Skip to content
This repository has been archived by the owner on Nov 30, 2024. It is now read-only.
/ BAKU Public archive

High-performance grasshopper plugin for simulating large number of particles using GPU acceleration.

License

Notifications You must be signed in to change notification settings

sean1832/BAKU

Repository files navigation

BAKU

High-performance Grasshopper 3D plugin for simulating large number of particles using GPU acceleration.

baku-demo-v0.1.0.mp4

⭐ If you like this project, please consider giving it a star!

💡 This project is early in development. If you encounter any issues, please report them in the issues section.


📘 About

This project started as a personal experiment to learn about GPU programming and high-performance computing. The goal is to create a high-performance particle simulation plugin for Grasshopper 3D using GPU acceleration. The project is named after the Japanese mythical creature Baku which is said to consume nightmares.

🌟 Features

  • Real-time simulation of large number of particles with Boids algorithm
    • 10k particles in real-time on a RTX 4090
    • Currently unoptimized (looping all the boids positions), expect better performance in future releases.
  • GPU accelerated using ILGPU
    • Up to 100x faster than CPU
    • Supports CUDA and OpenCL
    • Supports CPU fallback
  • Customizable boid and particle behaviors
  • Particle trails

📝 Planned Features

  • Spatial Hashing algorithm for performance improvement.
  • Attractor and repeller points
  • Obstacle avoidance
  • Boundary repulsion (currently particles are confined to a box)
    • Boundary repulsion is implemented since v0.1.0.

🖥️ Installation

  • Download the latest release from the releases page.
  • Unzip the file under the Grasshopper Libraries folder.
  • Unblock the DLLs by right-clicking on the files, selecting properties, and clicking the Unblock button.
  • Restart Rhino and Grasshopper.

📜 License

This project is licensed under the Apache License 2.0 - see the LICENSE file for details.

🤝 Contribution

Contributions are welcome! Please follow standard contribution guidelines.

🔗 References