A film has a runtime. However complex the edit, however many versions get tested, there's eventually one sequence of frames that an audience watches in a fixed order, at a fixed pace. A composer writing for that film is writing toward a timeline that already exists.

A game doesn't have that. A player might stand in a room for four seconds or four minutes. They might trigger a fight, retreat, wander back later, or never find the room at all. There is no fixed timeline to score — there's a space of possible timelines, and the music has to hold up across all of them.

This is the first thing that breaks when a composer moves from linear to interactive work: the assumption that there's a "right" length, a "right" arrival point, a "right" ending. There isn't one. There are many, decided in real time by someone else.

Two Ways Music Adapts

Interactive scores solve this problem with two related but distinct techniques.

Horizontal re-sequencing treats a score as a set of pre-composed segments — an exploration theme, a low-tension variant, a high-tension variant, a resolution — that crossfade into one another as the game state changes. The composer writes each segment to end and begin in ways that tolerate a cut at any transition point, similar in spirit to how a film cue has to tolerate an editor's trim, but happening live, driven by gameplay instead of an edit decision.

Vertical layering takes a different approach: instead of switching between separate segments, the score is built as a stack of synchronized stems — a bed layer, a percussion layer, a melodic layer, a tension layer — all locked to the same tempo and key, that get muted and unmuted based on intensity. Combat intensifies, and a layer fades in. The player retreats, and it fades back out. The underlying material never actually changes; what changes is which parts of it are audible.

Most modern interactive scores combine both: horizontal segments for major state changes (exploration to combat), vertical layers for continuous intensity within a state (calm combat to desperate combat).

What This Demands From the Composer

The practical consequence is that an interactive composer is writing modular material from the outset, not composing a piece and figuring out later how to chop it up. Stems have to share tempo and key by design. Segments have to have multiple valid entry and exit points, not just one. A melodic idea that works has to work whether it's heard for six seconds or six minutes, because the composer doesn't get to decide which.

This is a stricter version of a discipline that shows up anywhere music serves something else instead of standing alone: write less certainty into the material, and build in the seams where change can happen cleanly. Interactive scoring just makes the requirement explicit and structural, because the "editor" deciding when to cut is a real-time system with no judgment at all — it just does what it's told, exactly when it's told to do it.

The upside is that this constraint produces some of the most texturally rich material in modern composition — material designed from the ground up to recombine, layer, and evolve rather than simply play back. That's exactly the kind of raw material a stem-based texture library is built to provide.

Cinematic Resonance is a curated sample pack of tension-forward drones and textures, delivered as stems built for layering rather than straight playback. Explore it at sonalsystem.com

More on composition, sound design, and the business of music in the Frequencies archive.