Interesting.  I've switched to pre-compiling the cube corners and doing lookups, and this seems to have removed the trouble I was seeing with unwanted influence which prevented the screening of out-of-range balls.  With all of this screening in place now, the script will process ten balls in about a second, at a mesh resolution setting of 12 (which ends up generating a 24 X 24 X 24 box grid for the marching cubes).  So it's looking better, but still not particularly fast, if cases with hundreds or thousands of balls (or particles) are truly being considered.

I think I can get more speed with an octree scheme of some sort, although I'm not yet sure how that will impact the new pre-compiled data method.  I also have an idea of how to better control the final mesh density while avoiding any processing of empty spaces without any ball influences.  I'll see what I can do with that.

Thanks for the links, nruddock and ADP.  :thumbupboth:  I'm not prepared to commit to trying particle simulations, but more information and examples can definitely help me think about the idea more effectively.

===========================sigline======================================================

Cage can be an opinionated jerk who posts without thinking.  He apologizes for this.  He's honestly not trying to be a turkeyhead.

Cage had some freebies, compatible with Poser 11 and below.  His Python scripts were saved at archive.org, along with the rest of the Morphography site, where they were hosted.