Page 1 of 1

[GSoC 2016] Idea Discussion

Posted: Thu Feb 18, 2016 12:57 am
by spacegaier
GSoC2016Logo.jpg
GSoC Project Ideas
We are hoping to once again find a group of capable and motivated students that help us move Ogre even further. In order to guide this effort, we are in the process of compiling a list of potential projects from our development team point of view that we will share soon. However, we also want to poll ideas from the community.

If you have any potential ideas for additions / changes to Ogre (especially. 2.1 since that is the long-term future), please let us know here. If you are a student who wants to tackle a project, you at some point should create your own forum thread dedicated to your proposal and application (once we have been accepted as a mentoring organization for 2016).

UPDATE: Unfortunately, we were not accepted this year to the Google Summer of Code 2016.

Re: [GSoC 2016] Idea Discussion

Posted: Thu Feb 18, 2016 1:02 am
by dark_sylinc
Ideas from 2015 that still hold:
  • Port DualQuaternion Skinning to v2
  • Port Pose animations to v2
  • OpenGEX Importer
  • Level of Detail (LOD) tool
New ideas:
  • Port the GLES 3 renderer
  • Port the Metal renderer
  • 2D drawing interfaces. Right now we have Rectangle2D (which is not suitable for rendering large amounts of sprites) and Overlay (which has a lot of design issues with Ogre 2.1 and is quite hard to use / overengineered).
  • Remotery integration.
  • Support point light shadows (dual paraboloid mapping)
  • Support multiple cubemaps for transitioning between areas with our PBS
  • Add area light type
  • Sample showing how to access individual MSAA samples and perform explicit resolves. Make sure it works on all RenderSystems
Vulkan & D3D12 are out of the question unless the student can prove he is very talented. It is quite a significant endeavor.

Re: [GSoC 2016] Idea Discussion

Posted: Thu Feb 18, 2016 1:12 am
by Wolfmanfx
Ideas:

Re: [GSoC 2016] Idea Discussion

Posted: Thu Feb 18, 2016 11:33 am
by PhilipLB
Might be combined to have enough work for a full GSoC:
  • Make the Volume Component compatible with 1.10, 2.0 and 2.1.
  • Adding new volume sources is always good
  • Fix some small issues like always scaling 3D texture volume sources with a factor which was needed to display Acropora ( http://www.voxelogic.com/ ) exports. This factor needs to be configurable
  • Find and implement an alternative to Acropora.
I could mentor that.

Re: [GSoC 2016] Idea Discussion

Posted: Thu Feb 18, 2016 5:17 pm
by Transporter
  • Rewrite resource management of Ogre 2.1 to use integrated threading instead of boost/TBB
  • Scene editor for Ogre 2.1 (OpenGEX/terrain component)

Re: [GSoC 2016] Idea Discussion

Posted: Fri Feb 19, 2016 12:53 am
by Mako_energy
Since Transporter mentioned resource management, could we get the resource decoupling from GSoC 2013 finalized?
Other than that I like OpenGEX support and anything that gets 2.1 running on MacOSX.

Re: [GSoC 2016] Idea Discussion

Posted: Fri Feb 19, 2016 9:15 am
by syedhs
In the past, Ogre GSOC participation has been limited to graphics rendering functionalities.. how about tool? Will Ogre change its stance so that it will also include tool in GSOC? Examples are blender model support or world editor - revive Ogitor or just build a simple world editor, but easily extendable by lua script for an instance. The lack of such tools are seriously reducing Ogre value especially to newcomers.

Re: [GSoC 2016] Idea Discussion

Posted: Fri Feb 19, 2016 10:22 am
by al2950
All of the below are for 2.1 in my mind :D
  • Reflection system - Multiple cubemaps, reflection probes & SSR
  • Forward+ rendering & Tiled deferred rendering support
  • +1 for area lights
  • Volumetric lighting HLMS
  • +1 Remotery integration
  • Texture system re-write
  • +1 resource system
  • Tutorial suite for 2.1?
  • Ray traced shadows
  • Global illumination technique(s)
  • +1 mobile support in 2.1

Re: [GSoC 2016] Idea Discussion

Posted: Fri Feb 19, 2016 5:22 pm
by spacegaier
An initial "official" list has been created (which is also linking for additional inspiration to this thread). The dev team will extended it ove time as additional ideas garner traction here in this discussion:

http://ogre3d.org/tikiwiki/tiki-index.p ... ef_id=2189

Re: [GSoC 2016] Idea Discussion

Posted: Sat Feb 20, 2016 1:31 pm
by frostbyte
imho, establishing a strong ART pipeline( which is ogre's weak spot ) is a good customer attraction point
"official scene editor" is not a good idea- first you need a sceneFormat and anyway you can't realy top ogitor within time limits( took two years to write... )
Dergo, openGex support etc is a much better direction
since focus moved to ogre2.1 and since openGex is easy...
i would like to suggest adding/combining openGex support with MDL/OSL support or something similier( i don't realy know what's the trend )
http://www.mdlhandbook.com
https://github.com/imageworks/OpenShadingLanguage

offcourse i would also love to see all other suggestions materialize... optimizations/new renderers etc...any idea of how to prioretize?

Re: [GSoC 2016] Idea Discussion

Posted: Sat Feb 20, 2016 4:44 pm
by spookyboo
I do believe that a good art pipeline is needed, but I don't believe in a do-it-all-engine (anymore). Having building blocks with which you can build your own pipleline is the way to go. An separate openGex module is more useful than a tool that has it integrated. The trick is to identify all the building blocks and to glue them all together in your own tools (without jumping through hoops to make it possible).

Re: [GSoC 2016] Idea Discussion

Posted: Sun Feb 21, 2016 7:54 pm
by xrgo
al2950 wrote:All of the below are for 2.1 in my mind :D
  • Reflection system - Multiple cubemaps, reflection probes & SSR
  • Forward+ rendering & Tiled deferred rendering support
  • +1 for area lights
  • Volumetric lighting HLMS
  • +1 Remotery integration
  • Texture system re-write
  • +1 resource system
  • Tutorial suite for 2.1?
  • Ray traced shadows
  • Global illumination technique(s)
  • +1 mobile support in 2.1
this is exactly my wishlist! +1 to every item... plus every VR optimization in existence

Re: [GSoC 2016] Idea Discussion

Posted: Sun Feb 28, 2016 7:25 pm
by moejac
Implement order-independent transparency in Ogre, using A Phenomenological Scattering Model
for Order-Independent Transparency
by Morgan McGuire et al.

Re: [GSoC 2016] Idea Discussion

Posted: Tue Mar 01, 2016 1:24 am
by xrgo
tessellation with displacement maps!

Re: [GSoC 2016] Idea Discussion

Posted: Tue Mar 01, 2016 8:23 pm
by dark_sylinc
To avoid wasting potential candidate's time, we were informed yesterday that: We were not accepted this year to the Google Summer of Code 2016.

Re: [GSoC 2016] Idea Discussion

Posted: Tue Mar 01, 2016 8:30 pm
by al2950
dark_sylinc wrote:To avoid wasting potential candidate's time, we were informed yesterday that We were not accepted this year to the Google Summer of Code 2016
:( booooo! Do we ever get any feedback, or is it a simple yes/no?

Re: [GSoC 2016] Idea Discussion

Posted: Tue Mar 01, 2016 8:36 pm
by dark_sylinc
A simple yes or no, followed by "we have many applicants and limited slots" and "Hope you try again next time".
There used to be a scheduled IIRC session with feedback. Perhaps they do it again this year, but we haven't heard from them yet.

Re: [GSoC 2016] Idea Discussion

Posted: Tue Mar 01, 2016 10:08 pm
by hydexon
Probably Google thought OGRE is almost dying of inactivity.

Re: [GSoC 2016] Idea Discussion

Posted: Tue Mar 01, 2016 11:24 pm
by spacegaier
Given that no rendering engine got in this year at all, could also mean there is a switch in focus of Google.

Re: [GSoC 2016] Idea Discussion

Posted: Wed Mar 02, 2016 5:51 pm
by Thyrion
let's do our own summer of code! How much money do we need to collect!? :mrgreen:

Re: [GSoC 2016] Idea Discussion

Posted: Wed Mar 02, 2016 6:43 pm
by moejac
I second that idea :)

Re: [GSoC 2016] Idea Discussion

Posted: Wed Mar 02, 2016 6:44 pm
by moejac
Ogre Summer of Code - it could be an act to follow.

Re: [GSoC 2016] Idea Discussion

Posted: Wed Mar 02, 2016 6:56 pm
by Wolfmanfx
Its 3500USD per student :)