Making a fixed joint

[Waksra]

Hey, was wondering what would be a good way of making something similar to a fixed joint between a rigidbody and softbody?

I need both bodies to be able to effect each other which is why I initially went for the pin constraint, but it does not fix the objects to each other a rigidly as I would like.

The desired effect is to pin a softbody on top of a rigidbody, two balls stacked on each other. I'll be using the rigidbody to move around while the softbody follows around on top being wobbly. 

Any ideas on how to do this through script, or any way I can modify the pin constraint to make it completely rigid?

[Waksra]

An update here, I went with simulating the softbody in local space with the bottom part of it being fixed. 
This is not ideal however, as the softbody does not affect the rigidbody it is attached too.
I'm wondering if I need to supply more information on what I want to do in order to get a reply?

[josemendez]

An update here, I went with simulating the softbody in local space with the bottom part of it being fixed. 
This is not ideal however, as the softbody does not affect the rigidbody it is attached too.
I'm wondering if I need to supply more information on what I want to do in order to get a reply?

Hi there,

I'm assuming you want a to attach the softbody to a rigidbody, and have both affect each other (moving the softbody would result in the rigidbody following it, and viceversa).

The only way to achieve this is trough pin constraints. Pin constraints are formulated exactly as fixed joints are in a traditional iterative rigidbody solver, so they can be made more stiff by increasing the amount of pin constraint iterations (and/or reducing the simulation timestep) and making sure the mass ratio between the rigidbody and the pinned particles is as close to 1 as possible.

[Waksra]

Hi there,

I'm assuming you want a to attach the softbody to a rigidbody, and have both affect each other (moving the softbody would result in the rigidbody following it, and viceversa).

The only way to achieve this is trough pin constraints. Pin constraints are formulated exactly as fixed joints are in a traditional iterative rigidbody solver, so they can be made more stiff by increasing the amount of pin constraint iterations (and/or reducing the simulation timestep) and making sure the mass ratio between the rigidbody and the pinned particles is as close to 1 as possible.

Hey!

Yes, I want them attached, but they should also stay in exactly the same positions relative to each other, not move an inch. I have increased the iterations, the stiffness is set to 1 and the mass ratio is 1. But I did not see much difference, don't really know what I'm doing wrong. 
I have the ball on top of the rigidbody and when I hit play the softbody falls down onto the rigidbody before sliding down next to it. 
   
It then acts mostly like a spring joint with the pin constraint stretching as I move the lower body before dragging the upper body after. 

Any ideas what I could be doing wrong?
If you need any other info just let me know.

Thanks!