Blender wish list

Hi everybody,

We are 8 gratuate students motivated in developing a Blender Add-on for a school project.
We have some ideas like a crowd generator or procedural like trees, cities, … But we don’t really know what would be appreciated by the Blender community, maybe something very different from our ideas.

Do you know somewhere we can get a list of “wish-list” from the community ? Or give us some people to contact ? Or maybe just give us some ideas ?
We have one year to develop this add-on, so it can be something quite ambitious, not only a small feature. We also have a strong background in mathematics and theoretical computer science, so mathematical things don’t frighten us at all !

Sorry if this is the wrong forum, but I didn’t found any other which matches with my question.
Thank you in advance !

1 Like

Hi! Welcome on BA. You want necessarly to do an addon (you could work on the source code to add an internal feature…)? For Blender internal? Cycles? Or maybe the Blender Game Engine (my preference)?

You talk about crowd simulation… That would be a cool idea for the BGE (But I’ve no idea on how to design it).

You could work on pathfinding in BGE (new and faster algorithms) and obstacle avoidance (RVO2 or such things)
You could work on particle systems for the BGE… Realtime fluid, smoke, fire simulations. (Panzergame (a developper) already begun to work on it (and on terrain generation too))
Realtime reflections.
Tweakble night and day cycle for the BGE.
Plugin to convert games in webGL (no idea if it is possible)
Android, Windows Phone, ios game exporter.

I come back if I have other ideas :slight_smile:

You should ask current Blender developpers for such an ambitious project. You can join #blendercoders and #bgecoders channels on IRC (freenode) to contact them. Don’t hesitate to ask several times if you’ve got no answer :smiley:

8 students working on a project for Blender during 1 year: Great!

New render for the bge?

100% modern shader pipeline, with a android compatibility mode?

also a android blender player so people can sell bge games on a blender game marketplace?

Logic nodes for the game engine instead of bricks?

or

Multi touch support?

Sensor Set Property

Ray sensor if true set property target----------and-------------Steer to target

Message set property target ---------and--------Steer to target

Mouse over set property targetPosition---------------and--------------add object at targetLocation
mouse click---------------------------------------------------/

this would ‘unlock’ the potential of the current bricks

Thank you two for your ideas ! :slight_smile:
You look both very interested in improving the game engine, to be honest we forgot this part of Blender, so we will consider it seriously now !

To answer youle, the project is not necessarily an addon, we could look and modify the internal source code, no problem with that.

Thank you again for your ideas, don’t hesitate if you have other ones !
I will also go to #blendercoders and #bgecoders to discuss with current Blender devs.

Hi,

pretty cool idea of you guys to develop an add-on.
You can make a poll in a thread to see what is desired.

There are some city and tree generators around, but they are not all free, would like to see one.

Greetings

I don’t agree totally with Tiles :smiley: . Except that “addon” is not necessarly the best “design” to integrate a feature in the BGE. Usership is not necessarly a variable to take into consideration. Neither the fact that the bge “could” disappear one day (many prophets predict its vanishing since long years but it still there. In 2015, many new bge developers obtained commit rights for the BGE, so I’m not certain that the bge will disappear tomorrow…). Neither the fact bge is less powerfull than Unity or Unreal (On the contrary, the to do List of interesting things to make for the BGE to catch up Unity or Unreal is an argument to choose the bge). In my opinion, the main motivation should be the pleasure to code a feature that looks good to you 8. Furthermore, I think bug fixing couldn’t be an interesting 1 year project for 8 students. Realtime stuff is very stimulating and rewarding (Realtime simulations). Of course, many other things are possible. You can work both for the viewport and for the bge (that is not exclusive). And you can surely find very interesting stuff for Blender internal or cycles too… It depends on your own preferences (but for coders - that have not necessarly artistic skills (like me (amateur coder)) - I wouldn’t exclude bge at once.
But I don’t want to fight with you Tiles :smiley:

Wish list:

  • Alembic reader/writer to send data(sims, animations etc) to and from maya, houdini, other 3d softwares. This would help Blender to be integrated in studio environments.

  • Sculpt deformation: https://www.youtube.com/watch?v=yuiB1Rv75O8

  • Perhaps modifier that allows CODE execution (e.g python modifier(py might be too slow)). This would empower users to create very advanced functionality similar to KL with Fabric or VEX/PY in Houdini.

  • Tension map generation: https://www.youtube.com/watch?v=JHEpAbuq-t0

  • PBR Viewport rendering (additional layer like in Modo (Advanced Viewport))?***

If you are ambitious enough you can try to implement node based parametric modelling in Blender. I mean something like the plugin Railclone for 3DS max - with that you can make parametric assets by combining and deforming hand modelled parts. You can start with coding modifier nodes which is probably a huge task but it is listed in the development targets (http://www.blender.org/press/18-anticipated-blender-development-projects-of-2015/) and so you can be sure that you would get a support from the blender foundation.
There is already an addon implementing parametric modelling called Sverchok but it is more like a Grasshopper in Rhino which is a bit different aproach to parametric modelling than the RailClone.

Hello, i would suggest to implement an Online and Lan server, like this:

allow multiple users to work on the same canvas simultaneously.

something like this software do:
http://drawpile.net/Thanks for your time.

That’s a very large project, I wouldn’t recommend it for first - time projects

Sent from my SM-G920F using Tapatalk

I think you are right, is easy to dream.
i dont know any 3d software that have this feature.
Besides usefull it would put blender in other level.

You’re looking for personal opinions?

Possibly one of the biggest improvements that could come to Blender would be having a better, more intuitive weight painting blur brush.

As of the moment it’s much more difficult to weight meshes with separated parts than it should be. The new smooth tool(well, I guess literally it might be a variation of the old Blur tool) is a great new addition, but it still doesn’t help with the issue of blurring weights in dense mesh that’s separated into parts, especially spatially separated(but close) parts.

In addition, the tool is full of caveats and difficult to find/understand options. If you don’t know what you’re doing with it it’s easily capable of destroying hours of work very quickly with no chance of recovery due to the fact that when you blur you’re often clicking repeatedly, which destroys your undo list since(in terms of mesh editing) it works like object mode undo and not edit mode undo.

EDIT: Below, when I say “Z” I mean forward-back relative to the viewport screen, not up-down.
I’ll try to be succinct:
The problems:

  1. The weight painting blur brush has unlimited visual Z range. If you turn off Limit Selection to Visible the brush will blur everything in the path of the viewport’s Z axis vector(intersecting with the brush’s size circle), regardless of brush size(brush size only affects viewport XY range). There is an option to limit the effect using normals, but it’s often not enough.
  2. Options pertinent to the weight blur brush are scattered throughout Blender. Limit Selection to Visible is probably the best example; it’s not intuitive how it affects the brushes or why. To my knowledge you have to use controlled trial and error to really understand what’s going on/all the caveats.
  3. There’s no option to change whether or not the blur brush respects the weight of hidden vertices or not. Currently, if you have your cursor over a hidden vertex and blur, the hidden vertex will not be affected but its weight will still be used in the calculation for the new weights of affected vertices. This effect has its uses so it shouldn’t be gotten rid of completely, but it should be an option; oftentimes it’s very inconvenient and it’s also not easy to tell it’s even happening.

Proposed solutions:

  1. Add functionality to the weight blur brush to limit the actual spatial range it’s capable of affecting; a 3d falloff like sculpting brushes have. It might be a good idea to keep the XY and Z ranges separated, though.
  2. Build stronger association between the options in Blender which significantly affect the weight paint brushes and clarify them, preferably both in terms of UI and tooltip.
  3. Provide functionality which is currently lacking(explained above). Example: Add some kind of functionality to determine whether the blur brush takes hidden vertices into account in the blur operation.
  4. In addition, having weight painting undo work like edit mode undo would be pretty nice. One reason the blur tool can be difficult or risky to use is that it eats up your undos, unless you’re somehow capable of doing good weight painting without clicking a lot(I’m not).

I could go on for hours about this but I’ll wait for a response for now. Actually, now that I think about it, most of these are problems which affect all weight painting brushes in Blender.

These are just my ideas, I’m no professional, but I can defend them and I don’t think they’re unreasonable. Not sure how difficult it is to tackle them, beyond that you’re pretty much going to need C, which is beyond me still.

I apologize if I’m just ignorant of solutions to these things or available options, but I have been using Blender nearly every day for 3-4 years and to my knowledge not only do none exist but there’s no real source of information on a lot of the functionality I’ve mentioned, but that could have changed since I did my testing to learn.

@cgstrive You can already do that with shapekeys in Blender :slight_smile:

As for the wish list, I’d say anything that help Blender to be better integrated in a pipeline (Alembic, FBX…)
Or anything related to the “nodification” of Blender would be nice.

Anyway, as others already said, talk to the developers; they can give you a better scope of what’s needed.