EQ: Sculpting Frequency — Cutting the Mud, Boosting the Air, and Finding Every Instrument's Home

The Committee Decision

Jaylen Cole has been hunting a criminal for three days.

The "Static Bloom" session is in the best shape of its life. The static balance from Chapter 20 took ninety minutes and changed everything. The gain-staging pass from Chapter 21 killed every red light, parked the track averages around -18 dBFS, and left six clean decibels of headroom on the 2-bus. On his treated room's budget nearfields — the ones the Chapter 10 insulation panels finally let him trust — the verses sound close to right.

Then the chorus arrives, and a blanket falls over the song.

Not distortion. Not clipping — he checked; Chapter 21 made him the kind of person who checks. The blanket is subtler than that: the low half of the mix congeals into one warm, shapeless mass, and Demi's vocal, which sat perfectly in the verse, suddenly sounds like it's singing from inside a couch.

So he does what everyone does. He hunts for the muddy track.

He solos the bloom pad. Gorgeous — those detuned saws blooming through their filter exactly the way he designed them in Chapter 14. Not guilty. He solos Theo's acoustic guitar layer. Warm, woody, expensive-sounding — the 12th-fret placement from Chapter 12 doing its quiet work. Not guilty. He solos Demi's harmony stack. Smooth, rich, choir-soft. Not guilty. He solos the 808, the keys, the kick. Innocent, innocent, innocent.

Three days of this. Every suspect, interrogated alone, walks free. And the moment he un-solos and plays the chorus, the blanket comes back. At 1:50 a.m. on the third night he types the question into his phone, because that's where his real thinking happens:

how is every track fine and the mix still mud

Then he remembers something he wrote a long time ago, the night the listening switch flipped back in Chapter 4: my 808 and my sample are sitting in the same spot. can't tell where one stops. Same spot. He pulls a spectrum analyzer onto the 2-bus — eyes as a hypothesis machine, the way Chapter 4 taught — loops the chorus, and watches a smooth, complacent bulge sit there between 200 and 400 Hz like a sleeping animal.

So instead of looking for the one guilty track, he asks a different question: who's contributing?

The pad, blooming open right at the chorus, pouring detuned-saw warmth across the low-mids. The acoustic guitar, strumming full chords, its body resonance stacked in the same zone. The harmony stack — three of Demi, every one of her chest tones landing between 250 and 350 Hz. Each one depositing a perfectly reasonable amount of energy in a perfectly reasonable place.

He opens an EQ on the pad. One wide bell, 3 dB down at 300 Hz. The same move on the guitar. The same move on the harmony bus. Nine decibels of cuts, spread across three tracks, none of them dramatic enough to hurt anything soloed.

He plays the chorus.

The mix opens like a window. The blanket is gone. Demi's lead vocal steps two feet forward — and he never touched her fader. The 808 has edges again. The pad still blooms, the guitar is still warm, the harmonies still glow, and somehow there's air between them where there used to be felt. Total time for the actual fix, once he knew where to look: about sixty seconds.

The phone comes out one more time:

the mud was a committee decision. no one track did it. they voted.

If you've been with this book since the beginning, you've met this sentence before — on the very first pages, as a promise: your mix sounds muddy because three instruments are fighting for 200–400 Hz — here's the 60-second fix. Chapter 1 showed you the crime scene in the Kroger parking lot, where Jaylen's vocal sample dissolved into sludge on his cousin's car stereo. Chapter 4 gave the zone a name. This chapter keeps the promise. By the end of it you'll know exactly why the mud forms, why soloing can't find it, how to fix it in a minute, and — more important than any single fix — how to run the tool that decides, frequency by frequency, where every sound in your mix gets to live.

That tool is the equalizer. It's the most-used processor in mixing, the first one most people reach for, and the one most people use backwards. Let's fix that.

🏃 Fast Track: Short on time? Read "Frequency Real Estate," the filter vocabulary, "Cut First," "The Mud Map," and — non-negotiable — "The Matched-Loudness A/B Law." Do the Productive Struggle block when you hit it, then exercise C4 (the mud-committee fix) and the Project Checkpoint. That's the working spine: the model, the shapes, the doctrine, the founding fix, and the discipline that keeps you honest. Come back for the genre accents and the sidebar when your mixes start asking those questions.

🔬 Deep Dive: Read straight through, including the minimum-phase vs linear-phase sidebar. Then work both case studies with your DAW open — case study 1 is the kick/808 carve Chapter 13 promised you, logged move by move, and case study 2 is a complete spoken-word EQ chain you can steal. Do the full DAW Lab, especially C1 (the filter-shape tour) and C6 (the matched-loudness ritual). The companion volume, The Physics of Music, covers the harmonic-series science under all of this in its Chapter 7, and the masking psychoacoustics in its Chapter 5, if you want the physics at full depth.

The EQ You've Been Using All Your Life

Here's the good news: you are not learning a new tool today. You're being promoted to the precision version of one you've used since childhood.

Every bass and treble knob you've ever turned was an EQ — two fixed shelving filters, one aimed at each end of the spectrum. The "EQ" presets on your phone's music app: an EQ, obviously, usually a graphic one — a row of fixed bells pretending to be a smile. The tone knob on an electric guitar is a one-knob low-pass filter. And beyond the gear, the world EQs audio constantly, and your ear already knows every curve by heart. Music through an apartment wall: a low-pass filter — the wall absorbs highs and passes bass, which is why your neighbor's party is all thump and no melody, all night long. A voice on an old telephone line: a band-pass — the classic phone channel carried roughly 300 Hz to 3.4 kHz and threw the rest away, which is why telephone voices are all presence and no body, and why "telephone EQ" is now a production effect you'll meet in Chapter 17's bag of tricks. A pillow over a speaker. A hand cupped over your mouth. A car trunk between you and a subwoofer — Chapter 1's founding wound was, among other things, an EQ lesson administered by a Honda.

In Chapter 14 you met filters as sound-design tools: the low-pass filter was the workhorse of subtractive synthesis, and you spent a whole project stage sweeping cutoff and resonance by ear to design the "static bloom" pad. The filters in a mixing EQ are the same animals — same math, same shapes — wearing lab coats instead of leather jackets. Synthesis uses them to create a timbre. Mixing uses them to negotiate between timbres.

And Chapter 4 already handed you the map this chapter is drawn on. Six bands: sub below 60 Hz, lows 60–200 Hz, low-mids 200–500 Hz — the mud zone — mids 500 Hz to 2 kHz, high-mids 2–6 kHz, air above that. Back then, the bands were listening vocabulary: a way to name what you heard. Today they become a control surface. Every band you learned to hear is now a place you can reach into and change. That's the whole pitch of this chapter, and it's why Chapter 4 came first: an EQ in untrained hands is a steering wheel bolted to a car you can't see out of. You can see now. Time to drive.

One definition before we start driving, because the name is genuinely confusing: EQ stands for equalizer, a term inherited from early telephone engineering, where filters were used to "equalize" — flatten back out — the frequency losses of long cable runs. Nobody uses it that way anymore. In production, an EQ is a tool that turns a chosen range of frequencies up or down, and "to EQ" something means to reshape its tonal balance on purpose. Forget the flattening; remember the choosing.

Deciding Who Lives Where

Most beginners think EQ is a tone improver — a way to make each track sound better, one solo'd track at a time. That mental model produces exactly the mix Jaylen had: thirty-four individually gorgeous tracks and one blanket. The professional model is different, and it's the foundation everything else in this chapter stands on.

Frequency Real Estate: The Only Mental Model You Need

Your mix is a building with six floors — the six bands from Chapter 4 — and a fixed amount of floor space. Every sound you add moves in and occupies space on one or more floors, whether you planned it or not. A kick drum takes a big apartment in the sub and lows and keeps a small storage unit up around 3–4 kHz where its beater click lives. A lead vocal's chest tones rent in the low-mids, its core in the mids, its consonants and presence in the high-mids, its breath in the air band. A piano sprawls across almost everything. Nobody asks permission. They move in where their harmonics tell them to.

The building does not get bigger when you add tenants.

That's the entire problem of mixing stated in one sentence, and you've already met its enforcement mechanism: masking, from Chapter 4 — when two sounds occupy the same frequency neighborhood at the same moment, the louder one doesn't politely overlap the quieter one. It perceptually deletes it. Your ear's hardware runs a winner-take-most policy per band, with the distortion that masking spreads upward in frequency, which makes the crowded lower floors the standing aggressors against everything above them. In Chapter 4, masking was a listening phenomenon — the reason Jaylen's hats were audible at lower level in the Metro Boomin track than his own louder ones. Here in the mix, masking becomes the thing you manage daily, and EQ is your zoning law: the tool that decides which tenant gets which floor, who has to downsize, and who gets evicted from a floor they never needed.

So here is the reframe, and it's theme T2 wearing a tool belt: EQ is not about making tracks sound better. It's about deciding who lives where. A track EQ'd correctly for the mix often sounds worse on its own — thinner, duller, weirdly lopsided — because you've stripped it down to only the frequencies the song needs from it. The acoustic guitar in a dense pop chorus doesn't need its low body; the vocal and pad have that floor covered. Solo'd, the carved guitar sounds like a demo. In the mix, it sounds like clarity. The amateur EQs thirty-four tracks to sound individually impressive and gets a turf war. The professional EQs thirty-four tracks to fit, and gets a record.

Keep the real-estate model loaded for the rest of the chapter. Every technique that follows — the cutting doctrine, the high-pass discipline, the mud map, the pockets — is this one idea applied at a different scale.

🔄 Check Your Understanding

1. Why can a track that sounds worse solo'd be EQ'd correctly for the mix?
2. Masking spreads upward in frequency. Which bands does that make the standing aggressors, and against whom?
3. In the real-estate model, what does "adding a track" cost, even at a reasonable fader level?

Verify

1. Because the mix only needs part of each track's spectrum — the part no one else covers. Carving away the rest sounds thin in solo but creates space in context; the judge is the mix, never the solo.
2. The sub, lows, and low-mids — the crowded bottom floors — against everything above them: a bloated low end eats midrange clarity it never even occupies.
3. Floor space. Every new sound deposits energy across bands that are already occupied, raising the masking pressure on every existing tenant. The building doesn't grow.

The Filter Vocabulary: Four Shapes and a Width Knob

Open the stock parametric EQ in any DAW — and use the stock one; it's a precision tool, and the boutique ones mostly add color and price — and everything you see reduces to four filter shapes and three numbers. (Appendix E has each DAW's name for the stock EQ; Jaylen's lives in FL Studio as Parametric EQ 2, and yours is one menu away.)

HIGH-PASS (HP, "low-cut")             LOW-PASS (LP, "high-cut")
 0 dB ┄┄┄┄┄┄╭────────────────          0 dB ────────────────╮┄┄┄┄┄┄
           ╱                                                 ╲
          ╱  ← slope: 12 dB/oct                               ╲
         ╱     (6/18/24 exist)                                 ╲
 ──────┴┴────────────────→ freq       ────────────────────┴┴──────→ freq
       cutoff                                              cutoff
   passes highs, removes lows           passes lows, removes highs

LOW SHELF                              HIGH SHELF
 ╭────────╮                                          ╭────────╮
 │ +3 dB  ╰──╮                                    ╭──╯  +3 dB │
 ┄┄┄┄┄┄┄┄┄┄┄┄╰────────── 0 dB         0 dB ──────╯┄┄┄┄┄┄┄┄┄┄┄┄┄
   lifts (or drops) everything           lifts (or drops) everything
   below its corner, by a fixed amount   above its corner, by a fixed amount

BELL (boost)        BELL (cut)         SAME BELL, TWO WIDTHS (Q)
       ╭─╮                                 wide, Q ≈ 0.7    narrow, Q ≈ 8
      ╱   ╲         ─╮     ╭─            ╭───────╮              ╭╮
 ────╯     ╰────      ╲   ╱          ───╱         ╲───      ────││────
  +4 dB at center      ╲ ╱               gentle hill          ││
                        V  −4 dB         "tonal"              ╰╯ "surgical"

Caption: the four filter shapes of every parametric EQ — high-pass, low-pass, shelf, and bell — plus the Q knob that sets a bell's width.

The high-pass filter (HP) removes everything below its cutoff frequency and passes everything above — which is why some DAWs label it "low-cut," the same filter named for what it kills instead of what it keeps. Both names, one shape. Its slope sets how aggressively the lows fall away below the cutoff: 6 dB per octave is a gentle tilt, 12 is the everyday workhorse, 24 and steeper are guillotines. The low-pass (LP / "high-cut") is its mirror image, and you already own it — it's the synth filter from Chapter 14 and the apartment wall from ten minutes ago.

Shelves lift or drop everything beyond a corner frequency by a fixed amount, like the last step of a staircase that then stays flat — a plateau, not a cliff. The bass and treble knobs on every stereo you've ever touched were a low shelf and a high shelf. Shelves are tonal tools: broad, gentle, character-scale moves. "A touch more warmth" is a low shelf. "A little more sheen" is a high shelf.

Bells (your EQ may say "peak" or "parametric") are the scalpel and the paintbrush in one: a hill or a valley centered on a frequency you choose, as wide or as narrow as you choose. Three numbers define every bell: the center frequency (where), the gain (how much, up or down), and the Q (how wide).

Q deserves a paragraph, because it's the knob beginners ignore and professionals lean on. Q is the ratio of the bell's center frequency to its bandwidth — which means higher Q = narrower bell, the direction that trips everyone at first. You don't need the formula; you need the intuition: a low Q around 0.5–1.4 is a wide, musical hill that reshapes a track's character, touching an octave or more. A high Q from 4 up past 10 is a needle — it reaches in and addresses one ringing frequency while leaving the neighbors alone. A useful rule of thumb the rest of this chapter will keep earning: wide moves change tone; narrow moves fix problems. And a warning attached: narrow boosts with high Q don't only get loud at one frequency — they ring, adding a resonant, whistly quality you'll hear the moment you try it (exercise C1 makes you try it). That ringing is real physics, and the sidebar later explains where it comes from.

That's the whole vocabulary. Four shapes, three numbers. Every EQ move in the rest of this book — through mixing, into Chapter 29's vocal chain, all the way to mastering's quarter-decibel world in Chapter 32 — is spelled with this alphabet.

Cut First: The Subtractive Doctrine

Now the doctrine that separates the two kinds of EQ user. Subtractive EQ means cutting — turning down the frequencies causing a problem. Additive EQ means boosting — turning up the frequencies you wish there were more of. Both are legitimate; this book has no dogma, only consequences. But the order is not negotiable: reach for the cut first. When a track sounds dull, the amateur boosts the highs. The professional asks what's covering the highs — and usually finds something to remove. Three reasons, each one load-bearing.

Reason one: masking — the boost you want is usually a cut you owe. When the vocal sounds buried, it's rarely because the vocal lacks energy. It's because something else — the pad, the guitars, the cymbal wash — is sitting on the vocal's frequencies and masking it. Boost the vocal's presence and you've escalated the turf war: the vocal now masks the pad, so next pass you'll brighten the pad, which buries the hats, which... this is the amateur doom-loop Chapter 4 diagnosed, run with EQ instead of faders. Cut the pad where the vocal lives, and the vocal steps forward without getting louder, the pad loses nothing it needed, and the building gains floor space instead of losing it. Jaylen's chorus is the proof you've already watched: nine decibels of cuts bought him a vocal "boost" he never had to make.

Reason two: headroom — you spent Chapter 21 buying it; boosts spend it back. Every boost adds energy. Boost 3 dB here and 4 dB there across thirty-four tracks and your carefully gain-staged session creeps right back up toward the ceiling you cleared — track averages drifting hot above -18 dBFS, the 2-bus eating the headroom mastering was promised. Cuts move you away from the cliff. A subtractive mix gets clearer and quieter at once, and the monitor knob — not the channel fader — is the correct tool for getting the level back. (When you do boost, Chapter 21's plugin-gain-matching habit applies: pull the EQ's output trim down so the processed track returns to its staged level.)

Reason three: the boost-bias trap — boosting grades its own homework. Here's the psychology, and it's the sneakiest of the three. Every boost makes the track louder. And Chapter 4 burned the law into you: louder sounds better — not louder, better — fuller, clearer, more exciting, automatically, every time, to every human. So every boost you audition arrives pre-approved by your own auditory bias. Boost anything, anywhere, and your brain will report improvement. Cuts run the same bias in reverse: they make the track quieter, so every honest cut sounds, for the first two seconds, like a small loss. The result writes itself — beginners boost everything (it always "works"), never cut (it never seems to), and their mixes get louder, denser, and worse all afternoon. Cutting isn't acoustically magic. It's psychologically protective: it biases you against your own bias. And later in this chapter, the matched-loudness law will remove the bias from both directions entirely.

One honest boundary on the doctrine, because dogma is how good rules curdle: this is subtractive first, not subtractive only. A presence lift on a dull vocal, an air shelf on a great capture, a thump boost on a polite kick — all real moves you'll make, some of them later today. Make them after the cuts have done the structural work, make them wide and modest, and make them survive the matched-loudness test you're about to learn. Cut the problems. Then, if anything still needs celebrating, boost it gently and honestly.

High-Pass Discipline — and the 100 Hz Lemming

The single highest-value EQ habit, per second of effort, is the high-pass filter — and it's subtractive doctrine applied at the bottom of the building.

Almost every track in your session carries energy below its lowest useful frequency. Not musical energy — freight: mic-stand rumble through the floor, chair creaks, HVAC murmur, plosive thumps, the pad patch's sub-harmonic haze, the acoustic guitar's body bumping the mic. Individually inaudible. But the sub and low floors are where masking pressure is strongest and headroom is most expensive — big slow waveforms eat amplitude — and this freight stacks. Twenty tracks each contributing an invisible whisper of sub-100 Hz garbage equals one audible gray fog under your whole mix, stealing clarity from the kick and 808 who actually pay rent down there. So: high-pass everything that isn't bass. Vocals, guitars, keys, pads, hats, FX returns — each gets an HP filter clearing out the floor it doesn't musically occupy. On a thirty-track session, this one pass routinely does more for clarity than any other single move you'll make today.

And now the warning, because this advice has been repeated online so often it's mutated into a lemming march: do not set every high-pass to 100 Hz because a video told you to. The internet's "HP everything at 100" reflex produces a recognizable disease — mixes that are clean, polite, and thin, with no warmth and no weight, because thirty individually-harmless filters collectively amputated the low-mids' foundation. A male vocal's fundamental can sit near 100 Hz; an acoustic guitar's low E rings at 82 Hz; that "static bloom" pad's warmth starts well below where you'd guess. The number is not the discipline. The method is the discipline, and it's done by ear: play the track in the mix, sweep the HP cutoff slowly upward from the bottom, and listen for the moment the track starts to thin out or lose weight. The moment you hear it change, back off until it's whole again. Park it there. On one voice that's 70 Hz; on another it's 140; on a tambourine it's 400 and nobody mourns. Every track earns its own number — set by your ears, in context, not by habit. That's theme T1 in a single knob: the lemmings have the same filter you do; what they don't have is the listen.

🔄 Check Your Understanding

1. Name the three reasons cuts come before boosts, in one phrase each.
2. Your mix sounds clean but thin and gutless after your HP pass. What probably happened, and what's the by-ear method that prevents it?
3. A friend says "I never boost — boosting is wrong." What's this book's actual position?

Verify

1. Masking (the boost you want is a cut you owe someone else), headroom (boosts spend what Chapter 21 bought), and the boost-bias trap (louder grades itself as better, so boosts self-approve).
2. The 100 Hz lemming march: every HP set by number, not ear, collectively amputating real foundation. Method: sweep each cutoff up in context until the track thins, then back off to where it's whole.
3. Subtractive first, not subtractive only. Boosts are legitimate after the cuts do the structural work — wide, modest, and verified at matched loudness. No rules, only consequences.

🧩 Productive Struggle: Before this chapter hands you the map, go find the territory with your ears. Open your own mix-in-progress — or any rough mix you've made — and loop its busiest section. No analyzer; eyes off the screen. Three tasks, ten honest minutes. First: using Chapter 4's band vocabulary, write down which band feels most crowded. Second: name the three tracks you believe are contributing most to that crowding — and you may solo to listen, but notice (write down!) whether soloing actually helps you or misleads you. Third: predict, in writing, what would happen if you turned each suspect down 3 dB in only the crowded band. Keep the note. The next section is the answer key, and the gap between your guesses and the map is worth more than the map alone — and if your crowded band came out somewhere between 200 and 500 Hz, congratulations: you've just personally discovered the most common mix problem on Earth.

The Mud Map: Anatomy of the 200–400 Hz Pileup

Here's the section the book has owed you since page one. Why is it always the low-mids? Out of ten octaves of audible spectrum, why does nearly every amateur mix on the planet congeal in the same place — 200 to 400 Hz, the heart of Chapter 4's mud zone?

Because that's where everything lives. Look at where the energy of a typical session actually sits:

              sub      lows     LOW-MIDS      mids     high-mids    air
             <60 Hz  60–200 Hz  200–500 Hz  500–2 kHz   2–6 kHz    6 kHz+
kick           ███      ███       ▓ body                  ▒ click
808/bass       ███      ███       ▓▓ harmonics
bloom pad               ▓▓        ████         ▓▓           ▒         ▒
acoustic gtr            ▓         ████         ▓▓          ▓▓         ▒
keys                    ▓         ▓▓▓          ███          ▓         ▒
lead vocal                        ▓▓ chest     ███         ███        ▓▓
harmonies ×3                      ████         ▓▓           ▓
snare/clap              ▓         ▓▓ body      ▓▓          ███        ▓
hats                                                        ▓▓       ███
                                  ━━━━━━
                                  ⚠ the pileup: everyone deposits
                                    "a perfectly reasonable amount"

Caption: the mud map — every instrument's energy by band; no single track overloads 200–400 Hz, but nearly every track contributes, and the contributions sum.

Run the census. The fundamentals — the actual notes — of most pop instrumentation live in the lows and low-mids: the meat of a vocal's chest register, guitar and piano in their working ranges, the upper half of basslines, synth pads by design. Then stack the second deposits: every sound's low-order harmonics land just above its fundamentals — the 808 rooted at A1, 55 Hz, throws harmonics at 110, 165, 220 Hz, marching straight up into the zone (Chapter 1's harmonic-series recipe, sending its bill). Drum shells resonate there: snare body around 200–250, kick boxiness near 300, tom fundamentals throughout. Room sound recorded into every mic'd take piles up there too — small-room resonances love the low-mids, as Chapter 10's modes predicted. Result: 200–400 Hz is the one neighborhood where nearly every tenant in the session holds a lease.

Now add the two properties that turn crowding into mud. First, masking is at its cruelest here: the band is wide enough to hold many sounds and low enough that its congestion masks upward, smearing the mids and stealing definition from instruments that don't even live in the low-mids. That's why mud doesn't sound like "too much 300 Hz" to an untrained ear — it sounds like "everything is vague," a blanket over the whole mix. Second — and this is the insight that wasted three of Jaylen's days — the pileup is distributed. Each track contributes a modest, musically legitimate amount. Solo any one and it sounds rich, warm, professional. The crime only exists in the sum. Mud is a committee decision, and committees don't confess under solo interrogation.

Which means the fix is also distributed, and that's what makes it fast. You don't need one heroic 9 dB crater on a scapegoat track — that move leaves a hole the ear notices and the committee fills anyway. You need small cuts on the biggest contributors. Here is the book's founding promise, paid in full:

The 60-second mud fix. (1) Loop the muddiest section. (2) Confirm the bulge is low-mid: an analyzer on the 2-bus may corroborate, but trust the blanket-over-the-mix symptom your ears already know. (3) Find the committee's three biggest members by mute test, not solo: mute candidates one at a time and listen for whose absence clears the most fog — pads, rhythm guitars, stacked harmonies, keys, and dense bass harmonics are the usual ringleaders. (4) On each: one wide bell (Q around 1–1.4), 2–3 dB down, centered where that track crowds most — usually somewhere between 250 and 350 Hz; let your ear slide the center, since the guitar's deposit isn't at the pad's address. (5) Match loudness and A/B — the law arrives in a few pages. Done. Sixty seconds is the fix; the verification is the rest of your listen.

One honesty clause before you swing the scalpel, owed to Chapter 20's threshold concept: a mix problem is usually an arrangement problem wearing a disguise. If the chorus has five sustained instruments all holding chords in the same octave, the committee isn't an EQ problem — it's an over-cast section, and the mute button is the better surgeon (Chapter 16's subtraction pass, still on call). Jaylen's chorus had already survived the Chapter 16 mute test; every part defended its existence. What remained really was EQ's job: not too many parts — too many parts carrying low-mid weight at once. Carve when the parts all belong; cut parts when they don't.

🔍 Why Does This Work? Why does cutting 300 Hz on the pad make the vocal clearer — a track you never touched? Because audibility isn't a property of a track; it's a property of a relationship. Your ear's resolution in any band is finite, and what you perceive of the vocal is whatever clears the masking floor set by everything else in its bands. The pad's low-mid energy was raising that floor under the vocal's chest register; drop the floor 3 dB and everything standing on it rises 3 dB perceptually, for free, with no fader move and no headroom spent. This is also why the mud fix makes mixes feel louder while making them quieter in meter terms: un-masked detail reads to the brain as proximity and power. EQ's real product was never tone. It's audibility, redistributed on purpose.

Frequency Pockets: One Owner Per Zone Per Moment

The mud fix is triage. The system behind it — the thing professionals do from the first EQ move, so the committee never forms — is pocketing: deciding, for each band of the spectrum, which instrument is the featured owner at any given moment, and carving everyone else into supporting roles there. One owner per zone per moment. Everyone else visits; one tenant holds the lease.

The classic demonstration — promised to you, with a diagram and a deferred bill, back in Chapter 13 — is the kick and bass carve. Two instruments, one crowded basement. "Static Bloom" runs the handoff: the kick speaks the attack, the 808 carries the note, rooted at A1 ≈ 55 Hz. The EQ move that lets them coexist is a pair of complementary cuts — each instrument dips exactly where the other one reigns:

KICK channel EQ
 0 dB ┄┄╮        ╭───────────────────────────╮  ╭─╮ ┄┄┄
         ╲      ╱                              ╲╱   ╲       click ~4 kHz: kept
   HP 40  ╲    ╱   −3.5 dB @ 58 Hz              body ~105 Hz: the kick's home
           ╲  ╱    (clears the 808's throne)
            ╲╱
808 channel EQ
 0 dB ──────────────╮      ╭──────╮┄┄┄┄┄┄┄┄┄┄┄┄┄┄┄┄┄┄┄┄┄
                     ╲    ╱        ╲  harmonics 110–220 Hz: kept on purpose
   fundamental ~55 Hz ╲  ╱          ╲ (small-speaker survival — Chapter 26
   owns the floor      ╲╱            will make them glow)
                    −2.5 dB @ 105 Hz
                    (clears the kick's knock)

Caption: the complementary carve — the kick dips at the 808's fundamental, the 808 dips at the kick's knock, and each owns its pocket; full move-by-move log in case study 1.

Look at the logic, because it generalizes to every crowded pair in every mix you'll ever make. The 808 owns the sub floor around 55 Hz, so the kick — which doesn't need 55 Hz to be felt as a kick — donates that zone with a modest dip. The kick owns the punch zone around 100, so the 808 returns the courtesy. Each cut is small, wide-ish, and invisible in solo; together they turn a smear into a handshake. Both instruments end up louder where it counts without either fader moving, and the 808's harmonics at 110–220 are deliberately spared — they're the part a phone speaker can actually reproduce, the down payment on Chapter 26's saturation trick. (EQ gets the two of them out of each other's space; getting them out of each other's time — the moment-by-moment duck when both hit at once — is sidechain compression, and that bill stays parked where Chapter 13 left it: Chapter 28.)

Now run the same thinking up the building. Lead vocal and bloom pad both want 2–4 kHz? The vocal is the song; the pad donates a wide, gentle 2 dB dip there — a move so standard it has a nickname in a hundred studios: carving a hole for the vocal. Hats and vocal sibilance both want the air? Decide whose sparkle is featured. Acoustic guitar and keys both warming the low-mids? Pick the verse's owner and the chorus's owner — because that's the last clause of the rule and the one that keeps mixes alive: per moment. Ownership rotates. In the bridge, when Demi drops out, the pad can take the presence floor back; the verse guitar can keep low-mid warmth it must surrender when the harmony stack arrives at the chorus. For now, set the carves for the densest section — the chorus, almost always — and where verse and chorus truly need different maps, note it; Chapter 27's automation will let the leases change hands in real time, and Chapter 28's dynamic tools will make some handoffs automatic.

Write your map down. Six bands, your track's instruments, one featured owner per band for the big sections — the Project Checkpoint makes you do exactly this for your own song, and the discipline of seeing the zoning plan is half of what keeps you from boosting three tenants into the same floor at midnight.

🔄 Check Your Understanding

1. Why does the mud fix use small cuts on several tracks instead of one big cut on one track?
2. In the kick/808 carve, why does each instrument cut where the other is strongest?
3. What does "per moment" add to "one owner per zone"?

Verify

1. Because the pileup is distributed — each track contributes a reasonable amount and the crime is the sum. One crater leaves an audible hole and the committee refills it; several 2–3 dB cuts dissolve the sum while leaving every member musically intact.
2. Complementary carving: each donates the zone it doesn't define itself by, so both get louder where it counts without fader moves — the basement stops being a turf war and becomes a handoff.
3. Ownership rotates with the arrangement: the pad may own presence in the bridge that it donates to the vocal in the chorus. Set carves for the densest section now; automation (Chapter 27) re-assigns leases in time.

⚙️ Advanced Sidebar: Minimum-Phase vs Linear-Phase EQ

Somewhere in your EQ's settings hides a switch labeled linear phase, and audio folklore says it's the "cleaner" mode. Here's the honest version, in plain words.

Every ordinary EQ — called minimum-phase, the default in effectively every DAW — changes two things at once: the frequency balance you asked for, and, as an inseparable side effect, the timing of frequencies near the move. Boost a bell and frequencies around it get delayed by tiny, frequency-dependent amounts — phase shift. This isn't a defect; it's physics, the same cause-then-effect behavior as every analog EQ ever built, and it's where a high-Q boost's "ring" comes from: a narrow filter is a resonator, and resonators take time to start and stop. In normal mixing doses, the phase shift of a few gentle bells is inaudible as such — decades of beloved analog EQs prove it daily — though it's one more quiet reason the subtractive, low-drama doctrine ages well: big, steep, stacked moves mean big, stacked time-smearing.

Linear-phase EQ uses a different recipe: it delays all frequencies equally, shifting none of them relative to each other. The cost is threefold. It needs significant latency to work (it literally looks ahead), it burns more CPU, and — the genuinely strange part — it produces pre-ringing: a faint whoosh before sharp transients, because the filter's response is smeared symmetrically into the past as well as the future. Effect before cause. On a kick drum, a steep linear-phase low cut can put a small inhale ahead of every hit — an artifact minimum-phase can never make, since analog-style causality only smears after.

So when does anyone choose it? Mostly mastering — Chapter 32 territory — where EQ is applied to the summed stereo mix and to parallel-processing situations where phase-shifted copies would get recombined with originals and comb-filter against them. There, sameness of timing across a full mix can matter more than transient purity, and the moves are gentle enough that pre-ringing stays below notice. In an ordinary mix, on ordinary channels: minimum-phase, the default, is the right tool essentially always. The folklore has it backwards — linear phase isn't the clean mode; it's a different tradeoff, and now you know both price tags.

The EQ Pass in Practice

Model, vocabulary, doctrine, map, pockets. Now the bench work: the techniques you'll actually run, in roughly the order you'll run them, and the one law that makes every move honest.

Sweep and Destroy — and When to Stop

The mud fix and the pockets handle broad problems. But some problems are narrow: one ugly resonance — a boxy honk in a vocal recorded in a small room, a ringing overtone in a snare, a wolfy note in an acoustic guitar. You can hear that something is wrong without being able to name the frequency. Sweep-and-destroy is the searchlight.

┌──────────────┐   ┌──────────────┐   ┌──────────────┐   ┌────────────────┐
│ 1. BOOST     │ → │ 2. SWEEP     │ → │ 3. FIND      │ → │ 4. CUT         │
│ one bell,    │   │ drag it      │   │ the spot     │   │ flip gain to   │
│ +8–10 dB,    │   │ slowly       │   │ that honks / │   │ −2 to −4 dB,   │
│ Q ≈ 4–6      │   │ low → high   │   │ rings /      │   │ widen Q a bit, │
│              │   │ while looped │   │ hurts most   │   │ matched A/B    │
└──────────────┘   └──────────────┘   └──────────────┘   └────────────────┘
     ⚠ EVERYTHING sounds ugly boosted 10 dB. The sweep only nominates —
       the matched-loudness A/B convicts. Cut as little as solves it.

Caption: the sweep-and-destroy workflow — a boosted narrow bell is dragged across the spectrum as a searchlight, then flipped into a measured cut.

The method: insert an EQ on the suspect track, create one bell, crank it to +8 or +10 dB at a fairly narrow Q (4–6), and slowly drag its center across the spectrum while the track loops. The boost acts as a magnifying glass — when the searchlight crosses the resonance, the ugliness blooms: the honk honks harder, the ring rings louder, and your ear flinches at a specific spot on the dial. Stop there. Now flip the gain from a big boost to a small cut — minus 2 to 4 dB, rarely more — widen the Q a notch (surgical-narrow finds; slightly-wider fixes sound more natural), and listen in context.

And the warning, in the same breath, because this technique has ruined more tone than it's rescued: everything sounds horrible boosted 10 dB and swept. Every frequency of every instrument on Earth fails that audition. The sweep doesn't find "bad" frequencies — it finds candidates, and most candidates are innocent body and character. Beginners discover the sweep, hear ugliness everywhere, and dig a dozen narrow craters into every track — Swiss-cheese EQ, the signature sound of a mix that's been hollowed into a phasey, vowel-y ghost of itself. The rule that keeps you safe has two halves: only go hunting when your ears flagged a problem in normal listening first (the sweep confirms and locates suspicions; it never generates them), and the cut only survives if the matched-loudness A/B — two sections from now — proves the track better with it than without. One or two convictions per track is typical. Five means you're destroying, not sweeping.

Presence Without Pain: The 2–6 kHz Tightrope

The high-mids are where mixes are won and ruined, often in the same afternoon. Chapter 4 told you why this band is special: your ear canal's resonance peaks your sensitivity around 2–5 kHz — evolution's gift for consonants and alarm cries — so energy here reads as presence: closeness, articulation, attack, the vocal's diction, the snare's crack, the guitar's bite. The same physiology sets the trap: it's also where energy reads, with astonishingly little excess, as harshness — the glassy, fatiguing edge that makes listeners turn a mix down without knowing why. Chapter 1's car disaster filed its report from this band too: hi-hats that "shred glass" are a high-mid crime.

Walk the rope with three habits. First, respect the exchange rate: 1 dB at 3 kHz buys what 3 dB buys at 300 Hz. Moves here are small moves — if you're boosting more than 2–3 dB in the high-mids, the problem is almost certainly masking (something covering the presence that exists) or the source (wrong mic axis, dull take — Chapters 7 and 11 were the cheaper fix). Second, cut the harsh before boosting the present — subtractive doctrine at altitude. Harshness is usually specific: a resonance near 2.5–4 kHz on one offender (cheap-condenser presence peaks stack up fast — Chapter 7 warned you every budget "vocal mic" hypes the same zone). Sweep-and-destroy the offender narrowly, and clarity often returns with no boost at all. Third, when you do lift presence, lift wide and gentle — a 1–1.5 dB, Q ≈ 1 bell where the instrument's articulation lives, verified at matched loudness, on the featured instrument only. Presence is a pocket like any other: if the vocal, the snare, the guitar, and the hats all get a presence boost, you've reinvented the committee one floor up, and this committee bites. One more honesty note: some harshness is dynamic — fine most of the time, painful on peaks; sibilance is the classic case. A static EQ cut deep enough to catch the painful moments dulls all the others. Hold that thought for two sections.

Air, Honestly

Above roughly 8 kHz lives the air band — sheen, breath, the expensive shimmer that makes records feel finished. An air shelf — a gentle high shelf, +1 to +3 dB, cornered between 8 and 12 kHz — is the most seductive move in EQ, and this section exists because seduction needs a chaperone.

The honest rules. Air boosts amplify what's actually up there — and on a great capture that's breath, string sparkle, cymbal silk. On a budget capture in an untreated room, it's hiss, spit, mouth noise, and the brittle edge of a cheap capsule's top octave. The shelf is a magnifier, not a generator: if the air isn't in the recording, EQ turns up the absence. (When the top end genuinely isn't there to magnify, the real fixes live elsewhere: a better capture next time, or Chapter 26's saturation, which creates new high harmonics rather than amplifying noise.) Air flatters instantly — brighter reads as better even harder than louder does, so an unverified air shelf is the most self-deceiving move in this chapter; matched-loudness or it didn't happen. And air is fatigue's first victim: Chapter 4 taught you that listening fatigue dulls your high-frequency judgment as the session wears on — which is why mixes EQ'd at hour three grow air shelves like mold, and why morning ears so often pull them back off. Add air early in the session or not at all today. The professional air shelf is real, modest, on one or two featured elements — the lead vocal, maybe the hats or pad shimmer — and decided by fresh ears. The amateur air shelf is on everything, at hour four, and by Chapter 30 it'll be in the symptom table under "harsh."

🔄 Check Your Understanding

1. Why must the sweep only confirm problems your normal listening already flagged?
2. You want a brighter vocal. What two things do you check before boosting 4 kHz?
3. Why are air shelves the most self-deceiving EQ move?

Verify

1. Because everything sounds ugly boosted 10 dB — the sweep nominates candidates indiscriminately. Ears-first keeps you cutting actual problems instead of Swiss-cheesing innocent character.
2. Masking (is something covering the presence that's already there? carve the masker instead) and the source (mic choice/axis/performance — Chapters 7 and 11 fixes are cheaper than EQ). Then boost small, wide, and verified.
3. Brighter auto-reads as better, fatigue erodes your top-end judgment as sessions wear on, and a shelf magnifies whatever's up there — including noise. Fresh ears plus matched-loudness A/B is the only honest chaperone.

The Matched-Loudness A/B Law

Everything in this chapter so far tells you what to try. This section is how you know whether it worked — and it's the discipline that, more than any plugin, separates mixes that improve all afternoon from mixes that get sideways all afternoon. It has come up in fragments before: Chapter 4 discovered the bias, Chapter 21 applied the gain-matching habit to plugin chains. Here it becomes this book's formal law, because EQ is where it pays or punishes hardest:

The Matched-Loudness A/B Law: no EQ decision counts until you've compared processed against bypassed at equal perceived loudness. Louder is not better. Louder is louder.

Why a law and not a tip? Because of what you already know and cannot turn off. Louder sounds better — to you, to me, to every engineer with forty years in — and it doesn't arrive labeled as volume; it arrives disguised as quality: fuller, wider, clearer, more pro. Nearly every EQ move changes loudness. Boosts add it, cuts remove it. Which means every unmatched A/B is rigged before you toggle it: the boost auditions louder and wins unearned, the cut auditions quieter and loses unfairly, and a full afternoon of unmatched "improvements" optimizes your mix for loud instead of good — usually nine boosts deep into harsh, headroom-starved, and somehow worse than the static mix you started from. Every engineer learns this one the expensive way. The law is how you stop paying tuition.

The ritual, four steps, thirty seconds:

1. Make the move. Cut, boost, shelf — work your ears freely while shaping.
2. Match the level. Use the EQ's output/gain trim to compensate for what your moves added or removed, until toggling bypass produces no jump in size — only a change in character. By ear is fine and is the skill; toggle and trim until neither version feels bigger. (Many modern EQs offer auto-gain or loudness-compensated bypass — use it gladly, then verify by ear; it's estimating.)
3. Toggle blind-ish. Look away from the screen — the eyes outvote the ears whenever you let them watch — toggle a few times, lose track of which is which if you can, and ask one question: which version serves the song?
4. Issue one of three verdicts. Better — keep it. Worse — undo it, gladly; an undone move is knowledge, not failure. Different-but-not-better — and this is the verdict that builds professionals — bypass wins by default. Every active band costs something: a little phase shift, a little CPU, one more decision your future self must re-litigate. A move that can't beat bypass at matched loudness hasn't earned its slot. Most beginners' EQ moves, honestly tested, land here — which is why honest testers end up with less EQ doing more work. That's theme T2, enforced by procedure rather than promised by slogan.

Apply the law at every scale: each move, each track's whole EQ chain (bypass all bands — did the chain earn its life?), and the session level you'll hit in the Project Checkpoint. And notice what the law really is, because it's bigger than EQ: it's the only known antidote to your own auditory bias, and from here to the end of the book it governs everything — compression in Chapter 23, saturation's drive-and-match in Chapter 26, your master against references in Chapter 32. Trained ears plus an honest procedure beat golden ears plus self-deception, every single time. You can't buy this discipline, which is exactly why it's worth more than anything you can buy.

EQ Before or After Compression?

The question every mixer gets asked, and the rare one where "it depends" is the complete, correct answer — both orders are right, for different jobs, and naming the tradeoff now is a handshake with Chapter 23, where compression gets its full treatment. (For today, the one-line version: a compressor is an automatic level-rider that turns a track down when it gets loud, reacting to the signal you feed it — and that italic clause is the whole debate.)

EQ before the compressor when you want the compressor to react to the corrected sound. The canonical case is the high-pass filter: feed a vocal or a full bus into a compressor with its low-frequency freight still aboard, and the compressor — whose detector is dominated by big low-frequency energy — clamps down on every plosive thump and bass swell, audibly ducking the whole track in sympathy with rumble nobody can even hear. HP first and the compressor reads the music instead of the freight. Same logic for the mud: compress an uncarved track and the committee's energy drives the gain changes; cut first and the compressor regulates what matters. Hence the working default you'll see all over this book: corrective, subtractive EQ before compression.

EQ after the compressor when you want your tonal shaping left alone. Any boost placed before a compressor gets partially un-done by it: lift the vocal's presence 2 dB pre-compressor and the detector now sees a hotter signal there and compresses more, giving some of your lift back. Place the presence lift, the air shelf — the additive, flattering moves — after the compressor, and they sit on top of the leveled signal, stable and predictable.

Put the two answers together and you get the professional default — the EQ sandwich: subtractive EQ → compressor → additive EQ. Cut the problems so the compressor reads honestly; compress; then sweeten what's left. It's not a rule, it's a reasoned starting point — and you'll watch Chapter 29 build Demi's entire vocal chain on exactly this skeleton.

A Preview: EQ That Moves

One more tool deserves a name today, even though its chapter is ahead. Some problems are intermittent — the boom that only happens on certain notes, the harshness that only bites on the loud phrases, the sibilance from two sections ago. Static EQ faces an ugly choice there: cut deep enough for the bad moments and you've dulled all the good ones. Dynamic EQ resolves it — an EQ band that sits at zero and only dips when the problem frequency actually misbehaves, like a bell with a tiny compressor inside. It's the surgical middle ground between the EQ you learned today and the compression you'll learn next, and Chapter 28 covers it properly, alongside its cousins. For now: when you find yourself wishing a cut could apply sometimes, write the spot down and leave it. The tool exists, it's coming, and a de-esser — Chapter 29 — is the most famous member of the family.

Genre Accents: Same Physics, Different Verdicts

Everything in this chapter is physics and psychoacoustics — masking, headroom, bias — and physics doesn't care what genre you make. But the verdicts — how much mud is mud, how much air is enough, who owns the low-mids — are aesthetics, and aesthetics are genre contracts, exactly as Chapter 18 framed them. The mud map is universal. The sentencing guidelines are not.

A few honest accents. Hip-hop and trap protect the basement above all: the 808's zone is sacred, high-pass discipline on everything else is aggressive, and the vocal rides bright and close — carved-out, present, compressed forward. Mud tolerance: very low; the aesthetic is space and weight, not warmth. Rock is the great counterexample, and a vaccine against everything this chapter could turn dogmatic: a wall of guitars is low-mid energy — 200–400 Hz is where guitar power chords live, on purpose, as the genre's signature. Over-clean a rock mix to pop standards and you've gutted it; the move is carving small vocal-shaped and snare-shaped holes into the wall, not demolishing the wall. Pop is the most ruthlessly pocketed genre on Earth: every element carved to its lane, vocal presence maximal, mud tolerance near zero, air on the money elements. R&B keeps more low-mid warmth in the vocal than pop dares — chest tone is intimacy, the air silky rather than bright. Electronic and house make the kick/bass pocket the genre's actual heartbeat, with everything melodic high-passed startlingly high; folk and acoustic music (Wren & Hollow's home) preserve body religiously — gentle moves, real low-mids, because "warm and woody" is the entire product. None of this is a preset to copy. It's theme T4 doing EQ's homework: pull up your three references, listen band by band with Chapter 4's five-pass discipline, and let your genre's records tell you how its EQ verdicts run. The gap between their tonal balance and yours is your to-do list.

No Solo EQ: Context Is King

Last technique, and it's a prohibition — the rule that quietly governs everything above: don't make EQ decisions in solo.

The solo button lies to you twice. First lie: solo removes the masking environment, so a track carved correctly for the mix sounds thin and wrong alone — and five minutes of solo'd "fixing" will undo every pocket you cut, returning the track to gorgeous-but-fighting. Second lie: solo hides the crime you're committing — boost the acoustic guitar's low-mids in solo because it sounds rich, and you can't hear that you've re-convened the committee and buried the vocal until you un-solo, by which point your ear has re-anchored and the damage reads as normal. Jaylen's three lost days were this exact mechanism: in solo, every committee member sounded innocent, because the crime only exists in the sum — and the sum is the thing solo deletes.

The working rule: solo to find, mix to decide. Soloing is legitimate, briefly, as a flashlight — to locate a click, to run a sweep on a suspect resonance, to learn what a track even contains. The moment your hand moves toward a decision — does this cut stay? is this brighter version better? — the solo comes off, the whole mix plays, and the matched-loudness law presides. The mix is the only courtroom, because the mix is the only thing anyone will ever hear. Nobody streams your solo bus.

🔄 Check Your Understanding

1. Recite the three matched-loudness verdicts and what happens on each.
2. Why does an HP filter usually go before a compressor, but a presence boost after?
3. The solo button "lies twice." Name both lies.

Verify

1. Better → keep. Worse → undo, gladly. Different-but-not-better → bypass wins by default — an active band must earn its cost (phase, CPU, complexity), which is theme T2 as procedure.
2. The compressor reacts to what it's fed: HP first so the detector reads music instead of low-frequency freight; boosts placed before get partially compressed back down, so flattering EQ goes after — the EQ sandwich.
3. Solo deletes the masking context (correctly-carved tracks sound wrong alone) and hides the damage your "improvement" does to everyone else (the committee re-forms unheard). Solo to find, mix to decide.

Common Mistakes and the 60-Second Fixes

The diagnostic table below is this chapter's pocket edition of the emergency room Chapter 30 will build in full; Appendix A carries the complete frequency chart, instrument by instrument. First, the seven EQ mistakes that account for most of what goes wrong, each with its fast fix:

Soloed-perfection EQ. Every track polished alone; the mix worse than before you started. Fix: bypass every EQ (yes, all of them — sixty seconds of courage), play the mix, re-make only the moves the mix asks for, in context.

The everything-boost. A smiley-face curve on every channel, level creep everywhere, harshness by hour two. Fix: for each boost, try the equal-and-opposite cut on the masker instead — brighter vocal via duller pad — and gain-match what survives.

The 100 Hz lemming march. Clean, thin, gutless. Fix: re-sweep every HP cutoff by ear, in context, until each track is whole again; expect many to land lower than the habit said.

Swiss-cheese surgery. A dozen narrow cuts per track; hollow, phasey, vowel-y tone. Fix: bypass all bands, then re-fix with at most two or three wider, gentler moves — if it takes more, the problem is upstream (source or arrangement).

EQing the arrangement. Heroic carving trying to fit five chord instruments into one octave of space. Fix: Chapter 20's threshold concept — mute before you carve; the best EQ for a part the song doesn't need is the off button (Chapter 16's subtraction pass).

The unmatched A/B. Every move "helps"; the afternoon ends louder, harsher, and worse. Fix: the law. Gain-trim, toggle blind, three verdicts. Re-audit today's keepers tomorrow at matched loudness and watch a third of them lose.

Mixing the analyzer. EQing bumps you can see into curves you can't hear, on a display that flatters straight lines. Fix: analyzer for hypotheses, ears for verdicts — eyes off the screen at decision time (T1, always T1).

Caption: the problem-frequency quick reference — symptoms to neighborhoods to first moves; the full instrument-by-instrument chart lives in Appendix A.

🔄 Check Your Understanding

1. Your mix got worse as every track got "better." Which mistake is this, and what's the sixty-second first response?
2. A track needed five narrow cuts before it sounded okay. What does that count usually mean?
3. The analyzer shows a bump at 120 Hz you can't hear as a problem. Do you cut it?

Verify

1. Soloed-perfection EQ. Bypass every EQ in the session, play the whole mix, and re-make only the moves the mix asks for — in context, under the law.
2. The problem is upstream: a poor source (re-record, re-choose the sample — Chapters 7, 11–13) or an arrangement collision (Chapter 16's mute test). Swiss cheese is a symptom, not a technique.
3. No. The analyzer generates hypotheses; ears issue verdicts (T1). If a flagged bump survives an honest in-context listen as a non-problem, it's character, not crime — plenty of great records show "wrong" curves.

Project Checkpoint: Building "Static Bloom"

Where the project stands: Chapter 20 left "Static Bloom" statically balanced — thirty-four tracks, six buses, faders only — and Chapter 21 left it gain-staged, averages near -18 dBFS with six clean decibels of 2-bus headroom. This checkpoint runs the EQ pass, in the order the chapter taught it.

Pass one — high-pass discipline. Jaylen works bus by bus, sweeping each track's HP up in context until it thins, then backing off: Demi's lead parks at 90 Hz, the harmony stack at 120, Theo's acoustic at 85, the bloom pad at 95 (its sub-haze was real freight; its warmth starts higher than he guessed), hats and shakers way up at 300, FX returns at 200. The 808 and kick get no HP above their basements (40 Hz on the kick) — they are the basement. The mix gets quietly, unglamorously clearer; the 2-bus peaks drop almost a decibel for free.

Pass two — the committee. The hook's fix, formalized: wide bells, -3 dB at 300 Hz on the pad, -3 dB at 290 on the acoustic, -3 dB at 320 on the harmony bus — centers slid by ear, since each member crowds a slightly different address. Demi's lead steps forward, untouched.

Pass three — pockets. The zoning map gets written into the session notes: sub = 808; punch = kick; low-mid warmth = guitar (verses), pad (chorus, post-carve); mids = lead vocal with keys supporting; presence = lead vocal, snare second; air = hats plus a 1 dB vocal shelf that survived a matched A/B on morning ears. The kick/808 complementary carve goes in exactly as diagrammed — case study 1 logs every move and verdict.

Pass four — the audit. Every move re-tried against bypass at matched loudness. The day's tally: 31 moves made, 9 reverted (including two air shelves and a presence boost that were louder, not better), 22 kept — most under 3 dB. The mud is gone, and nothing sounds EQ'd. That's the goal state.

Your track. Run the same four passes on your own session: HP by ear (log each cutoff), find and cut your committee, write your six-band ownership map, carve your low-end pair, and audit everything under the law. Genre adaptations: hip-hop — settle the kick/808 handoff first; it is your mix. Rock — your guitars own the low-mids on purpose; carve vocal- and snare-shaped holes instead of cleaning the wall. Electronic — fix tone at the synth (Chapter 14's filters) before reaching for channel EQ; sound design is pre-EQ. Folk/acoustic — half-decibel moves, body preserved; your committee fix may be 1.5 dB, not 3. Next chapter, the compressors arrive — and they'll read every frequency decision you made today.

Cross-Chapter Connections

Backward. Chapter 1's harmonic series is why instruments trespass far above their fundamentals — EQ reshapes the recipe Fourier wrote, and the companion volume's Chapter 7 holds the deep math. Chapter 4 built this chapter's entire sensory apparatus: the six bands became a control surface, perceptual masking became the thing EQ manages, and louder-sounds-better became the law's reason for existing. Chapter 7's placement-beats-purchase and Chapter 11's mic technique are pre-EQ — the cheapest EQ move is the one capture made unnecessary. Chapter 10 drew EQ's hard boundary: you can't EQ away comb filtering or your room's nulls, because they aren't in the signal. Chapter 13 tuned the kick and designed the 808 handoff this chapter's carve completed. Chapter 14's filters returned in lab coats. Chapter 16's subtraction pass and Chapter 20's threshold concept police EQ's jurisdiction — carve parts that belong; cut parts that don't. And Chapter 21's gain staging gave EQ its operating room: headroom to preserve, and the in/out matching habit the law generalized.

Forward. Chapter 23 inherits the sandwich: your subtractive work feeds its compressors honest signal, and the EQ-before-or-after question gets its full answer. Chapter 24 will EQ the spaces themselves — high-passed, darkened sends that sit reverbs behind dry sounds. Chapter 26 picks up every "the harmonics aren't there to boost" thread: saturation creates what EQ can only magnify, and the 808's phone-speaker glow gets built. Chapter 27 puts your pocket map in motion; Chapter 28 delivers dynamic EQ, mid-side EQ, and the sidechain duck the low end is still owed. Chapter 29 assembles the vocal chain on the sandwich skeleton, de-esser included. Chapter 30's symptom table expands this chapter's quick reference into the book's emergency room. And Chapter 32 returns EQ for mastering — quarter-decibel moves, where the linear-phase tradeoff finally earns its keep.

Spaced Review

Chapter 21: Your EQ pass ended with six boosts averaging +3 dB scattered across the session. In gain-staging terms, what just happened, and what's the habit that repairs it?

Verify

Boosts add energy: track averages drift above the ~-18 dBFS habit and the 2-bus eats into the ~6 dB of pre-master headroom. The repair is Chapter 21's plugin in/out matching — pull each EQ's output trim down so the processed track leaves at the level it entered — or better, prefer cuts and turn the monitor up. Headroom is a workflow, not a one-time setup.

Chapter 20: No matter how you EQ, the chorus stays cluttered. What does Chapter 20's threshold concept tell you to ask before reaching for another filter — and which sixty-second test answers it?

Verify

A mix problem is usually an arrangement problem wearing a disguise: is the section over-cast, with too many parts holding the same register at once? The mute test answers it — mute candidates one at a time; if the chorus improves more from a mute than any EQ move achieved, the fix belongs to arrangement (Chapter 16), not to EQ.

Chapter 18: A friend insists "you always cut 200–400 Hz — that's the rule." Using Chapter 18's conventions-as-contracts idea, name a genre where that "rule" would damage the record, and state the principle that replaces the rule.

Verify

Rock (and plenty of soul and folk): guitar walls and warm bodies live in the low-mids by contract — over-cleaning guts the genre's signature. The principle: the mud map is physics, but the verdicts are genre aesthetics; your references (T4) define how much low-mid energy your record owes its listeners, and you carve holes for the vocal and snare rather than demolishing the wall.

What's Next

EQ decided where every instrument lives. Chapter 23 decides how each one behaves over time — because the other half of why pro mixes feel controlled isn't tone at all, it's dynamics: vocals that stay on a rail, drums that punch without flailing, basses that hold steady. The tool is compression, the most misunderstood processor in audio, and the chapter opens with Demi's vocal disappearing in verse two and a compressor putting her back on the rail — then Jaylen flattening his drums learning the same knobs. Before you go: do the exercises (the DAW Lab's matched-loudness ritual in C6 is this chapter's discipline made muscle), take the quiz, and run your own EQ pass while the mud map is fresh. The committee is in session on your mix right now. You know how to adjourn it.