User:NathanVegdahl/WeeklyReports/2023

2023

December 11-15

PR and Issue review

• Re-reviewed and approved Anim: Bake Channel operator #111263.
• Reviewed and approved Anim: Don't use keying sets when inserting keyframes during autokeying #115522.
• Tested and approved Fix #114981 : Add box, circle and lasso selection tools to weight paint + bone select mode #115807.
• Reviewed Anim: Add falloff shape property to Weight Paint context menu #116163.
• Investigated Very strange constraint jiggling not present in blender 3.6.3 #115915.

Development

• Landed Cleanup: rename bArmature.collections -> bArmature.collections_legacy #115931.

* And subsequent fix of the resulting build break: Fix: update collada export code to use the new bone collections array.

• Worked on hierarchical bone collections with Sybren.

December 04-08

PR and Issue review

• Reviewed and approved Fix #115657: Bone's tail is still selected after invert selection #115663.
• Re-reviewed and approved Anim: Action bake custom properties #113208.
• Reviewed Anim: Bake Channel operator #111263.
• Reviewed and approved Fix #115330: Animation not updating when cancelling modal operators in the Graph Editor #115810.

Development

• More work on and landing of Anim: store BoneCollections in a flat array #115354.
• Pair programming with Sybren on hierarchical bone collections.

November 27 - December 01

PR and Issue review

• Reviewed and approved Anim: Add "Add/delete" track operations to NLA context menu #114202.
• Reviewed and approved Fix: FCurve colors on bones #115517.
• Reviewed and approved Anim: Change how Only Insert Needed works #115360.
• Reviewed Anim: Action bake custom properties #113208.
• Reviewed Fix #44834: Add bone selection icon next to face and vertex selection in weight paint mode #115409.

Development

• Polished Anim: store BoneCollections in a flat array #115354 and took it out of WIP status.
• Implemented and landed Fix #115434: custom bone colors invisible in stick mode #115477.
• Implemented and landed Fix: allow overshoot numerical input in the Pose Breakdowner operator #115534.

Animation 2025

• Started working on UX design for the MVP data model.

November 20-24

PR and Issue review

• Another review of and approval of Anim: View FCurve of Property in the Graph Editor #114407.
• Reviewed and approved Fix #114828: Don't rebuild outliner tree when frame scrubbing #114845.
• Reviewed and approved Anim: Insert keyframes without keying sets #113504.
• Reviewed and approved Anim: Add basic autokey tests #115064.
• Reviewed and approved Anim: Restrict range of fcurve_to_keylist #114854.
• Reviewed and approved Cleanup: internally renaming NLA Channels to NLA Tracks #115011.
• Reviewed and approved Fix: Inserting keys no longer sets the FCurve channel color #115297.
• Gave some minor feedback on WIP: Anim: Refactoring NLA menu into more sensible format #114805
• A bit of discussion on Anim: Align new bones with the world axes #113138.

Development

• More work on hierarchical bone collections: finished changing BoneCollections to be stored in an array rather than a listbase.

Misc

• Gave feedback on glTF I/O feedback on bpy and technical open points.

November 13-17

PR and Issue review

• Reviewed Anim: View FCurve of Property in the Graph Editor #114407.

Development

• Worked on hierarchical bone collections.

Misc

• Helped translate for visitors from some Japanese studios.
• Then got sick again, and was out for the rest of the week.

November 06-10

PR and Issue review

• Reviewed and approved Anim: Unit tests for inserting keyframes #114465.
• Reviewed and approved Fix #114588: Graph Editor increment snap not working #114607

Development

• Continued work on hierarchical Bone Collections. Changed approaches after discussion with Sybren.

Animation 2025

• Discussed layer mixing behavior with Sybren. No hard conclusions yet.

Misc

• Filed issue Regression: snapping in graph editor with ctrl hotkey no longer functions properly #114588 on behalf of Hjalti.
• Joined meeting with Jacques where he gave an overview of gizmo support in Geometry Nodes.

October 30 - November 03

PR and Issue review

• One more review pass and tentative approval of Drivers: implement fallback values for RNA path based variables. #110135.
• Reviewed and approved Fix: NLA mapping not taken into account when framing FCurves #114325.

Development

• Implemented a fix for Regression: Crash on deleting NLA strip in tweak mode #109430 in #114331.

October 23-27

• Helped prepare the animation presentation for Blender Conference 2023.
• Blender Conference 2023.

October 16-20

PR and Issue review

• Additional review on and approval of Fix #113552: NLA Markers do not reset when length reaches 0.0 #110121.
• Reviewed Drivers: implement fallback values for RNA path based variables. #110135.

Development

• Started implementing hierarchical Bone Collections.

Animation 2025

• Made a detailed schedule for phase 1 of Animation 2025.
• Worked with Sybren on the Animation 2025 talk for BConf 2023.

Misc

• Made comparison images of Filmic and AgX for the Blender 4.0 release notes.

October 09-13

PR and Issue review

• Looked into more color model stuff for Bone Colors with Demeter.
• Reviewed Fix #113552: NLA Markers do not reset when length reaches 0.0 #110121

Development

• Hooked up the new Animation datablock to the dependency graph, to trigger animation evaluation.

Animation 2025

• Finished the draft schedule.
• Meeting with Sybren and Fiona about the schedule.
• Meeting with the studio animators a keying workflows and other misc. topics.
• Started learning Blender's depsgraph, and figuring out how the new Animation system "MVP" should get hooked up there.

Misc

• Started writing up the findings about key framing workflows.

October 02-06

PR and Issue review

• Read through and gave feedback on Proposal: Unify Active/Selected Bone Colors and set current Blender Theme Colors as Default #112943.
• Suggested some tweaks to the approach proposed in Proposal: New Bone Color Presets #112635.

Development

• Fixed AgX has banding-like issues on hyper saturated gradients #112018 in #113253.
• Translated an implementation of OkHSL from C to Python for Proposal: New Bone Color Presets #112635

Animation 2025

• Started putting together a draft development schedule for Animation 2025.
• Meeting with animators about selection sets, bones pickers, and keying sets.
• Meeting with Sybren to figure out how we can reliably dedicate more time to Animation 2025 work.

Misc

• Opened a devtalk thread for the issue brought up in WIP: Improved overlapped bone selection with a single click. #112654

September 25-29

PR and Issue review

• Did some investigation on Butterworth Filter is causing issues on geometry #112850.
• Gave feedback on Improved overlapped bone selection with a single click. #112654.

Development

• Finished fix for action.frame_range span always >=1 when there's only 1 key frame in it. #107030 in #112709.
• Added documentation to the Blender manual for the new B-Bone vertex mapping option.

Animation 2025

• Read through and gave feedback on Proposal: Object Parenting UX #112758
• Meeting with Blender Studio animators about bone colors.
• Meeting with Falk and Sybren about Grease Pencil integration in the new animation system.
• Meeting with Fiona and Sybren about Animation 2025 scheduling.

September 18-22

PR and Issue review

• Continued review of and approved Bendy Bones: implement a new curve-aware vertex to segment mapping mode. #110758.
• Reviewed and approved Anim: Graph Editor Push/Pull operator #112388.
• Reviewed and approved Fix #97628: Clear and Keep Transformation not working when keyed #112670.
• Reviewed Anim: Bake Channel operator #111263.

Development

• Finished up and landed Animation: Graph Editor Handle Selection #111143.
• Started working on a fix for action.frame_range span always >=1 when there's only 1 key frame in it. #107030 in #112709.

September 11-15

PR and Issue review

• Reviewed and approved Animation: Time Offset Slider #110540.
• Re-reviewed and approved Refactor: Rename functions and structs in keyframes_draw.cc #111510.
• Gave opinion/feedback on "Whole Character" keying set not working properly with autokey enabled #91941
• Reviewed and approved Fix: Graph Editor first key handles not drawn #112365
• Started reviewing Bendy Bones: implement a new curve-aware vertex to segment mapping mode. #110758

Development

• Landed Fix #111780: bone collections and undo don't work together #111965

Animation 2025

• Had a UI/UX powow with Sybren about the animation editor, resulting in some rough draft designs.
• Met with the studio animator about those rough draft designs.
• Met with Sybren and Brad to share and compare the different draft UI/UX approaches we've come up with for the animation editor.

Misc

• Took some time to improve my development environment, particularly around quickly switching between different build configurations.
• New desk!
• Helped interview potential new hire.

September 04-08

PR and Issue review

• Reviewed and approved Armature: apply new DualQuat scale handling to constraints and crazyspace. #111759.
• Reviewed and approved Fix: Animation Editor snapping not working when using rotate or scale #112088.
• Reviewed and approved Animation: Rename "Bake Curve" to "Keys to Samples" #111049.
• Reviewed and approved Animation: Scale Average slider for Graph Editor #111744.

Development

• Fix #111780: bone collections and undo don't work together #111965

Animation 2025

• Met with Hjalti to talk about the new animation system design.

August 28 - September 01

PR and Issue review

• Merged improvements from Alexander into the Bone Collections PR.
• Landed the Bone Collections PR.
• Reviewed and merged Animation: remove UI code relating to bone layers #111730.
• Reviewed and merged Fix: Bone collection renaming when empty string #111723.
• Reviewed Refactor: Rename functions and structs in keyframes_draw.cc #111510.
• Reviewed and approved Animation: Shear operator for Graph Editor.
• Reviewed and approved Fix: Blend to Ease keyframes property name #111745.
• Reviewed and merged Bone collection: Set first collection active #111733.

Development

• Misc improvements and bug fixes for the Bone Collections PR before landing.
• Misc fixes after the Bone Collections PR landed, including Fix: failing keymap test for ARMATURE_OT_assign_to_collection

Animation 2025

• Bone Collections are in! Group effort along with Christoph and Alexander.

Misc

• Attended meeting with and presentation from Alexander Wilkie about spectral rendering.

August 21-25

PR and Issue review

• Re-reviewed and approved Animation: Graph Editor - Don't draw keys if they are too close to each other #110788.
• Second functionality review of Bendy Bones: implement a new curve-aware vertex to segment mapping mode. #110758.
• Reviewed Animation: Add channel type options to the Bake Action operator #110903.
• Reviewed and approved Fix: Memory leak in Graph editor keyframe jump #111464.
• Reviewed Replace Bone Layers+Groups with Bone Collections #109976
• Re-reviewed and approved Fix #100718: NLA Hold Forward Inconsistency #109182, pending confirmation from Sybren to land.
• Re-reviewed and approved #111099 - AgX-Step5: Add AgX and its components.

Development

• Sybren walked me through his Bone Collections PR (Replace Bone Layers+Groups with Bone Collections #109976), in preparation for me hopefully landing it while he's away on vacation.
• Implemented custom property support for Bone Collections as an additional commit in Replace Bone Layers+Groups with Bone Collections #109976.

Animation 2025

• Finished writing up the rationale document for why we decided on unidirectional pointers for connecting animation and IDs.
• Met with Sybren and Brad about Brad's ongoing work on the Animation 2025 design document.
• Met with Sybren and Dalai about object nodes.

August 14-18

PR and Issue review

• Approved Animation: Match Slope slider #110567.
• Re-reviewed and approved Refactor: resolution_scale in graph_draw.cc #111037.
• Tested, re-reviewed, and approved LibOverride: keep some anim data from liboverride when it exists in both IDs. #110900.
• Tested, approved, and merged Fix #109081: Ensure swap strips do not overlap each other.
• Reviewed Animation: Graph Editor - Don't draw keys if they are too close to each other #110788.
• Re-reviewed and approved USD: Skeleton and blend shape import #110912.
• Created example file for new b-bone deform mode, demonstrating the discontinuity issue: Bendy Bones: implement a new curve-aware vertex to segment mapping mode. #110758
• Reviewed and approved Fix: Number input on slider #111269.
• Reviewed #111099 - AgX-Step5: Add AgX and its components.

Development

• Continued Animation: Graph Editor Handle Selection #108142 in #111143, with code updated to latest main.

Animation 2025

• Planned initial MVP for the animation system with Sybren.
• Meeting with Pablico about the proposed animation data model, to get feedback and ideas. Very positive response, and some good insights about workflows we need to make sure to accommodate.

Misc

• Put together a test file for Amelie, for checking if her work on grease pencil layers in the dopesheet unexpectedly impact other things.

August 07-11

PR and Issue review

• Further review of Animation: Move Snapping to Scene #109015
• Reported back about the two possible fixes for Swap Strips can delete strips if there is not enough space #109081. The animation & rigging module decided to go with #109980.
• Suggested a stop-gap fix for Child object rotates after parenting with 'Keep Transform Without Inverse' #109790. The longer-term fix would be something along the lines of what's being discussed in the devtalk thread about moving transforms to use doubles generally.
• Did an initial functionality-level review of Bendy Bones: implement a new curve-aware vertex to segment mapping mode #110758.
• Reviewed USD: Skeleton and blend shape import #110912.
• Did a cursory review of LibOverride: keep some anim data from liboverride when it exists in both IDs. #110900.
• Opened Armature layers can be put into an invalid state via the Python API #111043 after some initial investigation of the addon-repo issue #104535.
• Reviewed Refactor: Rename variable in graph_draw.cc #111037

Development

• Landed Fix #110169: nearly flat normalized fcurves jump around in editor.

Animation 2025

• Moved the animation-level constraints document to devtalk, since that's better for long-form discussion.
• Read through Brad's WIP animation system design document.
• Meeting with Brad about the design document so far.
• Finished the (super) basic animation-level-constraints prototype, built on top of the data model prototype by Christoph and Sybren.

July 31 - August 04

PR and Issue review

• Reported back about the module meeting discussion on Drivers: implement default fallback values for RNA path based variables. #110135.
• Reviewed Animation: Blend to Infinity slider #110567.
• Reviewed Animation: Blend To Ease Slider #110566.
• Reviewed Animation: Blend Offset Slider #110544.
• Looked at NLA: Swap strips fixes #110074. The author has two PRs open to fix the same bug in two different ways. Asked for clarification on the pros/cons of the two fixes.
• Reviewed Animation: Move Snapping to Scene #109015
• Reviewed and approved Fix #110161: Crash on dragging a Speaker NLA strip #110605
• Approved Animation: Graph Editor curve drawing performance improvement #110301
• Reviewed Animation: Common curve drawing for FCurves #110764

Development

• Confirmed Broken FCurve rendering #110169 as a bug, and submitted a PR with a fix in #110796.

Animation 2025

• Started work on an "animation-level constraints" Python prototype, built on top of the existing data model Python prototype from Christoph and Sybren. It immediately became clear that there are some likely prerequisites to a system like this working, and we'll need to consider the trade-offs involved with that.

Misc

• Continuing discussion about possibly moving transform matrices to doubles: https://devtalk.blender.org/t/moving-transforms-and-matrices-to-double-precision/30329

July 24-28

PR and Issue review

• Reviewed Animation: Graph Editor curve drawing performance improvement #110301
• Reviewed Anim: Armature drawing code refactor, introduce UnifiedBonePtr, other small improvements #110424

Development

• Fixed an overlapping IK constraint to zero influence/disabled gives weird behavior #88752 with Brecht's help in #110417.
• Tracked down the cause of Segfault pasting stale driver. #88482. Will work on a fix.

Animation 2025

• A little more brainstorming about animation-level constraints, control rig, and simulation layers.
• Wrote up a summary of the brainstorming I've been doing about animation-level constraints: https://hackmd.io/LwmMoFhqSRiKk3RWgIkhGA There's at least one promising direction, which would also potentially address other things in a unified way (switching control rigs, animation remapping, etc.). But it remains to be seen if it would actually make sense / be feasible.

July 17-21

PR and Issue review

• Summarized the discussion from the 9-year-old issue Graph Editor: (Auto-)Focusing Channels #43933, so people can more easily get up-to-speed with it, and so that in turn we can move forward with addressing it.

Development

• Started preparing a PR for Setting an overlapping IK constraint to zero influence/disabled gives weird behavior #88752. Will submit it first thing next week.

Animation 2025

• Finished editing and published the Animation Workshop Lessons blog post.
• More brainstorming about how control rig, animation-level constraints, and simulation layers could integrate into the new animation system design.

Misc

• Based on a recent issue, started a topic on devtalk about possibly moving transforms in Blender to use double-precision floats.

July 10-14

PR and Issue review

• Investigated Child object rotates after parenting with 'Keep Transform Without Inverse' #109790. It's due to floating point rounding error, so not exactly a bug. But it may be worth discussing this in a larger context, regarding whether it's worth taking a bit of a performance hit for higher precision transforms or not.
• Opened F-curve ghosting displays wrong values outside the initial view when ghosting was enabled #109920

Development

• Started investigating what the appropriate fix for Setting an overlapping IK constraint to zero influence/disabled gives weird behavior #88752 is, after Brecht identified the root cause.
• Fixed F-curve ghosting displays wrong values outside the initial view when ghosting was enabled #109920 in #109926

Animation 2025

• Began exploring ways for control rig, animation-level constraints, and simulation strips/layers to fit into the proposed animation data model.
• Gave some feedback on Sybren's draft blog post about the Animation Workshop.
• More work on my blog post about lessons from the workshop.

July 03-07

PR and Issue review

• Looked into Bake actions operator converting F-curves to linear #109528. No bug, the new behavior is intentional. Users can always set the keys back to bezier if they want, but linear is a more sensible default for bakes.
  • Having said that, in the process of looking into this and speaking with Brad Clark about it, it became clear that there are some related issues:
    • The fact that baking always bakes to linear curves now isn't documented. Not a bug, just a documentation thing.
    • The "Overwrite Current Action" option for baking leaves existing keys in place on a baked fcurve if they aren't overwritten by a baked key. This isn't a bug, but seems like undesirable behavior.
    • F-Curve ghosting seems buggy in certain ways. For example, it doesn't work properly for parts of the f-curve that extend before/after the view frame when turned on, among other things.

Development

• I was spinning my wheels too much on Setting an overlapping IK constraint to zero influence/disabled gives weird behavior #88752, so I documented the seemingly relevant parts of my investigation so far and poked Brecht about it. He managed to identify the issue, but there seem to be trade-offs in how to resolve it.

Animation 2025

• Wrote a "Workshop Lessons" blog post. It's about ways we think we can make workshops more effective in the future, based on our experiences at the animation workshop last week. The post isn't published yet, but is more-or-less "done", pending review.
• I did some thinking/investigation into pointer direction for hooking up IDs with Animation blocks in the new data model. We basically have three broad options (although there are sub-variants of each): Animation -> ID, Animation <- ID, and "strict" bidirectional Animation <-> ID. All-in-all, I think we just don't have enough information to make a good decision yet about which to go with. Specifically, I think it's going to depend at least partly on how control rig, animation-level constraints, procedural animation, simulation layers, etc. fit into this. And we haven't figured that out yet. Fortunately, most of the design decisions in the data model aren't fundamentally impacted by this, so we should be able to move forward regardless, just with the knowledge that we may need to change this aspect of it in the future.
• Talked to Sergey about how the dependency graph might need to be impacted for some of the things we want in the new animation system (e.g. control rig, animation-level constraints, animation metastrips, etc.). Main take-aways: control rig (perhaps surprisingly) might actually be the lowest-impact of these things, and things like animation-level constraints and metastrips might be the harder things to tackle. Still needs more investigation, however, and it will of course depend on how we actually approach these features.

June 26-30

Animation 2025

• We had the animation workshop!
  • Together with Sybren, Brad, Christoph, Demeter, Brecht, and a bunch of other people who popped in at various points, we have gotten *much* closer to a final data model for the new animation system.

June 19-23

PR and Issue review

• Reviewed NLA Hold Forward Inconsistency #109182

Development

• More work on tracking down and fixing Zero influence IK gives wired behavior when other IK moved the bone #88752. Eliminated a bunch of things that it's not, but still haven't figured out what it is.

Animation 2025

• Had a meeting with Midge Sinnaeve and Sybren about motion graphics and "weird" uses of Blender.
• Explored possibilities for how constraint/space/rig switching could work with the new data model proposal.
• Finals bits of workshop planning with Sybren for next week.

June 12-16

Development

• Worked on tracking down and fixing Zero influence IK gives wired behavior when other IK moved the bone #88752. Narrowed it down a little, but over-all no luck yet.

Animation 2025

• Meeting with Brad about animation data model requirements, pain points, etc.
• Meeting with Sybren and Jason Schleifer about animation layers and NLA workflows.
• Meeting with Ian Hubert about using motion capture in indie VFX pipelines.
• Impromptu meeting with a visitor about their animation workflow: they made a 7-minute animation in about 1 day.
• Put together a second data model draft proposal (high level).
• Put together a presentation for that new draft proposal.
• Meeting with Brad and Sybren about the new draft data model proposal.

Color

• More work on the tweakable tone mapper. Experimented with a different approach for desaturating colors as they blow out.
• A bunch of time tweaking how the tone mapper gets exported to OCIO. Lots of annoying fiddly stuff to avoid artifacts that aren't in the tone mapping operator itself, but rather are due to the lossy nature of trying to represent that operator in OCIO (OCIO is really limited in what it can represent analytically, so things are forced into LUTs, which is... fiddly).

June 05-09

PR and Issue review

• Further reviews and merged Armature modifier - Preserve volume + Scale #108134. The new updated solution from the author is really nice, and addresses the large majority of issues with DualQuat + scaling.
• Confirmed #87313 as still an issue, and marked as bug. Also provided more details on proper reproduction.
• Confirmed #88482 as still an issue, and marked as bug.
• Closed #87487 because it's the intended behavior.
• Reviewed #104673
• Closed #88689 as a current limitation of the dependency graph, not a bug.
• Confirmed #88752 as still a bug.

Development

• Addressed PR comments and merged #108548 (fixes #106801)

Animation 2025

• Met with Brad and Sybren about requirements for the data model.
• Met with Falk, Amelie, and Sybren about Grease Pencil and how it and Animation 2025 fit together.
• Met with Sybren about an alternative direction for the data model that he came up with. Really promising.
• Met with Dalai about dynamic overrides and how they might impact the direction we take Animation 2025's data model.
• Worked on thinking up and writing up use cases and user stories that will hopefully help us flesh out the requirements for the data model in a more grounded way.
• Looked into breaking changes we might want to make to Armatures for Blender 4.0 (e.g. bone layers -> bone collections).

May 29-June 02

PR and Issue review

• Looked into "Clean Keyframes" in Dope Sheet only works if channels are selected #87160. Definitely feels like something needs to change (either naming or behavior).
• Confirmed Blur brush(Weight Paint): Take edge distance into consideration when blurring weights #87238 as still an issue. Not a bug, but an undesirable behavior. Put on my todo list.

Development

• Fixed #106801 in PR #108548. Pending review (particularly needed because it's an unfamiliar area of Blender's code for me).

Animation 2025

• Worked with Brad Clark on brainstorming UX options for "where do keys go" and "what data are you working with" in an NLA-like system. This led to some promising new ideas for the animation data model.

May 22-26

(A very meetings-heavy week.)

PR and Issue review

• Confirmed Inclusive bounds for VertexWeightEdit? #82831 as still behaving the same way. Brought it up with the module. Seems like it makes sense to change the behavior as suggested.
• Reviewed Fix Armature modifier - Preserve volume + Scale #108134. Doesn't fully resolve the issue, but maybe still worth merging as a stop gap. To discuss with the animation & rigging module.

Development

• Added better code documentation for the Sobol sampler(s), describing best practices for using the different sample dimensions.

Animation 2025

• Meeting with other Blender Institute developers to discuss the proposed animation data model. Lots of good feedback. The data model definitely isn't done yet!
• Meetings with Brad and Sybren about the animation data model and how it relates to animation system UX.

Color Management

• Several meetings with the visiting Gimp, Inkscape, and Wayland guys about color management and how we can make all our software play well together in that respect. No concrete outcomes yet, but we definitely have general alignment on how things should be approached.
• Reviewed Wayland's WIP color management protocol. Basically no complaints. They've done an awesome job.

May 15-19

PR and Issue review

• Narrowed down #106801 to a minimal reproduction case.

Development

• More work poking at heat diffusion weighting failures.
• Worked on tracking down #106801, to actually fix it. No luck yet.

Animation 2025

• Continued work on the performance prototype, adding a basic model of ID Blocks and the ability to query the animation system for a specific ID Block.

May 8-12

PR and Issue review

• Proposed a design for #103802, based on the idea of simply respecting handle selection in the graph editor.
• Confirmed #76549 as still an issue, and marked as bug.
• Closed #85350 as not a bug, but with notes about why it is indeed confusing.
• Identified #86371 as already fixed in main, but still affecting the current LTS releases (which were referenced in the issue). Left a comment asking if anyone needs a backport to one or both of the LTS releases. Original reporter says they don't need a backport. Also discussed with the animation module and decided to not backport, because the backport would likely be involved, and the bug is easily worked around.
• Confirmed #84609 as still an issue, and provided a partial explanation of what's happening and how (for now) users can work around it.
• Reviewed and approved #107869.
• Reviewed and approved #107173 for Blender 4.0 (since it's a breaking change for the Python API).

Development

• Investigated a quick fix for #84609 (automatic weights failure), and although it fixed the reported issue, it also caused other problems. So it's not a viable fix. Following that, figured out (experimentally) a "fix" that allows automatic weights to succeed over-all, leaving only the problematic vertices unweighted. Certainly a better behavior, but I want to better understand what's going on and hopefully figure out a better fix.

Animation 2025

• Worked on animation data performance prototype. Some promising results, but I'm not confident it models current Blender internals well enough yet.
• Talked to Falk David about whether the proposed animation data model would work/conflict with the new Grease Pencil design. Conclusion: they'll work quite well together.
• Talked briefly with Rik about some of the ideas for Animation 2025. Definitely want to talk more with the animators! Great feedback and skepticism, which will help push the design to be really strong and useful.

May 1-5

PR and Issue review

• Confirmed #106410 and marked as bug.
• Confirmed #106542 and marked as bug.
• Closed #106980 as not a bug.
• Looked into #103802. It's not a bug, but probably needs discussion as to what the desired behavior is.
• Finished #107223 and got it merged.
• Identified #106340 as a duplicate (already fixed).

Animation 2025

• Finished the draft animation data proposal, and gave a presentation at the animation module meeting.
• Started a "performance prototype" of the proposed animation data system. One of the unknowns of the proposed system is how fast we can make it. So this is just a quick-and-dirty outside-of-Blender skeleton to get a broad-strokes idea of what the performance might look like. This is explicitly not to be the code of the actual system (e.g. it isn't even in the right programming language).

April 24-28

PR and Issue review

• Final review on #107089.
• Reviewed #107281.
• Communicated the tentative consensus on #107030.
• Identified #107286 as a duplicate.
• Reviewed #107173.

Development

• Testing and more work on #107223.

Animation 2025

• Created an outline of a concrete animation system foundations proposal. Still need to write/draw it up properly to share with the module.

April 17-21

• Started work on a draft proposal for Animation 2025's foundations.
• Got blender building and debuggable on my workstation.
• Hosted the animation module meeting for the first time!
• Found and fixed a crash in Blender! #107086
• Fixed #95855 "Graph editor - when keys on two different curves overlap, and one curve is active, trying to select the key on the active curve sometimes results in the key on the non-active curve being selected." in #107223 (pending review).