Christoph Hormann wrote:

> Tor Olav Kristensen wrote:
> >
> > Yesterday I took some time off from my old CSG-macros to
> > explore the bozo pattern.
> >
> > Here's an image from this journey.
> > (The camera was placed inside the pattern.)
> >
> > In this scene there are several hundred million spheres generated
> > by only one iso-surface.
>
> Don't tell me you counted them ;-)

Actually I tried to calculate how many there are.
I estimated the number of spheres in this scene to be about 10^10.


> ...
> It looks good, i suspect it's made with no_shadow modifier.

Yes you're right -My iso has a no_shadow modifier. But I believe now that
the sense of depth suffers from this.


> A flythrough animation would be great, but 2.6 hours/frame would be quite a
> lot.

It certainly crossed my mind too. But I would need a much faster PC in order
to plan a good path for the flight. It was a lot of work just finding a good spot
to place the camera in this scene.

I even thought of a new application for my Nelder-Mead simplex macros:

Maybe I could make them determine the best "forward direction" in which
to move, in order to reach a new spot with low field strength. If this were
repeated, one could determine a flight path through the areas with low field
strength and therefore few spheres.

I.e.: An autopilot for "flying" through the "opens" in pattern-worlds.

But this would be a lot of work :(   -so I guess I wont try it.


> BTW, you said it's a bozo pattern so you probably use a pigment function which
> is slower than the plain noise3d although it should lead to identical results.

Do you mean that the noise3d function also uses a kind of bozo pattern ?


Tor Olav
--
mailto:tor### [at] hotmailcom
http://www.crosswinds.net/~tok/tokrays.html

Post a reply to this message