering, emitting and absorbing).
Could you post your source code or give more specific info about your scene
(media settings etc.)?

Margus

Joshua Sullivan wrote:
> 
> I was fiddling around with media, trying to figure it out (who isn't)
> and I ran into an interesting issue. It would seem that all light
> passing through a turbulent media are warped. When I had a room with
> reflective walls, the reflection of everything on the far side of the
> media object (in this case, a fireball) is warped and crazy looking. Is
> there any way to introduce turbulence to a media object without messing
> up reflections? I know you can see through the media object without
> distortion, do the reflection rays follow a different series of
> modifiers? This isn't a particularly critical issue to me, mostly a
> point of curriosity.
> 
> Josh Sullivan

Post a reply to this message

Related(?) note:

I don't think I kept the source code, but I had a scene where there was a
media object (I forget what kind it was) with a spotlight going though it, and
oddly, the part that was lit by the spotlight seemed grainy--kinda like the
way coincident surfaces do. I didn't dig deeper into it, and I'm afraid I
don't recall any further details, but I'll experiment and see if I get it again.

-Mark R.

Post a reply to this message

I don't know if this would apply to the original question, but what you
describe sounds like an artifact of the monte-carlo sampling used for
media, which basically means random samples.  Not enough samples and you
get that grainy look.

This is the most noticable difference between halos and atmosphere and
media, since the previous 'volume' method seemed to take samples at
regular samples along the viewing ray which gave a smooth yet banded
look at lower sampling amounts.

-Mike

Mark Radosevich wrote:
> 
> Related(?) note:
> 
> I don't think I kept the source code, but I had a scene where there was a
> media object (I forget what kind it was) with a spotlight going though it, and
> oddly, the part that was lit by the spotlight seemed grainy--kinda like the
> way coincident surfaces do. I didn't dig deeper into it, and I'm afraid I
> don't recall any further details, but I'll experiment and see if I get it again.
> 
> -Mark R.

Post a reply to this message