dark        

Celeste composer Lena Raine talks video game music, philosophy, and upcoming projects

0

A closer look at how a soundtrack comes together

Celeste was a surprise indie hit when it released at the beginning of this year, and it’s now solidified itself as a bona-fide classic in the realm of twitch platformers. However, the incredibly tight gameplay and damn-near-perfect level design are only some of the factors that make it such a standout title.

Alongside a wonderfully-written story, which tackles themes like depression and anxiety in a way that never feels gaudy or pretentious, the music perfectly mirrors the themes and actions of the player. When you move from scene to scene, the music moves with you. When Madeline, the protagonist, is filled with anxiety, so are the sounds that surround her.

Most impressively, the soundtrack stands on its own, outside of the game, as a genuine album, yet it’s still impossible to fully remove it from Celeste. While the songs have their own story to tell, it’s all so perfectly scripted for each scene that I find myself getting very specific flashbacks to exact moments from the journey when listening.

Lena Raine, the composer for Celeste, put so much heart and feeling into each note. After years of hard work in the industry, she’s finally starting to receive some of the recognition that she deserves. Most people hear a song in-game and assume that the creation process is exactly the same as what you hear on the radio, but there’s so much more to it than that. I had the chance to chat with her last month about writing video game music and just what exactly that entails.

Kevin: In your experience, what's the biggest difference between writing music for a video game and more conventional music? On both a technical and philosophical level, it seems like there would be a great deal of variation between the processes.

Lena: In my eyes, the differences lie primarily in the intended purpose of the music as informed by the venue where it's experienced. Video game music can still be conventional music, but it relies on how you define conventional. One of my favorite methods of defining music comes from David Byrne's book How Music Works. He argues that music genres and styles arise out of the venue they're meant for. Chamber music was made, literally, for a chamber. Religious music is defined by the cathedrals it's performed in. It's slower than chamber music, meant to fill a huge resonant hall. Recorded music is defined by the analogue or digital mediums it's recorded for. To extend Byrne's reasoning, video game music is defined by the hardware and systems it's interacting with.

Everyone generally knows that early game music was defined by its hardware limitations. Various chip sounds, and their limited possibilities, created a palette to write with. Once everything moved to CD audio, however, there arose a common misconception that limitations were over. You could just write whatever you wanted. But even still, for games that wish to engage with reactive and dynamic presentations for their music, there are very real limitations to be taken into consideration.

How does your music react? How does a piece of music being reactive define its composition? Most AAA games have settled into an expected pattern for this. You know what music will probably sound like in an Assassin's Creed title when you go from exploration to combat. On the indie side, we're trying our best to work within those limitations and create scores that use dynamic implementation to the advantage of music. But that also means that game music is something to be experienced within the context of a game. It's defined because it happens while you're interacting with a game, and that's what sets it apart.

[This is easily my favorite song from the album.]

Most video game soundtracks fall into this trap where all of the songs sound like audio clips from the game. You went the extra mile and created something that stands alone as a full-on album. These are fleshed out songs that build into a separate story of sorts. More insanely, the instrumentation is actually interesting from a music theory and composition standpoint. This isn't just basic pop melodies either. You veer off into territory that reminds me of early Faust, Berlin era Bowie/ Warm Jets era Brian Eno, and some of the more poignant acts of today. Why go that extra mile? Why challenge yourself like that? Clearly, it's paid off in retrospect, but what was your reasoning going in?

I think it's extremely important to consider a game soundtrack and a game album two wildly different things. They should be treated separately, but also cared for simultaneously. It's actually a very important part of my process, especially with dynamic music, to create a listenable "album version" of each series of tracks I write for a game. I did this with every area I wrote for in Celeste, as well as my upcoming soundtrack for ESC. It's not just for my benefit, or the soundtrack's, but it also helps the team to hear my intention for a piece of music and how it evolves through play.

When you just write for a series of cues within a game, you risk losing the thread. It can be extremely easy to drop thematic content, write something that is too similar to a cue you just wrote, or is wildly out of place. By checking in and hearing music not just in the nonlinear context of a game, but in a linear and organized fashion, you can verify that what you've written for the game can be considered a complete and well-paced statement.

When compared to structurally complex music, a game level that has an interesting narrative flow will also require a soundtrack with a flow to match. But then it's not just looping wallpaper. It's movements within a larger whole. It's developing thematic material across movements. It goes back to what I mentioned earlier about venue defining the music. If I let myself use the mechanics, structure, and narrative of a level to help define the score, then the music is more interesting.

As for the second question: My general attitude is, why would anyone not challenge themselves? I can think of a number of reasons why you wouldn't (time, money, direction...), but I got into music because I wanted to write things no one else was writing. So I challenge myself because I feel I need to, and to prompt others to join in.

Speaking of audio cues... As someone with limited knowledge of game development, I was curious how that process actually works. Are the audio triggers in the game (when different segments of a song start up during frames of the game) something that you handled the direction of, as the composer, or is that something other folks on the team implement at their own discretion? How involved do composers typically get with the way that their creations are presented in-game?

It really depends on the composer and also how they're interacting with the team. Sometimes there's an audio engineer specifically tasked with planning out and implementing music files as they're delivered from the composer. Sometimes, and especially for indie projects, the composer is handling everything themselves, or they just hand it over to the designers to implement. I've had a number of different situations over the years.

For Celeste, I mostly planned out the cues and how I'd like them implemented. I then sent over the files and worked with our sound designer Kevin to get them into FMOD, the middleware tool we used for audio implementation. Then, Matt and Noel would take the triggers generated by FMOD and implement them in the level scripting. For my current projects, I'm doing all the audio setup myself, working on setting up files and how they trigger dynamically, and preparing them for designers to implement.

[Really, they're all amazing though.]

You say it's a "middleware tool," but, for the sake of folks like myself who are totally lost, what does that actually mean? Everything I've found has left me scratching my head when trying to understand its primary function. Is it just a software specifically designed for implementing audio in a game, via choreographing audio triggers, or does it have other functions as well?

Basically, FMOD is an implementation tool that works in conjunction with the game engine to handle how the game calls and deals with audio. Everything regarding the scripting of dynamic music to the way sound effects play back is handled by the tool (which could be FMOD, Wwise, and a few others, but the first two are the most common). It simplifies things for the programmer or designer implementing audio, so they can just put a simple call in the script -- thing plays or cues here -- and the audio tool handles the rest, from randomizing sound effects, to doing fancy fadeouts, or triggering dynamic music cues on the beat. 

I've always been curious about how a soundtrack is initially built for a video game. As a composer, are you just presented with a demo that gives a basic outline of the game and story, or is it a more reactive process where you're given segments of a game, in increments over time, to specifically write for? What sort of basic blueprint were you presented with when beginning to write for Celeste?

Again, it totally depends on the project. I love to work as hands-on as I possibly can. For Celeste, I had constant build access as the team was working on the game. I was in there, playing levels, even before any final level graphics were created. I'd test out mechanics, give feedback, and plan out how I'd use the structure and interactivity of each level to influence the structure of the music. In previous projects, I've been given a simple list of tracks that need to be written, as well as access to story outlines and artwork from the project to help guide my process. I'm not as big a fan of this, however, since it really makes the process so much less directed. If at all possible, I always want to be playing the game both before and after my cues are in there, to make sure I've got the right vibe and pacing.

One fun exception to this was when I worked at ArenaNet as a designer on Guild Wars 2. I often had the opportunity to write music for quests I was designing, which meant I was creating the content, determining how music would fit into the gameplay, writing the music, and then implementing the cues in my script. Having that kind of extreme hands-on experience really helped me a lot in learning the full iteration loop necessary to get music into a complex game.

[A screenshot from ESC. I have no idea what's going on here, but I'm into it.]

You mentioned your upcoming project called ESC. What the heck is it, and when are we going to hear more? As of right now, all I've really heard is that it's an interactive novel that you'll be composing music for and writing. Is this going to be something akin to visual novels, which focus specifically on narrative, and how will the change in gameplay elements affect your approach to writing music? Is the soundtrack going to be orchestral (like past works), electronic, or something entirely different? No matter what, I'm super excited to shove more of your ear candy into my brain, but what should fans expect?

The general way I've been pitching it is that ESC is an interactive novel about identity, roleplay, and dreams. It takes place within two character perspectives: a recreation of a text-based roleplaying MUCK, and a series of logs from a mysterious individual known as the Navigator. During the MUCK sections, you inhabit a character and type out all of her text input, but everything has been predetermined, a lot like the text chat segments in Superhot. I wrote and programmed the whole game, and of course composed the score, which is just under fifty minutes of music. I also teamed up with 2 Mello (composer of Read Only Memories, as well as a remix in Celeste itself) to do some really cool sound design and Dataerase, a super rad glitch artist, to do background art and graphic design for the project.

The soundtrack is primarily ambient electronic, but verges into a bit of IDM, trance, and some generally weird, glitchy shit. Because everything is focused on embellishing the words themselves, the music often functions more like sound design, until it really needs to crank up the intensity. I had a whole lot of fun writing both the words and music, so I'm hoping people check it out! It's in beta right now, and hopefully releasing later this Spring.

When you say that ESC is an interactive novel, do you mean something akin to Steins;Gate, Danganronpa, and Doki Doki Literature, or is it more like a text adventure? Also, you wrote, programmed, and composed the music for this project? Is this your first time ever helming a video game's development from all angles like that? How do you balance a passion project like that with your other work?

I'd say it's more along the lines of the short stories in Lost Odyssey, or the text-only segments in the NieR series. In fact, the way I present some of the game was extremely inspired by NieR: Automata. Similarly, the game is not short on surprises, and you might find that it is slightly more than I'm letting on.

The way I differentiate between ESC and a visual novel, is that something like Danganronpa or Doki Doki Literature Club focuses more on visually presenting the characters and dialogue. ESC is more about focusing on the text itself, whether it's a roleplay chat log or first-person prose. There's backgrounds, sound design, and music, but the primary focus is the writing.

And yes! I've primarily worked on other people's projects, but this is my first time stepping into the director's seat so to speak. I've got a lot of experience working part-time on my own projects, like my music work before going freelance, or writing a novel while working QA earlier on in my career. I'd say it's less balancing and more sacrificing my own free time to do something I'm passionate about. You could say that's poor work-life balance, but it's a bit hard to draw that line when my hobbies are also things I aspire to be professional at.

[I'm loving the art style for Date or Die so far.]

Celeste has been quite the success story, and you're now, rightfully so, one of the more prominently known composers in the industry. Outside of the upcoming project ESC, what's your next move? Seriously, the sky is the fucking limit for you. Do you ever see yourself pursuing a music career outside of games?

So, regarding pursing a music career outside games, that's something I already started with my Kuraine alias. I released a solo EP Singularity back in 2016, which is actually seeing a vinyl release alongside Celeste! I'm hoping to continue that project and release even more solo music beyond what I'm doing for games. I'd also love to delve into writing music for animation at some point. I have no strong desire to score live-action film, but I've always loved animation from around the globe and am extremely up for trying new things.

As for my next projects, I extremely wish I could dig into those publicly! Unfortunately, almost everything I've got going is under some pretty heavy NDAs. The one project I can talk about is Date or Die, a visual novel by a whole bunch of talented folks that I'm co-writing the soundtrack for alongside Christa Lee, who also contributed a remix to Celeste. Beyond that, I'm working on an unannounced indie game that is extremely charming, and I can't wait to talk about it. I'm also working on a AAA title that I cannot say anything about and is probably a long way off. Stay tuned!

Alright, Date or Die has my attention. What's it about, who else is involved, and when can I expect to hear more? Just going off the name, my brain is already starting to come up with hypothetical plots.

Date or Die is a visual novel about a group of individuals that have been kidnapped by a terrible individual known as The Host and forced to play a televised dating game with life-or-death stakes. It's a bit like The Bachelor meets Danganronpa. There is actually a fully-playable prologue demo available! The demo features Christa's music as of the time of its release, but we're both going to be hard at work to flesh everything out!

Thank you so much for taking the time to be a part of this! Seriously, you're doing exciting things in the industry, and it's always great to get some insight. Video game music is still in this wonderful phase where it's growing and rapidly evolving at alarming rates. It's a beautiful thing, and you're working magic like a freaking wizard. I have no idea how you're doing it, but, please, don't ever stop.

Thanks so much for the great questions! It was a pleasure!


You are logged out. Login | Sign up

 
 

 

TwitterRedditEmailFacebook
 
Kevin Mersereau
Kevin MersereauContributor   gamer profile

Bacon is delicious on everything. EVERYTHING. more + disclosures


 



Also on destructoid: Celeste   (12)   From our database:

  • Celeste sold best on Switch according to its creator - Chris Carter
  • You can enjoy the original Celeste in the new game - Peter Glagowski
  • Review: Celeste - Kevin Mersereau
  • Action platformer Celeste hits PC, PS4, Switch, and Xbox One soon - Jordan Devore
  • TowerFall creator's next game Celeste is looking sharp, coming in January - Chris Carter
  • Celeste is the next pure platformer to keep an eye on - Darren Nakamura
  • TowerFall creator's new game Celeste headed to Switch - Jordan Devore
  • TowerFall developer's next game is a mountain-climbing platformer - Jordan Devore
  • More related stories
    Filed under... #11 bit studios #Destructoid Originals #game music #Indie #Interview #Matt Makes Games #Music #PC #PS4 #Steam #Switch #Top Stories #Video Game Music

    READER COMMENTS LOADING BELOW...


    LET'S KEEP THE COMMUNITY GREAT


    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!



     
     
  •