Where it shows up
The same three-rule skeleton keeps reappearing wherever many simple parts self-organise without a boss. Swarm robotics uses it so drones hold formation and dodge each other from onboard sensing alone. Crowd and traffic simulations lean on separation and alignment to keep pedestrians and cars flowing. Film studios have rendered stampedes and bat swarms this way since the early 1990s. Zoom out and the pattern rhymes with active-matter physics: schooling fish, swarming locusts, and even the aligned spins of a magnet all show global order emerging from purely local coupling, no central plan required.
The knobs
- Agents: how many boids share the canvas. More of them thicken the flock and sharpen the murmuration; changing this reseeds the whole group from scratch.
- Separation: how firmly each agent avoids crowding. Turn it up for airy, well-spaced ranks; turn it down and they clump into a dense blob.
- Alignment: how eagerly agents match their neighbours' heading. High values give long, coherent streams; low values let the flock scatter into independent wanderers.
- Cohesion: the pull toward the local centre of mass. More cohesion gathers loose agents into tight balls; less lets clusters drift apart and dissolve.
- Speed: the maximum velocity, which sets how briskly the flock travels and how sharply it can turn.