Glossary/Benchmark/Stress test vs benchmark
Stress test vs benchmark
A stress test pushes a system to its limits to find instability; a benchmark measures performance under a controlled workload to produce a comparable score.
Also known as stress test, stability test
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 calibrated GPU raster load reporting frame time, pacing and decay; and a path tracer reporting samples per second. The paragraphs below say which metric carries this term.
The goals differ, and so do the workloads. A stress test wants maximum stress (often unrealistically so) and its pass condition is simply that nothing crashed or produced a wrong result over hours. A benchmark wants representative, repeatable work and its output is a number you can compare.
Confusing them leads to bad conclusions in both directions. Passing a synthetic power-virus stress test does not prove stability in mixed real workloads, which stress different execution units. And a benchmark score says nothing about whether an overclock is stable. Plenty of unstable configurations post excellent scores right up until they corrupt something.
SystemCheck is a benchmark with sustained-load characteristics, not a stability tool. It measures throughput, decay, and consistency under a fixed four-stage load of roughly three minutes. It is not a substitute for Prime95, OCCT, or memtest86 when validating an overclock or undervolt. Use those for stability, SystemCheck for measurement.
Related terms