Prepping for a mix session: Bars, Bit rate, and Knowing What to Print

The mix starts before any mixer opens the session, it starts with how the session was prepped. I've had files arrive in beautifully organised folders that take hours to untangle, and I've had "messy" sessions that took me ten minutes to get up and running. The difference is never tidiness. It's whether the person bouncing the tracks understands what a mixer actually needs.

Here's the short version: export everything from bar 1, export both wet and dry (separate folders), and don't strip out effects that are part of the sound.

Export from bar 1, every time

If a vocal comes in at bar 109, I still want that file starting at bar 1, silent until bar 109. Not trimmed, no offset and not labelled "starts at bar 109".

Why? Because the moment I have to think about where parts start, I lose the one thing that makes your multi-track session usable: everything should line up against everything else automatically. I need to drag your parts straight onto a new mix session and every part sits where it should in the arrangement. No mixer wants to think about where this particular kazoo overdub needs to come in, I want to push play and mix. It also protects you. If there's ever a question about timing, a re-comp or additional parts, it's all referenced against the same bar 1 as everything else.

While we're on the subject of not making extra work for anyone: the mix session runs at 24-bit/48kHz. So if your exported parts come out matching that, nothing has to be converted, resampled, or dithered before we start, and nothing sits around getting quietly degraded by a format change nobody asked for. Export at 24/48 and everyone's a winner.

Wet and dry, not "the good version"

Give me both. The dry, unprocessed signal, and the wet, fully-processed version, both starting at bar 1, both the same length. The same goes for reverbs and delays on their own aux returns rather than baked into the source track wherever possible. If I love what you did, I use the wet file, this is the case more often than not, but lead vocals I usually process from scratch. If something's not translating, or fighting another element you couldn't have heard in isolation, I've got the dry signal to rebuild from without asking anyone to reopen a session six weeks later.

But don't strip everything

There's a version of "always give me dry" that gets taken too literally, and it does real damage. Some effects aren't polish, they're the instrument. A synth patch that's built around a specific delay, a vocal doubling effect that's part of the phrasing, or distortion on a bass that defines what the part is. Pitch correction, whether it's used as a creative effect or a repair tool, is a production decision. Strip those out and you're not giving me a clean canvas, you're giving me a different, worse part.

The rule I like to live by is simple: if removing the effect would make someone say "that's not the sound anymore," it stays on the bounce. Effects that are about polish, glue, or taste at the mix stage, those get printed wet and dry. Effects that are the sound design itself get committed. When in doubt, ask, or just give me both and a note explaining what's what.

Naming track files

It sounds obvious but it isn't, apparently. Please don’t put the name of the session, BPM or artist name on the multi-track parts, I only need to know the track name. "Vox Lead," "Vox Dbl," "Vox Adlib 1," not "Audio 14," "Audio 22," "Audio 31." I will figure it out either way, but every minute I spend guessing what's what is a minute not spent mixing, and on a fixed quote that's not free time, you’re paying for the time.

One consolidated take per part where possible

If there are twelve vocal takes still sitting on separate tracks because you haven't committed, that's a production decision, not a mixing one and it's one I can't make for you, you are the producer. Commit to your comp, then send it over.

The recording engineer's role has moved, not disappeared

Worth saying plainly: in the DAW era, "the recording engineer" as a standalone role on most sessions has largely folded into the producer's job. Nobody's booking a separate person to just push record on a Logic session in a bedroom. What used to be split across a producer, an engineer, and often an assistant is now one person making every one of those decisions in real time, from mic choice through to how the take gets compiled.

Which is exactly why the state of a session tells me so much about the producer's work before I've heard a single note. Good gain staging, sensible track names, printed effects that were clearly chosen on purpose rather than left on from a preset, arrangement decisions already made rather than forty muted alternate takes still sitting there unresolved, that's not tidiness for its own sake. That's the producer having done their job properly. My job is to build on that, not to go back and do it for them. A session prepped the way I've described above is usually a good sign the rest of it was handled with the same care.

Don't forget the rough mix!