Hi all

This is a followup to a question I asked a few months ago about the modelling of
concentrator optics using POV-Ray.

I want to pursue the idea of making quantitative measurements of the
photons/rays passing through my modelled scene. The model is really simple:
just a bunch of mirrors and only a single reflection for each ray. I want to
find out the shape of the 'spot' that is formed when light reflects off an
imperfect mirror. I have previously done this using some commercial software
called OptiCAD but I would like to know if it can be done realistically with
POV-Ray instead.

The idea is to have a flat 'film' surface placed in the scene. This film would
do nothing except count the rays that pass through it, and record their
location, perhaps using a grid of 'bins'. Then, when a raytrace had been
completed, I would like to know how many rays had passed through each bin.

Such a thing can probably be hacked with POV-Ray by creating a lambertian
surface at the focal plane of my mirror, then 'looking at it' with the camera.
But this introduces extra complexity (extra surfaces, extra reflections,
possible multiple reflections) to the model that, for the sake of fully
understanding the process, I would rather avoid.

Can such a thing as a 'film' be added to POV-Ray? Is this idea compatible more
with the 'photons' techniques (which to me seem to be a bit of an approximation
to 'real' photo raytracing: the ideas of 'depositing' photons, and considering
'caustics' as entities in their own right doesn't entirely gel for me...)? Or
is it even perhaps compatible with the reverse raytrace 'from the camera' idea
somehow?

Any suggestions/thoughts would be very much appreciated.

Cheers
JP

--
John Pye
http://pye.dyndns.org/

Post a reply to this message