Chapter 15 Quiz
Closed book on the first pass — that's the honest score, and editing doctrine only protects you when it's loaded in your head, not your bookmarks. Multiple choice, then true/false with justification (the justification carries the points), then short answer, then one applied scenario that looks a lot like a Tuesday. Answers hide under each Verify fold; scoring guide at the end.
Section 1 — Multiple Choice (2 points each)
1. The two-stage comping workflow is: rate takes fast, then comp from the survivors. The triage stage exists primarily to:
A) Identify the single best take so the others can be deleted B) Shrink the field to a few takes using cheap, fast judgments — before fatigue degrades your hearing and before phrase-level work multiplies across material that never deserved it C) Find and fix the pitch problems before assembly begins D) Give the singer feedback on which evenings went well
Verify
**B.** Triage spends your fastest, truest judgment (the first reaction) on a cheap question — alive or dead? — so the expensive phrase-by-phrase work happens only on the three to five takes that earned it. Deleting is forbidden regardless (archive, never delete), and fixing comes after choosing: rate the performance, not the problems.2. The chapter insists the unit of comping is the phrase. The underlying reason:
A) Phrases are easier to see on the waveform display B) DAW take lanes can only select whole phrases C) A phrase is the smallest unit of a singer's intention — breath, energy, vowel color, timing, and pitch move as one connected gesture inside it D) Shorter units would create too many files on disk
Verify
**C.** Within a phrase, everything is connected: the launching breath sets its energy, the vowels share color, pitch and timing move as one gesture. Cut between phrases and you splice at a natural reset point; cut within one and you're splicing two different intentions mid-thought — survivable occasionally, never the default.3. You're splicing take 3's ending phrase to take 7's starting phrase, with an inhale between them. The correct edit:
A) Cut after the breath, keeping take 3's inhale leading into take 7's phrase B) Cut in the middle of the breath and crossfade the two inhales together C) Cut before the breath, so take 7's inhale survives, attached to the phrase it launches D) Delete the breath entirely — it's cleaner
Verify
**C.** A breath is a run-up: it rises into the downbeat of the phrase it *launches*, and its energy matches that phrase. Cut before the breath and take 7's inhale launches take 7's line, which is the truth. Glue take 3's inhale onto take 7's phrase and you've built a tiny lie the ear flags without naming. Deleting it outright trades a seam problem for an airless one.4. When a one-word rescue forces you to cut inside a phrase, the best splice point is:
A) The middle of the longest vowel, where there's plenty of material B) An unvoiced consonant — t, k, s, sh, ch, p, f — a pitchless noise-burst that masks the edit C) The exact midpoint between two words D) Wherever the waveforms happen to look similar
Verify
**B.** Unvoiced consonants are broadband noise with no pitch and no waveform continuity to violate; they sound nearly identical take to take and briefly mask everything around them — the edit hides inside the consonant like a cough in applause. Mid-vowel splices ask two different pitch gestures, vibratos, and vowel colors to fuse mid-flight, and they almost never survive scrutiny.5. Your finished comp sheet shows a different take chosen for nearly every consecutive phrase. The chapter's diagnosis:
A) Excellent — you've harvested the best of everything B) A likely Frankenstein: no connected energy arc, a singer who seems to teleport between moods; re-comp for longer runs, switching at section boundaries C) Fine, as long as every seam gets a crossfade D) A sign you need more takes
Verify
**B.** "A comp sheet that looks like a checkerboard is usually a comp that sounds like a séance." You're assembling the best *performance* — a sequence with an arc — not the thirty best phrases. Crossfades hide seams; they can't manufacture continuity between intentions that were never going the same place.6. A butt splice (no crossfade) produces an audible click because:
A) The DAW inserts a marker tone at unfaded edits B) The two clips have different sample rates C) The waveform value jumps instantaneously at the cut, asking the speaker cone to teleport — reproduced as a click or tick D) Unfaded edits always land off the grid
Verify
**C.** Audio is a continuous pressure wave; cut two clips together where their waveforms don't meet at the same value and playback *steps*. The speaker reproduces that discontinuity as a click — sometimes soft in the edit session, and reliably louder after downstream compression raises everything quiet.7. Crossfade length, by location: which pairing matches the chapter's habits?
A) Gaps/breaths ~10–30 ms; inside consonants ~3–10 ms; sustained material 50 ms and up B) Always exactly 10 ms — consistency is what matters C) Inside consonants 100 ms+; gaps as short as possible D) Length doesn't matter as long as the curve is equal-power
Verify
**A.** Comfortable room in gaps and breaths; short enough to live entirely inside the noise-burst on consonant cuts; long blends on sustained material. Numbers are habits, not laws — the law is the listen, soloed and in the track.8. During mixing, mysterious ticks keep surfacing on tracks that sounded clean during editing. The most likely culprit:
A) The converters are failing B) Unfaded edits from the editing pass, revealed as compression raises quiet material C) Phase issues between the overheads D) The session's sample rate changed
Verify
**B.** The cruelest property of the butt splice: barely audible in the edit session, perfectly audible three chapters later, after compression lifts everything quiet. Every engineer learns it the expensive way once — the fades-everywhere doctrine exists so it's only once.9. "In time," for a vocal or instrument edit pass, means in agreement with:
A) The DAW's editing grid, which is mathematically exact B) The track's actual rhythmic center — usually the drums — judged by ear in context C) The metronome click used at tracking D) The nearest sixteenth-note division
Verify
**B.** The grid is a measuring tool and a last resort, not a destination. When a track has a pocket (Static Bloom's drums sit deliberately behind the grid), grid-snapped parts are mathematically correct and musically rushing. Tighten to the groove, not the grid — snapping off for timing work.10. The timing doctrine says "move phrases before syllables" because:
A) Phrases are faster to drag than syllables B) Sliding a whole phrase preserves its internal micro-timing — the performance's anatomy — while per-syllable surgery multiplies seams and destroys the relationships that made it sound human C) Syllable edits are destructive in most DAWs D) Phrases always rush or drag uniformly, so nothing else is ever needed
Verify
**B.** A rushing phrase usually rushes *as a unit* — the singer came in early and the whole gesture came along. Slide the clip and the consonant placements and word-to-word timing survive intact. Carve it into thirty aligned slivers and you get thirty crossfades and a vocal with the rhythmic personality of a ransom note. (D is tempting but overclaims — occasionally internal correction is needed; it's the last resort, not a never.)11. Time-stretching (elastic audio) punishes which material worst?
A) Solo vocals, because pitch is so exposed B) Drums and transient-heavy material — stretch algorithms smear the attack that carries each hit's identity C) Sustained synth pads D) Spoken word
Verify
**B.** Stretching must invent or discard audio, and transients suffer first — snap smears into a soft "fff" onset, fatal on a snare. (Polyphonic material gets seasick fastest after that.) Hence the drum doctrine: cut-and-slide at the transient, heal gaps with crossfades in the decay; modest stretches on a solo voice, by contrast, can be genuinely invisible.12. In transparent pitch correction, retune speed is "the entire game" because it determines:
A) How many notes get corrected per pass B) Whether the human pitch gestures — vibrato, scoops, slides — survive while the note's center is pulled true (slow), or get flattened into robotic snapping (fast) C) The plugin's latency D) Which scale the correction targets
Verify
**B.** Slow retune lets the note's center drift toward true while vibrato and entry scoops live; fast retune kills the gestures and births the robot. Strength decides how *far* toward perfect; speed decides whether anyone can tell a hand was there. Documentary or costume — speed is the switch.13. The chapter's position on hard-tune (zero-ish retune speed, full strength, key-locked):
A) A confession of weak singing, acceptable only when re-recording is impossible B) A legitimate aesthetic — pitch quantization as an instrument — that actually rewards strong, athletic note-to-note singing, and whose only craft sin is being used accidentally C) Obsolete since the early 2010s D) Acceptable in hip-hop but dishonest in pop
Verify
**B.** Hard-tune quantizes the pitch *between* notes, so a deliberate, athletic performance plays the stairsteps like an instrument (T-Pain is the canonical proof), while a timid one is still timid, now in stairsteps. The failure mode isn't the robot — it's transparent settings pushed until they flicker, an effect wearing a documentary's clothes. Inaudible or intentional.14. What distinguishes note-level editing (the Melodyne lineage) from real-time correction?
A) It only works on drums B) It's faster but less precise C) It analyzes the recording and displays detected notes as draggable objects on a pitch-time canvas — so you can recenter one note while leaving its vibrato and drift intact, touching nothing else D) It corrects pitch and timing simultaneously and automatically
Verify
**C.** Instead of riding pitch live, it analyzes first — like a piano roll grown from audio. That makes it the most transparent-capable tool (correct exactly the three notes that hurt, by exactly enough) and the easiest place to over-polish one innocent drag at a time. The protecting discipline: keep an untouched copy one click away and ask of every move, "accident fixed, or intention sanded?"15. The group-edit rule for multi-mic sources ("what was captured together gets edited together") exists because nudging one mic of a kit independently:
A) Confuses the DAW's take-lane system B) Changes arrival-time relationships within a single sound, inviting comb filtering — a hollowed snare, a kit whose bottom drops out, worst when summed to mono C) Makes the session file larger D) Violates the crossfade doctrine
Verify
**B.** A multi-mic'd source is one sound at several capture points; the time relationships between them are the difference between full and hollow ([Chapter 12](../chapter-12-recording-instruments/index.md)'s lesson). Slide one mic alone and the close mic fights its own image in the overheads — almost fine in stereo solo, collapsing on a phone speaker. Edit groups cost one checkbox; the damage is nearly undiagnosable later.Section 2 — True/False, with Justification (2 points each: 1 for the call, 1 for the why)
16. Every clip boundary in the session — every track, every edit, every bounce, every chop — should carry a fade or crossfade.
Verify
**True.** No exceptions, no "it's probably at a zero crossing." Fades are free, instant, and non-destructive; clicks are forever and prefer to reveal themselves on release day. The doctrine covers the millisecond scale (hygiene); musical fade-ins and fade-outs are arrangement decisions for later chapters.17. Strip-silence should be run aggressively — the more gap-junk removed automatically, the better.
Verify
**False.** Set it conservatively and check its work: an aggressive pass beheads soft phrase entrances and amputates reverb tails and decaying word-endings. And gutting gaps to digital black backfires on exposed, intimate material — silence isn't silent; a sparse vocal with vacuum-packed gaps sounds weirdly dead between lines.18. The professional standard is to delete all breaths from a lead vocal.
Verify
**False.** Reduce, don't delete — breaths are phrasing, and a vocal stripped of them sounds airless and android. The working habit is pulling them down 6–10 dB so they read as punctuation, deleting only gasps that serve nothing, and keeping the run-up breath before a big phrase especially. (Genres that *do* strip breaths — hard-tuned rap, slick pop — do it as a deliberate plastic aesthetic: chosen, not defaulted.)19. Classical music, the genre with the strictest performance-truth culture, is also one of the most edited genres on record.
Verify
**True.** Modern classical releases are commonly assembled from multiple complete takes plus patch sessions, spliced at bar lines — a practice as old as tape and openly discussed in the industry literature. Glenn Gould argued in print that splicing was creative liberation. The cultures differ in what they're willing to be *heard* editing, not in whether they edit.20. A vocalist's timing deviation that repeats consistently across multiple takes should be corrected in the edit pass.
Verify
**False.** Consistent isn't an accident — it's a choice the performer's body made about the feel (Demi drags every pre-chorus; the drag *is* the lift). The diagnostic: repeats-and-serves-the-feel = intention, keep; one-take stumble sticking out from the performance's own pattern = accident, fix. The pass removes outliers from the pattern, not the pattern.Section 3 — Short Answer (4 points each)
21. Demi sang every fragment of her comp, and she still can't find the seams. Explain why, using the Chapter 4 threshold concept and at least one other capture or editing factor doing the hiding.
Verify
You hear with your brain, not your ears — the brain is a prediction engine that tracks phrase-level contour and actively *constructs* continuity unless the evidence contradicts it. Seams placed at gaps and breaths present each phrase as a complete, internally consistent gesture (the shape phrases always have), so the continuity hypothesis is never falsified. Supporting factors (any one earns the point): consonant cuts bury the splice inside broadband noise (masking); matched capture conditions — same mic distance, same room, same position — keep timbre continuous so nothing flags the boundary; crossfades remove the physical discontinuity that would otherwise click. The seam isn't inaudible because it isn't there; it's inaudible because the listener's brain paints over it.22. State the "inaudible or intentional" rule and apply it to three tools from this chapter: elastic audio, pitch correction, and an audible stutter edit in an electronic track.
Verify
The rule: processing must either be undetectable (the documentary) or committed enough to read as a deliberate style (the costume) — the half-audible middle fails as both craft and candor. Elastic audio: stretch until you can hear the warble and you've gone too far — back off, solve differently, or push all the way and own it as texture. Pitch correction: transparent settings backed off at the first flicker, or full hard-tune committed — never transparent settings pushed until they flicker on loud notes. Stutter edit: at full volume, rhythmically placed and repeated, it reads as style; timid and unowned, it reads as a mistake. Same law, three tools.23. Name the three damage-control habits and the shared principle behind them — and explain why the chapter frames that principle as the funding for boldness rather than insurance against it.
Verify
(1) Save-as version the session before the pass and at milestones; (2) archive losing takes — muted, recallable — never delete; (3) commit/render on copies, keeping pre-commit originals muted underneath. Shared principle: total reversibility — every decision can be unwound weeks later when you learn it was wrong. The framing: editors with no undo path edit timidly, and timid editing serves no genre; when nothing is at stake but minutes, you can comp aggressively, try the weird edit, hard-tune the folk song to hear it. Safety is what daring costs.24. Why did polyphonic note-level pitch editing arrive roughly a decade after monophonic real-time correction? Answer in terms of what autocorrelation does and what a chord breaks.
Verify
Autocorrelation answers "what is THE period of this signal?" — slide a copy of the signal past itself, find the lag where it best matches, and the repeating cycle's length falls out; fast and cheap enough to run live by 1997. A chord contains several interleaved periodicities whose harmonics overlap and even share frequencies, so there is no single best self-match — the honest answer is three answers. Solving polyphony meant abandoning find-the-period for decomposing the spectrum into families of related harmonics, inferring which partials belong to which note, allocating shared ones, and reconstructing each note as a separately movable object — an inference problem orders of magnitude harder, which is why it needed another decade of cleverness.Section 4 — Applied Scenario (8 points)
25. A friend texts you in a panic. Their band's single gets mixed tomorrow by an outside engineer, and tonight's final listen-through revealed problems. The session, as described: the lead vocal comp switches takes on almost every line and "feels like the singer keeps changing moods"; there are two audible ticks in the second verse and none of the session's edits have fades; all breaths were deleted from the first chorus only, "to clean it up," and that chorus now feels weirdly dead; the vocal tuning is on a transparent preset but flickers robotically on the loud notes; the acoustic guitar was snapped 100% to the grid even though the drummer plays consistently behind the beat, and now it feels like it's rushing; the snare close mic was nudged "tighter" by a few milliseconds without the overheads, and the kit sounds thin on a phone speaker; and to "keep things tidy," every losing vocal take was deleted from the session and the trash emptied. Your friend has three hours tonight. Write your triage: the order you'd work, what each fix involves, which problem is the most damaging per minute of work, and which one may be unrecoverable — plus the one-line prevention rule for each.
Verify
Strong answers triage by damage-per-minute and name the doctrines. A defensible order: **(1) Fades everywhere (~20 min, massive return):** apply default crossfades across all edits and outer-edge fades on every clip; hunt the two known ticks at sample zoom. Ticks get *louder* after the mix engineer's compression — cheapest fix, biggest disaster averted. Rule: every edit gets a fade, no exceptions. **(2) Un-nudge the snare mic (~10 min):** restore the close mic's original position relative to the overheads (or slide it back by the same amount) and lock an edit group — what was captured together gets edited together. The thin phone-speaker kit is comb filtering; nearly undiagnosable at mix time. **(3) Guitar timing (~30 min):** unsnap from the grid; slide phrases by ear against the drummer's actual pocket — tighten to the groove, not the grid. **(4) Tuning flicker (~20 min):** the forbidden middle zone; either back speed/strength off until inaudible (likely right for this band) or commit to the effect — inaudible or intentional, nothing between. **(5) Chorus breaths (~20–30 min):** breaths can't be un-deleted if the takes are gone, BUT the comp's source lanes may retain them under the comp, or breaths can be transplanted from elsewhere in the kept takes (a breath is a sound object; it moves whole). Reduce 6–10 dB, don't delete; that's the rule they broke. **(6) The checkerboard comp (the big one, time permitting):** re-comp for runs, switching at section boundaries — comp for continuity, not a highlight reel. Here's where the deleted takes bite: with only the comp's chosen segments surviving, re-comping options are crippled — **the emptied trash is the potentially unrecoverable wound** (check backups, the recording computer, cloud sync, the DAW's auto-saved versions before declaring it dead). Rule: archive, never delete — disk space is the cheapest thing in the studio. Full credit requires: fades first or second (cheap, catastrophic-if-skipped), the mono/phase fix identified as comb filtering from breaking group-edit discipline, the tuning identified as the middle zone, the timing fix referencing groove-not-grid, recognition that deleted takes (not the comp's mood swings) are the unrecoverable part, and a tone that diagnoses rather than scolds.Scoring
| Section | Items | Points |
|---|---|---|
| 1 — Multiple choice | 15 × 2 | 30 |
| 2 — True/False + justification | 5 × 2 | 10 |
| 3 — Short answer | 4 × 4 | 16 |
| 4 — Applied scenario | 1 × 8 | 8 |
| Total | 64 |
| Score | Reading |
|---|---|
| 58–64 | Editor. Your seams will be invisible and your sessions will be boring stories — the good kind. On to Part IV. |
| 48–57 | Solid. Re-run whichever doctrine cost you points; the fades-everywhere and groove-not-grid habits are the two that compound hardest downstream. |
| 38–47 | The concepts are landing but they're not reflexes yet. Do exercises C1 and C2 (the comp drill and the crossfade drills) before moving on — this chapter lives in the hands. |
| Below 38 | Reread with a session open and edit along — editing theory without a waveform in front of you is swimming lessons by mail. The four-take comp drill is the best second pass through everything. |