Mouse Polling Rate Test

The mouse polling rate test timestamps mouse pointer-movement events delivered inside its surface. It calculates the current interval-derived rate, a rolling average over recent intervals, a median rate, an overall delivered-event estimate, the highest valid observed interval rate, median interval, p95 interval, usable sample size, and duration. Very small or invalid intervals are filtered so the display never divides by zero or produces an infinite result.

The primary label is estimated browser-observed mouse event rate. Hardware polling rate describes how frequently a device reports state at another layer of the input stack. A web page cannot inspect the USB bus, device firmware, receiver scheduling, or every report that the operating system may combine before dispatch. Similar numbers can occur, but the mouse Hz test cannot certify that they represent the same process.

A rate in hertz can be estimated from time spacing: 1000 divided by an interval in milliseconds. The tool uses multiple summaries because one interval is noisy. Current rate reflects the latest valid interval, rolling average follows recent activity, and median rate derives from the middle interval in the stabilized sample. Overall estimate divides delivered event count by elapsed time, while peak records the fastest valid interval-derived observation.

The first few events are ignored in interval statistics so the act of entering the area and beginning motion has less influence. The tool also requires a minimum sample and duration before marking the result usable. A limited or insufficient label is not a low hardware score; it simply means the browser sample is too short or sparse for a steady interpretation.

Common rate numbers such as 125, 250, 500, and 1000 Hz are useful reference bands, but the browser cannot guarantee a device's configured polling rate. When a usable median falls reasonably near one of those values, the mouse rate checker may show a near-band label. That label describes numeric proximity in the delivered events only. It is not a manufacturer specification, USB protocol capture, or validation badge.

A 1000 Hz device can produce a lower browser-observed rate because motion, scheduling, power saving, coalescing, system load, display timing, a remote connection, or browser behavior limits delivery. A peak near a familiar band can also be misleading if the median and p95 interval are unstable. Use the median, duration, sample size, and repeatability together rather than choosing the largest number on screen.

Use a desktop mouse on a surface that permits continuous movement. Keep the tab visible, avoid touching the edge of the screen repeatedly, and move in smooth loops or broad horizontal and vertical strokes. Collect at least 25 usable intervals over roughly a second, preferably longer. Keep the connection mode, vendor profile, power state, browser, browser zoom, and operating-system settings fixed across comparison runs.

If you intentionally change a device polling setting, reset and take several samples before and after. Do not compare one peak from each run. Compare median rate, rolling stability, median interval, p95 interval, and sample duration. A consistent before-and-after shift is useful browser-level evidence, while overlapping variable samples suggest that the current environment or movement technique is limiting the comparison.

Pointer events may be aligned with rendering, combined, delayed by main-thread work, or exposed differently by each browser engine. Operating systems can apply power policies, accessibility transformations, remote-desktop processing, or input scheduling. Wireless receivers and Bluetooth modes have their own pipelines. The mouse polling rate checker sees only the final messages dispatched to the page and cannot isolate which earlier layer shaped them.

Rendered frame rate is also a separate measurement. A 60 Hz display does not automatically force every input path to 60 reports per second, but applications may process or render input on frame boundaries. Likewise, browser movement event rate is not wheel-event rate. Use this tool for pointer movement, the scroll speed test for delivered wheel-event density, and purpose-built native or laboratory equipment when a physical protocol measurement is required.

For an unexpectedly low or unstable sample, move continuously for longer, close heavy tabs, keep the browser foregrounded, and repeat. Check the official device profile, connection mode, battery, receiver placement, and power-saving settings. Try another browser and a reputable native utility. Change one variable at a time and avoid assuming that a single browser number proves a cable, sensor, receiver, or driver fault.

Timestamps and calculations stay in browser memory and the tool does not upload raw movement coordinates or intervals to perform the mouse polling rate test. No screen position history is retained by the tool. Continue with the full mouse test for button mapping or the double click test for press timing. Browser-observed delivery is useful diagnostic evidence, but it cannot certify the physical polling rate.