Chapter 30 Exercises

This set is the clinic's residency program: you've read the diagnostics, now you treat patients. You'll need your current mix (the one that's survived Chapters 20–29), at least two old mixes from before this book (the rougher the better — they're your teaching patients), your Chapter 4 reference playlist, your listening journal, and access to the testing panel: a car you can borrow, your phone, a pair of earbuds, and one quiet hour.

Three ground rules govern everything below, and they're the chapter's three disciplines wearing lab coats. First: diagnosis and treatment never share a sitting. Listening tours produce written symptoms, not fader moves — if you caught yourself adjusting something in the car, the exercise restarts. Second: matched loudness on every verdict. An A/B with a level mismatch is not evidence; it's a loudness preference with paperwork. Third: one change at a time, re-verified on the system that exposed the symptom. Exercises marked ★ have worked answers in the back of the book (Answers to Selected Exercises). Time estimates are honest averages. Do not run this set in one day — fresh ears are part of the apparatus, and several exercises explicitly require a night of sleep between diagnosis and surgery. Spreading the work across a week isn't a compromise; it's the protocol.

Part A: Conceptual Understanding

A1. The table, from memory. ★ (15 min) Without looking back, reproduce the eight-row symptom table: symptom name, what you hear, the address, the 60-second check, and the first fix. Then open the chapter and grade yourself cell by cell. Journal which cells your memory bent — those are precisely the rows you'll misdiagnose under pressure. Required follow-up: which three symptoms have no frequency address at all, and which axis does each live on?

A2. The oscillation trap. (10 min) Muddy and thin are described as opposite mistakes that cause each other. Explain the oscillation in one paragraph: why the cure for mud, overdone, becomes thin; why the cure for thin, overdone, becomes mud; and exactly which two pieces of the chapter's discipline break the loop. (Hint: one is an address plus a confirmation check; the other is a law you've been living under since Chapter 22.)

A3. The attribution test. ★ (10 min) A symptom shows up on one system and nowhere else. Write the exact protocol that decides whether it's your mix or the system's accent, explain why a mastered reference track is the right measuring instrument, and state what you do with each of the two possible outcomes. Use the word "accent" correctly at least once.

A4. The cost ladder, defended. (15 min) For each rung of the fix ladder — arrangement, source, edit, mix — name its cost profile, the kinds of problems that live there, and one concrete example from the chapter. Then state the rule that fires when you find yourself reaching for a fourth plugin on the same problem, and explain why the ladder climbs from the cheap end rather than the familiar end.

A5. Mute versus solo, the jurisprudence. (10 min) Explain why the solo button acquits committee crimes while the mute button convicts them — name the mechanism, not just the verdict. Then state the one diagnostic job solo legitimately keeps, and write the chapter's three-sentence working rule about mute, solo, and where decisions get made, from memory.

A6. Off the map. ★ (10 min) Lifeless, small, and cluttered never appear on the symptom-frequency map. For each: name its axis, explain why a spectrum analyzer is blind to it, give its 60-second check, and name the stage where its first fix usually lives. One of the three is described as "not a mix problem" at all — say which, and why that's the table's most important row.

A7. Tired ears prescribe. (10 min) List three distinct ways ear fatigue corrupts a diagnostic session — not just "you hear worse," but the specific failure modes the chapter names — and the scheduling rules that keep you a working instrument. Finish with one sentence: why was Jaylen leaving the Civic without opening his session protocol rather than patience?

A8. The honest meter. (10 min) For the null test: state what a clean null proves, what an audible residue actually is, the three uses the sidebar gives it, the two technical requirements that make it work, and the one thing a null can never tell you. Bonus: what does it mean when a chain won't null against itself twice?

Part B: 🎧 Listening Lab

Everything in this lab produces journal entries written in band-and-axis vocabulary. "Sounds off" is not a symptom; "low-mids piling up, chorus only, around 300" is. Chapter 4's bands are the address system — this lab is where they cash out.

B1. The archive autopsy. ★ (30 min) Open your two oldest mixes and run the full symptom table on each: top three symptoms per mix, each named with an address, a timestamp ("chorus only," "from the bridge on"), and a suspect list. Fix nothing — these are teaching patients, and you'll operate on the worst one in exercise C4. Keep both symptom sheets. A year of mixes diagnosed honestly will show you something the chapter promised: your mixes fail in characteristic ways, and the pattern is the most valuable thing in your journal.

B2. How the pros stay healthy. (25 min) Take one reference track from your Chapter 4 playlist and run eight listening passes — one per symptom row. For each row, write one sentence explaining why the symptom isn't there: "Not muddy because the low-mids belong to the piano alone and everything else is high-passed out of its way." "Not lifeless because the second chorus adds two layers the first didn't have." This inverts the table: you're hearing prevention in the wild, which is the same knowledge as cure, learned in a better mood.

B3. The three-differences drill, installed. ★ (15 min) Level-match a reference down to your current mix by ear — neither should feel louder. Thirty seconds of the reference, then the same section type of yours. Write exactly three differences, each in band or axis vocabulary. Then convert each difference into a fix-list item with a guessed stage (arrangement, source, edit, or mix). Finish with one sentence: why does the drill demand exactly three — not one, not ten?

B4. The full tour. (60 min, plus travel) Your current mix on all five tests: car, phone speaker, earbuds, quiet listen at conversation level, and the next-room test. Calibrate each system first with thirty seconds of a reference — learn the accent before you judge through it. Log symptoms per system, then run the attribution test on anything that appears only once: cross it off if the reference fails the same way. Deliverable: one consolidated symptom sheet, ranked by how many systems testify to each symptom. This sheet feeds C1 directly.

B5. Accent mapping. (20 min) Now reverse the experiment: play one reference (not your mix) across the same five systems and log what each system flatters, hides, and exaggerates. The car's smile, the phone's missing floor, the earbuds' microscope, the kitchen's skeleton. This sheet is permanent equipment — your personal footnotes to the translation-test matrix — and it's what makes every future tour faster: you'll know instantly which complaints are the system talking.

B6. The skeleton listen. (15 min) Conversation level, your mix, then a reference. Three questions, answered in writing: Does the vocal survive, or does your brain stop filling it in when the volume drops? Could you label verse versus chorus blind? Does anything you consider important simply vanish? Note which fix-list items this one free test surfaced — for most mixes it finds the balance and dynamics problems that louder listening flatters into invisibility.

B7. Fatigue forensics. (20 min) Moderate level, earbuds, ten uninterrupted minutes of your mix on repeat. Note when your ears first complain and which band the complaint lives in — fatigue almost always has an address, and it's usually in the 2–6 kHz neighborhood. Repeat with a reference at the same level. If your mix tires you measurably faster, you've located harshness that every thirty-second listen has been flattering. Harsh mixes demo well and live badly; this exercise is how you catch the lie before your listeners do.

Part C: 🎛️ DAW Lab

C1. The diagnostic pass. ★ (90 min, across two days) The Project Checkpoint, run exactly as written — this is the chapter's centerpiece, and if you do nothing else in this set, do this. Day one: the tour (three-system minimum plus the quiet test, or your full B4 results), symptoms written in band-and-axis vocabulary, nothing fixed. Day two, fresh ears: run the 60-second checks from the table on every symptom; build a ranked fix list, five items maximum, sorted by impact on vocal and groove; execute one change at a time at matched loudness; re-verify each change on the system that exposed the symptom; log every verdict — including the items that turn out to be deliberate choices rather than diseases. Deliverable: the fix log. Chapter 31 will not accept "it sounds good in here."

C2. One change at a time, proven. (45 min) Take three convicted fixes from C1 (or stage three plausible ones on an old mix). Version A: apply all three at once and bounce. Version B: apply them one at a time with a matched-loudness verdict between each, bouncing after each step. Now examine what each workflow taught you. In version A, suppose the mix got slightly worse — which fix did it? You cannot know, and you can't undo selectively without starting over. In version B you know exactly what each move bought or cost. Write the two-sentence moral and pin it where your impatient future self will find it.

C3. The check battery. (45 min) Run all five frequency-symptom checks on your current mix, even where you suspect nothing: the committee mute (your top three low-mid depositors, muted together, worst section looping), the boxy sweep (a -4 dB bell driven 300 to 600 Hz across the lead vocal), the harshness census (count every 2–6 kHz boost and every active saturation instance in the session), the HP bypass (every high-pass off for one chorus), and the room walk (loop the bassiest section and move around the room). Log all five results — including clean bills of health. Knowing what your mix doesn't have is half of diagnostic confidence, and the battery takes less time than one panicked forum post.

C4. Re-balance versus rescue. ★ (2.5–3 hours, plus the next morning) The single most instructive afternoon in Part VI. Take B1's worst patient. Save it as two new versions. Rescue path: sixty minutes maximum — symptom table, fix list, processing moves, your best honest effort at repairing the mix as it stands. Re-balance path: every fader to silence, then rebuild the static balance most-important-element-first per Chapter 20 — ninety minutes, faders and mutes only, before you're allowed to touch a single plugin. Bounce both, gain-matched. The next morning, listen blind on your phone and one other system: which translates better? Write the verdict plus a one-paragraph lesson about foundations. Most people's results match the chapter's claim that the re-balance wins. If yours don't, that's a real finding too — write down why, because it means the patient's foundation was healthier than its symptoms suggested.

C5. Null lab. (30 min) Four steps. (a) Duplicate any track, flip the polarity of the copy, play both, and confirm silence — you've verified the instrument. (b) Null your processed lead vocal against its bypassed self and describe the difference signal in band vocabulary: that residue is exactly what your chain adds and removes. (c) Null your 2-bus with and without its color chain. (d) Write the one-sentence law: what the null proves, and what it can never judge. Advanced credit: if something refuses to null at all, diagnose why — time-variant processing, or a plugin-delay-compensation problem — and note that the refusal is itself information.

C6. The confession audit. (30 min) Loop your busiest four bars. Mute the three parts you would defend least in court and journal what opens up — be specific about which symptom rows just improved. Then run the demote-don't-delete experiment on one of the three: half the level, darker EQ, narrower placement. Does the section still improve with the part present but demoted? Decide each part's fate — keep, demote, or mute — and write one sentence connecting the result to Chapter 20's threshold concept. The cheapest fixes in this entire chapter live behind that mute button.

C7. The log as equipment. (20 min) Build a reusable fix-log template with these columns: symptom / address / check run / convicted offender(s) / fix and its stage / verdict / system verified on. Backfill it from C1's pass. Then put it somewhere permanent — this document is Chapter 31's admission ticket, and three mixes from now, the stack of logs becomes the most personally relevant mixing textbook you own: a record of how your mixes characteristically fail, and what actually fixed them.

Part D: Synthesis & Critical Thinking

D1. Your signature diseases. (30 min) Across B1's autopsies and C1's pass, which symptoms recur in your mixes? Hypothesize the cause in your monitoring, your room, or your habits: bass-hyped headphones teaching you to starve lows? A 60 Hz null teaching you to overfeed them? Late-night sessions over-brightening by Fletcher–Munson logic? Arrangement maximalism feeding the mud committee? Write a prevention plan that names what changes at which stage of your next production — before mixing starts. Treating is this chapter; preventing is the career.

D2. The translation contract. (25 min) Write the half-page "definition of done" for your mixes: which tests, which systems, and what must survive — hierarchy, intelligibility, feel — stated specifically for your genre. (A trap mix's contract names the phone test and the harmonic ladder; a folk mix's names the quiet test and the vocal.) Then defend one line of it to a skeptic who says "it sounds great in my room — that's enough." Your answer should contain the chapter's sentence about who will never hear your mix.

D3. The car test on trial. (25 min) A producer friend argues that car tests are obsolete: streaming normalization levels everything, and most listeners are on earbuds anyway. Write the honest rebuttal — what the car uniquely measures (low-end truth at real volume in a real acoustic, fatigue over long listens, a consumer-tuned system's flattery, your margins above road noise) — and then the honest concession: which of the car's jobs phones and earbuds genuinely do cover. End by designing your own testing panel and justifying every member. There are no rules, only consequences; name the consequences.

D4. Diagnosis is ear training. (20 min) Explain the chapter's quiet thesis using Chapter 4's threshold concept: what does installing a symptom vocabulary permanently do to perception? Predict — concretely — what you will now hear in other people's mixes, in commercial releases, in the gas-station playlist, that you couldn't hear fourteen chapters ago. You will not be able to un-hear it. Discuss, in a few honest sentences, whether that's a gift, a tax, or both.

Part M: Mixed Practice

Interleaving earlier chapters — spaced retrieval is how this becomes permanent equipment instead of a chapter you once read.

M1. (Chapter 4 × 30) The quiet test's physics. Using equal-loudness logic: why does conversation-level listening strip away bass-and-treble flattery and expose the balance skeleton? Name which symptom rows this single test catches best, and explain why exactly those rows — what do a buried vocal and a flat energy curve have in common that mud doesn't share?

M2. (Chapter 10 × 30) Clearing the first suspect. Your mix booms at your chair. Design the two-part verification that convicts or acquits your room before you touch the mix, state what you do under each verdict, and explain why a null at your chair is worse news than a peak — including what a year of mixing against a null quietly does to every bounce you export.

M3. (Chapter 21 × 30) The headroom audit after surgery. Your fix pass walked two high-pass filters back down and added a body boost to the pad. What do you re-check on the 2-bus before Chapter 31's print, what peak-level habit are you protecting, and why does the print spec care? Connect: levels are relationships — what relationship did your fixes just renegotiate?

M4. (Chapters 16 + 20 × 30) The disguise, unmasked. Three of the eight symptoms route their first fix to the arrangement rung. Name them, name the shared instrument that catches all three, and explain how Chapter 20's threshold concept predicted this chapter's "confession rate" — why does the cheapest rung keep winning the top slot on honest fix lists?

M5. (Chapter 26 × 30) The phone test's physics. Why does a saturated 808 survive a phone speaker that a pure sine dies on? Name the mechanism (and the perceptual trick your brain performs with a harmonic ladder), name the stage that fix lives at on the cost ladder, and name the tradeoff ledger entry that Chapter 22's mud committee will eventually send you a bill for.