Capstone Project 1: The Original Single

The default path. One original track, concept to release-ready master, every pipeline stage executed and documented. This is the book's progressive project formalized into an assessed brief — if you've been doing the Project Checkpoints, you've already walked this road once with the chapters holding your hand. Now you walk it without them.

Overview & Outcomes

This project asks for one thing, whole: a single. Not a beat. Not a demo. Not "basically done except the master." A release-ready record with the paperwork that proves it — the same artifact Jaylen shipped at midnight in Chapter 39, scaled to your song, your room, your genre.

The song matters less than you think and more than you'll admit. What's actually being assessed is whether you can run the complete pipeline — concept, capture, edit, arrangement, mix, master, QA, package — making one defensible decision after another and writing those decisions down. Anyone can get lucky at one stage. The capstone asks you to be deliberate at every stage, in sequence, which is the job description.

By the end, you'll have demonstrated that you can:

• define a record's intent before touching a fader (Chapter 5), and hold the finished work accountable to it;
• plan and execute capture and programming that fix problems at the source instead of in the mix (Chapters 7–14);
• edit invisibly, arrange by subtraction, and defend every part's existence (Chapters 15–17);
• mix to translation rather than to your room (Chapters 20–30);
• print, master, and measure to modern platform reality (Chapters 31–33);
• assemble a deliverables package a distributor or collaborator could use without phoning you (Chapters 35–36);
• and document the whole run so your decisions are traceable to reasons, not moods (Chapter 19's hygiene doctrine, applied to your own thinking).

A word on how this differs from the Project Checkpoints you may have already done. The checkpoints taught stages in isolation, each with a chapter spotting you. The capstone assesses the pipeline as one continuous run, where your Chapter 16 subtraction decisions shrink your Chapter 22 EQ workload, and your Chapter 21 gain habits decide how much trouble Chapter 32 inherits. Stage-to-stage consequence is the thing no single chapter could teach, and the thing this project exists to make you feel.

One honest sentence about scale: done properly, this is four to eight weeks of real work. Budget like it.

Prerequisites

Chapters 1–33 are the working minimum — the full pipeline through loudness QA. Chapters 35–37 are required before the release-week milestones; you can start the project while still reading Part VIII, but don't arrive at the upload gate without it. Chapter 34 is optional. Chapter 38 is optional but recommended (the AI audit makes a sharp logbook entry). Chapter 39 is your map — read it twice, once before you start and once the week you finish.

The practical floor, gear-agnostic as always: a DAW you're fluent in, with your template built (Chapter 6); monitoring calibrated and its lies learned (Chapter 8); a room you've clap-tested and treated to budget (Chapter 10); ear training in progress (Appendix F); and the symptom vocabulary installed (Chapter 30), because you'll be your own first mix doctor.

If you built a track through the Project Checkpoints, choose: finish that track to capstone standard, or start fresh. Both are legitimate. Fresh is recommended if your checkpoint track was a learning mule you no longer love — the capstone goes better when you care — and it has a second virtue: it proves the pipeline is yours now, not the book's.

The Brief

Produce one original track, any genre, release-ready, and deliver the following six items. "Release-ready" has a precise meaning here — see What "Done" Means below.

1. The master. A release-ready stereo WAV at your session's native sample rate, 24-bit (include a dithered 16-bit alternate if you like), printed through your Chapter 32 chain. If you actually release it — encouraged, optional — the live link joins the package.
2. The loudness QA numbers. Measured, not estimated: integrated LUFS for the full track, maximum true peak (≤ -1.0 dBTP), and a one-paragraph justification of your target — the Chapter 33 argument made for your genre and your platforms, hedged the way platforms force ("as of this writing"). "It's probably fine" is not a number.
3. The session, in pro hygiene. Chapter 19 standard: named and colored tracks, version history with sane semantics (no "final_v3_REAL"), audio consolidated, plugin list exported, 3-2-1 backup evidenced. The grading question is reopenability: could a stranger navigate this session cold?
4. The stems package. Printed from zero, all files the same length, your bus architecture as the stem map — the six-bus standard (DRUMS / BASS / SYNTHS / GTR / VOX / FX) adapted to your track — wet/dry decisions stated, and a documentation sheet: sample rate, bit depth, tempo, key.
5. The production logbook. The signature deliverable. Every pipeline stage in order, and for each: what you did, the ONE key decision you made there, and the reasoning — the format of Chapter 39's walkthrough, written by you, about you. One page per stage, maximum. This is what makes craft assessable. Without it, the rubric can only grade your luck.
6. The release plan — or the release. Either a complete four-week release plan (Chapter 35's timeline doctrine plus Chapter 37's campaign built from your own production artifacts), or evidence of an actual release executed on that doctrine. A plan must be executable as written: dates, deliverables, the pitch window, a content calendar. Not vibes.

Milestones

The table maps the book's pipeline onto a 4–8 week run. Compress or stretch to fit your life, but keep the order — the sequence is the lesson. Treat each milestone as gated: don't open the next while the current one is failing.

Two milestones get under-budgeted by almost everyone: the edit (Milestone 4 always takes twice what you planned — schedule for that and you'll be on time) and the troubleshoot pass (Milestone 8 is where the record gets honest; don't let release excitement compress it). The mix gets two weeks not because the moves are slow but because fresh ears need scheduling — Chapter 20's session psychology is part of the assessment surface. And Milestone 6 is non-negotiable: a solo producer's errors have no natural predator, so Chapter 39 promoted the outside ear from courtesy to gate. Someone who isn't you hears this before it ships.

One habit turns this schedule into the logbook deliverable for free: close every milestone by writing its entry the same day — what you did, the one decision, the reason — while the reasoning is still true. Producers who batch the logbook at the end write fiction, politely. The rubric can usually tell.

Constraints & Variations

The brief is genre-agnostic on purpose. The gates are physics; genre is logistics. Use Appendix H to profile your lane's conventions and Chapter 18's audit to choose your compliance and your violations deliberately. One adaptation worth naming: a fully instrumental track is legitimate. The lead element — lead synth, sampled hook, melodic 808 — takes the vocal's seat in the Chapter 29 sense: it gets the full attention-chain treatment and the sits-everywhere verification.

The three reader archetypes. The bedroom producer is this brief's native species — Jaylen's road, end to end. The self-recording band member can absolutely run it solo, capture-heavy; but if another human's song or performance is the center of your work, Capstone Project 2 will assess you better. The podcaster or content creator is welcome here — this project stretches your music muscles — but if the show is your real lane, Capstone Project 3 points the same rigor at your actual output.

The collaboration variant. Add one guest performer — a vocalist, a guitarist — the way "Static Bloom" itself carried Demi's voice and Theo's guitar. This triggers two additions: the Chapter 19 feedback protocol run with a real human, and a split sheet executed before release (Chapter 36). Recommended if you have the human; it makes Dimension 5 of the rubric real instead of conditional.

Hard mode — the "constraint album" variant. Declare three binding constraints in your manifesto and keep them to the end. The menu: twelve tracks maximum in the session; one synth for every synthetic sound; one mic for every capture; nothing sampled that you didn't record yourself; stock plugins only; or the four-track bounce discipline — commit your bounces, no recall, the Chapter 5 Sgt. Pepper lesson lived rather than admired. Log each constraint's bill and its gift in the logbook. This variant exists because constraint is the fastest known route to identity (Chapter 40), and it's called the constraint album because the producers who try it tend to keep the rules and make four more tracks. You've been warned.

What "Done" Means

Done is not a feeling. Done is the gate list.

Gates 0 through 6 passed — song defended, mix translating, print to spec, master referenced, loudness measured, release QA run, package complete with metadata matching the splits — makes the project release-ready and assessable. Gates 7 through 10 are the released path: upload on the four-week lead, campaign loaded, release day executed, next track started. Either endpoint is acceptable for assessment; the gates you claim must be the gates you can evidence. The printable list with sub-items and sign-off lines is in Appendix G; Chapter 39 is the commentary.

The done-beats-perfect boundary holds here exactly as Chapter 39 drew it: the gates are the floor, the version-2 list is the ceiling deferred. A flaw that doesn't fail a gate goes on the list, and the project ships anyway. That's not a compromise. That's the discipline.

Keep that version-2 list as a live document from week one, not a guilty afterthought. Every "I should have" that doesn't fail a gate gets written down the moment it surfaces, which does two things at once: it protects the current record from endless reopening, and it hands your next track a head start — Chapter 40's template-iteration loop, already running before this project is even graded.

Assessment

Score yourself against the Capstone Assessment Rubric before anyone else sees the work — the gap between your self-scores and your assessor's is free ear training. Six dimensions, weighted, with a 70% pass gate, the same line the book's quizzes draw.

Know which deliverable feeds which dimension: the master and stems carry Technical Quality and Translation; the logbook carries Craft Decisions and Process & Documentation almost entirely. An undocumented good decision is scored as luck, because from the outside, that's what it is.

If you're working without an instructor, recruit a two-person panel: one producer-literate ear and one civilian. The producer scores the rubric; the civilian answers two questions only — "did anything annoy you?" and "would you play it again?" — which between them cover more of Technical Quality, Translation, and Identity & Taste than you'd expect.

Common Failure Modes

Every one of these has a chapter waiting for it. Diagnose, reopen, fix, return.

1. The endless demo. The arrangement never freezes; week six arrives mid-experiment. Fix: the mute test and a freeze date — Chapter 16 — and Chapter 39's done-beats-perfect doctrine.
2. Mixing the arrangement's problems. Three hours of EQ on a chorus with five mid-range parts. Chapter 20's threshold concept: a mix problem is usually an arrangement problem wearing a disguise.
3. Skipping the static balance. Plugins before faders. The most important ninety minutes of the mix are the cheapest — Chapter 20.
4. Gain-staging archaeology. The 2-bus at +4 dB with every fader below zero. Chapter 21's red-light audit, source to 2-bus.
5. The headphone-only mix. Passes your room, fails the car — the book's founding wound, relived. Chapter 30's translation protocol, plus Chapter 8's headphone lies and Chapter 25's mono fold-down.
6. Mastering into a number. The limiter pushed until the meter flatters, dynamics gone. Chapter 33: normalization turns it down anyway. Master for the music.
7. Printing through a limited 2-bus. The master has nothing left to work with. Chapter 31's print spec: 24-bit, ~-6 dB peaks, no 2-bus limiter.
8. The day-of upload. Editorial windows missed, pre-save impossible. Chapter 35's four-week doctrine exists because Jaylen learned it almost too late.
9. The logbook written at the end. Memory invents tidy reasons for messy choices. Write each stage's entry the day you close the stage — Chapter 19's hygiene applied to thought.
10. No outside ear. Solo means nothing stops a mistake except the gates — so use all of them, including the human one (Chapter 19).

Ship it. Then start the next one within fourteen days — Gate 10 is real, and Chapter 40 is waiting.