I went to a lecture on the evolution of sound effects in animation that’s got me thinking about what games can learn from how sound was created and used at places like Disney.
In the early days, all the sound for a film had to be performed at the same time. Music, effects, dialog, everything. The whole team got together in a room and recorded it start to finish, like a stage show.
In modern films, sound effects are often cut together from other sounds. To make a monster’s howl you might use a mix of, say, pitch-shifted breathing and a lion’s roar played backwards. But back then if you wanted an unusual sound you had to build yourself an instrument.
For example, to get the sound of rain they’d stick hundreds of nails in a barrel and fill it with Mexican peas. By turning the barrel at different speeds they could match whatever the intensity was of the rain on screen. For Tinker Bell the sound effect artist used a set of bells attached to his fingers, which he could “play” to create whatever sounds fairies make.
Because they had to be made in real time, the effects became a performance. Working with tactile props it was possible to create some amazingly subtle and interesting sounds that added life to what was happening on screen.
Animation and games have the same fundamental problem: they’re not real. Even if you can get the audience to ignore that fairies and space marines don’t actually exist, you’ve still got to make them believe in your fake little world (of pixels, drawings, etc). And that’s where sound effects can be a huge help. They’re a way to inject an authentically human presence into an otherwise coldly technical universe.
With sound effects that feel like a performance we can sense the human being that made them. Tinker Bell’s movement doesn’t feel like a set of bells, it feels like the actions of a living creature. Which is a good thing, because Tinker Bell herself doesn’t have any dialog in Peter Pan. And she doesn’t need any. We get enough of her personality just from the sounds she makes.
I wish games had more of that.
Instead, our sounds are often bland and rigidly mechanical. When a monster growls at me they sound exactly the same every time. Or at best, one of three growls plays at random. Sometimes that’s OK. I don’t need Mario to make a different sound every time he jumps. There, the sound is just a part of the game’s interface, a way of confirming the player’s action. But in most games that’s the only thing sound is used for. I think we should be asking sound to do more.
Some areas that come to mind:
- More distinctive sounds and a wider emotional range for those sounds would help in creating memorable characters and environments. Wrex from Mass Effect is a good example — his absurdly deep voice and folksy tone create an unusual mix that’s threatening and comforting at the same time, like Wrex himself.
- Gameplay based around sound feels like it’s been pretty much ignored (except, you know, Guitar Hero etc). For example, a game that asked you to pick a Terminator out of a crowd by listening for his metallic heartbeat would be a nice change of pace. Not only does it give you an excuse to create interesting sounds, it also gives players a reason to notice and appreciate them.
If I hear someone spray painting a wall and then come around a corner and see the wall, that feels a lot more convincing than if I actually watch the character spraying it, no matter how beautifully modeled and animated they are.
Plus, it’s cheaper!