The standard reference on surface codes, written as a tutorial rather than a terse paper. Where every industrial fault-tolerance roadmap ultimately points.
Surface codes
A two-dimensional topological code needing only nearest-neighbour connectivity, with a relatively forgiving error threshold near 1% — the leading candidate for real fault-tolerant hardware.
Why it matters
Almost every superconducting-qubit roadmap in the industry is built on the surface code, so its overheads set the qubit counts everyone quotes for useful machines.
After this you will be able to
- Explain the surface code layout, stabilisers and threshold
- Estimate physical-qubit overhead for a target logical error rate
- Follow current experimental below-threshold results
3 best places to start
Hand-picked and ordered. If you only have time for one, take the first.
A cross-referenced encyclopedia of classical and quantum error-correcting codes, each with properties, decoders and references. Indispensable once you are past the textbook codes.
Simulates million-qubit stabiliser circuits in seconds, which is what makes modern error-correction research computationally possible. The tutorial notebooks double as a surface-code course.
2 more resources
The standard fast decoder for surface codes, and the natural companion to Stim. Together they let you run realistic threshold simulations on a laptop.
The first convincing demonstration that adding more physical qubits made the logical qubit better rather than worse. Arguably the most important experimental result of the decade so far.
This unlocks
Topics that list Surface codes as a prerequisite.