Chapter 22 Key Takeaways

The Big Ideas

1. EQ is frequency real estate, not tone improvement. Your mix has six floors (Chapter 4's bands) and fixed space; every sound moves in where its harmonics dictate, and masking enforces a winner-take-most policy per zone. EQ is the zoning law: deciding who lives where, who downsizes, and who gets evicted from a floor they never needed. A track EQ'd correctly for the mix often sounds worse solo'd — the judge is the mix, never the solo (theme T2: the professional carves tracks to fit, not to impress alone).

2. The filter alphabet is four shapes and a width knob. High-pass and low-pass (cut everything beyond a cutoff, at a chosen slope), shelves (lift/drop everything beyond a corner — the tonal tools), and bells (a hill or valley at a chosen center). Q sets bell width, and higher Q = narrower. Wide moves change tone; narrow moves fix problems; narrow boosts ring.

3. Cut first — for three reasons. Masking: the boost you want is usually a cut you owe the masker. Headroom: boosts spend what Chapter 21 bought; cuts move you away from the cliff. The boost-bias trap: boosts make tracks louder, louder grades itself as better, so boosting always "works" and slowly ruins the mix. Subtractive first, not subtractive only — boosts are legitimate after the cuts do the structural work, wide and modest and verified.

4. High-pass everything that isn't bass — by ear, never by number. Sub-frequency freight stacks across dozens of tracks into fog and wasted headroom. But the 100 Hz lemming march (every cutoff copied from a tutorial) produces clean, thin, gutless mixes. The method: sweep each cutoff up in context until the track thins, back off until whole. Every track earns its own number.

5. Mud is a committee decision — the book's founding example, resolved. Nearly every instrument deposits a reasonable amount at 200–400 Hz (fundamentals, low harmonics, drum bodies, recorded room sound), the deposits sum, and upward-spreading masking turns the sum into a blanket over the whole mix. Solo acquits every suspect because solo deletes the sum. The 60-second fix: mute-test to find the three biggest contributors, then wide bells, -2 to -3 dB each, centers slid by ear around 250–350 Hz. Carve when the parts belong; mute when they don't (Chapter 20's threshold concept polices the boundary).

6. Pockets: one owner per zone per moment. Assign each band a featured owner for the densest section and carve supporting players complementarily — the kick/808 carve being the classic: each dips where the other reigns (the 808 keeps ~55 Hz, the kick keeps ~100, and the 808's 110–220 Hz harmonics get protected as phone-speaker survival). Ownership rotates with the arrangement; Chapter 27 will automate the lease changes.

7. The Matched-Loudness A/B Law is this chapter's discipline cornerstone (T1). No EQ decision counts until processed and bypassed are compared at equal perceived loudness: move → output-trim until the toggle changes character but not size → toggle blind → one of three verdicts. Better: keep. Worse: undo gladly. Different-but-not-better: bypass wins by default — an active band must earn its cost. The law is the only known antidote to louder-sounds-better, and it governs every comparison from here through mastering.

8. Order matters, and both orders are right. Subtractive EQ before the compressor (so the detector reads music, not low-frequency freight), additive EQ after (so compression can't take your flattering boosts back): the EQ sandwich, the skeleton of Chapter 29's vocal chain. Dynamic problems — sibilance, intermittent harshness — are the wrong job for static EQ entirely; dynamic EQ (Chapter 28) and the de-esser (Chapter 29) are scheduled.

9. The physics is universal; the verdicts are genre. Rock's guitar wall owns the low-mids by contract; pop pockets ruthlessly; trap protects the basement; folk preserves body. Your references (T4) — not this chapter — set how much warmth your record owes its listeners. And no verdict gets issued in solo: solo to find, mix to decide.

Numbers Worth Keeping (anchors, not gospel)

Mistakes This Chapter Should Prevent

• EQing tracks to perfection in solo, then wondering why the mix got worse.
• Hunting for the one muddy track when the mud is a committee (Chapter 4 warned you; now you own the fix).
• High-passing everything at 100 Hz because a video said so.
• Swiss-cheese surgery: a dozen narrow sweep-found cuts hollowing a track that had one real problem.
• Auditioning any EQ move — especially boosts and air shelves — without matching loudness first.
• Fixing an over-crowded arrangement with heroic carving instead of the mute button.
• Trusting hour-three ears with high-mid and air decisions.

🎚️ "Static Bloom" Status

The EQ pass is done, and the founding wound is closed. High-pass discipline swept the whole session by ear — Demi's lead at 90 Hz, harmonies at 120, Theo's acoustic at 85, the bloom pad at 95, hats way up at 300, the 808 and kick exempt — and bought almost a decibel of 2-bus headroom for free. The chorus mud committee (pad, acoustic, harmony stack) was cut 3 dB each around 290–320 Hz, and Demi's vocal stepped forward without her fader moving. The six-band ownership map is written into the session notes, and the kick/808 complementary carve is in — the 808 owns ~55 Hz, the kick owns ~100, the 808's harmonics are protected for the phones, and case study 1 holds the full ledger: 31 moves attempted across the session, 9 reverted under the law, 22 kept, almost nothing deeper than 3 dB. The mix no longer sounds EQ'd; it sounds zoned. Your track: the same four passes — HP by ear, committee cut, ownership map, low-end carve — every move matched-loudness verified. Next, in Chapter 23: the compressors arrive. EQ decided where everyone lives; compression decides how they behave.

The One Thing

If this chapter leaves you with a single sentence: mud is a committee decision and clarity is a zoning decision — find the three biggest contributors, cut each a few honest decibels, and verify every move at matched loudness, because louder is not better, louder is only louder.