Google announces this year's GSoC projects for Blender (10 projects)!

The list includes…

  • Sebbas - Mantaflow Integration: Single Code Fluid Solver (Using Mantaflow for fluid simulations)
  • DingTo - Cycles: Texture system improvements and better user feedback (improved VRAM usage and user feedback)
  • Inlinebool - Sculpting: toolset upgrade (new brushes and improved performance)
  • Lukas Stockner - Cycles: Implementing a Denoising Filter in Cycles (Integrated Denoiser)
  • João Araújo - Curve Editing: Improvements to Bezier Curves (curve editing features)
  • Severin - Scene Management: Layer Manager (ability to name, order and group layers)
  • tianwei37 - Tracking: Multi-view Reconstruction with Witness Camera for Blender (Witness cameras to strengthen estimation result)
  • Nathan Vollmer - Paint Modes: Performance enhancements with P-BVH and new/better tools for vertex & weight painting)
  • Peter Cassetta - Extensions: Package Manager Proposal (package manager to help users update and install addons)
  • SaphireS - UV editing: New and better tools

In short, there’s something that most everyone will benefit from, and there’s a good chance several of these will make it into master when considering that most of the larger projects went to those who are veterans with the Blender code and the rest have what appears to be more focused and smaller projects than in years past.

Obviously, not everyone’s wishlist can be fulfilled this summer, but there’s still some pretty good ones :slight_smile:

Well, at least my wishlist got fulfilled! :slight_smile:

I think I might be the only Cycles user more excited for the texture storage improvements than the de-noiser. If DingTo’s project goes according to plan, it means the end of the 5-float-texture limit, and texture-heavy scenes will use a fraction of the memory they do now.

Also, anyone got a link to the UV tools proposal? All I could find were (promising!) mailing list discussions.

First post amended, comment irrelevant

Best of luck to the students involved!

I hope some of these make it into master - ALL of them sound very useful and needed, but I suppose the reality is that the students’ work may not be entirely finished or ready for master when the GSoC ends. Personally I’m pretty excited about improved UV tools and layer management. I don’t really do any rendering, but the changes to cycles also sound like they will benefit a lot of users. Seriously though, each project sounds great.

@J_the_Ninja >

Good luck for all students!

Cheers, mib

All good projects this year :slight_smile:

Yeah, really good projects this year.

Good luck to the students, and have fun!

Good luck all! Looks great.

Yeah Layer Manager - I am very curious about how this one work will !!!

Am I the only one who thinks the outliner should be used as a template for the layer manager? I would even let it have it´s own windowtype, instead of being in the toolbar like now (the layer manager addon in trunk)…well, let´s see what Severin comes up with :slight_smile:

Since he is in the UI team, he will for sure coordinate everything with the others in the team to have a good long term plan that is consistent with other upcoming developments.

I hope the mantaflow implementation will be improved to one that doesnt require the hacky python export.
i have high hopes for this project because Sebbas has already spend quite some time on this project, and he doesnt have to actually implent the fluid solver. ‘Only’ implement mantaflow to blender in a stable way that is still easy enough to update (as mantaflow development goes on in the future).
i hope the mantaflow devs will accept code changes that will make implentation into blender easier.

just saw that the mentor on this project is N_T. Is this Nils Thuerey? (One of the authors on mantaflow, and overall genius)

Texture system improvements project - is it all about image textures - or does it also include procedurals i.e. expanding the capabilities of existing textures or adding new ones?

no Assets Management of course, Blender is alergic to that

Asset Management is already being worked on. Blender Coders Blog; Phabricator Design Task. Even if it weren’t being worked on, it is way too big of a project for a student to do over a summer.

It´s already being worked on by Bastien Montagne, so how exactly is Blender allergic to that?

Unfortunately the last I heard, Sebbas and Nils decided on continuing down the python path.

If the mantaflow implementation will be python based, it makes it chances of being included into master allot smaller :(. Lets hope for the best :slight_smile:
good luck to all gsoc students this year!