How did early game composers make music with so few channels?

Early composers worked within chips that offered roughly three to five voices, and they stretched those voices in several ways. They assigned each channel a role, made one channel switch between roles, used timbre and volume controls to vary the sound, and reused material to save memory. Much of the work was closer to programming than to writing on a staff. [1] [2]

The simplest approach was to treat each channel as one instrument. A common NES texture put a triangle wave bassline at the bottom, an arpeggiated square wave for harmony, and the melody on top, a style seen in Nobuo Uematsu's Final Fantasy score. Another had a driving bass under two square waves playing the melody in thirds or fifths, as in the Castlevania main theme. In the Super Mario Bros. theme, all three pitched channels first move together as three-part harmony, and the triangle later splits off to play a simple arpeggiated bass. [3] [1]

Composer Hiroto Saitoh describes a three-note machine where the first channel carried the melody, the second carried counter melodies, and the third alternated between bass and rhythm. [4]

One technique was channel sharing, where several parts take turns on a single voice. Changes in pitch, volume envelopes and pulse duty cycle helped listeners tell the parts apart. SNK's Yasumasa Yamada described placing drum sounds on another track when the bass was silent, and copying a melody to a second track, shifting it in time and lowering its volume to imitate a delay. Kondo said he used quick bursts of notes or very low pitches so a square wave sounded like something else. [1] [5] [2]

Chips generally expected instructions in a channel, pitch and duration format, which made it easy to turn a score into code, either by the composer or by a programmer. Yamada recalled typing hexadecimal values by hand for early arcade and Famicom games. Kondo said composers also programmed sound effects and spent more time adjusting sounds than writing notes. Memory was a further limit, and Kazumi Totaka described reusing phrases and removing notes one by one so that music and effects would fit. [1] [5] [2]