How Double Fine used their creative juices to survive


In their GDC session "Creative Panic: How Agility Turned Terror Into Triumph," Double Fine's Tim Schafer, Brad Muir, Lee Petty, and Nathan Martz talked about how they used a two-week break during Brütal Legend's production to focus on something entirely different, and regenerate their creative resources while doing so.

This process of Amnesia Fortnight led to the creation of various prototypes, four of which later saw the day of light as Trenched/Iron Brigade (Brad Muir), Stacking (Lee Petty), Once Upon a Monster (Nathan Martz), and Costume Quest (Tasha Harris, who couldn't make it to the session).

As Tim Schafer put it, it's becoming harder and harder to sign big games with publishers. Publishers are averse to new IPs, they want to own their IP -- which Double Fine learned the hard way with Brütal Legend -- and they tend to want to remove unknowns. At the same time, it's exactly those unknowns that drive Double Fine to create their games in the first place.

Amnesia Fortnight was inspired by director Wong Kar Wai's infamous production of his film Ashes of Time. The extended shooting period of this film, which was shot in the middle of the desert, exhausted both the crew and the actors. Eventually, Wong Kar Wai took a break from production to allow everyone to restore their health and minds. During this break, he conceived Chungking Express and Fallen Angels -- originally intended as a third story for Chungking Express -- which ended up becoming some of the director's most beloved work.

For Double Fine, their 2007 Amnesia Fortnight session (a second Amnesia Fortnight followed in 2009 after Brütal Legend shipped) consisted of two weeks where small teams would create working prototypes for complete experiences, using the adaptable Brütal Legend engine. They could go nuts with whatever they thought of, which lead to a morale boost for the teams and allowed employees to tackle new roles for a change. Little did they know at the time that this process would end up saving the studio.

After Brütal Legend shipped, work began on Brütal Legend 2. Because Double Fine had expected a sequel to be a safe bet, the sudden cancellation of the title by Activision meant that the studio's future had suddenly become very uncertain. However, they still had all these working prototypes from Amnesia Fortnight sitting around, which were already functional enough to demo and pitch to prospective publishers.


Lee Petty, art director for Stacking, talked about how German expressionist film (e.g., The Cabinet of Dr. Caligari, Nosferatu) inspired him to create the odd little title. These films were known for using highly stylized stage design to get the most out of their small budgets, and the genre's silent film treatment even ended up being used in Stacking to save on voice acting costs.

Eventually, Petty settled on the Russian dolls concept. They were a little bit creepy, but not too creepy to be useful for an ageless audience. The problem was, what can those dolls do in a game? One of the ideas was to allow them to perform Hat Combat, something that according to Petty had "clearly been done too much already." Another idea was to have them fight nut crackers of a German invasion during an industrialist era. In the end they settled on the stacking mechanic to keep things manageable.

The small budget the team had to work with meant they had to focus on the core idea of stacking, and relate everything in the game to this mechanic. It also meant they had to focus on creating a lot of personality for something that is hard to create a personality for, and to keep in mind that expectations for such a title would be high because of the Double Fine name. In a nutshell, they had to be as economic as possible with their resources and try to get the most out of what they had.

One of the ways they did this was by having one character model for the entire game, the Russian doll, and rely on different textures to create personality. The power of accessories came into force by adding things like hats and canes to characters to make them distinguishable from one another and add personality. Still, it was hard for the team to create personality for a wooden object. They played around with facial animations for a bit, but it ended up being way too creepy on a doll. Petty's advice to other designers was to torture your animators, because they are really clever. At one point, he asked them to create a dog doll who could scratch his ass on the carpet, and the animators pulled it off.

There were other challenges for Stacking, such as how to add replayability and accessibility to an adventure game. The focus on their small budget and having to be economic with their decisions led to the multiple solutions of the puzzles in Stacking. They allowed players to have multiple ways to solve a puzzle, so they wouldn't be denied progress by doing something that would make total sense within the game to do, and the option to unlock things by finding all the solutions added replayability for those who craved it.

One of the lessons the team learned from the Stacking prototype was that you should embrace limitations, as they ended up leading to creative solutions. Another lesson was to find the essence of your game, and to stick with that core essence no matter what. Something that was also a driving force for Once Upon A Monster.

Once Upon A Monster

Nathan Martz was inspired by Tsutomu Kuomo's LocoRoco, a game about music and happiness. It made him think about his childhood heroes, such as Bill Watterson and Jim Henson, and about making a game centered around the joy of making music and tapping into your inner self. It ended up becoming the prototype Happy Song.

Happy Song was to be an interactive musical toy where you could create your own personal happy song with the help of monster friends. Besides, Martz wanted something joyful to work on after all the time he spent on writing dismemberment code for Brütal Legend. After that title shipped, Happy Song was picked to develop further alongside Brütal Legend 2. However, when the time came to think about how to pitch such a game to "the suits," he felt it wasn't commercially viable enough as it was. The challenge became to adapt the prototype enough to get it signed with a publisher, without losing its core essence.

By this time, Kinect had released. A new platform with new opportunities and a strong demand for new mechanics, and a family oriented product to boot, made it the perfect match for Happy Song. The question became how to make that work for a game centered around the joy of music. In order to do just that, they changed it from the joy of music to the joy of interacting with others, but still with monsters.

This led to a renewed focus on interacting with characters. Multiple physical activities would let you interact with the monsters, while "the magic of Kinect" allowed for things such as monsters being able to see and interact with the player through the camera. Eventually they decided on turning this interaction into a cooperative experience as well, making all activities for two players so that parents and kids could play with each other while playing the game. Happy Song became Monster Party.

A short while later, Martz read about WB Games entering a partnership with Sesame Street in 2010. Of course, this was an almost perfect match for what Monster Party was already about. The idea of focusing on the joy of interaction was changed into the joy of helping others, to match the educational aspect of Sesame Street, and the game itself was turned into a collection of stories in a storybook to provide context for players.

While Monster Party previously had a disembodied narrator who would provide exposition and instruction, Once Upon A Monster turned Cookie Monster and Elmo into narrators players could interact and connect with more easily. However, they didn't want the game to become a pure cash grab or a game version of Tickle Me Elmo, but at the same time Sesame Street's educational aspect would be boring in a game if they focused too much on numeracy and literacy training. Instead, they focused on the show's emotional curriculum and the joy of helping others.

The lesson the team learned throughout the process was to know your vision for your game, to keep it simple enough to communicate (the product's experience), and to stick with it no matter what. Although the road from Happy Song to Once Upon A Monster was full of slight changes, the vision for what kind of game they wanted to make remained the same.

Trenched/Iron Brigade

Brad Muir's inspiration for Trenched came from the core mechanics in other games, like Mech Assault and the tower-defense genre. At the time, co-op also became increasingly popular, so they wanted to do something with that as well.

For their prototype, they envisioned a third-person action game in a steampunk Britain setting with mobile factories and some sort of metallic zombies, and it had a title so awful that Muir asked everyone in the session to never repeat it.

This posed a problem, though, as other Double Fine games are all very unique in their approach while the Trenched prototype was a mashup of existing mechanics and genre conventions -- not very Double Fine at all. It raised the question of what actually makes a Double Fine game unique enough to be identifiable as a Double Fine game. In other words, what makes it fit the Double Fine brand? Muir initially came up with three important aspects that people expect from a Double Fine game: story, funny cutscenes, and characters.

For the story (and setting), originality is key. As a team effort, a lot of the ideas for the prototype were pretty out there. The artists came up with Castle Robots (giant castles that were also robots), Bosch Bots (inspired by Dutch painter Hieronymus Bosch), and even Founding Fathers who would pilot robot heads based on their heads.

Eventually they settled on a makeshift World War I stetting with television set smoke monsters as enemies. And a boat with legs, because it's awesome. The WWI setting led to the creation of the mobile trenches, while the TV smoke monsters were later turned into the monovisions we've come to know and love/hate.

For the funny cutscenes aspect, Muir found that it's not the cutscenes per se, but the personality that matters the most, so he involved the whole team to add more personality to the prototype. One of the inspirations was found in the Man's Life magazine covers, where burly men fight for survival against tiny animals and some of the artists' own versions ended up in the game. Someone came up with the idea of creating an interactive version of the aircraft carrier deck, where people could run around and salute with a button. Despite his initial reservations, Muir let the team member create it within a few days and so many people ended up having fun with the silly saluting mechanic that they kept it.

The aspect of characters remained the same: characters are simply important. But how do you add characters to a game where you pilot mechs, which generally are just hulking robots with very little personality of their own? To this end they opened up the top of the mechs, scaled down their size a little so you could see your character on top of it, and as a result they could show your character doing little animations.

Frank Woodruff, the commander of the Mech Brigade, was to be confied to an Iron Lung. At some point they even had him in a Warhammer 40K Dreadnought or StarCraft Dragoon type of mech, but he ended up being a resident on the carrier he commanded.

What became evident to Muir in the process was that another aspect of the Double Fine brand is to surprise your fans. Nobody expected a shooter, so the act of actually making one fit the surprise aspect of other, less conventional titles, that the studio released.


The Double Fine folks at the session recapped their experiences of the Amnesia Fortnight experiment by stressing how the agility of their workforce led to the survival of their studio. By producing the prototypes which could later be pitched to publishers as demos, and turned into full games relatively quickly, Double Fine essentially managed to create multiple streams of revenue and stayed afloat instead of being forced to close when Brütal Legend 2 was cancelled. For them, diversification created opportunity instead of a sense of aversion towards the concept, and it taught them how important it was to clearly see, and stick to, the core principles of their games -- creativity being paramount among them.

This begs the question of how many other amazing little prototypes they still have lying around, awaiting the chance to find their way into the public's hands. Now that the Kickstarter success for Double Fine Adventure has been so tremendous, that game will become a far bigger enterprise than the studio had initially expected. Since Double Fine was able to release four downloadable titles in the same year, however, I wouldn't be surprised if we'll see some more Amnesia Fortnight games before Adventure is released.

You are logged out. Login | Sign up



Maurice Tan
Maurice Tan   gamer profile

or on the banner above! Hai! You may know me as Professor Pew in older times. If you see a cblog here from beyond September 2010 or something: that's from the cblogging days and not anything ed... more + disclosures



Filed under... #Developer stuff #Double Fine #GDC #Tim Schafer



You're not expected to always agree, but do please keep cool and never make it personal. Report harassment, spam, and hate speech to our community team. Also, on the right side of a comment you can flag nasty comments anonymously (we ban users dishing bad karma). For everything else, contact us!