The Producer Triad

On a Friday in mid-April 2003, Chris Parker played hooky.

His project, Baldur’s Gate III, had been canned. He and the team had spent days wracking their brains for a way to save it. They had hit on what seemed like an attractive solution, one based on precedent.

“Fallout was originally a GURPS title, and something happened with Steve Jackson and he pulled the license from them,” Parker remembered of the post-apocalyptic RPG’s original license and rule set. “We were thinking we'd do the same thing. We'd go and figure out a rules system, and it would have some different stats and whatever else we needed to work around the Dungeons & Dragons system. But that didn't sound like a whole lot of fun because we'd then have to invent the entire world behind it, which, oh, okay, that's a lot of work.”

Since Interplay seemed to have given up, Parker gave up, too. He spent that Friday on his couch playing video games. Hours later, the phone rang. “Feargus calls me up and says, 'Hey, I just wanted to let you know that I turned in my notice, like, twenty minutes ago. They're escorting me out of the building. You, Darren, and I should talk soon.' I said okay.”

Early the next week, Darren Monahan, a producer at Black Isle alongside Parker, gave notice to Interplay. Parker held off for a few more weeks. Interplay’s executives were beside themselves over Feargus Urquhart’s resignation and were on the hunt for any signs that he was attempting to poach staff from Black Isle. “Both Darren and I were offered Feargus' job from different executives at the company,” Parker said. “They asked us to stay and hold Black Isle together. We were like, 'Nah, it's cool,’ and thinking, We're going to go off and build a company on our credit cards,” he said.

One evening after work, Parker and Monahan met Urquhart at his house and began formal talks for starting a studio. All agreed that their venture needed to be a multi-project studio. “That meant we needed two people who could run a project, and a third person able to go out and get new business,” Urquhart explained. “That entailed two product managers, Darren and Chris, and a biz-dev guy, me.”

At Black Isle, Urquhart, Parker, and Monahan had were a package deal: All three spoke in accord, and working for one usually meant working for all three. “They’re the producer triad, we used to call them,” said Chris Avellone. “They make the calls on almost everything between them.”

“By and large, he kind of managed me just by letting me do my thing and come to him for help when I needed it,” Parker said of Urquhart’s management style at Black Isle. “That was how we worked together. Since we left, it's been very much the same thing: We were partners back then and we're partners now. I think we build off each other's strengths.”

Urquhart already had prospects for work. He had been building his BioWare rolodex since Shattered Steel and Baldur’s Gate, as well as contacts at other publishers. Their new company would start out as a work-for-hire studio, signing contracts to develop games based on a publisher’s specifications. That suited the producer triad just fine. Their best work at Black Isle had been the Infinity Engine roleplaying games based on D&D settings.

Before parting, the triad made an agreement. They would try running a startup together. If they did not have a source of revenue within six months, they would disband the company with no hard feelings and go their separate ways, each sharing the debt and helping pay it off.

Parker swung by a bookstore on the way home. He and his friends had been producers, but they had never been saddled with the nuts-and-bolts details of running a company such as codifying guidelines in employee handbooks. He staggered out of the store loaded down with books on starting a business, becoming a corporate secretary, and setting up and managing a human resources department. Urquhart was neck-deep in books on tax corporations to determine how best to set up their company, while Monahan researched business infrastructures and drafted a handbook.

Back at home, Parker consulted his wife. They had only been married a year, and he wanted her input before leaving his stable and well-paying job at Black Isle to risk his—and their—financial stability. “I remember she was just like, 'You know what? Just go do it. If you guys fuck up and fail, the worst thing that happens is you'll have to go find a job doing what you were doing before. Tens of thousands of dollars is going to suck, but it won't be the end of the world. Nobody's going to be cold, hungry, or dead. If there are a handful of guys who can start a game company, it's you guys. Go figure it out.'”

The first thing they had to figure out was how to get their company, yet unnamed, off the ground. “Between me, Chris, and Darren, we probably fronted $150,000 pretty much on credit cards,” Urquhart said. “We did the totally wrong thing: We financed the start of our company on our credit cards.”

A few nights later, Urquhart and Monahan drove over to Parker’s house to continue their chat. The southern California weather was balmy, so they relaxed on the patio until bugs chased them back inside. They needed to come up with a name before Urquhart could incorporate, and bandied suggestions. “They varied in terms of being humorous or more serious. I don't remember any of the good ones,” Parker said.

“We listed all these names, and we had Scorched Earth, Three Clown Games or Three Clown Entertainment,” Urquhart remembered.

“I remember the bad ones because we used to joke about naming it things like 'Three Clowns Studio,' stuff like that, which we refer to every now and then when we think we've done something amazingly stupid,” Parker added.

They winnowed their list down to eight names and adjourned for the evening. Parker and Urquhart showed their lists to their wives, and Monahan ran names by his girlfriend. “The one they all said was the ‘least dorky,’ I believe was the consensus, was Obsidian Entertainment,” said Urquhart.

“That was one we were favorable on anyway,” Parker agreed, “just because of [similarities between] 'Black Rock' and 'Black Isle,' and Obsidian has that loose association with the studio we were leaving. We liked that. And I always felt that Obsidian was, I guess, the least stupid of the names we came up with, so we just went with that.”

On the subject of the games they would make, the triad were, as usual, of the same mind. “From the outset, we were only pitching roleplaying games, and we were going to be the best at making roleplaying games,” said Parker. “That's all we were going to do. When we spoke to publishers, we pitched ourselves as 'the roleplaying guys.' We were the guys who built Black Isle Studios. That's what our core competency is, that's what you should hire us for, and we will go make amazing games for you in that genre.”

Urquhart had been a student of the industry. One of the chief reasons for Interplay’s worsening finances had been Fargo’s insistence on sticking with PC games when consumers were adopting consoles such as Sony’s PlayStation 2 and Microsoft’s Xbox in greater numbers. He was determined not to make the same mistake.

“By the time we started Obsidian in 2003, we really understood that the future was console [driven]. I'm not saying I personally thought consoles were the future of gaming,” Urquhart clarified. “It was just the reality of the industry. There was the much lauded, over 2001 and 2003, death of the PC as a gaming platform. If we wanted to make RPGs, we were going to do that on consoles.”

New Republic

The clock was ticking. Urquhart, Parker, and Monahan incorporated Obsidian Entertainment in June. That left five months to secure work, hire employees, and make games, per their agreement.

Building a team would be easy, when the time was right. Employees at Black Isle had been calling them to ask when they could join up. Well aware of the paperwork Urquhart had signed, the producer triad chose their words carefully. Stay employed at Interplay. Keep collecting health insurance. Obsidian existed only on paper. They had no work, and no money.

Chris Avellone had been ready to resign the moment he’d caught wind of Urquhart’s resignation. Urquhart had been the only manager he’d had in the industry, and he was fiercely loyal. Urquhart cautioned him to stay put. He didn’t even have a company for Avellone to join just yet. Avellone waited, working on Project Van Buren—also known as Fallout 3—until he got the all-clear.

“During the Interplay resignation interview, I brought up the Baldur’s Gate III cancelation and the other troubles the studio was having as my reasons for leaving,” Avellone said. “The HR department didn’t seem to be aware of these issues enough to understand my concerns or why it would matter to me, which was further proof it was the right time to leave. No one likes seeing their work flushed for careless reasons, especially years of work. I know they questioned me quite a bit about my resignation and if Feargus had approached me for work. They were less interested in why I was leaving versus if Feargus was poaching people.”

Avellone was not the only developer who departed. Around the same time, Chris Jones, one of Black Isle’s senior programmers, jumped ship to go work with the producer triad. Jones and Avellone joined Obsidian early on as co-founders, bringing the management team up to five. Jones would take the lead on technology. Avellone was a natural choice to head up design efforts. “We wanted to make role-playing games and see what Black Isle would have been outside of Interplay,” Avellone said.

Obsidian’s crew of five made Urquhart’s 450-square-foot attic their home base. They bought folding tables and folding chairs from Costco and swung by Fry’s, a boutique electronics store popular on the west coast, to pick up PCs and monitors. Urquhart, Monahan, and Parker carried the loads on their credit cards. They lugged their equipment upstairs where they unfolded tables and hooked up computers. Urquhart’s wife, Margo, then in her third trimester, listened from downstairs as the guys stomped around. “That was one of the reasons why we needed to buy office space: Feargus was about to have a kid,” Parker said.

There were other reasons to make Urquhart’s attic as brief a pitstop as possible. Folding tables bowed under the weight of the twenty-four-inch CRT monitors—around eighty-five pounds each—they had purchased at Fry’s. Parker went to Home Depot, had them cut six one-by-six slabs of plywood, and placed one piece of plywood on each folding desk for support. While the attic had electricity, there were no ethernet ports to connect to the Internet. Monahan had to run cables down through the floor.

“We got our first server, and it was loud,” Urquhart said. “Before she was born, it went into my daughter's room because it was this super loud Nas from Dell, and there was an ethernet connection in the room. We had this nice nursery with a crib and all this baby stuff, and this frickin' Nas just [roaring] the whole time.”

The attic wasn’t all bad. Aside from snarls of cables that made their inaugural office look like a ship from The Matrix, the space was air conditioned. They got to work writing up proposals for roleplaying games and firing them off to publishers. “That's a little bit nerve-wracking because you want to make something that's super cool and compelling, but it's hard to know if you're doing that,” Parker said.

Firing off pitches gave way to yet more ideas for pitches. One was an outline for a roleplaying game that crossed zombies with biohazards, a concept inspired by Capcom’s Resident Evil series. Publishers didn’t bite.

Suddenly, one of Urquhart’s contacts got in touch. Simon Jeffrey, LucasArts’ chief operating officer, had been a fan of Black Isle’s RPGs since Baldur’s Gate. He invited Obsidian’s co-founders to make a pitch. “We first pitched him on using the Dark Alliance engine to make a Star Wars game, basically an action-RPG with lightsabers. I still think that would be cool,” Urquhart said.

Jeffrey said he might have something better. The next day he phoned Urquhart and turned the tables by pitching Obsidian a project LucasArts had coming up. That project was Star Wars: Knights of the Old Republic II. Urquhart frowned. “To be honest, at that point I wasn't aware of what Knights of the Old Republic was because we were so busy. I knew it was something BioWare was doing, I knew it was an RPG, and I knew it was Star Wars, but I didn't have a lot more information than that.”

Urquhart conferred with Parker and Monahan, who immediately grew excited. Knights of the Old Republic was a third-person RPG where players would get to make choices that moved them closer to the Light or Dark side of the Force. Urquhart was intrigued. Most Star Wars games were action titles, and put players in the role of heroes such as Luke Skywalker. KOTOR would be set 10,000 years before the events of George Lucas’ prequel trilogy, letting players tell their own tale. He called BioWare co-founders Ray Muzyka and Greg Zeschuk, who encouraged Urquhart to take the project. Then he called Jeffrey, who asked Urquhart to draft a simple pitch just so LucasArts had one on record. They banged out a three-page document. Jeffrey accepted.

“I don't think it was make or break, but I will be absolutely upfront in saying that I don't know if we would be the company we are today if we did not get that deal,” Urquhart said of Knights of the Old Republic II. “It was high profile, it showed our strengths, it let us make something awesome, it was a sequel built on previous technology, and it just let us shine as a studio that was just starting.”

Obsidian’s triad saw only one problem. The original Knights of the Old Republic was still in development for the Microsoft Xbox, but BioWare and LucasArts wanted the sequel ready to launch within a year so the franchise would have momentum right out of the gate. Obsidian would be working on a sequel while the first game was still coming together, giving them a shaky foundation to build on.

Despite their history with the Black Isle producers, BioWare’s developers may have harbored doubts as well. Two developers flew down from Canada to Orange County, California, with the source code to Knights of the Old Republic for the guys to use. “These two guys had to come with the source code for their game that they're super, super proud of and excited about, and go up to Feargus' attic where we've got all of this garage tech set up, and give us all of their tech,” Parker recalled. “They just looked completely baffled. We were kind of baffled, too. We're like, 'Yeah, whatever. We're going to make the sequel to KOTOR, in Feargus' attic, working on plywood.'”

Parker’s confidence faltered when he booted up the source code on an Xbox development kit, a modified console designed to test works in progress. “I remember sitting on my bed and playing it, and I remember thinking, Holy crap. What have we signed up for? This game is so damn good,” he recalled. “We're never going to get this thing done in a year. So, yeah. Don't sign up [to make] a sequel until you've been able to play the first game, I guess. That's the lesson learned, there.”

Ownership

Signing contracts with LucasArts gave Obsidian’s co-founders the funds they needed to move out of Urquhart’s attic and into a small office in Santa Ana. The time had come to fill out their ranks for Knights of the Old Republic II ahead of the start of development that fall. Their first two employees were programmer Dan Spitzley and artist Aaron Meyers, both seasoned developers from Black Isle.

The company’s payroll was precarious. Like most developers, Obsidian would be paid by LucasArts according to milestones: Submit work—code, art, and playable builds of Knights of the Old Republic II—and receive a payment. Until milestones rolled in, the owners improvised.

“We were all taking money out of our bank accounts, putting it into the bank, running it through payroll, and paying taxes on that money to give it back to ourselves,” said Parker. “We started paying ourselves back after three milestones or so, so maybe by the time we got into 2004. I think it was all paid back by the following spring, so within a year we'd managed to pay back from what we put on our credit cards.”

Spitzley and Meyers slipped naturally into Obsidian’s culture. Not coincidentally, coming into work every morning felt like going into the old Black Isle wing of Interplay. One of the company’s earliest hires had no formal experience making games, but more than made up for that deficit with a passion for the types of products the team wanted to make.

“I was a huge fan of Baldur’s Gate, Icewind Dale, and Fallout. I'm a big roleplaying-game guy,” said Adam Brennecke.

Brennecke grew up a well-rounded gamer. He jumped from Black Isle RPGs to first-person shooters like Quake. Addicted to Quake 2, he experimented with level design by downloading free tools and assembling maps. By the time Quake 3 rolled around, his skills suffered. Other amateurs were painting detailed character models and environments that made his look like stick figures playing on scratch paper. Brennecke was no artist. Eyeing programming, he attended DigiPen University in Seattle and graduated with a degree in game programming. He scoured job boards on Gamasutra, an industry news site targeted at developers, and found a job ad for a small company called Obsidian Entertainment.

“At the time I interviewed, they didn't have any games announced. There were rumors they were working on KotOR II,” Brennecke remembered.

A few of Obsidian’s co-founders drove to the airport to pick up Brennecke. As the car pulled onto the freeway, they asked Brennecke if he had any questions for them. He asked what projects they were working on. They responded by asking if he could guess. “KotOR II?” he ventured. “Maybe,” they replied.

Back at the office, Brennecke impressed the guys during his interview. They gave up the goods: Yes, Knights of the Old Republic II was in development, and was due for release in 2004. Turnaround would be tight.

Brennecke started soon after and joined KotOR II’s programming team. He was hired on as an intern and scripter, a programmer in charge of coding routines for graphics and audio and defining data architecture. Despite his low position on the totem pole, Brennecke was amazed by the freedom he was given. “I did a lot of the cutscenes,” he said. “One thing they really give is ownership. If you want it, if you work hard for it, you can have it.”

“We work hard to give people the tools and ownership to make stuff, and not get in their way too much,” added Urquhart, who named ownership as one of Obsidian’s cultural pillars. “There are times we have gotten in their way, so I'm not saying we're perfect. But I want to say that if you ask most of the people who work here, who are not founders, that they would feel they were given a lot of leeway to do what they wanted to do, maybe more so than they would have gotten at other studios.”

Obsidian’s freedom came with challenges. BioWare had released Star Wars: Knights of the Old Republic that July, and wanted the sequel within twelve to fourteen months. KotOR II would run on the Odyssey Engine, another proprietary platform built by BioWare. Obsidian’s programmers found themselves buried in an unfamiliar technology environment on top of a growing list of tasks directly related to progress on milestones. Development hit several snags, including feature bloat, known in the industry as creep, and bugs that remained in the code until after release.

Besides working with new tech and driving hard to meet a tight deadline, the small size of Obsidian’s team contributed to KotOR II’s bugginess. “We were very, very small,” Brennecke said. “I think when I was hired, we had twenty-five people. That was pretty much the entire KotOR II team. It felt like a startup because it was a startup. We did whatever we could to make the best games possible. We never really had the big budgets our competitors had. So, I enjoyed trying to figure out interesting challenges, trying to figure out how to get the most out of every person on the team.”

Fans and critics seemed willing to overlook the more slipshod pieces and parts of KotOR II’s composition. The game sold 1.5 million copies in under two years. Many critics claimed it surpassed the original KOTOR’s writing and character development, extolling it as the best-written RPG since Black Isle’s Planescape: Torment and an experience hampered only by glitches and poor artificial intelligence.

“I think working on a Star Wars game was probably the most monumental thing in my career in terms of having grown up on Star Wars, and loving Star Wars,” Parker said. “Getting a chance to make Knights of the Old Republic II was sort of life-changing for me.”

By the summer of 2004, months before KotOR II shipped out to stores, eighteen of Obsidian Entertainment’s thirty-six employees hailed from Black Isle. Their pedigree made the team suited for their next project. Announced in April 2004, Neverwinter Nights 2 was an Atari-published sequel set in the Forgotten Realms, Black Isle’s old stomping grounds. BioWare had developed the first game on the Aurora Engine, a successor of sorts to the Infinity Engine. Neverwinter Nights sported an isometric view and real-time combat rooted in D&D’s 3rd Edition rules, and Obsidian would build the sequel on the same technology.

Monahan stepped up to produce Neverwinter Nights 2 with a team of ten, including Chris Avellone as creative director and Ferret Baudoin as lead designer. Near the end of the project, another Black Isle alum joined in. “I think Josh was either talking with Feargus about getting a job at Obsidian as a senior designer, but he'd been posting on our forums for a while,” Brennecke recalled of his first online brush with Josh Sawyer. “I was like, 'Who is this guy? He's posting all this stuff about Neverwinter Nights, and Dungeons & Dragons, and all this weird stuff.’ That's weird, having another game developer posting on your forums. He'd post a lot and was very opinionated about things.”

Brennecke was impressed by Sawyer. Based on the arguments, structure, and flow of his forum posts, he was either an expert in Dungeons & Dragons, or a very opinionated fan. His suggestions spoke to an understanding of game design that only an insider could have.

Sawyer happened to be both, and in desperate need of a creative outlet. He had stayed at Black Isle almost out of stubbornness. Fallout 3 was still trucking along, and he was determined to see it through. Interplay had other plans. Every month brought more departures and more rumors that Interplay’s management would shutter Black Isle.

When it became apparent that Interplay’s owners cared much less about Fallout 3 than he did, Sawyer cut his losses. “It was a hope-against-reason situation,” he said of the project. “There was no reason to believe this was going to work out, but I'd been wanting to work on Fallout for so long that I was like, Well, if it's not now, it's not going to be ever, or so I thought.”

Sawyer nearly went down with the ship. Citing loses of over $20 million in 2003 alone, Interplay closed Black Isle Studios that December. Unlike many of his peers, he did not seek out Urquhart, Monahan, and Parker right away. He and Urquhart had butted heads over design during development of Icewind Dale II. After Sawyer left, both he and Urquhart agreed that Sawyer should look for work somewhere besides Obsidian, at least for a time. Sawyer applied to a design position at Ensemble Studios where he hoped to work on the Age of Mythology strategy game. When he landed an interview, the manager who spoke with him insinuated that Sawyer’s resume made him better suited to roleplaying games. Sawyer thanked him for his time.

His next interview, at Midway Games, went much better for both parties. He would lead design efforts on Gauntlet: Seven Sorrows, an action-RPG, alongside Doom and Quake co-creator John Romero, the project director at Midway’s San Diego studio. Romero’s vision was to create a new type of Gauntlet that injected roleplaying and story elements into the traditional arcade, hack-and-slash formula. Sawyer liked the game’s direction. When he wasn’t working on design, he hopped on Obsidian’s forums to talk D&D with other users.

“I kept participating in those threads in part because I wanted to give my perspective of a developer who had worked on Dungeons & Dragons games,” Sawyer explained. “I did it without really asking anyone at Obsidian if they were okay with me doing it. There were certain times where I said, 'Look, guys, these things are pretty hard to implement regardless of what engine you're talking about. These things are hard; these concepts don't come across from tabletop into CRPGs.' Just debating interface things and stuff like that, mostly because I was into roleplaying games and I just wanted to talk with people about them.”

In the spring of 2005, Darren Monahan reached out to Sawyer and invited him to help guide development of Neverwinter Nights 2 as a senior designer. Sawyer hesitated. His team at Midway was in the trenches on Gauntlet: Seven Sorrows. He wouldn’t feel right leaving at such a crucial time—until managers released John Romero and commanded Sawyer and the rest of the team to sew up the game in time for a Christmas release a few months later.

His vision for Gauntlet compromised, Sawyer met with Urquhart. A few days later, Midway announced his departure. “Fearg and I got together and talked for a while. It had been a few years, and we both had a slightly better perspective with each other and just moved forward from there.”

Sawyer got up to speed quickly. Neverwinter Nights 2’s developers had overhauled much of BioWare’s Aurora engine to improve their game’s visuals as well as the tools artists and designers were using to integrate their content. Ferret Baudoin had flipped the script on the franchise’s story: Instead of starting as a powerful character, player-characters would be relative unknowns so that players could build themselves up into powerful heroes. (Baudoin left Obsidian prior to the game’s launch, and Sawyer stepped in as the lead designer.)

Critics were mixed on Neverwinter Nights 2 when it premiered in October 2006. Opinions varied on the story, but most critics concurred that, as with KotOR II, dumb artificial intelligence for companions and severe bugs hurt their enjoyment of the story and tactical combat.

Going forward, Obsidian’s penchant for crafting memorable but glitchy experiences became a bigger issue.

Four Strikes

For Obsidian’s developers, one benefit of working on licensed properties was the opportunity to play in some of pop culture’s most popular sandboxes. Forgotten Realms and Star Wars were two. Alien, the sci-fi-horror franchise created by writers Dan O’Bannon and Ronald Shusett and held by Hollywood powerhouse 20th Century Fox, was another.

“The Aliens RPG was a game that we were really, really excited about for two reasons,” said Parker. “One, it was going to be built on our own engine technology, which was going to be a big step for us. Two, we loved the Aliens license. That whole world, at least for most of the movies, is actually a really good, really cool thing. You've got this mix of science fiction plus horror, and you've got really cool aliens.”

Announced by publisher Sega in 2006, Aliens: Crucible was a third-person RPG for PlayStation 3 and Xbox 360 whose story was set between Alien, the first movie in the franchise, and its sequel, Aliens. Drawing on the franchise’s horror motifs, Crucible took place inside a space colony infested by Xenomorph aliens. Players would control a party of survivors, each with customizable attributes and unique characteristics that would force players to depend on each colonist’s strengths throughout the adventure.

Crucible challenged players to care about their party by making death permanent. One scenario might see a facehugger alien attack a colonist. The player would have to decide to put them out of their misery, stick them in a sleeper and wake them only when in dire need of their abilities, or wring every drop of usefulness out of them until they popped.

According to Urquhart, Aliens: Crucible had turned a corner just before Sega pulled the plug in the summer of 2009. Obsidian had jumped several technical hurdles and submitted a feature-rich prototype on a Friday. The following Monday, Sega cancelled the game. Sawyer asserted that Sega had cancelled it because Obsidian’s developers had been falling behind on deadlines due in part to building their in-house Onyx engine in parallel to designing the game.

“When it was cancelled, we had a full, playable demo,” Parker said. “We had party members that ran around, and different kinds of weapons and stuff like that, and there were obviously aliens, and bad people you could fight against. It was a pretty straightforward, science fiction/horror roleplaying game. That's one I'll be sad forever that we didn't get to make.”

“I was on Aliens, my favorite project here,” added Dimitri Berman, lead character artist at Obsidian since 2006. “I put a lot of work into that game and was very sad when it got cancelled. I believe that game had a lot of potential.”

At the same time, Obsidian juggled another project: Alpha Protocol, an espionage-themed RPG. Alpha Protocol had sprung from a lunch conversation between Feargus Urquhart and Jones. When Urquhart asked Jones what type of game he would like to make, Jones suggested an action-spy thriller with RPG flavor. “The first pitch was the action hero RPG. It very quickly came to this idea of making an action-espionage RPG based on James Bond, Jason Bourne, and Jack Bower,” Urquhart remembered. “That was the weirdest thing: While we were talking about this, we realized their initials are all J.B. We thought about it, and we think it's just the strength of the ‘Buh’ in their names, and the shortness of the names.”

Urquhart dashed off a three-page proposal for a game called Alpha Protocol. Players would control Michael Thorton, a break from all the “J.B.s” sneaking around the spy scene. “Sega really wanted to come up with a better name than Alpha Protocol, but I was pretty purposeful in the choice of the name from the standpoint of it was short, it sounds secretive, like, ‘Oh, it's a protocol!’” he said.

Sawyer liked the idea, but had some reservations. “I'm not going to say I was the only person who was nervous. I think the owners realized this was an area of concern, and we tried to hire up for it, but I think those were two big problems that plagued Alpha Protocol through its development: For a long time, there was not a great sense of clarity about what kind of spy game we were making, and what aspects of spying we were trying to emulate.”

Chris Parker and Chris Avellone came aboard Alpha Protocol’s team in 2008, approximately halfway through development—Parker to lead production, Avellone to head up creative. Both agreed with Sawyer that Michael Thorton’s story and abilities needed wrangling. To cull ideas, Parker and the designers took over a meeting room and covered a whiteboard in pitches for what they wanted to do as a spy. “We distilled those down and then said, 'Okay, Mike fights, he stealths, and he talks. Can we pretty much say that's all we need to do and not bother with all this other crazy crap?' The answer, basically, was yes.”

Alpha Protocol became a melting pot of concepts. The game’s story has double and triple crosses, torrid love affairs, and assassinations. To add tension to conversations, a dialogue system gives players four ways to respond to characters and only a certain amount of time to choose their response. Responses are short and sweet so players won’t get frustrated by having to read long sentences while the on-screen timer ticks down.

A personality-driven input system further streamlines Alpha Protocol’s conversation system. Players respond by pressing one of four buttons on their Xbox 360 or PlayStation 3 controller. Each direction always corresponds to an attitude: professional, suave, aggressive, or noncommittal. No matter the wording of any branch in dialogue, players can press the button they know matches up with their avatar’s disposition. Some NPCs can be befriended, bedded, or alienated, widening or narrowing the type of mission and plot flow going forward.

Alpha Protocol’s team desired to give players as much freedom in terms of how they played. As players dispatch enemies and gain experience, they increase their expertise in disciplines such as hacking, pistols, shotguns, and martial arts. One player might go in with shotguns and semi-automatics. Another might prefer to operate covertly, sneaking around, karate-chopping guards in the throat, and hacking locked doors.

For Parker, some scenarios proposed throughout development offered too much freedom. Set pieces became a sticking point. In one scenario, enemies patrol beneath the wing of a plane. When the guards walk below the wing, players can shoot the propellers, which fall off and kill anyone in their path. “That was kind of neat-o, but it was also a really elaborate, almost esoteric thing that almost didn't make sense, so we got rid of those,” Parker said.

Alpha Protocol released in June 2010. Response varied. Parker and the team pointed to trouble working with the game’s tech. “It was a lot of work,” admitted Michael Edwards, a senior programmer who specializes in graphics. “We were learning Unreal because it was our first Unreal game, and we had to wrangle that engine.”

While the team’s leads owned their mistakes, they believed stubbornness on Sega’s part contributed to the glitchy state of the game on release. “When I took over the project, we were looking to reboot the game,” Parker said. “You know what publishers don't like? Wasting money on a game that has to be [restarted]. So, they had every reason to be upset with us about that. But they did oftentimes do arbitrary things even in the face of all that.”

Far in development, the game’s producers informed Sega that they planned to cut mini games from the product. They weren’t polished, weren’t fun, and just didn’t belong. Sega’s representatives refused. Mini games had been part of Obsidian’s proposal, so mini games had to stay. “We crunched on the project to get something as good as we could, but at the end of the day it was a little bit janky,” Parker said of Alpha Protocol. “I think that's unfortunate, because fundamentally it’s really strong. I just think it fails on some execution levels.”

Obsidian’s team never paused to take a breath. It wasn’t an option: As an independent studio, multiple projects ensured there would never be any gap in work—or paychecks—after one game shipped. There had been setbacks. Dwarves, a licensed title based on Disney’s Snow White and the Seven Dwarves animated classic, had been in development for a year before Disney decided there was no need for a Snow White prequel and cancelled the game.

The company rolled on. As Alpha Protocol wrapped up, developers used the Onyx engine they had built for Aliens: Crucible as a framework for Dungeon Siege III, an action-RPG in the vein of Diablo. Another team led by Sawyer and Avellone was already at work on Fallout: New Vegas.

Sawyer’s dream of working on a Fallout game had been put on hold after he had left Black Isle. He hadn’t been willing to let it go in the years since. “To be honest, after I left and was at Midway, I wrote a Fallout pen-and-paper system to run tabletop games with people at Midway,” said Sawyer, project director on New Vegas. “I still really wanted to play around in that universe. Being able to make New Vegas, I thought, Wow, I never would have dreamed I'd have this opportunity.”

Fallout: New Vegas was a sequel to Fallout 3 by Bethesda Softworks, holders of the Fallout license after they won a bidding war against companies that included Troika Games, co-founded by Fallout creators Tim Cain, Leonard Boyarsky, and Jason Anderson. Sawyer and the team had to work within the limits of Bethesda’s Gamebryo engine. That motivated Sawyer to push New Vegas into new creative frontiers without straying too far from what had made Fallout 3 a bestseller.

“I knew the game would fundamentally play like Fallout 3,” he explained. “The fans of Fallout 3 don't want a radical departure. I thought about how we could make this feel very true to the west coast, California experience. I think Bethesda very wisely, because they knew the east coast, set it right where they were.”

New Vegas’s Mojave Wasteland combined the scope of Fallout 3 with the emphasis on player choice and openness that had come to define Obsidian’s games. Every quest can be solved in a multitude of ways, and choices have lasting consequences. Upon the game’s release in October 2010, many players found New Vegas’s companions more engaging than Fallout 3’s. They also appreciated its darker script that placed a great emphasis on morality—not just the player’s, but that of various factions they could interact with over the course of their journey.

By the end of 2010, many RPG fans had formed a love-hate relationship with Obsidian. On the one hand, the company’s roster of Black Isle alumni had fostered a reputation for creating deep RPGs brimming with captivating worlds, fantastic writing, rich characters, and choices with real consequences. In 2018, video game documentarians GVMERS published an investigation into Aliens: Crucible and lamented the game’s cancellation. “While this game […] would ultimately end up being cancelled at Sega’s behest, what information that has emerged about the game in the years following its demise suggests that Obsidian Entertainment may have had something truly special on their hands. More than just an RPG in an Alien mold, Aliens Crucible would have featured a macabre return to the franchise’s horror roots bolstered by Obsidian’s expertise in crafting meaningful player choice, making for what could have been an enticing proposition for both fans of Obsidian Entertainment and the Alien franchise alike,” per the documentary.

On the other hand, Obsidian also put out absurdly, frustratingly buggy games. Developers learned that many fans and critics had begun referring to their company as “Bugsidian” due to the inordinately high number of glitches—some minor and almost amusing, such as characters’ tendency to float above the ground in New Vegas; others, such as those that caused games to crash or erased the player’s progress, more severe—in every game.

“I would say that reputation probably started with Neverwinter 2,” Sawyer admitted. “KotOR II was pretty buggy, and Neverwinter 2 was also pretty darn buggy. Then having Alpha Protocol and New Vegas come out and be very buggy? That was four strikes.”

“There were things at Interplay that we took for granted, and became apparent when we switched to Obsidian,” Avellone admitted. “One issue was a lot of support we’d come to rely on, such as QA, audio, and IT, were no longer present, so many of those things we had to rely on from a publisher until we grew. Over time, we did get many of those positions filled until we could do support services ourselves, and Obsidian now has a marketing department, which is a good thing. Staffing QA adequately was a struggle throughout our years and lasts to the present day. Testing and fixing RPGs requires a lot of manpower to do properly, and despite the studio’s experience, the scope of the games often resulted in a lot of bugs.”

“There are certain things, in retrospect, that I would not have done,” Sawyer continued, speaking to New Vegas. “The Caravan didn't get done in a good state, the card game that we made up. I would have cut that. There were a few other features, like the disguise system, which was very cool but we didn't have code support for it so we did it all through scripting—that became very buggy; I probably would have rolled that back. I think that with New Vegas, it was a miracle that it got done, but, yeah, it was really buggy.”

To Obsidian’s credit, Dungeon Siege III—and, in 2014, South Park: The Stick of Truth—shipped in much cleaner states. One of the ways Obsidian’s co-founders achieved a higher level of polish on those titles was by pushing for control in areas of production such as quality assurance (QA) that were traditionally in a publisher’s purview.

“Really, what we did was, starting with Dungeon Siege III, we got together and said, ‘We've got to stop this. How are we going to do that? We've got to be more serious about it. We can't blame other parties. We just need to stop,’” described Urquhart. “We have to have better bug-reporting tools. We have to have QA working earlier. We have to make conversations about bugs a part of development and not the icky thing that happens at the end. I don't want to excuse us, ever, but—and I feel like I'm trying to caveat this—I hate to admit that there is a challenge with greater bugs in games that have gotten bigger and bigger, and more complicated. It is now more of a challenge to get bugs out of them.”

The Worst Day

When the TSA security guard motioned, Urquhart stepped out of the airport body scanner and to one side while his carryon bags were searched.

A guard unzipped one bag’s main cavity, looked inside, cocked an eyebrow at Urquhart, then  reached in and withdrew what appeared to be an Xbox 360. It was a large box, but with an extra chunk of plastic clasped to one side. Urquhart tried to explain.

“He knew what an Xbox was, but he said, ‘What's the other thing on it?’” Urquhart recalled, referring to the sidecar attachment. “I said, ‘Well, it's a dev kit. It's used to make games.’ He said, ‘But what's in there?’ I said, ‘It needs more memory and stuff so we can develop games on it.’ He said, ‘Can you take that part off?’ No, I couldn't. And then of course he wanted to know where he could get one.”

Urquhart’s Xbox development kit (XDK) was one piece of Obsidian’s greatest puzzle yet: Developing Stormlands, a massive, open-world RPG that would launch alongside Microsoft’s Xbox One in 2013. Microsoft had yet to issue Xbox One XDKs, so Obsidian put together a proof-of-concept demo using the Xbox 360 kit for Urquhart to show Microsoft executives during his trip in the early spring of 2011.

“That was the only time we put a lot of money into the demo of a game,” Urquhart said.

Urquhart was traveling to pitch the Xbox manufacturer on publishing Stormlands. The team-up looked perfect on paper. Obsidian had the know-how to create a sprawling RPG exclusive to Xbox, and Microsoft had the deep pockets to fund its development.

Noah Musler, a biz-dev guy at Microsoft and a friend of Urquhart’s, had arranged the meeting. When Urquhart finished, Musler caught up with him and the other Obsidian developers to walk them outside. “He said, ‘I don't want to get your hopes up, but I think this is very possible,’” Urquhart remembered of the conversation. “It was very shocking that we'd gone to step two with Microsoft. We moved forward and signed the project.”

Obsidian’s team was elated. After nearly eight years, the company was on the verge of getting the funding it needed to support a huge team capable of creating the mother of all Obsidian roleplaying games. “It was intended to be a big world, and intended to be everything we like to do in terms of narrative depth, story depth, companion depth, choice and consequence, all those things we've kind of built our house on, that was all stuff we wanted to do in Stormlands,” said Parker. “The idea was it would be a launch title on Xbox One, and it would be the most awesome game ever.”

Stormlands’ developers used the company’s Onyx engine as a base. “We were trying to build a more open-world game with that technology, rather than a top-down-style game,” said Michael Edwards. “We were working on weather systems, better tessellation for character detail. We were really trying to extend that engine.”

Obsidian went all-in on Stormlands, hiring dozens of programmers, artists, and designers. Staffing up so quickly and to such a large extent was risky. “One of the challenges was we were staffing the project with more people than what Microsoft was paying because we wanted to keep [our team] working and moving this forward as quickly as we could. We'd never done a first-party title before,” Urquhart explained. “We'd never done a launch title before. These are holy grails you have when you start a studio, so we invested in it.”

Work progressed for approximately nine months. Then Microsoft’s management changed hands. The new guard inspected Stormlands and proposed sweeping changes. “I'm not trying to blame them. They were the ones footing the bill, and they had every right to change it,” Urquhart continued. “But I think it created a lot of chaos. We were not making the same strides. We were not moving forward. There were disagreements on how to logistically make the game.”

One of management’s mandates was that Obsidian incorporate motion controls using the Kinect camera that would ship with every Xbox One console. Microsoft saw Kinect as the cornerstone of Xbox One interaction: Users would be able to launch and quit games, open software such as Skype, and watch streaming apps like Netflix simply by speaking or gesturing. “I think there was a disconnect between what they were expecting and what we could deliver,” said Adam Brennecke. “They were also at a point where they were really trying to push Kinect for their first-party features on Xbox One. We were struggling as a team—and they were struggling, too—to figure out how those features would fit into a core RPG. They didn't want to move forward with it for whatever reason.”

Another mandate was to incorporate raids on a grander scale than any online RPG to date. Microsoft dubbed the feature “the million-man raid.” Raiding, the practice of grouping with dozens of other players to clear an extremely challenging dungeon or boss, was becoming a popular pastime for players of online RPGs like World of WarCraft. Stormlands’ million-man raid would be even better, Microsoft execs claimed, because it would be spontaneous. Instead of organizing a raid, players may be playing the game, venturing through the shared world, when suddenly a massive creature bursts out of the ground. Players would have to drop what they’re doing and rush the monster. To reach it, they must penetrate a thick fog. That fog would be the game’s way of disguising behind-the-scenes matchmaking to break all the players into separate groups of forty people or more, each group facing off against the gigantic monster.

As they fight, each player’s Xbox One records the battle and uploads their footage to the cloud. At the end, an artificially intelligent editing program stitches together footage from every player’s console to create an epic and personalized movie of their raid.

Urquhart and the rest of Obsidian’s leadership did not know whether to take Microsoft’s “million-man raid” concept literally. No server in extant could handle the load of one million players, each in his or her own game instance. That would require servers to  keep track of movements, attacks, life meters, and on-screen population such as characters and items for each player’s game world.

To Urquhart, the budget Microsoft’s previous leadership had proposed became lost in a fog of its own. Obsidian and Microsoft had agreed to develop a project with certain features and at a certain amount. All the changes the Xbox team was proposing would increase that budget. Fine, Microsoft said. They would write bigger checks and Obsidian would hire more developers.

There were other obstacles, too. Microsoft wanted the game to run on Unreal Engine 4, but UE4 had not been implemented in Xbox One XDKs, so Obsidian’s developers were building out Onyx to incorporate more features from UE3.

Parker and Urquhart agreed that their mistake was failing to set realistic boundaries and goals with Microsoft. “I think in some ways, Microsoft wanted it to be every game it could be,” Parker said. “We tried to accommodate that, and tried to accommodate it too much. Rather than being focused on making the greatest game ever, it was evolving into all games ever. It became a very complicated.”

“Ultimately, what happened was, and I don't think through malicious intent on the part of anybody, but it became a gray area. [We were asking], ‘What are we really doing?’” said Urquhart.

On Monday, March 12, 2012, Urquhart’s phone rang. For a moment, Urquhart stared at it. He was expecting a phone call from Microsoft. They had yet to sign contracts for Stormlands. Thus far, all work had been preliminary. This phone call could be an executive calling to give the green light. Instead, Microsoft cancelled the game.

“Stormlands was cancelled right around my birthday. That was pretty awesome,” Parker said.

Urquhart called a private meeting with Monahan, Parker, Avellone, and Jones in a nearby coffee shop. There would be no more milestones, no more payments. The next day, Obsidian would have to lay off nearly half of their staff. Their agenda in that coffee shop, where no staff could accidentally wander into an office and catch wind of impending layoffs, was to decide who stayed and who could go.

“I don't want to say ‘survival of the fittest’ or anything like that, but we have to consider the strengths of the studio,” Urquhart said. “Those are the harsh, horrible decisions we have to make.”

“There's nothing really to do but say, ‘Okay, who can we immediately put on the other project?’” Parker said of the process. “‘Who can help us moving forward the best?’ And then, you just look at the rest of the people and try to figure out some way forward. It's not pleasant in any way. Those days, those weeks, were the worst days that I've ever experienced in my life.”

On Tuesday morning, Obsidian’s co-founders briefed everyone on what was about to happen and told them to wait in their offices. Brennecke sat at his desk, heart pounding, staring at the doorway. Chris Benson, the IT guy, went back and forth, but never entered Brennecke’s office. Benson had been told to send all employees who would be laid off to one room, where the company’s management would talk with them. The others, including Brennecke, went to another room to receive an update. “We were working on one other project, which was South Park, but we had nothing else going on at the time,” said Brennecke.

“We found out at the meeting that people had already been laid off,” Edwards said. “Feargus told us, ‘If you're still here, your job is safe.’ It was pretty traumatic.”

“It's horrible,” Urquhart stated. “That's easy for me to say: That for me, it's horrible. I see it as my social contract with people. That when I hire them, it's my responsibility to keep them employed. I failed, and it's horrible. A lot of times, we can't give them much or any severance, because we can either run the company into the ground, or we can lay people off.”

“It was the worst day at Obsidian,” Sawyer said. “There's no comparison to it. Around forty people got laid off, a project got cancelled. I was seeing tons of my friends, people I'd worked with for a long time, leaving the building. Feargus was wrecked. That was the sort of pit that we had to crawl out of to make Pillars of Eternity. That was the low point of Obsidian's morale, and the state of the company overall.”