GEARCHECK LABRepeat a fixed computation five times and compare how long each round takes. This helps reveal inconsistent browser performance under the current conditions.
READY — choose an action below
No measurement yet.
Each round executes the same deterministic arithmetic workload in a worker. Timing covers computation inside that worker rather than page rendering. Five rounds are separated by short pauses to keep the page responsive. The minimum and maximum show the spread in this run; they do not establish the cause of a slowdown. JIT warmup can make later rounds faster, while background processes can cause an isolated slow round. A checksum confirms the result was produced, not that the hardware is universally healthy.
This tool does not read CPU temperature, clock frequency, power draw or thermal throttling flags. Do not label a slower round as overheating. Browser updates and optimization change results, so cross-device comparisons need care. The workload is bounded and the worker is terminated when stopped, hidden or completed. No result is sent to analytics.
Device previews, recordings, selected files and report notes stay in your browser. Downloaded files remain on your device. The call check sends small HTTP probes to GearCheck Lab. The WebRTC test exchanges temporary connection descriptions through our server and contacts the disclosed STUN service only after you create or join a session. Ordinary page requests and usage measurement are described in our privacy policy.