"Bill Pragnell" <bil### [at] hotmailcom> wrote:
> "sooperFoX" <bon### [at] gmailcom> wrote:
> > Thanks Bill, that is a very amazing shape.
>
> I'm glad you found it inspirational. I can't take the credit for formulating it
> however; I found the equation on the net somewhere.

I thought as much, however it is still important that you worked out how to
represent it in POV!


> With regards to what Alain said about radiosity and photons - I've found that
> when using very high quality radiosity with glass objects, you can get caustics
> naturally without photons. However, they're not nearly as good as photon-mapped
> caustics. If you use the simple scene you posted before then I wouldn't bother
> with radiosity, it probably won't add much (although it will clobber your
> render time). But the light-dome method, or even just an area_light, would
> probably look pretty good!

I think you are right. The main reason I had radiosity was because I started out
with something that wasn't glass, and then got over-ambitious ;)

I love the variation in shadow tones only possible with radiosity (for solid
objects) but since it is glass I suspect it will have limited use now, since
the surface itself is completely transparent except for the specular
highlights.


> I tried to make a mesh version of the knot-blob by tracing rays out from its
> centre, but sadly it occludes itself in places so it didn't work well enough.
> I'll make a mesh version of a non-blobbed knot at some point - this should
> render a bit quicker.

I also thought about this a bit. Have you seen Kevin Loney's mesh approximation
macro?

http://www.econym.demon.co.uk/isotut/approx.htm

(The download link is in the yellow box at the bottom of the page.)

If you could somehow represent the shape as an isosurface you could use Kevin
Loney's (or indeed Jaap Frank's much optimised subdivision version) to save the
isosurface as a mesh and then render that.

Or, maybe it would be possible to get the macro to work on a blob (or any type
of
object).

I'm afraid both are beyond my mathematical understanding! But maybe you know a
friend who knows a friend... :)


- sooperFoX

Post a reply to this message