Interview: Sam Lake, creative director, Remedy
“We used surface capture for every in-game line – I don’t think any other game has done that before”
Remedy adopts a ‘ready when it’s ready’ approach to game development, focusing its energy on fewer projects and taking its time with its creations. Creative director Sam Lake may exude the same kind of thoughtful confidence, but his frenetic energy is a surprising contrast to the methodical Finnish studio’s outward appearance. Here, we discuss the groundbreaking aspects of Quantum Break and the steep learning curve they created. How have you enjoyed working with live action? We’ve been doing little tests with live action as a component [in previous games] and we wanted to take a bigger leap in that direction. It was challenging and there was a lot of learning along the way. Creating a big game like this is an iterative process. Not every idea you have will work, and part of getting to something good means you sometimes need to change direction. Whereas with the traditional live-action production method, once you’re beyond the script being locked, you’re in a very rigid schedule. So it took a lot of figuring out to get to the point where we had a good plan for creating both elements. Were there any seismic changes along the way? The original idea for the show was that it was disconnected. We’d have a game and a show, and while the show would run parallel during the same crisis and in the same place, we’d have a different set of characters going through their own storyline. Along the way we decided to be more ambitious and have crossover – I felt if we could do this with an interactive narrative then we’re doing something nobody has done before. What additional challenges did that bring about? We kept pushing the show schedule back. We agreed to shoot it as late as possible to when we’d got through prototyping on the game side, had levels in place and scenes that worked. Only at that point did we lock the script and the branching stuff. The facial performances stand out particularly. Coming out of Alan Wake, the whole team felt the story we told would’ve been better if the facial animations had been closer to the actors’ performances. So we made a firm decision at that point that we would push our technology and focus on that aspect. We created a surface-capture studio for facial expressions and built a scanning lab of our own for high-detail head captures. And then, of course, a lot of work went into the actual pipeline of translating all of that data into something dynamic so that the animators can tweak and polish things where necessary. How did the performance phase go? For the mo-cap sessions we had helmet cams, so all the facial data was already captured, but the detail level you can get today with that technology is not as high as we can pull off in the surface-capture studio. So we brought the actors back for all of the close-ups and re-did the performances – they watched the mo-cap performance and then acted it out again. And when our actors came to Finland to be scanned, we had them visit a dentist where we took moulds of their teeth so that we could get a model that we could scan and put into the game. We used surface capture for every in-game line as well – I don’t think any other game has done that before. How difficult was it to balance all of the time-power dynamics involved in the combat? It was a long journey! There were many prototype time power ideas added or abandoned, and slowly we got to where we wanted to be. Originally the game was just frozen scenes, which is kind of cool but gets old fast. But in our engine, for a different purpose altogether, we had this state-recording function for objects, where you could capture an object in various states. That led to the idea of broken time and individual time lines for different components in a scene. I’m not the technical guy, but there really is all kinds of incredible stuff going on with the special effects. Do any of the time-power ideas you abandoned particularly stand out to you? With the state recording we had an early experiment with letting the player record a sequence of gameplay and then rewind it and play again with the recorded version of the character. But that ended up creating a lot of gameplay repetition, and pushed the game into more strategic territory. For a different game, I still think that was a really cool idea, but we wanted this fast-paced adrenaline rush, and that idea was holding that back. How do you feel about concerns involving reading emails – that it can break up the flow of the game? I understand it perfectly, and it fascinates me: it’s a really interesting part of human psychology, this desire to not miss anything. But at the same time, it’s clearly not forced on you. If all of that was on your critical path, you’d be really frustrated and would want to start skipping it, whereas when it’s made optional through exploration, some players feel compelled to search for it and not miss it. The idea was always to add all of this in there because it makes it richer and deeper, but I feel it’s something for when you replay the game. On the first playthrough you’ll be skipping some of it, but then you can go back and dig deeper.