"Nekar Xenos" <nek### [at] gmailcom> wrote:
> I am working on making the paint macro able to use eval_pigment() instead
> of declaring a colour beforehand. Now I am able to render a complete scene
> this way.

This is a *really* interesting technique that you're developing. (I've been
curious to see what you might come up with, as a logical extension of your
blob-men renders.) You've managed to turn the '3D-painting' problem completely
around, and come up with a nice workable technique. Well done!

I've been giving some thought to an idea: a built-in POV-ray 'tool' that uses a
greatly-modified combination of trace and eval_pigment, to actually find the
color that has *already been applied* to an object in 3-D space. I can see all
sorts of uses for such a tool--like placing correctly-colored paint blobs on an
object's surface!

Currently, eval_pigment can only be fed a *pigment* (obviously) to work
with--disconnected from any object that it might be applied to. While that's a
very useful tool, it naturally has limitations. This new idea would be to 1)
find an object's surface in space; and 2) evaluate the color at a point there.
(My conception of this trick would probably require no less than a 'ray tracer
within a ray tracer' to pull off successfully; but I think the 'ingregients' for
the tool already exist in POV-Ray.) Of course, there are lots of details that
such a scheme would have to deal with-- 'modifications' to the base pigment such
as lighting, highlites, radiosity, reflection, transparency, etc. that combine
to create the visible color.

But it need not be that complicated. In essence, a tool for finding just the
unmodified color at a point on an object's surface, nothing more.

But the idea might be doomed from the start--based on how ray-tracing works in
general: Finding the point of color (doing the 'ray trace') would probably be
done during the 'rendering' stage, with the result then having to be fed back
into the 'parsing' stage to be worked with. That might not be possible in
POV-Ray.

Post a reply to this message