Is the Sculpt Vertex Paint in 2.92 release? I neither see any vertex paint tools in the sculpt mode nor the sculpt vertex colors in the mesh data tab.
Edit: (needs to be set explicitly)
bpy.context.preferences.experimental.use_sculpt_vertex_colors = True
Some Experimental features are deactivated when Blender enters the beta stage. If you install 3.0 alpha Sculpt Vertex Colors are present in the Experimental features.
Whoa… 3.0 already? 
Yeah, there will be no more 2.9x versions of Blender. 
I think that’s exciting, as the Blender devs will undoubtedly unleash some major new features upon us.
They could have done that by continuing to cipher releases 2.8x.
Sincerely, I feel like 2.93 will be a 2.86 and 3.0 will be like a 2.87.5.
But what people are expecting since start of 2.8 hype is a 2.89b.
I think that UX workshop will allow developers to sort a useful list of priorities among all their todos. But I doubt that they will solve all problems of consistency with a prolongation of only few weeks.
Whatever you marketing strategy is ; time to write necessary code, review it and clean it, is irreducible, if you already have correct methodology.
People will probably have valid reasons to complain about a lack of consistency in UX until Blender 3.3.
If you have a developer expanding Sculpt Mode by adding several dozens of tools, almost an hundred. UI should allow user to build a custom set of these tools available in toolbar, easily.
If you allow to flip toolbar/sidebar, header/footer regions, Adjust Last Operator panel should be moveable.
If you allow to move origins with a gizmo and to drag 3D Cursor, it should be possible to move and rotate 3D Cursor using gizmos.
If you allow to create cubic shapes that are not cubes using primitive, dimensions in adjust panel should be 3 values of array instead of a unique radius.
If you create an Extrude Manifold tool, you should allow this tool to delete geometry at the end of movement.
…
2.8x and 2.9x series solved a big part of 2.5x,2.6x,2.7x technical debt. But not all of it. And in many new fields, they increased it.
About sculpting, retopo, paint 3D coat features are lot better than Blender and it’s addons.
For example on painting you have layers, drawing shapes, bake andcreate curvature and AO maps, custom procedural materials ,curves drawing and many more.
Each have it’s best features, but 3Dcoat is all in one, sculpting, retopo , paint without having to buy addons and without needing other external software. That make it very valuable and faster production for many things outside of modeling.
I hope they don’t. I detest hype and unrealistic expectations. 3.0 vs 2.93 should be no bigger than 2.93 vs 2.90. I would love nothing better than for them to slow down and actually address long standing bugs and UI annoyances that still linger from 2.79 and fix the performance regressions and multires glitches and add decent brush management. All the “new” stuff too fast for marketing purposes around “3.0” is just adding more technical depth to an already massive pile of technical debt. It would not surprise me if 4 years from now Blender has a reputation for bloat, bugs, and half-baked features on par with Maya and Max even though it’s still actually far less bloated and buggy than either.
The roadmap for 2021 at least seems to not be quite as heavy with big marquee features and instead is more in the way of focusing on workflow (ie. resolving shortfalls in areas like animation and I/O as well as continuing the projects from last year).
That said, layers for texture paint is sorely needed, I don’t understand how the BF literally seems to be the only software developer in existence that does not get why it is useful (as even the Material Maker guy was able to code texture painting and add layers in six months, in his spare time).
Probably because Blender devs expects that users use GIMP/Krita/Photoshop to paint textures. I mean, if none of the artists from open movies pushed for this feature (since devs pays more attention from them, forums are far too noisy) probably is because of this.
Some decisions taken by devs really baffles people. But that’s Blender for ya, i guess. (BTW, i do agree, layers for texture ARE NEEDED)
As someone who loves 3dcoat dearly, and has been using it for (way) over 10 years: I’ve replaced it for retopo and unwrapping with Blender. With some addons, I think Blender achieves the same thing faster, even if part of that speed comes down to not having to go to another program.
For sculpting I use Zbrush, but 3dcoat undeniably has some really interesting tools there. I’d lament the confused sculpting UI, but… well I just compared it to Zbrush, and that’s not exactly a positive UX casestudy.
Painting though? While I use Painter for most things PBR, 3DCoat is the unbeaten champion.
But, one thing to mention re: it being the all-in-one: with the price of 3dcoat, you can buy a lot of addons for Blender. It’s fairly pricey, even if it’s warranted for how much it does.
I’m sure 3.0 will not only bring new features but also improve existing ones.
Seriously, you could paint textures in 3D View in Blender, several years before creation of software like Mari or Substance Painter.
They don’t expect users to paint textures in a 2D software.
They just bet on wrong horse. They released experimental branches about Ptex.
They thought that no UVs at all was the future. But nobody wants to be alone in the future.
Blender users who are making games or modest movie productions, never were interested in heavy Ptex layers that cannot be converted into an image format, easily exportable in any other software.
So, they worked on UDIM support with a delay compared to other software.
They worked on support of PSD format, during 2.7x series.
Blender can open and display them but not edit them.
Substance painter and armorpaint software were becoming popular during 2.7x series, just before 2.8 refactor.
They probably had in mind, for a second, that they could drop texture painting like they thought about it for VSE. And finally, they only abandoned BGE after seeing advancements UpBGE team had made.
2.8 early design neglected some parts of textures workflow but not really painting part.
No support of Cycles/EEVEE textures by modifiers and Freestyle.
No improvement thought for baking. No custom set of brushes management.
The weird idea to generate automated previews of brush based on settings and a categorization by tools like in photoshop.
But nobody can deny that EEVEE and ability to immediately see result of painting in viewport on material rendering is an improvement.
The support of complicated mapping nodetree is far better in 2.8. Viewport now supports any type of mapping where 2.79 was limited.
And during 2.8, Pablo came up with the idea to make painting on vertices : the main way to paint assets. With the promise of baking result as an image ; that could be edited in Texture Paint mode.
So, basically, we are still in a comparable situation as in 2.6x series when Nicholas Bishop was doing experiment with Ptex. Waiting for a baking solution + new Attribute Paint mode and Texture Paint mode improvements.
Everything that we already had is slightly better after 2.8,2.9.
But what we were missing is still absent.
It’s a mystery to me why texture paint (and UV editing) receives so little attention. One would think the texturing workflow would be a consistent PITA for virtually every Open Movie production. Yet, even in the rare case when devs focus on it, it’s like a quick hack of something basic and they jump back over to working on something with a much more limited use-case instead. At this point, even the video editor gets more attention lol.
Texture paint must not be a fun area to code or something.
I just want single pixel/floating point paintbrush size. It already had a patch and was slated for a release.
They just let it die on the vine for the sake of future perfection that has now delayed it for years. I do understand that technical fixes now can have consequences for the future but my users would be pissed if I told them I couldn’t do something now because I wanted to do something perfect, in the future, if I have time, maybe.
Yes, you can paint in Blender, the problem is that the paint level, is “mostly” at “Windows Paint” software equivalent nowadays (Not trying to discredit the artists, you can still do wonders with it, but “time is money” as they say). You can workaround the lack of several advanced painting features with nodes and other ways users have come during years, (and thanks to the reprojection feature you can do more) but in the end, the paint module of Blender is showing its age.
Now to be honest, external devs have tried in the past to come around this limitation, but the lack of interest of Blender Devs, and ultimately, the complexity of an apparent “easy” to tackle problem (also this have to go through the UI team, so expect more delays) made people abandon the effort. (the last one i think was KWD in 2020 or it was 2015??? i believe Fabio Russo tried to improve in 2012 i think??). Maybe it’s time to kick that hornet nest again??
And BTW, i’m only talking about Image Layers here… Paint improvements are another can of worms 
(Yes Zeauro, i’m at the age my brain has holes :P)
I have all but abandoned my favorite past time of painting in Blender because, well, they change the api and break all my own addons but don’t give anymore attention to paint though they stated they would. I get things done if I need to do them, but haven’t really been feeling the love lately. I can’t keep up with all the changes in sculpt anyway, so really it might be my own fault here and not something to blame elsewhere - I just had hoped that paint would get easier because they said they would create a new brush engine and give us something like more mainstream.
I still love Blender, just feeling a little like the kid that saw his siblings getting the candy while left empty handed.
You would think that since Blender uses OpenEXR under the hood for all images in process before save/export that Blender could actually use layered exr for image layers inside blender, or even better, adopt .ora or .xcf or whatever open standard they can easily adopt.
I know people want psd support, but that might be possible to handle if they can at least get image layers going like Gimp and Krita do.
PSD support would be a nightmare (even tho Adobe published an unfinished PSD spec, and for what i’ve heard, is full or errors and omissions)… but is what most people consider a “standard” nowadays… ora would be a better alternative. But yea… even then, people would still asking for psd…
I personally dislike the need to use .psd as those OSS that do give that option can’t include all of the fancy layer styles and layer groups I think. I like Krita for painting, but wish to use Blender only because there is so much more possible inside Blender.
Gpencil, Sculpting, Compositing, Object modifiers, Procedural textures and UV mapping, basic modeling of anything to include in a a painting as a brush… but latest versions don’t like me as much, and little things bug me liek how I can’t select handles on a curve in pairs at the same time to adjust the curve shape like I used to in earlier versions. I’m old and getting crotchety.
I had to buy one upgrade only very long time ago.
The price is really fair, and all in one means you don’t need to switch software.
About blender addons, the main issue is when the guy does no more maintain some addon, it’s a waste of money.
Whatever, waiting for Blender 3 release with new sculpting brushes like Insert Mesh, also the new geometry nodes.