Loading page
About this tool
Test every mouse button online, catch double-click chatter to the millisecond, count scroll wheel ticks and estimate your mouse polling rate in Hz.
This mouse test listens for the browser's mousedown and mouseup events inside the test box and lights up the matching part of the mouse diagram, covering left, right, middle, back (X1) and forward (X2) buttons. Each press records the time since the previous press of the same button; any gap shorter than your chatter threshold (80 ms by default) is flagged as an unintended double click, the usual sign of a worn switch. Wheel events are counted by direction, and a direction flip within 60 ms is flagged as possible encoder bounce. Polling rate is estimated from the median interval between raw (coalesced) pointer-move events. Everything runs locally.
Open Mouse Test on AltFTool — it loads instantly in your browser.
Click Left, Right and the Middle (wheel) button inside the dashed test box, then press the Back (X1) and Forward (X2) side buttons; each part of the diagram turns green once detected.
Single-click one button 20–30 times at a normal pace and watch the Chatter tile and the Buttons table — any press closer than the Chatter threshold (80 ms by default) is counted and shown in red in the Click log.
Scroll up and down slowly to fill the Wheel tile, then move the mouse in fast circles inside the box to read Polling rate in Hz with its peak and nearest standard rate.
Press Copy report to copy the per-button clicks, chatter counts, wheel ticks and polling rate as text, or Reset to start over.
Back and forward are captured and their browser navigation is blocked inside the test box, so you can press them repeatedly without leaving the page.
Every press of every button is timed against the previous one; the table shows click counts, the fastest gap and a chatter count, with the threshold adjustable from 30 to 150 ms.
Using coalesced pointer events, the estimate reflects the reports your mouse sends rather than one event per screen frame, and snaps to the nearest standard rate such as 500 or 1000 Hz.
The click log lists every press with its interval and highlights chatter in red, and Copy report puts a plain-text summary on your clipboard for a support ticket.
| Polling rate | Report interval | Typical devices |
|---|---|---|
| 125 Hz | 8 ms | Basic office mice, default USB rate |
| 250 Hz | 4 ms | Older and budget gaming mice |
| 500 Hz | 2 ms | Many wireless and mid-range mice |
| 1000 Hz | 1 ms | Standard for gaming mice |
| 2000–4000 Hz | 0.5–0.25 ms | High-end wired and wireless gaming mice |
| 8000 Hz | 0.125 ms | Esports mice; needs a fast CPU and USB port |
Browser timing has finite resolution and may round very short intervals, so readings above about 2000 Hz are approximate.
Click a single button at a normal pace 20 or 30 times inside the test box. Each press is timed against the previous one; any gap shorter than the chatter threshold is counted as chatter and highlighted in the log. A healthy switch shows zero chatter. Several flagged presses mean the switch is bouncing and registering one click as two.
Most often a worn or oxidised micro-switch: its contacts bounce on release and the mouse firmware passes the bounce on as a second click. Static build-up and low debounce settings in the mouse software can also cause it. Raising the debounce time, replacing the switch or a warranty claim are the usual fixes.
The default 80 ms catches typical switch bounce while leaving room for normal clicking. Switch chatter usually produces gaps under 30 ms, while deliberate fast double-clicks by a person are usually 80 ms or more apart. If you are a very fast clicker, lower the threshold to 50 ms to avoid false alarms.
Move the mouse quickly in circles inside the test box for a few seconds. The tool takes the median interval between raw pointer events and converts it to hertz. A 1000 Hz mouse should read close to 1000 Hz at speed; slow movement reads lower because the sensor only reports when it moves.
The mouse must be moving fast and continuously, a USB hub or wireless power-saving mode can lower the rate, and some browsers deliver at most one pointer event per screen frame. The tool notes when raw coalesced events are unavailable; in that case try a Chromium-based browser for a closer reading.
Yes. Browsers report them as buttons 3 and 4. Inside the test box their default back and forward navigation is cancelled, so pressing them lights the Back and Fwd keys on the diagram instead of leaving the page. If they never light up, the mouse software may have remapped them to another action.
Scroll inside the test box: each wheel event increments the up or down counter and the arrow on the diagram lights. If a slow, steady scroll occasionally registers the opposite direction, the tool counts a direction flip, which points to a dirty or worn scroll encoder. Tilt-wheel left and right are counted too.
Partly. Taps and two-finger scrolling are detected as left clicks and wheel events, but touchpads do not have side buttons, and smooth scrolling sends many small events rather than discrete ticks. Polling rate is measured for mouse input only, so the tile stays empty on a touchpad or touchscreen.
No. Events are processed in your browser tab and kept only in memory; Reset clears them. The single value saved is your chatter threshold, stored in this browser's local storage so it is remembered next time. Closing the tab discards the click log and every measurement.