Blood, Sweat, and Pixels
Page 18
By January 2013, they were all in on Shovel Knight, which was exhilarating, yet terrifying. They were elated to be working on their dream game, but they knew they’d be working on it for at least a year before they started to make any money. Getting funding from investors or publishers seemed like a bad idea. A publisher would demand creative control, and worse, a publisher would oversee marketing. Velasco and crew had a specific vision for Shovel Knight as a brand, one that they wouldn’t want to trust with a big publisher that was looking for profit above all else. The only real option was crowdfunding, not just to earn money for their game, but to start building a base of loyal fans. “We thought, Kickstarter is probably the best way we can grow a community that is behind the game the whole way through,” said David D’Angelo. And so, like Obsidian and so many other independent studios, the Shovel Knight crew went to Kickstarter.
For the next two months they started preparing for the Kickstarter, the whole team living off their savings accounts as they drew bosses, sketched out levels, and tried to figure out what Shovel Knight would look like. He would have to be both unique and instantly recognizable. Everyone on the team had grown up in the late 1980s and 1990s, when a mustachioed plumber with a red shirt and blue overalls had taken over North America thanks to Nintendo’s relentless marketing, and they wanted to do the same thing. They wanted Shovel Knight sweatshirts. Shovel Knight plushies. Shovel Knight magazines. “I wanted to make an eighties brand that created a character, where the name of the game was the name of the character,” said Velasco. “That’s a thing that didn’t necessarily happen so much anymore.”
When they’d finished designing Shovel Knight, he had light blue armor, not unlike Mega Man. He always carried a shovel in one hand. His face was always masked, hidden behind a helmet with a T-shaped hole in the middle. Two pearl horns protruded from both sides of that helmet. Not only was he easy to recognize, he was easy to draw: sketch out the T helmet and a couple of horns, give him a little shovel, and you’ve got yourself a Shovel Knight. He didn’t have a personality, which was also by design. “Shovel Knight can be all things to all people,” said Velasco. “To some people, he’s a cute knight going around. To some, he’s a super badass Dark Souls knight.”
Every day for the next two months, they all went to Velasco’s apartment and worked on Shovel Knight, with D’Angelo’s head watching them daily from a laptop monitor. (Although D’Angelo was technically still at WayForward, his mind was mostly on Shovel Knight.) They didn’t sleep much. They’d decided to build King Knight’s stage, Pridemoor, to share with fans as an early demo and proof of concept. That meant there was a ton of work to do: designing the level, conceiving enemy creatures, drawing sprites, animating everything, programming the physics, and oh so much more.
Helping fuel Velasco and crew was the fact that they were working for themselves, not for some other company’s contract. Nobody enjoyed crunching, but spending long hours on Shovel Knight felt more rewarding than it had to lose sleep for WayForward’s licensed games. “Before, we were talking about it on [nights and weekends], and now the fact that we got to do it full time felt refreshing,” said Ian Flood. “I remember calling my dad and saying, ‘Hey, I turned down a raise and quit my job.’ And his response was, ‘Why?’ It’s like, ‘Oh, we’re going to put together this project and it’s going to be really great and we’re going to put it up on something called Kickstarter, where we take donations.’ His response was, ‘Well, let me know when your begging site is up.’”
They called their company Yacht Club Games, perhaps because for a group of poor, tired, bleary-eyed developers, sitting in a one-bedroom apartment and working on Ikea furniture, it felt morbidly ironic.
And on March 14, 2013, they launched the Kickstarter. A brief trailer, scored by the well-regarded composer Jake Kaufman (who went on to write all of Shovel Knight’s music), showcased the art and levels they’d done so far. In a series of short clips, Shovel Knight would stab enemies and bounce around on his shovel. “We believe in Shovel Knight,” they wrote in the Kickstarter description. “We know that the final game is going to be great. We’re so sure of this that we quit our full-time jobs to work on it. We’ve poured a massive amount of time, money and effort into our labor of love, but it’s not enough to finish the game. In order to make Shovel Knight a reality, we need you!”
That “you” didn’t seem to be showing up, though. By the end of the week they’d raised only around $40,000 of their $75,000 goal, thanks to a handful of articles, including a feature on the gaming site IGN, and they were getting only a few thousand dollars a day.* As the Yacht Club crew kept powering through their ever-growing list of tasks, they started worrying that Shovel Knight wasn’t going to make enough money. Or, perhaps worse, that it’d reach the absolute minimum goal. “You’d just have Kickstarter open on a second monitor or background, whatever you’re doing,” said Nick Wozniak. “You’re constantly looking, trying to do the math, doing rough projections in your head.” Even though they’d asked for $75,000, they were all hoping for way more than that, because they thought they’d need closer to $150,000 to get Shovel Knight out the door. “If it hadn’t funded, we didn’t even have a backup plan,” said Sean Velasco. “I guess we could’ve tried to stay as a team and pitch Shovel Knight to publishers. I don’t know if they would’ve gone for it. It was kind of do or die.”
They did have a plan to get more attention, though: they were going to PAX East. Earlier in the year, Yacht Club had booked a booth at Penny Arcade’s annual Boston gaming convention, where they could show Shovel Knight to tens of thousands of fans, journalists, and industry colleagues. But they still hadn’t finished the demo. In the following weeks, the Yacht Club crew worked sixteen-hour days in hopes of completing King Knight’s stage for PAX, only taking breaks to hit F5 on Kickstarter.
At one point, nervous that they might not make it, the team panicked and came up with a backup plan. They’d set up some beanbag chairs and old Nintendo 64 games at their booth, then tell attendees to go home and fund Shovel Knight. “You can just come and hang out and we’ll talk about our Kickstarter,” said Velasco. “We were thinking that even a week or two before PAX. It was nuts.”
Even on the day they flew to Boston in late March, the Yacht Club crew was putting the final touches on their demo. Hours before their flight, they all met up at Velasco’s apartment to make some last-minute changes, like adding an arcade-style leaderboard so PAX attendees could compete over high scores. They’d printed out several thousand Shovel Knight flyers, but they couldn’t afford to ship them to Boston separately, so they each had to stick giant stacks of paper into their luggage, then weigh it all to ensure they wouldn’t have to pay extra at the airport.
Things got a little frantic. “About twenty minutes before leaving, I tried to make coffee and [Velasco’s] sink started spewing up grounds of coffee,” said Ian Flood. “I said, ‘Hey your sink is going the wrong way.’ He said, ‘There’s no time, we have to go to Boston.’ I [said], ‘OK great.’ And so we just left it, and it was a floaty, chunky mess.”
The five of them—Sean Velasco, Ian Flood, David D’Angelo, Nick Wozniak, and Erin Pellon—spent a total of roughly $10,000 on the PAX trip, with the promise that they’d each get their money back once the Kickstarter had funded. They all shared one room at night—“It was pretty horrible,” said D’Angelo—and worked the PAX booth during the day, preaching about Shovel Knight to anyone who came within a few feet.
Their demo was flashy and eye catching, with bright 2-D graphics that immediately drew people’s attention. Even from across the show floor, it was easy to tell what was happening in Shovel Knight: a little blue guy with a shovel was jumping around and hitting monsters. It looked like a modern take on an NES game, which appealed to plenty of nostalgic PAX-goers. “In just a few short days Shovel Knight has gone from a blip on my radar to one of my most anticipated titles of the year,” a reporter for the video game website Destructoid wrote during the show.
At PAX, David D’Angelo talked to other developers who had successfully used Kickstarter and got two major pieces of advice. The first was to update their Kickstarter every single day, so backers could actively participate and spread the word about Shovel Knight rather than simply waiting around for the game. Right after PAX, Yacht Club started running daily Kickstarter updates with art contests, character reveals, and stretch goals. At $115,000 they’d make one of the boss knights playable. At $145,000 they’d add a second. If they hit $200,000, they’d add a multiplayer battle mode, and if they somehow got to $250,000, they’d create a third playable boss knight campaign.
The second piece of advice was that Yacht Club should send the Shovel Knight demo to popular YouTubers and Twitch streamers. Internet articles about the game were one thing; letting potential fans see Shovel Knight in action was quite another, and when huge YouTube channels like the Game Grumps later played through the demo, they reached hundreds of thousands of people. During the last few days of the Shovel Knight Kickstarter, the funding skyrocketed, going from a few thousand dollars to upward of $30,000 to $40,000 every day. It was enough to make Yacht Club wish they could extend the Kickstarter campaign an extra week, but alas, the site’s rules wouldn’t allow it.
When the Kickstarter ended on April 13, 2013, Shovel Knight had raised $311,502, which was more than four times their campaign goal but wasn’t all that much money for a team of five people in Los Angeles. At the standard burn rate of $10,000 per person per month (which included not just salaries but equipment, legal fees, and expenses for the small new office they planned to rent), it would last them six months—maybe longer if they all took less money. Once they figured out how much they’d each need to pay bills and put food on their tables, they split up salaries accordingly. “Then we said we’d pay the difference for everyone afterward,” said D’Angelo. “We really had to make the most out of every dollar.” Knowing that they had to make eight stages, and that King Knight’s stage had taken them around a month, Yacht Club figured they could finish Shovel Knight in a year. It’d take from April to December 2013 to complete all the stages, with an extra three months as buffer. They’d have to finish the game by March 2014, or else they’d be out of money.
They also had to learn how to start a business, which was a time-consuming process involving signing up for health insurance, sorting out taxes, and finding a copyright lawyer who could help them protect the Shovel Knight IP. Eventually they decided to designate Tuesday as “business day” and spend the rest of the week making Shovel Knight, but the work became far more time consuming than they had expected.
Because their funds were so limited, the Yacht Club crew gave up on all semblance of work-life balance, knowing that if they didn’t crunch on Shovel Knight, they’d run out of money. That March 2014 deadline was sooner than it seemed. “It would be best for us to kill ourselves on this than to fail,” said Nick Wozniak. “The thing we sacrificed for that was ourselves. We knew we’d have to be working weekends. We knew we’d have to be working multiple-hour days. Time is a thing that you just give to the game.”
For Nick Wozniak, that meant drawing and animating the many pixelated sprites that made up Shovel Knight, usually based on concept art drawn by Erin Pellon. For Sean Velasco, it meant designing levels, creatures, and mechanics. For David D’Angelo and Ian Flood, it meant writing the code that kept Shovel Knight running and fixing the code that wasn’t working. For all of them, it meant trying to figure out how to make the game fun: iterating, fine-tuning, polishing, and holding daily meetings about pivotal decisions surrounding both their game and their company.
From the beginning, Yacht Club had made the unorthodox decision that nobody would be in charge. Sean Velasco was technically the director of Shovel Knight, and he led most of the meetings, but he wasn’t the boss. They followed a simple, yet radical rule: If anyone said no to something, they all had to stop doing it. Nothing would happen until the entire team agreed on a singular way to proceed. It was the democratization of video game design. “We knew from the beginning we wanted to be equal,” said Wozniak. “We saw ourselves as five partners, on paper and in act.”*
In practice, this meant they spent a lot of time debating granular details. If one member of the team didn’t like the way Shovel Knight’s arm moved when he stabbed upward, they’d all have to talk about it. If someone at Yacht Club insisted that you should be able to replay levels after beating them, that would turn into a weeklong discussion. If Sean Velasco loved the visual of a blue, armored knight holding a fishing rod, then damnit, he was going to keep fighting until Shovel Knight could catch trout. (Said Velasco: “I kept bringing it up, kind of as a joke. Hey, let’s do fishing, it would be stupid, it would be fun. Everyone’s like, ‘No, it’s dumb, it’s dumb, it’s dumb.’ Eventually I think I just wore them down.”)
This sort of chaotic multilateralism might not have worked at any other company, but for Yacht Club, it just seemed to click. Part of it was their size, which kept them nimble—if they were fifty people instead of five, it wouldn’t have been quite as easy to battle over every design decision. The other part was the chemistry they’d developed after years of making games together at WayForward. “When I give these guys my designs or when they give me their implementations of stuff, there’s so much that doesn’t even need to be said or doesn’t need to be specced out, and there’s this level of knowledge and trust,” said Velasco. “It’s like being in a band.”
None of this would have been possible if they didn’t have a single, unified vision. Everyone on the team knew that Shovel Knight was a two-dimensional, NES-style platformer with eight different stages. It wasn’t getting any bigger. The fundamental gameplay wasn’t going to change. Nobody was going to argue that they should turn Shovel Knight into an MMO or replace the down-thrust mechanic with a machine gun. There were no publishers or investors who could drop by the office and insist that Shovel Knight would look better if maybe he took off his helmet. Even when the team fought over creative decisions, they all agreed on the basics of where Shovel Knight should go.
If they were going to turn Shovel Knight into a massive franchise—make it the next Mario—then their first game had to be perfect. Throughout 2013, Yacht Club designed and argued and crunched, building out each of Shovel Knight’s eight stages gradually and meticulously. Shovel Knight had improved drastically from the demo they’d shown at PAX, and they’d added all sorts of cool graphic effects. They were all particularly enamored of parallax scrolling, a technique in which the background of a stage could move separately from the foreground, creating the appearance of depth. Now, when Shovel Knight walked on top of the golden towers of Pridemoor Keep, the pink clouds in the background would move along with him.
Sean Velasco had come up with a series of platforming design edicts, inspired both by his time at WayForward and by studying classic Nintendo games, that he used to envision every level in Shovel Knight. One edict, for example, was that the game should teach you how to get past each encounter fairly, not through dry tutorials but with actual gameplay. Say Velasco wanted to introduce the player to a new enemy: the plague rat, which would explode upon getting hit. If you could just walk up to a plague rat, hit it, and then die because you didn’t know it would blow up, you’d get mad at the game. A better approach would be for the game to teach you the rules first, maybe by having the plague rat run back and forth next to a clump of dirt.
“That way when you [hit] the dirt, the rat also has a good likelihood of getting hit,” said Velasco. “Then you can see how he blows up.”
One question Yacht Club had to debate over the course of development was just how challenging to make the game. Backers and fans would approach the team with different requests, some begging them not to make Shovel Knight too hard, and others asking that they not make it too easy. “Half the people who play this game are going to expect a really hard NES game, and that’s part of what made those games fun,” said David D’Angelo. “And half the people who play this game want the same feeling you get from an NES game but don’t want it to be that hard. So how do you balance those two things?”
One solution was to add a variety of useful items, like a propeller dagger that would let Shovel Knight soar over gaps and a phase locket that you could use to gain temporary invincibility. Another solution—and one of the more elegant ideas in Shovel Knight—was to make checkpoints optional. Every time you found a checkpoint in a stage, you could either activate it (so that you’d restart there if you died and you wouldn’t lose too much progress) or destroy it, gaining some treasure but raising the stakes for yourself.
By the end of 2013, all five of them felt like they were making a great game—they just didn’t know how much longer it’d take. That $311,502, which was ultimately closer to $250,000 after taxes and fees, had dwindled quickly. (They’d also received $17,180 on PayPal from extra pledges, but that didn’t help much.) Yet finishing Shovel Knight by March 2014 seemed impossible. “I’ve always been late on every game I’ve ever worked on,” said Sean Velasco, “and it’s because I always want to revise stuff and do it again, or do it better, or put more polish on it.” For months now they had been crunching, but there was still so much to do. The second half of Shovel Knight wasn’t smooth enough, and they’d made some drastic changes to the ending, which they’d have to spend time tweaking and polishing. “I would say if we were at WayForward, we probably would’ve shipped it in March,” said David D’Angelo. “It’s probably the difference between good and great, or something. We really went over every pixel in the game and made sure it was what we wanted.”