Rolling pointer frequency estimate
Estimate mouse polling rate from browser pointer events, with average, stable, and peak readings from a rolling sample window.
For complete results, use a desktop or laptop with a connected mouse.
Move the pointer continuously inside this area
Use quick circles for several seconds. The rolling estimate uses valid movement samples from the last second.
Collecting movement samples
Event source:pointermove fallback
This is a browser-based estimate. Browser scheduling, the operating system, device drivers, display refresh, power settings, and other software can affect the result.
It is not equivalent to measurement with dedicated USB or hardware analysis equipment.
The tester listens for pointerrawupdate when the browser exposes it and otherwise falls back to pointermove. It timestamps valid movement with performance.now(), rejects duplicate or implausible intervals, and calculates readings from recent samples.
Average is based on the mean valid interval. Stable uses the middle of the rate distribution with extreme samples trimmed when enough data is available. Peak shows the fastest accepted interval and is capped to avoid clearly exaggerated results.
A mouse may advertise 125, 500, 1000, or more Hz, but a web page does not observe the USB bus directly. Browser scheduling, coalescing, drivers, CPU load, power settings, display refresh, and motion speed can all change the estimate.
Polling rate is how often a mouse reports movement to the computer, usually expressed in hertz. A nominal 1000 Hz device aims to report about 1000 times per second under suitable conditions.
No. A browser measures the pointer events it receives after operating-system, driver, and browser processing. The result is a browser-based estimate, not a direct USB measurement.
Slow motion, event coalescing, power saving, CPU load, browser scheduling, wireless conditions, and display environment can lower the observed event frequency.
Use several runs and compare the stable reading under the same conditions. A single peak is more sensitive to timing noise and should not be treated as the device's guaranteed rate.
More diagnostics
Use each focused test to compare click, button, wheel, and movement behavior without uploading event data.
