Ask the Developer Vol. 14, Nintendo Sound Clock: Alarmo

• This article has been translated from the original Japanese content. 

• Some of the images and videos shown in text were created during development. 

In this fourteenth volume of Ask the Developer, an interview series in which Nintendo developers convey in their own words Nintendo’s thoughts about creating products and the specific points they are particular about, we’re talking to the developers behind Nintendo Sound Clock: Alarmo™, available from Wednesday, October 9th. 

First, could you briefly introduce yourselves?

Tamori: Hi, I’m Yosuke Tamori. I’m a member of the Entertainment Planning & Development Department, and as a producer for Alarmo, I oversaw the project as a whole. In the past, I was involved in the development of Splatoon (1) for Wii U and Nintendo Labo (2) for Nintendo Switch as a game software designer.

(1) A Wii U™ game released in May 2015. This is the first title in the Splatoon™ series, in which players engage in 4-on-4 ink-slinging Turf War battles. The team with the most inked turf wins.

(2) Released in April 2018. Nintendo Labo kits provide players with the tools to build and play with various Toy-Con creations—such as pianos, fishing rods, motorbikes, and robots—by combining cardboard, the Nintendo Switch™ system, and software.

Akama: Hello. I’m Tetsuya Akama. I’m a member of the Technology Development Department, and I joined this project as the director. I was originally a product designer. Recently, I was involved in hardware projects such as the Joy-Con grip for Nintendo Switch and the Joy-Con Wheel.

Ah, so two developers from different departments—one who creates game software, and one who creates hardware such as game systems and accessories—worked together to develop Alarmo. Could you give us a quick overview of the product?

Akama: Sure. This is Nintendo Sound Clock: Alarmo. In a nutshell, it’s an alarm clock that’s out of the ordinary made by Nintendo. What makes it so unusual is that it has a special sensor inside the device that detects a person’s movement, so it can tell whether they’re in bed or not. Using this sensor, Alarmo repeatedly tries to wake you up while you’re in bed to help prevent oversleeping. When you get out of bed, the alarm automatically stops. Another unique feature is that the music and sound effects from selected Nintendo games are used as alarm sounds, offering an experience that feels like you’re waking up in a game world.

Tamori: Simply by placing this alarm clock next to your bed, you can wake up in a pleasant or fun way, and it can automatically play music to help provide a relaxing environment when you get into bed. It also keeps track of how long you’ve been in bed sleeping and how long it took you to get out of bed after the alarm went off.

Nowadays, Nintendo mainly creates game consoles and software, so this product is a bit unorthodox, wouldn’t you say? How did this unorthodox project start?

Tamori: One of our in-house projects was to research motion sensor technology. As this sensor maintains privacy since it doesn’t use a camera, we had an idea for how it could be highly suitable for use in the bedroom, so we decided to kick off a new project to see what we could do with it.

How does a motion sensor work?

Akama: It’s generally known as a “radio wave sensor.” To put it very simply, it uses the reflection of radio waves to measure the distance from and speed of an object. Self-driving cars, drones, and robots are equipped with sensors like these to help avoid collisions. The key feature is that it can recognize very subtle movements, and unlike cameras, it doesn’t need to film videos, so privacy is better protected compared to cameras. Since it uses radio waves, it can be used in dark rooms. It can also detect movements even if there are obstacles, as long as the radio waves can pass through.

Tamori: This sensor can detect the movement of a person in bed even when they’re covered with a comforter, and it doesn’t need to use images or video like a camera. Due to those characteristics, we’d been discussing internally for some time that a motion sensor could also be suitable for use in a private space, such as a bedroom.

I see, so you started development from scratch while also making use of ideas that already existed. You mentioned that a motion sensor could be suitable for bedroom use, but was it clear from the beginning that you’d be making an alarm clock?

Tamori: Actually… not quite. We knew that the theme was to support people during their sleep, and we were looking for a way to achieve this. However, it was only after a lot of technical trial and error and building many prototypes that we decided to focus on alarm clocks.

So, you utilized the team’s existing knowledge, but there was a certain moment that ultimately made you settle on an alarm clock. Is that right?

Tamori: Well… The motion sensor used in this device detects a sleeper’s movements, and there are certain body movements that they make just before waking up. From an early stage, we thought that if we could get this sensor to detect the timing of these movements, people could use it to help themselves wake up feeling refreshed. But it was technologically quite difficult for the sensor to detect those movements accurately enough back then, so we continued experimenting for a while.

Then, our programmers found a way to use the sensor effectively, and there was a point when its responsiveness improved dramatically. They developed a system that detects approximately where a person is on a bed based on the distance and angle measured by the sensor. This made it possible for the sensor to capture the moment a person gets in or out of bed more accurately. It was at this time that we started discussing the possibility of developing an alarm clock that automatically stops as soon as a person gets out of bed.

Akama: From the early days of the project, we considered implementing a feature to wake yourself up, such as an alarm that stops around 10 seconds after you get out of bed or when you move your body. Although those were all useful features, we weren’t sure if they’d be enjoyable or satisfying, unfortunately. Back then, there were no game characters appearing on the screen. The time remaining until the alarm turns off was simply displayed on an on-screen bar, so you had to look at the screen to see the time remaining. Also, it would take a while for the alarm to stop even after you got out of bed. So, there were many challenges.

Then, as programmers made technical improvements, such as the enhanced sensor accuracy mentioned earlier, we started coming up with more and more ideas to make it fun and responsive. These included sound effects that go up in pitch when you move your body, or a fanfare that plays the moment you get out of bed. It was around this time that we finally began to feel that we could create a new product in the form of an alarm clock.

Tamori: When creating hardware from scratch, as long as you can establish the technology that supports your idea, the functions you can implement are, in a sense, unlimited, so you can do all sorts of things. We even tried to detect whether a person is starting to sit up in bed to determine if they’re awake.

Akama: Yeah, we tried things like detecting arm movements, dividing the bed’s area into left and right to detect the direction you roll over, and making the alarm stoppable by stretching. Basically, we kept making various prototypes. During the trial-and-error process, the knowledge and expertise of the staff who were involved in the early stages of motion sensor development were very useful. For example, we took advantage of ideas such as “stopping a device without touching it” and “reacting to body movements” in the final product.

Tamori: We were so fixated on the idea of not touching the device that there was a time when we challenged ourselves to make the user interface, such as setting the alarm time and music, controllable with hand gestures. Gesture controls made things a bit cumbersome, so we nipped that in the bud. (Laughs)

Akama: Oh yeah, that was such a hassle. (Laughs) But we really did go through a lot of trial and error.