Biosim4 is a biological evolution simulator written in C++ that models populations of virtual organisms competing for resources and reproducing according to evolutionary principles.
The simulator addresses the question of how complex behaviors emerge from simple evolutionary pressures. It works by creating a grid-based environment where agents with neural network-based brains sense their surroundings, make decisions, and interact with each other. Organisms can move, eat food, and reproduce, with their neural architectures and connection weights subject to mutation and natural selection across generations. The system tracks how populations adapt over time in response to environmental conditions and competitive dynamics.
The tool is suited for researchers and hobbyists interested in artificial life, evolutionary computation, and emergent behavior. It provides a sandbox for experimenting with how different environmental pressures, mutation rates, and selection mechanisms shape population characteristics. The simulator is particularly useful for those wanting to observe evolution in action rather than studying it theoretically, and for understanding how simple neural networks can produce surprisingly complex behaviors through iterative selection.
Development activity shows consistent engagement with the codebase. The project receives regular commits addressing bugs, performance improvements, and feature refinements. Pull requests are reviewed and merged, indicating active maintenance and openness to contributions. Issues are tracked and responded to, suggesting the maintainer actively monitors user feedback and questions about the simulator's behavior and capabilities.