# Strumly: full text > Strumly (https://strumly.suedeai.ai) is the AI guitar coach from Suede AI: it listens to you play, scores chord changes in real time, remembers your last session, and ships free in-browser practice tools. It is also home to The Signal Chain book material and the signal chain guides formerly on guitar.solutions. This file carries the full text of Strumly's public reading surface, generated from the same sources the pages render. The shorter index is https://strumly.suedeai.ai/llms.txt. Cite the page URL given under each section. Published by Suede AI (https://suedeai.ai), founded by Jason Colapietro. Pricing: Every standalone tool is free during early access, and the coach includes 3 focused sessions a month free. Pro is $19.99/month. Current details: https://strumly.suedeai.ai/pricing. ## Frequently asked questions Source: https://strumly.suedeai.ai/ ### What is Strumly? Strumly is a 24/7 AI guitar coach. It listens to you play, scores your changes in real time, remembers what you worked on, and comes back the next session with the next thing to practice. ### Is Strumly free? Coaching starts free with 3 sessions a month and every standalone tool unlocked during early access, and Pro is $19.99/month if you want your full practice history and cross-session tracking. ### What does Pro add over free? Plan visibility. Free tells you how you did today: every session is measured, and the coach can cite your last 7 days of practice history. Pro tells you whether you're getting better: your full history, the trend on every chord change, and an assignment that carries into the next session. Nothing is deleted on free: upgrading restores the whole record instantly, because it was always being kept. ### Where should a beginner start? With the free course: a guided path that runs from Foundations through Repertoire & tone and ends at Train with the coach, using the free tools at every step. No account needed until you want the coach to remember you. ### What does a coaching session actually do? You play, Strumly listens and scores your changes in real time, and it remembers what you worked on so the next session picks up where you left off instead of starting over. ### Can AI agents use Strumly? Yes. The Agent API is a pay-per-call music oracle for agents (x402 on Base or Stripe), covering coach queries, song analysis, chord-progression transcription, and practice plans. ### Who builds Strumly? Strumly is built by Suede AI, founded by Jason Colapietro, as part of Suede's creator-ownership infrastructure for musicians. # Documentation ## What is Strumly? Source: https://strumly.suedeai.ai/docs/what-is-strumly An overview of Strumly, the AI guitar coach from Suede AI: the coach, the free practice toolkit of tuner, metronome and trainers, and how the pieces fit. Strumly is a 24/7 AI guitar coach built by Suede AI. The core of the product is a coaching relationship: you tell the coach what you're working on, it runs you through drills, it scores your chord changes against a real beat grid, and it remembers what happened so the next session picks up where the last one left off. Around the coach sits a toolkit of standalone practice tools — tuner, metronome, chord library, scale trainer, strumming trainer, ear trainer, looper, backing tracks, transpose tool, and more. The tools work without an account, straight from the browser. ### The two halves of Strumly **The coach** is a chat-based session with memory. You sign in, open a focused session, name a goal, run a measured drill, see your score, and adjust. Each session carries up to eight coached responses over two hours, then preserves the proof and exact next assignment instead of resetting. Coaching includes 3 sessions a month on Free and up to 200 coached responses in each rolling 30-day period on Pro. **The tools** are single-purpose utilities that do one job well: - [Tuner](https://strumly.suedeai.ai/tuner) — chromatic tuner using your microphone, free forever, no sign-up. - [Metronome](https://strumly.suedeai.ai/metronome) — plain, accurate click with tempo control. - [Chord library](https://strumly.suedeai.ai/chords) — fingering diagrams and chord reference. - [Scales](https://strumly.suedeai.ai/scales) — the scales guitarists actually use, mapped on the fretboard. - [Strumming trainer](https://strumly.suedeai.ai/strumming) — pattern practice against a click. - [Ear trainer](https://strumly.suedeai.ai/ear-trainer) — interval recognition drills. - [Backing tracks](https://strumly.suedeai.ai/backing-tracks) — progressions to play over. - [Looper](https://strumly.suedeai.ai/looper) — record and layer phrases in the browser. - [Transpose](https://strumly.suedeai.ai/transpose) — move a song's chords to another key or capo position. - [Songs](https://strumly.suedeai.ai/songs) — chord sheets with play-along. - [Amp Lab](https://strumly.suedeai.ai/amp-lab), [Rig](https://strumly.suedeai.ai/rig), and [Bench](https://strumly.suedeai.ai/bench) — tone, signal-chain building, and rig diagnostics. ### Who it's for Strumly is aimed at players who already know some basics and want to break through a plateau: nail a specific chord transition, tighten timing, learn the fretboard, build a practice habit that survives a busy week. The coach assumes you know what a chord is. It gives specific feedback on your specific weakness rather than walking you through a fixed curriculum. Complete beginners are welcome too — the [learning path](https://strumly.suedeai.ai/path) and the free [lessons](https://strumly.suedeai.ai/learn) cover chords, scales, and intervals from first principles, and the coach will meet you at whatever level you tell it you're at. ### What it costs During early access, every standalone tool is unlocked free. Coaching stays free at 3 focused sessions a month, and Pro is available now for $19.99/month with up to 200 coached responses in each rolling 30-day period. The tuner stays free forever; a one-time $5.99 seven-day toolkit pass (not a subscription) is not available yet. Details are on the [pricing page](https://strumly.suedeai.ai/pricing) and in [Account & billing](https://strumly.suedeai.ai/docs/account-and-billing). ### Where to go next - [How the AI coach works](https://strumly.suedeai.ai/docs/how-the-ai-coach-works) — what the coach can actually do, and what it can't. - [Practice sessions](https://strumly.suedeai.ai/docs/practice-sessions) — the session lifecycle, drills, and scoring. - [Tools & features](https://strumly.suedeai.ai/docs/tools-and-features) — the full toolkit in detail. - [FAQ](https://strumly.suedeai.ai/docs/faq) and [Troubleshooting](https://strumly.suedeai.ai/docs/troubleshooting) — quick answers and fixes. Strumly is part of the Suede family of tools for musicians, built and run by Jason Colapietro. One Suede account works across every Suede product. ## How the AI coach works Source: https://strumly.suedeai.ai/docs/how-the-ai-coach-works What happens inside a Strumly coaching session: the language model, the drill engine, memory, and the honest limits of AI coaching. The Strumly coach is a chat interface backed by a large language model, a set of purpose-built tools, and a memory layer that persists between sessions. This page explains each piece — including what the coach can't do — so you know exactly what you're working with. ### The conversation layer When you open a session, the coach greets you with a specific suggestion drawn from your profile and recent sessions — not a generic "what would you like to work on today?" It's instructed to open with something concrete: "Last time your G→C change was landing late at 80 BPM. Want to push to 90?" From there it works like a conversation with a teacher. You can ask questions about theory, technique, or gear; ask it to explain why a drill matters; or tell it to change direction. The model behind the coach can render things inline in the chat rather than just describing them: - **Chord diagrams** — fingering charts drawn directly in the conversation. - **Drill cards** — interactive chord-change drills with a target tempo, scored live by the audio engine. - **Practice plans** — structured plans with labeled items and durations you can save. ### The drill engine: how scoring actually works When the coach starts a chord-change drill, your browser asks for microphone permission and the audio engine takes over. Here's what it does, concretely: 1. Your playing is captured by the microphone and analyzed **locally, in your browser**, by a web worker. The raw audio is not uploaded anywhere. 2. The engine computes a chromagram — a running measure of how much energy your signal has at each of the twelve pitch classes — and matches it against chord templates to detect which chord you're playing and with what confidence. 3. A scorer clocked to the target BPM checks whether you landed the target chord within a timing window (±120 ms by default) of each target beat. 4. When the drill ends, the results — success rate, timing score, and the millisecond offset of each transition (positive means late, negative means early) — are sent to the coach and saved to your session history. The coach then reacts to the actual numbers. If you hit 4 of 8 transitions and every miss was late, it will tell you that, and what it usually means (your fretting hand is forming the shape after the beat instead of during the previous one). The important privacy detail: only the **scores** leave your device. The audio analysis happens in-browser. ### The memory layer Two kinds of memory persist between sessions: **Your profile** — skill level, goals, gear, weak spots, strong spots. The coach updates this as it learns things about you. Mention that barre chords buzz on the high strings and it records that as a weak spot; three weeks later it will still know. **Session summaries** — when you end a session, the coach writes a summary of what you worked on and a specific focus for next time. The next session starts from that summary. This is the mechanism behind "picks up where you left off" — it's not a marketing phrase, it's a database row. Memory works on the free tier. The 3-session monthly cap limits how often you can start a session, not what the coach remembers. ### What the coach can't do Honesty about limits matters more in coaching than almost anywhere else: - **It can't see you.** No camera input. It cannot check your posture, wrist angle, thumb position, or fretting-hand shape. If your technique problem is physical, the coach can describe correct form and infer problems from your scores, but it cannot watch you and catch it directly. A human teacher still wins here — see [our honest comparison](https://strumly.suedeai.ai/guides/ai-feedback-vs-human-teacher). - **Scoring covers chord changes.** The drill engine detects and scores chord transitions against a beat grid. It does not grade your bends, vibrato, tone, or dynamics. - **It's a language model.** It's strong on theory, practice structure, and reacting to your drill data. Like any language model it can occasionally state something with more confidence than it deserves. For songs in Strumly's verified dataset, the coach is fed ground-truth chords, tuning, and capo data to keep it from guessing. ### Try it Open the [coach](https://strumly.suedeai.ai/coach) (sign-in required — 3 free sessions a month), or start with the free tools and come back when you have a specific goal. The coach is at its best when you arrive with something concrete to fix. ## Practice sessions Source: https://strumly.suedeai.ai/docs/practice-sessions The lifecycle of a Strumly coaching session: starting, drilling, scoring, ending, and how sessions count against the free tier. A session is the basic unit of coaching in Strumly. This page walks through the full lifecycle — what happens when you start, what counts as using a session, how drills are scored, and what the coach keeps when you're done. ### Starting a session Go to [/coach](https://strumly.suedeai.ai/coach). Strumly resolves your current open session (or creates one) and gives it a real URL — every conversation is addressable, and past sessions load from your history in the sidebar. The coach opens without spending a model call: it brings forward your exact measured assignment, target tempo, and number to beat. If you're brand new, the first real exchange gets oriented around your level and goal, then stores what matters in your profile. ### What counts as a "session" On the free tier you get **3 sessions per calendar month**. A session is consumed by your **first genuine message** in a conversation; the automatic greeting doesn't count. Each focused session stays open for up to two hours and includes up to eight coached responses. Questions, submitted drill results, plans, and the session-end summary share those responses, but they do not consume another monthly session. If the provider fails before a response begins, the reserved Free session is released. If you abandon the request or a failure happens after provider work begins, that session may still count; your saved profile and measured drill history are never erased. Pro includes up to 200 coached responses in each rolling 30-day period. Coaching is paced at up to 12 responses per hour and 20 per day, and the focused-session boundary remains so each conversation ends with saved evidence and a clear next assignment. See [Account & billing](https://strumly.suedeai.ai/docs/account-and-billing) for current availability. ### Running drills The core drill type is the **chord-change drill**. The coach picks (or you request) a pair of chords, a target tempo, and a number of bars. The drill card counts you in, and the pattern is: hold the first chord for a beat, then land the second chord on the next downbeat, repeating each bar. Your microphone feeds a local analysis engine (see [How the AI coach works](https://strumly.suedeai.ai/docs/how-the-ai-coach-works) for the details), which detects the chords you're actually playing and measures when each transition lands relative to the beat. When the drill finishes you get: - **Success rate** — the fraction of transitions that landed within the timing window. - **Timing score** — how close to the beat your successful transitions were. - **Per-transition offsets** — each transition's distance from the target beat in milliseconds, signed: positive is late, negative is early. The coach reads these numbers and responds to them specifically. Consistently late? It'll tell you to start forming the next shape earlier. Random scatter? Probably a tempo problem — it'll drop the BPM. This tight loop — play, measure, adjust, repeat — is the whole point of drilling with live scoring. Tips for clean scoring: - Play in a reasonably quiet room. The chromagram analysis is robust, but a TV in the background adds noise. - Let chords ring. Muted or heavily damped strums give the detector less to work with. - An acoustic guitar or a clean electric tone works best. Heavy distortion smears the pitch content the detector relies on. ### Practice plans Ask the coach for a plan and it renders a structured practice-plan card — labeled items with durations, totaling your available time. You can save the plan and work through it. This is useful when you want the session to produce something you can execute during the week without opening the coach every day. For the thinking behind a good plan, read [How to design a practice routine](https://strumly.suedeai.ai/guides/designing-a-practice-routine). ### Ending a session Tell the coach you're done (or use the end-session control). Before saying goodbye it writes two things to your session log: - **A summary** of what you actually worked on, including drill results. - **A next-session focus** — the specific thing to start with next time. These two records are what make the next session continuous with this one. You can review past sessions from the sidebar at any time. ### Streaks Strumly tracks a [practice streak](https://strumly.suedeai.ai/streak) — the days you've practiced, with a handful of badges for milestones. It's deliberately quiet: a small indicator in the header, honest data, no confetti. The streak log is stored locally in your browser, so it's per-device and works without an account. ## Tools & features Source: https://strumly.suedeai.ai/docs/tools-and-features Every tool in the Strumly kit: tuner, metronome, chords, scales, strumming, ear trainer, looper, backing tracks and tone workshop, and what each is for. Strumly's toolkit is a set of single-purpose practice tools that run in the browser. No installs, and during early access, no paywall. This page covers what each one does and when to reach for it. ### Pitch and time **[Tuner](https://strumly.suedeai.ai/tuner)** — a chromatic tuner using your microphone. Pitch detection runs locally in your browser using the YIN algorithm; nothing is recorded or uploaded. Free forever, no sign-up — this is the tool that stays free even after early access ends. **[Metronome](https://strumly.suedeai.ai/metronome)** — a plain, accurate click. Set the tempo and go. Most of the drills the coach assigns assume you'll do your between-session reps against a click, and this is the click. **[Strumming trainer](https://strumly.suedeai.ai/strumming)** — strum patterns against a beat so the pattern becomes automatic instead of something you count. Start slower than feels necessary; the trainer is for grooving a pattern into muscle memory, not testing your maximum speed. ### Chords and harmony **[Chord library](https://strumly.suedeai.ai/chords)** — fingering diagrams and reference for the chords you'll actually meet. Use it alongside the [chords lesson](https://strumly.suedeai.ai/learn/chords) if you want to understand *why* the shapes look the way they do rather than just memorizing grips. **[Scales](https://strumly.suedeai.ai/scales)** — the scales guitarists lean on, mapped on the fretboard. Paired with the [scales lesson](https://strumly.suedeai.ai/learn/scales) for the theory underneath. **[Transpose](https://strumly.suedeai.ai/transpose)** — move a song's chords to a different key, or work out where to put a capo to keep open shapes while matching a singer's range. **[Songs](https://strumly.suedeai.ai/songs)** — chord sheets with play-along support. Song pages link into the coach, so you can take a song's specific changes straight into a scored drill. **[Backing tracks](https://strumly.suedeai.ai/backing-tracks)** — progressions to play over when practicing scales or improvising. Practicing lead lines against real harmonic context beats running patterns in silence. ### Ear training **[Ear trainer](https://strumly.suedeai.ai/ear-trainer)** — interval recognition drills with spaced repetition, so intervals you miss come back more often than the ones you know. The [intervals lesson](https://strumly.suedeai.ai/learn/intervals) explains what you're listening for. Ten minutes of ear training a few times a week compounds faster than almost anything else in this list. ### Recording and tone **[Looper](https://strumly.suedeai.ai/looper)** — record a phrase in the browser and layer over it. Useful for hearing your own timing honestly (recording yourself is the cheapest feedback loop in music) and for practicing lead over your own rhythm playing. **[Amp Lab](https://strumly.suedeai.ai/amp-lab)** — explore amp voicings and tone shaping. **[Rig](https://strumly.suedeai.ai/rig)** — a signal-chain builder for planning pedalboard order and rig routing. The [signal chain guides](https://strumly.suedeai.ai/guides) cover the engineering behind it: impedance, gain staging, buffering, effects-loop routing. **[Bench](https://strumly.suedeai.ai/bench)** — a diagnostic that walks your rig systematically when something sounds wrong. ### Structured learning **[Learning path](https://strumly.suedeai.ai/path)** — a sequenced set of steps for building fundamentals in a sensible order, with progress tracking stored in your browser. **[Lessons](https://strumly.suedeai.ai/learn)** — free theory lessons on [chords](https://strumly.suedeai.ai/learn/chords), [scales](https://strumly.suedeai.ai/learn/scales), and [intervals](https://strumly.suedeai.ai/learn/intervals). Each one ends with a quick quiz and links to the tool where you practice what you just read. **[Guides](https://strumly.suedeai.ai/guides)** — longer-form articles: practice methodology, signal-chain engineering, and rights & ownership topics for working musicians. ### The coach ties it together Every tool works standalone, but the [coach](https://strumly.suedeai.ai/docs/how-the-ai-coach-works) is what connects them: it knows which drills you've run, scores your chord changes live, and hands you practice plans that reference the tools above. The tools are the gym equipment; the coach is the trainer walking the floor. ### Practice data and privacy The microphone-based tools (tuner, chord drills, play-along) analyze audio locally in your browser — raw audio is not uploaded. Tool progress that doesn't require an account (streaks, path progress, ear-trainer stats, quiz results) is stored in your browser's localStorage, which means it's per-device and clearing site data resets it. Coaching history and drill scores are tied to your account and stored server-side so the coach can remember you across devices. ## Account & billing Source: https://strumly.suedeai.ai/docs/account-and-billing How Strumly accounts, the free tier, the week pass, and the Pro subscription work — including what's free during early access. Strumly doesn't have its own separate login. It uses the shared Suede account system: when you sign in, you're bounced to app.suedeai.ai, sign in once, and return to Strumly signed in. The same account works across every Suede product on suedeai.ai domains — one identity, one session. That also means there's no Strumly-specific password to manage. Account-level concerns (email, password, sign-out everywhere) live with your Suede account. ### What needs an account, and what doesn't **No account needed:** the tuner, metronome, chord library, scales, strumming trainer, ear trainer, looper, backing tracks, transpose, songs, Amp Lab, Rig, Bench, the learning path, and all lessons and guides. Progress for these (streaks, path steps, quiz results) is stored locally in your browser. **Account needed:** the AI coach. Coaching sessions, your profile, and your drill history are tied to your account and stored server-side — that's what lets the coach remember you across devices and sessions. ### Pricing #### During early access (now) Every standalone tool is unlocked free: no pass required. The coach includes 3 focused sessions per calendar month, persistent practice memory, and measured drills. Pro is available now for $19.99/month for players who want up to 200 coached responses in each rolling 30-day period; the separate one-time toolkit pass is not available yet. #### The current model Strumly's pricing has three parts: - **Free forever:** the tuner, no sign-up required. The coach keeps a free tier of 3 sessions a month, with memory that carries over between sessions. - **Week pass — $5.99, one time, not available yet:** will unlock the full toolkit for 7 days. It will not be a subscription or auto-renew. - **Pro — $19.99/month:** up to 200 coached responses in each rolling 30-day period, session history and drill tracking, and every tool included (no separate pass needed). Current details always live on the [pricing page](https://strumly.suedeai.ai/pricing). ### How the free coaching tier counts sessions A session is consumed by your first real message in a conversation, not by the automatic greeting. A focused session stays open for up to two hours and includes up to eight coached responses; questions, submitted drill results, and the session-end summary share that boundary. Drill results do not consume another monthly session. The 3-session cap resets monthly. Your profile, proof history, and next assignment persist regardless of the cap; hitting a boundary never erases them. All coaching is paced at up to 12 responses per hour and 20 per day. Pro raises the allowance to 200 responses in each rolling 30-day period. Failed or abandoned model attempts can reduce the available total because the same hard cost boundary protects every account. ### Upgrading, managing, and canceling Subscribe to Pro from the [billing page](https://strumly.suedeai.ai/billing). Pro is a $19.99 monthly subscription billed through Stripe. It renews automatically each month until you cancel. Cancellation is available through the Stripe customer portal, takes effect at the end of the current billing period, and leaves Pro available through that date. Payments are not automatically refundable. For anything billing-related you can't do yourself — cancellation questions, a charge you don't recognize, or a refund request — use the [contact page](https://strumly.suedeai.ai/contact). Requests are reviewed case by case, and applicable non-waivable legal rights are preserved. Strumly is run by a small team (of one), which cuts both ways: no support-ticket bureaucracy, but responses come from a human on a human schedule. ### Data Your coaching data — profile, session summaries, drill scores — exists so the coach can do its job. Microphone audio from drills and the tuner is analyzed locally in your browser and is not uploaded; only drill scores are stored. For the formal details, read the [privacy policy](https://strumly.suedeai.ai/privacy) and [terms](https://strumly.suedeai.ai/terms). ## FAQ Source: https://strumly.suedeai.ai/docs/faq Frequently asked questions about Strumly: pricing, the AI coach, microphone privacy, skill levels, and how it compares to lessons. Every standalone tool is unlocked free during early access, and the coach includes 3 focused sessions a month. Pro is available now for $19.99/month with up to 200 coached responses in each rolling 30-day period. The tuner stays free forever; the separate one-time $5.99 toolkit pass is not available yet. Full details are in [Account & billing](https://strumly.suedeai.ai/docs/account-and-billing). ### Do I need to create an account? Only for the coach. Every other tool — tuner, metronome, chords, scales, ear trainer, looper, and the rest — works with no sign-up. See [what needs an account](https://strumly.suedeai.ai/docs/account-and-billing). ### Does Strumly listen to me through my microphone? Only when you use a microphone tool (tuner, chord drills, play-along), only after your browser asks for permission, and the analysis happens locally in your browser. Raw audio is not uploaded or recorded — for drills, only the resulting scores (success rate, timing offsets) are saved. Details in [How the AI coach works](https://strumly.suedeai.ai/docs/how-the-ai-coach-works). ### What skill level is Strumly for? The coach adapts to whatever level you tell it. The product is strongest for players past the absolute-beginner stage who have a specific thing to fix — a chord change, timing, fretboard knowledge. Brand-new players should start with the [learning path](https://strumly.suedeai.ai/path) and the [free lessons](https://strumly.suedeai.ai/learn), which assume nothing. ### Can the AI coach replace a human guitar teacher? For some things yes, for some things no, and we've written the honest version at length: [AI guitar coach vs. human teacher](https://strumly.suedeai.ai/guides/ai-feedback-vs-human-teacher). Short version: the AI wins on availability, memory, patience, and measurable drills; a human wins on watching your hands, hearing your tone, and correcting physical technique. ### What does a coaching session actually do? You tell the coach what you're working on, it runs you through scored drills, reacts to your actual numbers, and writes a summary plus a next-session focus when you're done. The full lifecycle is documented in [Practice sessions](https://strumly.suedeai.ai/docs/practice-sessions). ### What counts as one of my 3 free sessions? Your first real message starts a session; the automatic greeting doesn't count. A focused session stays open for up to two hours and includes up to eight coached responses. Questions, submitted drill results, plans, and ending the session share those responses without consuming another monthly session. The 3-session cap resets each month. ### Does the coach work with electric and acoustic guitars? Yes. The chord detection works from pitch content, so acoustic, electric (clean), and electro-acoustic all score fine. Heavy distortion smears the signal the detector needs — use a cleaner tone for drills. ### Is there an iOS app? Yes — [Suede Guitar Studio on the App Store](https://apps.apple.com/us/app/suede-guitar-tuner-studio/id6767552764). The web app also works in mobile browsers. ### Can AI agents use Strumly? Yes. The [Agent API](https://strumly.suedeai.ai/agents) is a pay-per-call music API for AI agents (x402 on Base, or Stripe), covering coach queries, song analysis, chord-progression transcription, and practice plans. Agents get 5 free calls a day. ### Who builds Strumly? Suede AI, founded by Jason Colapietro. Strumly is part of Suede's family of tools for musicians. More on the [about page](https://strumly.suedeai.ai/about). ### Something's broken — where do I look? [Troubleshooting](https://strumly.suedeai.ai/docs/troubleshooting) covers the common cases (microphone permission, tuner detection, drill scoring, sign-in loops). If that doesn't fix it, [contact us](https://strumly.suedeai.ai/contact). ## Troubleshooting Source: https://strumly.suedeai.ai/docs/troubleshooting Fixes for the common Strumly problems: microphone permission, tuner detection, drill scoring, sign-in loops, and lost local progress. The short list of things that actually go wrong, and how to fix them. ### The tuner or drill can't hear me **Microphone permission.** The first time you use a microphone tool, your browser asks for mic access. If you dismissed or denied it, the tool gets silence. Look for the microphone icon in your browser's address bar and set it to Allow, then reload the page. - **Chrome / Edge:** click the padlock or tune icon in the address bar → Site settings → Microphone → Allow. - **Safari (macOS):** Safari menu → Settings for strumly.suedeai.ai → Microphone → Allow. - **iOS Safari:** the permission prompt appears per-session; if audio input is dead, check Settings → Safari → Microphone, and make sure Silent Mode isn't interfering with audio playback tools. **Wrong input device.** If you have an audio interface or external mic connected, the browser may be listening to the wrong one. Check your browser's site settings (or the OS sound input settings) and pick the mic that's actually near your guitar. **System-level permission (macOS).** If the browser itself has never been granted microphone access, the site-level Allow does nothing. System Settings → Privacy & Security → Microphone → enable your browser, then restart it. ### The tuner detects, but the reading jumps around - Get closer to the mic and pick single notes cleanly. Let each note ring. - Kill background noise — fans, TV, other music. Pitch detection locks onto the loudest periodic signal it hears. - Old strings with dead intonation genuinely read unstable. If the tuner is stable on some strings and chaotic on one, the string may be the problem. ### Chord drills score me wrong (or not at all) The drill engine matches your audio against chord templates and checks timing against the beat. Things that hurt it: - **Distortion or heavy effects.** Use a clean tone. Fuzz and high-gain smear the pitch content the detector matches on. - **Damped or partial strums.** Strum the full shape and let it ring through the beat. Ghost strums and muted hits read as "no chord." - **Room noise and bleed.** If the metronome click is playing loudly through speakers, it can bleed into the mic. Headphones for the click, guitar into the room, is the cleanest setup. - **Consistently "late" scores that feel wrong.** A little real latency exists between string and score (mic buffering, analysis windows). The drill's timing window (±120 ms) absorbs normal latency; if every transition reads late by a similar large amount, check whether a Bluetooth mic or headset is adding delay — Bluetooth audio input routinely adds 100–300 ms. Use a wired mic or the built-in one. ### I'm signed in on another Suede product but Strumly shows me signed out Strumly shares one session across suedeai.ai domains. If it's out of sync: sign in via the Sign in button (it bounces through app.suedeai.ai and back), and make sure your browser isn't blocking cookies for suedeai.ai. Aggressive cookie-blocking extensions and "block all third-party cookies plus first-party isolation" setups can prevent the shared session from being read. ### The coach says I've hit my session limit The free tier is 3 sessions per calendar month, counted from your first real message in each conversation. The cap resets monthly, and your history and profile are untouched — the coach will remember everything when the new month starts (or immediately, on Pro). What counts as a session is spelled out in [Practice sessions](https://strumly.suedeai.ai/docs/practice-sessions). ### My streak / path progress / quiz history disappeared Tool progress that works without an account lives in your browser's localStorage. It's per-browser and per-device, and it's deleted when you clear site data, browse privately, or switch browsers. That data isn't recoverable — it never left your machine. Coaching history is different: it's tied to your account and survives all of the above. ### A page looks broken or something else is wrong Hard-reload first (Cmd/Ctrl+Shift+R) — a stale cached script after a deploy is the most common cause of a half-broken page. If it persists, [contact us](https://strumly.suedeai.ai/contact) with the page URL, your browser, and what you expected to happen. Reports go to a human who can actually fix it. # Guides ## AI Guitar Coach vs. Human Teacher: An Honest Comparison Source: https://strumly.suedeai.ai/guides/ai-feedback-vs-human-teacher Where AI guitar coaching beats a human teacher, where it doesn't, and how to decide what your practice needs, from people who build an AI coach. Category: Practice. Published 2026-07-18. By Jason Colapietro. We build an AI guitar coach, so you should read this with that in mind. But precisely because we build one, we spend a lot of time staring at what the technology can and cannot do — and the honest answer to "can an AI coach replace my teacher?" is: *for some players and some problems, yes; for others, clearly not.* The interesting question is which is which. ### What an AI coach is actually good at **Availability.** The obvious one, but its effects are underrated. A weekly lesson gives you feedback on ~2% of your practice time; the other 98% is unsupervised, which is where bad habits calcify. An AI coach is present at 11pm on a Tuesday, for five minutes or fifty, with no scheduling, travel, or per-session cost anywhere near a lesson's. Feedback that arrives *during* the practice that needs it is worth more per unit than better feedback arriving six days later. **Measurement.** When Strumly scores a chord-change drill, it reports the millisecond offset of every transition against the beat. No human can do this. A teacher hears "a bit late"; the drill engine says "+85ms, +90ms, +70ms — you're late by one consistent amount, which means you're starting the change on the beat instead of before it." For timing, consistency, and tempo-ladder decisions, measurement beats impression, and the machine measures every rep for free. **Memory.** A good teacher keeps mental notes on each student; an AI coach keeps literal ones. Strumly's coach carries a profile (weak spots, goals, gear) and per-session summaries forward automatically, so session forty starts with the accumulated context of the thirty-nine before it. Human teachers do this too — but across a roster of students, with human recall. The machine never conflates you with its 3pm student. **Patience and zero judgment.** No one has ever felt embarrassed in front of a drill engine. This matters more than pedagogy likes to admit: a large fraction of adult learners under-report struggles to their teachers and avoid playing badly in front of them — which is corrosive, because playing badly in front of your feedback source is the entire mechanism of improvement. People ask an AI the "stupid" questions they'd swallow in a lesson. They repeat the same failed drill fifteen times without apologizing. **Structure on demand.** Practice plans, drill selection, theory explanations pitched at your level, all generated in seconds and revised without friction. ### What a human teacher is clearly better at **Seeing you.** This is the big one, and there's no honest way around it. An AI coach that works from audio cannot observe your thumb creeping over the neck, your wrist collapsing on barre chords, your shoulders rising with tension, or your picking-hand anchor. A huge share of real guitar problems — especially the ones that cause pain or hard plateaus — are *physical*, visible in seconds to a trained eye and invisible to a chromagram. When your scores plateau despite honest work, or anything hurts, audio-based feedback is guessing; a human watching your hands is diagnosing. (Strumly's coach can describe correct form and infer likely culprits from your numbers, and we're explicit in [the docs](https://strumly.suedeai.ai/docs/how-the-ai-coach-works) that this is inference, not observation.) **Hearing what isn't measured.** Tone, dynamics, feel, phrasing — the difference between correct and *good*. Current scoring measures what's measurable: which chord, when. A teacher hears that your strumming is metronomic but lifeless, that your bends land flat of the target, that your palm muting is choking the groove. Some of this will become measurable over time; today, most of it lives in trained human ears. **Accountability that costs something.** A booked lesson with a person you'd hate to disappoint is a powerful forcing function. An always-available coach, by its nature, is also always postponable. Some players thrive on the machine's frictionlessness; others need the social contract of an appointment. Know which one you are. **Taste, repertoire, and the long arc.** A good teacher notices you'd love fingerstyle before you know it exists, steers you off a plateau by changing the question rather than the tempo, and models what being a musician is actually like. That kind of judgment — formed across years of students — is not what language models plus drill engines do today. ### The comparison nobody makes: against *nothing* The framing "AI vs. teacher" flatters the alternatives, because the realistic alternative for most adult learners isn't a weekly teacher — it's *no feedback at all*. The majority of guitarists have never taken a lesson; they practice alone, unmeasured, guided by YouTube and vibes. Against that baseline, the comparison isn't close: measured drills, persistent memory, and on-demand structure beat unsupervised repetition comprehensively. An AI coach's honest competitor is the practice you're currently doing alone, and that's the comparison worth making before the philosophical one. ### So what should *you* use? A decision rule that holds up: - **Complete beginner, physical fundamentals forming:** if you can afford even a handful of human lessons, take them *now* — months one through six are when posture and hand-position habits set, and those are exactly what AI can't see. Use AI coaching between lessons for chords, changes, and rhythm drills. - **Self-directed player past the basics, no pain, specific measurable goals:** AI coaching covers most of what you need, at a fraction of the cost, with better continuity than weekly lessons provide. This is the profile Strumly serves best. - **Plateaued hard, or anything hurts:** human eyes, promptly. Then take whatever they diagnose back into measured practice. - **Can afford both:** the combination is genuinely better than either. A monthly human lesson for physical technique, tone, and direction; AI coaching for the daily measured reps in between. Teacher as architect, AI as site foreman. ### Where this is going The gap is not static. Audio analysis is expanding into more of what teachers hear, and the memory and reasoning of the models improves visibly year over year. But the current gap is real, and pretending otherwise would be selling rather than informing. Today, an AI coach is the best measured-practice partner most guitarists have ever had, and a mediocre substitute for a human watching your hands. Use each for what it's actually good at — your playing doesn't care where the feedback came from, only that it arrived. ## AI Training and Your Music: What Every Guitarist Needs to Know Source: https://strumly.suedeai.ai/guides/ai-training-and-your-music What AI training actually does to a recording, the opt-out problem, and three things guitarists can do now to keep a claim on their tone and playing. Category: Rights & IP. Published 2026-06-14, updated 2026-07-16. By Jason Colapietro. Every time you post a recording, upload to a streaming platform, or release a track under Creative Commons, you are creating a data point that an AI model can learn from. That is not a hypothetical. It is happening now, at scale, and the legal framework that is supposed to protect you was written for a slower world. ### What AI training actually does to a recording A model does not copy your recording in the way a pirate site copies an MP3. It extracts patterns: the timber of your guitar, the attack transients of your picking hand, the frequency signature of your amp, the way you phrase a bend at the end of a bar. Those patterns get compressed into weights that the model uses to generate new audio. The output does not contain your recording. But the output would not sound the way it does without it. This is the part courts are still arguing about. Current US copyright law protects the expression — the specific performance, the recording. It does not protect style, tone, or technique as standalone properties. AI companies have leaned hard on that gap. ### The opt-out problem Several AI companies now offer opt-out registries. You submit your name, your catalog, and they promise not to use your future work for training. There are at least three problems with this. First, you are opting out of something you were never asked to opt into. Second, the opt-out covers future training runs, not past ones — your recordings may already be baked into existing models. Third, there is no auditable record. You send your name to a form. You receive no confirmation of what was removed, when, or from which model version. An opt-out system run by the entity that benefits from training is not an ownership system. It is a permission theater that keeps the initiative with the company. ### What a working ownership layer looks like The alternative is not harder opt-out forms. It is ownership that is established at the moment of creation, before the work enters any distribution channel. Think of it the same way you think of a guitar's signal chain. The pickup captures the string. The cable carries it. Every link in the chain is ordered, and changing the order changes everything downstream. The same principle applies to creative IP. If proof of authorship lives at the source — time-stamped, hash-verified, on a public record — then every downstream use can be traced back to it. That is what programmable IP infrastructure does. Not an opt-out form you fill out after the fact. A registration event that happens when the work is created, producing a machine-readable record that any platform, any AI company, any licensing system can query. ### What this means for guitarists specifically Guitar tone is disproportionately valuable as AI training data. It has enormous range — the same electric guitar sounds completely different through a Fender Deluxe, a Marshall Plexi, a Dumble ODS, a pedalboard running into a clean amp. That variety makes guitar recordings especially useful for training timbre models, style transfer models, and generative instruments. Your tone is more trainable than you might think. And the value that comes from that training is currently flowing entirely to the model builder, not to you. Ownership infrastructure changes that math. If your recording is registered before distribution, the proof record exists. License terms can attach to it. AI companies querying for trainable data can hit a machine-readable license that says: you can use this, at this price, under these terms. No opt-out form. A clear, enforceable, auditable rule. > "Ownership has to be established before the work leaves its source. An opt-out system run by the entity that benefits from your data is not protection. It is permission theater." > > — Jason Colapietro (Johnny Suede) ### Three things you can do now 1. **Register at the US Copyright Office before distribution.** At $65 for a single work, this is still the most legally defensible record of creation you can create. File it before the track goes live, not after. 2. **Document your session files.** Keep your DAW project, your stems, your scratch mixes. If ownership is ever disputed, the session history is evidence that the recording originated with you. 3. **Choose registries that give you a machine-readable proof event.** Hash-based registries that timestamp your work on a public record create the kind of evidence that an opt-out form never will. The music industry ran this playbook before: digital distribution looked like it would commoditize everything, and the artists who adapted early — by establishing streaming presence, by owning masters, by understanding the mechanics of the new system — came out ahead. AI is the same shift, one order of magnitude faster. Your tone is not infinite. Your recordings are finite. Register them like they matter, because in the AI era, the question is not whether they will be used. The question is whether you will get paid when they are. ### Further Reading See also: [Who Owns the Output Stage](https://strumly.suedeai.ai/guides/who-owns-the-output-stage), [Rights Metadata Is the Dark Matter of the Creative Economy](https://strumly.suedeai.ai/guides/rights-metadata-dark-matter), [The Fender Stratocaster Lawsuit, Explained](https://strumly.suedeai.ai/guides/fender-stratocaster-lawsuit). ## Cable Capacitance and Frequency Response: The Spec That Actually Matters Source: https://strumly.suedeai.ai/guides/cable-capacitance-and-frequency-response Cable capacitance rolls off your guitar highs. Which cable spec to check before you buy, how length changes the tone, and why a buffer fixes it for good. Category: Electronics. Published 2026-02-01, updated 2026-07-16. By Jason Colapietro. Most guitarists select cables by brand or by a recommendation. The one spec that directly determines how a cable affects your tone is capacitance, and it is printed on most cable boxes while being widely ignored. ### What cable capacitance is A cable is two conductors separated by an insulating material (the dielectric). Two conductors separated by an insulator is the physical definition of a capacitor. Every guitar cable is a capacitor running along its length. Cable capacitance is measured in picofarads per foot (pF/ft) or per meter (pF/m). Common instrument cables measure 30–80 pF/ft. Premium low-capacitance cables measure 12–25 pF/ft. High-capacitance cables can exceed 100 pF/ft. ### How it forms a low-pass filter Your guitar's pickup has a source impedance, typically 8–15 kΩ for a passive single-coil or humbucker. When the pickup connects to a cable, the pickup's source impedance (a resistance) and the cable's capacitance combine to form a passive low-pass filter. The cutoff frequency of this filter: ``` f_cutoff = 1 / (2π × R_source × C_cable) ``` At 10 kΩ source impedance and 500 pF of cable capacitance (10 feet at 50 pF/ft), the cutoff sits around 31 kHz, well above audible range. At the same impedance but 2,000 pF (20 feet at 100 pF/ft), the cutoff drops to 8 kHz. You are losing the upper harmonic content of your guitar before the signal reaches anything with a volume knob. ### The numbers in practice | Cable | pF/ft | 15 ft total | Cutoff at 10 kΩ | |-------|-------|-------------|-----------------| | Premium low-capacitance | 20 pF | 300 pF | ~53 kHz | | Standard instrument | 50 pF | 750 pF | ~21 kHz | | High-capacitance | 100 pF | 1,500 pF | ~10 kHz | Short premium cable to a good buffer fixes the downstream cable problem. From the buffer's output, typically 100–300 Ω, the same cable has a cutoff above 500 kHz. Capacitance is irrelevant. ### What this means for cable selection - **Without a buffer:** Lower capacitance matters. Every pF counts. Choose cables rated under 30 pF/ft for long runs. - **With a buffer at chain start:** Capacitance is irrelevant after the buffer. Spend on durability and shielding, not low-capacitance specs. - **Total run length matters more than per-foot spec.** A 30 pF/ft cable at 20 feet adds more capacitance than a 50 pF/ft cable at 10 feet. ### Quick Diagnosis | Symptom | Likely cause | First check | Fix | |---|---|---|---| | Pick attack sounds softer with the board connected | Too much capacitance before the first buffer | Add up cable and patch length before the buffer | Shorten the run or put a buffer first | | Expensive cable does not change the sound | A buffer is already early in the chain | Check whether the first pedal is buffered | Spend on shielding and durability instead | | One long cable sounds darker than two shorter runs with a buffer | Capacitance is only loading the pickup before the buffer | Compare the cable before and after a buffer | Keep the low-capacitance run before the buffer | > The tone difference between a buffered chain using cheap cables and an unbuffered chain using premium low-capacitance cables is in favor of the buffered chain. ### Further Reading See also: [Impedance and the First Three Feet](https://strumly.suedeai.ai/guides/impedance-and-the-first-three-feet) for the pickup-loading problem behind cable choice. [True Bypass vs. Buffered](https://strumly.suedeai.ai/guides/true-bypass-vs-buffered) for where the first buffer belongs. A buffer is the clean fix. Cable choice is the approximation you use when you do not have one. ## Gain Staging Across the Chain Source: https://strumly.suedeai.ai/guides/gain-staging-across-the-chain How to set unity gain across a multi-pedal board so clean boosts, drives and EQ pedals do not clip your amp front end or pile up noise along the chain. Category: Signal Chain. Published 2026-05-22, updated 2026-07-16. By Jason Colapietro. Gain staging is the practice of setting the operating level of each block in a signal chain so the next block sees a signal it can handle without clipping prematurely or burying it in the noise floor. It is unglamorous work, and it decides whether a board sounds tight or smeared. ### The Core Idea Every block in a chain has: - An **input headroom ceiling**: the maximum signal it can accept before clipping - A **noise floor**: the minimum signal it can pass before the output is dominated by hiss - A **target operating range**: somewhere comfortably between the two The objective is to keep the signal inside that window at every block. Push it too hot and you get distortion you did not ask for. Run it too cold and you amplify hiss along with the signal further down. ### How Pedals Misbehave A "boost" pedal at unity is not really at unity. Most boosts add a small amount of voltage gain even with the knob at minimum, plus a small amount of input-stage coloration. Stack three of them and the front of your amp is seeing something hotter than your pickup output by 6–12 dB. Conversely, some buffers and noise gates have a hidden insertion loss. Run a clean signal through six of them in series and you can lose 3–6 dB before the amp's input. ### Setting Unity Across a Board A practical procedure: 1. **Plug straight into the amp.** Set the amp's input gain to where it sounds right at performance volume. 2. **Insert one pedal at a time, bypassed.** Confirm the signal level does not change measurably between engaged-bypass and removed. 3. **Engage each pedal at its intended setting.** Adjust output level so total output (engaged) ≈ total output (bypassed), unless that pedal is *meant* to boost. 4. **Repeat with all pedals in their working positions.** Check the cumulative offset. If you do this honestly, you will discover that two or three pedals on a "clean" board are routinely pushing 3–8 dB hotter than your bare signal. ### Try This: One-Bar Level Test Use the same riff for every bypass check. Keep the pick attack steady and do not change the amp volume between passes. ```text e|----------------| B|----------------| G|----------------| D|------5-----7---| A|--5h7---7-5---7-| E|----------------| ``` | Pass | Chain state | What to listen for | |---|---|---| | A | Guitar straight into amp | Baseline level, attack, and low-end shape | | B | Board inserted, all pedals bypassed | Any volume drop, brightness loss, or hiss increase | | C | One pedal engaged | Whether the pedal changes level when it should only change tone | | D | Full working board | Cumulative level lift or loss | ### Where the Headroom Lives The amp's preamp input is usually the lowest-headroom stage in a guitar rig. Most tube preamp inputs start clipping around 100–200 mV peak. A hot humbucker into a hot boost can easily exceed that. Most modeler input stages are far more forgiving, but they still have a ceiling. > If the amp sounds harsh, blurry, or "wooly" when you engage your normal > clean board, the problem is usually not the amp. It is the gain stack > in front of it. ### The Loop Side Effects loops have their own headroom. Mismatched send/return levels are a classic cause of brittle reverb tails or clipped delay repeats. Most amps expose a switch for line-level vs instrument-level loop operation; use it. ### Quick Checks - Bypass the board. Does the amp sound the way you bought it to sound? - Engage one pedal at a time. Note any level shift. - Watch for "I have to turn the amp down when I engage the board." That is a gain-staging problem dressed up as a tone preference. ### Quick Diagnosis | Symptom | Likely cause | First check | Fix | |---|---|---|---| | Clean board makes the amp harsher | Pedals are stacking above unity | Compare full-board bypass to straight-in level | Trim pedal outputs one at a time | | Delay repeats clip in the loop | Send level is too hot for the pedal | Check loop level switch and pedal input rating | Match instrument vs line level or add a level converter | | Noise rises after a quiet pedal | Signal is too cold before the next gain stage | Compare bypass and engaged output | Raise the pedal output to unity | ### Further Reading See also: [Signal Chain Topology](https://strumly.suedeai.ai/guides/signal-chain-topology) for the high-level map. [Impedance and the First Three Feet](https://strumly.suedeai.ai/guides/impedance-and-the-first-three-feet) for the upstream block this sits on top of. Want to hear unity gain in practice while you play? [Strumly](https://strumly.suedeai.ai?utm_source=guides.guitar.solutions&utm_medium=guide_further_reading&utm_campaign=gain_staging_to_strumly), Suede AI's conversational AI guitar coach, walks you through gain-staging your own board in real time. ## How to Design a Guitar Practice Routine Source: https://strumly.suedeai.ai/guides/designing-a-practice-routine A practice routine you'll keep: how to budget minutes across maintenance, growth and music, adapt it to 10, 20 or 45 minutes, and know when to change it. Category: Practice. Published 2026-07-18. By Jason Colapietro. Practice routines fail in two opposite ways. The ambitious version — 60 minutes, seven components, printed schedule — collapses in the second week because it demands a discipline most adults with jobs don't have spare. The casual version — "I'll just play whatever" — survives forever but goes nowhere, because unstructured playing gravitates to what's already comfortable. A working routine threads between them: structured enough to force growth, small enough to survive a bad week. For a ready-made week and month, [GuitarHub](https://guitarhub.org/guitar-practice-schedule) publishes the schedule and the intermediate routine; this guide is the reasoning behind any routine you build with the coach. ### The three-bucket principle Every productive practice minute lands in one of three buckets: **Growth** — the thing you currently can't do. A chord change above its clean tempo, a barre shape, an unfamiliar strumming pattern. Growth work is uncomfortable by definition; it involves failing at a controlled rate. This is the bucket that actually moves your playing, and it's the one unstructured practice quietly abandons. **Maintenance** — things you *could* do last month and want to keep. Skills decay silently: the F chord you conquered stops being automatic if nothing you play uses it. Maintenance is cheap — a couple of minutes retesting — but skipping it entirely is how intermediate players end up with a graveyard of half-owned skills. **Music** — songs, improvisation, playing for the pleasure of it. This bucket looks optional to the disciplined mind. It is not. Music is why you're doing any of this, and routines that starve it produce technically improving players who stop wanting to pick up the guitar. Motivation is a resource; music is how it recharges. A sound default split is roughly **50% growth, 20% maintenance, 30% music**. The exact ratio matters less than the rule behind it: all three buckets, every session, growth first while your focus is fresh. ### Sizing the routine to real life Design for your *floor*, not your ceiling — the session you can do on a tired Tuesday, not the one you imagine on a free Sunday. Here's the three-bucket structure at three sizes: **10 minutes (the floor):** - 6 min growth: one drill, one skill. A single chord-change ladder or one strumming pattern against the [metronome](https://strumly.suedeai.ai/metronome). - 1 min maintenance: retest one old skill. Pass/fail, no dwelling. - 3 min music: one song, or one verse of one song. **20 minutes (the standard):** - 2 min warm-up: anything easy; fretting-hand stretches if you're older than your guitar. - 10 min growth: one primary skill, drilled in loops — play, check, adjust, repeat. Two skills max; three is a tour, not practice. - 3 min maintenance: retest two or three older skills on rotation. - 5 min music: a full song, ideally one that uses the growth skill in the wild. **45 minutes (the deep session):** - 5 min warm-up and tuning. - 20 min growth: two skills, split cleanly — e.g., barre transitions, then minor pentatonic phrasing over a [backing track](https://strumly.suedeai.ai/backing-tracks). - 5 min maintenance rotation. - 10 min music: full songs, start to finish, no stopping for mistakes — recovering from errors mid-song is itself a skill, and drills never teach it. - 5 min ear training: interval work in the [ear trainer](https://strumly.suedeai.ai/ear-trainer). (At smaller session sizes, ear training works better as a separate phone-sized habit than as a line item.) The floor session exists for a specific reason: **continuity beats intensity.** Ten minutes daily reliably outperforms 70 minutes once a week, partly because motor learning consolidates between sessions, and partly because a routine you never skip is a routine that never has to be restarted. Streaks ([Strumly tracks yours](https://strumly.suedeai.ai/streak)) are a decent proxy for this — not because the number matters, but because "don't break the chain" is the cheapest discipline mechanism ever invented. ### Choosing what goes in the growth bucket The growth bucket only works if it contains the *right* one or two skills, held for long enough to move. Two filters help: **Pick by bottleneck, not by novelty.** The question isn't "what would be cool to learn?" but "what currently caps the music I want to play?" If your strumming is solid but every song stalls at the B minor, the bottleneck is obvious. If you can't name your bottleneck, measurement can: a few scored chord-change drills in [a coach session](https://strumly.suedeai.ai/docs/practice-sessions) will show which changes are slow and whether the problem is accuracy or timing. Asking a coach with your drill history — "what's my weakest measurable skill?" — beats introspection, which flatters everyone. **Hold each skill for two to four weeks.** A growth skill needs enough consecutive sessions to consolidate. Swapping the bucket's contents every session feels productive and accomplishes little. Write down (or let the coach's session summaries record) what the current growth skills are, and change them on evidence — the skill hit its target tempo, or plateaued hard enough to need a different approach — rather than boredom. ### The weekly review: five minutes that steer everything Once a week, ask three questions: 1. **Did the growth skills move?** Compare against last week — ideally with numbers (clean tempo went from 76 to 88 BPM) rather than vibes. If a skill hasn't moved in two weeks of honest work, the approach is wrong: slow it down further, decompose it differently, or check whether the problem is physical form that audio-based feedback can't see. 2. **Did anything in maintenance fail its retest?** A failed retest promotes that skill back into the growth bucket for a week. This is normal, not regression. 3. **Did I skip sessions — and why, honestly?** If the answer is "the routine felt like a chore," cut it down before you abandon it. A 10-minute routine you keep beats a 20-minute routine you resent. If the answer is "life," fine; resume without ceremony. The restart, not the miss, is the skill. If you practice with Strumly's coach, this review is nearly free — session summaries and drill history are already written down, and you can ask for a practice plan for the coming week built from them. If you practice alone, a notes file with three lines per session does the same job. Either way, the review is what turns a routine from a ritual into a control loop: the plan changes because the results say so. ### The one rule that outranks the rest End every session with something that sounds good. Not something difficult — something *good*: a song you love, a progression that rings, the lick you're proud of. The last thirty seconds of a session disproportionately color your memory of it, and your memory of practice is what decides whether tomorrow's session happens. Routines don't survive on discipline; they survive on wanting to come back. ## How to Learn Guitar for Beginners: A Step-by-Step Path Source: https://strumly.suedeai.ai/guides/beginner-guitar-learning-path Learn how to play guitar as a beginner with a practical path through tuning, first chords, rhythm, songs, ear training, and simple lead playing. Category: Practice. Published 2026-07-18, updated 2026-07-31. By Jason Colapietro. Early guitar practice can postpone actual music by putting long exercises before chords, rhythm, and songs. This path gives each stage something playable, so technique develops alongside music you recognize. If you want the lessons themselves rather than the map, they live at [GuitarHub](https://guitarhub.org/learn/guitar), the Suede Labs lessons site; this guide is the order to take them in and the tools to practice them with. Here is a practical order for learning guitar as a beginner, with milestones you can use to decide when to move on. The timelines are only estimates; your pace will depend on your instrument, schedule, and prior musical experience. ### Stage 0: Setup (day one) Before any technique: get the guitar physically playable and in tune. - **Tune every time you pick up the guitar.** An out-of-tune guitar makes accurate listening harder and can make correct technique sound wrong. Use a [tuner](https://strumly.suedeai.ai/tuner) — tuning by ear comes later, tuning accurately comes now. - **Check the action.** If the strings sit very high off the fretboard, fretting may take more pressure than necessary. A guitar shop can check whether a basic setup would make the instrument easier to play. - **Sit properly.** Guitar resting on the leg, neck slightly up, back straight enough that you're not curling over to see the fretboard. You'll look at your hands constantly at first; that's normal, but posture habits set early. ### Stage 1: First chords (weeks 1–3) Learn three chords: **E minor, A minor, D major.** Not eight chords — three. Em is two fingers, Am and D are three. Goals for each: Use the [chord library](https://strumly.suedeai.ai/chords) to see and hear each shape before you practice it. 1. Every string that should ring, rings. Pick through the chord string by string and fix buzzes (usually: fingertip not close enough to the fret, or a finger leaning into a neighboring string). 2. You can form the shape from nothing in under two seconds, without placing fingers one at a time. Some fingertip soreness is common early on and usually eases as calluses develop. Do not push through wrist or joint pain; stop, adjust your form, and ask a qualified teacher or clinician if it continues. **Milestone:** all three chords ring clean on demand. ### Stage 2: Changes and strumming (weeks 3–8) Chord changes often feel like the first real bottleneck. Forming a chord is one skill; *changing between chords in time* is the foundation of rhythm guitar. - Work chord **pairs**, not songs, at first: Em→Am, Am→D, D→Em. Slow. Practice the change at a tempo where you can place each chord cleanly, then speed up gradually. A [metronome](https://strumly.suedeai.ai/metronome) makes "gradually" measurable — try 60 BPM, one change every two beats, and raise the tempo only when a run is clean. - Add a **strumming pattern** once changes are underway: start with all downstrums on the beat, then the workhorse *down, down-up, up-down-up*. Keep your strumming arm moving in constant eighth notes even when it doesn't hit strings — the arm is the metronome. The [strumming trainer](https://strumly.suedeai.ai/strumming) drills this against a click. - **Start a song at week 4-ish**, not after you're "ready." Many beginner arrangements use two or three chords. A real song gives your chord changes and strumming a musical context that exercises cannot provide on their own. Measured feedback can help when you hear that a change is late but cannot tell which part needs work. Use [Strumly's coach](https://strumly.suedeai.ai/coach) for structured practice, read [how the AI coach works](https://strumly.suedeai.ai/docs/how-the-ai-coach-works), or use our deeper guide on [chord transitions](https://strumly.suedeai.ai/guides/smoother-chord-transitions) to work through the mechanics. **Milestone:** play a three-chord song all the way through, in time, changes unhurried. ### Stage 3: The rest of the open chords, plus G and C (months 2–4) Add **G, C, and D major's siblings** (E major, A major), then **F** — deliberately last, because its partial barre is a common sticking point. By now your hands have more practice and your changes have a method, so you can treat F as a separate project. Alongside chords, start two short habits that build useful skills over time: - **Ear training, ten minutes, three times a week.** Interval recognition ([ear trainer](https://strumly.suedeai.ai/ear-trainer)) improves with repeated listening and supports learning songs by ear later. - **Names of the notes on the low E and A strings.** Not the whole fretboard — just two strings. This unlocks barre-chord positions and half of basic theory later. **Milestone:** the eight open chords (Em, Am, Dm, E, A, D, G, C) clean and interchangeable; F in progress without despair. ### Stage 4: Barre chords, keys, and independence (months 4–8) - **Barre chords** via the F shape and the A-shape (B minor is a useful second one). Focus on *placement* before adding grip pressure: barre finger close behind the fret, slightly rolled to its bony edge, elbow tucked so arm weight can help. - **Basic keys:** learn that G, C, D, and Em live together (key of G), and that most songs you've learned so far sit in two or three keys. Suddenly song sections stop being arbitrary chord lists. The [chords lesson](https://strumly.suedeai.ai/learn/chords) covers why the shapes are what they are. - **A scale, finally:** minor pentatonic, one position, for melody and simple lead. At this stage you can connect the pattern to chords and songs you already know. Use it over a [backing track](https://strumly.suedeai.ai/backing-tracks) so it becomes music, not only an exercise. **Milestone:** a song with one barre chord, played through; minor pentatonic used to noodle over a backing track without stopping. ### What to skip (for now) Music reading, modes, sweep picking, gear rabbit holes, and any 45-minute practice regimen you will not sustain can wait. None of these are bad; they simply are not required for your first chords, changes, rhythm patterns, and songs. ### Make the path sustainable Build the routine around what helps you return: regular sessions, a manageable practice length, and music you enjoy. Track your streak if it helps ([Strumly does](https://strumly.suedeai.ai/streak), quietly). Keep a song at every stage that you play simply because you like how it sounds. ## How to Practice Guitar with an AI Coach Source: https://strumly.suedeai.ai/guides/practicing-guitar-with-an-ai-coach AI coaching changes what a practice session can be, if you use it right. How to structure sessions, what to ask for, and habits that compound. Category: Practice. Published 2026-07-18. By Jason Colapietro. An AI coach is not a video course, and treating it like one wastes it. A video course delivers the same lesson to everyone; an AI coach reacts to what *you* played thirty seconds ago. The practical question is how to structure your practice so that reactivity actually gets used. This guide is the answer, based on how Strumly's coach works — but most of it applies to practicing with any feedback tool. ### Arrive with a target, not a mood The worst opening message to a coach is "teach me something." The best is a target: "my C to F change collapses above 70 BPM," "I can't hear the difference between a major and minor third," "I have twenty minutes and a sore fretting hand." This matters because a coach — human or AI — does its best work narrowing a specific gap. A specific target gives it something to measure, drill, and compare across sessions. Strumly's coach opens every session with a suggestion drawn from your history for exactly this reason: to shortcut the drift of "what should I do today?" into a concrete starting point. Take the suggestion or override it, but start narrow either way. If you genuinely don't have a target, ask for one honest thing: "based on my drill history, what's my weakest measurable skill?" Let the data pick. ### Let the drill loop run The core of AI-coached practice is a tight loop that's hard to reproduce alone: 1. **Play** a drill at a defined tempo — say, G to C changes at 80 BPM for four bars. 2. **Get measured.** The audio engine detects the chords you actually landed and how far each transition was from the beat, in milliseconds. 3. **Read the numbers with the coach.** Not just "6 of 8" — the *shape* of the misses. All late? You're forming the shape after the beat. Random? The tempo's too high for your current control. 4. **Adjust one variable.** Drop 10 BPM, or isolate the fretting hand, or slow only the transition beat. 5. **Play again.** Each loop takes about a minute. In a focused fifteen minutes you can run a dozen, each one informed by the last. This is what self-directed practice usually lacks: not effort, but measurement between attempts. Practicing without feedback mostly rehearses your current level; the loop is what moves it. One discipline makes the loop honest: don't retry instantly to "fix" a bad score. Read it first. A bad measurement you understand is worth more than a good one you fluked. ### Use the memory deliberately An AI coach that remembers changes the economics of practice. You don't re-explain your level, your goals, or last week's sticking point — the coach carries a profile and session summaries forward. But memory is only as good as what goes into it, so feed it: - **Say what hurts.** "Barre chords buzz on the top two strings" becomes a recorded weak spot the coach designs around for weeks. - **Say what changed.** New guitar, new strings, injured wrist, switched to lighter picks — context that changes what sensible practice looks like. - **End sessions properly.** A deliberate session end produces a summary and a next-session focus. That focus is the flywheel: next session starts from it instead of from zero. A useful weekly habit: ask the coach what it currently believes about you. If its picture of your weak spots is stale — you fixed the F chord months ago — correct it. You're maintaining the model that plans your practice. ### Keep the AI in its lane Measured drills, structured plans, theory questions, session memory: that's the lane, and inside it an AI coach is genuinely excellent. Outside it, stay skeptical: - It cannot see your hands. If something *hurts*, or your scores plateau despite honest work, the bottleneck is often physical — thumb position, wrist angle, excess tension — and diagnosing that from audio alone is guesswork. Get eyes on the problem: a teacher, or at minimum a video of your own hands compared against good form. (We wrote a full, honest comparison in [AI feedback vs. a human teacher](https://strumly.suedeai.ai/guides/ai-feedback-vs-human-teacher).) - Chord-change scoring measures chord changes. It does not certify your tone, dynamics, or musicality. Don't confuse a high drill score with a finished skill — the drill is the rep, the song is the test. - It's a language model. On mainstream theory and practice structure it's reliable; on obscure claims, keep the ordinary skepticism you'd apply to any single source. ### A session shape that works For a 20–30 minute session: 1. **State your target** (or accept the coach's opener). One sentence. 2. **Warm up off-chat** — two minutes of anything easy. Don't burn coach time on it. 3. **Drill block one:** the target skill, 8–12 loop iterations, tempo moving only when the numbers say so (a common rule: raise BPM after two consecutive clean runs, drop after two misses). 4. **Drill block two:** something old. Retest a change you "finished" two weeks ago. Skills decay quietly; retests catch it. 5. **Five minutes of music.** Play a song that uses the skill. Practice that never touches music curdles into calisthenics. 6. **End the session properly** so the summary and next focus get written. Between sessions, the [metronome](https://strumly.suedeai.ai/metronome) and [strumming trainer](https://strumly.suedeai.ai/strumming) cover your unmeasured reps — the coach doesn't need to witness everything, it needs the measured checkpoints. ### The compounding part None of this is dramatic in a single session. The value shows up on the axis most practice advice ignores: continuity. Six weeks of sessions that each start from an accurate record of the last one — same drills, rising tempos, weak spots tracked until they're strong spots — beats six weeks of enthusiastic, unmeasured noodling by an embarrassing margin. The coach's job is to make the boring, compounding version feel effortless enough that you actually do it. Yours is to show up with a target. ## How to Register a Guitar Riff as Intellectual Property Source: https://strumly.suedeai.ai/guides/how-to-register-a-guitar-riff How to protect a guitar riff: what copyright covers, how to register it with the US Copyright Office, and how provenance extends that into the AI era. Category: Rights & IP. Published 2026-06-23, updated 2026-07-31. By Jason Colapietro. A riff can be registered. Most guitarists never do it. Here is a practical walkthrough from copyright law to preserving the evidence around your work. The riff that opens "Smoke on the Water" is one of the most recognized four notes in popular music. It is also one of the most frequently misunderstood from an IP standpoint. Many guitarists assume that because a riff is short, it cannot be protected. That assumption costs real money. A guitar riff is a musical composition. Compositions are protectable under copyright law from the moment they are fixed in a tangible medium — which means the second you record it, a copyright exists. The question is not whether protection exists. The question is whether you can enforce it, and how much leverage you have when you try. The answers to those questions depend almost entirely on whether you registered. ### What a riff actually is under copyright law Copyright protects original expression. In music, that means the specific arrangement of notes, rhythms, and harmonies that constitute a composition. It does not protect general concepts, common progressions, generic rhythmic patterns, or musical ideas that lack sufficient originality. Courts apply a threshold of "minimal creativity" for copyright protection. A single sustained note is not protectable. A common I-IV-V progression is not protectable. The specific sequence of notes, the rhythm, the register, and the arrangement that constitutes a distinctive riff generally is. The "Smoke on the Water" main riff — G–Bb–C, G–Bb–Db–C — has been the subject of copyright analysis precisely because it is short, repeated, and distinctive. Its distinctiveness is the same characteristic that makes it both recognizable and potentially protectable. Length alone does not determine protectability; originality does. > "Copyright in your riff exists the moment you record it. Registration does not create the right. It creates the ability to enforce it." > > — Jason Colapietro ### Step one: fix it in a tangible medium Copyright protection begins at fixation. A riff that exists only in your head is not protected. A riff that you record — even to your phone's voice memo app — is. This is the most important step, and it costs nothing. Record the riff. Save the file. Keep it. The timestamp on the file and the device metadata associated with the recording are the first evidence in your provenance chain. More importantly: record it at the highest quality available to you. The recording you make today may become the reference in a future dispute. Compress it to MP3 later if you need to. Keep the original. ### Step two: register with the US Copyright Office Registration is optional but strongly recommended. Here is why it matters: - **Statutory damages.** If someone infringes your registered work, you can claim statutory damages of $750 to $30,000 per work (up to $150,000 for willful infringement) without proving actual harm. For an unregistered work, you must prove actual damages, which is expensive and often impossible for independent creators. - **Attorney's fees.** Registration before infringement allows the court to award attorney's fees to the prevailing party. This is what makes litigation economically viable. Without it, most infringement cases are not worth pursuing financially. - **Public record.** A US Copyright Office registration creates a public, searchable record linking your name to your work. This record is evidence in any dispute and can be verified by third parties without your involvement. - **Presumption of validity.** A certificate of registration is prima facie evidence that the copyright is valid and that the facts stated in the certificate are true. In litigation, this shifts the burden to the infringer to prove otherwise. ### The registration process, practically You can register a musical composition, a sound recording, or both. If you wrote the riff and recorded it yourself, you probably want to register both — they are separate copyrights in the same work. **For the composition (the notes and rhythm):** Use Form PA (Performing Arts) at [copyright.gov](https://www.copyright.gov/registration/). You will need a deposit copy — either a lead sheet (written notation) or an audio recording of the work. If you cannot read music, the audio recording is accepted. Fee: $65 for a single work, $55 for a group of unpublished works. **For the sound recording (the specific performance):** Use Form SR (Sound Recording). You will submit the audio file as the deposit copy. If you register both the composition and the sound recording in the same filing, use Form SR and check the appropriate box — one registration can cover both if the same claimant owns both. **Timing matters.** Register within three months of publication, or before any infringement occurs. "Publication" includes posting publicly online. If you post a riff to Instagram and then register six months later, you lose the statutory damages window for anything that happened before registration. Processing time currently runs three to twelve months for online applications. The effective date of registration is the day the Copyright Office receives a complete application, not the date of the certificate. So file early. > "The Copyright Office registration is your legal anchor. Your source files, agreements, and dates are the evidence around it." > > — Jason Colapietro, Suede AI ### Step three: preserve a verifiable evidence bundle Copyright Office registration creates a legal record. You should also preserve the source material that shows how and when the riff was made: the original recording, DAW session, scratch mixes, agreements, and file metadata. A content hash is a deterministic fingerprint: the same file produces the same hash. Keeping the hash alongside the original file can help show that the evidence bundle has not changed. A hash alone does not establish authorship or replace legal registration. For the AI context, preserve the original file and the records around publication and distribution. Future auditing methods may use content hashes as one signal, but a hash does not by itself prove that a model trained on a work or that a later output infringes it. This does not replace Copyright Office registration. The practical workflow is: - Record the riff. - Preserve the original file, session history, metadata, and a content hash. - File with the Copyright Office within three months of publication. - Keep the evidence bundle and registration certificate together. ### What this looks like in practice You write a guitar intro. It is two bars, eight notes, a specific rhythmic pattern that has a signature feel. You have heard similar progressions before, but this specific sequence, at this specific tempo, with this specific rhythmic placement, is yours. You record it to your interface, preserve the original file and session history, compute a content hash, and file with the Copyright Office before you post it anywhere. If a later dispute arises, you have a starting point: the legal certificate, the dated source material, and a content hash that helps demonstrate file integrity. Without any of those, you have a complaint and no evidence. The music industry has been running on complaints without evidence for seventy years. The outcome of that approach is documented. Suede's IP Registry is live at [registry.suedeai.ai](https://registry.suedeai.ai), with registration on Base and Avalanche mainnet. A record stores a file fingerprint, a wallet-signed creator claim, contributor metadata, and a public timestamp. It records provenance; it issues no licenses, settles nothing, and is not proof of copyright ownership. ### Further Reading See also: [Rights Metadata Is the Dark Matter of the Creative Economy](https://strumly.suedeai.ai/guides/rights-metadata-dark-matter), [The Fender Stratocaster Lawsuit, Explained](https://strumly.suedeai.ai/guides/fender-stratocaster-lawsuit), [Who Owns the Output Stage](https://strumly.suedeai.ai/guides/who-owns-the-output-stage). ## Impedance and the First Three Feet Source: https://strumly.suedeai.ai/guides/impedance-and-the-first-three-feet Where to put a buffer pedal, and why the first three feet of cable from your guitar do more to your tone than everything else on the pedalboard. Category: Electronics. Published 2026-05-24, updated 2026-07-16. By Jason Colapietro. One common and cheap-to-fix source of tonal degradation is the unbuffered cable run between the instrument and the first pedal. ### The Problem A passive guitar pickup has an output impedance that swings with frequency, typically peaking somewhere between 5 kΩ and 25 kΩ depending on pickup design.[^1] Any capacitance downstream forms a low-pass filter with that source impedance. The cable is the dominant capacitance in the chain. Common instrument cables often sit around 30–80 pF/ft, premium low-capacitance cables often sit around 12–25 pF/ft, and high-capacitance runs can exceed 100 pF/ft. A 20-foot cheap cable presents roughly 600 pF of shunt capacitance. Combined with a typical humbucker source impedance, this places the rolloff knee in the audible top end, sometimes as low as 4 kHz. ### The Math The treble rolloff frequency is approximately: ``` f₋₃dB ≈ 1 / (2π · Rₛ · C) ``` For a 15 kΩ source and 600 pF cable: ``` f ≈ 1 / (2π · 15,000 · 600e-12) ≈ 17.7 kHz ``` That looks safe. The trap is that the pickup is *not* a resistor. It is an LCR resonant network. The audible "sparkle" of a Stratocaster bridge pickup sits on top of a resonant peak somewhere between 3 and 6 kHz. A long cable's capacitance combines with the pickup's own inductance to drop that peak's amplitude and shift it lower in frequency, where it loses the perception of clarity.[^2] ### What a Buffer Does A buffer is a unity-gain amplifier with high input impedance (typically 1 MΩ) and low output impedance (typically under 1 kΩ). Place one early in the chain and: - The pickup sees only the buffer's 1 MΩ load, so there is no audible high-end loss - Downstream cable capacitance now interacts with the buffer's low output impedance instead of the pickup, so the same 600 pF cable now has its rolloff knee well above 100 kHz In other words: the buffer moves the bottleneck out of the audio band. ### When *Not* to Buffer Buffers are not universally good. Specific cases where you want the pickup running into the next device directly: - **Vintage germanium fuzz**: Fuzz Face and friends rely on pickup interaction for their volume-knob cleanup behavior. A buffer flattens this. - **Treble-bleed-style volume controls**: these are tuned for the specific impedance the pickup presents. Buffering changes the math. - **Guitarists who deliberately want darker, woolier top end**: sometimes that rolloff is the sound. ### Practical Rule of Thumb - Less than 10 ft of cable, premium quality: passive run is fine - 10–20 ft, mixed-quality cable: buffer at the front of the board - More than 20 ft, or running into a switcher with long internal paths: buffer at both the pickup end and downstream ### Recommended Topology Tuner pedals often have input buffers that remain active. Placing a tuner first solves the problem incidentally, provided the tuner's buffer is reasonable. Most modern ones are. ### Quick Diagnosis | Symptom | Likely cause | First check | Fix | |---|---|---|---| | Bright straight into the amp, dull through the board | Cable capacitance loading the pickups | Measure total cable length before the first buffer | Put a good buffer first or shorten the unbuffered run | | Fuzz loses cleanup when the tuner is first | Buffer before germanium fuzz | Move the fuzz before the tuner for one test | Put germanium fuzz before the first buffer | | Top end returns when the guitar volume is full up | Pickup and volume-pot loading | Compare 250 kΩ vs 500 kΩ guitar behavior | Match pot value, cable length, and first buffer to the pickup | ### Further Reading See also: [Signal Chain Topology](https://strumly.suedeai.ai/guides/signal-chain-topology) for how this block fits the full chain. [Gain Staging Across the Chain](https://strumly.suedeai.ai/guides/gain-staging-across-the-chain) for downstream level management. --- [^1]: Source impedance depends on the pickup's winding, DCR, and the frequency of interest. Treat published "DC resistance" as a rough proxy, not a number to design against. [^2]: This is the classic argument for why short, low-capacitance cables sound "brighter" than longer ones. The effect is real and measurable. ## Pedalboard Order: A Practical Methodology Source: https://strumly.suedeai.ai/guides/pedalboard-order-methodology Where each pedal belongs on your board: a practical guitar signal-chain order for dynamics, drive, modulation, delay and reverb, and when to break it. Category: Signal Chain. Published 2026-01-15, updated 2026-07-16. By Jason Colapietro. The canonical signal chain map, dynamics, drive, modulation, time, and amp, is a starting point, not a law. Understanding why the order exists is what tells you when to follow it exactly and when to break it with intention. ### Why the order exists Each stage in a signal chain expects a certain kind of signal. A compressor works best on the unprocessed dynamics of your picking hand. A fuzz circuit reacts to high source impedance from your guitar's pickups in ways it doesn't react to the low-impedance output of a buffer. A delay repeating a saturated preamp tone sounds different from a delay processing a clean guitar signal before the amp. The canonical order describes what sounds best to most guitarists most of the time. It is a description of observed results, not a specification handed down. The reason to learn it is that understanding the cause lets you deviate with purpose. ### Dynamics first: compressor and wah **Compressor:** Before drive, the compressor works on your actual picking dynamics. Heavy attack gets clamped; light playing gets lifted. The drive stage then receives a more consistent level, which makes it tighter and more controlled. A compressor placed after drive compresses an already-saturated signal. This is useful for peak control (taming spikes going into a power amp or PA), but it does not shape playing feel the way front-of-amp compression does. **Wah:** Before drive produces the classic rock and funk wah sound. The filter sweeps an unprocessed guitar signal, then the drive colors the swept signal. After drive, the wah sweeps an already-saturated signal. The resonant peak is smoother, less sharp, and more modern. Both positions produce working sounds. Know which one you want. ### Drive in the middle Overdrive, distortion, and fuzz usually belong after initial dynamics shaping and before time-based effects. The exception is placing a drive pedal in an amp's effects loop, covered separately in the effects loop guide. **Germanium fuzz:** Before the first buffer and before most other pedals. The fuzz input circuit uses your pickup's impedance as part of its tone network. Buffering before a germanium fuzz changes the cleanup and input response. **Silicon fuzz and modern distortion:** Less impedance-sensitive. More flexible positioning. **Stacking drives:** The first drive in the chain sets the character the second drive amplifies and reacts to. A mild overdrive feeding a distortion sounds different from the same distortion feeding the same overdrive. There is no rule for stacking order. Test the order deliberately. ### Modulation after drive Chorus, phaser, flanger, tremolo, and vibrato belong after your drive pedals. You want modulation applied to the shaped, driven tone, not to the raw signal that will then be driven. Chorus before drive: the modulated pitch variations get amplified and saturated alongside the fundamental. The result is thick and indistinct. Chorus after drive: the modulation moves on top of the already-formed tone. It's audible as a separate effect rather than embedded in the saturation. **Tremolo exception:** Tremolo before drive can produce a more aggressive, stuttering interaction. The amp or gain pedal responds differently to volume drops than to a sustained note. This is a legitimate creative choice. It is not the default. ### Time-based effects last Delay and reverb belong at the end of the chain, or in the amp's effects loop. You want the delay to repeat the complete, shaped tone, not to feed repeated notes back into your gain stage where they accumulate and saturate. **Delay before reverb vs. reverb before delay:** Delay into reverb: each delayed repeat receives its own reverb tail. The result can become a wash quickly. Reverb into delay: one ambient space, with discrete repeats emerging from it. That is cleaner and more controlled. Most guitarists prefer delay before reverb. Both work. Know what each sounds like. ### Pitch and octave: the positioning exception Pitch shifters and octave pedals track your fundamental frequency. Their accuracy depends on receiving a clear, unambiguous signal. **Before drive:** The shifter tracks a clean guitar signal. Pitch detection is accurate. Both the original pitch and the shifted pitch then go through your drive together. **After drive:** The shifter attempts to track a harmonically complex, saturated signal. Tracking degrades, with more tracking errors, less pitch accuracy, and worse results on chords. **Rule:** Pitch shifters go before drive, with the narrow exception of effects specifically designed to be driven (Octavia, vintage Boss OC-2, certain fuzz-octave circuits built around intentional instability). ### Volume pedal placement **Before drives:** Controls the level entering your gain stage. Backing off cleans up the drive because the pedal or amp responds to a lower input level with less saturation. This mimics rolling back your guitar's volume knob. **After drives, before time effects:** Controls the output of the drive section feeding your delays. Volume swells emerge into a full saturated tone with trailing delay. Common for ambient and post-rock applications. **In the effects loop:** Controls master output volume without affecting preamp saturation. Useful for controlling stage volume while leaving your amp's gain character intact. ### Building a specific board **Start with three pedals.** Compressor, drive, delay, or the three you actually use most. Learn exactly how they interact at this scale before adding anything. **Add one pedal at a time.** When a new pedal enters the chain, you're changing one variable. Test it in the canonical position first. Then test it elsewhere. Understand what changes. **Default to the canonical order. Break it with a reason.** Putting tremolo before your fuzz because you heard it on a recording is a reason. Putting it there without knowing why produces sounds you can't predict or reproduce. ### Try This: Order A/B Play the same bar through two orders. Keep every knob still except the order. ```text e|----------------| B|----------------| G|--7---7---5-----| D|----7---7---7-5-| A|----------------| E|----------------| ``` | Pass | Order | What changes | |---|---|---| | A | Compressor, drive, delay | Pick dynamics tighten before the gain stage | | B | Drive, compressor, delay | Sustain rises, but the drive texture flattens | | C | Drive, delay, reverb | Repeats stay distinct before the reverb tail | | D | Delay, drive, reverb | Repeats saturate and blur into the gain | ### Quick Diagnosis | Symptom | Likely cause | First check | Fix | |---|---|---|---| | Fuzz feels stiff | Buffer or pedal before a germanium fuzz | Put fuzz first for one pass | Keep impedance-sensitive fuzz at the front | | Modulation disappears | Chorus or phaser is before heavy gain | Move modulation after drive | Modulate the finished drive tone | | Pitch pedal glitches on chords | Tracker is after distortion | Move pitch before drive | Feed the pitch detector a cleaner signal | ### Further Reading See also: [Signal Chain Topology](https://strumly.suedeai.ai/guides/signal-chain-topology) for the full five-block map. [The Effects Loop](https://strumly.suedeai.ai/guides/effects-loop-serial-vs-parallel) for routing time effects after preamp gain. The canonical order is a sensible default for a reason. What you build on top of it is your chain, for your sounds. The methodology is: understand the default, understand the cause, deviate deliberately. ## Power Supply and Noise Floor: The Hidden Variable in Your Tone Source: https://strumly.suedeai.ai/guides/power-supply-and-noise-floor Which pedal power supply spec sets your noise floor: isolated outputs versus daisy chains, current draw, and how to track down and fix ground loop hum. Category: Power. Published 2026-04-20, updated 2026-07-16. By Jason Colapietro. Your power supply is often the least discussed piece of gear on a pedalboard and one of the first places to check when noise does not respond to EQ. A mediocre supply does not change your frequency response or your dynamics. It puts noise under the whole board until you fix it or stop noticing it. ### What noise sounds like Understanding what you're hearing narrows the diagnosis: - **60 Hz hum** (low, buzzy drone): classic ground loop. AC mains frequency finding a path through your signal chain. - **120 Hz buzz** (slightly higher, more "electrical"): AC rectification artifacts, common with unregulated or cheaply filtered supplies and daisy chains. - **High-frequency whine or clicking**: switching power supply noise bleeding into the audio path. It usually originates from digital pedals such as delays, loopers, and reverbs with microprocessors. ### Ground loops: the primary culprit A ground loop forms when two or more pieces of equipment share a ground at different electrical potentials. Current flows between them to equalize those potentials, and that current couples into the audio signal as audible hum. Classic setup: your amp plugged into one outlet, your pedalboard power supply into another on a different circuit. A signal cable connects them. If those two circuits have different ground references, which happens constantly in older buildings, rented venues, or with long cable runs, current circulates through your cable's shield and adds hum. Isolated power supplies reduce inter-pedal ground loops by keeping each pedal's power path separate. They do not fix a ground loop between your amp and the wall, but they reduce board-side loops and shared-rail noise. ### Daisy chaining: shared rail, shared noise A daisy-chain cable splits one supply output across multiple pedals. Every pedal on the chain shares the same power rail and the same ground reference. If any one pedal generates noise on that rail, from a digital circuit, poor internal filtering, or a failing capacitor, that noise appears on the shared rail and leaks into every other pedal connected to it. Digital pedals are the worst offenders. Delays, reverbs, and loopers running microprocessors generate high-frequency switching noise that bleeds through a shared daisy chain into analog pedals sitting next to them. > The minimum fix: separate digital pedals from analog pedals onto different supply outputs. Ideally, stop daisy chaining entirely. ### Isolated outputs: what they actually mean A truly isolated power supply contains separate transformer windings or individual DC-DC converter modules for each output. Each pedal draws from its own independent source with its own ground reference. There is no shared rail for noise to travel between pedals. This reduces inter-pedal ground loops and prevents digital switching noise from contaminating analog supply rails. It is the correct fix for any board mixing digital and analog pedals. All professional-tier supplies use isolation: Strymon Zuma, Cioks DC7, Eventide PowerMax, Voodoo Lab Pedal Power 2+. The affordable single-transformer designs with multiple tapped outputs are not isolated. They share a rail. ### The specs that matter **Current (mA):** Every pedal specifies a current draw. Your supply must provide at least that much per output. Running at or near the rated limit causes voltage sag and increased noise. Budget 20–30% headroom above each pedal's rated draw. **Voltage:** Most pedals require 9V DC. Some use 12V or 18V. Some designs benefit audibly from higher voltage: increased headroom, less compression at the power stage. Always verify the pedal's specifications before using a non-9V output. **Polarity:** Nearly all modern pedals use center-negative, 2.1mm barrel jacks. Center-negative means the tip of the barrel connector is negative and the sleeve is positive. Reversing polarity can damage a pedal immediately. Vintage gear may differ, so verify before connecting. **Regulated vs. unregulated:** Regulated supplies maintain constant output voltage regardless of load variation. Unregulated supplies sag under load, which some vintage-voiced pedals are designed around but most modern pedals are not. Default to regulated. ### Practical recommendations by rig scale **1–4 analog pedals, short chain:** A quality single-brick supply (One Spot Pro CS6 or similar) with one pedal per output is adequate if no digital pedals are present. Avoid daisy chains. **5–8 pedals, mixed analog and digital:** Isolated supply required. Voodoo Lab Pedal Power 2+ is the minimum. Route digital pedals (delay, reverb, looper) to their own isolated outputs, separated from your drives and filters. **Full board (8+ pedals):** High-current isolated supply with multiple voltage options. Cioks, Strymon Zuma, Eventide PowerMax. Plan outputs before purchasing because high-current isolated outputs are the limiting resource. ### Safety Boundary Do not lift mains safety ground to chase hum. Use proper isolation, power conditioning, or a qualified tech for amp-to-wall grounding problems. ### Quick Diagnosis | Symptom | Likely cause | First check | Fix | |---|---|---|---| | Low hum appears only with the full board | Ground loop or shared supply path | Run one pedal at a time from isolated outputs | Separate power paths and check the amp outlet | | Digital whine appears when delay is powered | Switching noise on shared rail | Put the digital pedal on its own output | Use isolated high-current output | | Pedal sounds compressed or starved | Current draw exceeds output rating | Compare pedal draw to output mA | Use an output with 20-30% current headroom | ### Further Reading See also: [Gain Staging Across the Chain](https://strumly.suedeai.ai/guides/gain-staging-across-the-chain) for separating noise from level problems. [Signal Chain Topology](https://strumly.suedeai.ai/guides/signal-chain-topology) for where power noise enters the full rig. A clean power supply resolves an entire category of problems permanently. It costs less than most of the pedals it powers. ## Rights Metadata Is the Dark Matter of the Creative Economy Source: https://strumly.suedeai.ai/guides/rights-metadata-dark-matter Rights metadata decides who gets paid, and almost nobody can see it. What it is, why it breaks, and what a working infrastructure layer looks like. Category: Rights & IP. Published 2026-06-14, updated 2026-07-16. By Jason Colapietro. It governs everything. Almost nobody can see it. Here's what it is, why it breaks, and what a working infrastructure layer looks like. There's a concept in physics called dark matter. It doesn't emit light. You can't see it directly. But it governs the structure of everything around it — galaxies don't form correctly without it, orbital mechanics break down, the math doesn't close without it being there. Rights metadata is the dark matter of the creative economy. You can't see it on the streaming platform. It doesn't appear in the credits when a song plays. It's not in the Spotify interface, not in the YouTube description, not in the TikTok audio clip. But every payment decision, every licensing deal, every royalty route, every sync placement — all of it runs through this invisible layer. When it's wrong, nothing works. When it's missing, no one gets paid. And most of the time, it's wrong. ### What rights metadata actually is When a song is created, several facts need to be recorded: - Who wrote it (the composition copyright) - Who performed it (the sound recording copyright) - Who produced it - Which publisher holds which percentage - Which PRO administers which writer share - What version this is (original, remix, cover, interpolation) - Whether any samples were cleared, and if so, from what - What license terms apply to sync, broadcast, streaming, or agent use This information is rights metadata. It's not glamorous. It sounds like bookkeeping. But it's the reason a track generates a royalty check instead of a query letter from a music publisher's legal team. The problem is that there is no canonical record. In the United States, copyright registration is optional. PRO registration is required for royalty collection but separate from copyright. Publishing rights are tracked by each publisher independently. Sound recording rights are tracked by labels — or not tracked at all for independent artists. A single song can touch a dozen separate databases that don't talk to each other. ASCAP doesn't automatically talk to Harry Fox. DistroKid doesn't automatically reconcile with BMI. The streaming platforms don't require metadata that resolves to a payable entity — they require metadata that allows them to play the track. Those are different things. > "The signal chain starts at the pickup and ends at the listener's ear. The IP chain starts at creation and ends at the bank account. Miss a link in either chain and you lose the signal." > > — Jason Colapietro ### The gap between generating and owning There is a gap in the creative economy that almost nobody talks about clearly. It's the gap between the moment you make something and the moment you own it in a form that can be licensed, monetized, and defended. For a self-taught musician — someone who figured out the guitar from YouTube videos and started recording in their bedroom — this gap is vast. They know how to make music. They have no map for what happens next. And the systems that are supposed to help them are designed for an industry that assumes you have a manager, a lawyer, and a publishing deal before you're releasing music. Most creators don't. And the creators who do have that infrastructure often find that the metadata their team submitted years ago is wrong, incomplete, or stored in a spreadsheet on someone's old laptop. This is not a technology problem. It's an infrastructure problem. And it's specifically an infrastructure problem for metadata. ### What AI made worse — and what it could fix Generative AI didn't create the rights metadata problem. The problem is older than streaming. But AI made it worse in one very specific way: it collapsed the time between creation and distribution. Before AI, a song took time to make. That time created friction — and friction, annoying as it is, is also a forcing function. You had to deal with the metadata questions at some point because the release process required it. AI tools generate music in seconds. The release pipeline is increasingly automated. The metadata questions get skipped entirely because there's no moment where someone is forced to stop and answer them. And if you're training AI on creative work, the rights metadata problem compounds in the other direction. The training data often carries no ownership information at all. The model learns from a corpus where provenance was stripped at the point of scraping. And the outputs of that model are released into the world with no chain back to the inputs. Rights metadata isn't just a creator problem anymore. It's a structural failure in the data layer of an entire industry. ### What working infrastructure looks like A working infrastructure layer needs to do four things: **1. Registration at the moment of creation.** Not weeks later when you're trying to pitch a sync. At the moment the work exists. This is a timestamp problem — who had what, when — and it's solvable with tools that exist today: content hashing, registry-backed records, provenance packaging. **2. Rights metadata attached to the work, not stored separately.** The current model treats the creative file and its rights information as two separate things. A working layer keeps the rights data attached to the work — or at minimum, keeps a verifiable pointer from the work to its canonical rights record. **3. License terms that can be read programmatically.** A PDF license agreement is not machine-readable. A working layer expresses rights in a form that platforms, agents, and systems can check at query time. This is the programmable IP concept: a license contract compressed into a format that travels with the asset. **4. Routing that reaches creators, not just rights holders of record.** The reason royalties disappear is not that the money stops existing. It's that the money can't find the person it belongs to, because the metadata that would route it is wrong or missing. ### What this means if you're a working guitarist today You don't need to wait for the industry to fix its metadata problem. You need to build a personal rights infrastructure that is independent of whoever distributes your work. - **Register your work.** US Copyright Office registration is $65 for a single work. It creates a legal record that predates any dispute. - **Create a provenance file for every release.** Who wrote it, who owns what percentage, what version this is, what samples or interpolations are present. - **Use a PRO.** If you're in the US, join ASCAP, BMI, or SESAC. This is required for broadcast royalties. - **Keep your own metadata.** Don't rely entirely on the streaming platform's metadata fields. They are designed for playback, not for rights resolution. > "Rights metadata is the dark matter of the creative economy. It governs everything. Almost nobody can see it." > > — Jason Colapietro, Suede AI ### The longer view The AI era is not going to make the rights metadata problem smaller. It's going to make it bigger, faster, and harder to resolve retroactively. The creators who understand this now — who treat registration and provenance as a baseline practice, not an afterthought — are building something that will matter enormously over the next decade. The dark matter of the creative economy doesn't have to stay invisible. It can be registered, attached, and routed. The infrastructure to do this exists or is being built. I wrote the full history of how the signal chain leads to this moment — from Leo Fender's bench to programmable IP — in [The Signal Chain](https://guitar.solutions), free at guitar.solutions. ### Further Reading See also: [Who Owns the Output Stage](https://strumly.suedeai.ai/guides/who-owns-the-output-stage), [How to Register a Guitar Riff as Intellectual Property](https://strumly.suedeai.ai/guides/how-to-register-a-guitar-riff), [AI Training and Your Music: What Every Guitarist Needs to Know](https://strumly.suedeai.ai/guides/ai-training-and-your-music). ## Signal Chain Topology: What Actually Goes Where, and Why Source: https://strumly.suedeai.ai/guides/signal-chain-topology What goes where in a guitar signal chain, from pickup to speaker, and the engineering reasons behind the standard order of pedals, amp and effects loop. Category: Signal Chain. Published 2026-05-26, updated 2026-07-16. By Jason Colapietro. The order of pedals in front of an amplifier is not a matter of taste. It is a series of engineering decisions about impedance, gain staging, dynamic range, and noise floor. This guide lays out the canonical topology, explains the reasoning behind each block, and flags the places where guitarists routinely make tone worse by reordering on intuition alone. ### The Five-Block Model Every guitar signal chain, whether analog, digital, hybrid, or modeler, can be reduced to five sequential blocks. The labels change. The physics does not. - **Source**: pickup output, instrument volume, tone control - **Front of Amp**: dynamics, gain, filtering before the preamp - **Preamp**: voltage gain, tone stack, the "voice" of the amp - **Effects Loop**: time-based effects placed after the preamp - **Power Section & Speaker**: current gain, speaker excursion, mic'd or DI'd output The five-block model holds whether you are running a Tube Screamer into a 1959 Plexi or routing a profile through a digital convolver. Skipping blocks does not eliminate them; it consolidates them into something you have less control over. ### Block 1: Source The instrument is a high-impedance, low-level voltage source. The pickup output sits between roughly 100 mV and 1.5 V peak depending on string excitation, pickup design, and electronics. Anything that loads the pickup excessively will lose treble before the first cable run is done.[^1] **Engineering decisions made here:** - Pickup type sets the impedance curve and resonant peak - Volume pot value (250 kΩ vs 500 kΩ vs 1 MΩ) shifts the high-frequency rolloff - Cable capacitance acts as a low-pass filter. Long, cheap cables audibly darken the signal. ### Block 2: Front of Amp Anything you put between the guitar and the preamp input belongs to this block. The order inside it is where most tone problems originate. The canonical front-of-amp order is: 1. **Buffer**: converts high-impedance pickup signal to low-impedance line 2. **Tuner**: needs a clean signal to track pitch 3. **Wah / Filter**: wants to see pickup-like impedance, hence early placement 4. **Compressor**: sets dynamic envelope before gain stages 5. **Overdrive / Distortion / Fuzz**: gain stages stack in the order placed 6. **EQ**: shapes the post-gain spectrum **Why fuzz lives early.** Vintage germanium fuzz circuits expect to see the guitar pickup's impedance directly. Putting a buffer in front of a Fuzz Face collapses its interactive volume-knob behavior into a flat clipper.[^2] ### Block 3: Preamp The preamp does three things: amplifies the signal from millivolts to volts, shapes it through the tone stack, and applies the first major nonlinearity. Most of what we describe as "the sound of the amp" lives here. > The preamp is not a transparent gain stage. It is a deliberately colored > voltage amplifier with intentional harmonic distortion, deliberate frequency > shaping, and a tone stack that interacts with all of the above. ### Block 4: Effects Loop Time-based effects such as delay, reverb, and modulation placed before the preamp get distorted along with the dry signal. This is sometimes desirable (slap delay into a cranked Marshall), more often not (reverb tails getting clipped into mud). The effects loop exists to place these after the preamp's gain and shaping but before the power section. A serial loop puts effects in the audio path entirely. A parallel loop blends a wet signal back with a dry feed. Pick based on whether you want the effect to replace or supplement the dry tone. ### Block 5: Power Section and Speaker The power amp adds current gain and contributes its own compression and harmonic content under load. The speaker is the final filter, typically rolling off above 5 kHz and below 80 Hz, and is mechanically reactive in ways that no modeler approximates without convolution. This block decides whether your tone holds together at performance volume. A great preamp into a flubby cab is a great preamp into a flubby cab. ### Where Guitarists Go Wrong The most common topology mistakes: - **Delay before overdrive**: turns repeats into a wash of distortion - **Buffered fuzz**: collapses the volume-knob interaction - **Compressor after gain**: squashes the harmonic content you just generated - **EQ as a band-aid**: covers up a bad pickup choice instead of fixing it upstream - **Reverb in front**: usually belongs in the loop ### Quick Diagnosis | Symptom | Likely cause | First check | Fix | |---|---|---|---| | Rig gets darker as pedals are added | Source loading before a buffer | Compare straight-in vs board-bypassed tone | Fix the source block before changing EQ | | Gain feels smeared | Too many stages are above unity | Bypass one gain stage at a time | Reset levels around the preamp input | | Ambience turns into mud | Time effects before main gain | Move delay and reverb after preamp gain | Use the loop or place time effects after drive | | Stage volume changes the feel | Power section and speaker are part of the tone | Compare bedroom and performance volume | Tune gain and EQ at the volume you use | ### Further Reading See also: [Impedance and the First Three Feet](https://strumly.suedeai.ai/guides/impedance-and-the-first-three-feet), [Gain Staging Across the Chain](https://strumly.suedeai.ai/guides/gain-staging-across-the-chain). --- [^1]: Pickup loading is a textbook RC low-pass problem. See the appendix in "Inductance, capacitance, and pickup response," to be published. [^2]: Roger Mayer and others have written extensively about the input impedance sensitivity of germanium fuzz designs. ## Smoother Chord Transitions: A Mechanical Approach Source: https://strumly.suedeai.ai/guides/smoother-chord-transitions Why chord changes stall and the four mechanics that fix them: anchor fingers, minimum motion, early formation and honest tempo, each with a drill. Category: Practice. Published 2026-07-18. By Jason Colapietro. Every guitarist has a change that won't cooperate. For beginners it's usually C to F or anything into G; for intermediates it's barre-to-open moves at speed. The frustrating part is that grinding repetitions often doesn't fix it — you can play a bad change five hundred times and mostly get better at playing it badly. Chord transitions are a mechanical problem, and mechanics respond to analysis better than to volume. Here are the four mechanics that govern nearly every transition, and how to train each one. ### 1. Anchor fingers: stop lifting what doesn't move Lay out the two chords and look for fingers that stay on the same string and fret — or the same string, different fret — across the change. - **C to Am:** the index (first fret, B string) and middle (second fret, D string) don't move *at all*. Only the ring finger travels. Most beginners lift all three fingers and rebuild both chords from scratch, tripling the work. - **Am to E:** the whole shape just shifts one string group. Same fingering geometry, different strings. - **G to C (the classic stall):** no full anchor exists, but the ring finger can act as a *guide*, sliding along the A string rather than jumping through the air. The drill: form chord one, then move to chord two **in slow motion** — five full seconds per change — watching your fretting hand and consciously keeping every anchor planted. Slow motion isn't a warm-up here; it's where the actual rewiring happens, because at full speed your hand replays whatever habit it already has. ### 2. Minimum motion: fingers hover, never fly Watch a good rhythm player's fretting hand and it barely seems to move. Fingers hover a few millimeters above the strings between chords. Beginners' fingers fly a centimeter or more off the fretboard, and every extra millimeter is time. Two sub-habits build this: - **Release pressure without releasing contact.** Between chords, let fingers unweight but stay touching (or nearly touching) the strings. Full lifts are almost never necessary. - **Move fingers as a group.** Practice forming the destination chord's shape *in the air* over the frets before landing it. If the shape arrives as a unit, landing is one motion; if fingers land one-by-one, each landing is its own timing error waiting to happen. The drill: "form and hover." Cycle between two chords without strumming at all — form chord one, unweight, morph the hand to chord two's shape while hovering, land. Ten cycles, then add the strum back. ### 3. Early formation: the change happens *before* the beat Here's the counterintuitive mechanic that measurement makes obvious: clean players don't change chords on the beat. They begin forming the next chord *during the previous beat* — often sacrificing the tail end of the old chord to do it. Listen to any strummed record closely and you'll hear it: the last eighth-note before a change is frequently open strings or a transitional smear, and it sounds fine, because the *next* chord lands exactly on the downbeat, which is what the ear actually judges. Beginners do the opposite. They hold the old chord to its last moment (it feels dutiful), then form the new one after the beat has passed. The result reads as "late" — and if you drill with scoring, it shows up as consistently positive timing offsets. This is exactly the failure pattern Strumly's drill engine surfaces: when every transition in a run lands +60 to +120 milliseconds, the diagnosis is almost always late formation, not slow fingers. The drill: strum a bar of chord one, but **leave the last upstroke as open strings** while your hand moves early to chord two. Deliberately, exaggeratedly early. It will feel like cheating. It's how the pros do it, and once the earliness is comfortable you can tighten it back until the transition is inaudible. ### 4. Honest tempo: the metronome is the rep counter The final mechanic is scheduling: transitions improve at the tempo where you *almost* fail, and nowhere else. Too slow and nothing is challenged; too fast and you're rehearsing sloppiness. A protocol that works, whether you run it with a [metronome](https://strumly.suedeai.ai/metronome) or a scored coach drill: 1. Find your **clean ceiling**: the highest BPM where you can hit the change eight times out of eight, on the beat. 2. Practice at **ceiling minus ~10%** for most reps — accumulating clean, confident volume. 3. Probe **above the ceiling** briefly each session — two or three attempts at +10 BPM, expecting misses. This tells your hands where they're going. 4. Raise the ceiling only on evidence: two consecutive clean runs at the new tempo. The discipline this protocol enforces is the one most self-taught players lack: tempo decisions made by results instead of by mood. It's also where measured drills beat self-assessment — your ear, occupied with playing, systematically overrates your timing. The millisecond offsets don't. A scored run in [a coach session](https://strumly.suedeai.ai/docs/practice-sessions) gives you the success rate and the lateness pattern per transition, which turns "I think it's getting better" into a number that either moved or didn't. ### Putting it together on a real change Take G to C, the most complained-about open change: 1. **Map it** (thirty seconds): ring finger guides along the A string, third fret to third fret. Index and middle cross to their new strings as a unit. 2. **Slow motion** (two minutes): five-second changes, anchor/guide planted, fingers hovering low. 3. **Early formation** (two minutes): bars of G with the final upstroke sacrificed, hand moving on the "and" before the change. 4. **Tempo ladder** (five minutes): find the ceiling, work below it, probe above it. 5. **Deploy in music** (three minutes): a song that uses G→C, played through without stopping regardless of misses. Songs are the test set; drills are the training set. Don't let the two blur. Twelve minutes, one change. Two or three sessions of this typically moves a stalled transition more than a month of unstructured strumming — not because the drills are magic, but because each rep is aimed at the specific mechanic that's actually failing. ## The Effects Loop: Serial, Parallel, and the Four-Cable Method Source: https://strumly.suedeai.ai/guides/effects-loop-serial-vs-parallel Serial or parallel effects loop, and when to use the four-cable method: how to wire your amp effects loop correctly and which pedals belong in it. Category: Gain & Dynamics. Published 2026-03-10, updated 2026-07-16. By Jason Colapietro. The effects loop is one of the most useful and most ignored features on a tube amp. Guitarists run their delay into the input, wonder why it sounds indistinct and washy, and conclude they prefer a dry amp. Often the loop is the missing piece. ### What the effects loop is Inside a tube amp, there are two major stages: the preamp and the power amp. The preamp receives your guitar's signal, applies your EQ and gain, and shapes the tone character. The power amp receives the preamp's output and delivers enough current to move a speaker. The effects loop sits between them. The Send jack taps the signal after the preamp has done its work. The Return jack injects signal back in directly before the power amp. A pedal placed in the loop processes the sound of your preamp, not the raw guitar. The tone shape, the gain character, and the amp EQ all enter the loop pedal's input. The loop pedal's output feeds the power amp directly. ### Why this matters for time-based effects Delay and reverb produce repeats and reflections that sound most natural when they're applied to an already-formed tone. If you run delay into the front of a driven amp, the amp's preamp processes both the original note and every one of its repeats. Each repeat is re-driven, re-compressed, and re-shaped. The delays become thick and indistinct. In the effects loop, delay receives the preamp's shaped output. Its repeats go directly to the power amp, bypassing the gain stage entirely. Each repeat is a clean copy of the processed note, not a re-driven version. The delays decay naturally with clear pitch definition. The same logic applies to reverb, chorus, and flanger. Applied after the gain stage, they create movement on top of a defined tone. Applied before, they feed into the gain stage and lose coherence. ### Serial vs. parallel effects loops A **serial loop**, the most common type, places your pedal completely in the signal path. The amp's signal goes entirely into the pedal's input, and the pedal's output returns entirely to the power amp. The pedal's wet/dry mix control determines the blend. A **parallel loop**, less common and found on some boutique and high-end amps, splits the signal into two paths: one through the loop pedal, one bypassing it. The amp blends them before the power amp stage using a Mix control. At 100% wet, you hear only the loop pedal. At 0%, you hear the dry amp path. Between them, you get both. The parallel loop is useful for reverb and delay because it preserves the amp's dry signal on one path while adding depth on the other. Some engineers prefer this because the dry path never touches a converter or a buffer. The tradeoff: the parallel loop's Mix control behaves poorly with drive pedals. You cannot blend a processed and an unprocessed signal cleanly when the processed signal contains heavy harmonic content. ### The four-cable method The four-cable method (4CM) routes a multi-effects unit or switcher so that some effects sit before the amp's preamp and others sit in the effects loop. The physical amp remains the gain stage in the middle. The routing: 1. Guitar → Multi-FX Input 2. Multi-FX Send → Amp Input 3. Amp Send → Multi-FX Return 4. Multi-FX Output → Amp Return With this wiring, effects assigned to the "front-of-amp" path in the multi-FX hit the preamp input. Effects assigned to the "loop" path process the preamp's output. Your physical amp's preamp stays in the signal path as the primary gain stage. This works with Fractal FM3, Line 6 Helix, Boss GT units, or a dedicated loop switcher. It gives you the real amp's preamp character while routing delay and reverb where they belong. ### Level matching Some effects loops operate at instrument level (-10 dBV). Others operate at line level (+4 dBu). A rack reverb designed for +4 dBu connected to an instrument-level loop will be quiet, noisy, or both. Check your amp's manual and your rack unit's specifications. Most tube amp effects loops are instrument level. Professional rack units often expect line level. A passive attenuator or a dedicated level converter resolves the mismatch. ### Practical placement guide - **Delay, reverb, chorus, flanger, phaser:** Start in the effects loop when the amp preamp supplies most of the gain. - **Compressor, wah, overdrive, fuzz, distortion:** In front of the amp. These shape the signal entering the preamp. - **Volume pedal for dynamics and cleaning up drive:** In front. Controls input level and preamp response. - **Volume pedal for master level control:** In the loop. Controls output without changing preamp saturation. - **Pitch shifters:** In front for tracking accuracy. The cleaner the input, the more accurate the pitch detection. ### Try This: Front vs. Loop Delay Use a short palm-muted phrase so the repeats are easy to hear. ```text e|----------------| B|----------------| G|----------------| D|--2-2---5-5-----| A|--2-2---5-5-----| E|--0-0---3-3-----| ``` | Pass | Placement | Settings | Expected difference | |---|---|---|---| | A | Delay before amp input | 350 ms, 25% mix, 3 repeats | Repeats hit the preamp and smear with gain | | B | Delay in the loop | Same settings | Repeats stay clearer because they bypass preamp distortion | | C | Reverb in the loop after delay | Small room, low mix | The repeat trail sits behind the driven tone | ### Quick Diagnosis | Symptom | Likely cause | First check | Fix | |---|---|---|---| | Delay repeats turn to mush | Delay is before a driven preamp | Move the delay into the loop for one pass | Keep time effects after the main gain stage | | Loop sounds thin or noisy | Level mismatch | Check -10 dBV vs +4 dBu on amp and pedal | Match levels or add a converter | | Parallel loop has phasey repeats | Wet/dry blend conflict | Set the pedal fully wet if possible | Let the amp handle the dry blend | ### Further Reading See also: [Pedalboard Order](https://strumly.suedeai.ai/guides/pedalboard-order-methodology) for what belongs before the amp input. [Signal Chain Topology](https://strumly.suedeai.ai/guides/signal-chain-topology) for where the loop sits in the full chain. The effects loop is a routing decision with audible consequences. Running reverb in the front is not wrong. It is a different sound. Understanding the difference is what makes the choice deliberate. ## The Fender Stratocaster Lawsuit, Explained Source: https://strumly.suedeai.ai/guides/fender-stratocaster-lawsuit What Fender won in its German Stratocaster body-shape ruling, the cease-and-desist letters that followed, and what it means for players and builders. Category: Rights & IP. Published 2026-06-11, updated 2026-07-16. By Jason Colapietro. Fender won a copyright ruling on the shape Leo drew in 1954, then started mailing cease-and-desist letters to the builders who kept it alive. One of them landed on John Mayer's guitar. ### What Fender actually won In March 2026, a court in Düsseldorf ruled that the Stratocaster body is not just a trademark but a "copyrighted work of applied art" under German and EU law. The case ran against a Chinese manufacturer that never showed up to defend itself, so Fender won by default. A judgment against an empty chair, but a real one: it makes selling or importing S-style copies across the EU a copyright matter, not just a branding dispute. Fender's law firm treated the ruling as a starting gun and began sending letters. ### The letters One reached LSL Instruments, a family shop in California that builds fewer than 500 guitars a year. The letter asked LSL to stop making S-style guitars, hand over sales data, destroy unsold stock, recall instruments already sold, and cover Fender's legal fees. LSL opened a GoFundMe to pay lawyers and wrote that the shop "poses no threat to them in any way whatsoever." Then the campaign reached PRS, and the target was the Silver Sky: John Mayer's signature model, confirmed by the Wall Street Journal in May 2026. PRS says it disagrees with Fender's assessment and intends to defend the guitar. ### Why the Silver Sky exists at all Here is the part that turns a legal story into a guitar story. Mayer spent the 2000s as the most famous Stratocaster player alive. He asked Fender for years to evolve the instrument with him, the feel of his early-sixties Strats with the flaws engineered out. He could not get it built. "I tried to do something like this with Fender," he said, "but I couldn't create enough enthusiasm within the company for my vision." He left in 2014. Paul Reed Smith picked up the phone, and two years later the Silver Sky shipped. The forums called it an overpriced Strat clone until players actually held one. The affordable SE version topped Reverb's year-end best-seller lists in 2022 and 2023, and Guitar World reports the Silver Sky has outsold the Stratocaster in recent years. Fender's answer to that scoreboard arrived this spring, on legal letterhead. ### What it means for players and builders If you own a partscaster or a boutique S-style, nothing on your wall changes. The pressure falls on small builders who sell into the EU, where importing a close copy now carries copyright risk, not just trademark risk. For some shops that is the difference between a product line and a closed door. Worth remembering: in 2009 Fender tried to register the Strat, Tele, and P-Bass body shapes as US trademarks and lost, in part because builders had copied those shapes in the open for half a century. The 2026 campaign is an attempt to re-run that fight through a friendlier law in a friendlier court. ### The bigger question Strip away the guitars and this is a fight about who owns a design, who gets to build on it, and what happens when a company reaches for lawyers instead of a workbench. That same question is landing on every creator right now as AI turns music, art, and identity into things that copy themselves. I wrote the full version of that argument, from Leo Fender's bench to the economics of AI, in a longer essay. [Read the full essay: Fender Is Trying to Kill John Mayer's Guitar. AI Companies Are Next. →](https://jasoncolapietro.substack.com/p/fender-is-trying-to-kill-john-mayers) ## True Bypass vs. Buffered: What Your Pedal Actually Does Source: https://strumly.suedeai.ai/guides/true-bypass-vs-buffered True bypass versus buffered pedals: which one preserves your guitar tone, how cable length decides it, and which pedals on your board need a buffer. Category: Electronics. Published 2026-05-15, updated 2026-07-16. By Jason Colapietro. Every guitar pedal makes a decision when it is switched off: does it put your signal straight through a wire, or does it keep running it through a circuit? That choice, true bypass vs. buffered bypass, is one of the most argued and most misunderstood topics in guitar gear. The argument is mostly settled. Here is why. ### What the switch actually does In a true bypass pedal, the footswitch routes your signal through a mechanical relay or switch connecting input directly to output. The pedal's circuit is completely removed from the path when bypassed. Step on it and the relay clicks; the effect engages. Step off and the wire is the entire pedal. In a buffered bypass pedal, the pedal's input buffer stays active even when the effect is switched off. Your signal enters the circuit, gets buffered (impedance-converted), and exits. The wet effect is bypassed, but the buffer remains in the path. ### The case for true bypass True bypass advocates have a real point: a bypassed signal should ideally touch only wire. No active stages, no op-amps, and no extra buffer coloration. A quality true bypass pedal removes its effect circuit from your signal when bypassed. This matters especially for pedals with noisy or reactive input stages. A mediocre buffer is worse than no buffer at all. True bypass eliminates that variable entirely. ### The problem with true bypass at scale Cable is a capacitor. Guitar cable has a capacitance rating, typically 25–100 pF per foot. When that cable sits between your high-impedance pickup (8–15 kΩ output) and the next thing in your chain, it forms a passive low-pass filter. The longer the cable run, the more high frequencies attenuate. Ten feet of cable is barely noticeable. Fifteen feet to your first pedal, plus patch cables between six true-bypass boxes, plus another fifteen feet to your amp, means you are running 30–40 feet of capacitance against a 10 kΩ source impedance. Your highs are measurably rolling off before the signal enters any effect. True bypass does not help this. It removes the buffer that would fix it. ### The case for buffering A buffer, a unity-gain circuit with very low output impedance, solves the cable problem completely. It accepts your pickup's 10 kΩ signal and outputs the same signal at 100–300 Ω. From that point, cable capacitance is acoustically irrelevant because the source impedance is negligible. This is why Boss has used buffered bypass for decades and why serious rigs use a dedicated buffer at chain start. A well-designed buffer is inaudible. A chain of true-bypass pedals plus long cable runs is not. ### The germanium fuzz exception Vintage germanium fuzz pedals, including Fuzz Face, Tone Bender originals, and their modern derivatives, expect to see the raw impedance of your guitar's pickups. Their input circuit uses your pickup's inductance and the cable capacitance as part of the tone-shaping network. Put a buffer before a germanium fuzz and it sounds wrong: thinner, stiffer, less organic. > A buffer before a wah lowers cable loading. A buffer before a germanium fuzz changes cleanup and input interaction. **Rule:** If you use a germanium fuzz, it goes first in your chain with no buffer before it. Silicon fuzz pedals are less sensitive and more flexible. ### Practical guidance - **Long total cable runs (>20 ft):** Buffer early. A clean, transparent buffer at chain start, or your first buffered-bypass pedal, solves this permanently. - **Short runs (<15 ft) with few pedals:** True bypass is fine. Capacitance loading is low enough not to matter. - **Germanium fuzz in the chain:** First pedal, before any buffer when possible. - **Stacking many true-bypass pedals:** The capacitance accumulates. Six true-bypass boxes plus cables between them adds up. One buffer at chain start fixes this. - **Boss-heavy rig:** Boss's JFET buffers are well-designed and largely transparent. An all-Boss chain is correctly buffered by default. ### Quick Diagnosis | Symptom | Likely cause | First check | Fix | |---|---|---|---| | Board sounds darker when every pedal is off | True-bypass chain has too much cable capacitance | Count cable length before first buffer | Put one good buffer near the front | | Fuzz cleanup disappears | Buffer is before a germanium fuzz | Put fuzz first and retest guitar volume cleanup | Keep that fuzz before the buffer | | One buffered pedal sounds worse than bypass | The buffer itself is noisy or poorly voiced | Swap in a known clean buffer | Use a better buffer or move it later | ### Further Reading See also: [Cable Capacitance and Frequency Response](https://strumly.suedeai.ai/guides/cable-capacitance-and-frequency-response) for the math behind long cable runs. [Impedance and the First Three Feet](https://strumly.suedeai.ai/guides/impedance-and-the-first-three-feet) for the first-buffer decision. The debate is not bypass type vs. bypass type. It is about whether you have a good buffer and where it sits. A great buffer is inaudible. An absent one with 30 feet of cable is not. ## What Happens to Your Guitar Samples in AI Training Source: https://strumly.suedeai.ai/guides/guitar-samples-ai-training Your guitar recordings are likely in AI training sets. How the scraping pipeline works, what rights you have now, and what an ownership layer changes. Category: Rights & IP. Published 2026-06-23, updated 2026-07-31. By Jason Colapietro. AI music models are trained on real recordings — yours included. Here's what the pipeline looks like and what your rights actually are. If you have ever posted a guitar riff to SoundCloud, uploaded a demo to Bandcamp, or dropped a loop pack on any royalty-free marketplace, that audio is almost certainly in an AI training dataset somewhere. Not maybe. Not hypothetically. The pipelines that scrape audio for AI training do not ask for permission, and most terms of service that platform users agree to explicitly permit it. This is not a conspiracy. It is an engineering reality. And understanding how it works — specifically how your guitar recordings travel from your hard drive into a foundation model — is the first step toward knowing what, if anything, you can do about it. ### How audio ends up in training data The pipeline typically works in four stages: **1. Scraping.** Large-scale crawlers pull publicly accessible audio from streaming platforms, music hosting services, sample libraries, and archival collections. The scraper does not distinguish between commercially released work, bedroom recordings, and free loops. If the URL is public and the file is audio, it goes into the pipeline. **2. Filtering.** The raw scrape is filtered for audio quality, format compatibility, and sometimes genre or instrument classification. Guitar-heavy content is particularly useful for training generation models because it is abundant, stylistically diverse, and carries clear tonal signatures that the model can learn to replicate. **3. Stripping.** Metadata is often stripped or truncated at this stage. The file name, the ID3 tags, the embedded album art, the original uploader's name — all of it may be dropped before the audio reaches the training loop. This is where provenance breaks. The model learns from your playing, but the chain back to you as the creator is severed. **4. Training.** The processed audio is converted to spectrograms or audio tokens and fed into the model. The model does not store your specific recording. It encodes the statistical patterns: the way a particular picking attack decays, how a stacked-humbucker midrange compresses, the timing feel of a certain strumming pattern. That signature — your signature — becomes part of the model's learned capabilities. Without attribution, without compensation, without notice. > "The model doesn't store your recording. It stores the pattern. But the pattern came from somewhere, and that somewhere is you." > > — Jason Colapietro ### What the law currently says Copyright law protects expression, not style. You own the specific recording you made. You own the specific composition you wrote. What you do not own is the general style of playing, the tonal characteristic of your guitar, or the statistical pattern of how you pick a note. This is not a new doctrine. It predates AI by decades. Sampling cases in the 1990s established that taking a specific recording without license is infringement. Style-copying cases — George Harrison, Led Zeppelin's "Stairway to Heaven" dispute — largely failed because courts drew a hard line between the specific expression and the general style. AI training sits in this contested middle ground. If a model was trained on your copyrighted recordings without a license, there is a plausible infringement argument at the training stage. Several major lawsuits are currently working through this question. As of 2026, there is no settled precedent. Courts are actively wrestling with whether ingestion of copyrighted audio for model training constitutes fair use. For independent guitarists, the practical reality is this: you have a copyright in your recordings from the moment they are fixed in a tangible medium. That copyright exists whether you register it or not. But enforcing an unregistered copyright is significantly harder. Statutory damages — the provision that makes copyright suits economically viable against large defendants — are only available for works registered before the infringement or within three months of publication. ### The sample pack problem is different If you have sold or distributed loop packs, sample libraries, or royalty-free guitar content, the analysis shifts. The license you offered buyers may have been broad enough to permit AI training use. Many royalty-free licenses include language about "any purpose," "commercial use," and "modification" that a determined AI company could argue covers training ingestion. This was not what most sample creators intended when they wrote those licenses in 2015. It is increasingly what the licenses technically say. If you have existing sample libraries available under broad royalty-free terms, the moment to review those licenses was yesterday. The moment after that is now. > "Royalty-free has never meant free for every possible use. AI training is a use that most 2015 license templates never contemplated." > > — Jason Colapietro, Suede AI ### What an ownership layer changes The underlying problem is that provenance breaks at ingestion. If your recording carries no machine-readable ownership claim — no registry record, no content hash linked to you, no metadata that survives stripping — then by the time the model is trained, there is no evidence connecting your work to the resulting capability. A working ownership layer changes this at the source. When a recording is registered with a verifiable content hash at the moment of creation, the hash travels with the work. If the audio is scraped, the hash is scrapeable too. If the hash is stripped, the stripping itself becomes part of the evidence chain. Future auditing tools — and regulators are beginning to require them in some jurisdictions — can use those hashes to trace what was used and who owned it. This does not prevent the scraping. It does not retroactively fix training datasets that are already built. What it does is establish a provenance record that survives the pipeline and can anchor a licensing or compensation claim. ### What you can do right now - **Register your recordings with the US Copyright Office.** SR registration covers sound recordings. $65 for a single work, $55 for a group registration of up to 10 unpublished works. This creates the legal prerequisite for statutory damages. - **Review any existing royalty-free licenses.** If you have sample packs or loops available under broad terms, consider whether an updated license with explicit AI training carve-outs is appropriate. - **Embed ownership metadata in your audio files.** ID3 tags, XMP metadata, and broadcast wave chunks can carry copyright notice, ISRC codes, and contact information. None of this is legally binding on its own, but it establishes intent and can survive in partial scrapes. - **Keep a timestamped evidence bundle.** Store the original recording, session files, ownership metadata, agreements, and a content hash together. That bundle gives a future ownership or attribution discussion something concrete to examine. The guitar samples going into AI training today are building the sonic vocabularies that models will deploy for the next decade. Whether you are credited, compensated, or simply erased from that process depends on what ownership infrastructure you put in place before the scraper finds your work. Basic tools for preserving source files, metadata, and hashes exist now. They are not expensive or complicated. They are simply not the default. Use the US Copyright Office where appropriate and preserve your own source evidence before the scraper arrives. Suede's IP Registry is live at [registry.suedeai.ai](https://registry.suedeai.ai) and will timestamp a fingerprint of that evidence on Base and Avalanche mainnet. It records provenance; it issues no licenses and is not a substitute for copyright registration. ### Further Reading See also: [Rights Metadata Is the Dark Matter of the Creative Economy](https://strumly.suedeai.ai/guides/rights-metadata-dark-matter), [AI Training and Your Music: What Every Guitarist Needs to Know](https://strumly.suedeai.ai/guides/ai-training-and-your-music), [Who Owns the Output Stage](https://strumly.suedeai.ai/guides/who-owns-the-output-stage). ## Who Owns the Output Stage Source: https://strumly.suedeai.ai/guides/who-owns-the-output-stage The output stage is where intent becomes sound. How Dumble, the signal chain, and on-chain rights infrastructure describe the same ownership problem. Category: Rights & IP. Published 2026-06-14, updated 2026-07-31. By Jason Colapietro. The output stage is the last point where a signal is still clearly yours. Howard Dumble understood this. AI broke it. Here is the architecture people are exploring to restore it. ### Dumble and the output stage Howard Dumble hand-wired each amplifier himself. He refused to sell to most people who called. If he did not believe a player understood tone at a molecular level, the conversation was over. He built maybe 300 amplifiers in his lifetime. Stevie Ray Vaughan had one. Carlos Santana had one. The rest of us read about them. The output stage is where the signal stops being internal and becomes external. It is where intent meets physics and the result is either yours or it is not. Dumble cared about this obsessively. He was protecting the relationship between a player's hands and what came out of the speaker. The output stage was the last point at which the signal was still clearly his and clearly theirs. ### The signal chain The signal chain as a concept is not complicated. String vibrates, pickup converts motion to voltage, amplifier shapes and magnifies that voltage, speaker converts it back to motion in air. The links are: transducer, preamp, power stage, output stage, transducer again. That is it. I spent years tracing that chain. The research became [The Signal Chain](https://guitar.solutions): four books, 46 chapters, the full history of amplifiers, effects, and the pursuit of electric guitar tone. Free at guitar.solutions. Because the knowledge should exist somewhere that is not paywalled. What I kept returning to while writing those books was the output stage. Not the components. The ownership question embedded in it. ### Then AI arrived In the last three years, AI generation models changed the practical meaning of "who made this." A player records a demo. A model trains on that recording without consent. The model now generates audio that carries the tonal fingerprint of the original player's hands, without any mechanism for the original player to know, assert, or benefit. The signal chain works exactly the same way it always did. The pickup still converts string motion to voltage. But the output stage — the point where creative work becomes a product that exists in the world — no longer has a clear owner. ### Infrastructure, not policy The core problem is not that AI generates music. Generative tools are not new; every effects pedal, every amp model, every convolution reverb is generative in some sense. The problem is that there is no layer in the current stack that answers three questions at the point of creation: who made this, under what terms can it be used, and where does value flow when it is. The signal chain has no ownership link. The output stage is unregistered. Suede's proposed design hashes a work, links a structured rights record to the creator's identity, and makes that record queryable. That design is the basis for USPTO patent pending 63/947,120: a method of binding a creative work's hash to a machine-readable rights object at the moment of creation rather than after the fact. The insight is timing. If you wait until a dispute to prove ownership, you are in a Marshall versus Fender situation, arguing about who copied whose circuit forty years later. If you register at creation, the record is contemporaneous with the act. ### The modern output stage The proposed settlement design places a rights record on Base, the Ethereum L2 that Coinbase built for low-cost on-chain transactions. A future agent or application could then license a work through x402, a payment protocol designed for machine-to-machine HTTP transactions using USDC. That architecture is a design goal, not a currently available Registry workflow. The tweed Bassman circuit that Marshall copied was a great circuit. It produced great amplifiers. But Leo Fender never saw a dollar of that, and the players whose tone defined that era never had a mechanism to say: this sound exists because of this work, and here is the record to prove it. They made the music. Someone else owned the output stage. That is the problem Suede set out to solve. The public IP Registry is live at [registry.suedeai.ai](https://registry.suedeai.ai), with registration on Base and Avalanche mainnet: a fingerprint, a signed creator claim, and a public timestamp, so the record exists before the argument does. It records provenance; it issues no licenses, settles nothing, and is not proof of copyright ownership. ### Further Reading See also: [Rights Metadata Is the Dark Matter of the Creative Economy](https://strumly.suedeai.ai/guides/rights-metadata-dark-matter), [AI Training and Your Music: What Every Guitarist Needs to Know](https://strumly.suedeai.ai/guides/ai-training-and-your-music), [What Happens to Your Guitar Samples in AI Training](https://strumly.suedeai.ai/guides/guitar-samples-ai-training). ## Why Session Musicians Are Losing Publishing Royalties in 2026 Source: https://strumly.suedeai.ai/guides/session-musicians-publishing-royalties Session guitarists built modern music and rarely see a publishing royalty. Why the system excludes them, how AI compounds it, and what changes. Category: Rights & IP. Published 2026-06-23, updated 2026-07-31. By Jason Colapietro. Session guitarists built the sound of modern music. The royalty system was never designed to pay them. Here's the structure of that failure and what changes it. Tommy Tedesco played guitar on an estimated 5,000 recording sessions. Hal Blaine drummed on more than 35,000 recordings. Carol Kaye's bass lines underpin records that have generated hundreds of millions of dollars in royalties. Most of what these musicians earned came from a flat session fee. They received no ongoing royalty as those records sold, streamed, and licensed. That was the deal. Work for hire. You play, you get paid once, the label keeps everything downstream. It was the standard arrangement for studio musicians from the 1950s onward, and it was explicitly structured that way by people who understood the value of what they were capturing and preferred not to share it. In 2026, the same structure is still running. It is just running in a new context, at a scale the original architects of work-for-hire could not have imagined. ### The two-track royalty system and why it matters Music royalties flow through two separate tracks that most people conflate: publishing royalties (for the composition) and master royalties (for the sound recording). When a song streams on Spotify, roughly half the payment goes to the rights holder of the master recording, and roughly half goes to the songwriters and publishers via mechanical and performance royalties. Session musicians are almost never on either side of this split. The master recording belongs to whoever paid for the session, typically the label. Session musicians signed away any ownership claim to the master as a condition of getting hired. The composition copyright belongs to the songwriter and their publisher. Session musicians are not songwriters by default — they play what they are given, and unless they contributed to the composition in a legally recognized way, they have no stake in the publishing. Performance royalties for the master recording do exist in some jurisdictions. In the United States, there are no terrestrial broadcast performance royalties for master recordings at all — a legal carve-out that was grandfathered in at the lobbying of broadcasters and has never been reversed. Digital streaming platforms do generate master performance royalties, but those go to the master rights holder, not the individual performers. > "The session fee was never the problem. The problem was that the fee was designed to be the entirety of the payment, forever, while the record kept generating revenue for everyone else." > > — Jason Colapietro ### Where AI makes the problem structurally worse Generative AI introduces a new layer of value extraction that the existing royalty system has no mechanism to address. When a music AI model is trained on session recordings, it learns the stylistic vocabulary of those recordings: the phrasing, the tone, the timing, the technique. That vocabulary becomes a capability the model can deploy on demand. Producers who previously hired session guitarists can now generate approximations of session guitar performance at essentially no marginal cost. The session guitarist's labor went into the training data that made this possible. The session guitarist receives nothing from the commercial deployment of that capability. This is not a hypothetical future scenario. It is the current operational reality of music production in 2026. The AFM (American Federation of Musicians) has negotiated some protections around AI use for union members. Non-union session work has no equivalent protection. Independent session musicians — the guitarists recording remotely through platforms like AirGigs, SoundBetter, and direct client relationships — are operating with no contractual protection against their work entering AI training datasets and displacing their future employment. ### The metadata gap at the center of it The royalty system cannot route money to session musicians it cannot identify. This is not a new problem, but AI has made it acute. Historically, session musicians were identified in liner notes and AFM contracts. Neither travels with the digital file. The MP3 of a 1972 recording that is streaming today carries no machine-readable indication of who played guitar on it. The session contract from that date may or may not be findable in anyone's archive. The musician may or may not still be reachable. For contemporary independent sessions, the problem is different but equally structural. A producer hires a session guitarist through an online platform, pays a flat fee, receives the files, and distributes the recording. The recording metadata that goes to the DSP rarely includes the session musician's name in any field that generates a payment or searchable record. There is no standard field for "session performer" that any major streaming platform routes money through. > "The metadata gap is not a technical accident. It is the predictable outcome of a system designed to capture value at the top, not distribute it at the source." > > — Jason Colapietro, Suede AI ### What session musicians can do now - **Negotiate explicit AI carve-outs in session contracts.** Any agreement that covers "all uses" or "any medium now known or hereafter devised" is broad enough to cover AI training. A specific carve-out requiring separate negotiation for AI training use is now a standard clause worth requesting. - **Register eligible reference recordings with the US Copyright Office.** Even if you cannot register the master you were hired to play on, you may be able to register practice recordings, demos, and reference takes you own. - **Document your session work with timestamped records.** Contracts, session notes, correspondence, and reference files with verifiable timestamps create the evidentiary foundation for any future attribution or compensation claim. - **Keep a verifiable evidence bundle for your catalog.** Store agreements, source files, credits, dates, and file hashes together so a future attribution discussion starts with records rather than memory. The session musician problem is not going to be solved by the existing industry structure. The same parties who benefited from work-for-hire arrangements for seventy years are not going to redesign those arrangements voluntarily. Any durable solution needs reliable records that connect performance to performer before the work enters a pipeline designed to strip that connection out. Build your own evidence trail now. The alternative is a future in which your playing is everywhere and your name is nowhere. Join the conversation on [Suede AI Social](https://suede.social). ### Further Reading See also: [How to Register a Guitar Riff as Intellectual Property](https://strumly.suedeai.ai/guides/how-to-register-a-guitar-riff), [Rights Metadata Is the Dark Matter of the Creative Economy](https://strumly.suedeai.ai/guides/rights-metadata-dark-matter), [Who Owns the Output Stage](https://strumly.suedeai.ai/guides/who-owns-the-output-stage). # The Signal Chain: free song lessons ## Purple Haze — the tone, the rig, and the lesson Source: https://strumly.suedeai.ai/book/lessons/lesson-purple-haze How Purple Haze gets its sound: Fuzz Face, Octavia, a Strat and a loud Marshall. The rig, a budget recipe, the theory, and original tab drills. From The Signal Chain by Jason Colapietro. Published 2026-06-10. #### The Rig & Signal Chain Hendrix's core rig here is one of the most documented in rock, and also one of the most mythologized — so let's separate what's solid from what's lore. - **Guitar:** A **Fender Stratocaster**, played right-handed-strung but flipped for a left-hander. That flip matters tonally: the staggered pole pieces and the slanted bridge pickup end up reversed, which subtly changes the high end. The bridge pickup does most of the work on the riff and lead. - **Fuzz:** A **Fuzz Face** — period units used germanium transistors, which are touch-sensitive and clean up dramatically when you roll back the guitar's volume knob. This is the single most important pedal in the chain. It's woolly, a little unstable, and reacts to your hands. - **Octave:** An **Octavia** (built by Roger Mayer), an octave-up fuzz that adds a ring-modulated, upper-octave ghost note. You hear it most clearly on the solo, where single notes high on the neck bloom into that piercing, alien doubling. - **Amp:** **Marshall** stacks — commonly cited as 100-watt heads pushed loud. Settings are usually described as bright and near-full, but treat any "exact" numbers you see online as approximate; the surviving documentation is thin. What's well-supported is the Octavia on the lead. Be skeptical of overly precise claims about "secret" settings — the magic here is mostly fuzz + a Strat + a loud Marshall + Jimi's hands. #### The Tone Recipe You can get ~90% there with accessible modern gear. - **Guitar:** Any Strat-style guitar with single-coils, on the **bridge pickup**. Humbuckers: split the coil or roll the tone back. - **Fuzz:** A modern silicon or germanium Fuzz Face clone, or any two-knob fuzz. A Big Muff works but is darker. - **Octave:** Any octave-up fuzz; engage it only for the solo. - **Amp:** Cranked-clean or lightly driven, set bright. Let the fuzz do the dirty work. - Fuzz pedal: **Fuzz 7–8, Volume to unity.** - Guitar volume: **8–9** for the riff, back to ~6 if it's too fizzy. - Amp: clean-ish. **Bass 5, Mids 6, Treble 7**, loud enough to feel it. - Guitar tone: **7–8.** Fuzz can get shrill; this tames it. - Pick attack: **medium-hard, edge of the pick.** The fuzz rewards aggression. Key move: practice rolling your guitar volume knob. Fuzz-cleanup is the secret handshake of this tone. STARTING POINTS · SCREENSHOT THIS AND STAND IN FRONT OF YOUR AMP · RANGES, NOT GOSPEL #### What's Going On Musically The song lives in **E**, but not plain major. Hendrix is working a blues/rock vocabulary built around the **E minor pentatonic** (E–G–A–B–D) and the **E blues scale** (add the b5, Bb), while the chords lean on **E7#9** — the famous "Hendrix chord." **Why E7#9 sounds the way it does.** A dominant 7 chord (E7 = E–G#–B–D) wants both a major 3rd (G#) and a flat 7 (D). The "#9" adds what is enharmonically **G** — the minor 3rd. So you're stacking a major 3rd and a minor 3rd in the same chord. That collision is exactly the sound of the blues: the vocal "blue note" frozen into a chord. Tense, gritty, unresolved — which is why it became shorthand for funk, psych, and hard rock. **The tritone.** The intro hook is the interval **E to Bb** — three whole steps apart, a tritone, historically nicknamed "the devil in music." It's the most dissonant interval in the octave and it refuses to sit still. Hendrix exploits that restlessness as a hook before the riff even starts. #### Signature Moves ##### 1) The tritone intro (E ↔ Bb) ``` e|-------------------------| B|-------------------------| G|-------------------------| D|-------------------------| A|--7~----6~----7~----6~---| <- Bb on the A string E|--0-----0-----0-----0----| <- E pedal underneath Bb A Bb A ``` That clash of open E against Bb is the tritone — dissonant on purpose. Add slow vibrato (~) and let the fuzz growl. ##### 2) The main riff (E blues vocabulary) ``` e|----------------------------| B|----------------------------| G|--------------------2-------| D|------2--4h5--------2-------| A|--0h2--------0------0-------| <- lands on an E7#9 stab E|--------------------0-------| ``` Pentatonic motion underneath, resolving to a stabbed E7#9 voicing. Pick hard; this is where the Fuzz Face earns its keep. ##### 3) The E7#9 voicing (the Hendrix chord) ``` e|--x--| B|--8--| <- G (the #9 / minor-3rd "blue note") G|--7--| <- G# (the major 3rd) D|--6--| <- D (the b7) A|--7--| <- E (the root) E|--x--| ``` Both 3rds (G# and G) in one grip — that's the entire emotional payload of the chord. Stab it, mute, repeat. #### The Drills ##### Drill A — 7#9 + R&B embellishment Builds the funk-comping reflex Hendrix used everywhere: hold the chord, decorate it with hammer-ons and muted scratches. ``` e|--x---x------x---x----------x---x------------| B|--8---8------8h8------------8---8------------| G|--7---7------7---7----------7---7------------| D|--6---6------6---6----------6---6----6/7-6---| A|--7---7------7---7----------7---7----7-------| E|--x---x------x---x----------x---x------------| > > > > slide tag ``` Tone: bridge pickup, fuzz at 6–7, guitar volume rolled to ~6 so the chord stays defined. Keep the strumming hand moving in constant 16ths; let the pick skim muted strings between hits. ##### Drill B — Tritone riff Trains the E-against-Bb tension move and locks in a swung, lurching feel. ``` e|--------------------------------------------| B|--------------------------------------------| G|--------------------7-----------------------| D|--------------------------------------------| A|--7~----6-----7-----7-----6-----7~----6-----| E|--0-----0-----0-----0-----0-----0-----0-----| Bb A Bb A Bb A (tritone) (resolve up to D then back) ``` Fuzz growling, guitar volume ~8 for bite. Let the open E ring under everything so the dissonance stays in your ear. The brief jump to D gives the line somewhere to breathe before it falls back into the clash. ##### Drill C — Octavia-style lead lick Trains high-neck single-note phrasing with bends, where an octave-up fuzz sings hardest. ``` e|--12b14r12----------12-15-12-----------------| B|-----------15--12----------------15b17~------| G|---------------------------------------------| D|---------------------------------------------| A|---------------------------------------------| E|---------------------------------------------| bend & release run down wail + vibrato ``` Engage the octave fuzz and use the bridge pickup. The effect tracks cleanest on single notes above the 12th fret, so keep the lines high and avoid full chords. Bend slowly and add wide vibrato — let each note bloom into its ghostly upper octave. #### Make It Yours The takeaways travel far beyond this one song. First, the **E7#9 grip** is a permanent addition to your vocabulary — drop it as a funk stab, a turnaround chord, or the last hit of any blues. Second, **fuzz-cleanup**: practicing your volume knob turns one pedal into a whole pedalboard, from clean-ish sparkle at 4 to full snarl at 10, all without tap-dancing. Third, the **tritone** is a tension tool you can sprinkle into any riff — pit a root against its b5 and let your ear feel the pull. Steal these ideas and you've got the DNA of a sound, not just a cover. - The **tritone clash** (E against Bb) in the intro — can you hear the dissonance refusing to resolve? - Both **3rds ringing together** (G# and G) when you stab the E7#9 — the "blue note" living inside the chord. - The **fuzz cleaning up** as you roll your guitar volume down — grit melting to clarity without your foot moving. - The **octave-up ghost note** blooming on the high lead lines — present on single notes, gone on chords. - Your **right-hand attack** driving the fuzz — dig in for the riff, ease off for the comping. ## Comfortably Numb — the tone, the rig, and the lesson Source: https://strumly.suedeai.ai/book/lessons/lesson-comfortably-numb The Comfortably Numb solo tone: Big Muff bloom, long delay, a clean Fender-style platform. The rig, a budget recipe, the theory, and original tab drills. From The Signal Chain by Jason Colapietro. Published 2026-06-10. This is the lesson where you stop measuring solos in notes-per-second and start measuring them in intention per note. Gilmour's outro solo is one of the most loved in rock not because it's hard to fret, but because every bend lands exactly on pitch, every note is allowed to bloom, and the whole thing climbs like a story with a beginning, a middle, and a roof-raising end. Your hands can probably play these notes today. The work is in the patience. #### The Rig & Signal Chain - **Guitar:** A **Fender Stratocaster** (his famous black Strat), typically on bridge or bridge-plus-middle for leads. Single-coils are central — that glassy attack is doing real work. - **Fuzz:** An **Electro-Harmonix Big Muff**, the violinist's secret weapon, for that endless, vocal sustain. The Muff is a fuzz, not a tube-amp overdrive — it compresses hard and sings. - **Delay:** Tape/analog-style **delay** for the wide, rhythmic repeats that make the solo feel three-dimensional. - **Compression:** A touch of **compression** (he's a known MXR Dyna Comp user across this era) to even out sustain and round the attack. - **Amp:** **Hiwatt** heads run clean and loud, a big uncolored platform so the pedals define the dirt. The signal philosophy matters more than the brands: dirt from a pedal, cleanliness from the amp, space from delay. That's the template for the whole genre of soaring lead tone. #### The Tone Recipe You can get ~90% there with very accessible gear. The point is the relationship between fuzz, clean amp, and delay — not boutique exotica. - **Guitar:** Any Strat-style single-coil guitar, bridge pickup. Humbucker? Roll the tone to ~6 and reduce gain — you're chasing glassy, not dark. - **Fuzz:** Any Big Muff variant or clone. In a pinch, a high-gain overdrive into a clean amp gets you in the area, with less bloom. - **Delay:** ~430–480 ms, feedback for 3–5 audible repeats, mix moderate. - **Amp:** Any amp set clean with headroom. A clean Fender-style channel is perfect. - Big Muff: **Sustain ~7, Tone ~4–5** (back off the fizz), Volume to unity. - Compressor: light — **Sustain ~4**, Output to taste. - Delay: **Time long, Repeats ~3–4, Mix ~3–4.** - Amp: clean and bright but not harsh; **Treble 6, Mids 5–6, Bass 5.** - Pick attack: **medium-soft, near the neck.** Let the note ring; vibrato from the fingers, not the pick. STARTING POINTS · SCREENSHOT THIS AND STAND IN FRONT OF YOUR AMP · RANGES, NOT GOSPEL #### What's Going On Musically The famous outro solo sits over a repeating loop in **B minor**: **Bm – A – G**. Your core scale is **B minor pentatonic**, rooted at the 7th fret. The magic ingredient Gilmour adds is **Dorian color**: a minor scale with one bright note raised, the **major 6th** (in B Dorian, that's G#). Hitting it over the A or G chords gives that hopeful, lifted, "not-quite-sad" sound that keeps the solo from feeling like a generic blues. Here's the chord-tone map you'll target — the whole secret to why his notes sound "right": - Over **Bm**: aim for B, D, F#. - Over **A**: aim for A, C#, E. - Over **G**: aim for G, B, D. When the chord changes, a great soloist changes their target note to match. That's why Gilmour's lines feel like melodies that belong to the song rather than scales sprayed on top. **Targeting chord tones** is the single highest-value skill in this lesson. #### Signature Moves ##### 1) The conversational opener (call-and-answer phrasing) ``` e|------------------------------------------------| B|--10b12r10--7~~~------------7--10b12~~~----------| G|---------------------9~~~-9----------------------| D|------------------------------------------------| A|------------------------------------------------| E|------------------------------------------------| ``` Phrases that pose a question (the first bend) and answer it (the resolution), with silence carrying the line. Let the delay fill the gaps. ##### 2) The soaring sustained bend ``` e|------------------------------------------------| B|--15b17~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~| G|------------------------------------------------| D|------------------------------------------------| A|------------------------------------------------| E|------------------------------------------------| ``` Push the 15th-fret B up a whole step to the target pitch, then sustain. Don't rush the vibrato in — let the note bloom first, then shimmer it. This is the Big Muff and compressor doing their job; your job is the vibrato. ##### 3) The climactic repeated bend ``` e|--15b17~~--15b17~~--15b17~~--15b17~~-------------| B|------------------------------------------------| G|------------------------------------------------| D|------------------------------------------------| A|------------------------------------------------| E|------------------------------------------------| ``` Repetition is the payoff. Every repeat must hit the exact same pitch — wandering intonation kills the climax. Dig in slightly harder each pass. #### The Drills ##### Drill A — Sustained bend-to-pitch with delay Trains intonation and patience. Set delay long with a few repeats. ``` e|------------------------------------------------| B|--15--15b17~~~~~~~~~~~~--15--15b17~~~~~~~~~~~~---| G|------------------------------------------------| D|------------------------------------------------| A|------------------------------------------------| E|------------------------------------------------| ``` Builds pitch-perfect whole-step bends. Trick — first fret the destination note (17) and listen, then bend the lower note and make it match exactly. Tone: Muff Sustain ~7, delay mix ~4 so the repeats reinforce your sustain. ##### Drill B — B minor pentatonic chord-tone targeting Trains landing on the right note when the chord moves. Loop Bm–A–G underneath (or imagine it). ``` (Bm) (A) (G) e|------------------|------------------|------------------| B|--10b12~~--10-----|--10--7-----------|------------------| G|-------------11---|--------9--7~~----|--7~~--9--7-------| D|------------------|------------------|------------9~~---| A|------------------|------------------|------------------| E|------------------|------------------|------------------| target: F#/D target: C#/A target: D/G ``` Builds hearing the chord under your line. The held notes are chord tones of each chord — that's why the line sounds "composed." Tone: a hair less gain than the climax so the note changes stay articulate. ##### Drill C — Long-phrase building (the narrative arc) Trains stamina and dynamic shape: start low and quiet, climb, and peak. One continuous four-bar idea. ``` e|-------------------------------------------------|-----------------| B|-------------------------------------------------|--15b17~~~~~~~~~~| G|--7~~--9--7----------9b11~~--9--------------------|-----------------| D|------------9--7--9-------------9~~--7------------|-----------------| A|-------------------------------------------------|-----------------| E|--7~~--------------------------------------------|-----------------| ``` Builds solo architecture — the discipline of saving your highest, loudest note for last. Play it three times without stopping; the high bend should feel earned, not early. Ride your volume knob up as you climb for a built-in crescendo. #### Make It Yours The lesson of "Comfortably Numb" transfers to any solo you'll ever play: **slow down, bend in tune, and target the chord.** Next time you take a lead break, give yourself a one-note rule for the first phrase — bend to pitch, hold it, add vibrato, and stop. Let the silence (and a little delay) do the talking. Then build, raising your register and intensity only as the section repeats, so you always have somewhere left to go. The Big Muff sustain is fun, but the real tone here is restraint: a clean amp, a held note, and the confidence to let it ring. Steal the architecture, not just the licks. - Bends that arrive exactly on pitch — no flat, no sharp, no creeping up to it. - Notes that bloom and sustain (fuzz + compression + clean amp), not stab and die. - Delay that adds width and rhythm without washing out your dry attack. - A genuine arc: the solo climbs in register and intensity toward a clear climax. - The repeated climactic bend hit identically every time — that consistency is the payoff. ## Pride and Joy — the tone, the rig, and the lesson Source: https://strumly.suedeai.ai/book/lessons/lesson-pride-and-joy The Pride and Joy shuffle decoded: Tube Screamer into a pushed Fender amp, heavy strings, right-hand technique. Rig, budget recipe, theory, tab drills. From The Signal Chain by Jason Colapietro. Published 2026-06-10. #### The Rig & Signal Chain - **Guitar:** SRV's "Number One," a hard-played early-'60s-spec **Stratocaster** with a fat neck. The defining feature is the **strings**: famously heavy gauges, commonly cited as.013s and up, tuned down a half step. Heavy strings are a big part of why his tone has that piano-like fundamental and refuses to sound thin. - **Overdrive:** A **Tube Screamer** (TS-808 / TS9 family), used less as a "distortion" box and more as a midrange-pushing, signal-fattening boost: it tightens the lows, bumps the mids, and shoves a clean-ish tube amp into early, touch-sensitive breakup. - **Amp:** **Fender** — a **Vibroverb** and/or **Super Reverb**, run loud and fairly clean-to-edge-of-breakup, letting the Tube Screamer and his right hand do the dirty work. - **Pickup:** **Bridge** for the rhythm bark; neck and middle for fatter lead moments. Settings are best treated as approximate. Avoid anyone who quotes you exact "secret" dial positions — those are guesses. #### The Tone Recipe You can get ~90% there with very accessible gear. - **Guitar:** Any Strat-style guitar with single-coils. Bridge pickup is home base. - **Strings:** Go up at least one gauge from your norm — try.010s or.011s. You don't need.013s to start, but heavier strings genuinely change attack and sustain. - **Overdrive:** Any Tube Screamer-style pedal. A "green overdrive" with a mid hump is the sound. - **Amp:** Any Fender-voiced clean amp pushed to the edge of breakup. - Tube Screamer: **Drive 3–4, Tone 5–6, Level 6–7.** Fatness and bite, not fuzz. The Level past noon is what pushes the amp. - Amp: **Bass 5, Mids 6–7, Treble 6, Reverb 3**, volume as loud as you can stand — power-tube compression is part of the feel. - Pick attack: **dig in.** This tone lives in the right hand — play hard, near the bridge, with a stiff pick. - Gain placement: most of the dirt from **amp and pick attack**; the pedal is the flavor, not the whole meal. STARTING POINTS · SCREENSHOT THIS AND STAND IN FRONT OF YOUR AMP · RANGES, NOT GOSPEL #### What's Going On Musically This is a **12-bar blues** in **E**, played as a **shuffle** — a swung, triplet-based feel where each beat divides into a long-short "dah-da" rather than even eighths. The "Texas shuffle" is a particularly driving, full version of that. The vocabulary is the **E blues scale** (E – G – A – Bb – B – D) over **dominant** harmony. Because each chord in a blues is a dominant 7th (E7, A7, B7), the underlying color is **E Mixolydian**. The magic of blues is rubbing the bluesy minor-3rd (G) and b5 (Bb) against the major-leaning chords — that friction is the sound. ``` | E7 | E7 | E7 | E7 | | A7 | A7 | E7 | E7 | | B7 | A7 | E7 | B7 | (last bar = turnaround) ``` What makes SRV special isn't the form — it's that he plays **bass, chords, and lead simultaneously**. The thumb and low strings imply a walking bassline while the upper strings stab chords and fills. A **double-stop** gives those fills their thickness. And everything is glued together with enormous **vibrato** — wide, vocal, as much a part of his signature as any note choice. #### Signature Moves ##### 1) The Texas shuffle engine A moving bassline under a chord stab — a boogie pattern that walks while the high strings answer. Keep it swung. ``` e|---------------------------------| B|---------------------------------| G|---------------------------------| D|-------2----2----4----5----------| A|---2----2----2----2----2---------| E|-0----0----0----0----0-----------| (let the low E ring, walk the A string) ``` A two-bar fragment of the boogie engine — thumb and low strings imply the bass while the pulse stays swung. Palm-mute lightly for punch. ##### 2) The turnaround The last two bars resolve the form and set up the repeat. This descending figure over E is a blues staple SRV peppers with grit. ``` e|----------------------------------| B|----------------------------------| G|--3-------2-------1-------0--------| D|--2-------2-------2-------2--------| A|--2-------2-------2-------2--------| E|--0-------0-------0-------0--------| E7 (chromatic descent) B7 ``` A classic turnaround shape — the top voice steps down chromatically over a held E, then you'd land on B7 to restart. ##### 3) Raked double-stop fill SRV "rakes" — dragging the pick across muted strings into the target double-stop so the notes arrive with a percussive scrrr-CHANK. ``` e|--------------5b6----------| B|--x--x--x-----5------------| G|--x--x--x------------------| D|---------------------------| A|---------------------------| E|---------------------------| (rake the dead strings) (bend the top note up) ``` The rake (xxx) is the run-up; the double-stop with a slight bend on the high E is the payoff. The dead-string scratch is the attitude. #### The Drills ##### Drill A — Shuffle independence The hardest part of this music: keeping a steady swung bass while stabbing chords on top. Thumb (or pick) hits the bass note; fingers and pick answer on the upper strings on the off-beat. ``` e|------------------------------------------| B|----3-------3-------3-------3-------------| G|----4-------4-------4-------4-------------| D|--2-------2-------4-------5---------------| A|-----------------------------------------| E|-0-------0-------0-------0----------------| BASS stab BASS stab (walk D string up) ``` Builds bass-and-chord independence plus a rock-solid swing. Bridge pickup, palm-mute the low E for thump, let the stabs ring. Start painfully slow — the groove is in the spaces. ##### Drill B — Raked double-stops up the neck Trains the rake attack and double-stop intonation across positions. Each rake (xxx) leads into a clean two-note hit. ``` e|--x--x--5/7~-----x--x--7/9~-----x--x--9/10~-----| B|--x--x--5/7~-----x--x--7/9~-----x--x--9/10~-----| G|------------------------------------------------| D|------------------------------------------------| A|------------------------------------------------| E|------------------------------------------------| rake slide+vib rake slide+vib rake slide+vib ``` Builds the percussive rake, double-stop sliding, and vibrato on TWO strings at once. Push the Tube Screamer's Level so the rake scratches bark. Keep both fingers parallel so the slide stays in tune. ##### Drill C — E blues lead with wide vibrato The vibrato exercise. The whole point is the shake — slow, wide, vocal. Don't rush past the held notes. ``` e|------------------------------------------------| B|-------------------8b9r8-----5~~~---------------| G|--7b9~~~-------7----------7----------------------| D|------------9-----------------------9~~~---------| A|------------------------------------------------| E|------------------------------------------------| big bend+vib blues phrase land + WIDE vib ``` Builds the SRV vibrato — wrist-driven, wide, and patient — plus full-step bend control. Use the neck pickup for a fatter lead voice. The held notes with ~~~ are the lesson: make them sing, don't just touch them. #### Make It Yours You don't need.013s, an Eb tuning, or a Vibroverb to absorb what makes this great. The transferable lessons are three: **play hard**, **swing everything**, and **commit to your vibrato**. Take the shuffle independence idea into any blues — even implying a single bass note under a chord stab instantly makes your rhythm playing sound like a whole band. Steal the rake as an accent on any double-stop fill; it turns a polite lick into a statement. And spend real time on a slow, wide vibrato — it's the single fastest way to sound like a player with a voice rather than a player running scales. - A **swung, triplet** pulse — never straight eighths. If it sounds even, slow down and re-feel it. - **Bass and chords at once** — the low strings walking while the top stabs answer. - The **percussive rake** into double-stops: that scrrr-CHANK of dead strings before the hit. - **Wide, vocal vibrato** on held notes — slow and deliberate, from the wrist, not a nervous wiggle. - **Bite with clarity** — a bridge-pickup tone that's gritty and thick yet lets every string ring. ## Smells Like Teen Spirit — the tone, the rig, and the lesson Source: https://strumly.suedeai.ai/book/lessons/lesson-smells-like-teen-spirit The quiet-loud detonation decoded: a cheap offset, a Boss DS-1, a clean amp doing the contrast. Rig, budget recipe, theory, and original tab drills. From The Signal Chain by Jason Colapietro. Published 2026-07-05. #### The Rig & Signal Chain Cobain’s live and studio setup was famously cheap and rugged, which is good news for you — none of this is boutique. - **Guitar:** A **Fender Mustang**, commonly associated with his Nevermind period, or a **Fender Jaguar**: short-scale offsets with single-coils, though he often swapped a humbucker into the bridge. Single-coils give the verse its glassy thinness; a hotter bridge pickup feeds the chorus. - **Distortion:** A **Boss DS-1**, and later the **Boss DS-2 Turbo Distortion**, is the most-cited dirt box. This pedal is the heart of the sound — a hard-clipping distortion he reportedly ran fairly aggressive. - **Amp:** In the studio, a **Mesa/Boogie** amp, with a Studio.22 commonly cited, paired with **Fender** amplification for the cleans; live he leaned on Fender combos and solid-state heads. Cleans came from the amp; grit came from the pedal slamming the front end. - **Mic:** Producer Butch Vig tracked guitars at Sound City, close-miking cabs with the usual workhorse dynamics. Exact mic placement is studio lore — treat specifics as approximate. Settings are best treated as ranges. The DS-1 is **commonly cited as** Dist around 7–8, Tone around 5–6, and Level to taste. Amps sat **approximately** clean-to-edge so the pedal did the clipping. Do not chase a “secret” patch — there wasn’t one. #### The Tone Recipe You can get roughly 90% there for very little money. - **Guitar:** Any offset, or even a Strat/Tele with a bridge humbucker, works. No Mustang? A Squier with a hot bridge pickup gets you close. - **Distortion:** If a DS-1 is unavailable, use any hard-clipping distortion rather than a smooth overdrive: a ProCo RAT, a Boss MT-2 dialed back, or a digital DS-1 model. - **Amp:** Any clean platform — a small tube combo or a modeler set to a clean Fender/Mesa-style voice. - Amp: clean. **Gain around 3, Bass around 6, Mid around 4, Treble around 6, Reverb around 2.** - Distortion pedal: **Drive around 7, Tone around 5**, Level matched so the volume jumps slightly when engaged. - Pickup: **neck or middle single-coil** for verses; **bridge** for chorus. - Pick attack: verses light, loose, and almost lazy; chorus hard, full downstrokes, with the strings allowed to rattle. - Gain placement: keep the amp clean and let the pedal be your distortion, so the switch does the quiet-to-loud work. The mid-scoop people chase isn’t extreme. The DS-1 is actually mid-present; the “scoop” is really the contrast against the hollow clean verse. Don’t over-scoop the amp or the chorus loses its punch. STARTING POINTS · SCREENSHOT THIS AND STAND IN FRONT OF YOUR AMP · RANGES, NOT GOSPEL #### What’s Going On Musically The entire song rides **four power chords**: **F – Bb – Ab – Db**. A power chord, written as a “5,” is just root plus fifth — no third — so it’s neither major nor minor, which is exactly why it sits so well under distortion. Spell those roots out and you get a sound centered on **F**, colored by the flatted scale degrees of F minor. The Ab and Db are the b3 and b6 of that world. The progression doesn’t resolve in a tidy classical way; it loops as a riff, which is the point. **Form** is textbook loud-quiet-loud, the soft verse / loud chorus architecture Pixies pioneered and Nirvana perfected: a clean intro, a verse stripped to a two-note figure and bass, a pre-chorus that builds tension on a held chord, and then the full four power chords with full distortion and full attack. The **solo** is the lesson’s secret weapon: Cobain plays the vocal melody of the verse almost note-for-note. It isn’t a flashy shred solo — it’s a sung line on guitar. When you don’t know what to play, play the melody. It always belongs. #### Signature Moves ##### 1) The four-chord engine ``` e|--------------------------------| B|--------------------------------| G|--------------------------------| D|--3----8----6----11-------------| A|--3----8----6----11-------------| E|--1----6----4----9--------------| F5 Bb5 Ab5 Db5 ``` The whole song in four shapes. Clean and palm-muted in the verse, slammed wide-open in the chorus — same fingers, opposite dynamics. ##### 2) The dynamic drop ``` e|--------------------------------| B|--------------------------------| G|--------------------------------| D|--------------------------------| A|--6----6----4----4--------------| E|--x----x----x----x--------------| PM PM PM PM ``` Reduce to a single dyad and dead-note scratches. The guitar almost disappears so the chorus can explode. ##### 3) The melodic solo idea ``` e|--------------------------------| B|--4----4----4----4----6----4----| G|----------------------------3--1-| D|--------------------------------| A|--------------------------------| E|--------------------------------| ``` Tracks the verse vocal contour. Don’t rush it — let each note land where a sung syllable would. #### The Drills ##### Drill A — Clean/Dirty Dynamic Switch Train the loud-quiet reflex. Play four bars clean and muted, then the same shapes wide open. If you have a footswitch, stomp the distortion on beat 1 of the loud bars. ``` (quiet) (loud) e|------------------------|------------------------| B|------------------------|------------------------| G|------------------------|------------------------| D|--3--3--6--6--8--8--6--6-|--3--3--6--6--8--8--6--6-| A|--3--3--6--6--8--8--6--6-|--3--3--6--6--8--8--6--6-| E|--1--1--4--4--6--6--4--4-|--1--1--4--4--6--6--4--4-| PM (clean) (open, hard downstrokes) ``` Builds dynamic control — the single most important skill in this song. Tone: amp clean, pedal off for bars 1–2, on for 3–4. The volume jump should feel violent. ##### Drill B — Power-Chord Shape Shifting Lock in clean position changes across the neck so the chorus never stumbles. All root-fifth shapes; minimize finger lift between them. ``` e|--------------------------------| B|--------------------------------| G|--------------------------------| D|--1--/3--5--\3--6--/8--6--\4----| A|--1--/3--5--\3--6--/8--6--\4----| E|--x---x--x---x--4--/6--4--\2----| ``` Builds smooth, accurate shifts and slide articulation under the chord. Tone: light crunch so you can hear cleanly whether every shape rings true. ##### Drill C — Sung-Line Soloing Teach yourself the “play the melody” move. Hum any short phrase, then find it on the B and G strings in one position. This original line stays in an F-centered minor pentatonic box. ``` e|--------------------------------| B|--6--6--4--6--8--6--4-----------| G|--------------------5--3--5--3--| D|--------------------------------| A|--------------------------------| E|--------------------------------| ``` Builds melodic phrasing and the discipline to play a vocal line instead of running scales. Tone: same as the chorus — let the distortion sustain each note like a held vowel. #### Make It Yours The real lesson here isn’t four chords — it’s that **dynamics are a songwriting tool, not just a volume knob.** Take any progression you already know and try the loud-quiet architecture: strip a section to two notes and a whisper, then detonate the next section with full chords and full attack. The contrast does the emotional work. Pair that with the “play the vocal melody” trick for solos and you have two ideas that will outlast any specific tone. - The volume and texture jump the instant the chorus lands — clean verse to thick distortion with no gradual fade. - Power chords with **no third** — full, open, neither major nor minor under the dirt. - The verse guitar nearly disappearing, leaving bass and a couple of dead-note scratches. - The solo singing the verse melody rather than shredding over it. - Loose, almost careless attack in the quiet parts vs. hard, committed downstrokes when it’s loud. # Print the Quiet: essays on tone ## Jeff Buckley’s Hallelujah: The Tone That Won the Song Source: https://strumly.suedeai.ai/book/print-the-quiet/ptq-01-jeff-buckley-tone-hallelujah Inside Jeff Buckley's Hallelujah: the 1983 Telecaster, the Quadraverb halo, Andy Wallace's comp at Bearsville, and why tone won Cohen's song. Print the Quiet essay by Jason Colapietro. Published 2026-06-13. ● Print the Quiet · 1 / 9 ## Print the Quiet How Jeff Buckley out-toned Leonard Cohen at Bearsville in September 1993, and why no master has matched that take on its own terms. By [Jason Colapietro](https://strumly.suedeai.ai/book), writing as Johnny Suede · A Suede AI Social long read · 14 min There is a moment, about three minutes and twenty seconds into Jeff Buckley's Hallelujah, where the vocal drops to almost nothing. You can hear the room. You can hear the small movement of his lips opening. The guitar arpeggio is so quiet that the noise floor of the tape becomes part of the music. And then, without lifting his right hand from the strings, he begins the climb back. No splice. No tuning. No bus compressor doing the work for him. Just a body in Bearsville, deciding what to do next. That moment — and the thousand small tonal decisions that frame it — is why a song written by Leonard Cohen and arranged by John Cale is, for two generations of listeners now, Jeff Buckley's song. It is not because Buckley out-wrote Cohen. He did not. It is because he out-toned him. This is a piece about that. It is in two halves. The first is a survey of Buckley's tonal world — the guitars, the amps, the throat, the philosophy. The second is a deep listen at the Hallelujah recording specifically, what was on the floor at Bearsville Studios in September 1993, what landed on tape, and why, more than three decades later, that take still feels like the only one. ### Part I — The Tonal Religion #### The blonde Telecaster, and the borrowed life it had Buckley's main electric was a **1983 Fender USA Telecaster** in butterscotch blonde — a one-year-only Top Loader, meaning the strings load through a plate on top of the body rather than ferrules through the back. The build gives the guitar a looser, almost-acoustic compliance under the right hand. Players describe it as having less snap and more give than a standard Tele. You can hear that compliance everywhere on Grace. The chord work in "Lover, You Should've Come Over" rings like a parlor guitar. The arpeggios on "Hallelujah" sustain in a way a hardtail Tele would not. The instrument's history is part of the tone too. Buckley's gear was stolen in a Los Angeles burglary in 1991. He drifted to New York, met Janine Nichols at St. Ann's Warehouse around the Tim Buckley tribute that summer — the show where Gary Lucas first saw him — and Nichols, the program director, lent him her Telecaster. It stayed lent for years. Eventually he traded her a Gibson L-1 acoustic for it. She had already added the mirror pickguard, reportedly inspired by Chrissie Hynde. Stage lights bounced off it. After Buckley's 1997 death the guitar passed through a UK collector and the Matt's Guitar Shop in Paris before landing, in 2020, with Matt Bellamy of Muse — who is on record about how directly Buckley influenced him. A few caveats live here. The bridge pickup was swapped for a Seymour Duncan stacked humbucker (sources disagree on whether it was the Hot Stack, Hot Rails, or Hot Lead Stack), but photographic evidence places the swap in **late 1994 / early 1995** — meaning the Grace studio takes were tracked with the original bridge single-coil. The Tele was also eventually refretted. Treat any blanket claim of "Hot Rails on Grace" with skepticism; the album's bridge bite is stock Fender. Around the Tele lived a small, deliberate stable. A **1967 Guild F-50** acoustic, loaned by Steve Addabbo, was the main Grace acoustic; Buckley taped over its soundhole to fight live feedback. A **Rickenbacker 360/12 Fireglo**, bought with his first Sony advance, was kept in Open G (DGDGBD) as a dedicated drone instrument — that is the chime under "Last Goodbye." A **Gibson L-1** parlor turned up in 1994. There is no credible evidence of a Strat in his core rig. #### Two amps, an A/B switch, and a rack reverb that did half the work The Buckley signal chain is mythologized into something baroque. It was not. It was two combos and a Morley. The clean side was a **Fender Vibroverb '63 Reissue** in brown tolex — chime, headroom, that slightly compressed midrange Fender voiced amps had before the brownface era ended. His own Vibroverb misbehaved during the Grace sessions, so Andy Wallace rented a second '63 Reissue from S.I.R. Studios for the album cuts. The dirty side was a **Mesa/Boogie Dual Rectifier Trem-O-Verb combo** — this matters, because almost every fan rundown online still says ".50 Caliber+ head," and the visual record across Live in Chicago and Glastonbury 1995 shows the Trem-O-Verb combo with the rack reverb sitting on top. The cabinet was a Mesa 4x12. The two amps were toggled by a **Morley A/B switch box**, so the same Telecaster could go from glassy chime to wall-fuzz attack inside a single song. Listen to "Eternal Life" or the final two minutes of "Grace" live — that is the Trem-O-Verb side of the switch. On the floor: a **DigiTech WH-1 Whammy** (the first-generation red one) for the octave-up swoops on "Mojo Pin" and the descending chaos on "Eternal Life"; a **Big Muff Pi** for sustaining fuzz; a **TC Electronic Stereo Chorus/Flanger**; a tuner. Most rundowns include a **Boss DD-3** delay, though primary documentation is thinner there than for the Whammy. The piece of gear that did the most invisible labor was the **Alesis Quadraverb** sitting on top of the Trem-O-Verb. Buckley used the "Taj Mahal" preset, modified with a longer decay, for the halo that hangs around the Hallelujah arpeggios and the suspended washes on "Grace." A great deal of what listeners hear as Buckley's "atmosphere" is, technically, one tweaked preset on a $400 rack unit. It is a useful corrective to anyone who thinks tone is a function of budget. He printed the Quadraverb. What you will sometimes see attributed to him — a Vox AC30 — is fan lore. The chime people credit to a Vox is the Vibroverb pushed clean. No contemporary photo, tech statement, or interview places an AC30 in his 1993–97 rig. #### The right hand The pedals matter less than what his hand did to the strings. Buckley played **hybrid** — pick held between thumb and index, with the middle, ring, and pinky fingers free to pluck simultaneously. On the cleanest passages — Sin-é recordings make this audible — he dropped the pick entirely and used thumb and fingers, harp-style. That technique is the actual source of the harp-like quality on "Lover, You Should've Come Over" and "Hallelujah." It is not a chorus pedal. It is four fingers and a thumb. His dynamics were almost entirely a function of right-hand attack rather than volume-knob work. He could let strings ring at near-zero, then dig in for sudden crescendos that surprise the listener's nervous system before they reach the ear. The "whisper to roar inside one phrase" is not metaphor; it is a specific physical skill, drilled into him from years of forcing rowdy New York cafés into silence by going quiet. The audience's stillness became part of his instrument. #### The voice The number you will see is **four octaves**, roughly E2 to G5 with extensions in both directions. Treat it as the accepted shorthand rather than a measured fact. What matters more is the architecture. Buckley worked his voice like a multi-stage amp. Chest-voice was clean. Head-voice sat at the edge of break. Falsetto was the saturated stage. He could move between them inside a syllable. The most-overlooked influence here is not Robert Plant — though Plant was foundational — it is **Nusrat Fateh Ali Khan**, the Pakistani qawwali singer whose cassettes Buckley collected in the hundreds and whose phrasing Buckley spent years internalizing. "He's my Elvis," he said in 1993. Listen to the final ninety seconds of "Grace," the ululations and microtonal bends and the wave-form crescendos that climb without arriving — that is direct qawwali architecture, ported into a rock vocal. The full tonal genealogy he gave for himself, often quoted: "The warped lovechild of Nina Simone and all four members of Led Zeppelin, with the fertilized egg transplanted into the womb of Piaf." Each name in that sentence is a tonal decision. Piaf was the drama-as-theology lineage; performing at L'Olympia in Paris twice in 1995 was a deliberate Piaf gesture. Simone was the phrasing and harmonic risk — he took her arrangement of "Lilac Wine" as a template. Plant was the upper-register wail used as percussion. Nusrat was devotion as form. Van Morrison was the improvisatory mantra coda — see any live "Kangaroo." In the studio he sang into a **Neumann U87** large-diaphragm condenser, the standard Bearsville vocal mic, through what was almost certainly an API preamp into the Studer A80 tape stack. He sang without headphones; monitors in the room; sometimes with six or twelve people sitting in to give him a small audience. He told Interview's Ray Rogers in 1994: "Words are really beautiful, but they're limited. Words are very male, very structured. But the voice is the netherworld, the darkness, where there's nothing to hang onto. The voice comes from a part of you that just knows and expresses and is." And, separately, on recording itself: "It's not like a live show where you play it and it just disappears into the air like smoke. It's like painting. Sound painting." Hold that phrase. It is what Bearsville got onto tape. ### Part II — Bearsville, and the Hallelujah That Still Breathes #### The room Basic tracking for Grace began at **Bearsville Studios in Woodstock, NY, in September 1993**, in **Studio A** — a large open tracking room across the lawn from the smaller Studio B mix room, where the SSL 4000 E lived. The lineup in the room was small: **Andy Wallace** producing, engineering and mixing (fresh off his Nirvana Nevermind mix and chosen for his ability to bridge heavy and atmospheric records); **Steve Berkowitz**, Columbia's A&R, the man who signed Buckley; assistant engineer **Chris Laidlaw**; the band — Mick Grondahl, Matt Johnson, Michael Tighe — present at the complex but not playing on Hallelujah, which is solo. Berkowitz, on the working method: "There was no plan. It was very loose, like a conversation in someone's living room." That looseness is the underrated story of the record. The takes were live. Buckley tracked guitar and vocal together, no headphones, monitors in the room, sometimes a small audience seated within sight. Wallace had figured out early that Buckley sang better without cans on his head. #### The take, honestly The romantic version is that Hallelujah is one take. It is not. Buckley cut **more than twenty full performances** at Bearsville. The released master is a comp Wallace assembled from somewhere between **three and five** of those takes — accounts vary, and Alan Light's 2012 book The Holy or the Broken, the definitive book on the song's life, captures Wallace and Berkowitz remembering the process differently. Wallace describes the comp as light — "pretty much straight-ahead, mostly one version with fixes." Berkowitz remembers a more substantial stitch. The honest reading sits between them. What is true, and what matters: it is **full-performance comping, not phoneme-level surgery**. Wallace was choosing between complete emotional arcs, not stitching syllables. That is a different category of editing than the modern major-label vocal, which is routinely assembled from dozens of takes at the word level and then tuned. Phrasing decisions in a real take are causally linked — the way you sing line four is a response to how line three just felt. Comping at the take level keeps those causal chains intact. The listener's nervous system, which is built to read those chains, hears them as continuous. A modern phoneme comp breaks the chain, and listeners cannot articulate the discontinuity, but they clock it. So: not literally one take. But emotionally and structurally a single performance. The hedge matters. The argument survives it. #### What was on the floor The guitar is almost certainly the **1983 Telecaster**, dry into the **Fender Vibrolux reissue** Wallace had specced for intimate work, with the **Quadraverb's "Taj Mahal" preset** providing the gentle halo around the arpeggios. There is a long-running counter-claim from assistant engineer Bryant W. Jackson that a Gibson hollowbody pass exists from the **overdub sessions** at Soundtrack in Manhattan; Jackson is careful to note he was not present at the Bearsville basic tracking. The defensible reading is that the released master is the Tele captured at Bearsville. A hollowbody version may exist on another tape reel, somewhere. The vocal mic was almost certainly a **Neumann U87**. Wallace has never definitively pinned it in print — this is one of the most-asked, least-answered questions about the record — but the U87 was Bearsville's standard large-diaphragm condenser, and it is the mic visible in the studio footage of these sessions. Some accounts cite a U67. Either is plausible; both are large-diaphragm tube-or-FET condensers in the same family. The arrangement is in **standard tuning with a capo at the 5th fret**, finger-arpeggiated in 6/8 out of G-position shapes — G, Em7, C, D, with a B7 at the turn — sounding in C concert. The figure was already deeply rehearsed before tape rolled: Buckley had been playing it at Sin-é and FEZ for nearly two years. The single most important tonal decision is one of subtraction. The released master has **no double-tracked vocal. No harmony stack. No synth pad. No band. No reverb wash big enough to obscure the consonants.** Compare with what else was being mixed in 1994: Oasis's Definitely Maybe, with Noel's guitars layered four deep; Nine Inch Nails' The Downward Spiral, Reznor as a deliberate maximalist of the sonic real estate; Green Day's Dookie, Rob Cavallo compressing the whole thing to a pop wall. Wallace and Buckley made the opposite record in the same calendar year. Ambient space around a Telecaster and a voice. #### The mix Wallace mixed Grace on Bearsville's SSL with his documented working approach: **parallel compression** — a clean fader and a crushed bus blended back — conservative on time-based effects, punch first, polish second. He has since walked the Hallelujah session publicly in the Mix With The Masters "Deconstructing a Mix #31" series. The headline points: the vocal compression on the master is **relatively gentle**. The dynamic range you hear is genuinely as wide as it sounds; it is not the illusion of dynamics produced by a clever limiter. The spatial sense comes from blending several reverb sources — a long sustain plate and a shorter, drier ambience — rather than one obvious tail. Wallace's choice was to extend the room, not replace it. Put differently: the engineer's job, faced with that source, was to get out of the way. Wallace got out of the way more than he often does. #### Cale, not Cohen Buckley learned the song from **John Cale's 1991 piano cover** on the Cohen tribute I'm Your Fan, not from Cohen's 1984 original on Various Positions. That detail does most of the explanatory work. Cohen's original is a synth-and-choir gospel arrangement that he himself often performed as a near-recitative; Cale's version is piano, sustain pedal, and a specific editorial choice — Cohen had faxed Cale fifteen pages of verses, and Cale picked the "cheeky" ones (the kitchen chair verse, the "shoot at someone who outdrew you" verse) that Cohen had only performed live. Buckley inherited Cale's lyric selection wholesale. The translation from Cale to Buckley is the second key editorial moment. Cale voiced the chords as block triads with sustain pedal — vertical, hymnal. Buckley arpeggiated them in 6/8 with the capo at the 5th fret — horizontal, breathing, pulsed. That arpeggio is Buckley's invention. It is the part of the song that is not in the score, not in Cohen's writing, not in Cale's reading. It is the part that is tone. Asked about the song's meaning, in the press kit that circulated around Grace's release, Buckley said something that has been quoted many ways since: "Whoever listens carefully to Hallelujah will discover that it is a song about sex, about love, about life on earth. The hallelujah is not a homage to a worshipped person, idol or god, but the hallelujah of the orgasm." He added, half-joking, to his manager Dave Lory and to several interviewers: "I hope Leonard doesn't hear it." (The widely repeated claim that Cohen called Buckley's version his favorite does not trace to a verified primary source. What Cohen did say publicly, after Buckley's death, was: "Jeff Buckley — it's a sad story." Treat the rest as fan canon.) #### Why it still breathes — thirty years on Grace was released on **August 23, 1994.** It was a slow burn. The album did not go gold in the United States until 2002. Hallelujah did not chart as a single until 2006. The song's life has three inflection points: **1997**, when Buckley died and a critical re-evaluation began; **2001**, when Shrek used John Cale's version in the film and Rufus Wainwright's on the soundtrack, lifting all versions on a rising tide; and **the streaming era**, in which Buckley's version — the most emotionally dynamic and recognizably modern of the recordings — won the algorithmic war. The receipts: Buckley's Hallelujah sits at over **430 million Spotify streams** as of mid-2026, more than double Cohen's 1984 original. The songwriter wrote it. The tone won the song. There are technical reasons for this, and they are worth taking seriously, because they describe a vanishing standard. **Dynamic range.** Grace was mastered on the eve of the loudness war. The original 1994 CD master is widely cited in mastering forums as DR11–DR13 — solidly in the healthy-dynamics band on the Dynamic Range Database's scale. The 2004 Legacy remaster compressed the master for retail loudness and lost what audiophiles describe as "clarity and air." For a comparison number: Taylor Swift's 1989 (2014) is DR6. AC/DC's Back in Black is DR12. The accessible version of this argument: a track with a DR6 master cannot let a vocal drop to a whisper and rise to a full-throated cry inside the same phrase, because the mastering chain has crushed the difference between Buckley's breath intake and the chord change. They are sitting at the same perceived volume. A DR12 master can. Grace is the receipt for what wide dynamic range buys you. **Pitch correction.** Auto-Tune launched in **1997, three years after Grace was tracked.** The Hallelujah vocal predates the entire correction paradigm. Listeners under thirty today have effectively never heard a major-label pop vocal that was not at least lightly tuned to grid. The micro-pitch wobble on the word "cold" in the second verse — the slight bend on "hallelujah" in the final chorus — these are not artifacts to be corrected. They are content. There is plausible neuroscience here (Molnar-Szakacs and Overy's work on the mirror neuron system suggests listeners' nervous systems track vocal sound as if simulating its production), but the simpler version is that more of the body shows up in an unpunished signal, and the body is what people respond to. **Single-performance feel.** As discussed: not literally one take, but a comp of complete arcs, not phonemes. The causal chain of phrasing is intact. Listeners hear the continuity. The combined effect is a record that, by current pop-mastering standards, is underprocessed. And the underprocessing is exactly what makes it durable. #### A note on the question every listener now eventually arrives at There is a contemporary version of this conversation worth saying out loud. As AI music generation becomes excellent at notes — at melody, chord progression, plausible timbre — the question that keeps surfacing is what, exactly, tone is, and whether it is the part of music that remains specifically human. Hallelujah is a useful case to think with. The score of Buckley's Hallelujah is Leonard Cohen's. The chord changes, the melody, the lyric — all available on a lead sheet. What is **not** in the score: the half-second hesitation before "the holy or the broken Hallelujah." The slight pitch drift on the word "cold." The way his thumbnail catches the wound G string at the second-verse pickup. The proximity of the U87 in front of his mouth, capturing breath as content rather than as noise to gate. The absence of reverb on the verses and the controlled plate bloom Wallace adds on the choruses. The Bearsville Studio A air, reflecting the guitar off whatever the room temperature was that day in September. The fact that Buckley was twenty-six years old, in his first record, with the entire weight of his father's reputation and his own self-imposed standard pressing on the take. That is what tone is. It is the irreducible record of a body in a room making a decision in time. It is the part of music that is not in the score. It is the part that no current model has been able to infer from the notes — because the information was never in the notes. It was in the breath. Brad Pitt, who has been a Buckley evangelist for nearly thirty years and produced the 2025 documentary, has said: "There's an undercurrent to his music. There's something you can't pinpoint." Chris Cornell, who covered Buckley's voice better than almost anyone, said: "He had a way of playing the most beautiful songs you've ever heard and singing them, and still with the way that he sang, create a bit of an uncomfortable edge to it if he felt like it. And he did that mostly with his voice." Matt Bellamy, who now owns the Telecaster: "He wasn't scared to be a high-voiced male. I think that helped me open up." The thing none of them quite name is the same thing. Tone is a value system. It is a set of decisions about what to leave in and what to take out — about whether to print the breath, the fingernail squeak, the room — and the value system on the Hallelujah master is: leave the human in. Take the chrome out. Wallace got out of the way. Buckley made a decision in time. The tape caught it. Thirty years on, no master has matched it on its own terms. If you are listening to it tonight, on whatever device, do this. Turn the lights off. Put one good speaker in the room. Stop scrolling. Listen for the breath at 3:20. You can hear the room. That is what won. Suede AI Social. A long read on tone, provenance, and the parts of music that don't fit on a lead sheet. ## The Loudness War, in Numbers Source: https://strumly.suedeai.ai/book/print-the-quiet/ptq-02-loudness-war-in-numbers Metallica's Death Magnetic hit DR3. The loudness war in numbers: LUFS, DR scores, the Waves L1, Bob Katz's K-system, and what Spotify changed. Print the Quiet essay by Jason Colapietro. Published 2026-06-13. ● Print the Quiet · 2 / 9 How the masters got crushed, who tried to stop it, and whether streaming normalization actually ended it. By [Jason Colapietro](https://strumly.suedeai.ai/book), writing as Johnny Suede · A Suede AI Social long read · 11 min The first piece in this series argued, through Jeff Buckley's Hallelujah, that tone is a value system — a set of decisions about what to leave in and what to take out. The deepest single decision in modern record-making is the one a mastering engineer makes about how loud to print the master. That decision, made wrong about a million times between 1994 and 2010, took the dynamics out of pop music. This is the piece that explains, in numbers, what happened. ### What loudness actually means Three measurements get used in this conversation. They are different, and the difference matters. **Peak** is the instantaneous maximum sample value of the audio file. A peak meter reads the tallest spike. On a CD or a 16-bit WAV, peak maxes out at 0 dBFS (decibels relative to full scale). One sample at 0 dBFS, or above the digital ceiling for a true-peak meter, can produce audible clipping if the playback chain reconstructs the wave between samples. **RMS** (root mean square) is a moving average of signal power over some window — typically 300 ms. It tells you, roughly, how much sustained energy is in the signal. Two different songs can hit the same peak but have wildly different RMS levels. The one with the higher RMS sounds louder. **LUFS** (Loudness Units relative to Full Scale, defined in the ITU-R BS.1770 standard) is the modern measurement. It applies K-weighting — a filter that approximates the ear's frequency response — and gates out silent passages before averaging across the whole program. The result is a single number that correlates with perceived loudness across different program material in a way RMS does not. Spotify, Apple Music, YouTube, and Tidal all use LUFS for normalization. A useful intuition: a record at -18 LUFS sounds quiet. A record at -14 LUFS sounds normal. A record at -10 LUFS sounds aggressive. A record at -6 LUFS sounds, on first listen, exciting, and on tenth listen, exhausting. **Dynamic range** (DR, the score reported by the TT Dynamic Range Meter and the Dynamic Range Database) is loosely the difference between peak and average. A record with DR12 has 12 dB of breathing room between its quiet moments and its loud ones. A record with DR4 is essentially flatlined — its average level is almost touching its peak the whole time. Higher DR is, by most measures, healthier sound. ### The war, in numbers The shift began in earnest in the mid-1990s and accelerated through the 2000s. Pre-CD-era pop and rock — late-70s through mid-80s — generally sat in the DR12 to DR14 range. By the late 1990s, average pop had moved to roughly DR8 to DR10. By the 2008 peak, mainstream releases were routinely landing at DR4 to DR6. In LUFS terms, the average integrated loudness of pop and rock masters shifted from around -18 to -20 LUFS in the early CD era to roughly -7 to -6 LUFS at the peak of the war. That is a swing of more than 10 dB — and in perceptual terms, each 3 dB doubles the apparent power. The technical enabler arrived in 1994: the **Waves L1 Ultramaximizer**, the first mass-produced digital brick-wall limiter with look-ahead. The L2 in 1999 made the workflow routine. Sound on Sound's later assessment was succinct: "for some, the sound of the 1990s and the 2000s was the L1 and L2." Suddenly any mastering engineer with a Pro Tools rig could shave the peaks off a master and raise its apparent loudness without producing audible clipping. The temptation was nearly irresistible. Labels heard the louder masters in A/B comparisons and assumed louder was better. The psychoacoustics agreed with them — the Fletcher-Munson equal-loudness contours, first measured in 1933 and refined into ISO 226, describe how the ear's frequency sensitivity is non-flat and changes with level. At higher SPL the ear hears more bass and more air. So the same EQ balance appears subjectively fuller at higher loudness, in a five-second comparison. This is the technical engine of the loudness war: a louder master wins every quick A/B by default, regardless of how it actually sounds over thirty minutes. The cost was paid by the music. Dynamics are emotional information. A song that gets quieter and then louder is telling you something about its emotional shape. Crush the master to a flat wall and that information goes missing. The quiet verse and the loud chorus sit at the same perceived volume. Every drum hit lands with the same force, because the kick that should hit hardest has been clamped down to the level of the snare ghost notes. The listener's nervous system, which is built to read dynamic change as meaning, gets nothing to read. ### The three canonical disasters **Metallica, Death Magnetic (2008).** The textbook case. Mastered by Ted Jensen at Sterling Sound; the Dynamic Range Database lists the CD master at **DR3**, against Master of Puppets at DR12. Audible clipping on the snare and the cymbals. Fans noticed within hours of release that the Guitar Hero III DLC version, which used a less-compressed pre-master sourced from the multitracks, sounded objectively better than the retail CD — reportedly about 10 dB quieter and noticeably more dynamic. The petition asking Metallica to remix the album passed 13,000 signatures inside weeks. Ted Jensen posted publicly to a Metallica forum: "In this case the mixes were already brick walled before they arrived at my place. Suffice it to say I would never be pushed to overdrive things as far as they are here. Believe me I'm not proud to be associated with this one." It is the rarest of mastering documents — the engineer disowning his own master. Andrew Scheps, asked years later, put it more bluntly: "My line is that the loudness war is over because I won. And that was the record that did it." **Rush, Vapor Trails (2002).** Original master = **DR5**. The 2013 David Bottrill remix and remaster = DR7. Geddy Lee, on record about the original: "We overcooked it. The mixes were really loud and brash. The mastering job was harsh and distorted." Alex Lifeson: "It was a contest, and it was mastered too high, and it crackles, and it spits, and it just crushes everything." The remix was not a touch-up. It was a confession. **Red Hot Chili Peppers, Californication (1999).** **DR4**. Distortion audible on the slide solo of "Otherside" and on the snare across the record. Mastered by Vlado Meller at Rick Rubin's request. The 2012 180-gram vinyl reissue, cut by Chris Bellman at Bernie Grundman Mastering from less-compressed sources, is the listenable version of the same record. A whole separate master, on a different medium, exists because the original master is — let's be plain — broken. For contrast, the triumphs of the same era and earlier. Donald Fagen's The Nightfly (1982), recorded on 3M's 32-track digital system and mastered by Bob Ludwig, sits at high dynamic range and is still the audiophile reference for clarity. Pink Floyd's Wish You Were Here original CD and the 2011/2025 remasters sit at DR11 to DR13. The Steely Dan catalog and the original Beatles masters routinely score DR12 to DR14. Grace — the Buckley album from the previous piece — sits in roughly the same band on the original 1994 CD; the 2004 Legacy remaster compressed it noticeably and is, audibly, the worse master. ### Bob Katz and the K-system The serious technical counter-proposal came in 2000. **Bob Katz**, mastering engineer at Digital Domain, presented "An Integrated Approach to Metering, Monitoring, and Level Practices" at the 109th AES Convention in Los Angeles. The paper proposed the **K-system**: three loudness scales (K-12 for broadcast, K-14 for pop and rock, K-20 for classical and film) calibrated to a standard monitoring level of 83 dB SPL. The idea was that if labels and engineers agreed on a reference loudness and consumer playback systems delivered consistent volume, the incentive to brick-wall masters for competitive loudness would evaporate. It did not get adopted. Labels had no incentive while the war was still profitable, and there was no consumer-side playback standard to enforce it. The K-system became a religious document — cited by every mastering engineer who took the problem seriously, ignored by most of the industry. Katz himself kept the faith. **Ian Shepherd**, the British mastering engineer, founded **Dynamic Range Day** in 2010 as an annual advocacy event. The argument was being made, persistently, that the war was a self-inflicted wound and that the music was the casualty. ### What streaming actually changed The reset that the K-system could not impose, the streaming platforms imposed by accident. Around 2014, Spotify, iTunes, YouTube, and the others rolled out loudness normalization. Today's targets: - **Spotify**: -14 LUFS integrated. Quiet tracks get turned up; loud tracks get turned down. - **Apple Music** (Sound Check): about -16 LUFS. Apple only turns down loud tracks; it does not boost quiet ones. - **YouTube**: -14 LUFS. Turns down loud tracks; does not boost quiet ones; never limits. - **Tidal**: -14 LUFS, album-normalized by default since the 2019 update influenced by Eelco Grimm's research. - **Amazon Music, Deezer**: roughly -14 to -15 LUFS. The mechanism is simple. A -7 LUFS brick-walled master gets turned down 7 dB by Spotify to hit -14. A -14 LUFS dynamic master plays through untouched. To the listener, both arrive at roughly the same average level. But the brick-walled master has already had its dynamics removed at the source; the dynamic master plays back with its breath intact. On the listener's speakers, after normalization, the dynamic master sounds louder and more alive. The incentive flipped. For the first time in twenty years, mastering loud was the worse strategy. Bob Katz declared the war won in early 2014. "The debilitating loudness war has finally been won. The last battle will be over by mid-2014." He was, on his own terms, right. The pure incentive to brick-wall is gone. Ian Shepherd has argued, persistently, that this is the wrong question. The platforms normalize, yes — but most chart releases still arrive at mastering already brick-walled by the producers and mixers, before the mastering engineer even sees the file. The normalized playback of a hot master is not the same as a dynamic master normalized to the same level. The dynamic information has to have been preserved at the source for the listener's ear to receive it. A -14 LUFS playback of a -7 LUFS source still sounds crushed; it just sounds crushed at a lower volume. The war as a practice persists even if the war as an incentive has ended. The honest answer in 2026 is that both are right. The incentive has flipped. The practice is changing slowly. Average DR of new pop releases sits around DR8 to DR10 — better than the DR4-6 trough of 2008, still well below the DR12-14 baseline of the analog era. The trend is upward. We are not back to The Nightfly, but we are also not living in 2008. ### The engineers, on the record The mastering community has been remarkably consistent about this since the early 2000s. **Bob Ludwig**, who retired in 2023 after fifty years and roughly twelve thousand mastered albums, recalled the Chinese Democracy sessions of 2008 in which he supplied three masters at different loudness levels and the band chose the most dynamic one: "Fan and press backlash against the recent heavily compressed recordings finally set the context for someone to take a stand and return to putting music and dynamics above sheer level." He named the war as the central technical fight of his career. **Emily Lazar**, of The Lodge, who has mastered records for David Bowie and Foo Fighters and won the Best Engineered Album Grammy: "The loudness wars are the bane of my daily existence." **Howie Weinberg**, formerly of Masterdisk: "I like nice loud files, but I don't want them crushed. There is a happy medium." **Andrew Scheps**, who mixed Death Magnetic and watched his work get crushed at the mastering stage: said in 2023 that his mixes "have gotten much quieter over the last 5-10 years." This is, in mixer-speak, an enormous statement. Scheps was famous for hot rock mixes. The reset has reached him. **Mandy Parnell**, of Black Saloon Studios, has built a practice around mastering for dynamics on analog chains — Sigur Rós, Björk, The xx, Aphex Twin. Her records do not crush, and they do not need to. ### Vinyl, and the strange new vinyl problem The vinyl revival was supposed to be the loudness war's natural antagonist. Vinyl has physical limits: the stylus has to track the groove without skipping; RIAA pre-emphasis boosts highs in cutting and rolls them off in playback; low bass eats groove width. A brick-walled digital master cannot be cut directly to vinyl without compromising stylus tracking and inner-groove distortion. The industry standard was, for decades, to cut vinyl from a separate, more dynamic master. That separate master is increasingly being skipped. Headphonesty's April 2025 reporting documented a pattern of cheap modern vinyl pressings cut directly from the loud digital retail master. The buyer pays for the vinyl premium, and gets, sonically, the same compressed master they could have streamed for free. The loudness war has migrated, by negligence, onto the medium that was supposed to defeat it. Audiophile labels — Mobile Fidelity, Analogue Productions, ECM, the better reissue programs — still cut vinyl properly from less-compressed sources. The mainstream catalog often does not. This is worth checking, if you care, before paying $40 for a record. ### The AI question A note on what the war looks like in 2026. AI music-generation systems (Suno, Udio, the new wave) trained on commercial pop have learned, by absorption, to output already-mastered audio at brick-wall loudness. The training data taught them that this is what a finished track sounds like. There is no separate "mix" or "master" stage; the model produces a finished-sounding artifact. Third-party AI mastering services (LANDR, eMastered) further compress this output by default. The result is that AI-generated tracks are, on average, even more dynamics-flattened than human-made commercial pop. This is a small thing and a large thing. It is small because individual AI tracks rarely get listened to at the depth where the loss of dynamics matters. It is large because the systems are training the next generation of listeners to accept compressed audio as the baseline. If your reference for "what music sounds like" is AI-generated content, you will not notice when a human record is given the same treatment, because you will not have heard the alternative. ### What this means for a Suede AI Social reader A short coda for anyone making records. If you are a producer, master quieter than your instinct tells you to. -14 LUFS integrated is the streaming target. Anything quieter is fine; the platform will turn it up. Anything louder will be turned down, and the dynamics you destroyed to be louder will not be restored. If you are a label or an artist, ask your mastering engineer for two masters: a streaming master at -14 LUFS with dynamics intact, and a vinyl master cut from a less-compressed source. If the engineer cannot tell you the LUFS or the DR of what they're delivering, get a different engineer. If you are a listener, the easiest test is this. Take a record you love that was made before 1990 and a current chart record. Play them at the same volume on the same speakers. Listen to the quiet moments. The old record will have moments that are quieter than the chart record's quiet moments. Those quiet moments are the dynamics that the modern record gave up to be louder. They are also, almost always, where the song's emotional information lives. The war is mostly over. The damage is partly reversible. The recipe is in every Bob Katz paper, in every Bob Ludwig master from 1976 to 2023, in every Andy Wallace mix that left room for the breath. Hallelujah, in 1994, had room for the breath. That is why it still sounds alive. Print the quiet. Master the quiet too. Print the Quiet is a Suede AI Social series on tone, dynamics, and the parts of music that don't fit on a lead sheet. Next: the religion of the compressor. ## The Religion of the Compressor Source: https://strumly.suedeai.ai/book/print-the-quiet/ptq-03-religion-of-the-compressor Why engineers worship the compressor: Fairchild to Distressor, parallel, sidechain, Clearmountain, Scheps, and the brick-wall crime of Death Magnetic. Print the Quiet essay by Jason Colapietro. Published 2026-06-13. ● Print the Quiet · 3 / 9 How the Fairchild, LA-2A, 1176, SSL G-bus and Distressor became gods, and why glue and breath are physical facts, not metaphors. By [Jason Colapietro](https://strumly.suedeai.ai/book), writing as Johnny Suede · A Suede AI Social long read · 11 min If you spend any time around recording engineers, you learn that the compressor is not just a device. It is a religion. They will argue for hours about it, in language closer to theology than electronics. This one breathes. That one glues. This one feels alive. That one is dead. They will spend $4,000 on a single channel of hardware because the input transformer is wound differently. They will discount a plugin shootout that they would have called definitive last week because today they tracked through an actual 1176 and remembered what the real one does. This is the third piece in the **Print the Quiet** series, and it is about that religion. What a compressor actually does. The lineage of the canonical units. The art of using one, and the crime of using one badly. And why engineers talk about it the way they do — because compression, more than any other tool in the studio, is where a piece of recorded music gets its body. ### The mechanism, plainly A compressor reduces dynamic range by attenuating signal above a chosen threshold according to a ratio. Five controls govern the result. **Threshold** is the level above which the compressor starts working. Below threshold, the signal passes untouched. **Ratio** is how aggressively the compressor reduces signal above threshold. A 4:1 ratio means for every 4 dB above threshold, only 1 dB comes out. Anything above 10:1 is effectively limiting — the compressor refuses to let the signal grow further. **Attack** is how quickly the compressor clamps down after the threshold is crossed. Fast attack catches transients (the snap of a snare hit, the consonant in front of a vowel). Slow attack lets transients through and only clamps on the sustain that follows — which preserves the snap but sacrifices peak control. **Release** is how quickly the compressor lets go. Too short, and the compressor "pumps" — you hear it grabbing and releasing in time with the music. Too long, and the compressor stays clamped, choking the next phrase. **Knee** is how gradually compression engages around the threshold. A hard knee is abrupt; a soft knee feels transparent and musical. There is also the detection method. **Peak** detectors react to instantaneous transients — the 1176 is a peak-style FET compressor. **RMS** detectors average signal level over a window and feel more like the way the ear perceives loudness — the dbx 160 and the LA-2A live in that world. That's it, mechanically. Every compressor is a variation on those parameters. The religion lives in the variations. ### The lineage A short canon, with the technical detail that matters to engineers and the cultural detail that matters to anyone trying to understand why pre-1995 records sound the way they do. **Fairchild 660 / 670 (1959).** Designed by Rein Narma at the Fairchild Recording Equipment Corporation. The 670 is the stereo dual-mono version, with 20 tubes (the mono 660 has 11). Both use the rare 6386 remote-cutoff tube. Geoff Emerick brought the Fairchild into the Beatles' chain at Abbey Road starting with Revolver — Beatles vocals, Ringo's drum bus on "Tomorrow Never Knows," piano. Used 660s now sell for $20,000 and up; 670s push past $40,000. Tony Visconti, who has used a Fairchild on most David Bowie records: "The Fairchild was the go-to black box of sonic glue for the entire drum kit or the bass guitar, or even both." It is the original color compressor. You hear it on records the way you hear weather in a memory. **Teletronix LA-2A (1962).** Jim Lawrence's electro-optical tube leveling amplifier. The mechanism is unusual: an internal element called a T4 cell converts the audio into light, the light hits a photoresistor, and the photoresistor controls the gain. Because optical attack is naturally slow and program-dependent, the LA-2A is the vocal compressor — slow enough to not interrupt phrases, even enough to feel transparent on long sustained notes. It has, famously, two knobs: peak reduction and gain. There is no attack control. There is no release control. There is no ratio control. It does what it does, and the engineer's only decision is how much. The simplicity is the religion. (Side note: the "1965" date often cited is the year Babcock Electronics acquired Teletronix's patents; the unit itself was introduced 1962.) **UA 1176 (1968).** Bill Putnam's FET (field effect transistor) compressor, peak-detecting, famously fast — attack times as short as 20 microseconds. The aggressive one. The lead-vocal-in-modern-pop one. The drum-bus one. Revisions matter: - **Rev A "Bluestripe"** (1967-68): no low-noise circuitry, gritty, distorted, the most aggressive-sounding rev. - **Rev C, D, E** (1970-73): the LN low-noise circuit added; the most revered revs and the basis of modern UA reissues. - **Rev H "Silverface"** (mid-70s): input transformer removed, replaced with an op-amp input — the cleanest-sounding rev. The 1176's defining party trick is the "all-buttons-in" / British Mode: pressing all four ratio buttons at once produces a non-linear, aggressive, harmonically rich squash. The unit was never designed to do this. An engineer pressed the buttons by accident, liked it, and the move entered the canon. The most famous compressor color in pop music is a happy hardware bug. **dbx 160 (1976, not 1971 — that's the year David Blackmer founded dbx).** The first commercially successful VCA (voltage-controlled amplifier) compressor. Hard-knee, fast, punchy. The drum-room sound of the late 1970s and 1980s — Hugh Padgham, Bob Clearmountain, anyone tracking rock drums with serious attack. The Blackmer Gain Cell, patented in 1973, is the technical foundation of the VCA compressor as a category. **SSL G-bus compressor.** First appeared on the SSL 4000 B (1976), refined on the 4000 E (1981), reached definitive form on the G-series in 1987. Built into every SSL console's center section. VCA-based, feed-forward topology, derived from a dbx 202 black-can VCA chip. Because every SSL console had one, every record mixed on an SSL from 1985 to 2005 had at least the option to glue the 2-bus with this comp. Most did. The "SSL sound" — the bright, cohesive, slightly-over-glued mid-period CD-era pop master — is partly this one chip. **Empirical Labs Distressor (1995).** Dave Derr, who had been an Eventide engineer working on the H3000, designed the Distressor as a digitally-controlled analog compressor with mode switches that emulate the behavior of an LA-2A, a Fairchild, an 1176, and tape. "Dist 2" emphasizes second-harmonic content (tube-like); "Dist 3" emphasizes third-harmonic (tape-like); "Opto" mode produces LA-2A behavior. The Distressor is the most-used outboard compressor in modern professional studios because it can play any role in a session. It is the Swiss Army of the rack. **Manley Variable Mu (1994).** EveAnna Manley's tube compressor, using the same remote-cutoff tube transconductance gain reduction mechanism as the Fairchild. Soft, slow, expensive. The mastering 2-bus and tracking-bus comp for engineers who want analog tube character without the Fairchild's price tag (which is, charitably, ridiculous). ### The craft There are two kinds of compression. Compression you hear: the pumping 1176 on a drum bus that you can audibly hear breathing in time. The released kick that lets the room come up between hits. The vocal that compresses you can hear working on the consonants. That is sometimes the point — Phil Collins's "In the Air Tonight" gated reverb on the drums depends on a compressor and a gate you can identify by ear. Compression you feel: the SSL bus comp doing 1 to 2 dB on the master and you don't notice it until you bypass it and the mix falls apart. The light Fairchild on a vocal bus that adds tube color and "glue" without changing the shape of any phrase. The Distressor on a bass guitar that locks the bottom end in place so the kick can punch through. The terms engineers use here — glue, breath, body, air — are not metaphorical. They are physical descriptions of what compression does to the cumulative envelope of a mix. Glue is when several sources start sharing the same dynamic envelope, so they feel like one event instead of five layered events. Breath is the audible release of a slow comp letting go between phrases. Body is the gain-reduction that pulls the sustained portion of a note up relative to the transient. Air is what you have left in the high frequencies when the compressor hasn't eaten them. The fastest way to learn the difference between compression you hear and compression you feel is to put a stereo bus compressor across the mix bus, set it to a 2:1 ratio with a slow attack and an auto release, and watch the meter. If the needle is moving 1 to 2 dB at the peaks, you are doing it right. If the needle is jumping 8 dB on every chorus, you are doing it wrong — you have replaced the song's dynamic shape with a mechanical breathing pattern. ### Parallel, sidechain, multiband Three techniques worth naming. **Parallel compression** — sometimes called the New York drum trick — runs a heavily-compressed copy of a signal on a separate channel and blends it back under the original uncompressed signal. The compressed copy supplies the body and the sustain; the uncompressed original supplies the transient snap. The blend gives you body without losing punch. Bob Clearmountain popularized it on Springsteen, the Stones, and Bryan Adams. The term "New York Compression" was coined by Bobby Owsinski in The Mixing Engineer's Handbook (1999), but the technique predates the term — Motown's "exciter" bus on vocals in the 1960s was a parallel scheme, and the internal architecture of Dolby A noise reduction (1965) was built on parallel encode/decode buses. Andrew Scheps formalized parallel compression into his whole-mix philosophy: a post-fader send of every track in the mix (except the drums) into a single stereo aux bus, smashed hard, blended back at low level under the main mix. He calls it the "rear bus" because, on the Neve quad console where he developed the technique, the unused quadraphonic rear bus became his parallel comp send. The technique is now standard practice in pop mixing. **Sidechain compression** uses one signal to trigger the compression of another. The most familiar version is the kick-side-chained-bass trick of house and EDM music — every kick triggers the bass to duck briefly, making space in the low end. The technique works equally well, and more subtly, in pop and rock: side-chaining a synth pad to the lead vocal so the pad ducks whenever the vocal is present. Side-chaining a reverb return to the dry source so the reverb only blooms in the gaps. Used judiciously, sidechain compression is invisible. **Multi-band compression** splits the signal into frequency bands and compresses each independently. Useful for problem-solving — taming a resonant frequency on a vocal, glueing a too-broad master, fixing a mix that should have been remixed. It is also the cardinal sin of mastering engineers who use it as a substitute for actually fixing the source mix. There is a saying among older engineers: if you need multi-band compression to make a mix work, the mix is wrong. ### The signature uses A short list of who did what, and why it matters. **Geoff Emerick on the Beatles.** Fairchild 660 on Ringo's drum bus on "Tomorrow Never Knows" (the loop-driven Revolver track) — the compressed kit becomes part of the song's drone. The same Fairchild on Lennon's vocals from Sgt. Pepper onward, giving the vocal that tube-glued forward presence that has never quite been replicated since. **Hugh Padgham on Peter Gabriel and Phil Collins.** The "gated reverb" sound of the 1980s — first heard on Peter Gabriel's third self-titled album (1980), then on Phil Collins's "In the Air Tonight" (1981, with Collins playing the drums) — was not actually invented as a reverb trick. It was an accident. The SSL 4000 console's talkback mic had a built-in heavily-compressing limiter (so studio chatter could be heard at low level), and Padgham discovered that routing the drum room into the talkback compressor and then through a gate produced an enormous, slamming, distinctly-1980s drum sound. The gated reverb is a compressor's accident. **Bob Clearmountain on Springsteen and Bryan Adams.** Parallel compression on drums, drum-sample reinforcement, fast 1176-style on lead vocals. Clearmountain's mixes from 1982 to 1995 essentially defined what a rock record sounded like on commercial radio. **Andrew Scheps on Adele's 21** (Grammy Album of the Year): "I probably did the least amount of processing I've ever done — one mix had only five EQs in total, a little bit of spring reverb, and a compressor across the bus, because it was just the sound of the tracks." The restraint is the credit. Scheps's record is, in part, what he did not add. **Chris Lord-Alge** (Green Day, U2, Aerosmith, Springsteen, hundreds of others): the most aggressive compression hand in mainstream rock. CLA is documented hitting 15 to 20 dB of gain reduction on a Bluestripe 1176 on lead vocals, post-EQ, with a 4:1 ratio on the master bus pulling another 7 dB. The result is hyper-real, radio-ready, slamming. You can love it or hate it; you cannot deny it works on commercial rock radio. **Tchad Blake** (Tom Waits, Sheryl Crow, Pearl Jam): the Shure Level-Loc compressor, originally designed for paging systems and police-band radios, as a deliberate distortion-as-compression tool. Blake discovered it on Tom Waits's Bone Machine. His own line: "The Level-Loc changed my life." The Level-Loc is awful as a transparent comp and astonishing as a character tool. The lesson: a compressor's job is not necessarily to be invisible. **Andy Wallace** (continuity from earlier in this series): on the Grace sessions, parallel compression on the drum bus, light Fairchild-style on Buckley's vocal, conservative SSL G-bus on the master at perhaps 2 to 4 dB of gain reduction. The dynamic range of the Grace master — DR11 to DR13 — is the receipt for what restrained, intentional compression buys you. ### The crime There are bad uses too. Worth naming. **Brick-wall mastering as compression abuse.** The loudness war (see the previous piece in this series) was, technically, compression and limiting deployed as a loudness tool rather than as an artistic tool. The mastering chain stopped trying to shape the music and started trying to make it louder. Metallica's Death Magnetic (2008) at DR3 is the canonical disaster. Ted Jensen, who mastered it, publicly disowned the result: "In this case the mixes were already brick-walled before they arrived at my place. Suffice it to say I would never be pushed to overdrive things as far as they are here. Believe me, I'm not proud to be associated with this one." The mixes — by Andrew Scheps, who knows better — had been compressed too hard before mastering even started. The compressor was being asked to do what only the song could do. **The LANDR sound.** Algorithmic mastering services default to heavy compression because heavy compression always sounds immediately better in an A/B with an uncompressed source — for the Fletcher-Munson reasons explored in piece #2. Documented analyses of LANDR and eMastered outputs show crest factors as low as 11 dB (very brick-walled), hyped high-mids, dense sub-100 Hz limiting. The compressor, in these tools, is no longer being driven by an engineer's ear. It is being driven by an algorithm optimizing for "sounds loud in three seconds." **Compression as masking.** Inexperienced producers learn, often the wrong way, that heavy compression can hide a thin vocal or an uneven drum performance. It can, briefly. The cost is that the body that was uneven shows up later — in the listener's fatigue, in the song's lack of depth, in the eventual realization that what you are hearing has been pressed flat to disguise problems that weren't worth disguising. Compression cannot fix performance. It can only smooth its surface. ### Why the religious language Why do engineers talk about compressors the way mystics talk about saints? A reasonable answer: compression is, more than EQ or reverb, the tool that most directly shapes the body of a recorded sound. EQ changes the spectral balance — you hear the difference in terms of frequencies. Reverb adds space — you hear the difference in terms of room. Compression changes the envelope of a sound through time, and envelopes through time are how the nervous system reads emotional intensity. A fast-attacked drum hit is exciting because the transient is sharp. A slow-attacked vocal is intimate because the sustain is brought forward. Compression is the tool that most directly controls how the listener's body feels what is on the record. When engineers say a compressor "breathes," they mean it lets the music's natural dynamic envelope show through with subtle reinforcement. When they say it "glues," they mean it pulls several sources into a shared envelope that the body reads as one event. When they say it is "alive," they mean it is not crushing the envelope but tracking it. These are physical descriptions, even if the language sounds spiritual. The compressor is the throat of the recording. It is the part of the chain that decides how the body of the sound arrives at the listener's body. That is why engineers talk about it the way they do. It is also why the loudness war was a tragedy on a human scale, not just a technical one. The throat of the recording got clamped shut. The body that the listener was supposed to feel never arrived. ### A coda for the working producer Three principles, from people who have spent their lives compressing things well. **Use less.** Most engineers, including Scheps, Wallace, and Clearmountain, will tell you they compress less now than they did fifteen years ago. The instinct, when in doubt, is to add more gain reduction. The right move, almost always, is to add less. **Listen to the release.** The single most-misset knob on a compressor is the release. A release that is too fast pumps. A release that is too slow chokes the next phrase. The right release lets go just before the next attack — which means the right release is, in some sense, defined by the song's tempo and phrasing. If your compressor's release is at one setting for the whole song, it is wrong for at least half of it. Modern compressors with program-dependent release exist for this reason. **Bypass.** The discipline that separates competent compressionists from religious ones: bypass the compressor every few minutes. If the music sounds worse, you are compressing well. If it sounds the same, you are compressing too much. If it sounds better uncompressed, you have compressed too much. The compressor is the most powerful tool in the studio and the most easily abused. Used well, it is what makes a record feel alive. Used badly, it is what makes a record feel dead. There is no neutral setting. The decision is always tonal, and tonal decisions are always moral. Print the quiet. Compress the body. Bypass to check. Print the Quiet is a Suede AI Social series on tone, dynamics, and the parts of music that don't fit on a lead sheet. Next: the lost art of clean. ## The Lost Art of Clean Source: https://strumly.suedeai.ai/book/print-the-quiet/ptq-04-lost-art-of-clean Essay 4 of Print the Quiet: what a clean guitar tone costs and what it buys, from "Sultans of Swing" to Larry Carlton's $50K Dumble and SRV's "Lenny." Print the Quiet essay by Jason Colapietro. Published 2026-06-13. ● Print the Quiet · 4 / 9 A clean tone is a courtroom where every pickstroke is on trial. Knopfler, Hendrix, SRV, and Larry Carlton paid the price; their records show what it buys. By [Jason Colapietro](https://strumly.suedeai.ai/book), writing as Johnny Suede · A Suede AI Social long read · 9 min Listen to the first eight seconds of "Sultans of Swing." There is no gain. There is no fuzz. Mark Knopfler is fingerpicking a red 1961 Stratocaster into a Fender Vibrolux, with maybe a touch of compression and almost nothing else, and the tone is already doing more storytelling than most records manage with a wall of overdubs. We have, as an industry, mostly forgotten how to do that. The default tone setting for a modern guitar is dirty. Most plugin amp sims open on a pre-distorted preset. Most modern rock records reach for gain inside two bars. The decision tree for what a clean tone is for — what it can carry, what it asks the listener to do, what it withholds — has thinned to the point where many young players have never deliberately worked on a clean sound for its own sake. This is the fourth piece in the **Print the Quiet** series. The first three argued — through Jeff Buckley, through the loudness war, through compression — that tone is a value system, a set of decisions about what to leave in and what to take out. This piece is about the most foundational subtraction: not adding distortion. What that costs the player. What it buys the song. ### What a clean tone actually is The technical definition is simple: a clean tone is one where the amplifier's output stage is operating below the threshold where the tubes (or transistors) begin clipping the waveform. The signal that goes in is the signal that comes out, amplified but not reshaped. The harmonic content is dominated by the fundamental and a few low-order harmonics produced by the natural non-linearity of the gain stages. In practice this is more complicated. Most "clean" tones on record have at least some compression and some EQ shaping. Many — Knopfler's, Larry Carlton's, John Mayer's — sit at the edge of breakup, where the amp adds just enough harmonic complexity to be sweet without crossing into recognizable distortion. The technical term for this is the amp's breakup point, and the most beautiful clean tones in the history of recording are sitting one or two volume-knob clicks below it. That is where the wood of the guitar, the fingers of the player, and the speaker's compression all show up clearly. Add any more gain and you start replacing those signatures with the amp's signature. Add any less and the tone goes sterile. What clean does not do: it does not hide. Distortion is a great forgiver. It blurs pitch, it smooths attack inconsistency, it fills the spectrum so that a missed note disappears into the noise. Clean is a courtroom. Every articulation, every uneven pickstroke, every dead-string buzz, every flat note is audible. Players who work in clean for their living tend to have technique that is almost embarrassing in its precision, because they have no choice. ### The tradition, briefly A short canon of records that demonstrate what a clean electric tone can do, with the gear context that matters: **Mark Knopfler — Dire Straits, Dire Straits (1978) and Brothers in Arms (1985).** A 1961 Strat (and later, a National resonator and a red Schecter), fingerpicked, into a Fender Vibrolux or a Music Man HD-130. Knopfler doesn't use a pick — his right hand is doing a Travis-derived thumb-and-fingers thing he learned from Chet Atkins and James Burton. The clean tone is the song's narrator. On "Sultans," the lead lines do the dialogue. **Hendrix — "Little Wing," "Castles Made of Sand," "May This Be Love."** Strat into a Marshall, but with the amp's master volume backed off so the preamp isn't clipping. A Vox Wah and a Uni-Vibe sit on the floor for chorusing-style modulation, but the amp is clean. The chord work on "Little Wing" — the way the thumb wraps around the low E to play the F# while the fingers do the upper-voice melodies — is a clean-tone clinic. The whole emotional weight of "Castles Made of Sand" depends on the tone being uncluttered enough that you hear the chord voicings as poetry. **Stevie Ray Vaughan — "Lenny," "Riviera Paradise," the clean side of In Step.** SRV is mythologized as a tube-melting overdrive player, but half his greatest playing was clean. "Lenny" is a 1965 Strat (the one his wife gave him) into a Vibroverb, with the amp's volume conservative and a Tube Screamer kicked off. The vibrato in his fingers is doing what other players try to do with a delay pedal. **Larry Carlton — "Room 335" (1978), and a decade of session work.** A 1968 Gibson ES-335 through a Dumble Overdrive Special — the famously hand-built Howard Dumble amp, $50K+ on the used market now. JBL D-130F speakers in a custom cab. The Carlton tone is the perfect rounded-but-articulate clean: full bottom, no flub, top end that sings without ice. It is one of the most-imitated guitar tones in the history of recording, and almost no one gets it because almost no one has a Dumble and the patience to dial it in for three hours before tracking. **Pat Metheny — anything from Bright Size Life through Travels.** A Gibson ES-175 into a Lexicon multi-effects unit doing chorus-and-delay duty, into a small Acoustic Image or Polytone amp. The Metheny tone is so liquid that listeners often mistake it for a synth. It is in fact a hollowbody electric jazz guitar played with a pick at the precise distance and angle that produces a perfectly round attack envelope. **Andy Summers — The Police, all three records.** A 1961 Telecaster Custom with a humbucker in the neck, into a Marshall JMP-50 kept clean, with a Boss CE-1 Chorus Ensemble and an Electro-Harmonix Memory Man Deluxe doing the spatial work. Summers' chord voicings — the suspended fourths, the open strings ringing under fretted thirds — only work because the tone is clean enough to let you hear all six strings as discrete events. **Keith Richards on "Wild Horses" and "Beast of Burden."** Open-G tuning on a Telecaster (the famous "Micawber," with the bridge pickup pulled out and a 1950s Tele pickup installed at the neck), into a small Ampeg or a Mesa Mark I, clean. The strumming on "Wild Horses" is so quiet and so present that you can hear the pick hitting the strings. That is what a clean tone makes possible. **Lindsey Buckingham — Fleetwood Mac, "Never Going Back Again" and "Big Love."** Almost the limit case for what fingerpicked clean electric can do. Lindsey doesn't use a pick — ever — and his right hand on a Turner Model 1 or a Rick Turner-built nylon-string is producing something that crosses classical guitar with folk-rock storytelling. These are guitar examples, but the principle generalizes. The clean piano on D'Angelo's Voodoo. The clean Rhodes on every Stevie Wonder record from Talking Book through Songs in the Key of Life. The undistorted bass on every Motown record. The clean trumpet on Kind of Blue. The undriven vocal on a Sinatra master. Every one of these is a tone that refuses to hide. ### Why we stopped doing it Several reasons converge. **The amp simulator default.** When players started learning on plugin amp sims — first Line 6, then Helix, Neural DSP, Tonex — they encountered a UI in which the presets are pre-distorted. Manufacturers know that gain sells; a clean preset on a demo sounds anemic compared to a high-gain preset doing chugs. New players learn the gain knob first. The clean tone is something they discover later, if at all. **The loudness war hangover.** When records are mastered loud (see the previous piece in this series), clean tones get crushed in the mix. Their natural dynamic range — the difference between a fingerpicked note and a strummed chord — is exactly what compression eats. Producers learned that distorted tones survive the mastering chain better. So they reached for gain to protect the part. The part survives loud mastering, but the song loses the quietness it was supposed to have. **Bedroom production**. A real clean tone requires a real amp, in a real room, mic'd at a real distance. Bedroom producers can't crank a Vibrolux at 4 in the morning; they DI into a plugin sim. Plugin sims are now extraordinary — Neural DSP's clean models are technically very close to the real thing — but they capture less of the room and the speaker, and the most beautiful part of a clean tone is the air the speaker pushes into the room. **The retreat from intimacy as a commercial position.** Modern radio has limited room for quiet. The Spotify algorithm rewards songs that maintain energy from the first second. A clean intro is, in the current attention economy, a risk — the listener might swipe. **The de-skilling of the right hand.** This is harsh but worth saying. Pick-attack technique, dynamic control, the ability to play one note quieter than the previous note — these are skills that take a decade to acquire. Distortion lets you skip the decade. The clean players above all spent twenty years on right-hand control before they ever cut a record. That apprenticeship has thinned. ### What clean costs the player Clean asks for things that most players do not want to give. It asks for tuning. A clean chord exposes any tuning drift; distortion masks it. The clean players above use locking tuners, change strings on a strict schedule, and check tuning between songs without breaking the take. It asks for consistent right-hand attack. The angle of the pick, the depth of the pluck, the position over the pickup — all of these directly shape the tone, and inconsistency in any of them is audible. Clean players spend hours working on attack consistency. Some of them — Knopfler, Buckingham — refuse to use a pick because they can't get the consistency they need from one. It asks for restraint. The hardest thing about a clean tone in a band setting is not playing more. Cleaner tones suggest fewer notes; the player has to leave space. Many otherwise competent guitarists physically cannot leave space, and you can hear it in their playing. It asks for amplifier discipline. The cleanest tones are produced by tube amps running at moderate volumes, where the output transformer is doing some natural sweetening but the tubes are not yet breaking up. That requires a real amp. It also requires a real-volume tracking session, which most bedroom situations cannot accommodate. ### What clean buys the song A few things, in our view, that nothing else can buy. **Air.** A clean tone occupies less of the frequency spectrum, leaving room for the voice, the bass, the drums. Most over-distorted productions feel cluttered because the guitars have eaten the midrange real estate. Clean guitars give the vocal somewhere to live. **Pitch as content.** Clean tones reveal note-bending, vibrato, and microtonal nuance. Distortion smears all of that into a continuous wash. If your player has an interesting vibrato — Knopfler's, SRV's, Hendrix's — you want it heard. **Dynamics as content.** A clean tone responds linearly to right-hand attack. Quiet plucks stay quiet; hard strums get loud. That dynamic information is part of the song's storytelling. Distortion compresses dynamics chemically, making every attack sit at roughly the same level. You can't tell the chorus from the verse in dynamic terms. **The room.** A clean amp in a room sounds like the room. The early reflections, the speaker pushing air, the cabinet resonance — these are part of the tone. A distorted amp masks all of that under the gain's noise floor. The room is the most expensive thing a studio has, and clean is how you put it on the record. **Authority.** This is harder to measure but worth saying. A great clean tone has a kind of moral authority on a record. It announces that the player is not hiding behind effects, that the part has been considered, that the song does not need to be loud to be important. Tea for the Tillerman. Pink Moon. Court and Spark. Blood on the Tracks. These records do their work in part because their tones refuse to oversell. ### The Hallelujah connection The arpeggios on Buckley's Hallelujah are a clean tone, and the whole tonal argument of that record turns on the cleanness. The 1983 Telecaster, in standard tuning with a capo at the 5th fret, fingerpicked, into a small Fender combo with the Quadraverb adding a controlled halo. No distortion. No gain. The amp is in its sweetest unbroken-up zone, the speaker is moving conservative air, the room is contributing reflections, and Buckley's right hand is doing all the dynamic work. That tone is the song. If Wallace had put Buckley through a Mesa Trem-O-Verb in dirty mode, the same chords would not have done what they did. Distortion would have replaced Buckley's right hand with the amp's voice. The quiet verses would have lost their quietness. The whisper-to-roar dynamic arc that defines the take would have collapsed into a constant medium-loudness. Clean is what made Hallelujah possible. Clean was the precondition for what Buckley wanted to say. ### A short note on AI The state of AI music generation as of 2026 is that the models are extraordinarily good at producing plausible distorted tones, and noticeably less good at producing convincing clean tones. The reason is informative: distortion is a high-information signal where small inaccuracies are masked by the noise. Clean tones are low-information signals where every detail is audible, and the details that matter — the speaker's compression curve at 3.7 watts, the pick's exact angle at the moment of attack, the room's reflection at 12 feet — are exactly the things that are not in the score. Clean is, currently, where the human keeps showing up most clearly on records. That may change. But it tells us something about what tone is and where it lives. Tone is what survives the subtraction of everything that can be added. Clean is the practice of subtracting until only the player remains. ### How to start, if you want to A short coda for players reading this. Plug a single-coil guitar into a small tube combo — a Princeton Reverb, a Vibrolux reissue, a Tweed Deluxe. Turn the amp's volume up until it is just shy of breakup. No pedals. Sit twelve feet from the speaker and listen to what the amp does to your right hand. Spend a week playing one chord progression — say, the Hallelujah progression — and listen to how the tone changes as you vary the pick attack. You will be embarrassed at first. The cleanness will reveal every inconsistency in your playing. The notes will not all sound the same volume. Your tuning will drift. Your pick will catch unevenly on the strings. That is the apprenticeship. That is the thing that has thinned. It is also the thing on every record that you have ever loved. Print the quiet. Print the Quiet is a Suede AI Social series on tone, dynamics, and the parts of music that don't fit on a lead sheet. Next: the room as instrument. ## The Room as Instrument Source: https://strumly.suedeai.ai/book/print-the-quiet/ptq-05-room-as-instrument Why Hallelujah, Rumours, and Back to Black owe their sound to the rooms they were cut in — Bearsville to Daptone — and what convolution reverb can't fake. Print the Quiet essay by Jason Colapietro. Published 2026-06-13. ● Print the Quiet · 5 / 9 Bearsville, Sound City, Muscle Shoals, Electric Lady, Capitol, Power Station, Daptone — and the one instrument your plugins still can't fake. By [Jason Colapietro](https://strumly.suedeai.ai/book), writing as Johnny Suede · A Suede AI Social long read · 11 min A microphone does not capture a guitar. It captures the guitar, plus everything that the guitar's sound does on its way to the microphone — the wood floor it bounces off, the plaster wall behind the amp, the ceiling thirty feet up. By the time a signal reaches tape or hard drive, the room has already played its part. The room is on the recording the same way a violin's body is in the violinist's note. You do not get one without the other. We have, over the last twenty years, mostly stopped using rooms. Bedroom production, plugin reverb, close miking into acoustically treated nearfield environments. There are good reasons for the shift — cost, accessibility, isolation. There is also a real loss, and the loss is the subject of this piece. The fifth in the **Print the Quiet** series. About the room as an instrument, and what specific records owe to specific rooms. ### The acoustic vocabulary A few terms, briefly, so the rest of the piece reads cleanly. **RT60** is reverberation time: how long it takes a sound, after the source stops, to decay by 60 decibels. A bathroom is around 1.5 seconds. A well-treated mixing room is intentionally short, around 0.15 to 0.3 seconds. A cathedral can run past 8 seconds. **Early reflections** are the first bounces, arriving within about 30 milliseconds of the direct sound. Your auditory system does not perceive them as separate echoes — it fuses them with the original sound and reads the result as size. Early reflections are why a vocal in a big room sounds big, even if you cannot hear a distinct reverb tail. **Late reverberation** is the dense, diffuse tail you do hear as a separate ambience — what most people mean when they say "reverb." **Room modes** are resonant frequencies whose wavelengths fit neatly into the room's dimensions. At those frequencies, sound builds up or cancels depending on where you stand. A kick drum can boom in one corner of a room and disappear in another. **Hard surfaces** (concrete, wood, plaster) reflect sound. **Soft surfaces** (carpet, drapes, fiberglass) absorb it. A high ceiling delays the first vertical reflection long enough that the ear hears it as ambience rather than coloration. A wood floor warms the high end. A concrete wall slaps it back. Same band, two rooms, two records. This is not an exaggeration. ### Bearsville Studios, Woodstock, NY The story has to start here, because Hallelujah was tracked here. Albert Grossman — Dylan's manager, Janis Joplin's manager, the Band's protector — bought a fifteen-acre farm in Woodstock in the mid-1960s and built a recording compound on it, opening in 1969. Studio A's tracking room is a converted post-and-beam barn, with the main room running approximately 35 by 35 feet and a ceiling that rises to roughly 35 feet at its peak. Imagine that volume in wood: a cathedral-sized space inside a wooden envelope. The reflections off the timber are warm and complex. The vertical decay is long but musical. The room is large enough to track a band live and small enough that the players can hear one another without headphones. R.E.M. cut parts of Out of Time and Automatic for the People at Bearsville. Patti Smith, Bob Dylan, the Band, Todd Rundgren as producer for what felt like a decade of records, Phish for Junta. And Andy Wallace, in September 1993, chose Bearsville Studio A for Grace because he wanted to track Buckley's band live in one room — bass, drums, Buckley playing and singing simultaneously, with the room contributing its barn-air as a fifth instrument. A close mic on Buckley's Telecaster captures the speaker. A room mic twelve feet back captures the speaker plus the barn: the early reflections off the timber that tell your ear how large the air is around the amp, the late tail that tells your nervous system you are listening to something happening inside a 35-by-35-foot wooden cathedral, not into a Shure-into-a-laptop close-mic. That room information is what makes Buckley's guitar and Buckley's voice sound like they are in the same world. It is what bedroom production cannot easily reproduce. It is also what makes the record sound, thirty years later, alive. ### Sound City, Van Nuys, CA A former Vox factory converted in 1969 by Tom Skeeter and Joe Gottfried into a recording studio. Studio A was famously ugly — brown shag wallpaper, a linoleum floor over concrete (the often-repeated "hardwood floor" detail is wrong; check any documentary footage). The room was rectangular, the ceiling was modest, the proportions were unremarkable. By every textbook of acoustic design, Sound City Studio A should have been undistinguished. It produced one of the most-recorded drum sounds in late-twentieth-century rock. The reason was the linoleum-over-concrete floor, the proportions of the room, and the **1973 custom Neve 8028 console** — one of only four ever built. The hard floor delivered fast, attack-heavy reflections back at the drums; the room dimensions produced a tight, controllable low-end response; and the Neve preamps gave everything that touched them a particular saturated forwardness that has become shorthand for "expensive rock record." Tracking the drums there involved minimal microphones, the kit dead-center in the room, the Neve doing the rest. The receipts: Fleetwood Mac's Rumours (tracked 1976-77, the most-listened-to drum sound of the 1970s). Tom Petty's Damn the Torpedoes. Nirvana's Nevermind. Rage Against the Machine's self-titled debut. Slipknot's first three records. Johnny Cash's American Recordings. Sound City closed in 2011 — the economics of a single-room high-end studio could not survive bedroom-production economics — and Dave Grohl bought the Neve 8028 console and installed it at Studio 606. Grohl told NPR: "There's a life to it. There's more to that board than just wires and capacitors and knobs and shit. There's something inside. It's meant to be used forever." The room is gone. The console — the room's other half — survives. The records are the receipt. ### Muscle Shoals: FAME and Muscle Shoals Sound There are two rooms here, and both matter. **FAME Studios** at 603 East Avalon in Muscle Shoals, Alabama, was Rick Hall's room — built into a former drugstore. Aretha Franklin cut "I Never Loved a Man" there in January 1967, with the Swampers — Roger Hawkins, David Hood, Jimmy Johnson, Spooner Oldham, Barry Beckett — as her backing band. The session lasted one day before personal tensions blew it apart, but the master had already been cut, and the master is the record. In 1969, four of the Swampers left FAME to open their own studio: **Muscle Shoals Sound Studio** at 3614 Jackson Highway in Sheffield, Alabama, in a converted concrete-block coffin showroom. The room was small, rectangular, rigid concrete-block on the walls. By every textbook this room should have been a nightmare — modal problems, ringing, no diffusion. What you hear on Wilson Pickett's "Mustang Sally," Percy Sledge's "When a Man Loves a Woman," Etta James's "Tell Mama," and the Rolling Stones' Sticky Fingers sessions in December 1969 — "Brown Sugar," "Wild Horses" — is what the room actually did. It refused to give the drums a tail. The kit hit and the energy bounced hard and fast and died. That tight, thuddy, dry attack is the concrete-block walls rejecting reverberation. The dryness is the tone. The Muscle Shoals records sound the way they sound partly because the room was the wrong room. Or, more accurately, because the wrong room turned out to be the right room for a kind of music that wanted dryness, attack, and immediacy. Acoustic design and acoustic accident converge here. ### Electric Lady Studios, NYC Jimi Hendrix and engineer Eddie Kramer commissioned architect John Storyk to design Electric Lady in the basement of 52 West 8th Street in 1968-70. It opened in August 1970, three weeks before Hendrix died. Hendrix's instruction to Storyk: no right angles. Storyk responded with curved walls (he has acknowledged Gaudí as an influence), porthole windows on the doors, ambient lighting tubes that change color with the music, and a sloped ceiling rising toward the control room. The acoustic personality is medium-live. Hardwood floor, carpeted walls in places, several reflective surfaces. Kramer's stated reference was Olympic in London. "Never too live nor too dead." Storyk has admitted the low-frequency absorption of the ceiling was "partially an accident" — a happy one. The catalog is overwhelming. Stevie Wonder cut Talking Book, Innervisions, and Songs in the Key of Life at Electric Lady, mostly to tape, with the room contributing the warm reflections that make those records sound the way they sound. Patti Smith's Horses. The Roots — D'Angelo's Voodoo engineered by Russell Elevado, with Elevado on record that "85 percent is live" — meaning the room is doing as much of the work as the players. More recently: Frank Ocean's Channel Orange, Taylor Swift, Lana Del Rey. Electric Lady is the most-booked classic room in the world in 2026 because what it does cannot be plugin-replicated. The medium-live curve, the wood floor's high-end character, the basement's low-frequency stillness — these are physical conditions. ### Abbey Road Studio 2 The room is 38'3" by 60'2", with a ceiling around 24 feet — tall, partly because the building started life as a Victorian townhouse and the studios were carved out of larger volumes. The walls carry absorbent paneling; the ceiling is a grooved paper-fiber treatment that softens the high reflections without killing them. EMI's REDD valve consoles and, later, the TG12345 solid-state desk lived in the control room above. Plate reverbs in another room and the famous EMT chambers fed everything from John Lennon's vocals to Pink Floyd's Dark Side of the Moon. The Beatles cut almost everything there. The tall volume is why a Ringo tom sounds big without being washy: the early reflections give size without producing a long late tail. The Beatles' drum sound is unrepeatable not because the kit was special — it was a standard Ludwig — but because the kit was in that room, and the room is part of the kit. Geoff Emerick describes the room's character in his memoir Here, There and Everywhere: the grooved paper-fiber ceiling, the absorbent panels, the cumulative effect of decades of orchestral recording having tuned the room without anyone setting out to tune it. The Studio 2 character is not designable from scratch. It is what fifty years of use did to the volume. ### Capitol Studios, Hollywood When the Capitol Records tower went up in 1956, Les Paul designed eight trapezoidal echo chambers thirty feet underground beneath the parking lot. Concrete walls. Room-within-room construction. Interior walls, ceilings, and floors shellacked to a hard reflective finish to maximize decay. Chambers 1 through 4 in the original 1956 build; chambers 5 through 8 added in 1969. Each chamber tuned individually by varying wall angles and ceiling slopes. Decay times up to roughly five seconds. The Capitol chambers are the reverb on Sinatra's In the Wee Small Hours. On Nat King Cole. On Beck's Sea Change. On Paul McCartney's Memory Almost Full. When you hear a Sinatra master and the voice has that warm, slow, blooming tail behind it, you are hearing a microphone in the studio plus a speaker in a chamber thirty feet underground driving the same signal through forty feet of shellacked concrete and recapturing it through a different mic. The chamber is the reverb, and the chamber is a building. Universal Audio's Capitol Chambers plugin samples those chambers via impulse response and is genuinely convincing. It is also genuinely not the same thing as putting a signal through the actual chambers and recapturing it, because impulse-response convolution is, by definition, a linear measurement, and the chambers behave slightly differently at different SPL levels in ways a linear measurement cannot capture. Close, not identical. ### Power Station, NYC Tony Bongiovi, with architects Stephen B. Jacobs Associates, converted a former Con Edison power substation at 441 West 53rd Street into a recording studio in 1977. Studio A is a pine-paneled 52' × 48' tracking room with 35-foot ceilings — the high ceiling preserved from the original substation vault. The pine walls and floor give the room a particular tonal warmth; the height delays vertical reflections; the combination produces the famously big, ringing, glossy drum sound of the high-1980s. Bowie cut Let's Dance there in 1983 with Nile Rodgers producing. Springsteen's Born in the U.S.A. Madonna's Like a Virgin. Chic's later records. The very-eighties slamming drums on every one of those records are partly the high pine ceiling and the wood paneling working together as one enormous tuned chamber, with Phil Collins-style gated reverb processing arriving at the master via that natural acoustic boost. The room renamed itself Avatar Studios in 1996, then Berklee NYC in 2017, and remains in operation. The acoustic character — the pine, the ceiling, the volume — is unchanged. ### Daptone House of Soul, Bushwick A counter-example. Gabriel Roth and Neal Sugarman built Daptone in a two-family converted house on Troutman Street in Bushwick, Brooklyn, on the deliberate theory that the slightly-bad — uneven floors, leaky isolation, exposed ceiling joists — is the sound of the records they love. All-analog signal chain. Tascam Series 85-16 tape machine. No Pro Tools. The Stax/Muscle Shoals records of the late 1960s were cut in compromised rooms because that's what the artists could afford. Daptone reverse-engineered the compromise. Sharon Jones and the Dap-Kings cut their catalog there. Charles Bradley. And, in 2006, Mark Ronson brought Amy Winehouse to Daptone to cut "Back to Black," "Rehab," "You Know I'm No Good," and "Love Is a Losing Game" with the Dap-Kings and the Antibalas players. The sound of those records — the small-room, dry, slightly-saturated 1960s Stax/Daptone signature — is partly the room being acoustically compromised in precisely the right way. The Daptone lesson is that "good" is contextual. A perfectly treated mixing room produces a particular kind of clinical truth. A small wood-paneled house produces a particular kind of soul-record truth. The room's job is not to be acoustically correct. Its job is to be the right kind of incorrect. ### Honorable mentions A few rooms that deserve their own pieces. **Hansa Tonstudios, Berlin.** The Meistersaal is a 650-square-meter ballroom with a 15-meter coffered ceiling, herringbone parquet floor, and tall windows that, in the 1970s and 1980s, looked toward the Berlin Wall. Bowie cut Heroes there. The famous "blooming" vocal on the title track was Tony Visconti placing three microphones at increasing distances — close, mid, far — and gating the more distant ones by volume threshold so they opened up only as Bowie sang louder. The room, that big, is what made the technique possible. **Real World Studios, Box, Wiltshire.** Peter Gabriel's 1987 conversion of an 18th-century mill into a recording compound. The Big Room is 2,000 square feet with a wraparound SSL console and millpond views. The room sounds like a mill — wooden beams, stone walls, a particular warm slow decay. **RAK Studios, London.** Mickie Most's St. John's Wood conversion of an ex-school building. Adele recorded much of 25 there. Arctic Monkeys. Pino Palladino. The rooms are small, wood-paneled, and acoustically excellent for vocals and small ensembles. **Headley Grange.** Not strictly a studio. A three-story Victorian poorhouse in Hampshire where Led Zeppelin recorded Led Zeppelin IV in late 1970 and early 1971. Andy Johns set John Bonham's kit at the foot of the three-story stairwell and hung two Beyer M160 ribbon mics two floors up. The compressed stairwell sound on "When the Levee Breaks" — through Helios F760 limiters with a Binson Echorec delay — is the building doing the work. No reverb plugin recreates that drum sound because the energy is being channeled vertically up a narrow column with hardwood treads and plaster walls. Headley Grange is the drum sound. ### What bedroom production loses, honestly It is fashionable, in some circles, to dismiss bedroom production as inferior. The honest version of the argument is more nuanced. What bedroom production gains: anyone can make a record. Anyone can experiment. Anyone can develop ideas in private until they're ready to bring them to a band or a studio. The democratization is real and important. What bedroom production loses: the room. Close-miked sources, plugin reverb tails. Modern convolution reverbs (impulse responses of real rooms, including Capitol's chambers) are genuinely good. They reproduce the late tail of a real room accurately. What they reproduce less well: the nonlinear behavior of a room at different SPL levels (a room sounds slightly different at 95 dB than at 75); the way the back of a guitar amp couples to a wood floor and the floor's vibration adds low-frequency information to the recording; the cross-modulation of multiple sources sharing one acoustic space; the way a singer in a real room subtly adjusts their performance because they can hear themselves in that room. A plugin reverb tail attached to a dry close-miked source is, mathematically, a tail attached to a source. A real-room recording is a coupled physical system. The nervous-system difference is real, but — let's be honest — often subtle. Closer to a wine question than a category error. Bedroom records can be extraordinary. They are different from room records, and the difference compounds across an album. ### Why this matters for the rest of the series The thread through Print the Quiet is that tone is the part of music that is not in the score, and that the parts that are not in the score live in the body — the singer's body, the player's hand, the amp's tubes, the speaker's cone, and the room. The room is the largest and most easily forgotten of these. When Hallelujah was cut at Bearsville Studio A in September 1993, the room was tracked alongside Buckley. The barn-cathedral volume is on the record as much as the Telecaster and the U87. Strip the room out and you have a different record. Replace it with a plugin reverb and you have a different record. The reason listeners describe Buckley's Hallelujah as "spacious" or "open" or "intimate" is that they are responding to thirty-five feet of wooden volume reflected back at the microphones. The room is the part of the recording that did the work for free. Bedroom production has, by necessity, replaced that work with software. The software is closer than it used to be. It is not the same. Print the quiet. Print the room. Print the Quiet is a Suede AI Social series on tone, dynamics, and the parts of music that don't fit on a lead sheet. Next: tape is a compressor. ## One Take, or Twenty Source: https://strumly.suedeai.ai/book/print-the-quiet/ptq-06-one-take-or-twenty Glyn Johns won't punch in. Pop comps at the syllable. Buckley's Hallelujah found the middle: 20 takes, 3 chosen, no fixing inside them. Print the Quiet essay by Jason Colapietro. Published 2026-06-13. ● Print the Quiet · 6 / 9 Why phoneme comping breaks something the ear can't name, and how Andy Wallace and Jeff Buckley found the honest middle at Bearsville. By [Jason Colapietro](https://strumly.suedeai.ai/book), writing as Johnny Suede · A Suede AI Social long read · 8 min There is a thing a singer does on take seven that they could not have done on take one. They have learned the song's geography. They know which line lifts and which line drops. They have started shaping the second verse as a response to the first verse, the chorus as the answer the verses were asking for. By take eleven they are getting good. By take eighteen they are committing — gambling, leaning into the risk that this take is the one, that they can let go and trust the shape they have built up underneath them. The take after the take they trust is usually worse. They tighten. They overthink. They start performing the version of the song they think they should sing instead of singing the song. A great engineer hears this coming and stops them. A great engineer also knows that the next take, after the bad one, is often the best — because the singer has scared themselves into letting go. This is what a take is, and it is why "did you keep all the takes?" is the most important question in any vocal session. The thing being made on tape is not a sequence of independent attempts at the same thing. It is a continuous emotional negotiation in time. Each take is a response to the take that came before it. The singer's body is learning. This is the sixth piece in the **Print the Quiet** series. It is about what happens to that negotiation when modern recording lets us comp at the syllable level — and about the honest middle ground that Andy Wallace and Jeff Buckley found at Bearsville thirty years ago. ### The Glyn Johns school Glyn Johns engineered, produced, or mixed records for the Stones, the Who, the Faces, the Eagles, Led Zeppelin (the Led Zeppelin and II sessions), Bob Dylan, Eric Clapton, and Joan Armatrading, among others. He has a working philosophy that has not changed for fifty years: capture the live performance, do not punch in, do not fix later. Either the take is good or it is not. If it is not, you do another take. You do not stitch. Johns has said, in interviews and in his memoir Sound Man, that he refuses on principle to use the punch-in. He has been mocked for this — it is, technically, antique — and he has been hugely influential because of it. The reason: punch-ins introduce a microsecond discontinuity that the ear may not consciously hear but the body clocks. The phrase that started before the punch and finishes after the punch is no longer one phrase. It is two phrases stitched together. Johns argues that listeners can feel it, and that the loss compounds across a record. The same logic explains his famously minimal drum miking — three microphones on the Stones' "Honky Tonk Women," four mics on the Who's "Won't Get Fooled Again" — which now bears his name as the **Glyn Johns method**. He wanted to capture the kit as one event in one room. Adding mics and gates and triggers, in his view, would have replaced the kit with a representation of the kit. The takes had to be the kit, in the room, at the time. This school has its inheritors. **T-Bone Burnett** records bands live to tape with vintage gear and lets the takes stand. **Daniel Lanois** is a comper, but he comps from full takes, not phonemes. **Rick Rubin**, perhaps the modern Johns, has said in many interviews that the first usable take is usually the best one, and that he stops singers from singing too many times — he wants them to commit before they get bored. **Steve Albini**, until his death, refused even to call himself a producer because the word implied imposition; his job was to capture what the band did, in the room they did it in. This is one pole. ### The phoneme-comp era The other pole is what mainstream pop has done since approximately 2002, which is when ProTools 5 made phoneme-level editing trivial and engineers learned to do it. The workflow is roughly this. The singer records ten to thirty full passes of the song over the course of a day or three. The engineer or vocal producer then opens all of them on adjacent ProTools tracks, lined up to a tempo grid, and assembles the released vocal by choosing the best version of each syllable — sometimes the best version of each consonant. The "ch" in church might come from take seven. The vowel from take twelve. The closing "sh" from take three. Each fragment is then re-tuned with Melodyne for note and timing, breath samples are removed or replaced with cleaner ones from elsewhere in the session, sibilance is de-essed, and the result is a vocal that is — by all measurable standards — perfect. Listen to almost any major-label pop release from the last fifteen years. The vocal is on the grid. Every consonant is clean. The vibrato is even. The breath sounds are present but they are selected breath sounds, not the breath the singer happened to take after the line. The pitch is correct. This is not a complaint about effort. Doing this well is a real craft. There are great phoneme-comp engineers, and the results, technically, are stunning. The complaint is about the causal chain. ### The causal chain of phrasing Here is what we mean. When a singer sings line four of a verse, the way they sing it is causally linked to how line three just felt. If line three caught them by surprise — too much breath spent, too much emotion in the wrong place, a bend that wobbled — they compensate on line four. They tighten the grip, or they let go further, or they re-attack the next vowel with extra commitment. The body is responding to itself in real time, the way an actor's third sentence in a monologue is informed by their first two. This is causation. Line four is because of line three. Now consider phoneme comping. Line three is taken from one performance; line four is taken from another. The line-four singer was not responding to the line-three singer. They were responding to a completely different version of line three, in a different take, where they were in a different emotional state, where the vibrato had landed differently, where the room was, microscopically, a different room. Listeners cannot articulate the discontinuity. But — and this is the point that matters — listeners' nervous systems are exquisitely tuned to causation in vocal sound, because evolution built them to read other humans' speech for causal coherence. When someone you know is upset, you can hear it across one sentence. The same circuitry is running when you listen to a record. Phoneme comping breaks the causal chain. Not always audibly, but always structurally. The result is technical perfection that listeners describe, vaguely, as feeling "off" or "cold" or "too much" — without being able to say why. The why is that the causal information has been removed, and their nervous system is reading the absence. Comping at the full-take level — Andy Wallace's method on Hallelujah, Glyn Johns's method on his rare modern jobs, Daniel Lanois's standard practice — preserves the causal chain inside each take. The comp lives at the seams between full performances, not inside individual phrases. The body that sang line four was responding to the body that sang line three on the same pass. The continuity holds. This is why a full-take comp can feel like a single performance and a phoneme comp, however perfect, cannot. ### Sinatra, briefly Frank Sinatra is the most quoted single-take legend in popular music. The story is that he walked into Capitol Studios in the 1950s, recorded each song in one or two takes with the band live, and walked out. The story is mostly true. Nelson Riddle, Billy May, Gordon Jenkins, the arrangers — they wrote the charts knowing Sinatra would not be doing them again. The band knew it. Sinatra knew it. There was, in some sessions, a third take held in reserve, but the released master was almost always the first or second pass. The technical reason this worked: Sinatra had spent twenty years on the road before he started cutting masters at Capitol. He arrived at the session with the song already inside him. He was not learning the geography. He had already learned the geography on three hundred nightclub stages. By the time tape rolled, he was reporting on a relationship he had already established with the lyric. This is not how modern pop is made. Modern pop is made by writing the song on the day, recording it on the day, and finding out what the song is during the recording. That process generates many takes because the singer is in the act of discovering the song on tape. There is nothing wrong with that workflow, but it is a different workflow, and it requires more comping. The honest version of the argument is not that comping is bad — it is that the modern workflow has necessitated heavy comping by changing the order of operations. What's been lost in the change is the apprenticeship. Sinatra's first take worked because he had already done two thousand takes, just not on tape. The modern singer's twentieth take is doing the apprenticeship and the master at the same time. ### The honest middle: Buckley at Bearsville Wallace cut about twenty full takes of Hallelujah over the Grace sessions. The released master is a comp of somewhere between three and five of those takes. Wallace and Steve Berkowitz disagree, in print, about how much stitching there was — Wallace remembers a light comp, Berkowitz remembers a more substantial one. Alan Light's book The Holy or the Broken captures the disagreement and does not resolve it. What is resolved is the structural fact: the comp was at the full-take level, not the syllable level. Wallace was choosing between complete emotional arcs. The seams between takes are at line boundaries or verse boundaries, not inside words. The causal chain inside each segment is intact. When you hear Buckley sing "the holy or the broken Hallelujah," the whole phrase is one performance — Buckley's body responding to itself across the breath, the bend, the resolution. That is the honest middle ground. Not romantically one take. Not the modern phoneme comp. Twenty performances captured, three-to-five chosen, glued at structural boundaries, no pitch correction (Auto-Tune didn't yet exist), no time correction inside the takes. The result is a recording that listeners cannot tell from one performance because, in the way the body reads it, it is one performance. ### What this means for the working musician A practical coda. If you are a singer or an instrumentalist working in a modern session, three things follow. **One: track in full passes, not punch-ins, whenever you possibly can.** It will cost you takes. It will also preserve the causal chain. Tell your engineer you would rather do twelve full passes than punch four bars at a time. The full passes will compound — by pass nine you will have learned the song in a way punching cannot teach you. **Two: ask your engineer to comp at the line level or the verse level, not the syllable level.** This is a conversation worth having before the session. Engineers will accommodate the request. The result will be a vocal that sounds more like you and less like everyone. **Three: resist the urge to fix the imperfect take.** The slightly-wobbly take is often the truer one. Listeners' bodies read the wobble as evidence that you were there. Punching it out replaces the evidence with a stand-in. The stand-in sounds correct and feels less. If you are a producer or engineer, one thing follows. **Ask whether the take you are stitching needed the stitch.** Sometimes it did. Often the unfixed take, with one questionable note, is the take. Listeners are forgiving of one off-pitch syllable inside a committed performance. They are unforgiving — at the level of the nervous system, even if not at the level of conscious thought — of a flawless performance assembled from twenty competing emotional states. ### A note on AI vocals It is worth saying out loud where this argument lands in 2026. AI vocal generation is now extraordinary at the phoneme level. The models can produce a clean, in-tune, emotionally plausible vocal performance of any song in any style in less time than it takes to read this sentence. What they have not yet learned to produce, well, is the causal chain across phrases — the because of relationship between line four and line three. Phrases in current AI vocals tend to be locally beautiful and globally disconnected, in the same way that early phoneme comps were. The body is not there to respond to itself. That may change. But it tells us, again, where the human is showing up most clearly. The human is in the causation. The human is in the line that is a response to the line that just happened. The human is in the body that sang both lines on the same pass and let one inform the other. Twenty takes, three chosen, no fixing inside the chosen ones. That is the recipe. Buckley wrote the receipt thirty years ago. Print the Quiet is a Suede AI Social series on tone, dynamics, and the parts of music that don't fit on a lead sheet. Next: tape as a compressor. ## Tape Is a Compressor Source: https://strumly.suedeai.ai/book/print-the-quiet/ptq-07-tape-is-a-compressor Why pre-1995 records sound warm and glued: tape was a compressor, saturator and head-bump bass booster running free on every track until ProTools. Print the Quiet essay by Jason Colapietro. Published 2026-06-13. ● Print the Quiet · 7 / 9 For sixty years, magnetic tape ran soft compression, harmonic warmth, and head-bump bass on every track for free. Then ProTools pulled it off. By [Jason Colapietro](https://strumly.suedeai.ai/book), writing as Johnny Suede · A Suede AI Social long read · 14 min The cleanest way to describe what happened to recorded music between 1995 and 2010 is this: a piece of processing that had been on every track of every record for sixty years stopped being on every track of every record. The piece of processing was magnetic tape. Tape is a compressor. Tape is a saturation device. Tape is a low-pass filter at hot levels. Tape is a head-bump low-end booster. Tape is a transient softener that operates the moment the signal hits the record head, on every channel, simultaneously, for free. When digital recording replaced tape in mainstream studios — the transition is conventionally dated to ProTools' arrival in 1991 and its mainstream adoption between 1995 and 2000 — that whole chain of automatic, free, distributed processing came off. Records made after the transition did not sound bad. They sounded different. The difference is what this piece is about. The seventh in the **Print the Quiet** series. ### What tape actually does to a signal A short technical primer, because the rest of the argument depends on getting this right. **Soft saturation of transients.** Magnetic tape has a non-linear transfer function. As input level rises past nominal operating level, the oxide's magnetic particles re-orient progressively until they cannot be re-oriented further. The signal does not hard-clip the way a digital converter does at 0 dBFS — it gradually rounds over. Engineers call this soft clipping or saturation compression. Every transient that hits tape at a hot level gets gently flattened. A drum hit's sharp attack becomes a slightly rounded attack. A pick-stroke on a snare becomes a softer pick-stroke. A vocal consonant gets a quarter-decibel of natural smoothing. This happens automatically, on every track, the moment the signal hits the record head. It is not a setting. It is physics. **Harmonic generation.** Tape produces a mix of even and odd harmonics, with second-order content prominent at moderate levels and third-order content emerging at higher levels. Different tape formulations (Ampex 456 from 1975, the 1990s-era Quantegy 499 and GP9, BASF 911, the modern RTM SM900) have measurably different harmonic profiles. The popular shorthand of "tape adds second-harmonic warmth" is partly true and partly mythology; the honest version is that tape adds some harmonic content, predominantly low-order, and listeners read that content as "warmth." **High-frequency self-erasure.** This one is underdiscussed. At high flux levels, a strong present signal can magnetically erase a previous one — self-erasure — and the practical effect is that tape cannot hold its highest frequencies at hot levels. Above roughly 10 kHz, the response gracefully softens as the level rises. This is the technical mechanism behind tape's natural HF roll-off at hot recording levels. Cymbal sibilance gets a small amount of softening for free. Vocal "s" sounds get gently tamed. The de-essing the engineer would otherwise have to apply has been partly done by the medium. **Head bump.** At 15 IPS — the standard recording speed of most 1970s-90s multitrack work — every analog tape machine exhibits a low-frequency rise (typically about +2 dB) around 50 to 80 Hz, caused by the playback head's geometry. This is the "fat low end" engineers chase. Kick drums, bass guitars, the lower register of male vocals all sit in this band, and tape gives them a small natural boost. At 30 IPS the bump moves higher in frequency and away from kick/bass; you trade punch for cleaner top end. **Wow and flutter.** Tape transport mechanical imperfections add subtle pitch modulation. On a well-aligned Studer A800, the spec is in the 0.05% range — barely measurable. But it is non-zero. Cumulatively, across an album, that micro-modulation contributes to the slightly organic character of tape playback that perfectly steady digital playback does not have. **Noise reduction systems.** Dolby A (1965), Dolby SR (1986), dbx Type I — all complementary encode/decode systems that reduce the noise floor while introducing their own subtle character. Dolby SR is widely regarded as the best-sounding NR system; dbx delivers more noise reduction but can audibly pump on low-frequency content. The presence of any of these in the chain is part of the record's tonal signature. Put together, tape is a multi-stage automatic processing chain: soft compression on transients, gentle low-order harmonic generation, gentle high-frequency rolloff at hot levels, low-end head-bump, micro pitch modulation, and characteristic noise treatment. Every track, every channel, every time. Free. ### The canonical machines A brief lineage, because gear-religion is part of the story. **Ampex 350 / 351 / AG-440** (1953-1967). Tube electronics through the 350/351, transistorized AG-440 from 1967. The Capitol Studios Sinatra-era machines. Quarter-inch two-track for masters; later AG-440 in 8 and 16-track configurations for tracking. The Capitol catalog from the mid-50s onward is the sound of these machines plus the Capitol chambers plus Voyle Gilmore producing. **Studer A80** (1970-1988). The modular solid-state workhorse, 8/16/24-track configurations. Over 10,000 units sold. The standard high-end European machine through the entire classic-rock era. **Studer A800** (1978). The first microprocessor-controlled 24-track two-inch machine. 900 pounds. The standard upgrade from the A80 in high-end studios from the late 1970s through the digital transition. Grace — the Buckley album from earlier in this series — is widely cited as a Studer A800 record, though primary confirmation from Andy Wallace on the exact machine is harder to pin down than the gear hagiography suggests. Treat the claim as well-supported and not airtight. **Otari MTR-90** (early 1980s). The mid-range professional machine; less expensive than a Studer, very widely deployed in working studios. **Sony PCM-3324** (1982) and **PCM-3348** (1989). The digital crossover machines. Donald Fagen's The Nightfly (1982) was cut on a Sony 3M digital system and is famously high-dynamic-range — proving early that digital, properly used, was not the enemy. **Tape formulations matter.** Ampex 456, released 1975, was the most-used master tape of the 1970s and 1980s — and is infamous now for sticky-shed syndrome, a binder-decomposition problem requiring tapes to be baked at low temperature before playback. Ampex/Quantegy 499 was hotter. Quantegy GP9 (released 1998) was hotter still, with an operating level 9 dB above standard. BASF 911 was the German alternative. The supply chain post-2010 has been fragile; the current professional tape supplier is **Recording the Masters (RTM)**, a Mulann subsidiary that took over the old BASF/EMTEC formulations in 2015 after RMG/RMGI went into receivership. There are essentially no other suppliers of professional analog tape stock in 2026. If RTM stops, the medium ends. ### Why pre-1995 records sound the way they do Now we can assemble the argument. Take a 1985 multitrack session. The drummer is playing into seven or eight microphones. Each microphone is being recorded to a separate track on a 24-track 2-inch tape machine. Each track is independently going through the tape stage — record head, oxide saturation, head bump, HF rolloff, harmonic generation. Then the drum bus gets bounced to a separate stereo pair on the same multitrack, and the bounce passes through tape again. Then the entire mix gets summed to a half-inch two-track master, and the master passes through tape a third time. Then the master gets EQ'd and mastered for vinyl or CD, and a fourth analog stage adds further processing. The cumulative effect is enormous. A drum hit that began as a transient with a 30-microsecond rise time has been softened, harmonically enriched, gently HF-rolled, and head-bumped four separate times before it reaches the listener. None of those processing stages were chosen by an engineer. None of them were dialed in. They were imposed by the physical medium. This is what every record from 1955 to roughly 1995 has in common. It is the unifying processing signature of the analog era. The "warm/glued" character of pre-1995 records is not nostalgia. It is the audible residue of an enforced physical processing chain that lasted forty years. When ProTools replaced tape, the chain came off. The drum hit's transient now arrived at the master with its original 30-microsecond rise time intact. The cymbal's high frequencies were not naturally rolled. The kick was not naturally head-bumped. The mix was not naturally glued. Every record after the transition had to add back what tape had done for free — through plugin processing, through software emulation, through deliberate distortion stages — or accept the new, sharper, less-glued sound that digital natively produced. The "digital sounds harsh" complaint of the late 1990s and early 2000s is sometimes dismissed as audiophile snobbery. It was not. It was an accurate report of a real loss. Three factors contributed: - **Loss of tape's automatic processing chain.** The strongest case. - **Converter limitations.** 16-bit was standard through the late 1990s. 24-bit converters from Apogee, Lavry, and Prism Sound became professional norms only between 1998 and 2002. Early digital converters did have real, audible limitations that have since been fully solved. - **Loudness war acceleration.** The Waves L1 brick-wall limiter shipped in 1994 (see piece #2 in this series). Average pop loudness moved from roughly -18 LUFS in the early CD era to -7 LUFS by 2008-10. The records were not just transitioning from tape to digital. They were also being crushed harder than they had ever been crushed. The two factors compounded. Modern converters in 2026 (Burl B2, Prism Lyra, Lavry Gold, Apogee Symphony) are transparent enough that "digital harshness" is no longer a credible technical complaint. The remaining issue is the loss of tape's automatic processing — which good engineers now restore with plugin emulation. ### The plugin landscape Tape emulation plugins are, in 2026, a serious category. The honest version of the question is whether they capture what tape does. **UAD Studer A800.** The flagship. Models four tape formulations (3M 250, Ampex 456, BASF 900, Quantegy GP9), three speeds (7.5/15/30 IPS), bias setting, calibration level, hiss, and hum. The Studer A800 emulation is widely regarded as the deepest digital simulation of analog tape behavior available. **UAD Ampex ATR-102.** The mastering favorite. Models the half-inch two-track Ampex ATR-102 mastering deck. Transformer modeling, head configurations, multiple tape formulas. The ATR-102 plugin is the one most consistently cited by working mastering engineers as a "game changer." **Slate Digital Virtual Tape Machines (VTM).** Models a Studer A827 and an Ampex MM1200. Less calibration depth than UAD; lower CPU; favored by engineers who want tape character without obsessing over the parameters. Dave Schiffman (Nine Inch Nails, Weezer) has said publicly he has not turned on his real ATR-102 since adopting VTM. That is the strongest practitioner endorsement of the "plugin is good enough" position. **Waves J37.** Models Abbey Road's 1-inch four-track Studer J37, with three EMI tape formulations. The only emulation officially blessed by Abbey Road Studios. **Softube Tape.** Algorithmic — not a specific machine model — engineered for transparency and CPU efficiency. **Newer entries**: Brainworx bx_townhouse, Wavesfactory Cassette (a deliberately lo-fi cassette emulator), Baby Audio TAIP (a neural-network model of a 1970s machine). The category is getting better every year. The honest assessment, based on null-tests reported on Gearspace and Production Expert: the best emulations capture 90 to 95 percent of the audible character of real tape. The remaining 5 to 10 percent is the cumulative-pass behavior on a real multitrack (where the same tape sees the signal multiple times across overdubs and bounces), the subtle inter-channel crosstalk of a 24-track head stack, and the non-stationarity of a real tape transport. Whether you can hear those things in a finished mix is genuinely contested. Dave Schiffman cannot. Steve Albini, before he died, could. ### The Jack White / analog-purist position Jack White cuts to tape at Third Man Records. Lazaretto (2014) and subsequent records were tracked to two Studer A800 machines with JFR Magnetic Science 8-track headstacks linked with timecode for 16-track work. White also cuts direct-to-acetate masters and famously edits tape with a razor blade rather than software. His argument, paraphrased across multiple Tape Op and Popular Mechanics interviews: "If you put it onto tape, for me it was always like, well, that's the way the drums are. They're on there. That got rid of fifty percent of the choices immediately. The number one rule of being an artist is knowing when to stop." The case here is not nostalgia. It is constraint as creative discipline. Tape forces decisions. You cannot have 47 takes if you only have one reel; you have to commit. You cannot endlessly comp phonemes if you are working on physical media; you have to choose a take and live with it. The constraint, White argues, is the tone. The decisions tape forces are the decisions that make a record itself. **Steve Albini** at Electrical Audio held a similar but distinct position. His was less about constraint and more about archival permanence: "I feel like my fundamental obligation as a recording engineer is to make a historical record. There is nothing today you can use as an equivalent digital master tape that you can put on a shelf and will still be there in 20 or 50 years." Albini's records were on tape because tape was the medium he believed would survive. He died in May 2024; his studio survives him. **Gabriel Roth at Daptone**: "Show me a computer that sounds as good as a tape machine and I'll use it." All tracking is to 8 or 16-track tape. No ProTools in the building. The Amy Winehouse / Sharon Jones / Charles Bradley records emerge from that policy. ### The case against tape worship The honest counter-position deserves a hearing, because it is real. **Bob Clearmountain** — one of the most influential mixers of the last forty years — prefers digital. His argument: tape kept changing the playback against the original take. Every time a tape was played, it lost a microscopic amount of high-frequency information through self-erasure. A track tracked dry on Monday and played back Friday after twenty rough mixes was a slightly different track. "Whatever the tape did that changed the sound was making a decision about the tone that wasn't his." For an engineer who valued precise control over the final mix, tape was a meddling middleman. **Tony Visconti**, despite producing nearly every major Bowie record to tape, uses Pro Tools at 96kHz/32-float floating-point for the bulk of his current workflow. "Both he and David found working on tape too slow." The romance of tape did not, in Visconti's case, outweigh the workflow advantage of non-destructive digital editing. **The supply chain problem.** Two-inch Ampex 456 is $300+ a reel when available. RTM is essentially the sole professional supplier, and the supply has been precarious since 2012. Sticky-shed syndrome turns half of the existing 1980s/90s tape archives into baking projects. The economics of an all-tape studio in 2026 are punishing. **The transparency point.** An engineer interviewed by Production Expert made an underappreciated argument: the original designers of high-end analog equipment were chasing transparency, not the saturation we now romanticize. The "character" of analog gear was a byproduct of design constraints — leakage, transformer saturation, tube non-linearity, tape's S-curve — that the designers were trying to minimize. What we now treasure as the analog sound is the residue of imperfection that the engineers of the era considered a flaw. This is a legitimate way to think about it. It does not diminish the result. It does relocate the romance. ### What this means for the tone-as-truth thesis The deep point of this piece, and of the series, is this: tape is a physical constraint that produces a specific sonic signature. The magnetic particles can only re-orient so far. The pole pieces have a geometry. The transport has mechanical limits. These constraints generate the texture. The texture is the tone. Digital recording before plugin processing had no such constraints. The signal hit the converter and was stored without alteration. Engineers had to add character back deliberately, decision by decision, plugin by plugin. The character became chosen rather than enforced. That is a different kind of recording. Not worse, not better — different. AI-generated music has no constraint at all. The model can produce any spectrum at any moment, with no physical medium imposing a signature. This is why AI tracks, even technically excellent ones, often sound either generic or weirdly perfect. The constraint is the part that gives a recording a body. The body is the part that listeners' nervous systems read as alive. Producers know this. Jack Antonoff is on the record about Taylor Swift's Midnights: "Midnights is quite synthetic... but actually very analogue. So much was going into tape." The synthesis is digital; the printing medium is analog. The reason is that the printing medium imposes a body that pure digital does not. The bigger lesson — and this is the Print the Quiet thread — is that tone, in recorded music, is the residue of constraints honored. The tape head's geometry. The room's reflections. The amp tube's plate curve. The singer's body. The compressor's release time. None of these are arbitrary parameters in a model. They are physical facts being captured. The capture is the tone. ### A practical coda For the working producer in 2026, a short list. **If you can track to tape, do it for the basic tracking pass.** A pass through a real Studer or Otari on the basic tracking is forty years of automatic processing for free. Bouncing back to digital for overdubs and mixing is the modern compromise. Daptone and Third Man use exactly this workflow. **If you cannot track to tape, use the best emulation plugin you can afford, on every track.** UAD Studer A800 on individual channels, UAD ATR-102 on the master bus, is a workable approximation. Slate VTM is a CPU-friendlier alternative. Use it light. Tape is not subtle, but it is also not aggressive. **Use the constraint principle even without tape.** Limit your takes. Limit your plugin chain. Make decisions early. Print your effects rather than leaving them on inserts. Bounce sub-mixes rather than carrying every channel through to mastering. The constraint discipline that tape imposed mechanically can be imposed deliberately. It is harder, because it requires you to choose, but the result is closer to what tape produced. **Listen to a Studer-tracked record and a clean digital record back to back.** Same speaker. Same volume. Notice what the Studer record has that the digital record does not. Then ask whether your records have it. If they do not, you might think about why. Print the quiet. Print the constraint. Print the Quiet is a Suede AI Social series on tone, dynamics, and the parts of music that don't fit on a lead sheet. Next: the vocal chain, 1994 to 2026. ## The Vocal Chain, 1994 → 2026 Source: https://strumly.suedeai.ai/book/print-the-quiet/ptq-08-vocal-chain-1994-2026 From Buckley's U87 at Bearsville to Adele's Sony C800G and Suno v5.5 — how the vocal chain went from five stages to fifteen, and what the listener lost. Print the Quiet essay by Jason Colapietro. Published 2026-06-13. ● Print the Quiet · 8 / 9 Jeff Buckley's Hallelujah passed through five stages. A 2026 major-label lead runs fifteen. Here's everything added between the mouth and the master. By [Jason Colapietro](https://strumly.suedeai.ai/book), writing as Johnny Suede · A Suede AI Social long read · 11 min In September 1993, when Jeff Buckley stood at a microphone at Bearsville Studios and sang Hallelujah, the signal path between his voice and the final master had roughly four to five processing stages. Microphone, preamp, light compressor, tape machine, mix-bus compressor. Maybe a touch of light mastering. That was it. In May 2026, when a major-label pop vocalist tracks a lead in a New York or Los Angeles studio, the signal path between their voice and the final master typically has fifteen or more stages. Microphone, preamp, tracking compressor, Auto-Tune or Melodyne, comp from a dozen takes, surgical EQ, de-esser, dynamic EQ, fast peak comp (an 1176-style unit), slow opto comp (an LA-2A-style unit), multiband compression, parallel saturation bus, two or three reverb sends, master-bus compressor, brick-wall limiter for streaming target. The chain has tripled in length over thirty years. Each added stage trades fidelity-to-source for control. The aggregate cost is the audible body of the performer. This is the eighth piece in the **Print the Quiet** series — about what was added, why, and what the listener lost in exchange. ### The 1994 chain Start with what was on the floor at Bearsville. **Microphone**: a large-diaphragm condenser, almost certainly a Neumann U87. Some sources cite a Neumann U67 — the public record is not definitive on which it was. Either way, the family is the same: a tube or FET cardioid condenser with a smooth, slightly forward midrange, captured cleanly without exaggerated presence boost. **Preamp**: the SSL 4000 E console preamps at Bearsville, possibly supplemented by an outboard Neve 1073 for some sessions. SSL preamps are clean and slightly forward; Neve 1073s add a touch of saturation and low-mid warmth. Either way, the preamp is contributing tone but not radical color. **Tracking compressor**: most plausibly an LA-2A or Fairchild on the way to tape, hitting one to three decibels of gain reduction at peaks. Andy Wallace has not, in any retrievable interview, pinned the exact tracking compressor for Buckley's vocal — treat this as era-typical inference. The compression on the way in is light. Wallace's stated preference was to capture the voice with most of its dynamic range intact and let later mix-side processing shape it. **Tape**: Studer multitrack at 30 IPS. The tape stage does its own automatic processing — soft saturation, gentle HF rolloff, head-bump on the low end, harmonic generation. None of this is chosen by an engineer. It is the medium. **Mix-bus compression**: SSL G-bus stereo compressor at about 4:1 with a slow auto release, driven to perhaps -4 dB of gain reduction on the peaks. Light. Glue without crushing. **Light mastering**: enough EQ to balance for vinyl and CD release, a small amount of additional limiting for the CD format, six to nine decibels of dynamic range left intact. That is five stages, end to end. The result is a vocal that breathes, with audible micro-pitch, breath, and consonant character all treated as content. The body of the singer arrives at the listener's body across a chain short enough that nothing was added that did not need to be added. ### The arrival of Auto-Tune (1997) The single most consequential addition to the vocal chain in three decades arrived in 1997. **Antares Auto-Tune** was released that year, designed by Dr. Harold "Andy" Hildebrand, a Rice-trained geophysicist who had built seismic-data autocorrelation tools for Exxon. The widely-told origin story: at a 1995 NAMM lunch, a colleague's wife told Hildebrand he should build a box that would let her sing in tune. Hildebrand initially dismissed it and returned to the idea months later. His own line, quoted across early profiles: "I never figured anyone in their right mind would want to do that" — referring not to the gentle corrective use of the tool, but to the extreme, vibrato-stripping snap-to-grid setting that would become iconic. That extreme setting got its public debut on **Cher's "Believe" (November 1998)**. Producers Mark Taylor and Brian Rawling, working at Metro Studios in London, set Auto-Tune's **Retune Speed parameter to 0** — instantaneous correction, no transition, no preservation of pitch movement. The result is the stepped, robotic vocal glide that defined a decade of pop. The producers attributed it publicly to a vocoder for years to protect the trick. Antares later codified the setting as the "Cher Effect" in the Auto-Tune 5 manual. **T-Pain ("I'm Sprung," 2005)** turned the effect into a genre signifier. For the next decade, hip-hop and pop production routinely used Auto-Tune as an audible aesthetic choice rather than a correction tool. T-Pain's response to the criticism that he could only sing in tune through the effect came in 2014, when NPR's Tiny Desk concert featured him singing his hits with only a pianist, no Auto-Tune, no correction. He could sing. He was keeping the effect anyway, on his own terms: "I'm keeping it, man. That's what made me. As long as people know what I can do without it." **Celemony Melodyne** arrived in 2001, designed by Peter Neubäcker. The technical distinction matters: Auto-Tune is real-time and frame-by-frame, snapping the input to a target scale. Melodyne is offline and blob-based — the engineer can grab individual notes after recording and edit their pitch, length, formant, and vibrato amplitude independently. Since the **Direct Note Access** update in 2009, Melodyne handles polyphonic material too. The reason Melodyne is "harder to hear" than Auto-Tune is that a skilled engineer can preserve micro-pitch detail and vibrato shape rather than snapping the performance to grid. Modern pop vocal production, by 2010, used some combination of both. The chain had gained one or two stages, and the gained stages were aimed at removing the human pitch information that Buckley's 1994 chain had preserved. ### The comping era (2001-2010) The other transformative software was ProTools. **ProTools 5 (2001)** and **6 (2003)** and **7 (2005)** made playlist-comping trivial. The workflow shifted from "record the take and live with it" to "record twenty takes and assemble the released version syllable by syllable." Each consonant, each vowel, each phrase could be drawn from a different pass. Industry chatter and pedagogical sources confirm **Max Martin's** well-documented practice: many takes, then comp the best syllables, often with placeholder dummy lyrics ("Burger King English") to optimize vowel shape before words are set. Dr. Luke, the modern Nashville assembly, and most major-label pop producers from 2005 onward adopted some version of this workflow. Producers documented as resisting the practice: **Rick Rubin** (live tracking, single takes, the famously short attention span for retakes — "I like the idea of getting the point across with the least amount of information possible"); **T-Bone Burnett** (puts up the vocal first, prefers live-tracked performances); **Daniel Lanois** (comps from full takes, not syllables); **Andy Wallace** in his pre-mainstream-comping era. But these are the exceptions, and they have been outnumbered. The structural cost of phoneme-level comping was explored in piece #6 of this series. The summary: phrase decisions in a real take are causally linked — the way you sing line four is a response to how line three just felt — and phoneme comping breaks the causal chain. Listeners' nervous systems read the discontinuity even when they cannot articulate it. The technical perfection that comping enables comes at the price of structural coherence. ### The 2010s: vocal as instrument A different and largely contemporary development was the use of the vocal itself as a designed sonic instrument, with effects chains that are part of the song's architecture rather than mix-stage additions. **Bon Iver's 22, A Million (2016)** used **The Messina**, a custom rig built by engineer Chris Messina at Justin Vernon's request. It is not, despite many secondary sources saying so, a Prismizer — Francis Starlite's Prismizer (used famously on Chance the Rapper's "Friends") was an inspiration but not the technology. The Messina is a laptop running custom software, plus hardware, that produces a real-time polyphonic harmonic effect with vocoder-adjacent behavior. Vernon performs through it live and tracks through it in studio. The effect is part of the song, not part of the mix. **Frank Ocean's Channel Orange (2012)**, engineered partly by Jeff Ellis with Malay producing, used Neumann U47 and Telefunken 251 vocal mics, sometimes a Shure SM58 deliberately chosen for a more intimate control-room feel. The chain on most tracks was Neve 1073 preamp into a Tube-Tech CL1B compressor or a Fairchild 670. Notably, Ocean re-tracked several vocals through a Studer A827 for analog tape saturation — a deliberate retro decision inside an otherwise modern chain. The vocal was not just being recorded; it was being processed as a designed sound. **James Blake's** documented approach: triplicate vocal layers — one dry mono center, two stereo duplicates with divergent effects (filter, distort, pitch-shift) — printed into the song architecture itself. The "vocal" is not a single signal anymore. It is a designed multi-layer construction. ### The contemporary chain in detail For a sense of what the modern vocal chain actually looks like, take Adele as a documented example. **Mics and tracking** (varies by song and session): Sony C800G or Neumann U67. Adele has used both, with engineer preferences varying by producer. The Sony C800G (released 1992, $10,000+) has a documented presence boost around 8 to 14 kHz — the reason hip-hop and R&B producers (Dr. Dre, Kanye West on "Through the Wire," Drake, Travis Scott) made it the standard mic of the era. It captures forward sibilance and air with no EQ required. Aggressive de-essing typically follows it. **Mix-side chain** (Tom Elmhirst, who mixed both 21 and 25, has been explicit about this in Sound on Sound interviews): at Electric Lady Studios, Neve 1081 preamp → Bluestripe 1176 → Tube-Tech CL1B. For mixing Adele in Los Angeles, the chain shifted to Neve 1066 → Bluestripe 1176 → Fairchild 660. The 1176 catches fast peaks; the Fairchild handles slow program-level leveling. On 21's "Rolling in the Deep," Elmhirst's full vocal-chain plug-in list included Waves Q6 and De-esser, Digirack EQ III, Lo-Fi, Trim, Pultec EQ, UREI 1176, Tube-Tech CL1B, plus spring reverbs and delays. That is eight processing stages on the vocal alone, not counting tracking and not counting the master bus chain. On "Hello," he layered AMS delay, an Eventide "Canyon" preset, plate reverb, and spring reverb. The CLA / Glyn Johns stack — 1176 into LA-2A in series — is now ubiquitous on lead pop vocals. The 1176 catches transients with its fast attack and fast release. The LA-2A handles slower program-level leveling with its opto-photoresistor circuit. Together they produce a vocal that sits forward in the mix with consistent loudness across phrases. **Chris Lord-Alge's published settings**: 1176 input 30 / output 18 / ratio 4:1 / attack 3 / release 7. That setting represents one of the most-copied vocal-compression configurations in modern rock and pop. ### The 2020s: the AI layer A further transformation, still in progress. **iZotope Nectar 4** (current at the time of writing) bundles vocal-chain decisions — EQ, compression, de-essing, pitch correction — behind an AI Vocal Assistant that proposes processing automatically. The engineer can override, but the default workflow is "let the algorithm decide." **Soundtheory Gullfoss** is an adaptive perceptual EQ. It analyzes source material in real time against a psychoacoustic model and re-EQs the signal continuously, second by second. The decision that an engineer would have made — what to cut at 400 Hz, what to boost at 8 kHz — is being made by the software, in real time, throughout the song. **Waves Vocal Rider** automates fader rides. The engineer's hand on the vocal fader has been replaced by an algorithm tracking the input level and maintaining a target output. **Synthetic-voice tools** have moved from novelty to production. Suno's v5.5 (released March 2026) includes a Voices feature: a user uploads 30 to 60 seconds of clean vocal samples and the model synthesizes new performances in that voice. Warner Music settled its lawsuit with Suno and Udio in November 2025; UMG and Sony lawsuits remain active. The conceptual question — and it is an open question — is what the noun performance even means when a vocal can be reconstructed from a 60-second sample. A possible answer: a performance is a body in a room making a decision in time. The Suno output is not a performance in that sense. It is a synthesis trained on performances. Whether that distinction matters to listeners, or whether it stops mattering once they get used to the new baseline, is the central uncertainty in popular music in the late 2020s. ### The micro-decisions that change emotional character Beneath the gross structure of the chain are a half-dozen small decisions that, individually, change a vocal's emotional read significantly. **Mic distance.** The proximity-effect literature confirms that cardioid microphones below about six inches produce a marked low-frequency rise from roughly 200 to 300 Hz down. Buckley sang at four to six inches into the U87. Modern pop vocals are routinely tracked at eight to twelve inches into the Sony C800G's sweet spot, which produces a flatter, less chesty vocal that needs no de-mudding and reads as less intimate. The distance is the decision. **Compression depth.** Two decibels of gain reduction preserves micro-dynamics. Eight decibels squashes them into a constant body. Modern stacks running an 1176 into an LA-2A in series routinely pull six to ten decibels of combined gain reduction on the lead vocal. The body of the singer arrives flatter than it left. **Reverb predelay.** Twenty milliseconds reads as room. Sixty to eighty milliseconds throws the reverb behind the consonant and reads as cathedral. Pop typically uses short predelay (twenty or under) to keep the vocal forward and immediate. Buckley's Hallelujah uses minimal reverb on the verses and slightly more on the choruses, with predelay set to feel like the room rather than a separate processed layer. **Auto-Tune retune speed.** Zero is the Cher Effect — instantaneous correction, no vibrato. Twenty to forty is the modern transparent setting — corrective but allowing some pitch movement. Higher values preserve human pitch movement at the cost of looser correction. The parameter is, arguably, the single most consequential setting in modern vocal production. It determines, more than any other knob, how human the vocal will sound on playback. **De-essing aggression.** Heavy de-essing flattens "s," "t," and "ch" — the consonant cluster that carries linguistic emotion and clarity. The Sony C800G's high-shelf presence makes aggressive de-essing nearly mandatory, which is the structural irony of the modern chain: the mic is chosen for its forward sibilance, then a de-esser is deployed to undo the sibilance the mic was chosen to capture. Add a stage, undo a stage, net zero gain in audible reality and net cost in the chain. ### The mixer philosophies, briefly A short tour of how named mixers approach the modern vocal. **Tom Elmhirst** (Adele, Amy Winehouse): Neve preamps fronts, 1176 into Tube-Tech or Fairchild stacks, heavy use of Soundtoys plugins — Decapitator on vocals for "presence and attitude," EchoBoy on every track, Crystallizer and FilterFreak for texture. **Manny Marroquin** (Bruno Mars, Rihanna's "Umbrella"): mixes Bruno Mars's vocal deliberately deeper in the mix to signal that Bruno is a member of his band, not an isolated frontman. Uses 1176 plus de-essers plus de-harshers plus parallel compression plus multiband. Cites Michael Jackson's records as the model for a vocal that is forward without being isolated. **Serban Ghenea** (Taylor Swift's Folklore and a documented hundred-plus number-one singles): minimalist by reputation, primarily in-the-box, Avid Q3 to filter lows, Metric Halo Channel Strip, Waves R-Comp, H-Delay, convolution reverbs. His stated philosophy: "the song first." **Mark "Spike" Stent** (Madonna, Beyoncé): SSL G-Channel front, 1176 into LA-2A layering, Waves H-Delay for time-based effects rather than reverb wash, outboard Lexicon plates for the few places where he wants real plate character. **Andrew Scheps** (Adele's 21, Red Hot Chili Peppers, Metallica's Death Magnetic): in-the-box, parallel-aggressive secondary chain blended under main, has said publicly that on one 21 mix he used only five EQs total plus spring reverb and a bus compressor. The restraint is the credit. ### The honest accounting The modern vocal chain is not worse than the 1994 chain. It is different. It optimizes for different things. The 1994 chain optimized for a recognizable voice carrying breath, micro-pitch, room, and consonant character into the listener's room. The 2026 chain optimizes for grid-perfect pitch, broadcast-loud RMS, playlist-context volume, translation across earbuds, and consistency with the production aesthetic of the algorithm-curated playlist. Those are different targets. The technical execution of the modern chain is sometimes spectacular — Tom Elmhirst's 21 mixes are master classes in modern engineering. The cost is structural. With every stage added to the chain, a small piece of the original performance is replaced with a piece of processing. By the fifteenth stage, the listener is hearing the chain, not the singer. The Buckley 1994 chain, by comparison, has four to five stages, all of them light. The listener hears the singer. There is also a survivorship-bias point worth naming. We compare the best 1994 records to the average 2026 record because we have already filtered the 1994 catalog through thirty years of remembering what was good. The average 1994 record was probably not better than the average 2026 record. Grace is, by any measure, not the average 1994 record. The honest version of the argument is: the best records of 1994 made decisions that the best records of 2026 mostly do not make. That is the indictment, and it is real. ### What this means for a working singer or producer A short set of practical positions. **For singers**: ask your engineer how many stages are on your vocal. If they cannot tell you, get a different engineer. If they can, ask which stages are necessary and which are habitual. There is usually one habitual stage you can remove. **For engineers**: count the stages. The first cut you make should be of a redundant compressor. The next cut should be of an EQ that is fixing what the mic captured because the mic was chosen for the wrong reason. **For producers**: track with the shortest possible chain. Add processing later if needed. The bias should be subtractive, not additive. The best vocals on record from any era are vocals that were captured cleanly and left mostly alone. **For everyone**: listen to a 1994 record and a 2026 record back to back at the same volume on the same speakers. The 1994 record will have moments of intimacy that the 2026 record does not. Those moments are not magic. They are what a four-stage chain produces and a fifteen-stage chain cannot. Print the quiet. Track the body. Cut the redundant stage. Print the Quiet is a Suede AI Social series on tone, dynamics, and the parts of music that don't fit on a lead sheet. Next: the closer. ## What's in the Tape That Isn't in the Score Source: https://strumly.suedeai.ai/book/print-the-quiet/ptq-09-whats-in-the-tape The closer to Print the Quiet: tone is the body audible inside a record. From Buckley's breath at 3:20 to AI synthesis: why the decision stays traceable. Print the Quiet essay by Jason Colapietro. Published 2026-06-13. ● Print the Quiet · 9 / 9 The closer: tone is the irreducible record of a body in a room making a decision in time, and that decision has to belong to someone. By [Jason Colapietro](https://strumly.suedeai.ai/book), writing as Johnny Suede · A Suede AI Social long read · 11 min This series began with Jeff Buckley standing at a microphone at Bearsville Studios in September 1993, singing a Leonard Cohen song into the air of a converted barn. It has worked through the loudness war, through the religion of the compressor, through the lost art of clean tones, through the room as instrument, through the comp-versus-take question, through analog tape and its physical constraints, through the steady lengthening of the recorded vocal chain over three decades. The argument running underneath all eight pieces has been the same. Tone is the part of recorded music that is not in the score. The chord chart of Buckley's Hallelujah is Leonard Cohen's. The melody is Cohen's. The lyric is mostly Cohen's, filtered through John Cale's editorial choices. What is Buckley's, and what is the reason listeners still seek out his version in numbers that beat the songwriter's original by two to one on Spotify, is everything that is not on the lead sheet. The arpeggio Buckley invented to translate Cale's piano figure to guitar. The capo position. The 6/8 pulse. The U87 four to six inches from his mouth in a 35-by-35-foot wooden cathedral. The right hand on the strings. The half-second hesitation before "the holy or the broken Hallelujah." The slight bend on "cold" in the second verse. The thumbnail catching the wound G string at the second-verse pickup. The breath at 3:20. None of that is on the lead sheet. All of it is the record. That is the thesis. This piece, the ninth, is the closer. ### The argument in one paragraph Tone is the irreducible record of a body in a room making a decision in time. It is the part of music that is not in the score because the score, by definition, abstracts away the body, the room, the decision, and the time. The lead sheet says G major. The recording says G major as played by this human, on this instrument, at this volume, in this room, at this temperature, on this day, by these hands, in this emotional state, with this breath, at this distance from this microphone, through this preamp, onto this medium. Every one of those qualifiers is tonal information. None of them is in the score. The recording is the sum of the score plus all of the qualifiers. The qualifiers are what we call tone. The qualifiers are also, currently, the part of music that no AI system has been able to infer from the notes — because the information was never in the notes. It was in the breath. ### What the series has actually established A short recap of the technical and historical points that, taken together, support the thesis. Anyone who has read the eight prior pieces can skip this section. Anyone arriving fresh: this is what the receipts say. **On dynamic range.** Grace was mastered in 1994 at roughly DR11 to DR13 — solidly in the healthy band. Average pop today sits around DR8 to DR10, an improvement from the 2008 trough of DR4 to DR6 but still below the analog-era baseline. Records mastered with dynamic range intact allow the listener's nervous system to read dynamic change as emotional information. Records crushed to a flat wall remove that channel of communication. The loudness war was a forty-year wound to popular music's emotional bandwidth. The streaming-era normalization flipped the incentive, but the practice is still changing slowly. **On compression.** Compression, used well, is the throat of the recording — the tool that most directly shapes how the body of a sound arrives at the listener's body. Used badly, it removes that body. The difference between breath and crush is a difference of judgment, not of equipment. Andy Wallace pulled two to four decibels of gain reduction on the Grace master bus. Modern pop pulls six to ten on the vocal alone before it touches the master. **On clean tones.** Clean is the courtroom. It exposes everything the player does, which is why mediocre players reach immediately for distortion. The clean Telecaster on Hallelujah is the precondition for the take. If the tone had been dirty, the same chords would not have done what they did. **On the room.** Bearsville Studio A is on the record as much as the Telecaster or the U87. The 35-foot ceiling, the wooden envelope, the early reflections, the late tail — these are physical conditions that no plugin reverb completely captures because no plugin reverb is a coupled physical system. Bedroom production has, by necessity, replaced room behavior with software. The software is genuinely close. It is not the same. **On takes.** Buckley cut more than twenty full performances of Hallelujah at Bearsville. Wallace assembled the released master from three to five of them. The comp was at the line and verse level, not the syllable level. The causal chain of phrasing inside each take is intact. That is structurally different from modern phoneme comping, which breaks the causal chain even when the result is technically perfect. Listeners cannot articulate the difference. Their nervous systems read it. **On tape.** Magnetic tape is itself a multi-stage automatic processing chain — soft saturation, low-order harmonic generation, gentle HF rolloff at hot levels, low-end head-bump, micro pitch modulation — applied to every track of every record from 1955 to roughly 1995, for free, without an engineer choosing it. The transition to digital in the late 1990s removed that chain. What we call "the warm sound of analog records" is the audible residue of that chain. Modern tape emulation plugins capture 90 to 95 percent of it. **On the vocal chain.** Buckley's 1994 chain had four to five stages between mouth and master. A modern major-label vocal chain has fifteen or more. Each added stage trades fidelity-to-source for control. The aggregate cost is the audible body of the performer. These are the receipts. Taken together, they describe a forty-year set of decisions in the recording industry that have, on net, removed body from records and replaced it with surface. The body is the tone. The surface is the chain. ### What AI is good at, and what it currently isn't A serious version of the AI conversation, without the marketing voice or the doomerism. What current AI music systems are genuinely good at: chord progressions, melodic plausibility, surface timbre approximation, lyric generation, structural arrangement (intro, verse, chorus, bridge). The 2025 work from Suno, Udio, Riffusion, and similar systems is, by any reasonable measure, excellent at the score. If the score is what you want — a competent chord progression, a passable melody, a serviceable lyric, a finished-sounding production — you can have one in three seconds. What current AI music systems are noticeably less good at: the qualifiers. The body in the room. The decision in time. The breath at 3:20. AI vocals generated as of 2026 tend to be locally plausible at the phrase level and globally disconnected — they have not learned the causal chain across phrases, the way line four of a verse is a response to how line three just felt. AI guitar tones, especially clean ones, tend to sound like averages of training data rather than like a specific player in a specific room with a specific instrument. AI mastering defaults to brick-walled output because the training data taught it that brick-walled output is what finished records sound like. The shorthand: AI is excellent at what is in the score and approximate at what is not in the score. The thing it is approximate at is what we have been calling, for nine pieces, tone. Will it stay that way? Probably not. The models are improving fast, and the gap is narrowing. By 2030, AI systems will almost certainly produce vocals with convincing causal chains across phrases. They will almost certainly produce guitar tones that are indistinguishable from a specific player in a specific room. They will almost certainly handle dynamic mastering with the judgment of a skilled engineer. The technical limitation is real and it is also temporary. But the interesting question is not the technical one. The interesting question is what tone means when it can be inferred from the notes — when the qualifiers can be generated, in a plausible way, by the same system that wrote the score. ### What tone means when the body is optional Here is the position this series will end on, for the record. Tone is the irreducible record of a body in a room making a decision in time. When the body, the room, and the decision can be synthesized from training data, tone is still tone — but the question of which tone, and whose, becomes a different question. It becomes a question of attribution, provenance, and chosen constraint, rather than a question of physical capture. This is a real shift, not a marketing one. The 1994 Bearsville recording of Hallelujah could only have been made by Jeff Buckley, in that room, on that day. The 2026 AI-generated recording of Hallelujah in the Buckley style could be made by anyone, anywhere, with a prompt. The technical capture is no longer the bottleneck. The bottleneck is now the decision: whose body, whose room, whose breath, whose constraint. What this series has tried to argue — and what we, at Suede Labs, think matters about it — is that the decision is the thing that has to be preserved, attributed, and honored, because the body that made it is the source of the tone. If we let the decision be untraceable, the tone becomes generic. If we honor the decision — if we trace it back to a specific person, in a specific moment, choosing specific constraints — the tone remains tone, even when it is synthesized. In practice, this means three things: **One: the people who made the decisions deserve attribution.** Cohen wrote the song. Cale chose the lyric edit. Buckley invented the arpeggio. Wallace caught the take. Each of those decisions is a contribution to the record we hear, and each of those contributions should be tracked. This is what programmable IP and creator-ownership infrastructure is for, in the most honest version of why it matters. Not as a token mechanism. As a way to keep the line from a piece of music back to the bodies that made the decisions inside it. **Two: constraints should be chosen, not erased.** The 1994 chain had four to five stages because Wallace chose a constraint set. The 2026 chain has fifteen because most of those stages were added without anyone choosing them — they accreted as defaults, as habits, as algorithmic suggestions accepted without consideration. The discipline of choosing constraints, the way Jack White chose tape and the way Daptone chose all-analog, is itself a tonal act. Every default in your DAW that you accept without thinking about is a small piece of tonal authorship you have given away. Every default you change is a piece of tonal authorship you have kept. **Three: the records that survive will be the ones that had bodies in rooms making decisions.** This is empirical, not aspirational. The records that last thirty years and become the records people put on at weddings and funerals are records with audible decisions inside them. Grace. Pink Moon. The Nightfly. Court and Spark. Astral Weeks. Blue. What's Going On. Revolver. The frictionless, decision-free, AI-defaulted track is unlikely to be the one anyone returns to in 2056. The track with a body audible inside it — synthetic or organic — is the one that has a chance. ### A small confession Let me be plain about something the previous pieces have danced around. This series has a position. The position is not that AI music is bad. The position is that the part of music that listeners respond to most deeply — the tone, the body, the decision — is the part that is hardest to fake and easiest to lose. We have lost it, partly, over thirty years of mastering loud, of comping at the syllable, of replacing rooms with plugins, of adding stages to the vocal chain until the singer is barely audible behind the engineering. AI did not cause that loss. The loss preceded AI by a long time. What AI raises is whether we will lose it further by default — or whether we will use this moment, as the technology accelerates, to remember what tone is and to design our tools, our platforms, and our value systems to keep it. The whole Print the Quiet argument is that tone is a value system. It is the choices about what to leave in (the breath, the room, the wobble) and what to take out (the redundant compressor, the unnecessary correction, the safety overdubs that protected nothing). Those choices are not technical. They are aesthetic, and aesthetic choices are moral. When a producer chooses to leave the breath in, they are making a small moral claim about what listening is for: that listening is a meeting between two bodies, and that the meeting requires the bodies to be audible. When the breath comes out, the meeting becomes a meeting between a listener and a surface. The listener can still enjoy the surface. Many surfaces are beautiful. But the meeting is different. Something specific has been removed. ### Coda — the breath at 3:20 Go back to Hallelujah, one more time. Three minutes and twenty seconds into the song, Buckley's voice drops to almost nothing. The verse before the final chorus. He is, by this point in the take, deep enough into the performance that his body knows where it is going. The guitar arpeggio continues underneath, the Telecaster's clean Vibrolux signal pulsing in 6/8, the Quadraverb halo extending the sustain. And in the small gap between the verse's last word and the chorus's first one, he takes a breath. You can hear it. The U87 four to six inches from his mouth picks it up plainly. The breath has shape. It is not silent. It is not a noise the engineer should have gated out. It is the sound of his body preparing for the climb back. It is part of the record. If you played that breath to a person who had never heard the song, and asked them what it was, they would say: someone is breathing. They would not say: that is a tonal artifact of close-miking with a tube condenser at a specific proximity in a wood-paneled tracking room with light analog tape compression. They would say: someone is breathing. That is what tone is. The sound of a body, audible inside a record. The thing that makes a listener, who has never met Buckley, feel like they have met Buckley. The thing that lets a song written by Leonard Cohen and arranged by John Cale, sung by a twenty-six-year-old in a barn in upstate New York in September 1993, become, thirty-three years later, the version that two generations of listeners return to when they want to feel that the music is real. You can synthesize the breath. The synthesis will get closer every year. The breath, generated, will eventually be indistinguishable from the breath, captured. When it is, the question we will face is not whether the synthesis sounds like Buckley. The question will be: whose decision was the breath? Whose body, in what room, made the choice to leave it in? If the answer is no one's — if the breath is a default that the algorithm added because the training data taught it to — then the breath is no longer a meeting between two bodies. It is a meeting between a listener and a probability distribution. If the answer is someone's — if a specific person, at a specific moment, made the decision to keep that breath in — then the breath is still a meeting. The body is still in the record. That, more than any technical detail in any of the previous eight pieces, is what Print the Quiet has been trying to point at. Tone is the part of music that is not in the score. The part that is not in the score is the body. The body is in the room making a decision in time. The decision has to belong to someone, and the someone has to be traceable, or the meeting is no longer a meeting. Print the quiet. Keep the breath. Trace the decision. That is the work. That is the receipt. Print the Quiet, a Suede AI Social series on tone, dynamics, and the parts of music that don't fit on a lead sheet, ends here. Thank you for reading. — JC