Understanding Threshold Map in Ordered Dithering.
Single-pass palette refinement and ordered dithering
Usually image color reduction is done in two independent phases: first a color quantization algorithm finds a palette, which is then used by a pixel mapping phase to assign every output pixel a color from the palette. Ignoring any possible dithering for now, the chosen palette color is the one with the shortest Euclidean distance to the original pixel color. But if we know both the old and new colors for each pixel, it’s possible to refine the palette.
Ditherpunk — The article I wish I had about monochrome image dithering — surma.dev
I always loved the visual aesthetic of dithering but never knew how it’s done. So I did some research. This article may contain traces of nostalgia and none of Lena.
Surface-Stable Fractal Dithering Explained
Dither it!
Free online image dithering tool. Floyd-Steinberg, Atkinson, Bayer ordered dithering, animated GIFs, and multi-image upload — processed locally in your browser.
The mathematical beauty of hyperbezier curves
The mathematical beauty of hyperbezier curves
Notes on Gamma
The nuance between seeing and perceiving, linearity and non-linearity, what EOTFs are – and how to draw nice smooth gradients against the odds
40x less overhead! Rateless Invertible Bloom Filters - Part 3 of 3
How Stable Diffusion works
Understand in a simple way how Stable Diffusion transforms a few words into a spectacular image.
Why Between The Hashmarks Left Substack Behind
Complete guide to samplers in Stable Diffusion
Dive into the world of Stable Diffusion samplers and unlock the potential of image generation.
Steam Community :: Guide :: Map Generation in Slay the Spire
In this brief guide I will try to explain the relevant information regarding Map Generation and maybe some new ways to look at the game....
Alex Cheema on Twitter / X
Excited to collab with @ashxhart on this!Why does prefill/decode disaggregation work on DGX Spark + Mac ?DGX Spark: ~350 TFLOPS@fp4, 128GB@273GB/sM5 Max MacBook Pro: ~70 TFLOPS@fp4, 128GB@614GB/sPrefill is compute-bound. Prefill on Spark runs ~5x faster than M5 Max.Decode… https://t.co/xl9l0ArRhs— Alex Cheema (@alexocheema) August 19, 2026
