On Wed, 31 Mar 2004 08:30:42 -0500, Christopher James Huff 
<cja### [at] earthlinknet> wrote:

> In article <opr5ofawzup4ukzs@news.povray.org>,
>  Phil Cook <phi### [at] nospamdeckingdealscouk> wrote:
>
>> Well I've gone through optics.pov which shows reflected beams etc.,
>> unfortunately without explaining how it works, and pulled together the
>> following scene with comments.
>
> Hmm. I'll try to make some more informative comments for the 3.6 version.

<silly grin> It's a great image, takes ages to do on my computer though; 
it was quite fun dissecting it down to the bare bones

>> If anyone can offer an explaination why
>> both refraction and reflection need to be on in the light_source;
>
> Because you use both refraction and reflection. Sort of. It seems that
> photons will only be generated if the target is the first object the ray
> from the light source hits. In this scene, the first object hit is the
> media container, which then blocks the mirror from getting any photons.
> I recently ran into this with a candle scene, where the offending shape
> was the flame container. You can sometimes partially work around it by
> making the media container shadowless, but that won't work for this
> situation. It's almost certainly a bug.

Well I'm glad it's not just me that has all this fun :) Hmm trying to 
follow that, so photons only penetrate through the media container (first 
target) if both reflection and refraction are on; that does seem weird as 
I thought refraction was just for caustics and there are none in the scene 
I posted just reflections, seems a lot of calculation is going on for 
something that as far as I can tell isn't used.

>> why the
>> light_source must be outside the media container in order to get a
>> reflected beam;
>
> No idea. I'll have to investigate that further.

It does explain why reflected beams don't show for a global media, which 
is what first stumped me as too why nothing seemed to work. I've also 
tried it for a point light in a hollow open cylinder as soon as the 
light_source enters the media though...

>> and why a cylinder light_source produces sharp edges in
>> the original beam with refraction on, I'd be most interested.
>
> You might be seeing the difference between the original light source
> beam, rendered by usual raytracing, and the beam of media-deposited
> photons.

Yep "media_interaction off" on the light removes the edges. However it now 
shows that when the light_source is inside the media container all beams 
vanish, so the original beam I saw must have been created solely by 
media_interaction. What fun <deep voice> By Grabtar's Hammer I *will* 
master the strange phenonemon that is media :)

--
Phil
-- 
All thoughts and comments are my own unless otherwise stated and I am 
happy to be proven wrong.

Post a reply to this message