Pentiment director: experimenting during a game's production is 'extremely risky'
Pentiment director and veteran developer Josh Sawyer has some advice for studios wanting to shorten development times, based on experiences he's had working at Obsidian Entertainment.
Speaking to Bloomberg last week, Obsidian's studio design director said that Pentiment took him and a team of about 15 people three years to make. It's a smaller game in scope than, say, Avowed or The Outer Worlds 2, both made by Obsidian. But Sawyer's philosophy around game development informed the project's schedule from the beginning.
"I've always been working on very tight deadlines," Sawyer told Bloomberg. "And so my approach to everything is usually to be as scrappy as possible and to reuse and be very careful about not reinventing things."
In particular, the director is always cautious about firmly separating pre-production and production. During the former, taking big swings and experimenting with ideas makes sense, as the team is essentially determining what type of game it is making, and how. This, however, shouldn't happen again during the actual production.
"If then in production, when you're supposed to be just making the game, you start changing a bunch of things and experimenting, that's extremely risky, and it's extremely disruptive," he said.
Sawyer tends to be fairly conservative when it comes to proposed changes. If something has already been figured out in his eyes, he's likely to avoid redoing it for theoretical low gains in X, Y, and Z aspects. The potential to screw up the rest of the project is, in his words, enormous. "You're going to waste a bunch of people's work. You're going to eat up a bunch of time."
Regardless, iteration is bound to happen, and there are a number of factors why a game's development cycle might end up going longer than expected. For the former, Sawyer says that being experimental during production is always an option, but it comes with the cost of adding anything from six months to a year and a half in some cases. Knowing where to draw the line and being able to identify when it is no longer worth doing is a judgment call, and an important one.
While making Pentiment, the team found that the third act wasn't holding up. This led the developers to rewrite huge chunks of the game even when they were in beta. This, according to Sawyer, was worth the extra time.
"Most of my projects will slip by three to six months pretty consistently," Sawyer said. "Things do change, and there are complications. And you prototype things, and they work well. Then you implement them, and they dont work the same way. Or Unreal changes something fundamentally in your pipelines, but you need to upgrade the rest of the engine to get all these other things."
Last year, Obsidian released Avowed, Grounded 2, and The Outer Worlds 2. According to Sawyer, this was the result of things not going as planned internally, which taxed the studio's resources. Avowed and The Outer Worlds 2 were delayed multiple times, and both games took more than six years to make. They also missed their sales forecasts, in part, because their budgets were inflated by that development time.
"Other people want to push things, experiment more," Sawyer said. "And that can lead to more interesting games and more interesting experiences. So what do you value more, right?"
