Irrlicht applies for Google Summer of Code participation

[hybrid]

Maybe you already noticed our new Wiki page Ideas for GSoC

This means, we're going to apply for participation in this year's Google Summer of Code. Google Summer of Code is a yearly program which brings together students and OpenSource developers. The (university) students are offered development tasks (e.g. those on our ideas list) and get money ($4500) from Google if they can finish the task in time (3 month, starting end of May).

Unfortunately, only a limited number of students is accepted, you have to apply in about two weeks at the GSoC page. At that point it should be known if Irrlicht is chosen as a mentoring organization.

If you're eligible for this program and you want to code for Irrlicht start to discuss possible tasks here, or just apply. You can also suggest your own ideas, but these should be discussed before the application. We're looking forward to your application

[BlindSide]

I'd be willing to try this. There's a lot I could do, what do you recommend?

*After looking at the wiki page*:

Hmm the shadowing one seems suitable for me. If Irrlicht is chosen to participate I will take that one right away.

Although the meshbuffer optimization and SColor/IImage projects look good too.

I could also make a new particle system with batching, point sprites and a new 24 byte vertex format (No Normals).

[hybrid]

I'd really love to see you working on shaders. You're probably also allowed to do more than necessary, it's just that we need to fix some things to make the fulfillment of the task assessable Maybe you can also combine shadows and post-processing (since you have a framework for both at hand )

[Halifax]

I'm interested in a couple, specifically the DX10 driver and the hardware occlusion culling techniques. At any rate, it's a shame that they only accept a few developers because a lot of the things on the list look really great!

[hybrid]

let's hope the best. We could manage around 4 students, unless some more mentors will show up (i.e. some dev team members find enough free time )

[Nadro]

All topics on list looks very good (useful improvments). I can help in folowing topics:
1. Hardware occlusion culling
2. Postprocessing framework
3. Generalized shader access including Cg
4. Advanced shadowing techniques
I've done some elements of this points for my engine. I plan release my code, but I need add some improvments for it.

This is my suggestion for realize these points:
1. Hardware Occlusion Query support (based on depth comprision, needs precomputed sectors meshes - very low poly)
2. Screen Space Quad Node + effects: HDR, Bloom, Blur, Motion Blur, Depth of Field.
3. Adding support for techniques, passes and other features of *.fx / *.cgfx files
4. Parallel-Split Shadow Maps with some type of "blurring" eg. PCF+Intelligent Screen Space Blur or Variance.

[Halifax]

1. Hardware Occlusion Query support (based on depth comprision, needs precomputed sectors meshes - very low poly)

I just have to say that I strongly don't agree with that idea. Many people aren't going to want to do a pre-process step before loading their levels into the engine, and it sort of doesn't fit the Irrlicht mentality.

I've only looked into this slightly. But my idea would be doing a hybrid approach with OctTrees, which gives you the hierarchal data. Then you can use the new extensions for conditional rendering, to keep the checking in hardware, and implement the frame lagged technique from GPU Gems, given that you can generally count on the coherency between frames (which isn't always true in some circumstances).

[sudi]

I would love to work on irrlicht as well with this project.

PostPrecessing
Animation System (File Format)
Meshbuffers

These are the topics i can see my self working on. I already made aPostProcessing and "AnimationSystem" for my own projects, the meshbuffer stuff would be something different but also something i would like to pull of.

[MickeyKnox]

I'm also interrested in applying to gsoc. One topic i'm especially interest in is the
"OpenGL-ES 2.x driver", because i want irrlicht to be able to run on the Pandora (http://www.openpandora.org).
However, i must admit, that i don't have GLSL knowledge on my chest, but i'm willing to learn

Another topic i'm also interested in is "File format advances or add-ons", if my absence of
GLSL knowledge is not acceptable.

[Halan]

I would like to apply too. GLSL interests me a lot so I would like to write a Shader-Driven OpenGL 3.X / OpenGL ES 2.X video driver. As I would like have something for my own project too. I think this also would bring Irrlicht to the next level as soon Fixed-Function Pipelines becomes obsolete.

I am a active developer/tester for Ubuntu Linux (I for example helped to pack Irrlicht and will push it into the 9.10 series hopefully) and also develop my own Network Library ( http://code.google.com/p/open-network/ ). Also I've been using Irrlicht since I started with C++ (just did some C before) in around 2004.

greetings,
Halan

[hybrid]

We'll know if Irrlicht is elected on March 18th. After that day, 5 days are left to discuss possible topics with us, clarify the project content, etc. From March 23rd to April 3rd you can apply as a student at GoogleSummer of Code 2009 site.
But hey, we have to wait for a positive result on the 18th first

[3DModelerMan]

Can someone add high dynamic range lighting to the list?

[hybrid]

We did not make it into the accepted organizations this year. So sorry, no way to work on Irrlicht projects with Google funding

[Nadro]

We did not make it into the accepted organizations this year. So sorry, no way to work on Irrlicht projects with Google funding

With or without google funding I will help in improving Irrlicht Now we have got very good TODO list and We should add to engine this features from list.

[Halifax]

We did not make it into the accepted organizations this year. So sorry, no way to work on Irrlicht projects with Google funding

Just curious, but do they give you a reason why you aren't accepted?