Mouse Tester

Move the pointer into the test pad and press each button once. Each one lights up on the diagram and stays marked, and the Buttons table reads 5 of 5 tested when all have registered. Then roll the wheel up and down and move the pointer around the pad. Select Reset to clear the counts.

Click the suspect button slowly, about once a second, at least 20 times. Every press is timed against the previous press of the same button. A gap under 80 ms is flagged as a possible unintended double-click (switch bounce), and the Flagged column counts how often it happened. A healthy switch clicked slowly is never flagged.

A deliberate fast double-click normally has 100 to 500 ms between its two presses, and Windows accepts up to 500 ms by default. A switch that bounces sends its second press within a few tens of milliseconds, faster than a finger can repeat. If you click as fast as you can on purpose, you may occasionally dip under 80 ms, so judge a button on slow single clicks.

Not necessarily. Some browsers and operating systems do not pass the back and forward buttons to web pages, and mouse software can remap them to keystrokes or macros that this page cannot see. Close that software or try another browser before deciding. Inside the test pad the page cancels the browser's usual back and forward navigation, so pressing them should not take you away.

While you roll the wheel steadily one way, every step should go the same way. The tester counts a reversal when a single step goes the other way, with at least two steps before it and one after it in the original direction, each within half a second. That jump is a common symptom of a worn or dirty wheel encoder. A healthy wheel rolled steadily shows 0.

Not exactly. It shows how many move events per second the browser delivers while the pointer moves inside the pad, and the highest figure so far. Where the browser offers raw pointer updates or coalesced events, the count includes them; otherwise it may be limited to the screen refresh rate. Move the mouse quickly in circles for the highest reading and treat it as an indication, not a certified polling rate.

Yes. A trackpad reports left and right clicks, and two-finger scrolling arrives as a stream of small wheel steps, so the step counts are much higher than with a notched wheel. Up and down are the directions the browser reports, so they are swapped if your system uses natural scrolling. On a phone or tablet with no mouse, the page says a mouse or trackpad is needed and ignores taps.

No. Button presses, wheel steps and pointer positions are read only while the pointer is inside the test pad. They are shown on the page and discarded when you select Reset or leave, and nothing is sent to a server. Outside the pad, clicks, the right-click menu and scrolling work as usual.