"We basically develop our games by failing," says Irrational Games' Ken Levine, who credits the unique creative output of his studio to an approach that sees teams explore many concepts before arriving at the strongest one. "We'll just throw things away all the time," he told Gamasutra, when we visited Irrational recently to check in with the team, including director of product development Tim Gerritsen, lead artist Shawn Robertson and art director Nate Wells. The team illuminated for us an exploratory process of creative discovery that doesn't fear risk, nor does it adhere strictly to concepts that haven't yet been proven to be ideal in context: "We try things, and are incredibly open to failing, and learning from that and moving on," says Levine. Although the team favors a collaborative creative approach, where new ideas on the table are fully explored and treated as valid, it's also one where no one team member can become particularly precious about or invested in a concept, even if they've invested the time and attention into building it out. At Irrational, everyone accepts that only those contributions that best serve the final result are kept and elaborated upon, and those that don't work -- or that stop working as development evolves -- must be easily detached and discarded. "Everybody has to get comfortable with throwing their stuff away," Levine explains. Adds Gerritsen: "Our philosophy of process is that process serves development -- it doesn't drive development. If the process isn't serving the goals of the company, then we change the process." Robertson says this can be a little tough in particular for artists who are new to working with Irrational, many of whom will invest time and polish into the construction of assets that later don't work in the game. "Some people take a little more time to get used to that," he says. Given Irrational's focus on strength of product over specificity of process, "you have to get comfortable working with the ground shifting underneath your feet a lot," Levine suggests. "At the end of the day, it's about, 'is this going to be awesome?'" Although for cost efficiency team members must endeavor not to take an excess of risk, ultimately "you can't care about sunk cost." This is the approach that led to such a positive advance reception for the studio's current major project, BioShock Infinite, which the team says is its biggest undertaking to date in spades. "For a long time, it looked like 'BioShock 1 in the sky,'" Levine says of the highly-anticipated 2012 title, which sees players cast as former Pinkerton agent Booker DeWitt as he navigates a sky-borne, Americana-infused city, Columbia, and strives to rescue a mysterious woman named Elizabeth. "It had a very European look; it was very art-nouveau," says Gerritsen. And overall, visual and tonal concepts for Infinite were too much like the starkly-Objectivist, submerged kingdom of Rapture as seen in original BioShock, Levine and his team told us. "That's when we came to the conclusion we had to do some aggressive change to make it a distinct look," explains Wells. That's when the idea to work with concepts of American exceptionalism came along, says Levine -- "all of a sudden the phrase, 'July 4, 1912' came into our head," he notes, describing "this idealized summer's day... all of a sudden, literally the clouds parted." Just as New Year's Eve 1959 in New York was a stylized lynchpin for Rapture, this new date helped spawn the creative tone for Infinite. "It was uncomfortable for us," describes Robertson of the initial transition. "Ken started literally pushing back the clouds -- 'bluer, bluer!' We were really uncomfortable with it, our initial reaction is it's a little bit cartoonish.... but when we saw it in context, that was our a-ha moment." "That searching and that failure was absolutely essential," notes Wells. "By spending some time muddling around... by failing, you find it. That was an idea that I don't think that we could have landed on; we wandered to it. You need that galvanizing idea, but you can't... just do nothing until you have it." "This isn't a studio that says, 'we're going to make a design doc on day one and build that'," Gerritsen agrees. "There are certain aspects that we still don't have 100 percent nailed-down."