by dean chapman on Jun 1, 2012 at 3:26:54 am
I'm trying to figure out the best technique to simulate cell division.
I've tried simply using a metaball, but my problem with using a metaball is that it seems that the objects inside a metaball need to travel a fair distance apart before they fully divide from eachother......and once they finally break apart, they are too far away from one another. I need to be able to organically split the object in half, so that when they split they are close enough to touch eachother.
I've also tried using the collision object....which seems like it could work. I have two spheres. Sphere #1 has a collision object as a child. The collison object has sphere #2 in the colliders area (set to inside stretch). Then, I just pull sphere #2 and it works. Everything is in a group inside of a null. The null has a material with transparency enabled. My problem here is that the transparency is affecting the 2 spheres seperately, instead of as one entire object. I try placing the material on the spheres directly, but i get the same problem.
I've also tried simpy using 2 spheres and a boolean object...but i get nasty creases where the two objects meet. Using a noisy material with displacement hides the creases.....but it still just doesn't feel as organic as say the metaball approach.
Any suggestions on this....or any better ways to go about this?
I've attached screengrabs of my metaball and collision attempts.
Re: Metaball problem.....limitations by David Alex on Jun 13, 2012 at 9:30:34 am
Have you tried messing around with the Metaball tag? I found that decreasing the strength in the tag, them playing with the Hull control in the Metaball kinda fixed it. I just quickly attempted it, but I think with more time you can get a decent result