2.80 almost ready what will Blender 2.81 bring?

In a few days from now blender 2.8 will be released.
I often use daily builds, so I wonder after 2.8

What will be the new roads for its development.
What kind of things we might see in blender 2.81
Things that are planned, or small wishes that are not out of reach
And dont depart to far from 2.8 goals, what are the next steps ?

There are some plans as seen in the GSOC anouncements, but often they are investigation and dont always end up in blender /cycles
I hope for new procedural textures, and cloth sim.

Though outside of there is also Godot integration (those guys currently develop all kind of things that could be done inside blender texturing etc).
Then there where talks about some new kind of fluid / smoke sims in the past.

Then there is everything nodes, and plans for particle code changes

And on graphicall.orgs there roulate builds that have vastly improved the speed of cycles, (there seams to be a new intel denoiser, and some more new math improvements for solving light path math.
( i think the past argue that cycles should not becomme to complex to
configure isnt from this time anymore (given it has an experimental
option button, if time can be reduced a lot, it be best for all of us)

However, roads and goals, ideas and changes not always make it to the grant plan, so what is likely going to happen in 2.81, and what would be the low hanging fruits for 2.82 (planned for near future but not next version).

I’m so used to the daily builds that I don’t know if I’ll start using the stable build or I’ll continue with the daily builds even after blender 2.80 …

obviously the dailybuilds after blender 2.80 official will initially be unstable for the addition of new features … while for blender 2.80 in recent months it has been largely a continuous bugfixing …

5 Likes

These are the things that will surely be developed internally …
As for the GsoCs it depends on the stability of them, but even these will be implemented as much as possible

Based on the feedback in the previous roadmap article, there’s enough to do for 100s of developers on projects the coming years. That means we will have to prioritise. What’s possible to start working on in Q2 is:

Overrides and asset management
Better support for large scenes or complex environments
Modifier nodes
Physics & real-time mode (for designing simulation and baking)
Cycles: reviewing/adding patches 

If the development fund grows beyond 30k:

Particles and hair nodes
Texturing tools and tools for procedural textures
Painting and Sculpting improvements
Better snapping and precision modeling
Cycles: denoising
Compositor
5 Likes

Well right bug fixing is a continuous process, releases are goals in which people define ideas of what should be in it and what not. After 2.8 they plan for a 3month release schedule i red.I supose “goal coding sprints” for next futures. Perhaps they let the team work on futures still to be done and add them per 3 months not sure how it will work out and … what will be added

UDIM :wink:


5 Likes

Reaction to the web link, some of those GSOC goals i think are closer on the horizon (nearly ready) then for example large scene support, and asset management (i think, never seen updates about it). But then that article was from december, and GSOC wasnt started yet. Oh well I’m curious about 2.81

I would expect uncompleted features in 2.80 to be quickly finished in 2.81.

  • Vertex color alpha painting and rendering
  • Adjust Last Operator panel that should not be moveable at an other place than stuck to left bottom corner.
  • Clean-up of UI in paint modes. Instead of repeating same basic settings in 4 different ways, dispatching of all of them in a pertinent way.
  • Baking system update. BakePasses patch from Lukas Stoeckner
    https://developer.blender.org/D3203
  • Improvements from Outliner branch making its behavior a lot more intuitive.

These things plus what was promised (UDIM and Static Oberrides) : that would be minimal targets to me.

They probably still have lots of ToDo’s for active tools and multi-object editing. That would not be surprising to see improvements and less limits about that in 2.81.
Most of dust put under carpet should be cleaned as quickly as possible.

Pablo Dobarro said what could be merged from his branch in his article in code.blender.org.
GSOC about procedural textures is also popular and goes well. I would not qualify that as experimental but as very useful.

Clement and Antonio will probably not abandon work on EEVEE and Grease Pencil.
So, there will be improvements and bugfixings compared to 2.81.
I would like to see a solution for Indirect Lighting for animation, for example.

I think there is no need to merge a big project in 2.81.
It would be great if 2.8x series could be release of a big project in even number and then after feedback, fixing and immediate improvement in odd numbered version that follows.

2 Likes

Mantaflow is a target for 2.81 and maybe also the fracture modifier. I hope the asset manager will be there with some presets for everything… that would be nice…

Undo performance improvments…

in my opinion, the general re-design of the ui is not yet complete … although I think it will be difficult to change it after blender 2.80 …

  • What still puzzles me is the space window of property panels that act as global panels, while in reality they are largely the functions connected to the 3d view space window … a unification of these two space windows that make blender more coherent logical I would like.

we discussed it here:

  • then I would like them to solve the problem of the overcrowding of the addons in the N sidebar on the right. (the current system of sorting addons is too cumbersome) I tried to reason here …
    https://devtalk.blender.org/t/customizable-filter-addon-combination-presets/7649/3?u=nokipaike

  • And finally I would like the design system of buttons, sliders, and panels to be improved in general … I can’t stand it when we enlarge the panels, they expand by taking up all the space instead of forming more columns.

I believe it’s been said that the asset manager is mostly complete, function-wise, it just needs a UI to be be designed and then it’s ready for use. Then there are Pablo’s sculpting tools that can hopefully be integrated by then.

On a side note it seams Brecht got different tasks this week which would give him more time for cycles (still my fav render engine). But not sure at what exaclty he will work on, though there are now various cycles variant builds, i hope they get some attention.

1 Like

Speaking of the rest of the 2.8x series, Nathan Letwory has been collecting feedback on what developers find lacking in the current plans and what can be done to keep people on board. Hopefully this means better and higher quality development going forward.
https://lists.blender.org/pipermail/bf-committers/2019-July/050017.html

Zeauro has already pointed out what could be good candidates for 2.81, but the devs. are going back to the 3 month cycle so there will only be 1-2 months to get an in-progress feature ready to go. We do have a larger core team now so “maybe” all of these will be ready to go by then.

6 Likes

That mailing list thread is extremely interesting, with lots of frank discussion from developers about where they think the 2.8 dev cycle might have gone wrong and how future development can be improved.

1 Like

Before any new features i would like to see serious improvement in viewport performance with heavy scene and hopefully they fix the time it take to save scenes!

Personally, I’d argue very loudly for INTEGRATED asset management. Im using an addon which is a bit better than OK, but not by any means great, and for example, only holds objects and materials, no particle systems or anything else. Its a good start, but I’d really like to see the integrated one

TIM

1 Like

Just out of curiosity, which add-on are you using for asset management?

they should make sure everything works perfectly before adding more complexity like what Pablo said, tackling the basic functionality that artists need is more important IMO and u can see the same points raised by the rest of the devs, so hopefully everything works out from 2.80 and on.

1 Like

Here is the Module owners todo parent task for all the modules, has some really nice plans in it I’m looking forward to.

Another thing I was looking forward to is the fast import/export gosc project, but it seems a bit quiet around those parts, anyone knows if that’s still going on?

2 Likes

Blender features I’m personally looking forward to the most are:

  1. Sculpt Mode improvements, particularly a perfected OpenVDB Remesh modifier implementation, putting it on par with ZBrush Dynamesh (but non-destructively).

And a really effective quad-poly auto-retopologizer like ZRemesher would be a true game-changer for Blender.

  1. Eevee and Cycles improvements:

I think Eevee requires too much hassle at the moment, having to fiddle with separate objects for GI and proper reflections, and adjust a heap of rendering, material, lighting and shadow settings. I’d love Eevee to become more straightforward, like Marmoset Toolbag.

Cycles should become more optimized for GPU, like E-Cycles, and preferably get a bidirectional path tracer for better caustics and GI in difficult areas of a scene.

  1. More procedural textures, such as a procedural scratches texture.

  2. A (short) fur shader, like an advanced velvet shader. This would allow you to skip the elaborate and rendertime-intensive particle-generated fur route.

  3. Everything Nodes, hopefully leading towards a Houdini-like non-destructiveness in modeling and other areas.

  4. I still miss some essential modifiers from 3ds Max, such as a UV mapping modifier that allows you to easily and non-destructively assign animatable and transformable UV projections (spherical, cylindrical, box, triplanar, planar, etc.) to one or more objects.

More examples of desirable modifiers:

  • Convert To Mesh modifier, so you can keep adjusting closed curve shapes or metaballs, non-destructively auto-convert them to a mesh, and add mesh modifiers on top.
  • Curve To Mesh modifier (so you can non-destructively auto-convert a curve to a mesh and add subsequent mesh modifiers).
  • Cap Holes modifier (would be great to auto-cap meshed curves), ideally including cap shapes, such as rounded.
  • Lattice modifier automatically creating a fitting and locally oriented lattice, also around a vertex selection in Edit Mode.

I would love as many modifiers as possible to have their own automatic influence cages and center pivots, in stead of having to separately create and assign a dummy object to a modifier.

  1. An object creation mode like 3ds Max Autogrid and ZBrush Insert Mesh, enabling you to easily create new objects on the surface normals of existing objects.

Now let’s hope that one or more of these wishes might reach Blender developers. :slightly_smiling_face: And you’re welcome to discuss.

7 Likes

Very nice to see that the experience of this past year has evolved the developers’ organization, I peeked at both links and read the various developer opinions, and I think I can say that if blender 2.80 was a revolution, what is coming it will be even more interesting, because I believe that they will be able to achieve better results in less time, precisely because now the phase begins when they will reap the fruits of the experience of this last year, avoiding the mistakes of the past and structuring their projects in the best …
Users have also entered into the mechanism of this development, and even users have self-instructed themselves to be more efficient with their feedback and suggestions …
From blender 2.81 onwards, it will be less tiring and there will be results far beyond imagination.

1 Like