Case Study 1 — Fourteen Gain Stages: One Vocal's Journey from Larynx to Earbud
A teaching anatomy. The characters are our fictional duo, Wren & Hollow; the chain is absolutely real — it's the path every recorded voice you've ever loved actually traveled. We follow one sung phrase from the moment it's muscle and breath to the moment it's pressure in a stranger's ear canal, and we stop at every place its level gets decided. Some of these stages belong to chapters you haven't read yet; treat those stops as previews — the point today is the count, and who controls each number.
It's a Thursday night in Asheville. Theo Park has the spare-bedroom rig running — one condenser on a stand, the 2-input interface, Reaper open — and Demi Wren is cutting a scratch take of a new song for their EP, a phrase she's been humming all week:
"Count the rooms in your father's house."
Seven words, about four seconds of audio. Months from now — after chapters of work this book hasn't taught you yet — a college student in Tucson will hear that phrase through one earbud while walking to class. Between Demi's larynx and that earbud, the phrase's loudness will be set, re-set, nudged, lifted, trimmed, limited, normalized, and adjusted fourteen separate times.
Here's every one of those times, what it's for, who controls it, and what goes wrong there.
The analog opening: stages 1–3
Stage 1 — The performance itself. Demi's breath support, how hard she leans into "count," how she lets "house" decay. The first gain stage in any vocal chain is the vocalist: a performer's dynamic range — whisper to belt — is wider than any knob downstream, and no stage after this one can add intention that isn't here. Who controls it: Demi. What goes wrong: everything, if the engineer treats a level problem as a knob problem when it's a performance problem. A timid take recorded "hotter" is still timid. (Chapter 11 is about producing the performance; it outranks every other stage in this list.)
Stage 2 — Distance to the microphone. Demi works about eight inches off the condenser. Level falls fast as distance grows — step back and the mic hears less of her and more of the room around her. Distance is a gain knob made of air, and singers ride it instinctively, leaning back on big notes. Who controls it: Demi's body. What goes wrong: a singer who wanders changes level and the direct-sound-to-room-sound blend from phrase to phrase — the kind of inconsistency no fader fixes later, because it's two things changing at once. Theo's fix is masking-tape feet marks on the floor. Engineers have used that fix for seventy years because it works.
Stage 3 — The capsule and its pad. The condenser converts pressure into voltage — the chain's first transduction — and its sensitivity decides how much voltage per unit of pressure. There's a -10 dB pad switch on the body for sources hot enough to overload the mic's own internal electronics. Demi is not one of those sources; the pad stays off. Who controls it: the mic's designer, plus one switch. What goes wrong: a forgotten engaged pad steals 10 dB and quietly forces extra noise-adding gain downstream — check the switches before blaming the preamp. (The reverse error — no pad on a screaming amp — distorts at the capsule, before the cable, where no one thinks to look.)
The decisive stage: 4
Stage 4 — The preamp. Theo's interface lifts the mic's whisper of voltage to line level: tens of decibels of amplification, the single largest level change in the entire fourteen-stage journey. He sets it the ritual way — Demi sings full phrases at performance level while he watches Reaper's meter, landing peaks around -15 dBFS. Who controls it: the engineer, once, on purpose. What goes wrong: the chapter you just read. Too low, and tonight's take hisses forever when it's boosted later; too high, and one passionate consonant clips at the next stage's cliff. This knob's setting is married to this singer, this song, this night — which is why Theo re-checks it when Demi says she wants to try the bridge "bigger."
Crossing the border: stage 5
Stage 5 — The A/D converter. The lifted voltage gets measured into numbers — Chapter 2's whole story, happening in a chip the size of a fingernail. The converter's calibration is fixed; it isn't a knob. But it's a gain stage in the sense that matters most: it's the stage with the cliff. 0 dBFS lives here, and anything the preamp sends above it is discarded before the file exists. Who controls it: physics and the interface designer. What goes wrong: nothing — if stage 4 did its job. Every clipped capture in history is a stage-4 mistake that stage 5 merely notarized.
The forgiving middle: stages 6–10
Now the phrase is a file: vox_scratch_take3.wav, safe in Reaper, in the roomy digital middle where level changes are mathematics — undoable, lossless, free. The next five stages will mostly happen months later, at mix time, but they're set on this same audio, so we count them now.
Stage 6 — Clip gain. Level adjustment painted directly onto the recorded region, before any plugin sees it. When the mix happens (Part V of this book), the first vocal move will be clip-gaining phrases toward consistency — the quiet "in your" lifted a couple of dB so that every later processor receives a steadier signal. Who controls it: the mixer, phrase by phrase. What goes wrong: skipping it. Feed a wildly uneven vocal to the processors downstream and each one misbehaves in its own way, and you end up turning knobs to fight a problem that one quiet, early gain move would have dissolved. Earliest stage, cheapest fix — the doctrine again, this time inside the DAW.
Stage 7 — The insert chain's gains. Every plugin on the vocal channel has input and output levels, and several exist to change level dynamically — Chapter 23's compressors come with "makeup gain" built into the name. Each insert is a gain stage; a vocal chain of four processors is four chances to creep louder or quieter without anyone deciding. Who controls it: the mixer — ideally with the matched-loudness discipline Chapter 22 will drill. What goes wrong: the oldest self-deception in mixing: every plugin nudges the signal a little louder, louder sounds better, and the mixer congratulates the plugin for what the gain did. (Chapter 21 declares war on this; consider this stage a preview of the battlefield.)
Stage 8 — The channel fader. The vocal's level relative to everything else in the song — the balance decision Chapter 20 calls the most important 90 minutes of the mix. Who controls it: the mixer, and later, automation (Chapter 27) will move it phrase by phrase like a tiny invisible engineer riding the take. What goes wrong: set-and-forget. A static vocal fader is wrong somewhere in every song, because the song changes around it.
Stage 9 — The VOX bus. In the full production, Demi's lead will share a group channel with her harmony stacks — one fader governing the whole vocal section's level against the band (the 6-bus architecture "Static Bloom" uses, mapped in Chapter 20). Who controls it: the mixer. What goes wrong: redundant processing stacked at both channel and bus level by someone who forgot which gain lives where — the audit habit from this chapter, applied inside the session.
Stage 10 — The 2-bus and the mastering chain. The whole mix's level, printed with headroom (peaks around -6 dB, no limiter — Chapter 31's print spec), then handed to mastering, where the last deliberate gain decisions are made: a limiter raising density, a ceiling set at -1 dBTP (Chapters 32–33 unpack every word of that). Who controls it: the mastering engineer — Theo and Demi will hire one for the EP; Jaylen will learn to do it himself in Chapter 32. What goes wrong: chasing loudness for its own sake. The next stage explains why that war ended.
The stages nobody tells beginners about: 11–14
The master is finished. Demi and Theo's control ends here. Four gain stages remain, and none of them belong to the artist.
Stage 11 — The platform's loudness normalization. The streaming service measures the track's overall loudness and turns it down (or occasionally up) to match its reference playback level, so listeners don't ride their volume between songs. As of this writing, the common default sits around -14 LUFS — and the details, the units, and the surprisingly happy consequences for dynamic music are Chapter 33's whole story. Who controls it: an algorithm in a server farm. What goes wrong: mastering decisions made in ignorance of it — crushing a master to be "the loudest," only to have the platform politely turn it down to match everyone else, keeping the crushed dynamics and discarding the loudness they were traded for. Theme three, at industrial scale: that tradeoff bought nothing.
Stage 12 — The listener's device processing. The student in Tucson has a phone with a volume limiter, possibly an "EQ enhancement" mode she doesn't remember enabling (Jaylen's audit found his; hers is still on), and a Bluetooth hop to her earbuds that adds its own conversion and lossy compression. Who controls it: the listener, mostly unknowingly. What goes wrong: nothing you can fix — only anticipate. This is why Chapter 30 makes you test mixes on real-world systems, phone speakers and all. The chain doesn't end at your studio; it ends out there.
Stage 13 — The volume rocker. The most-used gain stage on planet Earth. She's walking past a bus stop; she clicks it up two notches. Who controls it: her thumb. What goes wrong: from the artist's chair — nothing. You don't get a vote. Make the record survive being played quietly; Chapter 4 explains why quiet playback is the honest test of a balance.
Stage 14 — The earbud and the ear. The final transduction: voltage back to pressure, three centimeters from an eardrum. Even here, level is being decided — an earbud that's lost its seal leaks bass dramatically, ask anyone who's worn one half-out; the fit is a final, accidental EQ-and-gain stage. Then the eardrum moves, the cochlea translates, and four seconds of Demi's Thursday night becomes the thing it was always headed toward: a feeling in someone else's chest. Who controls it: anatomy. What goes wrong: philosophically, nothing is left to go wrong — every stage that could be controlled was, by someone, on purpose.
The ledger
stage 1 performance Demi intention set here, nowhere else
stage 2 mic distance Demi level + room blend in one motion
stage 3 capsule & pad designer check the switch before the blame
stage 4 preamp Theo THE knob — noise floor decided
stage 5 A/D physics the cliff; notarizes stage 4
stage 6 clip gain mixer evenness before processing
stage 7 insert gains mixer louder ≠ better, match loudness
stage 8 channel fader mixer the balance decision
stage 9 VOX bus mixer the section against the band
stage 10 2-bus + master mastering headroom spent on purpose
stage 11 normalization platform the war that ended (Ch. 33)
stage 12 device listener anticipate, don't control
stage 13 volume rocker listener survive quiet playback
stage 14 earbud & ear anatomy the destination
Fourteen decisions about one phrase's loudness. The artist's team makes ten of them; the last four belong to algorithms, thumbs, and anatomy.
Three takeaways worth carrying out of this anatomy:
First — the level story is mostly written early. Stages 1–4 happen in one room in one evening, and they fix the recording's noise floor, its headroom, its consistency, and its intent permanently. Stages 6–10 can only arrange what stages 1–4 captured. The mix is a re-teller, not an author.
Second — every stage answers to a different boss. The singer, the engineer, the mixer, the mastering engineer, a server, a stranger's thumb. Professional audio is the discipline of making your stages so right that the stages you don't control can't break the music. You can't reach the student's earbud fit. You can absolutely make a vocal that survives it.
Third — count your own version of this chain. Your song will travel one just like it. The Project Checkpoint's map covers your stages 2 through 10 or so; this case study is what the rest of the road looks like. Fourteen stops. Now you know every one by name — which means, for the first time, none of them is mysterious.