Flow-Field Particles illustration

Flow-Field Particles

A flow field assigns a direction to every point in space, and here that field comes from smoothly varying Perlin noise. Thousands of particles are dropped in and simply follow the arrows, curving and bending as the field does, leaving trails that braid into silky ribbons and rivers. Because the noise is coherent, nearby particles flow together, so the swarm self-organises into streams without any interaction between particles at all. Drag the noise scale, the particle speed, the field’s drift over time and the trail fade, choose a palette, and paint fresh particles with the mouse. It is a generative-art staple built from one noise function.

Runs 100% in your browser — simulations are computed locally on your device.

Notes

  • Each particle just samples the noise field for a direction and moves, with no forces between particles — the flowing structure comes entirely from the field being smooth.
  • Perlin (gradient) noise is coherent: values at nearby points are similar, so the field has no jumps and neighbouring particles peel into the same streams.
  • Slowly evolving the noise over time makes the whole field drift, so the ribbons breathe and reroute instead of freezing into a fixed pattern.
  • Runs 100% in your browser — simulations are computed locally on your device.