Editing User:Whatever/GSoC15/proposal
From BRL-CAD
User account "Whatever" is not registered. Please check if you want to create/edit this page.
Warning: You are not logged in. Your IP address will be publicly visible if you make any edits. If you log in or create an account, your edits will be attributed to your username, along with other benefits.
The edit can be undone.
Please check the comparison below to verify that this is what you want to do, and then save the changes below to finish undoing the edit.
Latest revision | Your text | ||
Line 1: | Line 1: | ||
+ | Hello. I am interested in ray-tracing, and GPGPU. I plan to work on OpenCL acceleration of the BRL-CAD librt rendering pipeline. | ||
+ | =Preliminary Schedule= | ||
+ | * 1 week - ellipsoid (ell), arb8 (arbitrary polyhedron) shot routines in OCL | ||
+ | * 1 week - to refactor dispatcher, shoot, optical renderer to process many rays in parallel in C when rendering an image or block | ||
+ | * 3 weeks - grid spatial partitioning in OCL | ||
+ | * 2 weeks - to code for GPU side database storage of OCL implemented primitives | ||
+ | * 3 weeks - to port the compute intensive or critical parts of the dispatcher, boolean evaluator, optical renderer to OCL | ||
+ | * 1 week - torus (tor), truncated general cone (tgc) shot routines in OCL | ||
+ | * 2 weeks - bot (bag o' triangles) shot routine in OCL | ||
+ | |||
+ | Code should be implemented in C and OpenCL. The boolean weaving should be implemented in the 3 week segment where the boolean evaluator will be reimplemented in OCL. bot implementation is near the end because it might be possible to reuse some of the spatial partitioning code in it. |