In 1965, an Iowa physics student working on satellite signals built what one person has called called "the world's first digital synthesizer." This video unpacks how James Cessna’s “Arbitrary Waveform Generator” worked, and then recreates it as a Disting NT plugin, with a space ambient track to show it in action (patch notes included).

I'm curious to see how people here would classify this strange instrument. It was digital in the sense that it builds a waveform out of discrete amplitude steps and used a digital sequencer to scan them at high speed, like later wavetable or digital oscillator designs. But it still relied on analog amplitude controls. And if it was digital, was it really the "world's first digital synthesizer"?

All sequencers use digital electronics to move between each step of the sequence, does that make an analog synth with a sequencer "digital"? One of the most common modular synth tricks is putting audio-rate signals into inputs not "meant for" audio rate inputs, so does putting an audio-rate (or higher) square wave into the Clock In of a sequencer fundamentally change what the synth is?

Yes, you're right, sequencers use digital logic to advance steps, and have done so since the 1960s. But what's distinctive about Cessna's approach is that he discretized the waveform itself, encoding its shape as a table of amplitude values. That's clearly digital synthesis in the sense of constructing signals from numbers. Also, his implementation of pitch tracking was computational. But I think ultimately the distintinction between digital and analog is moot: even on analog systems, we are thinking digitally when we use on-off gates or when we use a sample and hold to transform continuous voltages into stepped ones.