여러 기기를 동시에 맞추는 타이머·스톱워치 활용법

When several devices must agree

One timer is enough for most things. But when the point is that everyone starts and ends at the same moment, separately pressed timers cause trouble.

• Classes and group work. When the teacher's screen and the students' screens show the same time, the "is time up?" questions stop.
• Group fitness. In an interval class, instructor and participants see the same work and rest signal so transitions line up.
• Presentations and meetings. Presenter, chair and audience share a sense of time and switch together when a stage changes.
• Relays and split timing. Start, handoff and finish points must record against the same start instant for splits to be meaningful.
• Study and remote work. Group Pomodoro: everyone focuses and rests on the same rhythm.

Why clocks disagree between devices

It is three effects stacked together.

1. Human timing differences

Even with "ready, set, go", reaction speed varies and one person cannot tap several devices at once. Whoever presses first gains or loses, which undermines fair comparison.

2. Clock offset

Device clocks are not identical. With automatic time off or long without correction, they can be seconds apart. Leaving "ring at 10:00:00" to each device's own clock makes them ring at different moments. Devices commonly reduce this by periodically syncing with a time server.

3. Network latency

When the host sends "start now", the message takes time to arrive: short on a local link, longer through an internet relay, and different each time. Starting on receipt makes each device start at a different instant.

How a clock offset is measured

NTP, the Network Time Protocol, uses four timestamps: T1, when the client sent its request (client clock); T2, when the server received it (server clock); T3, when the server replied (server clock); and T4, when the client received the reply (client clock). RFC 5905 defines:

• Clock offset: theta = 1/2 x [(T2 - T1) + (T3 - T4)]
• Round-trip delay: delta = (T4 - T1) - (T3 - T2)

A worked example. The client sends at 100.000 s; the server receives at 100.130 s on its clock and replies at 100.131 s; the client receives at 100.061 s on its own clock. Delay = (100.061 - 100.000) - (100.131 - 100.130) = 0.060 s and offset = 1/2 x [0.130 + 0.070] = 0.100 s. This client's clock is about 0.1 s behind the server. The calculation assumes the outbound and return delays are equal, so an asymmetric path leaves some error; real systems therefore measure several times and trust the sample with the shortest delay.

Once each device knows its own offset, an absolute instant such as "start at server time X" means the same thing to everyone.

Host-authoritative sync design

The key design choice is what you transmit.

• Streaming remaining time. If messages are late or lost, the display jitters.
• Sharing one end time. Tell everyone when it ends once; each device counts down to that instant on its corrected clock. A brief disconnection does not make the digits jump, and you resync only on events such as pause or added time.
• Start works the same way. Share "start at instant X" instead of "start now" so even a late arrival lines up. Running a local 3-second timer on receipt leaves you off by the propagation delay.
• Host authority. One device holds control and the others follow, which prevents two people pausing and resuming at once and leaving states inconsistent. Some designs let the host role be handed over or shared.

On-site fallbacks

No sync app removes every variable on site. For anything important:

• Rehearse on the real devices in the real place. A classroom with isolated Wi-Fi may not allow nearby discovery.
• Check settings: automatic date and time on every device, notification permission, Do Not Disturb or Focus, the silent switch and battery.
• Know a second path. If nearby connection fails, join by room code through the internet relay.
• Keep a human backup. A non-host with a separate timer lets the event continue if the app misbehaves.
• Use sound. In a noisy room do not rely on the screen; use a whistle or a loud call for transitions.
• Double the end alarm. Confirm the end alert is scheduled locally ahead of time so it fires even if the connection drops.

What to tell participants beforehand

However good the tool, trouble at entry eats into your event time. A day before, or ten minutes before the start, send a short notice covering: which app to install and the link; how to join (scan the QR or type the 6-character code) and that local-network, camera and notification prompts should be allowed; whether to keep the screen on and whether the alarm sounds with the screen off; and who to contact if connection fails. With a large group, connecting only the host device and one big display or tablet, and letting everyone else follow a loud signal, is often more practical than connecting every phone. The tighter the capacity limit, the more this simplification helps stability.

Two sample sessions

HIIT group class (40 minutes)

1. Ten minutes before class the instructor creates a room as host and shows the QR. Participants scan, join and tap ready.
2. Build a sequence: for example two steps, 40 s work and 20 s rest, run for 8 rounds with the repeat setting (the editor has a repeat field; check in your app version whether you can attach separate warm-up and cool-down steps). All screens switch together at every step.
3. After a 3-2-1 lead-in the instructor taps start and focuses on coaching during rounds.
4. If participants exceed the default capacity (2 devices) the host's device needs Pro. It may be more practical to connect only the instructor's device and one large display than every phone.
5. Review the session history afterwards.

Presentation plus Q&A (15 minutes)

1. Presenter and moderator join a room and create a "10 min presentation + 5 min Q&A" sequence; the total reads 15:00.
2. Make the moderator host so the presenter can just talk, and the host handles an extension with +1:00.
3. Enable a one-minute warning step so the wrap-up point is signalled.
4. When the sequence moves to Q&A every screen changes at the same moment, so the audience sees the boundary clearly.

Sharing one countdown with TimerLync

TimerLync, made by SETLOG, is a synchronized timer for iPhone, iPad and Apple Watch. The host sets the countdown and creates a room; participants join by code or QR, and the host controls start, pause and add one minute. Because it shares a single end time, the documentation says the display stays steady through brief connection trouble. Default capacity is 2 devices based on the host, rising to 8 with the one-time Pro unlock. Participants never need to pay.

Limits: Android is not supported (iPhone, iPad and Apple Watch only). With a default capacity of 2, connecting a whole classroom to one room needs Pro, and 8 is still the ceiling. Some behavior, such as watch alerts, must be verified on real devices, so test on your own hardware before a big event. Step-by-step usage is in the TimerLync guide.

Recording from one start with LapLync

LapLync, by the same developer, is a synchronized stopwatch, the opposite direction. It counts up from zero, devices share the start instant and a lap marked by the host appears on participants' screens at once. Join with a 6-character code or QR; nearby devices connect directly and distant ones through an internet relay. Free use allows a 2-device room; Pro adds rooms of up to 8, CSV export and unlimited history. It suits relays, split timing and team training, where several people's times are compared from one start point.

Limits: the app is for iPhone and iPad, and since nobody can promise perfection in every environment, an important race deserves a rehearsal. History retention and room size differ between free and Pro. See the LapLync guide for details.

Sources and further reading

1. IETF RFC 5905, Network Time Protocol Version 4 (offset and delay formulas) (https://datatracker.ietf.org/doc/html/rfc5905)
2. TimerLync guide · LapLync guide (app features and capacity per project documentation and app descriptions)

FAQ