When a game studio requests gameplay underscore, it’s often best to provide several versions of the same loop so that they have the option have mixing in real time even if they don’t have the budget for a full adaptive/interactive score, so the game ends up with more music than the developer has actually commissioned. Here’s an example of how I did this on a recent project.

This was a quick project that required just one additional 30-second loop for a mobile game. Although the client could have picked this up from a stock music source, it’s often just as inexpensive (or even cheaper) to request custom music, which also allows the client some creative input into how the piece should sound, and can ensure that no other game or video has the same exact loop (for at least a reasonable period of time, if not a buyout).

It gets incredibly boring to hear the exact same 30-second loop endlessly, and funny things happen when players experience this: sometimes they don’t even realize that the looping bothers them, and instead blame the art or story for being dull; other players know it’s the music, hate it, and the poor composer gets blamed for being uninventive. Regardless, the entire game suffers.

Simply bouncing a loop into multiple mixes is one way to avoid this. The game engine or script can then randomize between this set of loops somewhat, which works because the loops all fit together (the tail of one matches the head of the other) because it’s really the same piece of music, although it should go without stating that each variation has to make good musical sense on its own and not just be a section stem.

It’s also affordable for developers, since these additional loops are not new pieces of music and don’t usually count toward the price per minute fee (I personally don’t charge for them at all unless there’s some significant new compositional — not production — content added to one of the versions, and even when there is new composition it usually comes at a reduced rate compared to new music).

Here’s a full 30-second loop which employs orchestra, electronic sounds, choir, and a tiny bit of foley-based sound design:

Here’s the same loop mixed with only the string section and some light percussion:

By mixing the first loop and the second loop in a sequence, the piece sounds like a repeating phrase with variation rather than the same monotonous loop, especially since the strings version has less of a crescendo and provides a few moments of creepy breathing space with the violin harmonics.

At the extreme is a somewhat ambient version of the loop with most melodic content removed:

I provided another ambient version which is more subtle (essentially just a creepy pad) mixed from the same piece, but you get the idea.

The middle ground between the extremes means making very slight changes, and this works really well when those kinds of variations are mixed into the array of available loops; for example, here’s a version of this same loop with most of the brass parts removed:

I was sure to provide versions without the choir as well (as an aside: some people really hate the sound of choirs in anything, similarly to the way plenty of people either despise or love the sound of distorted guitar power chords in anything).

The game engine or script can take all of these loops and either cycle them randomly or play them based on some game events (perhaps a certain area of a level gets the ambient creepy stuff, while the brass section crescendo version is played when a boss is revealed, or somesuch).

You end up with something that is slightly different every time the game is played, in as much as the sequence of selected loops is slightly random and/or event-influenced, even though it’s really just one tiny 30-second piece of underscore, and even though you haven’t employed any fancy interactive music tools.

The changes are subtle, but they are effective when they’re heard in a sequence, especially when the loop is as short as this one is.

Summing up: The intent is that this underscore becomes a lot more interesting to the game developer and to the game player because it’s mixed into multiple versions and ends up spawning several minutes of slightly randomized or event-influenced music instead of sounding like the short single loop that it actually is. As a composer, you enable the creation of a musical sequence at runtime, while the game is played, that is non-linear and can’t be captured on a soundtrack or duplicated in something like a film (since it changes slightly each time, even for a simple loop such as this one). And this usually comes at no additional financial cost to the client, thanks to the ease with which these versions can be created digitally, and at no burden to the programmer, since no DSP or other code-level audio changes are required.

Obviously, though, there may be footprint issues to contend with: If the game is destined for iPhone deployment, for example, you would seldom want multiple loop files to eat up footprint and take your client’s game over the 10 MB over-air deployment limit when there are so many other art and code assets to bundle in that same space. In this case, the benefit of delivering multiple versions is that the client at least has several loops to choose from when deciding which one(s) to package.

There are many more interesting things that can be done with more complex projects, but even low-tech, low-cost games can benefit from this bit of loop variation and sidestep an obvious monotony pitfall. And since it means acquiring several pieces of finished custom music and commissioning only one, it’s often a lot more game-effective and cost-effective for game developers who otherwise would look to stock music sources for these types of low-budget projects.