Using Game Music Strategically

Historically, music has been a somewhat neglected part of game design. Not in terms of raw music quality--there have been many, many excellent soundtracks over the years--but in design terms. Too many developers have made poor decisions with regard to their music. They misuse what could be a huge advantage to a game's quality. Some lessons:

Make sure players actually hear the music:

Any soundtrack, no matter how good, is going to be wasted effort if players don't actually listen to it, right? This seems like an obvious lesson, but it's one many developers have ignored. Look at Crackdown, for example. From its wikipedia page: "Crackdown features over 100 tracks of electronica and sample-based music by a number of independent and video game musicians, including Amon Tobin, Atlas Plug, Celldweller and Hybrid. Music supervisor Peter Davenport was in charge of selecting the music for the game, a task that took three years to complete." Pretty impressive, right?

Well, it would've been if I could have properly listened to the music. The problem is that this (by all accounts, wonderful) soundtrack only plays while you're in vehicles. But a good 90% of the game will be spent by most players on foot, since the game's heavily focused on jumping, running, shooting, and brawling. This means that for the vast majority of the time the players won't hear music. Crackdown's a fairly short game in any case, which means that it's very unlikely that anyone will even hear all the music once, let alone enough for it to make an impact.

F-Zero GX had a similar problem. Each of the game's thirty or so racers had their own individual, diverse, excellent theme songs. Instead of putting the music in the actual racing gameplay so that players could listen to it, the designers decided to squander it by tucking it away in a submenu where players would maybe spend 15 minutes (and that's being generous).

Both of these games had missed opportunities in terms of music. Crackdown didn't get the huge acclaim for its soundtrack that the GTA games did, and F-Zero GX was widely considered to have an inferior soundtrack to its predecessors. Both judgements were unfair, but completely understandable. The developers could have solved the problem with very little development time.

Make sure players don't hear the music too much:

Generally, most people get sick of most music after a while, no matter how good it is. Listening to a song two or three times in a row is fine. After ten times, it'll start to get quite repetitive. After twenty times, it'll probably drive you nuts.

Despite that, lots of games ignore that and bombard players with the same tracks over and over. Western RPGs like Oblivion can have 100+ hours of gameplay, but they're lucky to exceed 2 hours of actual music. That means a lot of repetition, and even the best tracks start to get irritating after a while.

Japanese RPGs usually mitigate the problem of too little music. They often have their composers make 3-4 hours of music or more, sometimes by hiring more than one composer for a game. Unfortunately, while they might have enough quantity for their music, they get the distribution wrong, so the player still ends up hearing the same songs again and again. For example, look at JRPG combat. A huge chunk of most JRPGs is generally spent in battles, and yet relatively few have the sense to mix up the music playing during these fights. Occassionally the music will change for fighting bosses, but overwhelmingly the player will be subjected to one or two tracks repeating ad nauseam. It's even worse when the game has short random encounters and players only hear the first minute of the track, rather than the whole thing, over and over.

Final Fantasy Tactics is a notable example of a game that uses music very strategically (the music is also of great quality, incidentally). It has lots of tracks (made by at least two composers), almost all the tracks get a chance to be heard a few times, and they are distributed well--there are many battle tracks, for example.

Make sure the music fits the game:

This can be done in a lot of ways. Basically, try to integrate the music well. That can mean using dynamic music (historically, the Lucasarts iMuse system in their classic adventure games was a good example) so that the music matches the action on the screen, or it can mean simply playing appropriate tracks rather than just randomly picking one from the whole pool of available music (for example, making sure character themes actually play when the respective character is on-screen).

One neat trick is to play around with the source of the music. You can just play it normally, or you can try and make the music diegetic--that is, part of the game world. For example, you could pipe the music through stereos or loudspeakers placed in the level.

Use selective silence: Having a more-or-less continuous soundtrack is probably a fine approach for the vast majority of games, but selective silence can be a viable tactic as well. Half-Life 2 doesn't have background music for most of its gameplay time. Instead, the songs kick in at key moments to heighten emotion. The benefit of this approach is that musical silence can be immersive (real life generally not having a soundtrack); it also makes the player more aware of sound effects and increases the impact of the music when it does finally play.

Ultimately, thinking strategically about music costs very little time and effort, and can potentially yield large dividends in terms of the game's immersiveness, emotional impact, and variation. Having lots of great music for a game is one thing; using it in such a way as to maximize its effectiveness is another.