View Full Version : Rigging 2 objects to be able to follow eachother?
10-04-2007, 11:39 PM
I have two objects, that i need to follow eachother. ObjA follows ObjB for a while, then i need to blend that off, and blend another constraint on so that ObjB follows ObjA for a bit aswell. How can this be done?
Note that this is for a rig, so deleting parts of a rig or whatever is not an option. The solution needs to be blendable. At the moment i have 2 parent constraints (A>B & B>A), and even though the constraints never overlap (they never both have a weight above zero at the same time), i am getting cycle errors. Can anyone give me some ideas? Thanks :)
What if you had two more objects, C and D, that both A and B are constrained to (instead of constraining them to each other).
A&B weighted to C. B blends over to D and A continues following C. After a while A blends over to D as well.
I think that would solve the cycle problem, but would it fit into the rig?
10-05-2007, 02:41 PM
It would not fit the other stuff in the rig, but in theory it would fit in the rig itself. I'll need to do some testing to see how its fully done. Im confused though, where does B blend to? Because, in the end, something(s) needs to be constrainted to both hands. There needs to be a way to constrain HandA to HandB, and also HandB to HandA, so won't there always be a cycle error with this type of thing? (or atleast warning, and potential error)
10-05-2007, 06:11 PM
Jaz Tigan wrote a plug-in based on ILM's Siggraph 2006 presentation and code on Bi-Directional Constraints though I don't believe it is blendable (might have been updated but pretty sure it's a switcher) to allow the parent->child relationship to also behave child->parent. If you search the Maya forums here the post should still be around.
10-05-2007, 06:59 PM
It would be nice if Maya was smarter about this if you wanted it. Ie, if i had two hands pinned to eachother, at equal values, they simply would not move. HandA would cancel out HandB, and etc. Where as if HandA was to follow HandB at 100%, and HandB was to follow handA at 75%, you would not be able to move HandA, but you would be able to move handB, with 75% diminished returns.
10-09-2007, 04:17 PM
It would but unfortunately using a direct acyclical graph doesn't allow for that (hence the cycle errors). The bi-directional constraint is quite a bit of a hack to really make it function, Motionbuilder has a top-notch implementation of the bi-directional constraint though.
10-09-2007, 04:17 PM
This thread has been automatically closed as it remained inactive for 12 months. If you wish to continue the discussion, please create a new thread in the appropriate forum.