(Simulation with 10K particles)
(Simulation with 25K particles)
A simple boids simulation done using Cuda and C++.
Techniques implemented :
- Brute Force approach (Each boid iterates over every other boid).
- Uniform grid approach (Each boid iterates over boids only in neighbouring spatial grids).
- Coherent grid approach (Similar to Uniform grid, but memory access is more linear (fewer indrections required)).
Starting code / template obtained from : https://github.com/CIS565-Fall-2022/Project1-CUDA-Flocking.
Performance Analysis
