Divinity: Original Sin is the game Larian Studios waited 15 years to make

"We always wanted to make an old-school PC RPG with multiplayer," says Swen Vincke, the founder of Larian Studios. "We never managed to sell it to a publisher in the past; every time we proposed, it was refused." I've called him to ask after Divinity: Original Sin, Larian's recently-released isometric cooperative PC RPG. It's proven remarkably successful despite -- or perhaps because of -- its nostalgia-tinged design, which draws inspiration from German pen-and-paper roleplaying games as well as '90s-era PC RPGs like Baldur's Gate. The Belgian studio Kickstarted Original Sin (to the tune of just under $1 million) last April despite competition from prominent, concurrent Kickstarter campaigns for isometric RPGs like Richard Garriott's Shroud of the Avatar and inXile's Torment: Tides of Numenara. Larian managed to beat them both to market, launching Original Sin on Steam's Early Access service in January before bringing it to retail at the end of June. Now it's the top-selling title on Steam and Larian's fastest-selling retail release to date; the studio is well on its way to recouping the roughly $4 million it spent on the game's development. "We never expected it to be this successful," says Vincke. "It's all due to Kickstarter and Early Access."

Cancelled, but not forgotten

Larian Studios has repeatedly tried to finagle co-op and multiplayer options into its previous projects, including Original Sin predecessor Divinity II, but the cost of QAing that multiplayer content always caused publishers to mandate its removal. This constant struggle against publisher expectations eventually drove the staff of Larian Studios to pursue independent development, in part so they could start a project they'd been trying to make for fifteen years. "Not many people remember, but one of our first games, which was cancelled during production because the publisher went bankrupt, was a game called The Lady, The Mage

and The Knight," says Vincke. "We were trying this same multiplayer concept, and ever since it was cancelled we’ve been trying to remake that game." Cancelled in 1999, The Lady, The Mage and The Knight was designed to be a three-player cooperative RPG based on the popular German role-playing game franchise Das Schwarze Auge, or "The Dark Eye." Players could switch between the game's three protagonists or bring in other players to control them, and during cooperative play every character could attack, initiate dialog and generally take agency in the world. Vincke claims that many of the core cooperative multiplayer elements of Original Sin were based on The Lady, The Mage and The Knight. "If you're doing a quest like killing The Butcher in Diablo, I want to be able to go kill Cain while you're doing it," says Vincke. "That's what we originally wanted, and that's what you can do now in Original Sin. Honestly I’m surprised that nobody else has done this until now, at least in the way that we’ve done it." Except he's not, really; when I ask him about why he thinks open-world multiplayer RPGs are rarely attempted, Vincke laughs. "Because it’s a nightmare to make! It’s literally a QA nightmare to make."

The value of developing in public: the world is your QA team

Larian mounted one of its largest-ever quality assurance testing phases on Original Sin, but much of that work was done voluntarily by Kickstarter backers and Early Access players. This side benefit of public development was what made it possible for Larian to ship Original Sin, though they made a few mistakes along the way. For example, Larian initially required Kickstarter backers to pay more for the privilege of accessing alpha or beta versions of Original Sin -- a common crowdfunding tactic -- then instantly regretted it when it came time to start checking for bugs. "After we finished the Kickstarter and started working we realized that was the most stupid

thing we’ve ever done," says Vincke. "We had like 20,000 people who were willing to play this game and would probably give us the feedback we needed to test all the various gameplay permutations." Larian wound up giving alpha and beta keys to everyone who backed the game on Kickstarter, offering refunds to any backers who paid for the privilege. "It was a huge QA operation -- almost 70,000 testers -- but we didn’t have to spend as much money as we would have had to do otherwise, say if we had a traditional business arrangement with a publisher," says Vincke. But is there anything that he misses about working with a traditional publisher? The marketing and distribution support, or perhaps the regular pattern of milestone-based development? "No." answers Vincke, flatly. "No, there’s nothing I miss. I always wanted to be in a one-on-one direct relationship with our players -- we even put a global chat into our game while it was in Early Access, so we could talk directly to the players in the game." Larian had to disable the global chat after Original Sin left Early Access because there was "too much profanity going on," but Vincke recalls using it to ask and answer questions from the game's backers during development, which seems to suit him and his team better than the more formal milestone approval systems employed by many publishers. "With a publisher, you send a milestone and you’re lucky if someone reads what you’re saying, then you get back your thing and if it says ‘Milestone accepted’ you get your check and off you go," says Vincke. "Or you don’t get your check for this or that reason, and sometimes you have to spend time on the most ridiculous things or spend a lot of time arguing, because there are processes on the publisher side that have to be gone through, and by the time those processes are done you could have just implemented the bloody feature in the first place."

What developers can learn from Larian's experience

Vincke pitches crowdfunding to me as a better business model, one that affords developers more room to fail in creative ways while working on games like Original Sin because they aren't in danger of sinking their studio by missing out on critical milestone payments. "Milestone-driven development doesn’t have real room for failure," says Vincke. "Maybe your studio can get by with some failures because you have built a financial buffer, but how the hell are you supposed to know up front exactly how much of a safety net you’re going to need?" By pitching Original Sin on Kickstarter, instead of to a publisher, and then transitioning

it to Steam's Early Access service, Larian successfully funded the game's development without having to answer to anyone but its players. But explaining complex development decisions -- like the removal of the game's day/night cycle system, a key gameplay feature that had been promised in the Kickstarter pitch -- to its backers proved tricky. "We had to admit defeat," says Vincke, as the team realized that Original Sin would need to include things like night-time missions, separate NPC schedules and beds in order to simulate day/night cycles. "We’d basically have to make two games at that point, and that was just impossible." Vincke says the studio's fans responded well enough, offering criticism alongside some suggestions on features they'd like to see implemented in lieu of the day/night system. The Original Sin team actually took some of those suggestions to heart. "Pretty much the recipe that any developer has to follow is just 'listen to your alpha players'; you have to be able to filter and select things that are doable and don’t ruin your entire production, but there’s golden knowledge there," says Vincke. "It can save you days of meetings on a single feature, because they debate it for you and you often get a very clear analysis of that feature -- you just have to implement it." Going forward, Larian expects to rely on crowdfunding for its future projects -- though Vincke says his studio has learned a lot from its first brush with Kickstarter. "Don’t do anything physical," says Vincke, when I ask him about recommendations for his fellow developers who are thinking about using Kickstarter. "I would never again do all the boxed stuff, and I regret that we spent so much time on everything related to making a physical release happen." The studio wound up devoting a significant amount of resources and time to printing discs, shipping boxes, and getting Original Sin translated and age-rated in multiple territories prior to release. Vincke tells me he ignored good advice to focus on developing a digital game in English and only worry about things like localization after your game is released. "At the time I answered him by saying ‘you know we’ve been doing this for quite some time, we’ve released so many RPGs, we can deal with this, blah blah blah.’ And it’s true, we have done this several times, and it’s always been miserable! Here too, it was miserable again," says Vincke. "I will definitely try to listen to my own advice next time."