Mouse Polling Rate Test

Polling rate is how many times a second your mouse tells the computer where it is. At 125 Hz that is once every eight milliseconds; at 1000 Hz once every millisecond. It sets a floor on how quickly cursor movement can respond to your hand, which is why it matters for tracking a moving target.

The setting lives in the mouse firmware and no browser can read it, so this tool measures it from the outside: it timestamps incoming position reports and looks at the gaps between them. Move at a steady speed and those gaps cluster around the true report interval, and one divided by that interval is the rate.

There is a complication that makes naive versions of this test wrong. Browsers coalesce pointer movement, delivering a single event per rendered frame no matter how many reports arrived in between. On a 60 Hz display that would cap any measurement at about 60 Hz regardless of the mouse. This tool uses the coalesced-events API where it exists, which hands back the individual reports inside each batch and lets rates well above the refresh rate be resolved. When that API is unavailable the page says so, and the figure should be read as a floor rather than an answer.

Two other things shape the result. Movement speed matters, because a stationary mouse sends nothing and a slow drift sends less than a fast sweep, so the test only counts while you are actually moving. Background load matters too, since a busy CPU delays event delivery and widens the intervals. The headline number is therefore a median across many intervals rather than an instantaneous reading, with the fastest observed rate reported separately as the best case your setup managed.

Read a result close to a standard rate as confirmation that your configuration is working. Read a low result as inconclusive: it tells you the browser did not receive reports faster than that, which is not the same as your mouse being incapable of sending them.