Cage opened this issue on Apr 03, 2010 · 610 posts
Cage posted Thu, 06 May 2010 at 2:47 PM
Quote - I think you're off to a great start with the MetaBalls, granted there are issues/limitiations to deal with.
MetaForm was a great plugin back in it's day, certainly was great for animated fluid/smoke/flame effects. I keep thinking it used a basic directional particle(sphere) emitter prop or emitted from the surface of a figure, with particle lifetimes, direction/spray radius controls, plus collision witn non particles etc.
Looking forward to seeing what you manage to come up with. I'm not expecting another MetaForm, I think the scripts you've already created are awesome contributions, still amazed these are free.
Would be cool if some of the other gurus joined you and pooled resources(I see you caught BB's eye there), wonder if PhilC or face_off could offer any input.
So, whats next after MetaBalls?! Perhaps a fracture system for dynamicly destructable objects/scenes.. Hah.
I think a couple of fracture system scripts already exist, actually. Ockham has Exploder, and PhilC may have something as well, IIRC. Assuming I've understood your suggestion.
Some of what MetaForm does seems feasible to me. I started exploring isosurfaces because I wanted to be able to remesh an existing geometry, which would be something like what MF does with the field generation. I'm not sure whether that would be using a distance field (which potentially could be quite slow, from what I've read), or some kind of particle system.
I could also see developing the basic process into a formula object generator, to create primitives from isosurface algorithms. I could imagine implementing boolean operations between objects, assuming some kind of distance field method can be worked out. I'm less confident about animation and dynamics, which are still mysterious to me right now.
The main problem I could foresee in trying to develop anything really complex from what I have would be speed. With only three balls contributing, the processing takes three to eight seconds on my system, at a resolution level adequate to avoid holes in the resulting mesh. A distance field or particle field of any kind could involve the equivalent of hundreds or thousands of contributing balls. That could be quite slow, using interpreted Python instead of a compiled extension for the calculations.
I've thought about it a bit, and I can see two or three ways I can try to optimize my current metaball process for speed. I'm not sure how much that might help. The ultimate capabilities of the script, or any derivatives, would be dependent on how fast or slow the process ends up being, as well as my limitations as a script-writer and my more profound limits with respect to understanding complex math.
Really, it's a bit remarkable that I've gotten this far. :lol: I'll certainly see where else things can lead.
===========================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.