A great recording is mostly decided before a plugin is ever opened — in the choice of microphone, where you point it, and how you route the signal. This guide draws those invisible things: what a polar pattern actually looks like, why moving a mic two inches changes everything, how a compressor bends the signal, and how to reamp, punch in, and comp a take. Workflow notes are for Logic Pro X, but the physics is universal.

Microphone Types

Every mic converts air pressure into voltage, but how it does that shapes its sound and what it's good for. Three families cover almost everything you'll record.

Dynamic

A diaphragm moves a coil through a magnet — rugged, needs no power, handles brutal volume. Forgiving and fat. Slower transient response tames harshness.

Reach for: loud guitar cabs (SM57), snare, toms, live vocals, kick (large-diaphragm dynamics). Bulletproof and cheap.

Condenser

A charged capsule whose capacitance changes with sound — fast, detailed, sensitive. Needs +48V phantom. Large-diaphragm = warm & flattering; small-diaphragm ("pencil") = accurate & precise.

Reach for: studio vocals, acoustic guitar, overheads, room, anything where detail matters.

Ribbon

A thin metal ribbon suspended in a magnetic field — smooth, dark, natural. Naturally figure-8. Vintage ribbons are fragile; never send phantom power unless it's a modern active model.

Reach for: taming bright amps & brass, room mics, anything digital-harsh you want to soften.

TypeSoundPhantom?SPL / durabilityClassic examples
DynamicPunchy, fat, forgivingNoVery highSM57, SM7B, RE20, MD421
Large-diaphragm condenserWarm, present, flatteringYes (+48V)ModerateU87, TLM103, NT1, C414
Small-diaphragm condenserAccurate, fast, airyYes (+48V)Moderate–highKM184, C451, SM81, Pencil pairs
RibbonDark, smooth, vintageUsually no*Low (fragile)R-121, M160, RCA 44

*Modern active ribbons require phantom; classic passive ribbons can be destroyed by it, especially through a mis-wired cable. When in doubt, phantom off.

Set your interface as the Input Device in Settings ▸ Audio. Arm a track, click the Input Monitoring (headphone) button to hear the mic through Logic, and enable +48V on your interface for condensers. If you hear a slappy delay, turn on Low Latency Mode or lower the I/O Buffer Size while tracking.

Polar Patterns

A polar pattern is a map of where a mic is sensitive. The shape shows how loudly the mic picks up sound arriving from each direction — front is up (0°), rear is down. Where the shape pinches inward, the mic rejects sound. Choosing a pattern is choosing what to keep out as much as what to let in.

0° FRONT
Omnidirectional
Hears everything equally. Most natural, no proximity effect. Great for rooms & when the source is the whole space.
0° FRONT REAR REJECT
Cardioid
The workhorse. Full front, deep rear null — isolates one source, kills the room behind it. Has proximity effect.
0° FRONT small rear lobe
Super / Hyper-cardioid
Tighter than cardioid, best off-axis rejection — but a small rear lobe. Max isolation on a loud stage or a bleedy kit.
0° FRONT side null
Figure-8 (Bidirectional)
Front & back equally, dead on the sides. Native to ribbons; the key to Mid-Side stereo & clever bleed rejection.

Multi-pattern condensers (like a C414 or U87) switch between these with a capsule selector — one mic, many jobs. The side nulls of figure-8 are the sharpest rejection in audio: aim a null at the thing you don't want and it nearly vanishes.

Placement & the Proximity Effect

Mic position is the most powerful tone control you own — and it's free. Two ideas govern it: distance (close = detail + bass + isolation; far = air + room + balance) and angle (on-axis = bright & direct; off-axis = softer, darker).

SOURCE on-axis · close bright · bassy · dry off-axis · far airy · softer · roomy closer → more bass farther → more room
The same mic, two placements — before any EQ.

Proximity effect is the bass boost that directional mics (cardioid & friends) get as you move in close — inches from a cardioid can add several dB of low end. It's why radio voices sound huge, and why a thin acoustic guitar gets boomy up close. Use it as a tool: back off to thin out, lean in to fatten. Omni mics have no proximity effect, so they stay even at any distance.

The 3:1 Rule

Using two mics on one source (or nearby sources)? Keep them at least 3× as far apart as each is from its source. It keeps their bleed out of phase-cancellation range, so the blend stays full instead of hollow.

Check Phase

Two mics on one source (top+bottom snare, DI+amp) can fight and cancel bass. Flip polarity on one (the Ø button) and keep whichever sounds fuller. Mic a source from front and back? The back mic almost always needs the flip.

High-Pass Early

Roll off rumble below the source's real range at the mic pre or first plugin — ~80 Hz on vocals, ~100 Hz on guitars. Removes handling thumps, footfalls, and AC hum before they eat headroom.

SourceMic & patternWhere to put it
Lead vocalLDC, cardioid6–10" away, slightly above mouth angled down; pop filter a fist away. Back off on loud phrases.
Acoustic guitarSDC, cardioidAim at the 12th–14th fret, ~8" back — not the soundhole (boomy). Add a second at the bridge for body (mind 3:1).
Electric guitar ampDynamic (SM57)1–2" off the grille. On the dust-cap edge = bright; toward the cone center = brighter; toward the speaker edge = darker & warmer.
Drum overheadsSDC pairSpaced or XY above the kit; measure equal distance to the snare from each so the center stays solid.
Bass guitarDI + optional amp micDI for clean low end; a dynamic on the cab for grit. Blend, checking phase.
Use the stock Gain plugin's Phase Invert (Ø) on multi-mic tracks, or nudge one region by samples to time-align. Logic's Channel EQ has a one-click Low Cut (the leftmost band) — engage it on every mic track. Save a go-to mic chain as a Channel Strip setting so it's one click next time.

Stereo Miking

Two mics, arranged deliberately, capture a space: the width of a piano, a kit, a room, a guitar. Each technique trades width against mono-compatibility — whether it still sounds right collapsed to a single speaker (phones, laptops, clubs). Here's the geometry, from above.

90°, capsules together SOURCE is up ↑
XY (coincident)
Two cardioids, capsules stacked at ~90°. Moderate width, rock-solid mono — no phase issues. The safe default.
17 cm 110° spread
ORTF (near-coincident)
17 cm apart, 110° — mimics human ear spacing. Wide & realistic, still fairly mono-safe. Great on acoustic & room.
0.3–3 m apart both face source
Spaced Pair (A/B)
Two mics far apart. Huge, lush width — but weakest mono (arrival-time differences). Watch the 3:1 spacing to the source.
MID →front SIDE (fig-8) ↔ decode M±S
Mid-Side (M/S)
Cardioid (mid) + figure-8 (side), decoded into L/R. Width adjustable after the fact and perfect mono — the side simply drops away.
Record Mid and Side to two mono tracks. To decode: pan Mid center. Duplicate the Side track — pan one copy hard left, the other hard right, and invert phase (Ø) on the right copy. Group the two Side tracks so one fader raises/lowers both: that fader is now your stereo width knob. (Or drop the stock Direction Mixer plugin on the pair and flip it to M/S mode — one plugin does the whole decode.)

DI & Reamping

A DI (direct input) captures an instrument's clean, unamplified signal straight to the interface. Reamping is the trick that pays it off later: send that clean DI back out through a real amp (or pedals), re-mic it, and record the result — so you can perfect the tone after tracking, play intact, without the performer in the room again.

① TRACK THE CLEAN DI Guitar DI box Interface DAW: DI track ② SEND IT BACK OUT THROUGH AN AMP DAW: DI out Reamp box Amp Mic DAW: new amp track reamp box = line level → instrument level

The little box matters: an interface output is line level, but an amp wants a high-impedance instrument level signal. A reamp box does that conversion (and lifts ground hum). Feed an amp a raw line output and it'll sound thin and wrong.

Put the clean DI on a track and set its Output to a spare physical interface output (e.g. Output 3–4) instead of Stereo Out — patch that into the reamp box. Arm a new audio track listening to the mic on the amp, and record. Prefer to stay in-the-box? Skip the hardware entirely: drop Amp Designer or Pedalboard on the DI track and audition amps instantly — same "commit later" freedom, no cables.

How Compression Actually Works

A compressor is an automatic, lightning-fast volume fader: when the signal gets louder than a line you set, it turns it down. That's it. The result is a smaller gap between loud and quiet — so the whole thing can sit louder and steadier in a mix. The two pictures below are the entire concept.

1:1 (no comp) threshold output → input → 4:1 above
The Transfer Curve
Below threshold, output = input. Above it, every 4 dB louder in becomes only 1 dB louder out — that's the ratio.
input level gain reduction (how hard it clamps) ◂attack release▸
Attack & Release, in Time
Attack = how fast it clamps when the loud part hits. Release = how fast it lets go after. These two shape the feel.
ControlWhat it doesTurn it…
ThresholdThe level where compression startsDown to catch more of the signal (more compression).
RatioHow hard it clamps above threshold2:1 gentle · 4:1 firm · 8:1+ heavy · ∞:1 = limiting.
AttackHow fast it reacts to a loud hitSlow to let transients punch through · fast to tame peaks & clicks.
ReleaseHow fast it recovers afterFast = lively/pumping · slow = smooth/glued. Try to breathe with the tempo.
KneeHow abruptly it kicks in at thresholdSoft = gentle & transparent · hard = aggressive & obvious.
Make-Up / GainAdds back the level you compressed awayUp until compressed & bypassed feel equally loud (fair A/B).
Mix / Dry-WetBlends compressed with untouched signalThe heart of parallel compression — see below.

Set It in 4 Moves

1. Ratio to 3–4:1. 2. Lower threshold until you see 3–6 dB of gain reduction on the loudest parts. 3. Set attack/release by feel — punch vs. smooth. 4. Make-up gain, then bypass-compare at matched loudness.

Parallel Compression

Crush a copy hard and blend it under the untouched signal (the Mix knob, or a parallel bus). You get density and sustain without losing the transients & life of the original. Magic on drums, bass, vocals.

Gain-Match, Always

Louder sounds "better" for ~10 seconds, then fools you. Every compression judgment must be made at equal loudness — or you'll keep choosing "more" for the wrong reason.

COMPRESSOR STUDIO VCA −4.0 dB GR GAIN REDUCTION THRESH−18 dBRATIO4:1ATTACK12 msRELEASE180 msKNEE0.5GAIN+6 dB
Logic's Compressor (Studio VCA circuit), recreated — dial ratio ≈4:1, lower threshold until the meter shows 3–6 dB of gain reduction, then make up the level.
Logic's stock Compressor is seven compressors in one — the Circuit Type menu models classic hardware: Studio VCA (clean, versatile — a great default), Studio FET (aggressive, 1176-style, fast & punchy on drums/vocals), Classic/Vintage VCA (SSL bus glue), Vintage FET, and Vintage Opto (LA-2A-style, slow & smooth — beautiful on vocals & bass). Same knobs, different character. Use the built-in Mix knob for instant parallel compression, watch the gain-reduction meter for your 3–6 dB, and turn on Auto Gain for a fair bypass comparison.

Punch-In & Overdub

Overdubbing is adding a new part alongside what's already recorded (a harmony, a solo, a tambourine). Punching in is surgical: drop into record for just the flubbed word or bar, replace it, and drop back out — the rest of the take is untouched. The player performs into it live, so it stays musical.

PUNCH IN PUNCH OUT playback RECORD (replaces) playback only this slice is re-recorded

Pre-roll = Musical Punches

Always start playback a bar or two before the punch so the player can catch the groove and land in time. Punching cold never feels right.

Comp vs. Punch

Punch replaces audio destructively-ish in the moment. If you're unsure, record a new take over the section instead and comp the best of both — safer, more options.

Loop While You're There

For a stubborn phrase, cycle the section and let the player throw down several passes back-to-back into a take folder, then keep the best.

Two ways. Autopunch: enable the Autopunch button in the control bar (or Cmd ⌘ + Ctrl + P to show it), drag the red autopunch region over exactly the bars to replace, set a pre-roll, hit record — Logic only records inside that region. Quick Punch / on-the-fly: just tap Record (R) during playback to punch in and R again to punch out. Set Recording ▸ Overlapping Recordings to Merge or Create Take Folder so nothing you already have is lost, and turn on pre-roll (Record settings) so you always ease in.

Comping the Perfect Take

Nobody nails it in one pass — so record several, then comp: assemble one flawless performance from the best moments of each. Great verse in take 2, killer chorus in take 4, one perfect word in take 1 — stitch them into a "composite" that never actually happened live.

TAKE 1 TAKE 2 TAKE 3 COMP one perfect performance, stitched from all three
When overlapping recordings make a Take Folder, click it open to see every pass stacked. With Quick Swipe Comping on (the default), just drag across the part of each take you want — Logic highlights your choices and seamlessly crossfades the edges. Build the comp by swiping the keeper sections top to bottom, audition, then Flatten the folder to bake it into one clean region. Right-click the folder for Export/Flatten options, and keep the folder unflattened until you're sure — it's non-destructive up to that point.

Capture it right at the mic and the mix is half-done before you start. Everything after — EQ, effects, the blend — is easier when the raw recording already sounds like the thing you heard in the room.