Project info
- Developer Philip Gooch
- Showcase year 2021
- Programme Computer Game Applications Development
Synopsis
This project compares different approaches to optimizing Smoothed Particle Hydrodynamics (SPH). These techniques can be applied to any particle systems that involve collision detection. There are single threaded and multi-threaded applications. Spatial data structures are utilized to address the quadratic time complexity involved in collision detection between particles. This project makes use of Compute Unified Device Architecture (CUDA) for general purpose Graphics Processing Unit (GPU) programming and OpenGL for rendering.
The aim of the project is to see how many particles can be included in the SPH simulation whilst maintaining 60 frames per second, for real-time interactivity.
Optimization techniques for Smoothed Particle Hydrodynamics
Optimizing a Smoothed Particle Hydrodynamics simulation using spatial data structures and GPU acceleration with CUDA.
Motivation
I was interested in Craig Reynold's "boids" algorithm. I was frustrated at the limited boids I was able to compute. Looked into ways of optimizing it. Discovered SPH and got interested in that. They are both N-body simulations. I very much enjoy optimizing code so I thought this would be a great subject to study for my honours project.
Credits
鈥淥ptimization techniques for Smoothed Particle Hydrodynamics鈥� is a 2021 Digital Graduate Show project by聽Philip Gooch, a Computer Game Applications Development student at 尤物视频.聽聽聽聽聽聽
Find out more 聽聽聽聽聽
听听听听听听听听听听听
