Ah so the tears come from the handles interfering with the autogrouping as we discovered during beta... which surprised me that Poser let that happen since I deliberately omitted the geometry of the handles in the mannequins to avoid this. Pointless thus.

However there is a fairly simple workaround. You basically fool poser into using simpler autogrouping from a previous session. It sounds annoying but its not that hard.

You actually just rig it twice: once without the handles (just select the essential bones like chest, abdomen, hip etc, and NOT the handles. With the dress, you even ignore the legs), don't transfer the morphs and do autogroup. The first session just autogroups the mesh for you, don't bother with joint zones and morphs therefore. The second time round you use the result from the first session and rig it again to the mannequin, but this time round don't autogroup (because that was already done) but transfer the morphs, any bone / handle you like, and of course the joint zones / weight maps. This is also described in the manual.

Anyhoo I hope Smith Micro are watching because this can be fixed to work better imo. Same goes for this whole process. I'm sure many things here can be sped up eh?

@digitell: oops @ the manual saying to conform it. Indeed not even needed. Will change that.

"The fool is not the one that does something foolish, but the one that does nothing at all."