Testers
Mouse Tester
Test mouse buttons, scroll wheel, double-click speed and polling rate in the browser before you blame a bad match on your hardware.
This one needs a mouse. On a touchscreen the checks below have nothing to read.
Double-click test
Double-click this pad.
- Switch chatter — under 25ms
- Registers as a double-click — 25–500ms
- Two separate clicks — over 500ms
Scroll wheel test
Hover here and scroll.
- up —
- down —
- left —
- right —
Polling rate test
Sweep across this pad.
A mouse fails in a small number of recognisable ways, and each of the four checks above isolates one of them. Work through them in order and you will know whether the hardware is at fault before you spend anything on replacing it.
Mouse button test: reading the result
Every button you press should light up and stay marked as working. Five is the maximum any browser can report: left, middle, right, and the two side buttons. The middle button is the one people most often misjudge, because browsers use it to open links in new tabs — a middle button that seems broken on a normal page is frequently working perfectly and simply being consumed by the browser before you notice. On the pad above, that behaviour is suppressed so the press can be read directly.
If a button registers here, the switch, the cable and the driver are all fine and any problem you are experiencing lives further up in software. If it never registers, the signal is not arriving at all, and the fault is physical.
Double-click test: the one that matters most
This is the fault people actually arrive with. A mouse button contains a mechanical switch whose contacts bounce briefly as they close, and part of the switch's design is to settle before that bounce becomes visible. Age, dust and spring fatigue lengthen the bounce until the mouse reports it as a second press.
The threshold on the page is twenty-five milliseconds. No one produces two deliberate presses that close together — competitive click-speed records sit an order of magnitude above it — so a gap that short is mechanical rather than human. Click normally, several times, on the button you suspect. If a chatter reading ever appears without you intending a double-click, the switch is on its way out. A single deliberate double-click landing in the middle band is exactly what a healthy mouse looks like.
Polling rate test: what the number really measures
Polling rate is how often the mouse tells the computer where it is. The complication in a browser is that pointer events are normally delivered once per frame, so a naive counter would measure your monitor rather than your mouse and report every device on earth as 60Hz or 120Hz.
The check above avoids that by reading the coalesced samples — the events the browser folded together before handing them over — each of which carries its own timestamp. That is the mouse's real cadence. It is still an estimate taken through several layers of software, so treat it as confirmation that a rate is roughly what you expect, not as a calibrated instrument. If the number sits between two standard rates, the tool says so rather than picking the closest one.
Scroll wheel test: encoders fail one direction first
A scroll wheel turns a small notched encoder, and encoders rarely fail evenly. The usual symptom is one direction reporting fine while the other stutters, misses notches, or occasionally jumps backwards — a page that creeps upward while you are scrolling down. Watching the counts on the strip above is the quickest way to confirm it: scroll ten notches each way and compare. Ten and ten is healthy. Ten and six, or a count that increments in the direction you are not scrolling, is a worn encoder or dust in the wheel assembly.
Horizontal counts staying at a dash is normal and not a fault. Only a tilting wheel reports sideways at all, and most mice do not have one.
When the mouse turns out to be fine
A clean sweep across all four checks moves the problem elsewhere. Pointer stuttering with a healthy report rate usually means the surface, the sensor or a wireless channel fighting with other 2.4GHz traffic. Clicks that register here but not in a game are usually a polling or input-mode setting inside that game. And a mouse that behaves differently on battery is a power-management profile throttling its USB port rather than any fault in the mouse itself.
Questions
- Why does my mouse double-click when I only clicked once?
- Almost always a worn or contaminated micro-switch. The contact inside a mouse button physically bounces when it closes, and the switch's job is to settle before the mouse reports anything. As the contacts oxidise or the spring weakens, that bounce lasts long enough to be reported as a second, separate press. You can see it on the double-click check above: a gap under 25 milliseconds is far shorter than any human can produce deliberately, so if one appears when you meant to click once, the switch is failing rather than your hand being unusually fast. Cleaning with contact cleaner sometimes buys months; replacing the switch is the real fix.
- How do I test my mouse polling rate in a browser?
- Move the pointer across the polling rate pad above in long, steady sweeps for a second or two. The tool reads the individual movement samples your mouse sent, times the gap between consecutive samples, and takes the median. The median matters more than it sounds: any real sweep contains pauses and direction changes, and averaging would let a single hesitation halve the number you see. It needs at least sixteen samples before it will show anything, which is a fraction of a second at any modern rate.
- Why does the polling rate test show less than my mouse is rated for?
- Three usual causes, in order of likelihood. The sweep was too short or too slow, so the samples measured your hand rather than the hardware — move faster and further. The mouse is not running at its advertised rate because the rate is a setting in its own software rather than a fixed property, and many ship set to 125Hz out of the box. Or the browser is throttling: a background tab, a power-saving profile or an overloaded page all reduce how often movement is delivered. A reading that lands between two standard rates is reported as exactly that rather than being rounded to the nearest marketing number.
- My side buttons do nothing in this test — is my mouse broken?
- Probably not. Browsers only see five buttons, and the two side buttons are the fourth and fifth of them. If your mouse has extra buttons beyond those, its own driver software usually sends them as keyboard shortcuts or macros instead of as mouse buttons, and no web page can see that. The check is only meaningful for buttons you know are meant to be back and forward. If those two register nothing here but work elsewhere on your system, the driver has remapped them.
- Can a browser test tell me whether my mouse is dying?
- It can answer some questions definitively and others not at all. A button that never registers, a wheel that reports one direction but not the other, and a switch that chatters are all faults this page can show you conclusively, because they are failures of the signal reaching the browser. What it cannot assess is anything the operating system smooths over before the browser sees it: sensor jitter, lift-off distance, cable intermittency that resolves within one sample, and acceleration curves applied by your driver. A clean result here means the mouse is reporting properly, not that every part of it is healthy.