Typeface
955

I wanted less from my smartwatch.

I switched from an Apple Watch to a Garmin because I wanted a watch that felt more like a tool and less like another phone. Most available faces still put far more information on screen than I wanted.

My requirements were specific. I wanted four-digit 24-hour time, the day and date, and an exact battery percentage. Everything would be white on black. I did not want a colon, icons, fitness rings, weather, animation, seconds, or anything that required a tap.

With only three values on screen, small choices became obvious. Font width affected the largest possible time size. Vertical spacing affected how quickly I could read the face. Update frequency affected how much work the watch performed throughout the day.

The hierarchy is simple on purpose.

The time is by far the largest element. The date sits above it, and the battery sits below it. I can glance at the face and find the right value without scanning a grid of complications.

Settings without turning the face into a dashboard

I compared several font and battery layouts before choosing the default. The finished face includes five custom typefaces, four date sizes, four battery sizes, optional color controls, and an optional percent sign. The settings preview updates on the watch, so the user can judge a choice on the real display instead of guessing from a list of names.

The face redraws often, but most of the data does not.

Garmin can call the face several times after a wrist gesture. The Forerunner 955 still needs a complete frame each time, but the strings behind that frame usually have not changed.

The normal update path makes one clock query. It reformats the time when the minute changes and rebuilds the date once per hour. Battery updates take a different route. The app subscribes to Garmin’s battery complication when it starts, caches each value Garmin publishes, and requests a screen update only when that value changes.

When the face becomes visible, it refreshes the cached battery immediately so the first glance after charging is current. If the complication subscription is unavailable, the app falls back to reading System.getSystemStats() once per hour. Changing whether the percent sign is shown only reformats the cached value and does not trigger another battery query.

// Simplified rendering strategy
onStart() subscribeToBatteryChanges();

onBatteryChanged() {
  cacheBatteryValue();
  requestUpdate();
}

if (minuteChanged) formatTime();
if (hourChanged) formatDate();

draw(cachedDate, cachedTime, cachedBattery);

The simulator was not enough.

Connect IQ handles fonts, resources, packaging, and updates differently from a web project. The round display also gives every line of text a hard width limit. I used the simulator while iterating, then installed each serious version on my Forerunner 955 to check spacing and readability on the real screen.

One attempted battery optimization looked reasonable in the simulator: skip a draw when the visible values had not changed. On the watch, that could leave the screen black after a gesture. The hardware expects a complete frame on every update callback. I changed the design so every callback draws the full frame, while the code avoids repeating clock formatting, date lookup, battery polling, and font loading behind it.

Most of the work came down to small decisions

• Preparing custom fonts that rendered consistently on the watch.
• Making the time as large as possible without clipping near the curved edges.
• Adding color options without cluttering the default face.
• Finding a blank-screen bug that appeared only after installing the optimized build on the watch.
• Separating sensible optimizations from battery claims I had not measured.

It now runs on the watch I wear every day.

Typeface 955 now runs on the watch I wear every day. It gives me the three things I check most often. I do not have to open a widget, dismiss anything, or search through extra data.

I can explain why the code avoids unnecessary formatting and resource work. I have not completed a controlled battery comparison against a stock Garmin face, so I do not claim a measured battery-life improvement. That test is still on my list.