Interval Timer

Built to be heard, not watched

This is the one tool here that is normally used with the screen out of reach and often out of sight: propped against a water bottle at the end of a mat, on a bench across the room, face up on the floor while you are above it doing something else. Every design decision follows from that.

The last three seconds of every phase get a tick each, so a change arrives as a countdown you hear rather than a moment you have to catch. The change itself has its own sound, and the two directions are different: a rising figure going into work, a falling one going into rest. After a round or two you stop needing to check which is which.

When you do look, the phase name is the largest thing on the screen after the number, and the colour carries it independently: vermilion for work, teal for rest and for the get ready phase. The bar underneath fills across the current phase, which answers how much of this is left faster than reading digits does. The line under the number says what is coming next, or that this is the last round, which is the thing you most want to know halfway through a set.

Setting it up

Four numbers: how long to get ready, how long each work phase runs, how long each rest runs, and how many rounds. They are saved in your own browser, so the setup you use is there next time.

For Tabata, the protocol from Izumi Tabata's 1996 study is twenty seconds of work and ten seconds of rest, eight times, which is four minutes. It was originally tested on stationary bikes at an intensity most people would not recognise as sustainable, which is worth knowing before treating four minutes as an easy workout.

For general HIIT, forty seconds on and twenty off across eight to twelve rounds is a common starting point and is the default here. For circuits with stations, set the work phase to your station length and the rest to however long it takes to move, which is usually more than you think. For EMOM style work, set work to sixty and rest to zero, and use the round count as your minute count.

Rest can be zero. A rest of zero is skipped entirely rather than appearing for a single frame, so the timer runs work straight into work with a round change between them.

Get ready, skip, and reset

The get ready phase exists because the gap between pressing start and being in position is real, and a timer that begins the first work round while you are still walking back has already cost you the round. Ten seconds by default. Set it to zero if you start from the phone.

Skip moves to the next phase immediately, which is what you want when a round is interrupted or when you finish early and would rather not stand waiting. Reset takes the whole workout back to the start. On a desktop the space bar starts and pauses and R resets.

Why the timing does not drift

Across a long workout, small errors add up. A timer that counted by adding to a total on every tick would be visibly wrong by the end of a twenty round session, because the browser is entitled to slow those ticks down whenever it wants to save power.

This one works from deadlines instead. Each phase records the moment it should end against your system clock, and every reading is the difference between that moment and now. A phase cannot come out long because the browser was busy, and the workout cannot end late because of an accumulation of small delays. It also means the timer keeps running correctly if you switch to music or a message mid session, which on a phone is not an edge case, it is what happens.

When another tool fits better

For a single stretch with no rounds, use the timer. For focus sessions rather than physical effort, the pomodoro timer is the same idea at a different scale and keeps count of the cycles. If you want to know how long something took rather than to be driven through it, the stopwatch counts up and records laps.