The End of the Conversation... or the Beginning?
My latest project was contributing sound and music to cutscenes in A Virus Named Tom. After putting in an insane -- though ultimately blissful -- number of hours orchestrating music I was given notes that basically required me to start over. Though disappointed, I've learned this sort of request results in an even better draft that overcomes my favoritism of the first.
I dug into the next draft, and true to form created something I liked even better... and got a request for something more energetic. I tried a different approach, creating three abbreviated sketches of what I thought were energetic cues, hoping one would be worth developing further. I got back a brief response saying that none of the examples grabbed the developer's attention.
Now I was feeling tapped out creatively, dreaming of sending back the cutscene with a soundtrack of white noise and being done with it! I didn't want to burn a bridge, as I sincerely enjoyed working on this game.
Still, between my frustration and knowing there were several other cutscenes to score, I suggested to the developer he might want the other composer in the project for this cue. I felt if I couldn't nail it in all the previous drafts then maybe my mojo wasn't right for it.
I braced for an angry response, perhaps an accusation I was sabotaging his project by playing the Diva Card! But instead he was very cool -- actually a bit confused, as he felt we'd barely gotten started in our musical discussion, where I thought we were at the end of it.
We started talking about previous scenes I'd scored and how I need to achieve similar dark tones -- clarifying what "dark" meant, as we had different interpretations on the definition. He sent me cues from notable games illustrating his points.
 A Virus Named Tom
No, I didn't nail the next take perfectly, but the communication energized me. I didn't feel the dreaded burnout of having to create yet another draft that may or may not work. The final draft came together quickly, which ultimately turned out to be the best version that's currently in the game. It was the free exchange of ideas and the open, non-combative manner of our communication that allowed us to maneuver through the roadblock into creating the best audio that served the game.
Lessons learned here?
1. I don't know if this is essential in every case, but asking the developer for audio examples (songs, YouTube, etc.) that represent the mood he wanted really helped me nail down what he was after. Sadly, it has taken me multiple experiences with other leads to make this sink in. Not so much because either us are stubborn; more often they don't want to lock me into ripping someone off, and I want them to think I'm professional enough to handle any challenge they throw at me so they can focus on other things.
Even if they don't want me to rip off anything specific, I can still get an idea of what game music they like. I once went through an insane amount of failed work creating Celtic music for a designer, which I had interpreted as his request. Only when he later sent me a clip from the Pirates of the Caribbean film did I realize he wanted something way more Hollywood and heroic.
2. Developers usually aren't musicians, so there are times you need to clarify terminology when working to understand what they need from your work. I've had folks talk about "chords" when they meant "melodies" -- those things you can whistle. They ask for something dramatic and my delivered music is "too sad". But sad is dramatic, no? The burden of understanding what the dev team wants is usually going to be on the contracted audio expert. Ask questions!
3. Composers working on games need to make clear what sort of work is involved in creating music for a client. The developer of A Virus Named Tom didn't know the many hours I spent writing and orchestrating different mockups to the same scene. Given all the plates he was spinning, I totally understand that he had no reason to.
I should have spoken up at rejection of the first draft and asked what sort of instrumentation (i.e. sounds) should be heard in the music. Violins? Techno beats? In one instance I spent time developing electronic grooves to complement what the other composer was doing, realizing later that I was really expected to handle the more vintage, orchestral-themed music. I went down the wrong path!
|
"I've lost gigs explaining to indie developers three minutes of epic orchestral music created in my home studio is far too much work for $100, yet my rate is far cheaper than hiring a full orchestra for tens of thousands of dollars!"
I don't think most who aren't involved in music understand how much time it takes not just to compose something, but to *learn* how to compose it, to find inspiration, and to constantly revise a piece and build on it until you have something that's actually good... and that's not even counting the recording, mixing and so on, which also takes potentially dozens or even hundreds of hours. There's an understanding that music "just happens", and those who don't play it themselves rarely understand how much knowledge work is involved as well as intuitive creative skill.
You offer good advice for folks working with audio-illiterate devs. However, that's the first time I've heard anyone say that film sound folks have it better than game sound folks. If there was a competition to be the most disgruntled industry audio personnel, I'm pretty sure the film folks would put up a good fight.
"This differs from filmmaking, where by necessity sound is essential on the first day of shooting..."
You're right, of course, that the recording of production dialogue on a film set is absolutely a necessity, but it's generally considered a necessary evil. It's a bit of a stretch to imply that because dialogue gets recorded during production that the use of sound in film is anything other than an after-thought in the vast majority of productions. Randy Thom’s writing on this issue is insightful:
http://www.filmsound.org/articles/designing_for_sound.htm
"musical ideas are often planned early in the process. For example, composer Trent Reznor sent film director David Fincher music while he was shooting The Girl With the Dragon Tattoo."
Citing one exceptional example doesn't make it the norm. There are examples from games too, such as Austin Wintory (a free-lancer, not an in-houser) working on the score for Journey for much of its development (the results of which, I think it’s fair to say, speak for themselves), but that doesn't make it the norm in games either, right?
I think there’s a danger that some people might read your article and think “game audio development = terrible, film audio development = pure ecstasy”, which is both untrue and simplistic. There’s a massive spectrum of audio development experiences to be had in both industries. If we were to generalise then the closest we could get to a truism is “people are generally ignorant of all aspects of audio other than perhaps a naïve understanding of music as an emotional signifier, blinkered by their own limited musical experiences” – the industry at hand is, arguably, irrelevant.
Rob Bridgett’s thoughts (inspired by Randy Thom’s) on the collaboration between game audio personnel and the other development disciplines are relevant to this discussion:
http://www.gamasutra.com/view/feature/130733/designing_a_nextgen_game_for_sound.
php
It’s quite tempting to dismiss Rob’s "manifesto" as only being applicable to AAA game development. But “indie = audio ignorance, AAA = collaboration-gasm” is equally untrue and simplistic – brilliant collaborations and great work happen at all development scales and budgets, even if they are the exception.
But the main thing I’d like to draw your attention to from Rob’s article is this:
“Another advantage in game sound is that the game development process, and the roles of game development staff, are more mercurial than those working in the movie development process - the motion picture industry, particularly at the Hollywood level, has solidified into very rigidly defined jobs at very specific times whereas the game development model is still so young, experimental and flexible in terms of structure that it gives a good chance of getting this right before the development process solidifies.”
On a related note:
http://www.criticalpathproject.com/?v=38403565
For now, there is more opportunity in game audio than in film audio. That’s why I think the film folks would win the “most disgruntled industry audio personnel” competition. And, frankly, they’re welcome to that prize :)
I've found that the iterative process is really necessary for producing great results. If an artist wasn't willing to put in those iterations, then I can't imagine even approaching that same level of excellence. Iteration is communication in a sense., between designer and artist. That said, the more of that communication about feel, aesthetics, and experience that happens in actual words, the less waste there will be.
One tactic that I've picked up for both sound and art is to ask for 2-3 different pieces of really rough work that takes the same time to produce of a single iteration. By favouring parallel over series iteration, the differences and nuances that are created really speed along the collaborative process.
Something else that I've started doing is trying to ensure that communication and production lines stay open between all teams. For example, in my tactics game, I've got four separate factions, each with different look and feel. While developing the primary gameplay prototype using paper and canvas, we were developing the broad story for the world and the four factions. There wasn't much cross-over at that point. Eventually, we began to have problems with our writing and concept art teams; what did the factions actually do, look like, feel like? As we headed into producing our first build, the game design team had some time to work on specific strategies and abilities for each faction, referring back to the writing for inspiration. With a few more iterations of those new mechanics, we were back on track; we had defined a strategic feel for our factions and so we could now use that to fuel our story, art, and sound.
In all honestly, it was a fairly messy process and I would certainly do a better job of it next time. I do believe that it is key to ensuring that your narrative and gameplay levels are in sync - which many companies seem to ignore for the most part. From what I've come to understand from talking to my sound and art guys, it's fairly common for sound to come in at the very end o f the development process and then art somewhere in the middle. Maybe that works for some teams. I'd really love some advice on how to steam-line a team in a fashion similar to how I've been doing it.
Back to the point though, I think it would be greatly beneficial to talk to other members of the production team in those contract situations. Usually the artists will have a richer vocabulary and also be in tune with the artistic vision of the project. Adding them to the conference call with the creative director might just make all the difference. Having that conference in the same room together - with youtube access - will certainly help as well. Being able to hear and discuss something together is invaluable.
Speaking of which, I can't wait for my next audio session with my own lead.
I love Trent Reznor.
As a gamedev/software-engineering student, you make me realise that audio may be harder than it first appears, and have inspired me to put more effort into discovering the in's and out's.
Thanks for that.
If a sound designer is brought in early, I would expect them to play the latest builds as they come out and work on a living sound asset list with the design team, billing within a budget on a per-asset basis if needed. It's good know up front what's expected, but a more customized, flexible and communication driven approach is needed than a hands-off contractor approach if we as an industry expect audio to take a more prominent role.
The message for developers seems clear:
Think hard about what you actually want from your sound and audio, and write that stuff down, and re-read it and tweak it to try and make it clear and precise. Then give it to your sound designer, and after they read it, *ask them questions about it* to confirm that their understanding of the requests seem to be aligned with yours. You may be busy, but if you don't want the sound designers to waste a lot of their time (and your money) making the wrong assets, you need to engage with them enough to efficiently steer them towards the result you want. If you reject an asset they made because it doesn't feel to you like it "fits" with the rest of the game, then its a sign that they don't quite understand what you want, and you need to give them clear feedback that they can act on.