Motion sculpting in Blender: a different approach to animation, and my BCON26 talk

[notthatnda (Nacho de Andrés)]

Hi everyone, I’m Nacho de Andrés, also known as not that NDA. I’m an animator and Blender add-on developer, and I presented “Motion Sculpting: A New Animation Paradigm” at this year’s Blender Conference.

I’d like to open a discussion about the approach behind the presentation and hear how other animators would use it.

The idea is to make animation more tangible: see movement across time, interact with it in the viewport, and shape the motion while keeping the character’s performance in view. I’m interested in bringing the editing process closer to the things we judge when watching animation: poses, arcs, spacing, and the relationship between movements.

The full presentation is here:

Motion Sculpting: A New Animation Paradigm — BCON26

A lot of the design comes down to control. An animator might want to change one moment, reshape a section of movement, or make an adjustment that carries through a longer passage. The challenge is making the scope of that edit understandable, so you can judge both what you are changing and what you want to preserve.

There’s also more to demonstrate than I could fit into the talk. The workflow supports quadrupeds and other animal and creature types out of the box. I didn’t get to show those applications during the presentation, and I’d like to explore them in follow-up demonstrations.

I’m developing these ideas through Motionator. I’d be happy to discuss the implementation, current limitations, and the decisions behind the interaction.

Which animation task from your own work would you most want to see demonstrated with this approach? A concrete example would help me make the next demonstration useful.

For background, I also develop AniMate Pro, a full-featured Blender animation toolkit, alongside companion add-ons for rig control, transforms, and scene management:

AniMate | Blender Animation & Rigging Add-ons

[FinSalot]

This is some really awesome stuff! Genuinely some revolutionary work. My immediate thoughts are on how it’d interact with already existing rigs, say if I had a custom face rig or something similar- would I be able to swap between the two and animate with both?

[sozap]

Hey this is super cool Nacho !
I guess you are familiar with the ephemeral rig system and even though it’s something I’m following from a distance this looks quite related to your work !

I find super cool to try to think animation in a different way not by adding complexity but rather finding ways to make it much simpler and think more like a 2D animator would do…

It could be interesting to see how this would apply to a face, or how you would deal with stuff like picking an object on a table and giving it to someone else. If that’s not the case already I’m sure those can be figured out…

Maybe a final production rig would be a mix of that simple FK rig with more complex parts like the face that may not benefit that much from onion skin and motion path, but it’s already a great step forward IMO. I don’t think animators take the time to learn all the functionalities a production rig is having, so as riggers we add a lot of things thinking this is going to be useful but probably only a fraction of it is actually used…

Also I’m wondering how that would fit in a cartoony animation workflow where there is more squash and stretch, bendiness to the animation. Anyway since this is more the area ephemeral rigs are exploring and this seems similar to your approach I’m sure it’s not a limitation !

Last question, how much of that could make it’s way into vanilla blender ? Is there a way to support current way of animating but also allowing this workflow, or you have to decide to use that approach for your project and then half of current animation tools get irrelevant, or the other way around ?

Is making those tools available by default something you would be interested in or for now you prefer to explore those ideas by developing and selling addons ?

Because another problem we face with those new ways of thinking animation is to have them actually be used in production. It’s quite a bold move to switch all the animation tools of a project to something like that because every animator is used to the current way of working.
So even though the system is functional it could take years before it’s adopted. Having it integrated as part of blender, used in open movies or developed as part of blender lab, is probably a first step to make that “testing period” shorter and a bit reassuring for people to adopt that on some projects.

If that’s not already the case I would consider contacting Daniel Martinez Lara, getting in touch with Raf Anzovin is likely going to spark more ideas and probably also the animation module of blender.
Anyway, maybe it’s not your plan to have that in blender and that’s fine it’s already a good step forward !