PDA

View Full Version : Stretchy IK method question


donShole
10-09-2005, 05:16 PM
I was wondering, which is preferable and why for stretchy ik setups: joint translation or joint scaling. Thanks in advance :).

NolanSW
10-09-2005, 06:07 PM
Kind of depends what you're going for. Scaling will keep the volume of the geometry where as the translation will taper the geo.

Sean

donShole
10-10-2005, 08:51 AM
Kind of depends what you're going for. Scaling will keep the volume of the geometry where as the translation will taper the geo.

Sean

Thanks for your response Sean:). I am interested in maintaining volume.

Dichotomy
10-10-2005, 03:31 PM
if I remember correctly, when they needed a stretchy IK for motion capture, the mo-cap didn't record scale attribute information, so they devised the method of using the translation for the stretch. Scale is easier, in my opinion, though translation isn't much harder.
-2 cents-

Matt Leishman
10-10-2005, 05:23 PM
www.andy-phillips.com/blog.html

go here, and in his thursday august 25th entry, andy has a nice little step by step on how to set up some simple squash/stretch with translation as opposed to scaling, a much safer route if you ask me. oh, and its an FK setup, not IK.

NolanSW
10-10-2005, 10:46 PM
Actually Matt is right. Translation is the better way to go. But lets say you are using a single arm chain and you want to have stretch on the IK as well as the FK. No prob if you use a three arm setup. Simple solution is have a two skeleton setup. Rig controls the deform skeleton. Connect the two with point and orient constraints. Then you can use both scale and translation to stretch the joint however the deform skeleton will just translate because of the contraints. Curious if this seems like a proper solution.

Sean

boogiboy
10-11-2005, 10:43 AM
Scaling always much better than translatin joints.

NolanSW
10-11-2005, 03:01 PM
Enlighten us as to why?

donShole
10-11-2005, 03:14 PM
Interesting posts, thanks everyone. One advantage of scaling over translation that I've seen is that you need little or no weight painting, because the influence areas are stretched or compressed. Whereas with translation, the influence areas are moved, so the mesh shears unless the weights are painted compensate.

NolanSW
10-12-2005, 12:03 AM
Issues I've had with scale as opposed to translation is scale can do funny things to IK.

Dichotomy
10-12-2005, 03:48 PM
I never said which was "better", just that scale is easier, don't need all the multDivide nodes. I actually use translation (learned it first, and just kinda prefer it).

CGTalk Moderation
10-12-2005, 03:48 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.