Issue link: https://digital.copcomm.com/i/1524863
RL: Yes, there are a lot of different media in the game. It's quite a patchwork of in-world pre-rendered, live-action, game-rendered scenes. Getting them to blend and flow was high on the priority list and took a bit of logistics to work out. I broke down the different scenes into types, based not on the media content, but their narrative significance. From an audio post perspective, live-action versus game-rendered scenes didn't make much difference. However, their significance to the story and how we wanted the player to interpret them was more important. Roughly speaking, we broke it down into these sections. Traditional cinematics: i.e., tell the story from the in-game world perspective. These could be FMV, live action, or game rendered. They needed to sound as close to the game as possible. Robert Eklund and Joakim Kristensen from Red Pipe headed up the cinematics team. I had them working closely with our in-game sound designers, sharing assets and most importantly, testing all their work directly in-game. We kept the render pipeline as close to the games' final output as possible, e.g., all the SFX stems were rendered as 7.1.4 and the music and dialogue were matched to the game processing. Then there were the TV show sequences, like the Mr. Door talk show. Narratively, these are separate, contained scenes, and we wanted them to sound different to the game. For these, I had the guys from Finnish audio company El Camino do the sound. They worked separately from the main game team, so they weren't influenced by the game sound. The render specs were also different—stereo renders instead of 7.1.4. We were going for that authentic TV show sound. Then there is a whole host of video content that plays back from inside the game world. These were just treated as part of the game world sound design and were handled by the game sound team. All the audio was playing back through Wwise, so all the sounds were going through the same mix and render pipeline. This meant we could control all the transition points individually and allow, for example, the game ambience to run through a cinematic if needed to help ease the jump between media. Sonically, getting it all to blend was just a lot of manual tweaking and playing with fade times. The biggest overall challenge was getting the audio-visual sync to work 'OK.' When the game finally came together with all the content, we started having issues with audio dropout and buffering issues. Under the hood, there is a lot of video content being stitched together, and level loading, and once all the content goes in, we are really pushing the hardware—it's quite often that the audio ends up being the symptom of another underlying issue, and the issues manifest themselves differently depending on the hardware. For instance, on Xbox, the sync issues were noticeably worse. It took us quite a while to figure it all out. RR: The player also spends a lot of time in the new "Mind Place" where Saga Anderson pieces together the game's mysteries. How did you design the soundscape for this abstract mental space? RL: There is an excellent GDC talk by our senior technical designer Josh Bell called, "In Your Head: Audio-Driven Everything in Alan Wake 2 and its 'Mind Place.'" I highly recommend checking it out. He goes through the design and technical implementation of this feature. When Josh was running through the talk with us, it was nice to see he adopted a lot of the methodology and practices we'd discussed when approaching the sound design. The first which formed the design basis was the idea of 'concentration levels.' The player has different interaction depths like in a UI menu, and on the audio side, Josh designed these as Saga going deeper into her meditative state. The second was the use of 'action verbs' when thinking about the sound design for each of the profiling sessions. I got the idea for using this method for helping sound designers after reading Judith Weston's book Directing Actors— the idea is to find the right action verb that describes the feeling you're after. With Josh, we went through the characters and started mapping out what it were their core motivations during the profiling. RR: On the topic of directing actors—can you walk us through the process of directing the disturbing vocal performances for the enemy characters, the 'Taken' and 'Fadeouts'? RL: How we should approach the enemy character dialogue was quite an undertaking. Sam Lake, our creative director, wanted the lines from the enemies to tie directly to the narrative of the story—which is a great idea, but enemy dialogue needs to communicate the state of M OT I O N P I CTU R E S O U N D E D I TO R S 99