DevicedialKeyboard, mouse and reflex tests
The input chain from key press to pixel, measured one link at a time — how many keys register at once, how often the mouse reports, how fast you actually are, and where the milliseconds go.
Aim trainer
Speed and accuracy trade against each other, and target size moves the score more than either.
CPS test
Most people are slower than they assume — and the part they blame on the mouse usually is not the mouse.
Double click test
One press, two clicks. There is a number that separates a worn switch from a fast finger.
Keyboard tester
A cheap board gives up at two keys; the usual USB report carries six. Where does yours stop?
Mouse tester
Five buttons, two wheel axes, and one number the browser is not allowed to know.
Mouse polling rate test
1 ms between reports is 1000 Hz. Whether your browser can even see that is the interesting part.
Reaction time test
Part of every reaction time on the internet is the monitor. This one says how much of yours.
Typing speed test
Speed without accuracy is slower than it looks. See both numbers from one 60-second run.
Everything in this group sits on the same chain: a switch closes, a device reports it, the browser is told, and a pixel changes. When something feels wrong — a missed key, a stutter, a click that lands twice — the useful question is which link is responsible, and each page here isolates one of them.
Start with the hardware. The keyboard tester lights up what actually registers, including the three-keys-then-a-fourth case that exposes ghosting, and the mouse tester does the same for buttons, wheel and movement. The double click test is the one worth running on any mouse over a year old: it counts the double reports that a worn switch sends for a single press, in milliseconds, so the verdict is a measurement rather than a suspicion. The polling rate test then asks how often that mouse reports at all.
Then measure yourself. The typing test gives words per minute and accuracy on the standard five-character definition of a word. The reaction time test reports milliseconds and keeps every attempt, because one lucky run is not a reaction time. The CPS test counts clicks inside a window you choose, and the aim trainer puts hits per minute and accuracy together, which is the pair that actually improves — either alone is easy to fake by moving slower or by spraying.
Frequently asked questions
Why does my keyboard drop a key when I hold three down?
That is ghosting, and it is a wiring limit rather than a fault. Membrane keyboards share rows and columns across a matrix, so certain combinations become ambiguous and the controller discards one press rather than invent a keystroke you did not make. Keyboards sold as n-key rollover give every key its own path and never do it. The keyboard tester shows it live: hold three keys, add a fourth, and watch which one fails to light up.
My mouse clicks twice when I click once. Is it broken?
Probably, and the double click test will say so with a number. A worn micro-switch bounces as it closes, and the two falling edges arrive so close together that the operating system reports two clicks — that is switch chatter, and under about 50 ms apart it is mechanical rather than human. Occasional chatter can sometimes be cleaned or contact-treated away; steady chatter means the switch is at the end of its life.
Why does my 1,000 Hz mouse measure 500 Hz here?
Usually the port or the cable rather than the mouse. A hub, a long extension or a USB port sharing bandwidth with something busy will halve the effective report rate, and some wireless receivers negotiate down when the link is noisy. Browsers also coalesce pointer events, so the browser reading is a floor rather than a ceiling — if the page says 1,000 Hz you certainly have it, and if it says 500 the next thing to try is a different port straight into the machine.
Can this page tell me my mouse DPI?
No, and any page that claims to is estimating. The browser is handed movement in pixels after the operating system has applied its own pointer speed and acceleration, so the raw counts per inch never reach it. What can be measured is the ratio between physical movement and on-screen movement, which is enough to compare two settings on the same machine but is not a DPI figure and is not treated as one here.
What counts as a good reaction time?
For a visual cue and a click, most people land between 200 and 300 ms, with practised players around 150 to 200 ms. Averages matter more than a best run, because a single fast attempt is often an early guess rather than a reaction — the test keeps the whole set so you can see the spread. Bear in mind that a slow display adds its own delay to your figure: at 60 Hz a frame is 16.7 ms, so some of what you measure is the monitor rather than you.
How the test works
requestAnimationFrame and
event timestamps — which means a slow display contributes to your figure, and the page names that
contribution instead of hiding it.Sources: