3

four UV sphere tetrahedral particle system, central polar vertex alignment snap

The idea is to snap the four (selected) polar vertices to each other while scaling the particle (meeting snap point, in this case, is the coordinate center). Eventually I'd like to make the sphere radius equal to the length of the tetrahedron edge in such a way to align the vertex grids so that boolean operations can be implemented. Can it be done this way or is there a better alternative ?

Basically in order to make this enter image description here

Boolean didn't turn out well in the depicted orientation of the spheres, even though the vertices appeared to be aligning.

t8ja
  • 465
  • 2
  • 12
  • 2
    Not sure to get the point. You want to merge the 4 spheres in order to obtain the figure that is rotating? What's the link with particles? – lemon Jul 10 '19 at 12:12
  • I used the particle modifier to position the spheres (rotate them all simultaneously) , and I managed to make the shape, it was quite a while ago, so i'd have to repeat it to know exactly what I had to do... it's described to an extent in the "answer" I gave myself. I'm sure I sued the wrong terms, because at first I was trying to use the "duplivert" option, but realized I couldn't orient the spheres... or did I... (I know the boolean modifier made things more difficult at one point) – t8ja Jul 11 '19 at 15:54
  • I've read it, but still do not what you try to achieve here... And a question: does this figure have a name? – lemon Jul 11 '19 at 15:57
  • 1
    Knowing what I know now, I realize the spheres could be positioned individually. Also the dupli objects (spheres) can probably be made "real" and THEN positioned individually. So really I'm asking you, how would you do it ? Or more generally, how would you make a sphere grid, that can be subjected to boolean operations, because verices are aligned to allow that ? – t8ja Jul 11 '19 at 16:06
  • Here are more shapes that can come from that – t8ja Jul 11 '19 at 16:08
  • http://theu.one/tuo/wp-content/uploads/2017/12/19429665_10213505150448493_8575390339140040262_n-1.jpg – t8ja Jul 11 '19 at 16:08
  • http://theu.one/tuo/wp-content/uploads/2017/12/cropped-Icon_one.png – t8ja Jul 11 '19 at 16:09
  • 1
    is that of some help? https://i.stack.imgur.com/gPKgM.gif – lemon Jul 11 '19 at 16:13
  • Very nice, but that's a step beyond creating the resultant object/shape itself. How would you turn (a non beveled, I see it's beveld in the gif) tetrahedron into a ReuleauxTetrahedron (convince blender that it's the same object in order for it to extrapolate the intermediate forms, presuming Blender can actually do that using a predetmined algorythim of soem sort (which could then theoretically be changed to yield some cool intermediates)? – t8ja Jul 11 '19 at 16:20
  • 1
    ok, I get it now. Leander has done it here https://blender.stackexchange.com/questions/136607/turn-2d-mesh-into-a-3d-mesh-reuleaux-solids/136624. Keyword was Reuleaux! – lemon Jul 11 '19 at 16:25
  • You opened a can of worms by showing me that gif. LOL Now I wan't to know how to morph between any two objects, irrespective of the method used to make them. – t8ja Jul 11 '19 at 16:28
  • 1
    simply a shape key from a subdivided tetrahedron and itself turned into a sphere. – lemon Jul 11 '19 at 16:35
  • This shape would be integrated https://www.youtube.com/watch?v=BHJBlnUzwQU&t=1m15s – t8ja Jul 11 '19 at 16:44
  • Ok. I need to check out examples of sape key usage. Does' the initial object have to be subdivided (have the same number of vertices) ? How did you turn it into a sphere ? – t8ja Jul 11 '19 at 16:49
  • I get it now. It's subdivided BECAUSE you planned on turning it into a sphere. So the animation is actually only missing a first step where the subdivision happens, in odrer to show a full transformation from a perfect tetrahedron to a sphere. – t8ja Jul 11 '19 at 16:53
  • Still, how did you do it ? Did you point project vertices onto a sphere ? (though I don't know how that would be done, but I can imagine it) though that option would require Blender to register a sphere as a mathematical ideal. – t8ja Jul 11 '19 at 16:58
  • 1
    we can't extend a discussion so much here normally. But some advanced technics are around this https://blender.stackexchange.com/questions/80868/transforming-morphing-one-objects-shape-into-anothers – lemon Jul 11 '19 at 17:03
  • I'm pretty sure it can't do that because it can't do this ? https://blender.stackexchange.com/questions/143416/vertex-at-curve-circle-intersect – t8ja Jul 11 '19 at 17:03
  • Let us continue this discussion in chat. – t8ja Jul 11 '19 at 17:08
  • 1
    Thanks for that link, in the discussion. I got a bit sidetracked in the meantime. – t8ja Aug 07 '19 at 06:04

1 Answers1

0

With some inspiration from this answered question, though some settings must have been changed because I can't find the other size determining value, he pointed out, I managed to find the right Render>size "0.25" which makes the particles the same size as the original Dupli Object and takes the selected polar vertices to the same spot as I wanted. result

and to do the eventual How to convert particles to mesh? and How to Make Multi User Objects Single User? is required

t8ja
  • 465
  • 2
  • 12