Input toolsPERIPHERAL TEST WORKBENCH
Keyboard test
Every key, combination and hold — with the input behind it.
Keyboard test area
104 physical positionsPress and release keys to test — detection starts automatically across this page. Controls keep their normal keyboard behavior. Escape pauses capture after release.
Mac extended uses the existing demo layout, not a specific Apple product. Exclude positions your keyboard does not have.
MEASUREMENTS
BROWSER INPUTHELD NOW
TEST PROCEDURE
Press and release each target key. Coverage counts complete pairs, not repeats.
0 / 104 target keys observed
The target checklist below lets you exclude missing or browser-reserved keys. Exclusions are explicit and appear in your report.
Choose the keys first, then hold them together. A missing key is only missing from the best observed simultaneous set.
Press once at a time, slowly. OS auto-repeat is counted separately. Short gaps are candidates to reproduce, not a switch fault diagnosis.
Arm the test, hold the selected key, then release after the target. No repeat events is normal for modifier keys.
Target checklistPress and release every included key
Checkbox = included in coverage. Status distinguishes Held, Observed, Not seen and Excluded. Fn is usually handled inside the keyboard and cannot be tested here.
Simultaneous key timeline
Latest 12 seconds · no smoothingEvent register
LATEST 8 EVENTS · MEMORY ONLY| Time (s) | Event | Code | Location / flags |
|---|---|---|---|
| Press and release a key on your physical keyboard to begin. | |||
Optional raw event export
Raw key codes and timing can reveal what was typed. The JSON report contains summaries, without the ordered raw event log. CSV is an explicit opt-in for the retained events below.
Coverage confirms browser-received down/up pairs only. It does not certify hardware health. OS shortcuts, Fn and some media keys may not reach the page.
How to test & what the numbers mean
Start typing to test
Detection starts as soon as this page has focus. Use your physical keyboard anywhere outside controls; no activation click is needed. Dropdowns, checkboxes, buttons and editable fields keep their normal keyboard behavior. Switching to a control, another window or a hidden page interrupts active attempts and clears held states without inventing keyup events. Detection resumes automatically when this page regains focus outside controls; interrupted attempts must be started again.
Tab, Space and arrow defaults are suppressed during capture so these positions can be tested. Escape is recorded, then pauses capture so you can navigate controls with the keyboard. Click Resume capture to continue; manual pauses remain in effect when you switch windows. You can also use Pause capture at any time.
Read the physical positions
code identifies a physical position, location distinguishes left/right modifiers and the numpad. US keycap legends are a reference; a different character language or IME can produce different characters. Composition events are flagged, without recording typed text.
Choose ANSI/ISO, full size/TKL and Windows/Mac labels to match your device. The Mac extended model is illustrative. Missing special keys can be excluded from coverage; a key not seen by the browser is not automatically broken.
Interpret the evidence
Observed means a complete down/up pair. The peak simultaneous count only describes combinations tried. A key absent from a chosen combination may be intercepted by the OS. It cannot prove a hardware rollover limit.
Hold stability reports browser-visible release timing and repeat events, not electrical continuity. Auto-repeat depends on system settings. Fast release-to-press sequences warrant a guided retest; this page cannot measure hardware latency, polling rate or switch travel.