SlotLayer 0
Measure what the machine can actually do
Run a protocol whose answer is already known, on the hardware, and read the machine's own performance off how far the result falls short. Nothing here computes anything a user wanted — the point is that the answer is known in advance, because that is what makes the shortfall a measurement.
A programmable device — its qubits, its native gate set, its connectivity and its measurement — plus how many circuits and how many shots you are willing to spend, and the confidence level the answer has to be established at.
A number characterising the hardware, the protocol that produced it, and the statistical confidence it holds at — never an answer to a computational problem, because no computational problem was posed.
This one, drawn
From Physical qubits to Number about the machine
A circle is an object you are holding. Each line between the two ends is one recorded way through this slot; where a way is built from smaller slots, those are its own lines. Circles are named on hover, and each one is a link.
Nothing drawn here has a recorded way through it that this figure leaves shut. See it on the map
Why this is a layer
What a reader is choosing between is **what the number is about**: one method isolates a gate set from the rest of the machine, the other refuses to. Randomized benchmarking reports the average error rate of a gate set, from the decay of fidelity over random Clifford sequences, deliberately independent of everything around it. Quantum volume reports the largest random square circuit the whole machine can actually execute — gates, connectivity, crosstalk and compiler together — and Cross et al. introduce it precisely because they hold that the first kind of number does not predict the second: "performance of isolated gates may not predict the behavior of the system. Methods such as randomized benchmarking, state and process tomography, and gateset tomography are valued for measuring the performance of operations on a few qubits, yet they fail to account for errors arising from interactions with spectator qubits." So the two do not merely differ in what they measure — one of them is an argument about the other's category, which is a stronger condition than a layer needs and the reason this one is not a topic tag.
Ways to do this
2 methods recorded
- Quantum volume from random square circuits
Run random circuits that are as deep as they are wide, and ask how often the machine returns one of the outputs that should be more likely than the median. Widen and deepen together until it can no longer beat that bar; the last size it managed is the number.
- Randomized benchmarking over Clifford sequences
Apply a random sequence of Clifford gates, then the one gate that undoes all of them, and see how often the machine comes back to where it started. Lengthen the sequence and the return probability decays; the decay rate is the average error a single gate costs, and preparation and measurement errors fall out of the fit rather than contaminating it.
Routes that skip this layer
No recorded route avoids this step.
This is a step inside
Nothing in this graph needs this as a step, so it is where a reading starts.
In the Atlas
No record in the Atlas covers this yet. The catalogue is circuits and primitives; this part of the literature is not in it.