Hi,

I have been away from Poser a bit due to workload. Yesterday I was thinking: If there are weightmaps .... why are there at all actor groups? After all the assignment of a poly to an actor is identical to just a weightmap entry with 1.0.... so... why not get rid of all the groups, and use weightmaps for topological assignment of vertex to bone.

Does this already work? I heard about single group meshes earlier, but do not know how that evolved.

Second: Why not have morph targets for rigging weightmaps? If I remember, there was/is a valueAdd/Sub/Mult etc. stack for joint spheres... so its basically all there already. If we could ERC the combination of weight maps (slave that to dials), we could use that for further improving the genesis-i-fication of figures. It would ENTIRELY remove the need for a neuter base and treat all extremization morphs equally as an unmorphed base.

But maybe that all works already and I just dont know. It would be cool this way, the whole grouping issue could disappear. Grouping would be dynamic and ERCable... and so would be the rig. Hm.Hm.

LOL..I like how you think out loud ;).

Laurie

According to phantom3D, the actor divisions are actually better to rig.  As I understand it, they also split up the weight maps in a way that somehow makes more sense internally and functionally.  :unsure:  They do split up morphs.  Every morph would need to cover the entire figure geometry, without the actor division.  Without the actor splits, a Python script would have to iterate over the entire mesh to process a few vertices in what is currently one actor.  :scared:  In fact, get rid of actor geometries and huge parts of Poser's Python API would need to be expanded or revised to accommodate the new setup.

But keeping them means keeping the welds, and the welds are a PITA and a mystery.  :lol:

Apparently the single skin figure type is intended solely for export as a certain weight-mapped format.  Or something.  Maybe shvrdavid will happen along and explain it again.  Anyway, if you try converting a figure to single skin using the menu option, you'll see that bulge maps apparently stop working, among other things.  I guess that export format doesn't support bulge weights, so those aren't supported in single skin.  Very limited utility, that single skin format has.

One could make a single mesh figure using body handles for the skeleton.  That could work, AFAIK.  But you'd run into the potential problems I noted above, and probably others.  :unsure:

Hm. you are rightIguess it would not be very backwards compatible. However, with regard to performance, one could query the figure for an actor map, asking which global vertices the actor map currently containes. t wouldbring another indirection, but one wouldnt have to process the entire figure - only the verts tha belong to the current actor map. bulge, hm, so bulge is not defined byjoint settings only.. i shall look out for shrdavid. We need anopensource cr2 interpreter :D