A simple CUDA Monte Carlo path tracer that runs at about 19 ms/sample.

  Interesting things that were implemented include:   Subsuface Scattering, Stream Compaction, Material Sorting, Intersection Caching

  GPU Rasterizer written in CUDA that runs at >60 FPS

  Implements tile based rasterization and scanline rasterization along with standard features like depth testing via mutex locks; texture mapping; backface culling; point, line, and triangle rasterization;

Visualize 1.6 million particles running at 60fps

  GPU Flocking Simulation written in CUDA

  Tested on: Windows 10, i7-7700HQ @ 2.8GHz 32GB, GTX 1070(laptop GPU) 8074MB

  Implementaion of the Stream Compaction technique and a performance analysis of the same.

  Stream compaction is a technique that performs the following: given an array and some condition, creates and returns a new array that contains only those elements from the original array that satisy the condition, while preserving the order of the elements.