2021-11-11 Animation & Rigging Module Meeting

The main goal of this meeting will be a status update on & discussion of ongoing work.

The meeting will be open for everybody interested to join on Google Meet (link below).

People present are referred to by first name for brevity. Others are referred to by full name.

Present: Carlos Eduardo da Silva Barreto, Chinny Brynford-Jones, Christoph Lendenfeld, Colin Marmont, Daniel Salazar, Jason Schleifer, Kevin Burke, Luciano Muñoz, Maxime Casas, Sybren Stüvel

Links

Since the Last Meeting / Announcements

  • Sprite Fright has been released! Our last meeting was cancelled due to the theatrical pre-premiere of Sprite Fright.
  • Colin (@Kdaf on Blender Chat) introduces himself. He’s a highschool student, discovered Blender few years ago. Has done some bugfixes, interested in helping out with animation development.
  • Chinny (@Chinny on Blender Chat) introduces himself. He’s product manager for character animation in Unity, works in Montréal, focused on games. Started as demonstrator in Softimage 25 years ago. Blender is big part of the Unity pipeline, so he wants to become more familiar with the people & process.

Short-term goals

New short-term goal, which can be seen as a follow-up to the just-landed : new motion paths should not default to the 1-250 frame range, but to something more sensible for the current situation (like preview range, scene frame range, or last-used motion path range).

  • Colin will work on this by creating a design task. Once the details are fleshed out, he’ll also work on the implementation.

Demo by Maxime

Demo of a bulging reduction option for the armature modifier, by Maxime

  • Chinny & Jason react enthousiastically.
  • Chinny: does it make a difference when bones are really close together, or if the weighting is spread across multiple bones?
  • Maxime: using weights to find which vertex “belongs” to which bone(s); this is limited to two bones per vertex. This is an expensive step, could be precomputed in a way.
  • Chinny: how does this work with FBX exports? Is it going to export as regular deformation?
  • Maxime: That depends on the skinning method used by the software importing the FBX. To get the correct deform you’d have to export with Alembic or USD.
  • Chinny: Rigify has a third bone between the thigh and the shin bone, to correct bulging of the knee. If you use that set-up it conflicts with the bulge reduction, as it’ll try and correct for the wrong bone. This could be resolved by having explicit mappings (for example with other weight/vertex groups or face maps). This also applies to closely-spaced fingers.
  • Daniel: how expensive is it, performance wise?
  • Maxime: the performance impact is quite small, because it’s just computing two distances for each vertex.
  • Some discussion about the user interface, which leads to a discussion on the presentation of skinning methods in Blender.
  • Daniel suggest implementing Bone Glow
  • Jason: maybe a good idea to look into machine learning methods? That might even work better than Bone Glow when trained well.

Demo by Colin

Colin demonstrates his timeline-scrolling feature. In short, it allows scrolling of time-based editors such that the playhead stays inside the editor. There are multiple choices to affect the behaviour. All the module members agree it’s a good idea. He’ll make a design task to flesh out the user interface.

  • Christoph: there is a patch for playhead snapping options as well; maybe the two ideas could be integrated in one “playhead options” group.

Patch Review & Decision Time

We have 58 open patches in total. During the meeting we should visit a few, to at least accept or reject them on a conceptual/desirability level.

Patches introduced since the last meeting:

Older patches, so we can catch up with reviewing:

Other patches are left for future meetings.

Other Topics

Daniel presents a design issue with the weight painting. He wants some way (but is still not sure of the solution either, hence the discussion) to only show the weight paint colours when there are actually vertices using that colour. For example, when a vertex has 0.5 value, it’s shown in green. In his proposal, this green shouldn’t be shown when interpolating across a face.

  • Sybren: on a face with vertices of weight 0.0 and 1.0 on opposite sides, some points across the face do have this 0.5 value, even though there is not a vertex that has that value.
  • Daniel: that’s my point – I want to see the difference between “interpolated” and “vertex has this value”.
  • Daniel: maybe the current “Zero Weights” overlay can be expanded to have similar options for “Full Weight”?
  • Daniel will make design task so this can be fleshed out.

Next Meeting

The next meeting will be on Thursday 25 November, 18:00 CET/Amsterdam time. Again it will be open for everybody who’s interested. The provisionary meeting agenda will be linked in the #animation-module channel before the meeting.

7 Likes

Even if i’m new to animation, i can join this meeting? No restriction?

I believe so, yes. You can be present without participating even, and just listening.

Yup, that’s what I meant with “The meeting will be open for everybody interested” :slight_smile:

Do expect me to ask a few questions about yourself, just out of interest. Feel free to say you want to stay anonymous, if that’s the case, nothing’s mandatory.

1 Like