PDA

View Full Version : Bullet Kinematic Body - Not working?



RPSchmidt
12-02-2019, 01:31 PM
I have a standard male figure that I have rigged using Genoma. The figure is animated and is using Kinematic bullet physics with shape set to mesh.

I have a "cloth" object (a cape) using Deform dynamics.

When I calculate the dynamics for the animation, the dynamics outline of the kinematic animated body doesn't change at all from the frame 1 T pose, i.e. the figure is animated to jump and move, but the dynamics outline is maintaining the initial T pose vertice outline and only moves in x,y,z space. The dynamics outline doesn't bend or change shape at all with the animated figure.

The manual and other sources here say that kinematic dynamics are used for animated objects, but instead, it's acting like a rigid body as far as the dynamics goes.

For whatever reason, the kinematic dynamics aren't following the movement of the mesh at all. As a result, it looks like a statue in the T position moving around in space and the deforming object (the cape) is acting like that's what it is.

The kinematic dynamics are not actually following the mesh.

What am I doing wrong here?

Thanks in advance for any suggestions!

prometheus
12-02-2019, 03:05 PM
I have a standard male figure that I have rigged using Genoma. The figure is animated and is using Kinematic bullet physics with shape set to mesh.

I have a "cloth" object (a cape) using Deform dynamics.

When I calculate the dynamics for the animation, the dynamics outline of the kinematic animated body doesn't change at all from the frame 1 T pose, i.e. the figure is animated to jump and move, but the dynamics outline is maintaining the initial T pose vertice outline and only moves in x,y,z space. The dynamics outline doesn't bend or change shape at all with the animated figure.

The manual and other sources here say that kinematic dynamics are used for animated objects, but instead, it's acting like a rigid body as far as the dynamics goes.

For whatever reason, the kinematic dynamics aren't following the movement of the mesh at all. As a result, it looks like a statue in the T position moving around in space and the deforming object (the cape) is acting like that's what it is.

The kinematic dynamics are not actually following the mesh.

What am I doing wrong here?

Thanks in advance for any suggestions!

I can not track it down without animation preview, or the scene content uploaded here...too many variables that can cause it.

Ztreem
12-02-2019, 03:50 PM
It should be a kinematic body if it's a rigidbody but animated by translation and rotation. If the object is deformed by bones then it should be a deforming body. Try these settings an see if it helps.
146336

RPSchmidt
12-02-2019, 03:56 PM
It should be a kinematic body if it's a rigidbody but animated by translation and rotation. If the object is deformed by bones then it should be a deforming body. Try these settings an see if it helps.
146336

Thank you, Ztreem... I will give that a try, although I was hoping not to have to use deforming.

I don't need limbs wibble wobbling all over the place, I just need to be able to animate the body and have the deformed "cape" object collide with it.

I will give your settings a try!

Ztreem
12-02-2019, 04:03 PM
Thank you, Ztreem... I will give that a try, although I was hoping not to have to use deforming.

I don't need limbs wibble wobbling all over the place, I just need to be able to animate the body and have the deformed "cape" object collide with it.

I will give your settings a try!

Then I think the best way is to attach simple collision shapes to the body just for the cape. Just parent nulls to the bones and make them collision shapes like spheres and size them so they fit the body as good as possible.

PS. with the settings I gave you the deforming body should not wiggle around it should look the same as by deformed by bones. the only difference should be that other bullet bodies can collide with it.

Ztreem
12-02-2019, 04:35 PM
Here is a quick test with the same settings. I havn't used Lightwave once the last year, but 20+ years of hammering the software did surely leave a memory in my brain. :D
This is what I got, the sphere is deformed by bones. Even the rigid body cubes does not deform the sphere even if it is a deforming body. (just give the deforming body with bones the highest denisty)
146338

jwiede
12-02-2019, 04:51 PM
This is what I got, the sphere is deformed by bones. Even the rigid body cubes does not deform the sphere even if it is a deforming body. (just give the deforming body with bones the highest denisty)

Hang on, are you saying that when "deforming body" sphere wasn't highest density, the rigid body cubes were also causing the "deforming body" sphere to deform? That sounds off.

Ztreem
12-02-2019, 05:03 PM
Hang on, are you saying that when "deforming body" sphere wasn't highest density, the rigid body cubes were also causing the "deforming body" sphere to deform? That sounds off.

yes, when the cubes had higher density than the sphere they deformed the sphere a little.

Ztreem
12-02-2019, 07:12 PM
One more just for fun. Now I must do some real work.:D
146339