Go to file

MitchellHansen 1ed6a622bc That was a really dumb bug! Octree was being generated mirrored because I second guessed myself on a for loop		7 years ago
assets	Added new screenshots to show off imgui. As well as a quick delve into the blinn function to add the inverse square law	8 years ago
include	Whoops	7 years ago
kernels	Saving before breaking changes	8 years ago
notes	Hunting down some bugs and verfiying correct oct-tree traversal, not quite there yet but close	8 years ago
shaders	Got some geometry up and rotating. It will be pretty trivial to pass in	8 years ago
src	That was a really dumb bug! Octree was being generated mirrored because I second guessed myself on a for loop	7 years ago
.gitignore	Added a quick printout of the hardware info. Running into a problem choosing between platforms, going to abstract CL out into it's own class and hide all that logic	8 years ago
CMakeLists.txt	Added macOS compatability again	8 years ago
FindSFML.cmake	Added a quick printout of the hardware info. Running into a problem choosing between platforms, going to abstract CL out into it's own class and hide all that logic	8 years ago
README.md	Added new demo, tweaked some values to produce better lighting	8 years ago

README.md

About

This project is an implementation of an experimental "from scratch" volumetric rendering engine. Using OpenCL this program is able to hardware accelerate Woo's Fast Voxel Traversal Algorithm which traverses through a Sparse Voxel Octree as described by Laine et al. Light is simulated using the Blinn-Phong shading model along with a recursive ray tracer style of shadow tracing. Individual voxels are also textured using a user provided texture atlas and voxel data.

Build

In order to build this project you must have Cmake, GLEW, SFML, and the OpenCL development libraries installed / downloaded. You're also going to need OpenCL compatable hardware that supports the cl_khr_gl_sharing extension. The VS2015 build is working, Linux compiles but I don't have hardware that works, and Mac should compile and run, but I don't have access to a Mac.

Note on GLEW: macOS takes care of most of the "extension wrangling" so GLEW is not required when compiling for macOS
Note on cl_khr_gl_sharing: The cl_khr_gl_sharing extension, or cl_APPLE_gl_sharing for macOS, is only supported on certain hardware. Further, cl_khr_gl_sharing is not supported on intel CPU's running linux. You might be able to install and use beignet to get around this. The program will warn you if your CL device doesn't support this extension.

Screenshots

Video demo:

Screenshots: