Baseline
A baseline is a recorded reference result captured under known conditions, giving you something to compare against after any hardware or software change.
Also known as reference run, baseline measurement
Measured by SystemCheck
Part of the four-stage run: a 60-second sustained CPU load reporting throughput, decay and consistency; a memory stage reporting bandwidth and latency; a GPU raster load at a fixed resolution reporting frame rate, pacing and decay; and a path tracer reporting samples per second. The paragraphs below say which metric carries this term.
Without a baseline you cannot answer the only question that matters after a change: did this help? Memory of how a machine felt is unreliable, and comparing against someone else's result introduces differences in hardware, ambient temperature, background load, and driver version all at once.
A good baseline is captured deliberately: no background applications, known ambient temperature, machine already warm or already cold as a stated condition, driver version recorded. Re-capture it after any deliberate change so the reference stays current, and keep the old one so you can see drift over time.
SystemCheck's report includes measurement revision, GPU backend, and validity signals precisely so a saved report functions as a proper baseline. Capture one when a system is new and healthy; it becomes the reference for detecting degradation years later, when a raw number in isolation would tell you nothing.
Keep the baseline with its context. Note whether the machine was plugged in, which power mode and maker profile were set, which browser and version ran the test, and roughly how warm the room was. A later result can only be read against the baseline if those things match or the differences are known. Two baselines are better than one: one taken cold and one taken after the machine has been working for a while.
Related terms