1.8k words | Dan Hollick
Rasterisation and anti-aliasing.
Rasterisation asks which pixels a mathematical shape covers; anti-aliasing preserves partial coverage at its edges.

Rasterisation asks which pixels a mathematical shape covers; anti-aliasing preserves partial coverage at its edges. The apparent simplicity comes from a set of carefully chosen representations, transformations and physical assumptions working together.
Coverage
Edge pixels receive a value proportional to the amount of the pixel covered by the ideal shape.
This is one part of a longer chain: vector edge becomes sample grid becomes coverage becomes smooth pixels. The useful abstraction hides the physical work, but the underlying constraints still shape the software built above it.
Sampling
Extra samples reduce jagged edges by measuring detail at a finer scale before averaging.
The implementation is full of compromises. Precision, speed, storage and energy rarely improve together, so practical systems choose the errors people are least likely to notice.
Hinting
Small text may adjust outlines to align important strokes with the pixel grid.
Once this layer is visible, familiar design conventions stop looking arbitrary. They are accumulated responses to the capabilities and limits of the machinery below.
A visual study based on the original chapter. Text is condensed and rewritten.