Tutorials Audio

A History of Recorded Sound: From Wax Cylinders to Streaming

History · ~20 min

Overview

Every convention a modern audio creator takes for granted (the LUFS target you mix to, the reason a CD holds 74 minutes, the concept of a 'take') is a residue of a hundred-plus years of recording technology history, each era solving the previous one's constraint and creating a new one. Understanding that history isn't trivia. It explains why current formats and workflows look the way they do, and gives useful perspective on where they might go next.

What You Need

No equipment required. This is a reference and context guide, not a hands-on tutorial

Steps

Mechanical recording: phonograph and gramophone (1877-1920s)

Thomas Edison's 1877 phonograph recorded sound as physical grooves cut directly by sound-wave vibrations into a tinfoil, then wax, cylinder, with no electronics involved at all. A horn focused sound onto a diaphragm connected directly to a cutting stylus. Emile Berliner's flat disc gramophone (1887) proved more practical to mass-produce than Edison's cylinders and became the dominant format, establishing the flat disc lineage that runs all the way to the modern vinyl record. Recording during this era required performers to play directly and loudly into a horn: there was no amplification, so the physically loudest instruments and voices had a real, practical advantage.

Electrical recording and amplification (1920s-1940s)

The introduction of the microphone and electrical amplification in the mid-1920s replaced the acoustic horn with a genuinely sensitive transducer, dramatically expanding usable frequency range and dynamic range, and for the first time letting quiet instruments and subtle vocal performances record clearly. This era also introduced radio broadcasting as a parallel medium to recorded discs, establishing live broadcast conventions (level control, the studio engineer role) that would later transfer directly into music recording practice.

Magnetic tape and the birth of editing (1940s-1970s)

Magnetic tape, refined from wartime German technology and commercialized after World War II, was the first medium that could be physically cut, spliced, and re-recorded, meaning, for the first time, a recording could be edited after the performance rather than needing to be captured perfectly in one take. This single change created modern studio practice: multitrack recording (layering parts recorded at different times), the producer as a creative role distinct from the performer, and effects like tape delay and reverb created by physically routing tape through record and playback heads.

Digital audio and the compact disc (1980s-1990s)

Digital recording (sampling analog sound as discrete numeric values) eliminated tape hiss, generational loss from copying, and physical wear, at the cost of requiring entirely new theory (sample rate, bit depth, the Nyquist theorem) that analog engineers hadn't needed. The compact disc, introduced commercially in 1982, was deliberately specified to hold enough data for a single uninterrupted recording of Beethoven's Ninth Symphony (roughly 74 minutes). An anecdote that's more folklore than confirmed fact, but the format's actual capacity genuinely was set close to that length, and it shaped album-length conventions for over a decade.

The internet, MP3, and streaming (1990s-present)

Perceptual audio compression (the MP3 format, standardized in the early 1990s) made file sizes small enough for the era's internet connections by deliberately discarding audio information human hearing is least sensitive to. A lossy tradeoff the CD-era engineer would have considered unacceptable, made necessary by bandwidth rather than audio quality goals. Streaming, arriving roughly two decades later, changed the economic and technical model again: loudness normalization (the direct ancestor of the LUFS targets creators mix to today) exists specifically because streaming services needed a way to make user-uploaded, wildly inconsistent content play back at consistent perceived volume without engineers manually managing it.

Pro Tips

Next time you set a LUFS target, remember it exists because streaming platforms needed an automated answer to a problem tape-era engineers solved manually, one mix at a time.
The concept of a fixed album length (roughly 40-80 minutes) is a physical-media artifact, not an artistic law, streaming has already started dissolving it, and it’s worth questioning whether your own release format still needs to honor it.
Multitrack recording, which every modern DAW assumes as a given, didn’t exist as a workflow until magnetic tape made it physically possible. It's genuinely a 20th-century invention, not an eternal truth about how music is made.

Knowledge Base

What You'll Learn

Each era of recording technology solved a real constraint of the previous one, and the residue of that history is still embedded in the tools and conventions creators use today.

A Quick Timeline

1877: Edison's phonograph, first mechanical sound recording.
1887: Berliner's gramophone disc. The flat-disc lineage that leads to vinyl.
Mid-1920s: Electrical recording via microphone and amplifier.
Late 1940s: Commercial magnetic tape, recording becomes editable.
1982: Compact disc, digital audio reaches mass consumers.
Early 1990s: MP3, perceptual compression makes digital audio internet-portable.
2010s-present: Streaming, loudness normalization and on-demand access replace ownership of physical or downloaded copies.

Why This History Still Matters to a Modern Creator

Format conventions (album length, track structure), mixing conventions (loudness targets, the very concept of 'mastering' as a final loudness/tonal pass), and even vocabulary ('take', 'track', 'bounce', 'mixdown') all carry the fingerprints of specific historical technical constraints. Recognizing which conventions are genuinely still functional versus which are legacy habits from a format that no longer applies is a useful, practical lens, not just historical trivia.

What's Likely Next

Spatial and object-based audio (Dolby Atmos Music, similar formats) is repeating a pattern from every prior era: a new format arrives promising fidelity or immersion gains, and workflow conventions take years to catch up. See our companion piece on the future of media creation for where this and other current shifts may be headed.

Where This Fits

This guide covers one specific part of the history of media technology. The wider picture — how each link in the chain from capture through transmission to display was actually built, who built it, and why the credit so often landed somewhere else — is in A History of Broadcast Technology: The Chain From Capture to Screen, which frames the discipline as a whole and links out to the detailed guides underneath it, including this one. If you are starting from scratch rather than solving a specific problem, read that first and come back here.

FAQ

Q: Why do CDs hold about 74 minutes of audio?
A: The format's storage capacity was specified during its joint development by Sony and Philips in a way that comfortably fit a full recording of Beethoven's Ninth Symphony. A widely repeated origin story that's likely simplified but reflects a real engineering target close to that length, which then shaped album-length conventions for years afterward.

Q: Why do streaming platforms normalize loudness instead of just playing files at their original level?
A: Because user-uploaded and label-delivered content varies enormously in mastered loudness, and without normalization, listeners would need to manually adjust volume constantly between tracks. LUFS-based normalization automates the consistency that tape and vinyl-era listeners never had, and that CD-era listeners still dealt with manually.