Rise of the Synthesizer
This section on the making of Switched-On Bach paints a remarkable pitcture of the early days:
In a word, synthesizers in 1974 sucked. Sure, their vintage cred looks cool from 2015, but all synthesizers in 1974 were monophonic, which meant they could only produce one note at a time. That was a major headache if you were Wendy Carlos and you had made it your mission to include a composition such as Johann Sebastian Bach’s “Brandenburg Concerto No. 3” on “Switched-On Bach.” Because her Moog was monophonic, Carlos had to play the notes for each of the concerto’s nine stringed instruments—as well as the harpsichord part—one at a time. Worse, Carlos was forced to play each note in each of the chords any of those instruments might be required to produce one at a time, too.
As if that limitation were not hobbling enough, early synthesizers, including the Moog, were notoriously bad at staying in tune, which meant Carlos typically had to work in bursts—often lasting no more than 5 seconds at a time—before the tone she had found by twisting one knob this way and another that way had degraded. Once a clean burst was recorded, the tape would be rewound, cued up, and the next burst would be added in real time. It was a painstaking procedure, requiring endless takes. In retrospect, that a project like “Switched-On Bach” was completed at all is something of a miracle.
The article is a bit of a ‘long read’ but is totally worth it.