Image Dithering
Enter text or upload an image, pick a dithering algorithm and color palette, then download the result.
Input
Output Size
Dithering
Palette
Preview
128×128 preview
Download
Favicon
How this works
The input — text rendered via your system font or an uploaded image — is drawn onto an HTML Canvas. The resulting bitmap is then reduced to a limited color palette using dithering — a technique that simulates colors outside the palette by arranging the available colors in patterns that trick the eye at a distance.
Floyd-Steinberg
Error-diffusion dithering. For each pixel, it finds the nearest palette color, then pushes the rounding error to neighboring pixels (right and below) using fixed weights (7/16, 3/16, 5/16, 1/16). Produces organic, noise-like patterns.
Atkinson
Error-diffusion variant from early Macintosh software. Distributes only 6/8 of the error across 6 neighbors (each getting 1/8). The intentional loss drives flat areas to pure palette colors, producing a high-contrast, airy look.
Ordered (Bayer)
Threshold-map dithering. Each pixel is offset by a value from a repeating 4×4 matrix before finding the nearest palette color. No error is spread between pixels. Produces regular, grid-like patterns.
Everything runs client-side in your browser. No data is sent to a server.