In article <web.40bb4a79e36afb23a5b82adb0@news.povray.org>,
 "eric_degiuli" <eric.degiuli (at NOSPAM ) utoronto.ca> wrote:

> I am interested in what the difference is between rendering media with
> scattering with radiosity on/off.  I have looked through the source to see
> how the scattering procedures work, but I don't understand the relationship
> between them and radiosity.  Does radiosity include multiple scattering?
> What is the difference between the physical propagation of rays through the
> media?  I know this is vague, but perhaps someone will understand my
> confusion.

Media does not scatter scattered light...that is, it is unaffected by 
radiosity, a brightly lit object will not cast a glow on surrounding 
media. Scattering media is illuminated only by direct light or by 
deposited photons, both of which are attenuated by passage through the 
media.
However, scattering media does affect radiosity...a bright area of media 
will illuminate nearby surfaces. [1] And radiosity does simulate 
multiple levels of scattering, the default being 3. You can use the 
recursion_limit parameter to control this, but going above 3 or 4 is 
usually a waste of processor time.

[1]: However, when I tested this, I found problems when the surface was 
inside the media container. This was in MegaPOV, not a recent official 
version, but it may exist there as well.


> Also, some of the documentation on scattering doesn't make sense.  It says
> that having extinction any value other than 1 is not physical, but
> according to the definition given, this simply means that the ratio of
> scattered to absorbed rays is 1.  In normal radiative transfer methods,
> there is a dimensionful quantity kappa which describes the amount of
> scattering per unit mass of media. 

The documentation is not entirely correct. Values less than 1 are not 
physically correct...transmitted light + scattered light adds up to more 
light than hits the media. Values > 1 simulate media that absorbs as 
well as scattering light.


> Also, the optical depth is
> dimensionful.  In POV ray there are no units involved (as is common in
> computer science) so it appears that extinction is a parameter that
> incorporates unit conversion as well as the traditional extinction
> parameter.  Is this correct?

It does not do any unit conversion. The appropriate values depend on the 
dimensions and light intensities in your scene. I'm not really certain 
what you're asking here...

-- 
Christopher James Huff <cja### [at] earthlinknet>
http://home.earthlink.net/~cjameshuff/
POV-Ray TAG: <chr### [at] tagpovrayorg>
http://tag.povray.org/

Post a reply to this message