Imagine the following solution:

To enable a "full data" format (unclipped floating point values for color,
normal, depth etc.) you would write a special keyword in the global_settings
block or maybe specify it in the command line.

Together with the POV-Ray raytracer engine would be a post-process engine.
You control post-processing through a separate and very simple scripting
language which you write in separate files with a different extension
(.ppp?).

You could use the pp-engine in at least two ways:

1) You could call it from within your POV-script file (by specifying it in
global_settings). This way the PP-script would be automatically run after an
image is rendered, or after each frame in an animation are rendered.

2) You could run it directly. This would require that you'd somehow specify
which input images (in "full data" format) to process.

programmers could also write their own pp-engines that would read and
process "full data" files.

Also, in the raytracing engine could be a command-line switch to disable
rendering (so only parsing is done). That way you could render an image or
animation once (with post-processing), and if you find out you want to make
changes in the post-processing you could render it several times with
different post-process options without having to re-raytrace every time.

You could even use the I/O features in POV-Ray to write the pp-script (or
parts of it) and pass on variables to it that way.

Rune
--
3D images and anims, include files, tutorials and more:
Rune's World:    http://rsj.mobilixnet.dk (updated June 26)
POV-Ray Users:   http://rsj.mobilixnet.dk/povrayusers/
POV-Ray Webring: http://webring.povray.co.uk

Post a reply to this message