Dungeons & Dragons became a cornerstone of gaming culture by providing players with dice, sheets of paper, and guidebooks that teased the imagination—all the tools they needed to build their own worlds. Influenced by all-night D&D sessions, the video game developers at Black Isle Studios and BioWare had a thought: Leave the dice-rolling to computers, letting players focus solely on creating characters and embarking on unforgettable adventures.
The result was Baldur's Gate, a computer roleplaying game (CRPG) featuring breathtaking scenes, compelling characters, dozens of quests, and deep tactical battles. As financial turmoil plagued their parent company, a small team of developers broke away from Black Isle and set out to create unforgettable adventures of their own.
From the early days of Fallout and Baldur's Gate, to the formation of Obsidian Entertainment and the company's fateful crowdfunding campaign that averted financial ruin, Beneath a Starless Sky explores the making of the Infinity Engine CRPGs and the critically acclaimed Pillars of Eternity franchise.
•Originally published as a Shacknews Long Read
•Based on over 40 hours of interviews with developers from Black Isle and Obsidian
•Go behind-the-scenes to witness the creation of the celebrated Infinity Engine CRPGs: Baldur's Gate, Planescape: Torment, Icewind Dale, Baldur's Gate II, and Icewind Dale II
•Relive the dramatic moments that led to Obsidian's record-setting Kickstarter, and the making of 2015's Pillars of Eternity and 2018's Pillars of Eternity II: Deadfire
In 2017, I wrote Rocket Jump, still considered the flagship "Long Read" at Shacknews. In 2018, the Shack team and I were determined to outdo ourselves, and Beneath a Starless Sky was the result. The book was written over eight months of interviews and a studio visit to Obsidian. I like to think the end result speaks for itself, but you can be the judge. – David L. Craddock
"The scale of this is amazing. Holy shit, seven months of interviews, and it just shows everywhere."– 2018 Kunkel Awards Nominee: Excellence in Feature Writing
"As always, excellent and in-depth. The master of all things documenting video games."– Shacknews Reader
JOSH SAWYER WALKED down the hallway to Adam Brennecke's office and told the junior programmer he had found a bug. He was playing the latest build of Neverwinter Nights 2 and noticed that the camera failed to maintain its pitch when players switched between companions.
Brennecke scoffed. What Sawyer was saying was impossible.
"I'm like, 'Believe the impossible, dude," Sawyer said. "It's on my monitor. Right now. This is not a distant recollection. It's here right now."
"I knew, as a programmer, whenever I had to go talk to Josh about anything, I had to have a lot of ammunition to defend my point," Brennecke stated. He assured Sawyer that the camera was working perfectly because he had just checked it.
Sawyer grinned. Brennecke was a junior programmer and a graduate from DigiPen University. Like a lot of programmers fresh out of school, he was naïve and displaying a confidence bordering on arrogance. "I said, 'Okay. How much will you bet?'" Sawyer recalled.
Brennecke wagered one penny. Sawyer accepted. They left Brennecke's office and went back to Sawyer's computer. Sawyer sat down and moved his mouse. Brennecke stared. Sure enough, the camera's pitch had changed from one campaign to the next. Stunned, he told Sawyer he had no idea why the bug was happening. Neither did Sawyer—which, he pointed out, was why he had asked Brennecke to take a look in the first place.
Also, he added, Brennecke owed him a penny.
"It was a little humbling," Brennecke admitted. "I learned the lesson of not to promise things as a developer, because things can change so rapidly in game development day to day. He was teaching me a lesson, but it was pretty fun."
Sawyer made sure Brennecke never forgot their wager. He named his winnings The Penny of Broken Promises and, along with a florid account of the bet, framed it and dated it August 17, 2006.
A few weeks later, Sawyer went to Brennecke with another problem. Neverwinter Nights 2's portraits were failing to render at a proper level of detail. A portrait's appearance should change relative to the player's distance from the portrait: more detail up close, less detail further away. Once again, Brennecke guaranteed that Sawyer was not, could not be seeing what he claimed to be seeing. Once again, Sawyer dared him to back up his words by putting a penny on the line.
Once again, Brennecke handed over a penny. This one, Sawyer proclaimed, was The Penny of Broken Guarantees.
Two months passed. Neverwinter Nights released to favorable reviews, and Brennecke and Sawyer moved on to Aliens: Crucible. While carpooling one morning in November, they passed a Hardy's restaurant. Brennecke noted the smiling star logo shared by Carl's Jr., and mentioned that Hardy's had always used the star. Sawyer corrected him: He could not be sure when Hardy's had adopted the star, but it had not been a permanent fixture.
Brennecke amended his stance: Hardy's had used the star logo for at least ten years. He wagered a penny.
The first thing Sawyer did when he got to his computer was google Hardy's and Carl's Jr. Then he went to Brennecke to share his news: Hardy's had been purchased by Carl's Jr. parent company CKE Restaurants in 1997—nine years ago.
The Penny of False Guarantees went up on Sawyer's wall on November 8, 2006.
"It's great, though, because he learned," Sawyer said. "I remember we were working on New Vegas, I think. He said, 'That's not happening' to whatever bug I found. I said, 'Are you sure? Do you want to promise me that?' He said, 'I am... very confident.' I said, 'That's good, because it means you're not absolutely sure and there's a possibility you are incorrect.'"
"Whenever I need to talk to Josh about anything, I have to make sure I have a lot of reasons, well-developed reasons, for why I want to change something, or why I think something is not going correctly," Brennecke added. "It's a challenge, but I've found that Josh is right ninety percent of the time, which is really good for any developer. He has a good track record in his games, and his design sensibilities are really good. I've learned over the years to have ammunition to defend my position, or to just let him do what he wants to do, because he's usually right."
IN THE SPRING of 2012, Obsidian prepared to make several wagers. The company had been burning through cash to support its largest staff to date. With Stormlands cancelled and the staff gutted, the co-founders estimated Obsidian had until September to sign a game.
"The situation is dire for every independent studio on any given day. Was 2012 one of the more dire times at the studio? Absolutely," Feargus Urquhart said. "There were going to be points where we were going to have to make decisions about moving forward knowing that we can't keep spending money on things that aren't going anywhere. It was pretty dire."
Everyone else at Obsidian pounded pavement. "We had agents reaching out to different publishers to see what types of games they were interested in, and we would try to tailor our pitches to those publishers just to see if they would sign anything, get anything going with actually discussing projects and moving forward without anything substantial," said Brennecke.
Part of drafting pitches was researching the market and talking with contacts to find out what publishers wanted. "There were six, seven, eight-week stints where we were working on pitches for a new idea," said Rob Nesler, art director at Obsidian. "Companies are always interested in what we do, but it's also about crafting the right thing for them and what's going on in their world."
Sawyer detested sending pitches, but this situation made the process even worse. Obsidian's layoffs had made headlines on virtually every gaming website and magazine. That gave publishers an advantage. "Negotiating with anyone when they know they have you up against the ropes is awful," Sawyer admitted.
Obsidian's reputation as a maker of engrossing RPGs based on licensed properties worked against them. Before a project started, the team needed time to dig into a property so they could learn about it. Time was in short supply. "It is very frustrating because you might do a lot of work to put together a pitch doc, and they might not even read it, or you might get a ten-minute meeting where an executive looks at it and says, 'Yeah, we don't really want an RPG' and you're like, 'Oh. That's cool.' There's a lot of that," Brennecke added. "It is really grueling for the team to go through that process. It feels like there's a lot of wasted work."
Brennecke wanted to take a different tack. While working on Stormlands, he and designer Nathaniel Chapman had joked about Obsidian funding a game on Kickstarter, a crowdfunding platform where users pledge money to help fund a good or service in exchange for rewards. The higher their pledge, the more goodies they receive when a funded project is finished. Independent developers had funded games through Kickstarter in modest numbers, raising a few thousand or ten thousand.
Chapman and Brennecke rapped about crowdfunding a game, then forgot about it. They were joking, killing time while working on Obsidian's epic RPG. Then Tim Schafer put consumer-funded projects on the tip of every developer's tongue.
In February 2012, the legendary designer behind point-and-click adventure classics such as Full Throttle and Grim Fandango launched a Kickstarter for Double Fine Adventure, a game in the vein of his past hits. Schafer and his company, Double Fine, set a funding goal of $400,000. They ended up raising $3.4 million, setting a Kickstarter record.
One month later, Brian Fargo's company, inXile, announced Wasteland 2 as a Kickstarter project. Fargo had turned to Kickstarter as a Hail Mary. He had pitched several big-name publishers including Microsoft and THQ on reviving the Wasteland franchise from its roots as an Apple II game after he acquired the license. With a publisher's deep pockets, Fargo believed a sequel could enjoy the sort of commercial success Bethesda had achieved by resurrecting Fallout.
Microsoft turned him down. They had BioWare and Mass Effect, at the time exclusive to Xbox 360 and Windows. THQ, on the verge of financial collapse, was looking to create an original game, not dig up what executives viewed as a moldy-oldie.
"The business had fragmented such that everybody had their own unique perspective," Fargo explained. "I had one company say, 'We only do licensed products.' Another said, 'We only do products that we think can become billion-dollar franchises.' And, 'We only do multiplayer games now.' It wasn't like back in the '90s when we were all kind of doing the same thing. Things were very, very [specialized]. There was zero interest from anybody. Really, Kickstarter was a last-ditch effort for us to make those kinds of roleplaying games."
Fargo and inXile plowed ahead and set a Kickstarter funding goal of $900,000 to make Wasteland 2. They hit the number in two days. By the time its month-long campaign ended, Fargo and inXile had raised over $2.9 million.
Adam Brennecke and Josh Sawyer had watched game developers make headlines by tapping into the grassroots crowdfunding movement. "It wasn't just one success, it was repeated successes," Brennecke said. "While we were doing this pitch process, we kept on saying, 'Why don't we do a Kickstarter instead?'"
"We were like, 'Oh my god, this is a thing. We could actually do this,'" Sawyer recalled.
The landscape of Obsidian had changed by June 2012. Stormlands had been cancelled and developers were frantically sending off pitches to publishers. Brennecke and Sawyer approached the co-founders and ran the idea of crowdfunding by them. Urquhart and his peers demurred. "One of the things about running an independent studio is it's so much about perception: about how publishers see you, how your fans see you, how your employees see you, how you feel about your own company," Urquhart said, explaining his hesitation. "When it came to crowdfunding, I have to admit I was skeptical. If we put this idea out there, and we were having a tough time as a studio, what happens if it falls flat?"
Brennecke and Sawyer were undeterred. They approached management a second time. For a second time, management shot them down. Chris Parker backed up Urquhart and Darren Monahan. He knew the dangers of succumbing to industry trends. "There was this whole thing with Xbox Live Arcade," Parker remembered, "and then everybody was like, 'What you've really got to do is make this game that costs $8.99 on Xbox Arcade, and everything will be great.' Okay, and then everybody makes a game like that, and maybe nobody likes yours. You see it today with VR: 'Man, if you're not making VR, then you're going to be left in the dust.' Okay, there's some cool VR stuff out there, but it's not changing the industry."
Privately, Parker was intrigued by the enthusiasm around Kickstarter. Four months earlier, he had talked with Brennecke and Sawyer about crowdfunding and had even put together a proposal. Urquhart and Darren Monahan had voted him down—co-founder democracy in action. Parker had acquiesced then, and again when Urquhart and Monahan turned down Brennecke and Sawyer, because that was his job: He towed the company line with his partners. But the Kickstarter bug had bitten him.
After getting the thumbs-down yet again, Sawyer and Brennecke decided the time had come to make their biggest wager yet. This time, they were on the same side: If Obsidian refused to capitalize on crowdfunding, they would quit and launch a campaign on their own.
Josh Sawyer had one foot out the door when word of his imminent departure reached Chris Parker. Obsidian's co-founder sprinted across the studio to his office and asked Sawyer why he wanted to leave. Sawyer spelled it out. He and Brennecke wanted to make their game, and they wanted to crowdfund it. Parker was sympathetic, but explained that the co-founders needed everyone pitching to publishers.
Sawyer clarified his position. He simply believed this was something that someone at Obsidian should do, even if it wasn't him. Someone like Adam Brennecke, for instance.
Parker considered. If Brennecke was allowed to focus on the Kickstarter, would Sawyer continue pitching to publishers? Sawyer said he would.
Brennecke and Sawyer sat down for one final meeting with management. Brennecke stressed that now was the ideal time to crowdfund a project. With Stormlands cancelled, they had an opening.
"I give both of them an incredible amount of credit for really pushing and pushing me on it, and getting me to be okay with it, because it's been an amazing thing for the studio," Urquhart said. "I think the tipping point for me was, in some ways, I had to get more comfortable with it, just with the concept [of crowdfunding]. And I think it was also tied to the fact that we were not getting anywhere with a lot of the other pitches we had out to publishers. My partners and I had to make that decision to really trust Josh and Adam. If you mix that all together, we said, 'Okay. Let's do it.'"
Exhilaration coursed through Brennecke. With the go-ahead from management, he could begin building a crowdfunding campaign. Now all he needed was an idea for a game to fund. He had just the one: An Infinity-style RPG with isometric view, companion characters with unique backgrounds and personalities, real-time-with-pause combat, a party of adventurers, memorable stories and decisions that had lasting effects on characters—all the trimmings that had made Baldur's Gate and its descendants instant classics.
"I felt that was a very strong choice as the game to make, and the game that would succeed on [Kickstarter] just because of our roots with Black Isle and having a lot of people at the studio who worked on Planescape: Torment and Icewind Dale," Brennecke said. "I thought that would be a great way to enter crowdfunding, and a good idea [on its own]. I felt that when people heard it, that would be enough for people to open their wallets and give us their fifteen to twenty dollars."
Sawyer was on board. He missed the type of exploration that had been unique to RPGs like Baldur's Gate and Icewind Dale. One of his colleagues described playing them as moving through a diorama. That hit the proverbial nail on its head. "With 2D isometric games, even though a lot of people really don't like the look, it is a different experience," Sawyer explained. "It's you're looking into this little box that was created for you to peer down into. The way of navigating those spaces, the way you explore them, the way you fight in them—it's a different feeling."
Many of their peers agreed. Like Brennecke, senior designer and writer Eric Fenstermaker had grown up playing Baldur's Gate and its descendants. "I liked the challenge of working with a lower budget and seeing what could be done with the approach to the narrative to compensate for a lack of cinematic cutscenes. After having scripted—in the programming sense—dozens of cutscenes Neverwinter Nights 2, which was built on an engine that was built on another engine that was hopelessly nondeterministic and absolutely the worst imaginable framework for trying to create cutscenes, the notion that we might just write the actions rather than script and animate them was appealing."
Senior designer Bobby Null saw a hole in the market. "I thought, Nobody's making those games. Of course we should do it. There's got to be a market for it because we all love those games."
Rob Nesler missed the satisfaction of playing armchair general. "I like having a gathering of characters, of companions with interesting personalities, and having them move through the world and kill monsters and get loot. Certainly the character interactions and the thoughtful decisions that need to be made for me to feel right about how I'm playing—that's a layer of complexity that I'd forgotten about. It was quite wonderful to be reminded that that was going to be a possibility again."
Urquhart was sympathetic to Brennecke's and Sawyer's desire to create the next generation of classic RPGs. More than that, he missed those games, too. "You can definitely have a party in Mass Effect, Dragon Age, and games like that, and I love those games. But it's different in Infinity Engine-style games: How you have that control, that tactical control over a party. The other thing that's really cool about those games is the 2D-rendered [artwork] aspect. You can have this really immersive world and graphics, tied to having that Dungeons & Dragons style of party, on-screen all at once."
Pitching Obsidian's Kickstarter project as an Infinity Engine-style RPG made sense. Its foundation dated back to Baldur's Gate in 1998. What Brennecke needed next was a hook.