Luca Redwood, creator of 2012 breakout hit 10000000, didn't come from a games development background.
The success of 10000000 came as a complete surprise, vindicating the sleepless nights over the year he spent juggling making the game, spending time with family and working his full time job developing finance software.
Redwood seems to have a natural talent for games development although he puts it more modestly. Multiple times during our interview he identifies as a "blagger", which seems to be his way of saying he's learning and he goes.
Indeed, his studio EightyEightGames, is all about creating games to accommodate his own lifestyle and tastes.
Whether blagging or not, the success of 10000000 allowed him to move into full time games development and work on his next project, the fiendishly addictive You Must Build a Boat.
The animals went in
You Must Build a Boat was originally designed as an update for 10000000, but that got out of hand and it became downloadable content, which got out of hand and became an expansion which got out of hand and became a fully blown sequel.
In this way, You Must Build a Boat builds on the concept of 10000000, blending match-three gameplay with light RPG elements, and ties it all to a base building system where you have to - you guessed it - build a boat.
There is no fail state and gameplay comes in few minute bursts, perfect for someone with a family and work commitments. Players can take all the time they need.
The move into full time game development meant that Redwood could take his time too, which was both a blessing and a curse. Work on You Must Build a Boat lasted three years.
You Must Build a Boat was originally designed as an update for 10000000, but that got out of hand.
Without a deadline the project dragged on longer than he intended, but Redwood learned a few things from working on 10000000 that saved him some headaches with You Must Build a Boat.
"When I made 10000000, I had no inkling of how it would take off, so everything was hardcoded for mobile devices," Redwood begins.
"So when I went to PC, it took a long time to unpick all those flaws and make it work.
"With You Must Build a Boat, I knew I would have it on PC from day one, so I avoided a lot of that pain.
"I've got two main reasons for wanting to have it on PC: first, I want as many people playing my game as possible; but secondly, it means more sales opportunities.
"With a small studio with next-to-no costs like mine, if I'm on lots of platforms, it means I only need to sell a few copies a day to still be in good shape."
Taking the helm
Obviously releasing simultaneously on PC and mobile comes with its own set of challenges. Both control systems have to feel right, despite their differences.
However, the trickiest part for Redwood was gameplay balancing.
"You Must Build a Boat is incredibly time sensitive with a lot of time pressure, and I found that the relationship between input methods wasn't linear," he recalls.
"At the start of the game, touch-based players would be faster at making matches, but by the end, once mouse-based players had got used to the input they were slightly faster.
A big part of You Must Build a Boat is character progression.
"There's a lot of tweaking under the hood when it comes to difficulty, where it gets massaged by a curve - for PC it's slightly easier at the start, and slightly more difficult at the end."
Nailing the core has always been most important to Redwood. Although he designed every aspect of the game, from its pixel art visuals to the bleepy upbeat soundtrack, it all started with an idea.
"I always liked the idea having one side of a game controlling another, but glueing them together ended up being difficult," recalls Redwood.
A big part of You Must Build a Boat is character progression. Each run will net the player gold with which to buy upgrades with and completing objectives will unlock new characters to sell them upgrades back on their boat.
The challenging part of design here came in making sure good players couldn't just overpower their character and plough through the game.
"As you might guess, I'm very good at this game by virtue of playing it for thousands of hours, so my perspective is off," says Redwood.
"I came up with a list of targets (see above), then had people play and balanced it as close as I could, but there is still a lot of variability."
Need a bigger boat
Much of this variability comes from giving the player options. In the match-three gameplay, tiles are matched to an action - swords are for physical attacks, staffs for magic, keys for unlocking chests, shields for defense - so there's a fair bit of freedom in how players grow their character.
Rigorous playtesting was the only answer.
The tiles weren't always tied to specific actions, however.
One of my goals is to make games that can be played in small snatches of time, but that small go will progress the game.Luca Redwood
"I liked the idea of having 'generic' tiles that could perform literally any action, but it ended up being important to have the actions available on the board be 'verby'," Redwood says.
"Matching a sword can only do a limited array of things and still make sense, but by tightening the scope, it connects the two halves of the game together."
Because of a handful of smart design decisions, everything else just fell in place from here. At their core, RPGs are about getting stronger and faster, so it made sense to not bar player progress with a fail state. A happy side effect of this is that it made the game more mobile-friendly.
"One of my goals is to make games that can be played in small snatches of time, but that small go will progress the game a little bit," says Redwood.
"It wasn't an intentional 'I want to make a no-fail-state game', rather that the requirement that a 30-second play must advance the game necessitated it."
Despite this, many players will struggle to play in 30 second bursts, because You Must Build a Boat is damn addictive. Part of this could be down to the constant progression, but Redwood believes he owes a lot to the way objectives are displayed.
Before each run, players can activate a set of objectives - open three chests, kill five enemies, etc - and each one unlocks a special reward; usually a new crew member for your boat.
"There's a finite number of these, and they vary quite a bit, so It's interesting to see them," says Redwood.
"When you complete one, It means you've got a slot available to view the next one, which you'll want to see.
"The only way to see that objective is to go to the helm and add the objective, and at that point you've got a run all queued up, with the objectives locked in. The run itself has already been loaded asynchronously in the background so there's next to no load time… why not just have one more go?
My strategy is to keep plugging at it, put something in, see if it's fun, if not throw away.Luca Redwood
"Even in the event you don't complete any objectives you'll probably have gained a bit of money, so you can get a little upgrade and do even better next time."
Like the pixel protagonist of his game, also Redwood is growing better with each design decision. Part of that learning experience is knowing what to cut.
"I'm not particularly good at designing video games, so my strategy is to keep plugging at it, put something in, see if it's fun, if not throw away," he reveals.
"There are at least another three or four games on the cutting room floor. I switched between all sorts different matching mechanics - or none - turn-based or real time. At some point I even get rid of the running along the top and changed it into a pet battler.
"The biggest thing cut was the Choose-Your-Own-Adventure style exploration, which preceded a dungeon run and your choices customised the run in interesting ways [ screenshot attached ]. I loved the system, it had hundreds of encounters with thousands of possible outcomes, but it just wasn't right for this game."
Another thing Redwood didn't feel was right for You Must Build a Boat was the free-to-play business model.
He admits that You Must Build a Boat may have made more money if he had taken this route, but says that the option of bypassing the journey with money would have "short circuited" the entire game.
"I don't have any problem with free-to-play, but most free-to-play implementations and games are bad," says Redwood.
"Free-to-play itself doesn't make a game bad, but I strongly believe it would have made this game very bad."
Indeed Redwood's other game Smarter Than You uses the model as a means to connect players all over the world. It's about what fits, what feels right for the project.
The whole design process has been about Redwood following his own intuition. Even the sound design work is his own. Of course, he says he blagged it.
"Tiles were originally made of wood because that fit the theme better, but I couldn't get the sound effect of them breaking right, so In the end they changed to recordings of flowerpots and glass smashing, mixed with a bit of a thud which was satisfying," he says.
"Tiles falling down and reaching their home position is my favourite - the sound of poker chips falling on top of each other is quite satisfying, so I used that sound. But that didn't sound nice enough, so each tile moves at a slightly randomised speed so they arrive at different times and get a nice ripple effect.
"I had a couple of goals with the music - when running the dungeon it had to be interesting and melodic, but the main thing was it had to be fast and driving to keep people running through the dungeon.
"The music back on your boat, however, should be slower, upbeat, and above all constant - I liked the idea of having different music for each zone, but I preferred the idea of "Home" being constant regardless of where you were."
Again, the key for designing the sound was experimentation, and settling once it felt right. It's like the art style - lots of games are opting for the retro look these days. But in the case of You Must Build a Boat, it wasn't exactly a stylistic choice.
Redwood explains: "Possibly the greatest trick I ever pulled - unintentionally - was that this art style was intentional; big-pixels-with-high-color palette, and ugly. It's not intentional, it's literally the best I can do. I'm not an artist.
Sale away with me
Of course, talking about the game in the past tense suggests that its success was preordinated.
That wasn't the case. Leading up to release of You Must Build a Boat, the realisation started to dawn on Redwood that most of 10000000's profits, and three years of his time, had been committed to the gamble that was his new game.
iOS and Android are currently duking it out for top spot. Steam sales are great, but lower.Luca Redwood
He didn't get much sleep in the months leading up to release.
In the back of his mind, Redwood was telling himself he should have put the profits from 10000000 into his mortgage and gone back to 9-5 work. Luckily, his fears were unfounded, and he hit his sales goal within 10 days of release, with mobile revenues dominating.
"There's a piece of popular wisdom that you shouldn't make paid games on Android, " says Redwood.
"I haven't found it to be the case at all - iOS and Android are currently duking it out for top spot, Steam sales are great, but lower.
In fact, I don't think they'll ever reach even a third of what 10000000 sold on Steam. You Must Build a Boat is definitely a better game, but the PC market has changed a lot in the last few years."
And it looks like the market might have changed again by the time we see something new from Luca Redwood and his one man studio - he's currently at sea without a map.
"I have some less glamorous stuff to do - localisation for You Must Build a Boat, an update for Smarter Than You - but beyond that I'm in a bit of a funk," Redwood concludes.
"I've had around five ideas that I really love that I think have legs for my next project, and I can't commit to which one to go for, so am doing nothing instead.
"Watch this space."