Skip to main content
All experiments
Generative
Live

Generative ASCII

Typographic ASCII art generated from particle attractor systems. Each page load produces a unique piece with rarity tiers, like pulling a shiny card.

A particle simulation runs against one of four attractor types: Lorenz, Rossler, spiral, and flow field, depositing brightness values onto a grid. Each cell's brightness is mapped to a character from a density-sorted palette, producing ASCII art that reflects the attractor's topology.

PreText measures exact character widths in the target monospace font, ensuring the proportional-to-mono mapping preserves spatial structure. The seed is derived from the timestamp, so every page load is unique. A deterministic RNG means the same seed always produces the same piece.

Rarity is assigned by a hash of the seed: 75% common, 17% uncommon, 7% rare, 1% mythic. Rarer pieces get longer reveal animations, colored borders, and glow effects. Hit 'New pull' to generate another.

Highlights

  • Four attractor types: Lorenz, Rossler, spiral, flow field
  • Unique piece every page load, deterministic from seed
  • Rarity tiers with progressive reveal animation
  • PreText character width measurement for spatial accuracy

Stack

  • TypeScript
  • React
  • PreText
  • Canvas Math