Cineversity

Is it possible to have an object bend dynamically? I attached a quick keyframed animation I made to illustrate the result I’m trying to get.
I want both the sphere and the tall cube to be dynamic, so that I wouldn’t have to keyframe either of them. I don’t know if the cube would have to be a rigid body, or some kind of soft body, but is this even possible?
I tried rigging the cube with dynamic IK joints, but they will only interact with an object that’s set to be a collider body. I would need the sphere to be a rigid body.
Thanks!

Whoops. Didn’t realize mp4’s weren’t allowed. I’ll just zip it.

Well I can say it is possible, but not with the same level of control as the example you provided…especially the motion.
The motion in your example has a lot of character, this is most obvious when the sphere make contact and is able to hold onto the tip of the rectangle.

Now as far as doing it, I like the dynamic IK approach as you can still get some control over it, what would be the issue is having the sphere dynamic and a collider body? this should work, although will not provide any force on the sphere.

Otherwise you need to look into the softbody settings…you can use vertex maps to change the strength of the softbody effect, and also the mass…you would want to do this so that you can help the base maintain its shape.
One thing that could make setup easier would be to utilize the mesh-deformer so that dynamics on a low rez cube can be transferred to a higher res cube, saving on calculation / vertex weight painting.

I think in the end, if you only have a few, you could probably animate faster by hand than time spent trying to find a suitable dynamic solution.

Hey, thanks for the reply, Patrick.

I’ll play around with the vertex maps for the softbody. I agree that it would be faster to just animate by hand, as it didn’t take me very long to keyframe my example, but I had an idea to have a continuous stream of spheres going by from an emitter. This isn’t anything crucial really, just experimenting.

Hey Intaglio,

Please have a look to the attached scene. This is certainly the simplest way to create a project with many elements reacting to each other. Perhaps Patrick might share some ideas, as he went really deep into the Hair options.

I have used the “cube” here as instance in Hair. Note that the size of the “Instance Cube” is based on the hair material. This needs sometimes a fix, which means the transparency in the Hair-material needs to be set (otherwise the Hair renders as well. The cube is in need to be set as well for the renderings to “invisible”.

I have set the Gravity to zero in the Hair Parameters. Perhaps you like to explore this option with all the parameters in the Attribute manager.

Your example showed as well an affect to the sphere (to loose energy during impact), that might be a little bit more work. But perhaps this here is a starting point.

All the best

Sassi

Hey Sassi,

Thanks for the sample scene! This definitely looks promising. I’ll look into this technique as well.

Yeah, hair would actually be a pretty good option too. The hair series does cover using source geometry as each individual hair…you could also utilize hair to hair collisions…it is still approximated…but as the good Dr. always says…it’s what looks right, not what is right.

Thank you Patrick.

===========

I believe deeply in things like montage and editing. I believe in the language of camera/lenses and the expression of camera-moves. I could go on with this for a while, but I think I made my point. The animation by itself is as much believable as we create with all tools and options we have. [Synergy!] Digitally we have pretty much all these tools available now (compared to the analog time).

Having said that, in the moment we have an animation that comes very close to what we need, all filmic tools will support it, if executed well toward our target. If we take instead a non filmic approach, with a surveillance camera view perhaps (which can be filmic, but not without context), then the un-editied footage might suffer, if not pain-stackingly simulated to the smallest detail in 3D/4D.

We are (to my believe) visual artists here, and with our tools we (should) do magic. That’s all. :o)