A Java framework that can simulate a variety of particle systems. It uses space partitioning and parallelization to achieve a significant speedup.

It can be configured to use the rules of Particle Life, a particle system showing life-like behaviour.

Check out github.com/tom-mohr/particle-life-app to see this framework in action.

In 2014 or earlier, Jeffrey Ventrella created Clusters. In 2018, @HackerPoet aka Code Parade released a video, describing the rules of Ventrella's particle system. This video popularized the idea and introduced the name Particle Life, referring to Conway's Game of Life.

This project requires Java 16 or higher.

This repository is intended to be included in other projects as a Gradle dependency. Add the following to your build.gradle file:

repositories {
    maven { url "https://jitpack.io" }
}

dependencies {
    implementation "com.github.tom-mohr:particle-life:v1.0.1"
}

You can now import the framework:

import com.particle_life.*;

This allows you to use the Physics class, which is the most important class of the framework. You can initialize it with the rules of Particle Life as follows:

Accelerator particleLife = (a, x) -> {
    double rmin = 0.3;
    double dist = x.length();
    double force = dist < rmin ? (dist / rmin - 1) : a * (1 - Math.abs(1 + rmin - 2 * dist) / (1 - rmin));
    return x.mul(force / dist);
};

Physics p = new Physics(particleLife);

To run the simulation, you have to choose between two updating schemes:

1. Synchronous updating:

   while (true) {
   
       p.update(0.02);  // run the simulaton
   
       // simulation step completed
   
       if (button.pressed()) {
           p.settings.wrap = false;  // modify state of physics now
       }
   }

2. Asynchronous updating:

   p.startLoop();
   
   while (true) {
   
       if (button.pressed()) {
           // simulation could be running right now
           p.enqueue(() -> p.settings.wrap = false);  // modify state as soon as the simulation step is completed
       }
   }
   
   p.stopLoop();

The second approach (asynchronous updating) is recommended if you have a UI that can take user input while the simulation is running.