← What’s Lost with a Look · all measurements
The measured qubit’s own loss per readout
Runs. Seven IBM processors, 4–6 October 2026, three reads each (job IDs in the data file). A measured qubit is read out 1, 2, 3, 4, 6 or 9 times; its remaining contrast is converted to the time its own decay curve would need to lose as much.
What holds
- Time-equivalent loss per readout, in units of the stated readout duration (two estimators; means of three reads): kingston 1.661 / 1.677, marrakesh 1.516 / 1.509, fez 1.540 / 1.600, pittsburgh 1.340 / 1.360, boston 1.348 / 1.333, phoenix 1.351 / 1.323, aachen 1.371 / 1.351.
- The conversion assumes a single-exponential decay curve. Measured in the same campaign, 62.4–85.3 % of qubit-copies sit below their own 300 µs single-exponential extrapolation at 30 µs (26 jobs on six processors, 4 to 7 October, see controls; this page’s data cover 12 of those jobs, on marrakesh, kingston, pittsburgh and fez), so the conversion is approximate.
- At equal elapsed time, with no assumption about the decay curve, the comparison is from ibm_fez (11 October, decay under repeated readout) and the 22 October readout-instruction run (fez and marrakesh; kingston not decided; page).
Run it
Download the script and the files in data/ into one folder (data in data/), then python3 verify_measured_qubit.py. Standard library only; it prints each figure beside the recomputed value. The compact data were cut from the raw results; the raw files and the reduction scripts are available on request.
Files: verify_measured_qubit.py · data/measured_qubit.json · data/ruler_check.json