PRE-PREP
If you are communicating with client prior to receiving stems, please direct them to my list of demands here: https://www.rossbfortune.com/mixing/how-to-send-multitracks-or-stems-for-mixing/
If they have any questions, they’re welcome to ask me by email or phone.
MIX PREP
This list is written in order. Completing any of the steps out of sync is likely to cause you exponential headaches down the line (as I’ve repeatedly learned the hard way…). Good luck, brave adventurer, and if you have any questions, just let me know. You can download my latest mix template here.
File Management
The folder hierarchy for each song should look like the screenshot below.
Create an artist folder
Create a song folder with the name of the song, prefixed numerically 01, 02 etc according to the order we received it.
Add Misc folder to song folder
When you download the parts, prefix the folder name with the date that the parts were sent by the client (reverse format please, e.g 220618 for 16th June 2022) and a sensible name to identify them if it’s not obvious from their folder name (e.g 220618 Parts from Charlie).
Sometimes new versions of vocals or other parts will arrive weeks later, so these also need to be organised and named accordingly within the Misc folder.
The session folder goes inside the song folder – see next step
Create a new session from my template.
(Please save my template to your User Templates (File > Save As Template…), making sure to tick Include Media when you save the template, so that the blip file is included).
All sessions should be interleaved, 24 bit / 48kHz (unless a higher sample rate has been requested – check the sample rate and bit depth of whatever files have been delivered if in doubt). Files are often delivered at 44.1kHz, but I prefer to work at 48kHz.
Save the new session as the song title, within the song folder.
Use Import Audio to import the raw audio tracks from the client (ensure sample rate conversion is ON if importing 44.1kHz files)
Always tick Copy Audio on import
Move the newly imported audio channels to the bottom of the session, ignoring most of the template for now.
Set Inputs on new audio channels to No Input.
Set the outputs to =MIXBUS=
The I/O settings included in the template should automatically route new tracks to =MIXBUS= by default, but double check just in case.
Make sure they’ve included a ref mix! If you can’t see one, check with me in case I received it by email separately.
Set Session BPM according to ref mix
Make sure vocals are dry and split out into individual tracks – one track per vocal.
If they’ve only included wet vocals, tuned vocals, or the vocals are dramatically EQ’d or compressed, check with me to decide the best approach. I might ask you to chase up with the client.
The ideal situation is to have individual raw dry tracks for each vocal, a separate folder with their tuning on, and separate 100% wet stems for any effects they’ve been using (and/or screenshots with their FX settings). This isn’t always possible, but one can dream…
Compare stems to ref mix to make sure there are no parts missing and that the structure and arrangement are EXACTLY the same.
It’s helpful to match the volume of the ref to the multitrack (the ADPTR AB plugin has a good automatch function).
If there are any arrangement discrepancies in the rough vs the parts, make a note of each before moving on (e.g. guitar missing 1:10 – 1:21, vocal take different 2:10 – 2:15)
Please listen carefully to everything all the way through, especially vocals. Sometimes the ref will have a different arrangement, different lyrics, or use different takes in the vocal comp. Artists be nuts.
If the arrangement is all correct, but the ref is sonically very different to the multitrack, please flag it with me and I’ll take a listen and see if it’s worth chasing.
Use Insert Time / Cut Time to have the first significant downbeat of the song land on Bar 5 (or Bar 9 if there’s a longer intro – this is a visual thing for when the session is zoomed out – downbeats should land on the lines)
Add Section Markers
Please use the following format: intro – v1a – v1b – pc1 – c1 – post1 – v2a – v2b – pc2 – c2 – post2 – m8a – m8b – c3a – outro – end
v1b = halfway through verse 1. Sometimes the verse is short enough that you won’t need the v1b part.
Remove any superfluous numbers from track names (eg kick.01_02) or unhelpful suffixes (e.g. _bip)
Rename tracks sensibly (if necessary) e.g. Kick In, Kick Out, Snare Top, Snare Btm, Bass Synth 1, Bass Synth 2, Ac Strum, Gtr Pick, Mando, Juno Chords, LV, Ch Dbl 1, Vrs Dbl 2 etc. Names that make it clear what kind of content the track contains.
Organise BVs / Harms / Stacks
Harmony lines will sometimes arrive with inconsistencies. Eg, Harm 1 might contain a major 3rd harmony in the verse, but switch to a unison in the chorus, while Harm 2 starts as a unison then switches to a major 3rd in the chorus. It’s easier to balance and generally be sane when everything is on tracks that make sense, so in this example the Maj 3rd lines could all be put on the Harm channels, while the unisons could go on new Ch Dbl channels etc.
Colour Code the tracks. See track folders for reference: Drums light blue, bass brown, gtrs green, keys purple, lead vocals pink, backing vocals hot pink…
Save As… Song Name_01 Split Mono
Logic and Ableton tend to export mono files as stereo files, which use 2 voices in Pro Tools and doubles plugin CPU usage, so we can split these tracks in to mono and get rid of the “stereo” source.
First check if the stereo file really is mono throughout (beware, sometimes a mono instrument will have a stereo verb/delay automated somewhere in the track!). Use the Downmixer plugin to invert the phase of one side (tip: you can save this as the User Default to save time on each instantiation) to be 100% sure – if it plays back completely silent throughout (check the output meter in the Downmixer plugin), it’s definitely mono. If you can hear or see any signal, do not split the channel.
If it’s definitely mono, use Split Mono, then delete the original stereo track and the unused .R mono track. Keep the .L track.
Centre the pans on any split mono tracks
Use clip gain to raise the level of the mono audio clip by exactly +3dB
Sometimes hard panned backing vocals or gtrs etc will arrive on a stereo track. In this case, split silence, then delete the stereo track and whichever mono track is silent. Leave the pan on the remaining mono track where it is. You don’t need to clip gain the resulting clips in these cases.
Use Batch Rename to remove all resulting .L / .R suffixes
Group the audio tracks
Name this group according to how you’ve named the folders of parts within the Misc folder in the Finder ((e.g 220618 Parts from Charlie).
Select all of the audio tracks and Use Batch Rename to add _ORIG as a suffix (e.g Kick In_ORIG). This is so we have a clearly marked playlist with the original audio before we do any editing etc)
While still grouped, make a duplicate playlist.
They should now all have a _ORIG.01 suffix. Use Batch Rename to Replace the _ORIG.01 suffix with nothing (leave the field blank) so we’re back to nice track names.
Gain Structure – Clip Gain
Check that Pre-Fader Metering is enabled.
You’ll probably find it easiest to work with settings as below in Pro Tools preferences.
Loop a chorus (or a loud part of the song where all/most instruments are playing)
With the Group still enabled, select all of the audio clips in all audio channels
Enable Show Clip Gain Lines ( Ctrl Shift – ). Assuming they’ve exported their mix so that the summed individual channels sound exactly like the ref mix, it’s easiest to keep those relative balances as a starting point, by adjusting all clip gains together.
We’re aiming to have each individual channel metering somewhere around -18dB to -12dB (ideally -16dB, but raw vocal channels are often much quieter than drums, for example, so don’t worry if they’re not perfect). With the metering preferences set as shown above, the color break at -16dB will make this task easy.
Make sure to adjust the clip gain of all clips in the session simultaneously so that the relative levels remain intact!
Gain Structure – Faders
With Group still enabled, set the volume faders of the audio channels to -6dB. The =MIXBUS= channel should be metering around -16dB now too, although you might need to trim the channel faders even more (try -7dB, or –8dB, or any whole number). If it’s metering a couple of dB below –16dB, don’t worry.
Use Strip Silence to remove silent sections on everything except for live drum mics (I’ll figure out what needs keeping and what doesn’t for live drums)
It’s a good idea to maximise the clip audio zoom (cmd + alt + ] ) before you do this, so that you can see any printed reverb tails or fade ins to make sure they won’t be cut too early.
These settings should usually do the trick: Threshold -75dB, Min Strip Duration 200ms, Clip Start Pad 400ms, Clip End Pad 1500ms
If there’s non-musical noise or hum between phrases, generally speaking you can just chop it out, but be wary not to remove any of the actual audio!
Bonus Speedy Gonzalez tip for Strip Silence:
Go to the first channel you want to strip, and click in the timeline after the end of the longest audio clip in the session
Hit Shift + Enter to select from there back to the start of the track
Now you can keep the mouse hovering over the Strip button, then use ; to move down to the next track. You can repeat the click then ; sequence to whip through all of the tracks like a mad ting.
Tidy up clip tops and tails with fades, making sure you don’t remove any transients at the start of clips, especially drums/perc (3ms Fade In and Out using Batch Fade should be fine, and we’ve left a 400ms pad when stripping silence).
Save the session (you should still be on Song Name_01 Split Mono)
Save As… Song Name_02 Cleanup
DeNoise Guitars or other noisy tracks
Not always needed, but RX Spectral DeNoise is great here.
DeClick Lead Vocals (or any prominent vocal tracks – if in doubt, just ask)
Duplicate the lead vocal playlist.
Rename the new playlist LV_Dclk
Open RX DeClick in AudioSuite (see setting below – handy to save these as defaults) and run it on the LV_Dclk channel
Go back to the main LV playlist and select playlist view
Insert a compressor on the lead vocal channel (BF-76 or most 1176 emulations are good for this job). Set the attack around 1 o’ clock and the attack at its fastest and aim for between 10-20dB of gain reduction on the meter. Insert an EQ after the compressor and add a shelf from 3k upwards, boosting around 6dB. This will make clicks more obvious.
Now listen carefully through to the lead vocal. Wherever you hear unpleasant clicks or mouth noises, paste from the Mdc playlist. If the click hasn’t been removed from the Mdc playlist, try the Dclk playlist. If that doesn’t work, add a marker labelled LV CLICK and note that there are remaining clicks in the comments.
Time Correct Vocal Tracks
Check with me what kind of timing we want to do. It could be just aligning consonants, maybe moving single words that stick, or going super tight with VocAlign.
Make a new playlist with suffix _Time (Ch Dbl_Time)
Remove Breaths from Stacks & other BVs (but leave the breaths in for lead vocals!)
Use clip gain or edit clips to remove breaths.
Add fades to avoid clicking
Tune Vocals
Use AutoTune graphic mode to tune the LV track
Route the output to a new track, and call this LV AT
Move any post AutoTune plugins from the source track over to the new print track
Set the output fader of the source track to unity. Copy previous fader setting to the new track if necessary.
Print the output of AutoTune to the new track, listening carefully for any glitches.
Hide & Make Inactive the source track
Save As… Song Name_03 Route
Move the tracks into their relevant folders and route accordingly.
The drum routing is pretty involved (soz) – Kick In and Kick Out should be placed in the =DRUMS= / Kicks folder and routed to Kicks.
Do same for snare mics, in the Snares folder, routed to Snares.
Hats, toms, overheads, and crush mics go in next (in that order). Route these to Drums GRP.
Any room mics go in =DRUMS= / Rooms, routed to Rooms
Bass guitar and synth basses go in =BASS=.
Anything that isn’t guitars or strings can go in the =KEYS= folder, and routed to =KEYS= (don’t worry if they’re not actually keyboards – any miscellaneous instruments can go in here).
Electric guitars, Acoustic Guitars, Piano, Strings etc are pretty self-explanatory…
If there’s just one Lead Vocal, use the LV1 folder and routing. If there are two, I’ve included a second LV path in the template for the second lead vocal.
You can double check that all your routing is correct by just hitting play. If all goes according to plan, it should sound the same as before, and you can use the ref to make sure nothing has disappeared into the dusty crevices of the template.
Save the session (you should still be on Song Name_02 Route)
Save As… Song Name_04 READY FOR ROSS
Thank you, you are rad.