CUDA Boid Simulation

I implemented a boid simulation in CUDA, leveraging GPU parallelism.

Table of Contents

1. Results Showcase
2. View the Source Code
3. Refereneces

Results Showcase

Below are GIFs showcasing different implementations with 5000 boids. The naive implementation is the least efficient, resulting in the lowest FPS. In comparison, the scattered grid implementation demonstrates a notable improvement, increasing the FPS by 100. Finally, the coherent grid implementation proves to be the most efficient, achieving an impressive FPS of over 925.

Naive Implementation

Scattered Grid Implementation

Coherent Grid Implementation

---

View the Source Code

Below is the portal to my GitHub repository. I have included a performance analysis and additional details in my README.md.

References

1. https://developer.nvidia.com/cuda-toolkit