-
JayG 1,164 posts
So i've got an animated face, it's a deformable mesh, with full head and facial animation.
It has earrings hanging from the ears that are two seperate objects linked to the head joint. Right now they are just stiff and following the head in an obviously very unrealistic way. So I'd like to be able to simulate them based off the head movement so they look normal and pivot on the ear lobes.
My first instinct was to try a point-to-point constraint between the earring and head geo. This didn't work out because if the head geo isn't being simulated it seems reactor just kind of ignores it. I did some tests with a regular box constrained to a sphere. If the sphere had it's own animation the constraint wouldn't work properly, but if I simulated the sphere normally it works fine. I can't simulate the head geo since it's an animated mesh so this is where I got stuck.
In order for the constraint to work it needs both parent and child in a rigid body collection, and the constraint solver only allows you to choose an RB collection, so using a DM Collection was out.
This seems like it should be easy....but we can't get it to work, any ideas?
-
ivanisavich 4,196 posts
You should be able to just do the point constraint thing....did you make sure your head geo was set to "unyielding" (and/or, the object that the point constraint is following, like the sphere in your sphere test)? That's how to tell reactor that the object has keyframes, otherwise reactor will just use the first frame of the simulation as the reference for all other frames. The unyielding setting is in the properties dropdown.
-
JayG 1,164 posts
Interesting I just redid the test at home and it seems to work with my sphere test. Must be a world units thing...
I set the collision tolerance to 1 and it worked decently.
I'll give it another go tomorrow and get a better idea of what's going on.
-
JayG 1,164 posts
I got it working finally. It ended up being a World Scale issue. None of the parameters were changing anything so I thought maybe I had the process wrong but once I got the world scale to the correct value everything else fell into place. Thanks!
-
ivanisavich 4,196 posts
Ah ok, glad you worked it out
