Buttons & chords
Choose Buttons, then Start checker. Press each button on the active pad and release it. Try a chord, such as holding Primary while pressing Secondary. Separate mousedown and mouseup observations retain each button transition; a mouse pointer's pointerdown/pointerup alone do not describe every change in a chord.
Primary, Middle, Secondary, Back and Forward are standard browser button roles, not guaranteed physical labels. OS mappings can swap them. Some side buttons are reserved for navigation, remapped to keys, or intercepted before the browser. Their absence here is inconclusive. Additional vendor buttons are not detected or fingerprinted.
One intended click, an explicit boundary
- Choose Repeated clicks and the button you intend to use. Begin attempt focuses the pad and opens your boundary.
- Make one deliberate click, without intentionally clicking twice. Wait if needed; there is no automatic time threshold.
- Choose Finish attempt. Review down/up order, complete pairs, down→up holds, release→next down gaps and down→next down intervals. Repeat slowly in a fresh attempt if the sequence surprised you.
A browser dblclick gesture normally follows two click sequences under browser/OS gesture rules. It is shown separately, not counted as a switch fault. Multiple pairs inside your declared single-click attempt are an observation to investigate: extra finger actions, remapping, software and hardware may all contribute. The browser cannot distinguish these causes or recover electrical contact bounce filtered out by firmware.
Do not choose a universal 0, 2 or 4 ms setting from these intervals. Follow your exact model's documentation. Our mouse settings and click-latency guide explains why filter labels and measured delay are different things.
Hold continuity & dragging
Choose Hold & drag, select a button and Begin attempt. Press and keep holding while moving. Watch the held state, then release and Finish attempt. An up followed by another down during one intended hold is visible in the sequence, but the page cannot know whether your finger truly remained pressed.
Releasing outside the pad, within this window, closes the recorded hold. A missing release inferred from the browser button mask, pointer cancellation, focus loss, hidden tab or leaving the browser area while held stops collection and marks the unfinished attempt incomplete. Clearing held state on interruption is not a fabricated mouseup or a detected switch dropout. Returning does not resume automatically.
Scroll deltas, not wheel notches
Choose Wheel and Start checker, leaving the pad focused. Wheel events over that pad are kept local so the page stays still. Scrolling outside the pad or while stopped remains normal. Modified gestures, including Ctrl/Meta zoom and Shift scrolling, keep browser behavior and stop the checker.
The log reports deltaX, deltaY and deltaZ with the event's deltaMode: pixels, lines or pages. Unknown modes remain unknown. Positive Y is the browser's down delta, negative Y up; positive X is right, negative X left. These numbers are not physical notch counts, calibrated travel or a guarantee that content scrolled. Pixel, line and page magnitudes are never summed together.
Inspect a wheel reversal candidate
Slowly scroll in one intended direction, then inspect any sign reversal. Compare several deliberate trials. Zero deltas are not reversals, each axis is separate, and the baseline restarts after Stop, an interruption or a unit change.
Deliberate changes of direction, trackpad inertia and gesture processing can produce a reversing delta. A wheel encoder issue is only one possible explanation; the counter cannot diagnose one. No automated sequence on this page proves a physical defect or a repair.
Input, timing & privacy limits
Keyboard and touch operate the controls, but are not labeled physical mouse-button evidence. Mouse-class browser input can itself come from a trackpad, accessibility tool or automation. Script-dispatched samples are labeled in the log. Browser timestamps may be rounded, equal or out of order; unresolvable intervals display as unknown. None starts at physical switch actuation.
The checker stores only bounded observations in this mounted page. No uploads, persistence, device identifiers or third-party checker scripts. Clear session or leave this page to discard them. General site page analytics are separate from the checker; see privacy.
Continue with the right question
For you plus your setup, try the visual reaction test. For aiming practice, use Train · Arena. Neither isolates mouse hardware latency. Read input lag vs reaction time before comparing scores or replacing hardware.