Exporting for a mix - Your process?

Wondering whether anyone has gotten this process down to a fine (and easy) art?

I have 70 tracks of audio in a project i just finished tracking. Obviously along the way i’ve had to create rough mixes for people to hear what’s been done and how something *might sound in the context of a mix. So there’s automation, compression, sends etc.

I am not a skilled mixer so i get others to do that and they tend to use Pro Tools so i always have to give them stems.

Wondering whether anyone here does this (i assume they do) and what process they go through to get the best quality stems exported. Obviously using batch export is handy here but how much do you disable in the way of inserts, sends and automation before exporting?

Thanks some opinions here would be hugely handy

Thats a very complicated question!

A simplified answer may be along the lines of…

Most tracks should probably be exported as 24 bit original samplerate unmodified waves, unless there is good reason not to.

Maybe an effect is the sound or automation is providing the desired result.
You may have created a sound by a number of articulations that is essential to the whole, these should be kept.
Or provided for a reference.

Nothing worse than receiving a bunch of wavefiles that have closed down all your options.
But as said any essential modifications are well worth keeping.

So no simple answer, it all depends and is entirly at your mercy/decision.

What Split said…and:

  • make sure no levels are too high (ie: clipping);

  • Make all stems the same length (start to finish of track) so that the engineer can simply line up the starts of all waveforms and will know everything will be in the right place;

  • Make a note of the tempo and the key of the track if possible.

  • A more obvious one – Allow some empty space at the start and end of the track to make sure nothing gets cut off, (eg Reverb tails)

Thanks. But knowing which sounds and plugs you want to commit to is kind of a given… I wanted to hear from anyone who does this a lot and might have a simple check list they go through (or even useful stored macros) to make the job a bit easier.

For what it’s worth I will share my method for preparing a project for a mix and perhaps anyone who is interested can give some feedback or point out any insights they might have? I’ll go by type of track as this can affect how I treat it.


  • Cross-fade all edits
  • Remove (or disable Read / Write) on any channel automation
  • Take off any volume/gain changes or envelopes added to events (select all the events and set to 0 db on volume in info bar)
  • Centre-pan all tracks and set faders to unity volume
  • Remove group assignments – send all channels out through main output buss
  • Use batch export to select each drum channel for export and export to a folder on desktop

Often I record electrics using 2 mics on a cab. These go on to two audio tracks which are then routed to a mono group channel (since electric guitars are mono in my projects). This means I need to export the output of a mono group for each guitar.

  • Blend to taste, the two channels feeding the mono guitar group – often ends up being 2:1 ratio of SM57 to condenser
  • Centre pan both audio channels
  • Centre pan mono guitar group and set to unity gain
  • Do this for all guitars and their mono groups
  • Batch export all mono groups to same folder on desktop

Once a good vocal comp has been crossfaded and listened to in solo (for any dodgy edits, breaths, croaks, moans etc.) it is bounced and placed – (rather than replaced) “in front” of the component parts in the bottom lane. It is then tuned and listened to again in solo for any obvious artefacts. Then

  • Disable inserts compression and EQ (as these are almost always applied using built in plugs anyway so no great shame)
  • Route out through main buss
  • Bypass sends and automation
  • Set fader to unity and centre pan
  • Batch export in mono

Stereo stuff
Stuff like stereo synths I’ll record directly onto a stereo audio track so these can be treated like the mono ones and batch exported but in stereo. Acoustic guitars or any other stereo channels (eg stereo FX busses I might like a lot) formed from two or more mono channels I export using the solo buttons on channel faders and the main stereo out. This can be cumbersome…

…In fact the whole process is cumbersome (as must be reading this!) and obviously fraught with human error because it is so manual. For this reason when I have finally exported all stems I open up a new project, import them all and check all is well, nothing is clipping and nothing is too quiet.

Phew! Now I can refer back to this post whenever I need a check list! I hope this has helps somebody, somewhere – someday!

It’s hard to tell just what experience anyone with a general request has, sounds like you have a reasonable process worked out!

I agree with everything above except for combining the two guitar mics.
I have strong opinions about multi-micing and phase, and I hate receiving comb filtered fuzz guitar tracks.

BTW, after noting how hard it is to export everything, maybe you should just mix it yourself, in half the time!
At least it would sound the way you had visualized it yourself.

Some people would simply multi mic that way they don´t receive comb filtered fuzz guitars…

As someone who does mixing…I don’t want your stems. I want your tracks…sans any plug ins…I want the WAV files, timestamped–WITH any editing DONE. If I edit, it’s to remove a noise or other sonic reason…Otherwise, if there are ten vocal lead tracks in those 70, I want ONE of them–the one you’re comfortable with as “the” lead vocal as an artist.

I don’t produce. Remotely…it’s a losing battle. Content=not me. Every time I bend my own rule I end up doing revisions where ALL I’m doing is “hey, can you edit out this chord?”…“hey, can you fade X before I play that thing I no longer like?” Fail. Not my job. Here’s the rule, you send me what you’ve got, and I will do my damnest to keep as much of the content IN as I can. No promises.

Anyway, If you sent me a drum stem, I would refuse it. If you knew how to mix drums as well as I do, you wouldn’t be sending them to me to mix. I question someone who wants to mix your stems. I mean…is it inexpensive? Like really inexpensive?

So the process should be do a “save as” so you don’t mess up your working mix…reset the mixer…bypass all plug ins…and do two batch exports–one of the mono sources with “mono downmix” selected…and one of the stereo sources. Alternatively, you could set up mono buss, assign all mono track to it, and batch export in one pass (if the track is assigned to mono buss, Cubase will export it as mono. I personally think this is a bug. If I select a mono track/channel to export, I dont give a crap what it’s sent to in the mix, I want a mono file…but, that’s a grumble…I’m sure they do it because people insert mono to stereo plug ins and want to render…

Anyway, that’s how you get the tracks out for mix. I don’t want a stem except in very specific cases…you’ve done all kinds of virtual instrument orchestral stuff–I’d prefer a stereo mix of it–that may be 30 different channels. I don’t want to conduct your virtual orchestra. Or I’ve had people do sort of keyboard stacks that they use a bunch of different VIs but it ultimately is one stack playing a part…just give me the mix of the components in one…but, as a rule, I don’t welcome stems.

BTW…this process is important for backup, too…for those mixing themselves. Do NOT assume if you save the Cubase folders you have what you need. That is proprietary stuff. Take a few minutes–export timestamped, broadcast WAV. You’ll thank me ten years from now.

This is a discussion for another thread, but if you’re multi-micing, you ARE comb filtering, it’s physically impossible to do otherwise.

Yeah, I definitely don’t want the artist blending two mics on whatever monitoring set up they have. Give me the two tracks. I may or may not use them both…may get panned differently…may get continuous phase processing on one to make the filter less filter-y on the fundamental…lots of mix options that I would NOT want them deciding and then having me “mix” it.

It would be like this. “I need help painting my house…I’m thinking of a kind of muted earth tone…what do you think–I went ahead an bought this red paint.”

There are very few straight sonic advantages in DAWs. Gates–manual or look ahead…and the ability to have virtually unlimited tracks so that you can print both mics rather than making the on the spot call (at home often in headphones or lousy monitoring)…

I never said otherwise. But surely if you get “comb filtered fuzz guitars” you’ ve probably not done it very well. And surely you don’ t record a drumset with only one microphone, but simply take care and use your ears…

Quote: “you’ ve probably not done it very well.”


But thats just the sound I wanted, then some mixer went and time aligned them and made it sound all big and full again… Bstards :laughing:

But thats just the sound I wanted, then some mixer went and time aligned them and made it sound all big and full again… Bstards

LOL- That’s the best argument for mixing it yourself.
You can make it sound as bad as you please!