Chapter 23 Quiz

Q: A compressor in one sentence: A) A device that makes everything louder B) An automatic fader that turns the signal down when it crosses a line, by rules you set C) An EQ for dynamics D) A device that removes transients

B. Everything else in the chapter elaborates that sentence: the five controls answer five questions about where the line is, how firmly it pushes, how fast it grabs and lets go, and how you restore level to compare honestly.

Q: Threshold at -20 dBFS, ratio 4:1, hard knee. A peak arrives at -8 dBFS. The output peak lands at approximately: A) -8 dBFS B) -20 dBFS C) -17 dBFS D) -11 dBFS

C. The input is 12 dB over the threshold; at 4:1 that 12 dB of overshoot becomes 3 dB of overshoot. Output: -20 + 3 = -17 dBFS, which is 9 dB of gain reduction.

Q: Which pair of controls does the chapter call "the actual lesson," and why? A) Threshold and ratio — they set how much gain reduction happens B) Makeup gain and knee — they set the tone C) Attack and release — they decide *when* gain reduction happens, which shapes envelopes and groove D) Input and output — gain staging is everything

C. Threshold and ratio are arithmetic — policy. Attack and release are time — and time is where transients, punch, and groove live. The same threshold, ratio, and GR reading can produce punch or mush depending entirely on the time controls.

Q: A compressor with a slow attack (~30 ms) on a snare makes it punch *harder* because: A) Slow attack adds gain to the transient B) The transient passes untouched while the body behind it gets clamped — more contrast between hit and tail reads as more punch C) Slow attack boosts high frequencies D) It doesn't — compressors can only reduce punch

B. The compressor never makes anything louder; it turns down everything except the hit. The crack now towers over its own body. Punch is contrast, and a slow attack manufactures contrast.

Q: Jaylen's drums went flat with the compressor set to fastest attack because: A) The ratio was too low B) Fastest attack clamps during the transient itself, removing the front edge that carries the "hit" information C) The threshold was too high D) FL Studio's compressor is low quality

B. The vise. Clamping within the first fraction of a millisecond turns down the crack along with the body — the hit loses exactly the part that made it a hit, on every drum, all song.

Q: A release set much too slow on a drum loop causes: A) Distortion on the hi-hats B) The compressor to be still recovering when the next hit arrives, shaving the following hits and flattening the groove C) Louder transients D) Pumping in time with the kick

B. The loud moments steal level from their neighbors. Energy drains out of the spaces between hits, and the loop feels tired without an obvious culprit.

Q: A very fast release on an 808 or bass line risks: A) Phase cancellation B) Audible distortion, because the gain starts riding individual waveform cycles instead of notes C) Nothing — fast release is always safer D) Comb filtering

B. A 50 Hz wave takes about 20 ms per cycle. A release much faster than the wavelength is no longer responding to musical events — it's modulating the gain inside single cycles, which is distortion by definition.

Q: The healthy GR working zone for *glue* on a drum bus is roughly: A) 1–3 dB, swaying with the backbeat B) 6–10 dB, pinned C) 0 dB — glue compressors shouldn't move D) 12 dB on every hit

A. Individually inaudible, collectively cohesive. The needle should sway gently with the groove and come home to zero. Pinned needles are the squash story.

Q: Which GR-needle "story" indicates squash? A) The dance — dips and full recoveries between phrases B) The twitch — occasional flickers on rare peaks C) The pin — the needle never returns to zero D) A needle at zero throughout

C. If the needle never comes home, the music never gets to be itself — nothing breathes. First fix: raise the threshold until recovery appears somewhere musical, then re-judge at matched loudness.

Q: Why does 1–3 dB of shared gain reduction on a drum bus read as "one kit"? A) It boosts the room mics B) Every element now dips and recovers together — shared micro-fate mimics the physics of sounds sharing a room, and the brain files them as one performance C) It removes phase differences between tracks D) It centers everything in the stereo field

B. Glue is shared consequence. When the snare cracks, the whole group ducks a hair and releases together, dozens of times a bar — micro-correlations that are individually inaudible but statistically unmistakable to the brain.

DataField.Dev

Chapter 23 Quiz

Closed book first pass — that's the honest score. Multiple choice, then true/false with justification (the justification is most of the point), then short answer, then one applied scenario: a broken mix on the table and you holding the GR meter. Answers hide under each Verify fold; the scoring guide is at the end.

Section 1 — Multiple Choice (2 points each)

1. A compressor in one sentence:

A) A device that makes everything louder B) An automatic fader that turns the signal down when it crosses a line, by rules you set C) An EQ for dynamics D) A device that removes transients

Verify

**B.** Everything else in the chapter elaborates that sentence: the five controls answer five questions about where the line is, how firmly it pushes, how fast it grabs and lets go, and how you restore level to compare honestly.

2. Threshold at -20 dBFS, ratio 4:1, hard knee. A peak arrives at -8 dBFS. The output peak lands at approximately:

A) -8 dBFS B) -20 dBFS C) -17 dBFS D) -11 dBFS

Verify

**C.** The input is 12 dB over the threshold; at 4:1 that 12 dB of overshoot becomes 3 dB of overshoot. Output: -20 + 3 = -17 dBFS, which is 9 dB of gain reduction.

3. Which pair of controls does the chapter call "the actual lesson," and why?

A) Threshold and ratio — they set how much gain reduction happens B) Makeup gain and knee — they set the tone C) Attack and release — they decide when gain reduction happens, which shapes envelopes and groove D) Input and output — gain staging is everything

Verify

**C.** Threshold and ratio are arithmetic — policy. Attack and release are time — and time is where transients, punch, and groove live. The same threshold, ratio, and GR reading can produce punch or mush depending entirely on the time controls.

4. A compressor with a slow attack (~30 ms) on a snare makes it punch harder because:

A) Slow attack adds gain to the transient B) The transient passes untouched while the body behind it gets clamped — more contrast between hit and tail reads as more punch C) Slow attack boosts high frequencies D) It doesn't — compressors can only reduce punch

Verify

**B.** The compressor never makes anything louder; it turns down everything *except* the hit. The crack now towers over its own body. Punch is contrast, and a slow attack manufactures contrast.

5. Jaylen's drums went flat with the compressor set to fastest attack because:

A) The ratio was too low B) Fastest attack clamps during the transient itself, removing the front edge that carries the "hit" information C) The threshold was too high D) FL Studio's compressor is low quality

Verify

**B.** The vise. Clamping within the first fraction of a millisecond turns down the crack along with the body — the hit loses exactly the part that made it a hit, on every drum, all song.

6. A release set much too slow on a drum loop causes:

A) Distortion on the hi-hats B) The compressor to be still recovering when the next hit arrives, shaving the following hits and flattening the groove C) Louder transients D) Pumping in time with the kick

Verify

**B.** The loud moments steal level from their neighbors. Energy drains out of the spaces between hits, and the loop feels tired without an obvious culprit.

7. A very fast release on an 808 or bass line risks:

A) Phase cancellation B) Audible distortion, because the gain starts riding individual waveform cycles instead of notes C) Nothing — fast release is always safer D) Comb filtering

Verify

**B.** A 50 Hz wave takes about 20 ms per cycle. A release much faster than the wavelength is no longer responding to musical events — it's modulating the gain inside single cycles, which is distortion by definition.

8. The healthy GR working zone for glue on a drum bus is roughly:

A) 1–3 dB, swaying with the backbeat B) 6–10 dB, pinned C) 0 dB — glue compressors shouldn't move D) 12 dB on every hit

Verify

**A.** Individually inaudible, collectively cohesive. The needle should sway gently with the groove and come home to zero. Pinned needles are the squash story.

9. Which GR-needle "story" indicates squash?

A) The dance — dips and full recoveries between phrases B) The twitch — occasional flickers on rare peaks C) The pin — the needle never returns to zero D) A needle at zero throughout

Verify

**C.** If the needle never comes home, the music never gets to be itself — nothing breathes. First fix: raise the threshold until recovery appears somewhere musical, then re-judge at matched loudness.

10. The purposes taxonomy says a compressor should be hired for exactly one of five jobs. They are:

A) EQ, balance, width, depth, loudness B) Leveling, punch, glue, density, FX C) Attack, decay, sustain, release, hold D) Vocal, drums, bass, bus, master

Verify

**B.** Leveling (evenness over time), punch (transient contrast), glue (bus cohesion), density (thickness/sustain — usually parallel, [Chapter 28](../../part-06-advanced-mixing/chapter-28-advanced-mix-techniques/index.md)), FX (the compressor as a sound). One compressor, one named job; two jobs means two compressors.

11. Why does 1–3 dB of shared gain reduction on a drum bus read as "one kit"?

A) It boosts the room mics B) Every element now dips and recovers together — shared micro-fate mimics the physics of sounds sharing a room, and the brain files them as one performance C) It removes phase differences between tracks D) It centers everything in the stereo field

Verify

**B.** Glue is shared consequence. When the snare cracks, the whole group ducks a hair and releases together, dozens of times a bar — micro-correlations that are individually inaudible but statistically unmistakable to the brain.

12. Pumping is:

A) Always a mistake B) Audible gain-reduction recovery — a rhythmic surge after loud events — which is a failure when unchosen and an aesthetic when chosen (French house built a genre on it) C) A hardware-only artifact D) Another name for clipping

Verify

**B.** The mechanism is identical either way; the difference is the choosing. The deliberate version — driving the compressor from a different signal — is sidechain territory, Chapter 28.

13. The bounce test sets which control, and how?

A) Threshold — lower it until the meter pins B) Release — exaggerate GR temporarily, sweep release until the needle bounces in time with the groove and gets home before each hit, then restore the threshold C) Attack — fastest position, then back off D) Makeup — match it to bypass

Verify

**B.** By ear, with the meter as a witness: down on the hit, home just in time for the next one. Then raise the threshold back to the job's working zone — the release stays right.

14. The optical compressor became the historical default for vocal leveling because:

A) It was the cheapest design B) Its photocell physics produce a slow-ish attack and a two-stage, program-dependent release that happens to match the shape of human phrasing C) It has the fastest attack available D) It adds the most distortion

Verify

**B.** Nobody designed that release curve — the light cell just does it, and the behavior flatters voices. Case study 1 tells the full story; the behavior, not the brand, is the lesson.

15. A limiter is:

A) A compressor with an effectively infinite ratio and near-instant attack — a ceiling nothing crosses B) A gentler compressor for mastering C) An EQ with a steep slope D) A noise gate in reverse

Verify

**A.** ∞:1 above the threshold: the output does not rise. Its legitimate homes are peak safety and mastering's loudness stage ([Chapter 32](../../part-07-mastering/chapter-32-mastering-tools-techniques/index.md)) — not your 2-bus while you're mixing in Part V.

Section 2 — True/False with Justification (2 points each: 1 for the call, 1 for the why)

16. A compressor only ever turns signal down, so it cannot make a sound more punchy.

Verify

**False.** The "only turns down" half is true; the conclusion doesn't follow. With a slow attack, the compressor turns down everything *except* the transient — increasing the contrast between hit and body. Punch is contrast, so punch increases even though no sample ever got louder (until makeup gain, which lifts everything equally).

17. Two compressors showing identical 4 dB GR readings, with identical thresholds and ratios, will sound essentially the same.

Verify

**False.** The meter shows *how much*; attack and release decide *which part of the envelope* pays the 4 dB and what happens between events. One setting yields punch, another mush, another pumping — same depth, different time. This is the threshold concept in meter form.

18. With makeup gain applied, a leveling compressor effectively makes the quiet phrases louder even though it never touches them.

Verify

**True.** The compressor catches only the loud phrases, shrinking the loud/quiet gap; makeup gain then raises the whole track. Net: loud parts roughly restored, quiet parts lifted. That's why Demi's whisper got closer — the rail works on both ends.

19. Pumping on a record always indicates an engineering error.

Verify

**False.** Unchosen, it's a failure mode (too much GR with audible release). Chosen, it's a legitimate aesthetic — French house made the surge the groove's engine, and modern EDM features it deliberately. Pumping is only a mistake when nobody decided it should be there.

20. Because major streaming platforms normalize playback loudness, crushing a mix's dynamics to "win" on loudness gains essentially nothing there — the platform turns it down to match everything else.

Verify

**True** (as of this writing — platforms change their numbers, but normalization itself is the established norm). The slammed master and the dynamic master play back at comparable loudness; the slammed one just arrives with its punch already spent. [Chapter 33](../../part-07-mastering/chapter-33-loudness-wars-streaming/index.md) signs the full treaty.

Section 3 — Short Answer (4 points each)

21. Name the five compression jobs and give the GR working zone for each.

Verify

Leveling: 2–4 dB on the loud phrases, nothing on the quiet ones. Punch: 3–6 dB on the hits, full recovery between. Glue: 1–3 dB on a bus, swaying with the groove. Density: 6–10+ dB, normally on a parallel copy blended under the dry ([Chapter 28](../../part-06-advanced-mixing/chapter-28-advanced-mix-techniques/index.md)). FX: whatever the art calls for — the meter becomes art direction.

22. Explain why hardware compressors can never truly act before a transient, and how digital lookahead gets around it — including the cost.

Verify

Hardware lives in causality: the detector can't respond to a signal that hasn't happened, so even "instant" attack reacts during the event, and clamping fast enough to catch the rest means bending waveforms (distortion, especially on lows). Lookahead delays the audio path a few milliseconds while the detector listens to the undelayed copy — the clamp can begin before the transient reaches the gain stage. The cost is latency: free-ish in a mix thanks to plugin delay compensation, hostile while tracking because performers feel the delay.

23. Your friend's vocal still lurches after compression, so they keep lowering the threshold. State the hierarchy they're violating and the correct order of operations.

Verify

[Chapter 21](../chapter-21-gain-staging/index.md)'s clip-gain-first hierarchy. Rogue words and wildly hot phrases are *level* problems: fix them with clip gain on the regions (and the fader for overall placement) so the compressor only has weather to handle, not lightning. Then a gentle leveler (2:1–4:1, 2–4 dB on loud phrases) finishes the job. Asking a compressor to fix clip-gain problems forces deep GR, which buys squeeze without buying evenness.

24. Why does the chapter insist you name ONE job per compressor instance, and what's the prescribed move when a source genuinely needs two jobs?

Verify

Because each job wants different settings — punch wants slow attack and recovery between hits; density wants fast attack and deep GR — and one instance splitting the difference does both jobs badly. The move: two stages, each doing its one job gently (serial — two 2–3 dB stages beat one 6 dB stage, the principle [Chapter 29](../../part-06-advanced-mixing/chapter-29-mixing-vocals/index.md)'s vocal chain is built on), or move the second job to a parallel path ([Chapter 28](../../part-06-advanced-mixing/chapter-28-advanced-mix-techniques/index.md)).

Section 4 — Applied Scenario (10 points)

25. A bandmate sends you their "compressed but worse" rock mix with these notes, and you can hear every problem in the bounce: (a) the lead vocal still disappears in verses and barks in choruses, even though their vocal compressor shows 8 dB of GR; (b) the snare, compressed at fastest attack, sounds weaker than bypassed; (c) the drum-bus compressor's needle sits at -7 dB and never moves; (d) the bass has a fuzz nobody recorded; (e) everything was judged with makeup gain cranked +6 over bypass, and they insist it "obviously sounds better compressed." Diagnose all five: name each problem, the responsible control or habit, and the fix — and state which order you'd work in and why.

Verify

Order: fix the judging first, then work the mix hierarchy (vocal, then drums, then bass — [Chapter 20](../chapter-20-what-is-mixing/index.md)). **(e) first — the honest A/B.** +6 dB unmatched makeup means every decision was made through the louder-sounds-better bias. Match bypass and engaged levels on every instance before re-judging anything; expect several "improvements" to evaporate. **(a) Vocal:** 8 dB GR that still doesn't level = a clip-gain problem wearing a compression costume (and likely wrong time constants). Raise the threshold, clip-gain the lightning-strike words and quiet-verse regions by hand, then re-set the leveler for 2–4 dB on loud phrases only (3:1, 10–30 ms attack, auto release). **(b) Snare:** fastest attack is clamping the crack — the vise. Slow the attack to 15–30 ms until the crack passes, release timed to recover before the next backbeat, 3–6 dB on hits. **(c) Drum bus:** the pin — squash, not glue. Raise the threshold until GR sways at 1–3 dB with the groove and returns to zero; 2:1, slow attack, auto release; makeup matched. **(d) Bass fuzz:** release faster than the bass wavelengths — the gain is riding cycles. Slow the release until the fuzz disappears; keep 3–5 dB at 4:1 with a medium-fast attack so the pluck survives. Full credit requires the matched-loudness point leading, correct control named per symptom, and the hierarchy order with a reason (the vocal is the mix's spine; and re-judging honestly may delete some compressors entirely — deletions are decisions too).

Scoring

Section	Points available
Multiple choice (1–15)	30
True/False + justification (16–20)	10
Short answer (21–24)	16
Applied scenario (25)	10
Total	66

56–66: You can hear the difference between glue and squash and you know which knob made it. Go do the Project Checkpoint with confidence. 44–55: Solid on the how-much controls; re-read "Attack and Release: The Actual Lesson" and redo exercise C1 — the time knobs aren't yours yet. Below 44: No shame — this is most readers' hardest tool. Replay the Productive Struggle sweep with the chapter open, then take this again in two days. The compressor rewards exactly this kind of stubbornness.