Skip to main content

Tool dossier

keyboard-testerVerified

About

Updated 2026-10-17

Online keyboard test for dead keys, ghosting, N-key rollover, chatter and double typing. Test laptop and mechanical keyboards, and flag keys that stop working.

Related Tools06

Mouse Test Online: Double Click, Scroll, Polling Rate & CPS

Mouse DPI Test: Measure Real DPI & Verify Sensitivity

Gamepad Tester: Xbox, PS5, PS4 Buttons & Polling Rate

FPS Mouse Sensitivity Converter & Calculator: CS2, WARDOGS

Stick Drift Test: PS5, Xbox & Joy-Con Drift, Dead Zone

Reaction Time Test: Click & Measure Response Speed Online

Features12

  • Full key sweep: press each key and the on-screen keyboard lights up in real time, so any key that stays dark is a dead key, and a full pass takes about a minute
  • NKRO tracking: shows how many keys are held at once plus the session maximum, so you can tell 6KRO from true N-key rollover at a glance
  • Chatter detection: an adjustable 10-150 ms threshold flags repeated keydowns of the same key in orange, separating switch bounce from OS key repeat
  • Break contact detection: catches a keydown with no matching keyup during a hold and highlights it purple, so a 5 second long press exposes it
  • Ghosting check: hold several keys together and watch the counter for drops, separating matrix limits from a single failed switch
  • Long-press waveform: a rainbow strip inside the keyboard case shows hold duration at about 60fps and breaks visibly on release
  • Key event log: millisecond timestamps, event type, hold duration and anomaly labels, clearable in one click
  • Extra key detection: media, brightness and other non-standard keys appear as green keycap badges so you can verify those too
  • Win/Mac layout toggle: relabels Win/Alt/Ctrl as Cmd/Opt/Ctrl so Mac users can match their keycaps
  • Typing practice: Free mode measures natural speed, Words mode compares against your own text or a .txt file, with live WPM and per-character accuracy
  • Combo tracking: hold 2 or more keys and their names appear live with a combo history, verifying the keyboard matrix
  • Debounce analysis: shows the millisecond gap between repeat events of the same key, for tuning a mechanical keyboard's debounce setting

How to Use10

  1. Click anywhere on the page to focus the tester, then press any key; the on-screen keyboard lights up in real time.
  2. Sweep every key once: a lit key means it registered, and a full pass finds every dead key in about a minute on both laptop and mechanical keyboards.
  3. Test rollover: hold W, A, S, D, Shift and Space together and watch the counter; if the count drops while keys are still held, you have hit the rollover limit.
  4. Test for chatter: press one key slowly, once per press, and read the event log; a single press that logs two keydowns is chatter. Adjust sensitivity with the threshold slider (10-150 ms).
  5. Test for break contact: hold a key still for 5 seconds; an extra keyup or a re-press in the log is break contact.
  6. Test for ghosting: hold several keys at once and see which ones fail to light up; those positions are matrix conflicts.
  7. Read the key log: the bottom panel timestamps every press and release with hold duration and anomaly labels. Use Clear to reset it.
  8. Measure typing speed: pick Free mode for natural typing, or Words mode to load a .txt passage and watch WPM and accuracy update live.
  9. Switch layout: the Win/Mac toggle relabels the modifier keys as Cmd/Opt/Ctrl so they match a Mac keycap.
  10. Switch to an English input method before testing: an active IME intercepts keystrokes and makes the tester look unresponsive.

Frequently Asked Questions20

How do I use an online keyboard test, and how do I confirm every key works?

Click the page to focus the tester, then press each key and watch it light up on screen. A lit key means that key registered, and a full sweep takes about a minute; every key that stays dark is a dead key. Switch to an English input method first, because an active IME intercepts keystrokes.

Some keys on my keyboard stopped working. How do I find the cause?

Work from software to hardware, in four steps. First switch to an English input method and retest, to rule out IME interception; second, move the keyboard to another USB port and reseat it, to rule out contact problems; third, retest on another device or browser to see whether the fault follows the keyboard; fourth, plug in an external keyboard to separate a faulty laptop keyboard from a system-wide input problem. If the key stays dark after a port and device change, the switch or its ribbon cable is the likely fault.

The keyboard tester shows nothing when I press keys. Why?

In most cases the page does not have focus, or an input method is intercepting the keystrokes. Click an empty part of the page and press a key again; if nothing happens, switch to an English input method (Win+Space on Windows, Control+Space on macOS) and retest. A third cause is a background browser window: key events are not delivered to a page that is not in the foreground, so click the window title bar first.

How do I test for keyboard chatter, and how do I fix it?

Press one key slowly, once per press, and read the event log. A single press should log exactly one keydown; two or more keydowns with no keyup in between is chatter, also called switch bounce. The threshold window is 10-150 ms, and a lower value is stricter. To fix it, clean the switch with contact cleaner and click it rapidly 30-50 times; if it still repeats, the metal dome is worn, so a soldered keyboard needs a switch replacement, while a gaming keyboard may let you raise its debounce delay to 4-8 ms in the vendor software.

How do I test for break contact, and does replacing the switch fix it?

Hold a key for 5 seconds or more and watch the event log. A healthy hold logs exactly one keydown and one keyup; an extra keyup or a re-press in between is break contact. Break contact means the switch contacts are not closing cleanly, and replacing the switch does fix it, but only on a hot-swap socket; on a soldered board the switch has to be desoldered, so try contact cleaner first.

How do I test N-key rollover, and how do I know if my keyboard is full NKRO?

Hold W, A, S, D, Shift and Space together and read the rollover counter. A steady 6 means full NKRO; a count that falls below 6 while the keys are still held means 6KRO, which is the standard USB keyboard limit. Some keyboards offer full NKRO only in wired mode and drop back to 6KRO over wireless, so test both.

What is the difference between keyboard ghosting and NKRO?

Ghosting drops inputs; NKRO registers them all. Ghosting is the keyboard matrix deliberately discarding signals to avoid false ones, so some keys fail to light up when several are pressed together, while NKRO (N-key rollover) means every key registers no matter how many are held. You test both the same way: hold several keys at once and look for a key that stays dark or a counter that drops. Membrane keyboards ghost routinely, while most mechanical and gaming keyboards support full NKRO or at least 6KRO without ghosting.

What keyboard latency is normal, and can a browser measure it?

A browser cannot measure real keyboard latency, and no browser readout in milliseconds tells you whether your keyboard is normal. Latency measurement needs a physical reference point: you have to film the keycap press and the character appearing on screen with a high-speed camera and compare the frames, which a browser has no access to. What this page does measure is whether every key registers, the rollover limit, chatter and break contact, which are the most common keyboard faults. For latency you need dedicated hardware or a vendor tool.

Can this tester measure keyboard polling rate?

No. A mouse reports its position continuously, so coalesced pointer events reveal its report interval, but a keyboard only sends a report when a key changes, and the event stream carries no heartbeat, so the browser cannot derive a rate. On Windows the keyboard is also claimed by the system's exclusive driver, so WebHID cannot open it either. Polling rate (commonly 125-8000 Hz) only affects how fast the device itself processes input; browser key events arrive after the OS and driver have done their work.

Why do some keys not show up in the keyboard test?

Fn, some media keys and system shortcuts never reach the browser, and that is a browser limitation rather than a keyboard fault. Fn is handled by keyboard firmware and is not forwarded by any operating system; media keys such as volume and brightness are often intercepted at the laptop driver level; and Ctrl+Alt+Del, Win+L and Alt+Tab are claimed by the OS first. The test: if the key works in normal typing but not here, it is being intercepted; if it fails in normal typing too, suspect the switch or its connection.

How do I test a laptop keyboard?

Exactly like an external keyboard: click the page to focus the tester, then press each key and watch the on-screen layout light up. Laptop keyboards usually have no number pad and compressed arrow keys, but all standard keys register normally. Two notes: Fn combinations such as Fn plus brightness are handled by firmware and will not appear, and some laptops respond more slowly on battery power, so retest while plugged in for a cleaner result.

How do I test a mechanical keyboard for dead or failing switches?

Mechanical keyboards have an independent switch per key, which makes them ideal for key-by-key testing. Sweep the on-screen keyboard to find keys that stay dark, then retest each suspect key: press it slowly, once per press, and a single press should log exactly one keydown, with anything extra being chatter. On a hot-swap board you can swap in a known-good switch of the same type; if the key recovers, the original switch was faulty, and if it stays dark, the socket or PCB trace is at fault.

How do I test a TKL or 60% keyboard?

On a compact keyboard you only test the keys that physically exist, and you do not need to fill 104 positions. A TKL (87-key) board has no number pad, and a 60% layout also drops the dedicated arrow cluster and the function row, so those positions staying dark is normal, not a fault. Focus on whether every key you do have registers, and whether the Fn layer, for example Fn plus a number key firing an F-row key, triggers correctly.

Can I measure my typing speed with this tester?

Yes, through the typing practice section, which tracks words per minute, accuracy and character-level errors live. Free mode measures your natural typing, while Words mode compares you against your own text or a .txt passage for precise accuracy. It is a browser-based estimate that is useful for tracking your own progress, not a certified typing exam; for formal testing use a dedicated typing test site.

Where can I see a keyboard layout diagram, and why does the on-screen keyboard not match mine?

The on-screen keyboard in the test area is an interactive layout diagram. Keys light up as you press them, so you can use it to locate a position directly. If it does not match your physical keyboard, the cause is usually layout differences: laptops are often 87 keys or fewer with compressed arrows, and a Mac keyboard places Cmd and Opt where Win and Alt sit on a PC board. Use the Win/Mac toggle to relabel the modifiers so they match, though key positions themselves do not move.

Why do different keyboard test websites give different results?

Different sites read the same data but use different implementation details. Those are whether the page recognises e.repeat events, what chatter threshold it uses, and whether it handles coalesced events. The same keyboard producing different chatter results across sites usually means the thresholds differ, not that the keyboard changed. Firmware-level questions such as polling rate and exact matrix behaviour need a desktop tool like Switch Hitter on Windows, since a browser cannot reach them.

Does this tester work with Hall effect or magnetic switch keyboards?

Key events work, magnetic-specific settings do not. A Hall effect keyboard sends standard key events to the browser, so key registration, chatter, break contact and N-key rollover all test normally here. Actuation point (magnetic switches go down to 0.1 mm versus about 2 mm on a mechanical switch) and Rapid Trigger live in the vendor driver, for example Wooting or DrunkDeer, and the browser cannot read them. An over-aggressive Rapid Trigger setting does show up in the event log as extra keydowns or keydowns with no keyup, the same signature as chatter, so you can check whether your parameters behave sanely.

Can I test a MacBook or Mac keyboard with this tool?

Yes, as long as macOS delivers the key events to the browser, which covers built-in MacBook keyboards, Magic Keyboards and external PC keyboards. Use the Win/Mac layout toggle to relabel Win/Alt/Ctrl as Cmd/Opt/Ctrl. Two honest caveats: Fn is handled by firmware and never reaches the browser, and macOS intercepts shortcuts such as Cmd+Tab and Cmd+Space before the page sees them, which does not mean those keys are broken.

My keyboard repeats characters when I hold a key. Is it hardware or a system setting?

Both can cause it, so check the system first. On Windows, go to Settings, then Bluetooth and devices, then Keyboard, and look at the character repeat delay and repeat rate; a very short delay with a fast repeat rate means a held key repeats quickly by design, not because of a fault. To rule out hardware, press the key slowly, once per press: a single press should produce exactly one keydown, and a second keydown with no keyup between them means the switch is chattering. Test each suspect key three times; any repeat without a release points to switch wear.

How is chatter detection different from the OS key repeat setting?

System repeat is software sending signals by design; chatter is one press misread by the switch as several. This tool ignores OS repeat events (e.repeat) entirely and instead measures the gap between consecutive non-repeat keydowns of the same physical key, flagging anything below the threshold window (adjustable 10-150 ms) as switch bounce, a hardware fault.