Q&A: Wrist-mounted game design on the Apple Watch

Apple is expected to start shipping its Apple Watch device to people who preordered it this week, and shortly after those people strap it on they'll be heading to the App Store to download games and apps that work with their new toy.

In the Apple App Store they'll find a slew of games designed specifically for the Apple Watch from established studios like WayForward (Shantae and the Pirate's Curse), Bossa Studios (Surgeon Simulator) and more.

But the field of iOS watch game design is still rough and relatively uncharted. Games for the Android smartwatch equivalent, Android Wear, only started cropping up late last year; prior to that, novelties like the GCE-Game Time watch accounted for the bulk of wrist-mounted game design.

Now developers are taking a fresh look at what it takes to design a great smartwatch game, so I rang up Bossa's Vince Farquharson and Rob MacKenzie this week to get their take on the topic. Farquharson serves as Bossa's COO while MacKenzie works as a designer, and they both had quite a bit to say about the studio's approach to designing Spy Watch (or Spy_Watch if you want to get typographically exact), an Apple Watch launch game that aims to draw players into a fictional game of espionage.

We wound up covering a fair bit of ground, touching on topics like porting smartphone games to smartwatches (short answer: don't) and how the design of immersive "background games" may change if players have a communication device with limited inputs strapped to their wrists at all times. I've taken the liberty of transcribing our conversation and publishing an edited version below.

Why make a game that’s limited to the Apple Watch?

Farquharson: It’s an interesting question, and I understand where you’re coming from. For us, Bossa has kind of a history of working with new hardware. We were one of the first people on the Oculus Rift, one of the first on the HTC Vive, we’re working with Improbable -- we sort of love new gadgets and fun stuff to play around with, and Apple Watch is clearly in that same category.

Of course, when we start on a game we don't first decide what the game is and then try to map it to a new device -- that's really counter-productive. What we do is we look at the hardware and what it's strengths are, then we work out what the right experience is that will take advantage of the features that are unique to that device.

For example, the closest thing to the watch is the phone, right? So I think you're going to see lots of games coming from phone to watch, but for us that was very much not the right approach. When it comes to the usage cases, you look at the phone and people maybe play a game on it once every one or two hours, for five or ten minutes at a time. 

Now the watch, you don't use it like that. You maybe interact with it every 15-20 minutes, for 3-4 seconds. The whole device has been built to serve that need extremely well. It's the ultimate notification/quick-response device. So what game type is the perfect fit for that? I'll tell you the answer we found: there isn't one. 

There really isn't any kind of game that works like that. So if you try to port a phone game, nine out of ten ideas you have just will not work on this watch.

But one of the things you'll see when you use the Apple Watch is that it's very good for, like if someone sends you a text message like "do you want to meet down at the pub later" it's very easy to just quickly say "yes", or if your health app tells you something that's really important, you can very quickly respond to that notification. It's very good at notifications.

So we figured maybe there's a game in that -- a game that's a bit like when you talk to people on your watch, but we make up something fictional. This person's off doing something really cool and they need your help in some way. So you interact with them in the same you you're interacting with your friends, but you're actually telling the game what to do. So it was a small step from there to saying well you're a spy, and there's this other spy that's off doing cool things but he needs your advice. And that was really the germ of the idea, if you will; we really went to town off that.

So you designed the game to be played purely on the watch, or did you have to use the phone as well?

F: Oh no, you play it completely from the watch; it's a watch game. You can play it from the phone as well -- we're calling that "backwards compatibility" -- but we strongly encourage people to play this game on the watch.

It's what we're calling "background gaming", where you don't decide you're going to play a game and go play for a certain length of time. It's just in the background, events in the game are running in real-time, and it all comes back to you via notifications. The watch will literally tap you on the wrist, like someone's tapped you with a finger, and you look at it to see an encrypted message come in. 

You pick one of two responses, and the whole thing is dynamic -- it will never send you messages more often than about once every twenty minutes, but if you interact a lot it will start sending you more messages and if you don't interact because you're in meetings or wherever than it will send fewer notifications.

What you describe as "background games" must present interesting design challenges -- it reminds me of games like Extrasolar and Majestic. How do you design something like that to respect player lives -- to keep from waking players at 2 AM, for example?

F: Well, we had to do a lot of iteration and testing to nail it down. What we learned is that different people like to interact at different frequencies, and if you pick one static system you'll never get it right.

So we had to go dynamic; depending on who the player is, how they like to interact and even how their day is going, the schedule of interactions will change. Our game will initiate interactions at a maximum frequency of once every 20 mins; if you don't interact with the agent, he'll just start making his own decisions. If he's waiting outside an embassy in a flower bed, he's not going to wait for 9 hours for you to tell him to garrote a guard or sneak in undetected.

Getting all that to feel real, to feel like an actual conversation, was a real challenge. We got that wrong a lot before we got it right.

More broadly, what practical design challenges did you encounter in working with WatchKit, and designing a game played entirely on the wrist?

F: Not many, I think because we just focused entirely on the watch. Had we not taken the approach we've taken, we'd have had loads of problems. 

If we'd tried to just port some great phone game to the watch, that would have been a nightmare. But we didn't do that, we just looked at the watch and said 'What can it do well, and what can we plug into?' So it's got very good notifications, let's make a game with notifications.

So I think my advice for avoiding design challenges is to design a game around what the platform's strengths are, then design something that fits to those strengths.

MacKenzie: The thing we constantly had to ask ourselves was, "does this belong on the watch? Yes, we can do it -- but does it belong on this platform?" 

We could have made the game a hell of a lot more complex than it is, but then it would have meant that you're interacting with the watch a lot longer, at which point the question becomes "Why is this on the watch, and not the phone?" If you don't have an answer, maybe you should be making a phone game.

F: [The Apple Watch] is a powerful device if you use it the right way, but it does require you to deliver a game that is very different from other games. It's a whole new idea, that you can't just "choose" to play our game -- the game integrates itself into your life, so for the next 5-6 weeks you're basically live role-playing. You're pretending you're the agent and this is all real, so you can't choose to play a mission whenever you like; you have to wait for the game world to interact with you.

I don't know, I think we've managed to find a very good design match for this hardware.

What makes the hardware itself appealing? In light of the crowded nature of other game markets, do you see this as an opportunity to boost your game's discoverability by getting in early?

F: Yeah, I guess that's one benefit of being early. That's not really the intention though; our intent is to make interesting games on interesting devices. We get really excited at all these new possibilities. I think there are lots of developers who are equally excited.

The whole company is basically based off game jams. For everything we create, we've probably gone through 30 or 40 things you'll probably never hear about. I Am Bread came out of this process, Surgeon Simulator came out of it, and so on.

So Watch Spy came out of a jam?

F: Yeah, some aspects came from one jam and some from another. We took the original ideas and married them together with what we learned about the platform.

The original idea we had, was this idea that it would be quite cool on a watch if you were talking to someone else. Right? We originally thought of loads of cool people you could talk to, but the spy idea seemed best because it matched so well to the idea of a super-cool watch.

But you know, we thought, if we do this right it could be a genre as well as a game. We do have other ideas we could do in the same style, if people like it; we have at least two other complete games we could ship on the watch.

But we decided to get behind Spy Watch first, and the whole thing is totally experimental: you never quite know how people will take it, and that's the exciting part. That's also the scary part.