koobas.hobune.stream
The Science Behind Unity
Hallo everyone,
I am looking very targeted for papers which served as close reference for the way Unity chose to implement certain algorithms. For instance I would be interested to read the paper(s) behind
Occlusion Culling (this one I am most interested in!)
Frustum Culling
Lightmapping (Single, Dual and Directional)
the Terrain module
DX11 Tesselation
Shadows
SSAO and Screen Space Ambient Obscurance
Deferred and Forward Lighting
Would be awesome if you guys links me to one or the other paper :)
Thanks a lot!
Sean
Unity Forums is a better place to ask this kind of question. Unity Answers is structured for answering single, specific technical questions.
I'm sorry my help on the forum wasn't what you needed. I'm not sure that algorithm implementation in Unity is something that will be discussed publicly. Also, for things like DX11 tessellation I imagine Unity just calls into DX11 code, so has nothing specific.
Graham, no need to apologize. I appreciated all replies even though they weren't what I was looking for. I'm sure exact implementation details are not publicly known, but I thought it is roughly known which publications served as reference for the implementation of certain techniques. that usually isn't a top secret thing. For instance I thought it was known on which algorithms Occlusion Culling (the Umbra 3 tech) is roughly based on. Is it e.g. some kind of hierarchical occlusion culling?
I don't know exactly how Occlusion Culling works, however I do know how some of the other stuff works:
-View Frustrum Culling works using the bounds of the mesh and the camera properties to deter$$anonymous$$e weather to render the mesh
-Lightmapping probably works using Radiosity and/or Photon Rendering, but I'm not sure on this one
-The Terrain $$anonymous$$odule is just a dynamic mesh with certain optimisations, nothing special
-Tesselation is done using Standard DX11 functionality
-Shadows depend on the rendering method, Differed (Not Deffered) or Forward. I suggest looking up how it's done in OpenGL (either method), there are a ton of resources online!
-Same goes for Screen Space Ambient Occlusion (not Obscurance). I remember doing some research on it and finding some papers (just google it), but at the time I didn't understand it too well ;)
