Prepare your mix
for mastering.
A great master starts with a well-prepared mix. Follow this guide and your files will arrive ready to work on — no delays, no surprises.
Pre-mastering checklist
Run through this before you export. The details behind each point are below.
1. Headroom, peak levels & digital dynamics
The 0 dBFS ceiling
In digital audio, 0 dBFS (decibels relative to full scale) is the absolute upper limit for signal level. Going past it causes clipping: the peaks of the waveform are flattened, adding harsh, odd-order harmonic distortion that changes the tone, balance and character of your mix.
Bit depth and dynamic range
A 24-bit recording has a theoretical dynamic range of 144 dB (24 × 6.02 dB), across 224 possible amplitude levels. Healthy signal levels use more of that resolution and keep the signal well above the noise floor before it reaches the mastering chain.
Headroom is not loudness
Peak headroom doesn't decide how loud your mix feels — that depends on its crest factor and density. Typical commercial masters sit between −16 and −9 LUFS; pre-master mixes usually measure between −25 and −18 LUFS.
For example, a dense pop mix with a 15 dB dynamic range (quiet passages at −18, peaks at −3 dB True Peak) will measure much louder than a classical piece with a 60 dB range (quiet passages at −60, peaks at −3), even though both peak at the same level.
The headroom you leave is space for the mastering engineer to do corrective EQ, multi-band dynamics, analog summing and limiting without pushing processors into digital saturation.
Target peak levels
| Standard target | −3 to −6 dB True Peak |
|---|---|
| Extended target | −6 to −9 dB True Peak |
| Meter | dB True Peak meter (ITU-R BS.1770 compliant) |
Why True Peak?
A standard sample-peak meter only reads the individual digital samples. When a converter rebuilds the continuous analog waveform, it often overshoots between samples. A True Peak meter measures those inter-sample peaks, so your 3–9 dB of headroom is real.
2. Master bus processing & gain staging
What's acceptable on the master bus
Subtle master bus processing can be part of a mix's character, and deliberate creative choices — like sidechain pumping — are completely fine. What isn't: a brickwall limiter slamming the mix to −5 LUFS. It destroys transients and dynamics, and mastering a squashed file is like icing an overcooked cake.
| Processing | Maximum / status |
|---|---|
| Limiting | Not allowed — disable before export |
| Compression | Up to 3 dB gain reduction |
| EQ | Up to 3 dB boost or cut |
| Harmonic distortion / saturation | Up to 1% THD |
| Stereo wideners | Not allowed — disable on the master bus |
Don't mix into a limiter
Mixing through a master limiter to hear "release loudness" means every fader, EQ and compressor decision is made against the limiter's gain reduction. When the limiter comes off for mastering, the balance falls apart: dynamics shift, the tonal balance changes and peaks blow past the headroom target.
Mix from start to finish with no limiter on the master bus. Get density, punch and balance from channel and bus processing alone.
Fixing clipping the right way
If removing master bus processing makes the output clip, don't pull down the master fader — leave it at 0 dB. Select all your track faders and lower them together until the output sits between −3 and −6 dB True Peak.
Sending references and alternatives
- Homebrew reference: send the clean, unlimited mix for mastering. If you like, add your own heavily limited version alongside it — purely as a listening reference.
- Two versions: not sure if your bus compression or EQ is helping? Send one version with it and one without.
3. File export standards
- Native resolution: export at the exact sample rate and bit depth of your session (e.g. 24-bit / 44.1 kHz or 24-bit / 48 kHz). No conversion means no conversion artifacts.
- Never upsample: converting a 44.1 kHz session to 96 kHz adds filtering, not fidelity.
- Uncompressed WAV only (linear PCM).
- No lossy formats: MP3, AAC and Ogg throw away audio data permanently and add artifacts.
Dither: off
When you export at your session's native bit depth (e.g. a 24-bit session to a 24-bit WAV), turn dither off. Dither only belongs at the final mastering stage, when the word length is reduced — for example from 24-bit to a 16-bit delivery file. Adding it earlier just raises the noise floor.
4. Buffers, trims & decay tails
Start: about one second of space
Don't start the file right on the downbeat. Leave roughly a second before the first sound. Room tone or preamp noise in that second lets us take a noise profile for clean-up if needed — just tell us whether it was left on purpose or should be muted.
End: the full tail, plus two seconds
Don't rely on "bounce to end" or the project length — they regularly cut off reverb and delay tails. Set the export range by hand: solo the ending, listen until every tail has fully died away, place the end marker there, then add two more seconds.
Clean up before you bounce
- Cut or gate non-musical sounds: stick clicks and count-ins, vocalist movement, instrument handling, floor thumps.
- Add short fades and crossfades on every edit to avoid pops and clicks.
5. Phase, stems & common mistakes
Check your mix in mono
Phase cancellation weakens the low end and makes parts vanish on mono playback. The usual causes:
- Stereo wideners and Haas-effect plugins on tracks or buses, which cause comb filtering.
- Fake double-tracking — duplicating one take, panning hard left and right and delaying one side. Record two real takes instead, or check mic phase alignment at the source.
Sum the whole mix to mono and listen on a single speaker. If guitars or synths drop away or the mix turns thin and hollow, find the problem channels with a correlation meter and fix them before you bounce.
Unsure about the vocal level? Send two stems
Listening fatigue makes vocal balance hard to judge. If you're in doubt, export two stems:
- Vocals — all vocal tracks with their processing and effects on.
- Instrumental — everything else, with processing and effects on.
Both stems must start and end at exactly the same points, keep full tails, and be exported with your master bus processing still active. That lets us fine-tune the vocal-to-music balance without losing the glue of your mix. A vocal + instrumental job is charged at the PH Enhanced rate, $80 per track.
Questions about preparing your files? Read the FAQ or get in touch before you order.
Mix
ready?
Order online and upload your files — the 48-hour clock starts when everything's in.
