What I'm trying to do is create a few bouncing jello cubes. I can make a jello cube. And I can scale it to make it bounce. I can't do both.
Scaling it requires Soft Body Goal to be enabled, but enabling Soft Body Goal glues it in place and prevents it from actually bouncing around the scene.

Bring in more of these objects, and you have another issue in that they don't want to collide. You can sort of fix that last issue by tacking on a Collision modifier but it seems a bit buggy. And I'm also back to not being able to scale the cube. As user3800527 suggested I tried tacking on a Rigid Body to that, but that just led to some really weird behavior.

As far as trying to make the soft body animateable, I've tried just plain animating the scale of the object as well as animating the Soft Body Edge Spring Length property, but those don't appear to affect the simulation while it's taking place. Soft Body Goal is able to slightly fix this, but when you turn that on you're trading that for the soft body's freedom of movement.

Here's an example blend file showing some of what I've tried so far, layers 1-3 have different things I've tried, and all of them have the bottom cube scale from 1 to 2 from frame 20 to 23. http://pasteall.org/blend/index.php?id=43836

So is something like this even currently possible in blender?

  • 1
    $\begingroup$ Could you post a blend file or screenshots to help demonstrate your problem? $\endgroup$
    – JakeD
    Commented Oct 3, 2016 at 2:51
  • $\begingroup$ Sure thing, I've updated my first post with example animations and a blend file with examples of what I've tried on layers 1-3. $\endgroup$ Commented Oct 5, 2016 at 3:22

1 Answer 1


Yes this is possible

First create a floor cube large wide but not that high. Make it a Passive rigid body. (so it wont fall down its the floor)

Next a bit above it add a cube, subdivide it a few times. First add rigid body collision (convex hull, source deform). Next add softbody to this cube.

It will drop and woble on the floor.

oh and for better results use damping on soft body. such cubes can bounche on eachother as well if you copy them shift D

  • $\begingroup$ That almost works, but not quite. It acts like there's now two objects, an invisible rigid body and a visible soft body, and the two are separate. With Soft Body Goal on, the soft body acts attached to the rigid by a rubber band, which leads to some odd behavior. Turning Soft Body Goal off gets us back to the soft bodies ignoring each other again. $\endgroup$ Commented Oct 5, 2016 at 3:04
  • 1
    $\begingroup$ Looking at it i think you used subdiv modifier as well (apply it), also check your colision object type (use detailed mesh). and you might put in more calculation time (as when they colide) adding mass and damping more can help as well. .. Note goals is there for objects who naturaly spring back, ea sponge. $\endgroup$
    – Peter
    Commented Oct 6, 2016 at 13:20
  • $\begingroup$ Well that pretty well solves the "jello cubes don't like to interact" issue. Thanks! Now all I need is to figure out how to animate the scale of a soft body object during the simulation. Right now it only seems to respond to scale and property changes if it's not currently being simulated. $\endgroup$ Commented Oct 7, 2016 at 2:09
  • $\begingroup$ not exactly sure what your after, but pressing i (loc scale) creates a key on time line that locks place and current scale. move timeline, go to the objects property (location rotation scale) change scale and press again i in the scale property itself, as this time you wouldnt want to lock position (i asume). $\endgroup$
    – Peter
    Commented Oct 8, 2016 at 19:49
  • $\begingroup$ That's exactly the problem. Soft Body Simulations don't care about that. Give it a try, keyframe a soft body cube's scale. Hit play. It'll stay scaled at 1 unit through the simulation. I'm not sure how to describe it, but it seems like soft body sims basically rip the vertices off the object and animate them. Select the object during animation and you can see the origin of the object getting left behind. $\endgroup$ Commented Oct 9, 2016 at 6:17

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .