The Bit-Width Compendium

A field guide to processor bitness

Colophon

About this compendium

A small, independent reference built by people who could never quite agree, over coffee, on whether the 68000 was 16-bit or 32-bit — and decided to write everything down.

Why it exists

“Bitness” is the sort of specification everyone quotes and few examine. It appears on console boxes, in processor datasheets and in casual argument, standing in for speed, sophistication and generational prestige. But the number hides a genuinely interesting tangle of engineering decisions — register width against bus width, arithmetic reach against addressing reach, cost against capability. This site exists to pull that tangle apart patiently and in plain language, so that the next time someone insists a machine is “really” n-bit, you can ask the only question that matters: which width do you mean?

How it is made

The compendium is a static website — hand-written HTML and a single stylesheet, no frameworks, no build step, no client-side JavaScript. It is deliberately lightweight so that it loads instantly, reads comfortably on any device, and will still work in a decade. There are no advertisements, no trackers, no cookies and no analytics; we do not know who visits, and we like it that way.

Corrections & sources

Computer history is dense with edge cases, and we would rather be right than tidy. The Resources page lists the encyclopedic and institutional references we lean on, and where an article oversimplifies for the sake of a clear narrative, we try to flag the simplification rather than hide it. If you find an error, assume it is ours and not the historical record's.

On the name. The domain nods to the 8-bit and 16-bit machines that made computing personal — the eras when a single byte was a resource to be budgeted and a 64 KiB address space felt like room to roam. Everything since has been a story of that room growing, one doubling at a time.