2 years ago developers planned to make procedural texturing (with layer system), everything was fine, people were discussing how this system should look like.
A year passed, almost no progress, developers said that this project is still a goal.
Another year passed, not a rumor about this project…
I think this project was canceled in favor of other projects for blender, like projects to improve animations, etc.
Probably, what’s your point?
It’s still planned but yeah they started with improvements in the brush system…
You should always take announcements like these with a grain of salt, it’s what developers wish to work on, but it’s never that simple.
In a commercial project this would be announced quite later, once they are sure to deliver. But with blender we learn about these once it’s just a meeting note, and a lot of things can get delayed or even fail to be delivered, but on this particular task it’s not abandoned yet…
Indeed, the article was posted in February 2022.
Texturing is one of the strategic targets for 2022. The first part of the project involves upgrading 3D texture painting performance, quality and tools, and is planned to be worked on in the coming months.
The second part of the project is a new procedural and layered system to create textures, aimed at PBR shading. That part of the project will start later this year, and we are sharing the initial design now for feedback from the community.
The first part mentioned was partially released in June 2022, in 3.2, with
3D painting improvements, limited to Color Attributes.
In January 2023, targets were announced.
And in the “there’s more” section of secondary targets, it was mentioned.
The procedural texturing project – while not having a concrete roadmap and resources yet – is still a goal.
Then, the module became orphan, in 2023. Unique developer in charge of Sculpt & Paint module was fired.
But, in other areas, most of primary targets for 2023 were postponed to settle new basics of 4.x series.
In Blender 4.0, principled shader was not the only shading node modified.
Noise and Voronoi textures were changed.
So, for 2024, 4 primary targets mentioned were already targets for 2023. (Extensions Platform, GPU-based Realtime Compositor, EEVEE Next, Brush Assets).
Vulkan passed from being a primary target to become a secondary target.
And procedural texture system is still a secondary target.
Medium and long-term list
- Blender Apps (unknown yet, we lack resources for it).
- Layered texturing (with a bit of luck it starts 2nd half 2024).
- Move to Vulkan (might be 2nd half 2024, depends on GPU Projects).
Things did not progress as well as expected in 2022 and 2023. But projects were not canceled. They were just delayed.
We can expect to see work done to obtain 3D painting for texture, when Brush Assets project will be finalized.
We can expect work on Layered Texturing to start, when EEVEE next basis will be consolidated.
You should take whatever Blender Foundation say about their plans with a grain of salt. Those are only plans, not a hard, contractual obligation.
That blog post isn’t a promise. Back then Blender treated Blog as basically feedback thread. They used to post designs they had, they would create feedback thread and see what people thought. But because people, like you, took it as promise, got excited and thought it was planned project they announced this year that they’re changing communication format.
Since 2023 blog posts are not made about design feedbacks and they’re only posted on devtalk forum. And only approved and scheduled projects make it to the user-communication channels.
Layed Texture design, not only was not planned for immediate future, but was also just not good. It didnt answer all the questions, it didnt fit into node editing workflow well. It proposed using entire different node editor. It was just trying to come up with solutions, but failed.
So design isn’t there yet. In 2024 official goals are finishing existing projects - Dyntopo and 3D painting brush and performance improvements (which is honestly bigger than layered texturing). From 2025 we can start thinking about layers again, but I wouldn’t put my money on it.
Welcome to my waiting for the updated and improved Hair/Cloth simulation system. As far as things currently stand, I won’t be surprised if the current simulation tools are still the only option in 2-3 years time.
As for texturing, save yourself the pain/wait and just get Substance Painter via Steam or 3DCoat Textura. With either of those tools, it pretty much doesn’t matter what is or isn’t added to Blender for texturing/painting.
Blender is notorious for delaying or even forgetting its projects.
It’s very bad right now, Blender knows that too, which is why the roadmap for 2024 is projects has to be completed.
Layered texturing is still to be started in 2024, so you have to wait.
I find it more amusing that the menu itself has 21 default hotkey assignments.
I am sure if they got rid of it someone would complain that they actually use it.
(works for sculpt brushes too)
You could also say this for Microsoft or Google… …and finally everyone is free to pay for a better suited product… buy then… blender is not a product… nobody has to pay for it… ( and i also do not understand some people claiming that blender is not usable at all… )
But still… in the 3D area… what other versatile program would you suggest ?? ( which also is not a one man show… so that someone may suffer of it’s possible discontinuation )
BTW, does this thing work for material textures or only brush textures?
That was working for everything, related to Blender Internal render engine, in 2.79.
So, it was meant to work with BI world, BI materials, Brushes, Freestyle Line Styles, and compositing.
With abandon and removal of BI render engine, in 2.80, that never had been useful for Cycles and EEVEE.
And that stopped to work for Freestyle.
Freestyle now use EEVEE/Cycles texture nodes.
That never worked for old modifiers.
GN modifiers can use EEVEE/Cycles texture nodes.
That only works for brushes and compositing.
In 2.79, we could reuse same procedural texture, almost everywhere. That was a BI texture datablock.
Now, a procedural texture sharing depends on type of nodetree.
You can create a nodegroup, that can be shared between shading nodes (materials, world, line styles).
But you cannot create a nodegroup, that can be shared between shading nodes and geometry nodes. Although nodegroup is made of same texture nodes, present in shading nodes or geometry nodes.
You are forced to copy/paste content of nodegroup, instead of copying/pasting nodegroup node, to create another nodegroup, that will not be synchronized.
If you want to keep synchronization, you have to convert a GN nodegroup into an attribute or use multiple drivers.
And brushes, compositing nodes are using different old BI textures.
I have only ever used it for brushes, (and not very often).
It does not work with material textures.
I can get the texture into the compositor (add texture) but it does not seem to do anything (I could be doing it wrong)
As zeauro says it is a relic from the past,
one day it will be depreciated totally, maybe when they do the new work on brushes 
That’s unfortunate. Even if redundant with material nodes, could be good quality of life feature. Or even include texture-specific operations such as blur
for every time something more or less pans out as expected (original eevee, grease pencil, cycles x, geo nodes), there are just as many things that don’t come to fruition, get completely rethought as something else, or become much longer of a project than anticipated (animation 2020, particle nodes, eevee next).
i do think there is something to be said about developers maybe presenting things as an area of exploration rather than a roadmap.
in particular i think the new animation system should be thought of that way. with all the new ideas proposed about how to handle something as complex as animation, i’d wager a lot of what sounds good on paper during brainstorms will shake out a little differently in practice (just my gut).
and procedural texturing probably falls in the same category.
I’m really surprised procedural texturing hasn’t made greater strides over the years.
I started my 3d modelling and texturing journey using Imagine 2.0 on the Amiga, way back in 1992 (actually I started with VU-3D on the ZX Spectrum - but that didn’t really have texturing).
Imagine 2.0 already had a decent set of procedural textures, and when Imagine 3.0 came along and expanded the set (especially with an addon procedural texture pack - which I believe is based on the IFW texture library).
Things don’t really seem to have moved on all that much - which is a shame. I much prefer designing materials using procedural textures.
There was a flurry of activity a couple of years ago around using SDFs to expand the procedural texture toolset - and was showing promise and there were experimental builds and everything:
But it appears to have died a death. It’s a shame because this SDF style approach, building textures form simpler components and transforms really fits well with the whole node based/geonodes paradigm.
Another feature that procedural textures are missing is loops. Secrop created a loop node for cycles using python a few years back and geonodes has (I believe) just had one added. This is a very powerful feature and it would be nice to have out of the box.
Some of the new functionality in the Voronoi procedural texture (e.g. fractal Voronoi), could have actually been achieved by adding a texture loop node instead - with the advantage that it would have worked across any of Cycles procedural textures, not just Voronoi. I actually did experiments along these lines in the thread above - with each subsequent layer of texture having a increasingly smaller scale.
Do you happen to have your work in Imagine you can show?
