Surface code memory, distance 3 on QCCD — every design, ranked

Pick a measure to rank by; click a dot to open a design.

About this board

Three rounds of the distance-3 rotated surface code's syndrome extraction (eight ancillas, reset and reused every round), then the data read out; ranked by the logical error rate per round under the noise model built from the device's physics. Every design ran the same experiment, compiled by the reference compiler; its hardware program was turned into a noisy stim circuit by the noise model qccd-noise@1 (a text, on the Evaluation page) read over each device's own physics, and sampled (up to 200,000 shots or 200 logical errors, decoder pymatching, fixed seed). Ranked by the logical error rate per round, lower is better; click a dot to watch the ions and read the stim program that was sampled. Each dot is a design; pick what to rank by, click a dot for its numbers and the steps in the design tool where each claim can be checked.

Click a dot to see the design's numbers, its claim, and the steps to check it on.
Submit a design

Coloured dots are submitted designs, each credited.

Disqualified — not ranked

Not ranked: each breaks a rule or the task’s hardware budget.

designfamilybreaksR10
Surface code memory, distance 3 on grid5x5 — a small planar lattice
40 traps on the wires of a 5x5 junction grid. LER per round 4.12e-04, 95% interval 3.6e-04 to 4.7e-04 (210 errors in 170,000 shots)
latticeover budget: dacs 1600 > max_dacs 128passed
Surface code memory, distance 3 on grid9x9 — the baseline lattice
traps on the wires of a 9x9 junction grid. LER per round 4.88e-04, 95% interval 4.3e-04 to 5.6e-04 (205 errors in 140,000 shots)
latticeover budget: dacs 5760 > max_dacs 128passed
Surface code memory, distance 3 on cylinder 6x4 — four concentric rings with spokes
the torus drawn as a chip: 42 traps, no crossings. LER per round 5.21e-04, 95% interval 4.5e-04 to 6.0e-04 (203 errors in 130,000 shots)
torusover budget: dacs 1680 > max_dacs 128passed
Surface code memory, distance 3 on random 4-regular graph, 16 junctions
32 traps on a random expander: short paths, no drawing. LER per round 4.51e-04, 95% interval 3.9e-04 to 5.2e-04 (203 errors in 150,000 shots)
random graphR20 R21 over budget: dacs 1280 > max_dacs 128passed
Surface code memory, distance 3 on torus 4x4 — a periodic lattice
every row and column a closed loop, 32 traps. LER per round 4.71e-04, 95% interval 4.1e-04 to 5.4e-04 (212 errors in 150,000 shots)
torusR20 R21 over budget: dacs 1280 > max_dacs 128passed
Surface code memory, distance 3 on chain — a line of traps
one linear trap
lineover budget: dacs 1728 > max_dacs 128not run