Click Speed Test

This page is a click speed, or CPS, test: it measures how many valid clicks or space-key presses you can make within a fixed time window. It is mainly used by gamers, mouse enthusiasts, and anyone curious about short-burst hand speed. The result is clicks per second, total clicks, duration, and recent records, which makes it easy to compare several rounds of practice. It is not a typing words-per-minute test and does not evaluate text accuracy. CPS can be affected by mouse switch feel, debounce behavior, browser event handling, keyboard use, fatigue, and test duration. Very high numbers may reflect hardware or input quirks, so compare results under the same settings rather than treating one score as an absolute ability measure.

CPS (clicks per second) is defined as `total_clicks / duration_seconds`. The page measures time with `performance.now()` rather than `Date.now()` because `performance.now()` returns a monotonic high-resolution timestamp in fractional milliseconds that is not affected by NTP corrections or the user changing the system clock mid-test. The duration is captured once at start and once at the natural end of the window, so a 10-second test always divides by 10.000 rather than by a drifting wall-clock difference. Click capture binds to `mousedown` (or `pointerdown`) instead of `click` because `mousedown` fires before the browser's 250-300 ms double-click delay and lands on the first physical button press, which is what hardware-style CPS expects. Each event records `event.timeStamp` and increments the counter. Keyboard parity uses `keydown` with `event.repeat === false` so a held-down Space key cannot generate auto-repeat events at the OS key-repeat rate (typically 30 Hz on Windows, configurable on macOS) and falsely inflate the score. Touch input uses `touchstart` with `event.preventDefault()` to suppress the synthetic `mousedown` that would otherwise double-count on mobile. The physical ceiling sits well below where most people imagine: ordinary single-clicking on a consumer mouse plateaus at 6-10 CPS, primarily because the mechanical switch debounce window (Omron D2F: ~8 ms, Kailh GM: ~3-5 ms) caps how fast successive presses can register. Jitter clicking (locking the forearm and tensing the click finger) reaches 12-16 CPS, butterfly clicking (alternating two fingers on one button) hits 16-22 CPS, and drag clicking (frictioning the finger across the button) can spike above 30 CPS but requires hardware that does not debounce aggressively. Records persist via `localStorage.setItem('cps_history', JSON.stringify(records))`, keeping the last 50 entries scoped to the origin.

• Timing uses `performance.now()` (monotonic, sub-millisecond), not `Date.now()` which can jump when the system clock is corrected.
• Click listener is `mousedown` (or `pointerdown`), not `click`, to avoid the 250-300 ms double-click delay and count the first physical press.
• Keyboard parity uses `keydown` with `event.repeat === false` to ignore OS auto-repeat events (≈30 Hz on Windows) from a held key.
• Mouse switch debounce (Omron D2F ≈ 8 ms, Kailh GM ≈ 3-5 ms) is the usual hardware ceiling on raw single-click CPS.
• Touch input handles `touchstart` with `preventDefault()` to suppress the synthetic 300 ms-delayed mousedown on mobile browsers.
• Typical bands: normal click 6-10 CPS, jitter 12-16, butterfly 16-22, drag 25+; records persist to `localStorage` (no network).
• CPS values are only comparable at identical durations; a 5 s burst usually scores 15-30% higher than a 30 s endurance run because fatigue dominates the longer window.