PLife

A particle simulation written in C. Particle Life explores the concept of emergence, where complex structures arise from simple rules. Inspired by Tom Mohr's video on Particle Life simulations.

Basic Usage (without runner)

The provided executable may not work on your architecture. If this is the case, you will need to compile the program in the provided build folder. We recommend using Conan + CMake, config files for which are provided.

./build/plife simulation.txt

Modify simulation.txt to configure simulation properties. Ensure that if you specify n quantities, the attraction matrix is n x n. We recommend a viscosity of 0.4 and a repulsion strength of 4.0 based on testing.

• Epochs: Number of simulation frames to render
• Quantities: Number of each particle type in simulation
• Attraction: Attraction matrix, in same order as quantities are defined
• Viscosity: Slows particles down as they move to prevent chaos
• Radius: How close particles need to be to interact
• RepulsionStrength: How much particles push back on eachother when getting too close (to minimize particle overlap)

Example simulation.txt:

#Epochs
1000
#Quantities
50 50
#Attraction
0.5 -0.2
-0.3 0.8
#Viscosity
0.4
#Radius
50
#RepulsionStrength
4.0

Using runner.py

The runner.py script automates the creation of simulation.txt with customizable attributes and runs the PLife simulation. Currently epochs, viscocity, radius, and repulsion strength cannot be modified with flags directly. If you wish to modify these, do so in the runner.py script directly. runner.py only supports creation of uniform particle quantities.

Note that the plife executable is still required to use the runner.

Basic Usage

python3 runner.py -s SIZE -n NUMBER

• -s SIZE: Number of particle types (required).
• -n NUMBER: Number of particles per type (required).

Example:

python3 runner.py -s 5 -n 200

Optional Flags

• -a, --assortativity: high or low.
• -d, --degree: Degree distribution (integer between 1 and size of matrix -s).
• -c, --clustering: Clustering coefficient (0 to 1).
• -k, --skew-symmetry: high, medium, or low.
• -r, --reciprocity: Reciprocity level (0 to 1).
• -p, --sparsity: Proportion of possible edges (0 to 1).

Examples

High Assortativity

python3 runner.py -s 5 -n 200 -a high

Specifying Sparsity

python3 runner.py -s 5 -n 200 -p 0.2

Applying Skew Symmetry and Reciprocity

python3 runner.py -s 5 -n 200 -k high -r 0.7

Combining Attributes

You can combine any number of attributes. However, some attributes will contradict each other, so note the following conditions.

• Sparsity and Degree Distribution: If both are specified, degree distribution takes precedence.
• Reciprocity and Skew Symmetry: Reciprocity is applied after skew symmetry and can override it.

Notes

• Modify build/simulation.txt to manually adjust particles, the attraction matrix, or epochs.
• The simulation.txt file can be renamed; specify the correct filename when running the simulation. However, runner.py uses build/simulation.txt.
• If you wish to modify the particle simulation code (src/main.c), you must set up your Conan environment or link dependencies manually. Conan is a package manager for C/C++ and is used in this project to resolve the SDL2 and ZLIB dependencies. Please refer to the Conan official documentation.