Hardware Test Bench

Mouse test

Click, scroll and move. Every button lights up, the wheel reports each notch, and the chatter check catches a switch that clicks twice.

Move the mouse to begin

Test pad

move, click and scroll inside the box

Move the pointer in here

0, 0

Right-click and middle-click menus are suppressed inside this box so the buttons can be read. The rest of the page behaves normally.

Buttons

0 of 5 proven
L R 4 5
  • Left
  • Right
  • Middle
  • Back
  • Forward

Scroll wheel

idle
Notches up
0
Notches down
0
Sideways
0
Last delta
Reversals
0

Scroll slowly in both directions. A wheel that jumps backwards mid-scroll has a worn encoder.

Double-click check

Suspected chatter
0
Shortest gap
Average gap

Anything under 50 ms is faster than a human can click, so each one flagged is the switch bouncing rather than you.

Movement

indicative

Hz reported while moving

Pointer type
Buttons held
none
Movement step
Display scaling

Browsers group pointer events to the display refresh, so treat this as a health check rather than a measurement. A rate that collapses while you move is the finding that matters.

Why chatter is the fault to look for

A worn micro-switch sends two clicks from one press, and it is the most common way a mouse dies. It shows up as files opening when you meant to select them, or a drag that drops halfway. The switch is a cheap part, and several manufacturers cover this fault for years.

Side buttons are usually software

If the thumb buttons do nothing here, check the manufacturer's utility first. Logitech, Razer and Corsair software often binds them to macros or profile switches, which never arrive as ordinary button events. Reset them to forward and back, then retest.

What a web page cannot measure

DPI and raw sensor counts are gone before the browser sees them: the operating system has already applied pointer acceleration and display scaling. The movement trace is still good for spotting jitter and a sensor losing grip on a shiny desk.

How to use the mouse test

Right-click and middle-click menus are suppressed inside the test pad so the buttons can be read properly. Everywhere else on the page behaves normally.

  1. Click every button in the pad

    Move into the test pad and click left, right and middle, then press the thumb buttons if your mouse has them. Each one lights on the diagram and ticks off in the tally.

    A button that clicks physically but never lights is a failed switch or a broken solder joint, and on side buttons it is often just a driver profile that has unbound them.

  2. Scroll, and scroll back

    Roll the wheel slowly in both directions, then press it in for middle click and tilt it sideways if it tilts. Each notch shows up as a step.

    A wheel that occasionally jumps the wrong way has a dirty or worn encoder. Scrolling slowly makes it obvious in a way that fast scrolling hides.

  3. Run the chatter and rate checks

    Click the same button twenty times, deliberately and slowly. The chatter check flags any pair of clicks arriving too close together to be human.

    Then sweep the mouse steadily across the pad to measure polling rate. Higher is smoother, but what matters most is that it stays steady rather than collapsing.

Questions people ask

Double clicks you did not make, side buttons that ignore you, and what polling rate really tells you.

What is double-click chatter and how do I test for it?

Chatter is a worn or dirty switch sending two clicks from one press. It is the single most common mouse fault, and it shows up as files opening when you meant to select them, or dragging that drops halfway.

Click one button twenty times slowly and watch the chatter counter. Any interval under about 50 milliseconds is faster than a human can click, so each one flagged is the switch misbehaving rather than you.

Can I fix a mouse that double clicks?

Often, yes. Many switches fail because of oxidation on the contacts rather than mechanical wear, and working the button hard a few hundred times sometimes clears it temporarily. Contact cleaner applied inside the switch works better and lasts longer.

The permanent fix is replacing the switch, which is a cheap part but needs soldering. Check the warranty first, because several manufacturers cover this fault for years and it is a known defect on some popular models.

Why don't my side buttons register?

The usual cause is software rather than hardware. Manufacturer utilities from Logitech, Razer, Corsair and others often bind side buttons to macros or profile switching, and those never reach the browser as ordinary button events.

Set the buttons back to their default forward and back actions in that software and retest. If they still do nothing here but work elsewhere, the browser is filtering them; if they do nothing anywhere, the switch has failed.

What is polling rate and does a higher one matter?

Polling rate is how often the mouse reports its position, measured in Hz. 125 Hz means eight milliseconds between updates; 1000 Hz means one.

Above about 500 Hz the returns diminish sharply for most people. What matters more is consistency: a rate that swings wildly points at a failing cable, a congested USB hub, or wireless interference.

How accurate is the polling rate shown here?

Treat it as an indication rather than a measurement. Browsers coalesce mouse events and tie them to the display refresh, so a 1000 Hz mouse often reports lower here even when it is working perfectly.

Use it to compare a mouse against itself over time, or to spot a rate that collapses during movement. For an exact figure, use a dedicated desktop utility.

Does this test measure DPI or sensor accuracy?

No, and no web page can measure DPI reliably. The browser reports movement in CSS pixels after the operating system has already applied pointer acceleration and display scaling, so the raw sensor counts are gone before they arrive.

The movement trace is still useful for spotting jitter, stutter and a sensor that loses tracking on a reflective surface.

Why doesn't the right-click menu appear?

It is suppressed inside the test pad on purpose, because the context menu would open over the diagram and swallow the button event before it could be read.

Right-click works normally everywhere else on the page. Move the pointer outside the pad if you need the menu.

Can I test a trackpad or a trackball?

Yes. Both report the same events as a mouse, so buttons, scrolling and movement all register. Trackpad gestures arrive as scroll events, which is why two-finger scrolling shows up here.

Polling rate is not meaningful on a trackpad, since the numbers reflect the gesture engine rather than a sensor.

My scroll wheel skips or jumps backwards.

That is a worn optical encoder inside the wheel, and it is a common failure. Dust on the encoder wheel gives the same symptom and is worth ruling out first with compressed air.

Some mice have a free-spin mode that disengages the notched mechanism, so check you have not toggled it accidentally before assuming a fault.

Is my pointer movement being tracked?

Only inside this page, only while it is open, and only to draw on screen. No coordinates, clicks or timings leave your browser, and nothing is stored.

Test the rest of your setup

Every tool runs in this tab, reads your hardware directly through the browser, and stores nothing.