Mouse Test

Click here to start testing

0 of 5 buttons tested

How it works

1. Click Into the Diagram to Focus It. The test area only registers input once it's focused, so page scrolling and browser shortcuts stay untouched everywhere else.
2. Press Every Button at Least Once. Left, right, middle, and the two side (back/forward) buttons if your mouse has them — each lights up on press and stays marked once released. Any button past those five that reaches the browser is added to the list as you press it.
3. Click a Few Times on Any Button You're Worried About. A handful of clicks is enough for the chatter check to tell a worn switch from a normal double-click.
4. Scroll Up, Down, and Side to Side. Notch count and the size of each notch appear live, along with a warning if the steps look uneven.
5. Run the Polling-Rate Measurement. Press the dedicated button, then move your mouse in circles for a few seconds — this needs its own short measurement window, unlike the other checks.
6. Reset to Test Again. Clears every result without needing to reload the page.

How to Read the Diagram

Each target has three states: neutral (not tested yet), bright (held down right now), and marked (pressed and released at least once). The goal is simple — turn every button and wheel direction into a marked one.

A Double-Click Failure vs. a Real Double-Click

Most "double click testers" measure the time between one press and the next press. That number can't actually tell a worn switch apart from a fast, deliberate double-click — both can be under 200ms.

This tool measures something more specific: the gap between releasing a button and pressing it again. A human finger physically has to lift and come back down, so even a very fast intentional double-click leaves a real gap there. A chattering switch has no finger to lift — it fires a second full press-and-release on its own, usually within 5-15 milliseconds of the first release, because worn contacts inside the switch briefly reconnect on their own after opening.

What Polling Rate Is, and What Your Browser Will Actually Let This Page See

Polling rate is how often your mouse reports its position to the computer, in times per second (Hz). A 1000Hz mouse reports a thousand times a second — smoother motion, lower input latency. No browser API reports this number directly; it has to be inferred from how closely spaced the position updates arriving at the page actually are.

• Full precision: the browser delivers every raw update — the most accurate estimate this page can produce.
• Reduced precision: the browser bundles several updates together before delivering them, and this page unpacks that bundle to recover most of the detail.
• Not available: the browser only delivers updates capped to the screen's own refresh rate — at that point, any number reported would actually describe your monitor, not your mouse, so this tool refuses to show one rather than guess.

Why This Tool Won't Guess Your Mouse's DPI

DPI (dots per inch) is how far the cursor moves on screen for a given physical distance the mouse travels. You'll find pages online that print a confident-looking DPI number after a short mouse wiggle — that number is not a measurement, because a browser cannot see it.

Every pointer-move value a browser receives has already passed through your OS's pointer-acceleration curve (macOS and Windows both apply one, and it's non-linear) and been divided by your display's pixel density. The physical distance your hand moved is simply not information the page ever has access to. This tool reports what it can verify — button state, click timing, scroll steps, and polling cadence — and skips DPI rather than dressing up a guess as a result.

Scroll Wheel Encoders

A mechanical scroll wheel moves in discrete steps called notches, produced by a small encoder inside the mouse. As that encoder wears, it can start skipping a notch under normal use, or report an uneven step size — both feel like "the wheel scrolled too far" or "nothing happened" when you scroll normally.

This tool counts vertical and horizontal notches separately, reports the typical size of one notch, and flags it when notch sizes vary more than a healthy wheel should. Chrome and Firefox report notch size differently from each other by design — that's a difference in how each browser measures the same physical click, not a sign anything is wrong.

Side Buttons and Browser Navigation

The two side (thumb) buttons on many mice are wired by the browser itself to back/forward navigation — that mapping happens above the level any website can fully override. This tool does its best to keep you on the page while testing those buttons, but depending on your browser, pressing one may still navigate away.

Mice With More Than Five Buttons

A browser never sees your mouse — it sees numbered button events handed to it by the operating system, and that numbering runs out early. Left, right, middle, back and forward are the five with standard names; a sixth or seventh index only arrives if your mouse sends it as a plain mouse button. This tool keeps whatever does arrive: press a button beyond the standard five and it joins the list as “Button 6”, “Button 7” and so on, graded by the same chatter check as the named ones.

• Thumb wheel: a second, sideways wheel under your thumb reports as horizontal scrolling — exactly what tilting the main wheel produces. Both are counted here under horizontal steps.
• Gesture and mode buttons: usually consumed by the mouse's own software (Logitech Options+, Razer Synapse, SteelSeries GG) and turned into keystrokes or app commands, so the browser never learns a button was pressed at all.
• Free-spin and high-resolution wheels: send far more, far smaller steps than a ratcheted wheel. This tool recognises that pattern and only checks the slower, ratcheted part of your scrolling for uneven steps, instead of calling a working wheel faulty.

Frequently asked questions