Chapter 13 Exercises

Work these roughly in order: Part A loads the vocabulary, Part B retrains your ears on records that already solved these problems, and Part C — the chapter's real payoff — turns both into muscle memory in your DAW. Exercise C1 assumes you kept the Productive Struggle loop from the chapter; if you deleted it, rebuild it in five minutes before you start (96 BPM, eight bars, quantize 100%, flat velocities — making it bad is easy). Time estimates are honest averages. Selected exercises (marked ★) have worked answers in the back of the book (Answers to Selected Exercises). The Listening Lab deliverables go in your listening journal from Chapter 4 — this chapter adds a permanent new column to it: where does the feel live?

Part A — Conceptual Understanding

A1. The four chairs. ★ (10 min) From memory, name the four roles of a groove, the typical sounds that fill each, and the one-line symptom you feel when each role is failing. Then state the hierarchy rule that governs how much timing freedom each role gets, and explain in two sentences why the rule runs in that direction and not the reverse.

A2. The fortune cookie, upgraded. (15 min) "Humans are imperfect, so imperfection sounds human" is the fortune-cookie version of this chapter. Write the real version: explain the three actual mechanisms by which a perfectly quantized, flat-velocity beat loses a listener — the prediction mechanism, the meaning mechanism, and the skeleton mechanism — in your own words, one short paragraph each. Include Charles Keil's term for the productive friction between players, and what it implies about a grid.

A3. Swing math, then swing law. ★ (15 min) A tutorial filmed at 140 BPM swears by 58% swing. Your session runs at 84 BPM. (a) Compute the off-sixteenth's delay in milliseconds at both tempos (work from the chapter's anchor: at 96 BPM an eighth-note pair lasts 312.5 ms; scale from there). (b) Explain why the same percentage feels heavier at the slower tempo. (c) State the by-ear law for setting swing, word for word if you can, and explain what the number is for once you've stopped sweeping the dial.

A4. Two ways to fail. (10 min) Describe the flat grid and the humanize lottery as failure modes: what does each one get wrong about how a real performance deviates? Then define the thing that lives between them — structured deviation — using three adjectives the chapter attaches to it, and give one concrete example of each adjective in a drum pattern (e.g., what does "directional" look like on a clap?).

A5. Two senses of velocity layers. (10 min) Your snare library triggers a soft skin-tone thud at velocity 45 and a cracking rimshot at 120; separately, you've programmed your hat line with accents at ~112, body at ~85, and ghosts at ~40. Name the two senses of "velocity layers" in play, say which is the instrument and which is the performance, and explain how the map you write exploits the layers the library recorded. Bonus sentence: what can you do for a single-layer sample so harder hits still get brighter? (Chapter 14 will hand you the tool; the chapter named the trick.)

A6. The fix ladder, defended. ★ (10 min) Your kick and 808 are smearing into one boom. List the four rungs of the fix ladder in order, and write one sentence per rung on what it does. Then defend the ordering: why does arrangement outrank EQ, and why do the mix tools (Chapters 22 and 28) come last even though they're the famous ones?

A7. One-shots vs. loops, from memory. (10 min) Reproduce the tradeoff table: control, feel, cost, editing, best-at — one line each for one-shots and loops. Then describe the top-loop doctrine in two sentences, including the approximate high-pass move that turns a full loop into a top, and the two integration rules (one about swing, one about dirt).

A8. The forum reply. (15 min) A post reads: "Sampled 4 seconds of an old soul record for my new single, pitched it up 3 semitones so it's basically unrecognizable. Under six seconds is fair use anyway, right? Dropping it on streaming Friday." Draft a four-sentence reply in this chapter's spirit: name the two separate copyrights in play, kill the two folklore safe-harbors the post leans on, state the lab-versus-release distinction, and point them to where this book treats clearance properly (Chapter 36). No scolding — diagnose, then fix.

Part B — 🎧 Listening Lab

Your reference playlist from Chapter 4 stays open all lab. Listen at your marked reference level from Chapter 8. The deliverable for every exercise is journal lines, and the meta-skill is always the same: stop asking what sounds are these and start asking where does the feel live.

B1. The swing safari. ★ (30 min) Build a five-stop playlist that walks the swing dial from straight to drunk, and rank the stops by ear before you read the suggested order. Widely available picks: Billie Eilish's "bad guy" (machine-straight pulse, feel from sparseness), Daft Punk's "Around the World" (four-on-the-floor, almost no swing — the grid as hypnosis), Frankie Knuckles' "Your Love" (house lilt living in the hats and bassline, kick straight), A Tribe Called Quest's "Electric Relaxation" (the MPC-era head-nod zone), and MJ Cole's "Sincere" (UK garage — swing as the genre's entire identity). Substitute freely from your own library, but keep five stops spanning the range. For each: where does the swing live (which role), and where is it absent? Deliverable: your ranked list with a one-line "where the feel lives" note per track, plus the one that made your neck move first.

B2. Ghost-note hunting. (25 min) Three classic kits full of ghosts: James Brown's "Funky Drummer" (Clyde Stubblefield), The Meters' "Cissy Strut" (Zigaboo Modeliste), and Aretha Franklin's "Rock Steady" (Bernard Purdie). Loop a verse of each and transcribe just the snare: mark the cracking backbeats as capital X, the ghosts as lowercase dots, bar by bar, for two bars. (Headphones help; ghosts hide under everything by design.) Then find one modern programmed track in your genre with no ghost notes at all and play it immediately after "Cissy Strut." Deliverable: your two-bar transcriptions and one sentence on what the ghostless track feels like by comparison — rolling versus marching is the vocabulary to beat.

B3. Programmed or played? ★ (30 min) Build yourself a six-track blind test (or trade playlists with a friend so it's genuinely blind): mix three programmed-drum tracks and three human-played tracks from your library. Two free pre-made pairs that work: Prince's "When Doves Cry" (programmed — a drum machine played like an instrument) against Led Zeppelin's "When the Levee Breaks" (played, in a stairwell — Chapter 12's old friend); and D'Angelo's "Playa Playa" from Voodoo (played by a human deliberately imitating machine-era drag) against any modern trap single (programmed, imitating nothing). For each track, vote programmed/played, then list the tells you used: velocity variation, fill repetition, physically impossible parts, timing texture, cymbal behavior. Deliverable: your votes, your tells, and — most useful — the track that fooled you and why. Expect to be fooled at least once; the border is the lesson.

B4. 808-tuning recognition. (20 min) Cue three tracks built on tuned, sliding 808s — Future's "Mask Off," Kendrick Lamar's "HUMBLE.," and Travis Scott's "SICKO MODE" all work, or substitute from your library. For each: hum the 808 line until you can sing its root movement, then find one slide and describe it (up or down, fast or slow, landing where). Now the contrast test: dig through free beats on any upload platform until you find one where the 808 fights the song's key — untuned low end is common in amateur uploads, and hearing the clash once labels it forever. Deliverable: three sung-and-described basslines, one documented clash, and a one-line answer to: how could you tell the clash was tuning and not level?

B5. The Amen lineage. (20 min + one passive week) Play The Winstons' "Amen, Brother" (1969) and find the drum break (it arrives just past the middle — you'll know it). Listen three times: once for the whole kit, once for just the snare placement, once for the ride. Then hear what forty years of producers built from those bars: any classic jungle or drum & bass track will do — Goldie's "Inner City Life" and Shy FX's "Original Nuttah" are canonical entries. Deliverable now: one paragraph on what the break has that a programmed loop of the same pattern wouldn't. Deliverable after a week of normal listening: a tally of how many times you caught the Amen (or any chopped break) in the wild once you knew its face.

B6. Hat dialects, blind. (20 min) Have a friend (or a randomized playlist) play you one-bar excerpts from six genres — trap, house, boom-bap, pop, rock, R&B/neo-soul — and try to ID the genre from the hats and backbeat placement alone, before the melody gives it away. For each: name the dialect features you used (rolls and triplet switches? offbeat opens? dust and heavy swing? backbeat on 3 versus 2-and-4?). Deliverable: your six IDs with evidence, and your score. Four or better means the genre table stuck; below that, replay the chapter's hat-language tour with the records playing.

Part C — 🎛️ DAW Lab

C1. The flat-loop rebuild. ★ (60–90 min — the chapter's centerpiece; do this one even if you do nothing else) Open the Productive Struggle loop: 96 BPM, eight bars, kick/clap/sixteenth-hats/808, quantize 100%, flat velocities. Save it as drums_v0_flat and bounce it — that's your "before." Now rebuild in five committed stages, bouncing after each so the progression is auditable:

1. Timing decisions (v1). Kick stays essentially tight (downbeats dead-on; nudge any syncopated kicks by ±3–6 ms by hand if your ear asks for it). Drag the clap late — sweep between +8 and +25 ms while the loop plays, eyes off the numbers, and stop where your neck stops you. Write down the number you chose.
2. Swing (v2). Sweep your DAW's swing control across the hats only — kick and clap excluded — from 50% up toward the mid-60s and back, eyes closed. Land it. Log the percentage and compute the off-sixteenth delay in ms.
3. Velocity architecture (v3). Build the hat skeleton: accents ~100–125 on the downbeats, body ~70–95 on the "&"s, ghosts ~25–55 on the e's and a's. No two adjacent hits identical. Let the last three steps of each bar lift slightly into the next downbeat, and let the whole pattern breathe in a slow two-bar wave.
4. Ghosts and conversation (v4). Add snare or rim ghosts (velocity 30–50) tucked between backbeats — two or three per bar maximum. Resist the urge to fill every empty step; the empty chairs are load-bearing.
5. Punctuation (v5). One small turnaround variation in bars 4 and 8 (an extra kick, a displaced clap, an open hat) — make bar 8's slightly bigger. Mute-test it: if muting the variation costs the transition nothing, replace it with something that earns its seat.

Finish with the A/B that makes it real: v0 against v5 at matched loudness, three honest listens. Deliverable: six bounces, plus a journal entry logging every number — clap offset, swing %, velocity ranges — because those numbers are the first draft of your pocket. (Chapter 40 will ask for them back.)

C2. The swing-dial experiment. (25 min) Take your C1 loop at 96 BPM and print six versions: hats at 50%, 52%, 54%, 57%, 62%, and 66.7% swing, everything else frozen. Shuffle the bounces so you can't see which is which, listen blind, and rank them by neck. Then — this is the point — change the session tempo to 140 BPM and audition 57% again. Log what happened to the feel and why (your A3 math predicts it). Deliverable: your blind ranking, your home-zone guess for 96 BPM, and one sentence on what the tempo change proved about percentages versus milliseconds.

C3. The velocity-only drill. ★ (25 min) New pattern: one bar of sixteenth closed hats, one hat sample, every hit locked to the grid at 100% quantize. The rule: you may touch nothing but velocity — no timing, no swing, no sample swaps, no FX. Program three versions: (a) rolling — strong-ghost-medium-ghost with a lifting tail, (b) marching — accents on every downbeat, near-flat elsewhere, (c) urgent — accents shifted onto the offbeats. Bounce all three and A/B. Deliverable: three bounces and a one-line verdict on the chapter's claim that accent patterns beat new samples. If the three bars don't feel like three different drummers, your ghost-to-accent spread is too narrow — widen it (try 40+ values of separation) and re-bounce.

C4. Layer and tune the kit to the key. (40 min) Using your own track's key (Jaylen's session runs A minor; yours is yours): (a) Build a layered kick — one transient layer for the knock, one body layer for the weight. High-pass the transient layer until the overlap stops adding mud; run the polarity-flip test and keep the fatter combination; if both flips sound thin, nudge one layer's start a few ms and re-test. (b) Build a layered backbeat — clap over snare — and pitch one layer by semitones until they read as one drum instead of a duet. (c) Tune what sings: loop your kit against your track's chord pad and pitch-check anything with a singable sustain — kick body, toms, tonal percs — by semitones until the wrestling stops. Leave what knocks. Deliverable: a session note logging every semitone move and which polarity-flip won, plus a bounce of the kit grooving against the pad.

C5. The 808 line, keyed and choked. (30 min) Load an 808 with audible sustain. Find its fundamental (tuner, spectrum analyzer, or by ear against your pad) and tune the sample so its MIDI note is honest. Program a bassline on your progression's roots — for an A-minor session that's the Chapter 9 sketch's territory, down around A1 — then make the two decisions that separate pro low end from soup: trim or soften the 808's attack so your kick owns the first ~20 ms, and choke every note to end just before the next low-end event. Add exactly one slide at a phrase boundary. Bounce a "ringing" version (no chokes) and a "choked" version and A/B on headphones and small speakers. Deliverable: both bounces and a note on where the ringing version fell apart first.

C6. Top-loop integration. (30 min) Take any full drum loop you have the right to use (a licensed pack loop or one you record yourself — even pens on a desk works). High-pass it around 200 Hz until its kick weight is gone and it reads as a top. Layer it over your C1 programmed kick and clap. Two problems will appear; solve both: match the swing (extract the loop's groove as a template if your DAW offers it — Appendix E names the feature per DAW — or nudge your hats until the argument stops), and mind the dirt (if the loop's snare pokes through against your backbeat, chop or duck those hits out). Deliverable: a bounce of the hybrid, plus the mute test — does the top earn its seat, or is it soup?

C7. Chop and flip. (40 min) Choose a two-bar source you can legally play with — your own Chapter 12 recordings are clearance-proof crates. Run all three chop methods on it: transient slicing, grid slicing, and manual chops at musically chosen moments. Map eight chops to pads or keys and play a new phrase the original never contained — re-order, repeat, choke chops against each other. Then resample: print your flip as new audio and build one more pass on top of the print. Deliverable: the original, the flip, and the resampled flip, plus one sentence on what committing to the print changed about your decision-making.

Part D — Synthesis & Critical Thinking

D1. "Drum machines have no soul." (25 min) Write two honest paragraphs. Paragraph one: defend the bumper sticker — what the session drummers of the early '80s correctly heard in factory-preset machine rhythm, using this chapter's mechanisms. Paragraph two: prosecute it — why the observation indicted the settings rather than the machine, with house and techno as exhibits A and B (genres where the grid's pitilessness is the aesthetic) and the programmable parameters where soul actually lives. End with one sentence stating when you would choose quantize 100% on purpose.

D2. The constraint sprint. ★ (60 min) Run the MPC sidebar's lesson as an experiment. Self-impose: one kit (eight sounds maximum), sixteen steps per pattern, eight allowed velocity values, one hour, no undo past the last bounce. Build a complete beat. Then write a half-page debrief: which constraint hurt, which one helped, how many decisions you made compared to a normal session, and whether the result has more identity than your average unconstrained sketch. (T2's sharpest form: infinite options produce the flat grid faster than they produce a sound.)

D3. The Winstons memo. (20 min) "Amen, Brother" gave drum & bass its DNA, and G.C. Coleman — the drummer — famously saw essentially none of the money. Write a half-page memo to your future self establishing your own sampling policy: what you'll do freely in the lab, what gates a release (the two copyrights, the clearance-before-traction rule), which safe lanes you'll favor (licensed packs, your own recordings, interpolation with composition clearance), and what you owe — legally and otherwise — to the players inside the records you flip. Chapter 36 will give you the legal map; this memo is the ethics draft it'll be laid over.

D4. Re-dialect the beat. (45 min) Take your finished C1 groove and translate it into two other genres from the hat-language table without changing the kick and clap sounds — only patterns, placement, swing, and velocity. Suggested pair: house (kick to four-on-the-floor and dead straight, offbeat open hats, humanity upstairs only) and boom-bap (tempo down to ~90, heavy swing, dusty texture, snare pushed late). Deliverable: two bounces plus a three-line analysis per version: which of the four roles changed, which survived untouched, and what that says about where genre actually lives.

Part M — Mixed Practice

Interleaving earlier chapters — spaced retrieval is how this sticks.

M1. (Ch 2 + Ch 13) The SP-1200's beloved crunch came from pitching records up to fit ten seconds of memory, then pitching back down to play. Using Chapter 2's vocabulary — sample rate, aliasing, and what transposition does to both — explain where the grit physically comes from, and why the same artifact is a flaw in a vocal session and a flavor in boom-bap. (Tradeoffs, not rules.)

M2. (Ch 4 + Ch 13) ★ Explain "sustain is real estate" using Chapter 4's masking: why does an 808 tail ringing into the next kick cost more clarity per millisecond than the same overlap would up at 2 kHz? Then connect Chapter 4's loudness law to sample browsing: why must candidates be level-matched before you're allowed to have an opinion?

M3. (Ch 12 + Ch 13) The polarity-flip test on a layered kick is Chapter 12's multi-mic physics indoors. Explain the connection: what do two mics on one snare and two samples on one kick track have in common, what exactly does the flip switch invert, and why is calling it the "phase" button — as everyone does — technically the wrong word? (Chapter 12's sidebar has the distinction if it's fuzzy.)

M4. (Ch 9 + Ch 13) Chapter 9 taught humanization on the chord pad; this chapter taught structured deviation on drums. Write the four-line upgrade memo that unifies them: what's wrong with ±20 ms random jitter on any part, what "directional and consistent" looks like on a pad versus a clap, and which element of a session should stay the most machined — and why that answer is the hierarchy rule wearing different clothes.

M5. (Ch 6 + Ch 13) Your drum cast is now nine tracks deep. Apply Chapter 6's session hygiene: write the naming scheme and color code you'll use for the nine, state which bus they route to in your template's six-bus skeleton, and explain in two sentences why future-you — staring down ~34 tracks at mix time in Chapter 20 — will care that you did this tonight rather than then.