Composing
for Interactive Music
You're
walking slowly down a hallway too dark to be certain of what, if anything
lies in the shadows. You're being very cautious, because your health
is very low, and there are no health-ups in sight. (While "death"
in a video game is never the end, it is a major inconvenience and quite
frustrating.) As you go you get the vague feeling that there's danger
ahead. "Of course", you think, games always throw danger at
you in dark places. But there's more to it than that: You feel worse
about this hallway as the seconds go by, your character creeping along
with only a pixel or two on his health meter. Is it the darkness? Perhaps...
but it's been dark all along. Why do you feel worse about this now?
Finally you realize why. The place sounds worse. The music has gone
all spooky and dissonant. You're just thinking "cool" as a
huge blue Orcus with knotted muscles and oozing lips rips your character's
head off. You've just had the interactive music experience.
The
example above is more than speculation. It's in games now. The idea
is simple from a programming angle: if the distance between you and
an enemy is less than x, then send a flag to the music to change its
state. The composer can set the music up to make it sound more dangerous
and spooky. Should be easy! It is easy, for the programmers. For the
composer this is a challenge and a can of worms of unusual depth. This
challenge is the subject of this article.
Let
me clarify: By "interactive music", I don't mean simply music
being used in an interactive application like a game. I mean music that
responds to the state of affairs the user is experiencing. And beyond
mere switching from one song to another, true interactivity implies
more of an interaction between the music and the gamestate, rather than
a direct interaction between the music and the player. Few games ask
the player to manipulate the music (notable exceptions being Parappa
the Rapper and Toejam & Earl). But more and more games
are asking the music to respond to the state of the game, i.e., what's
going on.
We
want interactive music, "we" being the players and the developers.
As players we want it because it makes the total environment feel more
real, more immersive. As developers we want it because it is a unique
new tool for communicating with the player, one which is sometimes the
only one available. It offers us the ability to communicate the nature
of the situation with something other than visuals, something that doesn't
take up any extra RAM. (Once you've allotted for the interactive audio
driver)
Above
all else is that ephemeral joy we get from the way that music can orchestrate
a situation. Like walking with headphones, if the music and the listener
and the situation all match well, a stirring sense of meaning is imparted
to the experience. We feel elevated and glorified. At its best, interactive
music does this even more than static music, because it seems to know
what has happened. The emotional tone follows our experience.
A
Brief History of Interactive
Music
The
history of interactive music has been surprisingly brief, and a relatively
small proportion of games has had it. It's a hard thing to do. (More
on why later) When video games started, they could only make a few basic
tones. The sound capabilities were not sufficient to justify hiring
an actual composer to compose actual music. Usually a programmer who
had a few spare hours would write something up in a text file as direct
commands to the tone generator. Someone in that position wouldn't be
thinking of interactive music, they'd be thinking of getting that chore
out of the way. Nonetheless, even in the primitive 70's the interactivity
was there. Space Invaders, and then Asteroids both employed a simple
device to aurally communicate urgency to the player: they speeded up
the sound. Space Invaders was 4 bass notes going down, while Asteroids
was merely two. (The world's shortest melody!) Yet those of us old enough
to have played these games in their heyday remember that the feeling
was more intense as the wave of enemies or asteroids reached its end
because that loud bass synth was going so fast...
By
the mid-80's advances in electronic musical instrument technology had
started to benefit the video game market, and producers wanted to have
"real" music. They wanted music that sounded like actual songs,
songs you'd hear on the radio, the farthest possible cry from simple
tone-generator patterns. Most music for games tended towards either
incorporating digital recordings (in arcade games, at least) or higher-quality
synthesis and digital soundfile instruments played by MIDI scores. Hiring
composers became justifiable. This trend towards music that sounded
more like what people thought of as music, music they'd hear on the
radio, was a big step away from interactivity. Music for listening is
inherently non-interactive. The compositional traditions of western
(and most) music require that the final state of the composition is
known and fixed. Music that works well with a game, especially music
that is ready and able to change itself unpredictably is likely to be
boring if listened to by itself. With producers looking for good songs
(as opposed to soundtracks) and composers coming from listening music
backgrounds, interactivity was almost impossible and nearly forgotten.
Often
this problem has been solved by recording several different songs as
digital soundfiles and simply playing them back, while switching between
them to respond to the gamestate. The obvious limitations of course
are that it's technically difficult to do, the gaps while switching
are jarring and unpleasant, and from a musical standpoint, it's coarse.
You don't have the option of subtlety, and you are limited to only a
few choices because of space considerations, both memory and storage.
Then there’s the question of whether the game you’re composing for can
handle constant disc reads to get new music!
Still,
this method is still used today with some success, depending on the
type of game. (Sports games are more amenable to this treatment, for
example)
A
concurrent issue changed the course of music styles in games. When I
first started, I was convinced that video game music should be subtle
background sounds that only vaguely suggest musical themes and melodies,
with occasional musical gestures asserting themselves at moments when
the events of the game justified it. I wanted to make scores that at
times would be only background sounds. Low tones, wind, occasional notes...
I came into the arena thinking this way because as I played video games
throughout the 80's, I came to be very annoyed with the fully crafted
songs that became prevalent in them. The problem was that the better
the song was, the worse it was for the game! Good songs that repeat
themselves over and over are a kind of psychological torture that only
our modern culture could produce. Almost every one of us has a bad memory
of a game we liked but whose music could not be turned off, or was an
arcade game. By the time my gaming switched primarily to PC games, I
took my place among the armies of gamers whose first act in a game was
to find the off switch for the music.
But
there was some resistance to "less musical music" within the
industry. Producers wanted to impress people with the high quality of
each component of the game. For the music, this meant "songs",
in the classic sense of the word. It took some time before it sunk in
that the game as a whole was suffering. In 1994, reviews were still
ready to criticize the soundtrack if it didn't sound like music in the
traditional sense. An innovative composer working on a game could get
the product a bad paragraph in its review by creating a score that was
less musical but fit better in the game.
And
finally, the tools weren't there. Music that changes in response to
gamestate has to be set up to respond to signals from the game program
itself, and this just isn't in the MIDI spec. Code for interactivity
has to be written, and programmer time is as precious as plutonium.
The
desire for interactive game scores was strong enough to overcome them,
however. The critics got used to it, the tools got written, and the
composers learned how to make it work on their end as well. This last
is what I want to talk about now.
________________________________________________________
Conceptual
Problems of Interactive Game Scores
|
|
|
|
Features
