Chronometer / Stopwatch

Math

Stopwatch with start, pause, lap and reset at millisecond precision for timing tasks, workouts and code execution, with records that survive tab reloads.

Stopwatch Ready
00:00:00.00

About Chronometer / Stopwatch

When you want to time a process, a brew, or a workout set without installing an app, the browser stopwatch is the handiest option. This tool is a high-precision web stopwatch that displays down to centiseconds (hundredths of a second), with start, pause, resume, and reset, plus lap marking while running. Each lap interval is computed automatically, with the fastest and slowest laps highlighted to compare pacing. Timing advances purely from local timestamps, so it works offline, uploads nothing, and keeps running until you refresh the page. One misuse is treating the result as exact: browsers reduce timer precision for backgrounded pages, so timing that accumulates short-interval callbacks drifts and can be noticeably off when the page returns, which makes periodic resync necessary for long measurements. The second is confusing splits with cumulative time — splits alone cannot rebuild a total, since one missed mark breaks the chain, while a running total cannot be decomposed, so keep both. Browser timing suits measurements a person observes, such as how long an operation took or pacing a talk; it does not suit high-precision benchmarking, which needs a monotonic clock and must account for timer precision limits. If results feed a performance comparison, fix the conditions, meaning the same device and load with no background tabs, or the numbers are not comparable. Timing is also lost when the page closes, so anything worth keeping must be recorded externally. Fix three things when results feed a comparison: the same device, the same load, and the page in the foreground with no competing tabs. Varying any one of them makes the figures incomparable, and the deviation is invisible in the numbers themselves. For long-term records, store the timestamp with each result so an anomaly later can be attributed either to changed conditions or to a genuinely slower target. A final premise is the difference between elapsed and actual work time: measuring a user action includes time when the person steps away or switches tabs, whereas measuring computation should take the difference inside the same synchronous block to avoid timer precision effects. Deciding which is being measured avoids results that are large for no clear reason and makes anomalies easier to explain. Compared with instrumenting code, browser timing wins by not touching the subject: operation duration can be measured without adding hooks, which suits observing an existing third-party page or a manual process. For long-term collection and comparison, instrumentation is more reliable because it fixes the sampling points, removes human variation and aggregates centrally.

How It Works Internally

Timing that accumulates short-interval callbacks drifts, because browsers reduce timer precision for backgrounded pages. Compute elapsed time from the difference between the start instant and the current instant, and resync when the page returns to the foreground. Splits and cumulative time are different measures: the first evaluates one stage, the second the whole run.

InputCumulative 10 minutes with a last split of 3
OutputTotal 10 minutes of which the last stage is 3; splits alone cannot rebuild a total, since one missed mark breaks the chain

Measured on This Page

Time baseTiming uses Date.now() deltas rather than accumulating interval ticks, so browser throttling of background tabs cannot lose time
Lap mathLaps of 1.64 s and 1.14 s matched the rhythm; each lap is current total minus previous total
Fastest and slowestWith more than one lap the fastest and slowest are tagged automatically
PauseThe display held still for a full second while paused and resumed from the same point
Fixed hereFive lap-area strings (N lap(s), Lap N, total, fastest, slowest) were hard-coded English even on the Chinese page; now on i18n

In the measured run the two laps were 1.64 s and 1.14 s and the display held still for a full paused second. Laps are stored as cumulative totals, not per-lap values, and the per-lap figure is derived by subtracting the previous total — the two representations cannot be swapped: storing only per-lap values makes the total at an arbitrary moment unrecoverable, while storing only totals makes laps unreadable. This page stores totals and derives laps, which serves both readings.

How this was measured: Drove the page on a fixed rhythm: start, a lap at about 1.5 s, another at about 1.0 s, then pause; read the main display and both lap entries, then confirmed the display freezes while paused and resets to zero. · Measured 2026-10-29

How to Use

  1. Open "Chronometer / Stopwatch"
  2. Click "Start" to begin timing
  3. Click "Lap" to record each split while timing continues
  4. Click "Stop" to finish, then compare the fastest and slowest laps
  5. Click "Reset" to clear the timer for the next run

Use Cases

  • Process timing — Measure how long a manual operation actually takes from start to finish.
  • Interval training — Use laps to record each set during a workout and compare fastest to slowest.
  • Pour-over coffee — Control pour and bloom seconds precisely to reproduce a steady brew curve.
  • Talk pacing — Mark sections while rehearsing to learn how long each segment runs.
  • Smoke checks — Roughly record an operation response time to spot obvious lag.
  • Reaction timing — Use the lap feature to measure user reaction times or frontend interaction latencies during UX testing.
  • Pomodoro tracking — Time work sessions and breaks with the stopwatch to track focus intervals throughout the day.

FAQ

How precise is the timing?

It shows centiseconds (0.01s). Lap and total times come from system timestamps, and the display refreshes about every 31 milliseconds, smooth enough for the eye.

Does it keep counting in the background or when locked?

The total is based on real timestamps, so even if the tab is backgrounded or the timer is throttled, it shows the correct total on return without undercounting.

Does a lap record split time or total time?

Tapping lap stores the elapsed total at that moment, and the tool derives each lap interval from those, then flags the fastest and slowest laps.

Can I resume after pausing?

Yes. Pause keeps the current elapsed time, and starting again continues from that point; reset clears everything including all recorded laps.

Are records kept after closing or refreshing?

No. The stopwatch state lives only in the current page session and clears on refresh or close. Note times manually before resetting if you need to keep them.

Timing drifts once the page goes to the background. Why?

Browsers reduce timer precision for backgrounded pages to save resources, so a timer that accumulates short-interval callbacks drifts steadily. Compute elapsed time from the difference between the start instant and the current instant rather than summing intervals, resync against system time periodically for long measurements, and re-check when the page returns to the foreground so you do not keep recording with the offset.

What is the difference between lap splits and cumulative time?

Cumulative time is the total from the start, while a split is the interval between two consecutive marks, and they answer different questions: splits evaluate a single stage, cumulative shows the whole run. Keep both, because splits alone do not reliably reconstruct a total (one missed mark breaks the chain) and a running total cannot be decomposed back into stages, leaving nothing to check against afterwards.

Does the timer stay accurate when I switch tabs?

Browsers throttle or suspend background tabs, but a timer based on timestamp deltas (not accumulating intervals) stays accurate when you return. Test it: leave for a minute, come back and compare with the system clock. setInterval-based implementations drift.