Skip to content

Latest commit

 

History

History
75 lines (53 loc) · 2.46 KB

README.md

File metadata and controls

75 lines (53 loc) · 2.46 KB
Raw

Pysics

This project is a 2d particle simulation implemented in Python, featuring optimized collision detection and resolution. The simulation includes rendering capabilities to visualize the particles and their interactions.

Features

  • Particle generation with random initial positions and velocities.
  • Gravity and wall constraints.
  • Collision detection using spatial partitioning for efficiency.
  • Collision resolution with impulse-based physics.
  • Rendering support to visualize the simulation.

Installation

  1. Clone the repository:

    git clone https://github.com/noahfarr/pysics.git
cd pysics

  2. Install the dependencies:

    poetry install

  3. Activate the virtual environment:

    poetry shell

Usage

To run the simulation, you need to have a renderer implemented. This project already has a RaylibRenderer implemented.

  1. Ensure you have the renderer set up and import it in the simulation script.

  2. Create a script to run the simulation:

    from pysics.render import RaylibRenderer
from simulation import Simulation

if __name__ == "__main__":
    renderer = RaylibRenderer((800, 600))  # Set the window size as needed
    sim = Simulation(n_particles=100, dt=0.01, renderer=renderer)
    sim.simulate(total_timesteps=1000, render=True)

All of these options can be set in the config.py file.

  1. Run the script: 
    python main.py

Project Structure

  • simulation.py: Contains the Simulation class and related logic.
  • particle.py: Contains the Particle class with physics properties and methods.
  • render.py: Contains the Renderer and RaylibRenderer classes for visualizing the simulation.
  • config.py: Contains a dataclass to configure the parameters of the simulation.
  • pyproject.toml: Configuration file for poetry, listing dependencies and project metadata.

Optimization Techniques

  1. Spatial Partitioning: Efficiently detects potential collisions by dividing the simulation space into a grid.
  2. Vectorized Operations: Utilizes NumPy for efficient mathematical operations.
  3. Reduced Function Call Overheads: Simplifies and inlines frequently called functions.

Contributing

Contributions are welcome! If you find any issues or have suggestions for improvements, please open an issue or submit a pull request.

License

This project is licensed under the MIT License. See the LICENSE file for details.