Simulator and hardware comparisons
Keep idealised, noisy and hardware runs distinct so comparisons remain interpretable.

The situation.
A circuit behaves differently when moved from a simulator to a hardware backend. Rather than combining results into one headline score, preserve execution context and examine how compilation, sampling and device conditions affect the observation.
A practical approach.
- 01
Fix the experiment
Define the circuit, measured quantities and comparison protocol.
- 02
Capture context
Record backend, compilation settings, sampling choices and relevant execution metadata.
- 03
Repeat and compare
Analyse repeated runs and explain which conditions differ between environments.
What to validate
- Label simulation and hardware results clearly.
- Avoid claiming a trend from one run without examining variability.
Typical outputs
- Comparison protocol
- Versioned run records
- Results with uncertainty and limitations
Example measures
- Variation between repeated runs
- Agreement with the chosen reference
- Completeness of execution metadata
Agree a baseline and acceptance criteria for the specific project.
Illustrative scenario and conceptual imagery. Scope, deliverables and acceptance evidence are agreed for each engagement; example measures are not achieved results.
