This March the GDC held their first-ever Showcase event. This online gathering provided the game development community a chance to get together and share expert knowledge in the time window usually occupied by the full-fledged Game Developer Conference in San Francisco. Completely free of charge, this event featured talks from the GDC Vault: a repository of lecture videos from the long history of the conference. During Showcase week, GDC curated a selection of lectures from their illustrious history and spotlighted those talks in live-streams accompanied by enthusiastic text discussions in an accompanying chat box. One of my lectures from a previous GDC event was featured during this GDC Showcase, and I was happy to participate in the chat discussion, answering questions and providing additional resources. My lecture was entitled, "Homefront to God of War: Using Music to Build Suspense," and you can watch the entire video of my lecture for free at this link in the GDC Vault.
While the videos remain a part of the GDC Vault, those chat discussions from GDC Showcase are no longer available in any form. I found the chat conversation during my lecture session to be lively, intelligent and tremendously worthwhile, so I preserved the text of the discussion and I'd like to share portions of it here. As we all know, these sorts of text-chat discussions don't really allow for lengthy answers, and often the questions fly by so fast that there's little time to elaborate on ideas. With that in mind, I thought I'd expand on some of the topics brought up during my GDC Showcase session. You'll see that I've organized this article under topic headings, quoting the original chat excerpts and then adding a few additional thoughts to flesh things out.
Before we begin, I'd like to express gratitude to the chat participants whose thoughtful and stimulating questions are included here. Many thanks to Marcelo Manga, Mark Peterson, Mikel Dale, Michael Heller, Misha Samorodin, Navid Lancaster, Joseph Tate, Tremaine Williams, Yigit Koc, and Nicole Lazzaro!
Now, on to the chat discussion!
Marcelo Manga: How much time do you spend on planning the music for a game?
Winifred Phillips: Hey Marcelo! The more time, the better, but we're not often given loads of time to prepare. A lot of the time, we're just asked to hit the ground running. But I like to have time to research and think about what I'm going to do.
Every project is different, as is every development team. As game composers, we don't often get the chance to join the team during the early phase of game development. When we do, it's a rare treat. We're able to contribute in a meaningful way while a project is in its embryonic state. We get to interact with the team while they're engaged in big dreams and starry-eyed ambitions. We can suggest the role that music might play, the style and genre of the musical score, and perhaps even strategize regarding implementation and dynamic structure.
More often than not, we're brought in late in development, and there may not be much time for us to get our bearings and start delivering compositions for integration. Over the years, I've learned to grab whatever preparation time I can get at the beginning of a project and fill it with research, experimentation with instruments, reading about history or culture... anything that can help me to be inspired.
Whether we have the luxury of time or not, research and preparation are crucial at the beginning of a project, and we have to make at least a modicum of time for those efforts before we plunge into the full music composition/production schedule. If you'd like to read more about the process of preparing to begin a project as a game music composer, you'll find much more about this in my book, A Composer's Guide to Game Music (Chapter 7: Preparation and Workflow).
Mark Peterson: Curious about how important a classical music composition background is in this field?
Winifred Phillips: I think this sort of background is important if that's the kind of music you're going to be creating. But a lot of game composers come from the world of pop music, EDM, etc. These require a very different background and education.
Game music composition encompasses a wide variety of genres and styles. A fairly long tradition exists of famous movie composers with highly classical backgrounds, so naturally that impression extends into the video game industry as well. But so many game genres call for popular musical styles that fall outside the "classical" box. Racing games demand pounding EDM soundtracks. Sports games embrace rock, R&B, rap, country, and more. Family games may encompass folk and pop styles, along with more soft-edged electropop qualities. All these musical genres represent opportunities for video game composers without a traditionally classical background.
Mikel Dale: Hey Winifred, how do you make a track as interesting as possible within the confines of interactive music? Keeping the same key/harmony/tempo indefinitely?
Winifred Phillips: I understand what you mean, but I try to resist staying in the same key too long. Game music already has issues with feeling repetitive, so we don't want to introduce any additional elements that feel repetitive.
Michael Heller: On the flip side of foreshadowing combat to come, I often feel the tension is lifted unnaturally when the music changes to a calmer, lighter tone before I would otherwise have realized that the immediate threats are neutralized. What are your thoughts on this? Any tips for avoiding it?
Winifred Phillips: Interesting point. I think that the combat and ambient music needs to be married well in terms of overall tone and mood. It shouldn't feel drastically "happy" and "angry" or the changes will feel artificial.
These two questions are interesting when considered together, because they bring up an important point: when do we aim for a unified musical tone in our game music? And when do we opt for lots of dynamic contrast? These are obviously competing objectives. As game composers, we want to inject lots of drama and variety into our music, if only to combat the repetitive nature of musical scores for video games. Let's consider an example. Say we're asked to compose a two minute combat track that loops indefinitely. We'll certainly want the body of that composition to feel internally diverse, rather than internally repetitive. If the track is going to loop, we don't want to be repeating a track that is also internally repeating. Harmonic modulations, subtle tempo shifts and novel melodies/motifs can help to introduce fresh ideas to a repeating track.
But on the other hand, should this diverse combat track also be sharply differentiated from its associated ambient track? Let's say that this two minute combat track is interrupted when the player completes the fight, and the music immediately returns to its low-keyed exploration ambience. This is inevitably going to introduce a sharp change in mood. We can address this problem in three ways. First, we can make sure that the action and ambient music share some common stylistic elements, such as motifs, instrumentation, and other qualities inherent in the overall genre. Second, we can keep the mood of the ambient music relatively neutral, so that it doesn't feel emotionally jarring when contrasted with the combat track. Finally, we can compose a transition element that will help one track lead smoothly into the other. A short combat stinger might be used at the beginning of an action sequence, allowing the inclusion of a musical build to bridge the ambient music with the combat track. Also, a short combat outro segment might trigger at the completion of the action sequence, with a falling action that diminishes the energy and allows the score to return to understated ambience once more.
With that in mind, here's a question that falls into related territory:
Misha Samorodin: With suspense, is it better to lead the player into an uneasy situation with an escalation or a more sudden burst? Say a player is approaching enemy territory?
Winifred Phillips: I think both these techniques are important, because one doesn't work without the other. One is the setup, and the other is the payoff.
Let's think about this issue from the vantage-point of a mystery author. The surprising twist in a mystery story is never truly a complete surprise. There are always hints sprinkled into the narrative, and the tone of the prose shifts to prepare the reader for the revelation. If a minor character just suddenly shows up and declares that "the butler did it," the reader will feel no sense of satisfaction from that. Likewise, if the identity of the murderer is revealed years later while the main characters are attending a birthday party, the revelation will feel forced and disappointing. It's better if the identity of the murderer becomes known in a moment of emotional crisis, and that the revelation feels like an "aha!" moment because the reader remembers the clues that led to the discovery.
These principles apply to our work as game composers as well. If players are exploring a dangerous area where monsters may suddenly attack, the music should telegraph a sense of lurking dread. Like a clue in a mystery story, this unsettling musical atmosphere escalates the tension until the monster lurches out of the shadows and strikes. Then, the music can execute a sudden burst to heighten that moment of revelation when the enemy shows itself.
Misha Samorodin: I suppose most of the music is composed directly with the game footage, or is there ever any improvisation?
Winifred Phillips: I like to pay attention to gameplay when I'm composing. Games have their own momentum, and I want to make sure I'm composing music that compliments the visual kinetics.
Mikel Dale: Do you often write sound effects? I can imagine you closely collaborate with sound designers for games this big! As you said, a lot of this crosses into sound design so was curious?
Winifred Phillips: I sometimes use sound effects in my music (such as the Doppler effects in Speed Racer). It can help my music marry better with the sound design.
Both of these questions relate to the concept of aural kinetics, so I thought I'd address them together. Music is very good at provoking a physical reaction in listeners. As I mentioned during my lecture at the GDC Showcase, music is proven to provoke some of the physiological reactions we associate with adrenaline rushes. Because of this, music can help players physically internalize the kinetics of the action they are seeing onscreen. The rousing qualities inherent in music produce a state of emotional suggestibility, and the game can then insert itself into that suggestible state and tell the player why this arousal has occurred. In my work as a game composer, I like to reinforce this phenomenon by working sounds into my music that directly relate to the action of gameplay. For instance, in my music for the Speed Racer video game, I included engine roars and Doppler effects to simulate the experience of high speed racing. These sounds help to inform players of the reasons why they are experiencing heightened tension or excitement. The music seems to be saying, hey! Does this feel exciting? It's because we're driving at over 300 miles per hour! Can you feel those turns? Doesn't it feel fast? Ideally, we want players to experience some internal physical echo of the kinetics they're seeing onscreen.
Navid Lancaster: How do you translate the Director's vision to your music?
Winifred Phillips: Lots of meetings and communication is key. If we understand each other, then I have a good chance of meeting the director's vision.
Joseph Tate: As a composer, what do you find are the best notes/directions a developer/director/etc can give you when creating game music?
Winifred Phillips: When we're communicating with the director, we want to make sure we understand music in the same way. Exchanging reference music and discussing musical style in that context can be really helpful.
These two questions approached a similar topic, so let's look at the issue in a little more depth. Whenever we're hired to compose music for a game, we'll inevitably be communicating with a specific liaison chosen to relay our assignments and answer our questions about the project. This might be a dedicated audio team member like an audio director or sound designer. On the other hand, it might be a producer, game designer, creative lead, or other non-audio team member. It's vitally important that we find a way to communicate clearly with the team about the music they want us to deliver, in order to determine the style and compositional qualities they want us to execute. However, every team communicates differently, and their understanding of music may vary wildly. One team may include members proficient in music, who communicate with an expressive musical vocabulary. Another team may have no idea how to talk about what they want from the musical score. Personally, I like to approach every project with the assumption that the team and I don't share any kind of musical vocabulary that will be universally meaningful. When they say they want exciting music... does that mean fast? Loud? Aggressive? Or is it edgy and avant-garde, challenging expectations? Or is it suspenseful and subtle, keeping players on edge? Exciting could have lots of definitions. Instead, I like to start by trading musical examples back and forth in the form of reference tracks. While language sometimes fails us, we always know what we like about music when we hear it.
Tremaine Williams: Since everyone is working remotely these days. What kind of information would be best to send to a composer in order to properly display the direction of the product?
Winifred Phillips: For the question about working remotely – communication again is key. Often we're working remotely as independent contractors anyway. These are really important skills to have.
One of the unusual aspects of being freelance composers in the video game industry is our ability to work remotely with teams around the world. This has been true for a long time. Game developers aren't shy about employing tech into their day-to-day work lives. While Hollywood may still cling to the idea of composers living in LA and working with production companies in person, game developers are perfectly happy with long-distance relationships. I've worked with teams that are less than an hour's drive away, and I've worked with teams on the other side of the world. The working relationship with both has been fairly similar. In terms of what information is most needed to be a successful remote team member... that differs from project to project, and team to team. Ideally, the more information, the merrier. I love receiving design documents, detailed music asset lists, reference tracks, concept art, gameplay capture video, and even builds of the game in progress (when that's possible). But sometimes materials like these are unavailable because the game is in flux, and we just have to forge ahead with whatever guidance our team liaison can provide. In any case, it's important that we ask questions and keep lines of communication open, remembering to be flexible and to adapt to the interpersonal dynamics of the development team. If they are a chatty bunch that enjoys lots of virtual meetings, that&a