Peripheral Test
Check your mouse and keyboard right in your browser: polling-rate estimate, button chattering detection, keyboard rollover and key-repeat rate — free, no sign-up, nothing leaves the page.
What this checks: Five input checks that run in your browser: a mouse polling-rate estimate (click the target as fast as you can), a button hold test for chattering, a scroll-step count, a keyboard rollover check on an on-screen key grid, and a key-repeat rate measurement.
Everything runs on this device in your browser. No sign-up, no tracking, no data stored — results live in this page only and are gone when you leave it.
Browser-level estimate — HID/USB sampling and event-queue limits apply; use vendor tools where precision matters. Every number on this page is a coarse signal from the browser's event stream, not a calibrated measurement.
Mouse
Polling rate (Hz) estimate
Click the target as fast and steadily as you can, then stop. Click at least a few dozen times.
Browser-level estimate — HID/USB sampling and event-queue limits apply; use vendor tools where precision matters.
Button hold test (chattering)
Press and hold the primary button until you stop the test. One clean press should produce exactly one event.
Scroll step count
Scroll the mouse wheel a few times over the target, then stop.
Keyboard
N-key rollover check
Hold as many keys as you can at the same time — e.g. hold W plus the key above and below it, or a whole row. The grid lights up what your browser receives.
Same-machine key-grid check, not an NKRO certification: the browser only reports what it receives, and the keyboard itself may still ghost.
Key-repeat rate
Hold any single key until you stop the test.
Help
How to use this check-up
- Mouse — polling: click the target as fast and evenly as you can, then Stop. The app snaps the median event gap to common polling rates and shows a confidence level.
- Mouse — chatter: hold the primary button through the test. Any extra event while you held still is a suspect chattering event.
- Mouse — scroll: rotate the wheel a few times, then Stop, and compare the count to what you expect from your OS settings.
- Keyboard — rollover: hold several keys at once (two fingers on one key is the classic ghosting test) and watch the grid and the peak count.
- Keyboard — repeat: hold one key for a few seconds and read the repeat rate your OS delivers.
Good to know
- Fast mice merge events: above roughly 1000 Hz the browser and OS start coalescing events, so the estimate plateaus — a low 'Hz' reading on a fast mouse is usually the event queue, not the hardware.
- For a definitive polling rate, use the vendor's software or a dedicated hardware analyzer; this page is a sanity check, not a calibration.
- NKRO depends on the keyboard's controller and the OS driver stack. If the browser shows fewer keys than you pressed, the grid tells you what reached the browser — check the vendor tool next.
Use cases
When people reach for this check-up:
-
Unboxing a new mouse
Verify the advertised polling rate ballpark and that both buttons register cleanly before you need the vendor software.
-
A mouse that double-clicks on its own
Run the hold test: if one pressed-and-held button fires extra events, that is the chattering signature — RMA or replace.
-
Choosing between two keyboards
Press rows and multi-key chords on the key grid and compare what the browser receives from each keyboard, on the same machine.
-
New keyboard setup
Confirm keys register, nothing ghosts in simple two-finger holds, and the key-repeat rate matches your OS settings before your first day of use.
-
Pre-purchase sanity check for a used device
Run a quick browser check before committing to a used mouse or keyboard: registration, chatter and rollover all show up in a few minutes, with no software install.
Frequently asked questions
- How accurate is the polling-rate estimate?
- It is a browser-level estimate, not a calibrated measurement. Browsers coalesce fast input events and the OS applies its own sampling limits, so above roughly 1000 Hz the reading tends to plateau. Use vendor tools where precision matters.
- Why does my fast mouse read as 500 Hz or 1000 Hz?
- When a mouse polls faster than the display refresh and OS event loop can keep up, the browser merges events. A lower reading than the spec is usually the event queue, not the hardware — treat it as a rough signal.
- Does the chattering test break the button if it's bad?
- No. The test only records click events while you hold the button; it cannot add wear. If you detect chatter, a dedicated chattering test tool and the manufacturer's warranty are the next steps.
- Is the NKRO result a certification?
- No. It is a same-machine key-grid check: the browser reports the keys it receives, and the keyboard's controller or driver may still ghost under full-load scenarios. Vendor tools decide the real NKRO number.
- Is anything uploaded or stored?
- No. Everything runs on this device in your browser: no sign-up, no tracking, no data stored. Results live in the page only and disappear when you leave it.