"Samuel Benge" <stb### [at] hotmailcom> wrote:
> High! :-)
>
> This is just a quick test of an idea I had for underwater air bubbles.
>
> There's a point set of 4D vectors containing bubble positions and sizes. For
> each
> bubble there is a blob. For each blob all other points are tested, and if
> another
> bubble comes close to the current bubble, a blob is subtracted from it. A
> duplicate blob is also subtracted from underneath each bubble. Parse times
> aren't
> so horrible for low-count point sets (<250), and the render times aren't bad at
> all.
>
> The first part of the basic idea came from someone from these newsgroups, but I
> can't remember who it was :(
>
> The initial render (/the/ render) took less than 50 seconds with max_trace_level
> set at 10. In-POV post-processing luminous bloom and chromatic aberration
> effects
> were applied. Textured fog was also used.
>
> The arrangement of the bubbles was the hardest part, as there needs to be a
> logical progression of large->small bubbles.
>
> There's more that can be done, such as having the larger bubbles capsizing into
> numerous smaller ones below, but I'm not quite sure how to script that yet...
>
> ~Sam

This looks very nice.  You were able to get that bubbles rushing to the surface
look, as opposed to suspended air bubbles that basic spheres look like.

-tgq

Post a reply to this message