whimera

Keyboard Test · Keyboard Inspector

Verify every key press.

Test key registration, rollover, chatter, hold stability, and timing in one session.

Keyboard layout

Press each key once to complete the coverage check.

104-keyWindowsCoverage 0/104 · 0%
UntestedDetectedActiveSuspectFailed

Live status

Last event

Waiting for input

code / key / location

Response timing

N/A

Avg N/AP95 N/AJitter N/A

Multi-key

Held now

0

No keys held

Peak

0

Maximum simultaneous inputs.

Rollover

Not tested

Hold at least two printable keys together.

Diagnostics

Chatter

Monitoring

Confirms after 3 same-key re-presses ≤20ms within 8s.

Hold stability

Waiting for input

About Keyboard Test

A browser-level view of keyboard behavior.

Keyboard Test maps the keydown and keyup events delivered by your browser to a 104-key layout. It exposes event values, hold transitions, simultaneous input, chatter suspects, and browser response timing in one local session. It does not read keyboard firmware, USB reports, hardware scan rate, or physical switch latency.

Runs locally in your browser. No account required.

What Keyboard Test measures

  • 104-key coverage with live states plus key, code, location, and event type

  • Current, average, P95, jitter, and recent event-to-frame timing

  • Event-driven 0/1 hold waveform and confirmed chatter suspect detection

  • Held count, session peak, and observed KRO or NKRO display

How to read the results

  1. Use the keyboard map to confirm coverage, then compare Last event with the physical key you pressed.

  2. Hold keys and try deliberate combinations to inspect the 0/1 hold signal, simultaneous peak, and displayed rollover class.

  3. Repeat any suspicious input, compare timing only on the same setup, and reset the session before testing another keyboard.

Technical notes

What is the difference between key and code?

Key is the interpreted value after the active layout and modifiers are applied, for example, a or A. Code identifies the browser's standardized physical key position, such as KeyA, regardless of the printed character.

What does location tell me?

Location distinguishes Standard, Left, Right, and Numpad variants. It is especially useful for paired modifiers such as Shift, Control, Alt, and Meta, plus keys on the numeric keypad.

What do KRO and NKRO mean in Keyboard Test?

KRO is the largest group of printable keys observed together in this session. Keyboard Test shows an exact value from 2KRO to 6KRO, then labels a result above six as NKRO. This is a browser observation, not a certification of the keyboard's USB protocol or every possible key combination.

How is chatter identified?

Keyboard Test treats a same-key re-press within 20ms as a candidate, then marks the key as suspect only after three candidates occur within eight seconds. It uses each keyboard event's timestamp and ignores held-key auto-repeat, reducing false positives from intentional fast input and delayed event handling.

Is the hold waveform sampled?

No. It is event-driven: keydown moves the signal to 1 and the final keyup returns it to 0. The horizontal axis is elapsed browser time, so the trace scrolls continuously without inventing intermediate samples.

Is Response timing the keyboard's hardware latency?

No. It measures from the browser receiving a keydown to a later animation frame. Browser scheduling, display refresh, system load, and the input path all affect the result, so it is most useful for comparison within the same setup.

Does keyboard test data leave my device?

No test data is uploaded. The session stays in memory in this browser tab and is cleared when you reload the page or choose Reset session. Avoid entering passwords or other sensitive text into any keyboard tester.