In the 1990s, creating dance music was a vastly different process than it is today. For artists working in home studios like mine, the setup typically revolved around an Atari 1040ST computer, a collection of vintage analog synthesizers – often partially broken – Akai and Emu samplers, Roland drum machines, and an Alesis Quadraverb for effects. Recording was a multi-stage affair. We’d start by capturing ideas on a DAT machine, then periodically visit a demo studio to transfer tracks to ADATs for better quality. Finally, every year or so, we’d get the chance to work in a professional studio, laying tracks down on an SSL console to real tape.
Looking back at those old demo tapes from the 90s, it’s fascinating to consider what defined their sound and how the production process itself contributed to it. Of course, modern digital tools offer immense advantages in dynamic processing and noise control, which undeniably shape the sonic landscape. However, beyond these technical advancements, the very nature of the 90s writing process, at least for me, was fundamentally different.
Consider a simple bassline created on a synthesizer like the SH101 back then. The notes themselves might be sequenced by the SH101’s internal sequencer or the Atari, but that was largely where the programmed control ended. There were no detailed parameter articulations, no automated patch changes or mutes, no programmed send effects or volume rides, and certainly no automated routing. Instead, we had to manually manipulate parameters in real-time during each recording take. If a take wasn’t quite right, we’d simply record another pass. Even with more advanced MIDI synthesizers like the Roland JX3P, our automation capabilities were limited, often to just basic patch changes. Sending complex Sysex automation envelopes to modulate parameters was simply not feasible within our workflow.
This hands-on approach resulted in a distinctly human quality within the music. When a filter opened, the contour was shaped by the subtle nuances of a human hand turning a knob. Any variations or changes were imbued with the inherent imperfections of human interaction.
These limitations profoundly influenced how we wrote, arranged, and mixed tracks. Today, we have the luxury of meticulously crafting a track, compiling numerous takes to achieve a “perfect” performance – stacked, wide, nuanced, and intricately micro-edited. In contrast, back in the 90s, we typically recorded the entire song to tape in a few takes, focusing on capturing the best overall performance. We’d subtly tweak filters and envelopes, manually adjust levels by ear and eye, and then compare a few takes to decide which version felt strongest. “Version 3 feels like the one,” we might say.
Working in a professional studio with an SSL console offered some automation, but it was still rudimentary compared to modern digital automation. Micro-editing and tape splicing to achieve hyper-detailed arrangements were simply not part of the workflow. Think about the sheer amount of time spent today splicing takes, crossfading sections, layering sounds, and meticulously automating parameters within a modern DAW. That intensive digital manipulation undeniably contributes to the sonic character of contemporary music. The 90’s dance music sound, born from these technological constraints, carries a different kind of energy – one that’s undeniably human and inherently imperfect, yet full of its own unique charm.
