Check the [overview](https://devtalk.blender.org/t/sculpt-paint-texture-module-meeting-overview/33567) thread for more information about the meeting.

## Present
* Sean Kim
* Julien Kaspar
* Hans Goudey
* Jean-Silas Moor
* Dalai Felinto

## Announcements
N/A

### Commits

* [Fix #148835: Sculpt: Scrape/Fill Multires threading crash](https://projects.blender.org/blender/blender/commit/26ff5d53f93)
* [Fix #148837: Scene brush size incorrectly set](https://projects.blender.org/blender/blender/commit/5199cde4edd)
* [Template: Update Sculpting template](https://projects.blender.org/blender/blender/commit/d4167bfcd7f)

*For a full set of changes, see the main tracker. The above is a curated list of fixes and changes for the module.*

## Meeting Topics

### Development Updates
* High Severity Issues
* [Sculpt brush radius change preview doesn't match with new brush size when radius unit is set to scene](https://projects.blender.org/blender/blender/issues/148837)
* [sculpt mode—Scrape/Fill—crash exit](https://projects.blender.org/blender/blender/issues/148835)
* Roadmap
* Multires
* Farther off on [prototype](https://projects.blender.org/blender/blender/pulls/140792) than thought from last meeting. Updated the PR description with the current technical state. Making progress once again.
* Planning to do some cleanup now that the basis is changing less so Hans can review.
* 5.1 Targets
* [Sculpt: Add Geometry Nodes "texture" evaluation to brushes](https://projects.blender.org/blender/blender/pulls/133123)
* Currently behind an experimental option in the PR.
* [Sculpt: Scene Project Brush](https://projects.blender.org/blender/blender/pulls/136342)
* Planning on adding offset, what else needs to be done from an artist POV?
* Questioning the usefulness of the "relative" option as being built in even with the anchored brush preset
* **Conclusion:** Keep option, don't expose it for now, future work can handle making a mesh filter so that behavior works a bit more as expected

### General Discussion

For planning for 2026, this is the *module's* point of view of potential projects to work on in a **non-exhaustive** prioritized list. Additionally, this only concerns work that would be expected to take upwards of one month of dev time to complete or would otherwise require coordination between modules.

### 1. 3D Texture Paint
* Relevant Links: [PBVH image texture painting technical design](https://projects.blender.org/blender/blender/issues/96223)
* Estimate: 3 ~ 6 months, 2 devs?
* Status: Needs Milestones Defined, Started

Improve existing severe texture painting performance and reduce technical debt by replacing with the sculpt brush engine. Allows for a true 3D brush instead of purely projection painting. This is part of the larger work to introduce a "Paint Mode" and remove the distinction between Vertex Painting and Texture Painting.

* Possible risks:
* Current code uses a AoS structure, meaning that large parts of it will likely need to be rewritten.
* Image Datablocks have their own undo system
* Deprecation of existing Vertex Paint & Texture Paint modes, if done as part of the project, will likely have a large bug-fixing period following this project release

#### Discussion Notes
* Dalai suggests being reducing the scope of this down to the minimal feature parity with Texture Paint and not worry about introducing a new mode or other specific Paint Mode features to avoid having too many unknowns.
* **Conclusion**: Module is aligned on idea.

### 2. Sculpt Layers
* Relevant Links: [Sculpting layers design](https://projects.blender.org/blender/blender/issues/132533)
* Estimate: 6 ~ 12 months, 1 person
* Status: Needs Technical Design, Needs Milestones Defined, Not Started

Allow for a shape-key like system of blending displacements in a non-destructive manner while sculpting.

* Possible Risks:
* Impact to existing multires performance
* Needs involvement with Core module about key data management topics
* Needs involvement with UI module

#### Discussion Notes
* Estimation seems high, still lacking the minimal technical design and ideal end state.
* **Conclusion**: Sean and Hans to hold meeting to figure out technical details this week.

### 3. Brush Management
* Relevant Links: [Assets: Support editing/overriding essentials assets](https://projects.blender.org/blender/blender/issues/129053)
* Estimate: 1 ~ 3 months, 1 person
* Status: Started

There are two main topics after the brush asset changes that remain pain points for users:
* Inability to edit essential assets (commonly the smooth brush) without digging into overrides
* Inability to change & persist values on a per-project basis

Would benefit mesh painting, grease pencil, curves sculpting, and image painting.

* Possible Risks:
* Needs involvement with Asset System module for review and potential design questions

#### Discussion Notes
* No comments

### 4. Multires Attribute Support
* Relevant Links: N/A?
* Estimate: 3 ~ 6 months, 2 people
* Status: Needs Technical Design, Needs Milestones Defined, Not Started

The only attribute currently supported for multires on a per-high res vertex level is the sculpt mask. Notably, this means that color and UV attributes are not supported. May enable multires node tool support and general better integration of Multires with Geometry Nodes.

* Possible Risks:
* "One way door" with domain topic, design deserves extra scrutiny because of this.
* Possible Blockers:
* General Multires performance (time, memory) issues may need to be improved before extra support is added to avoid degrading current user experience.
* May make more sense to finish Paint Mode work prior to introducing yet more areas that would need to be migrated.

#### 5. Discussion Notes
* Technical idea here is still ambiguous.
* **Conclusion**: Sean and Hans to go over this topic as well this week.

### Brush Alpha Texture System Rework
* Relevant Links: [Sculpt: Texture System Refactor Design](https://projects.blender.org/blender/blender/issues/135752), [Improved Rake functionality](https://projects.blender.org/blender/blender/issues/98070)
* Estimate: 3 ~ 6 months, 1 person
* Needs Milestones Defined, Not Started

Better control and performance for fine detail sculpting.

* Possible Risks:
* Prior implementation had many small issues that were not obvious until users had the chance to test the code

#### Discussion
* Hans brings up that the two topics aren't closely tied together
* **Conclusion** Sean will revisit before finalizing document & isolate this to the rake texture work for now

### 6. Flexible Brush Keybind Mapping
* Relevant Links: [Paint: Brush Toggle Improvements](https://projects.blender.org/blender/blender/issues/147313)
* Estimate: 1 ~ 3 months, 1 person
* Needs User Design, Not Started

Currently, functionality such as smoothing or alternate brushes is hard coded on a per-brush basis. It would be nice to make this user-defined and more generally inform the user about these toggles. This has some potential crossover with tablet work and keymap work. Would benefit mesh painting, grease pencil, curves sculpting, and image painting.

* Possible Risks:
* Needs design

#### Discussion
* Julien brings up that this is hardcoded to not even the specific essential smooth asset for the Paint brush.
* **Conclusion**: Sean can look into this smaller topic outside of a project context

### 7. Geodesic Tools & Brushes
* Relevant Links: [Sculpt Mode: Improve Expand performance](https://projects.blender.org/blender/blender/issues/102586)
* Estimate: 3 ~ 6 months, 1 person
* Needs Literature Review & Prototype, Not Started

The current approach to determining the geodesic distance between any two points on a mesh's surface inside is a naive implementation. Implementing a faster algorithm would help the existing complex brushes & tools in Sculpt mode and may be extensible to Geometry Nodes.

* Possible Risks:
* There are a number of different research papers that can be used as a basis here. It is unclear if any of them will be sufficient for performance or official integration

#### Discussion
* No comments

### General Feedback
* Sean to refine list further after meeting with Hans this week before further review later in Q4.
* Add Multires ongoing work to the top, since that's still in progress and make issue for Multires work for it to be linked on the landing page.
* Hans to write up the Dyntopo related idea for a localized remeshing so that the module can have further discussions about a concrete idea.

## Other Topics
### Sculpt Mode Painting Color Space
* Related to bug report [#148926](https://projects.blender.org/blender/blender/issues/148926)
* Julien brings up that painting on a mask in Sculpt mode does not blend perceptually as one would expect.
* There is some difficlty here with vertex paint being used to paint non-color attributes
* **Conclusion**: Further work and research will happen in the PR.

-----

<details>
<summary>Detail Section Template</summary>
</details>

<details>
<summary>Devtalk Date Format</summary>
[date=2024-01-23 time=17:00:00 timezone="Europe/Amsterdam"]
</details>