There are many ways to approach a mix, but one way that made a huge difference for me early on, and has served me well ever since, is to think of mixing in terms of mixing musical parts, rather than mixing tracks. What I mean by “tracks” here is the individual strips of audio onto which we capture each mic, instrument, or line input. So, for example, we might have 8 tracks of drums, or 2 tracks of a keyboard, or 2 tracks of a multi-miked guitar.
To me, when we’re working at the track level, we’re dealing with an arbitrary by-product of the recording process: the technical manifestation of the musical performance in a linear, physically structured way. That’s super left-brainy, and doesn’t really have a lot to do with the original intent of the music.
Recording is (justifiably) all about the track level: we want to make sure that each discrete input is properly captured at optimal levels, with no noise, etc. But mixing should be all about the part level – about returning to the music: we want to make sure that each part is playing its intended role in conveying the intention of the song, so that the end result connects with the listener in an emotional way. That’s right-brainy stuff, and where the music comes from in the first place. That right-brain connection is an emotional thread that we want to maintain from the moment of creation right through to the moment of listening. If we haven’t maintained that thread, then we haven’t done our job completely.
So, working at the part level forces us to think in terms of music and not technology. What I mean by “parts” here is the individual and collective musical components, i.e., the drum part, or the keyboard part, or the guitar part. By the time we move from thinking about tracks to thinking about parts, we’re thinking about single things vs. the multiple pieces from which they’re made.
Once we get to the mixing stage of production, we shouldn’t be thinking of “mixing 40 tracks of music,” but rather, we should be thinking in terms of “mixing 17 musical parts” (or however many parts the song happens to have). This subtle change in mindset automatically makes us approach the mix from a completely different perspective – in my opinion, the right perspective – for serving the song, the songwriter, the performer(s), and the listener.
An extension of the part idea is the notion of sections. Sections are – again, musical – groupings of individual parts, such as the string section, the horn section, the rhythm section, or the backing vocal section. Since we create and perform these things as musical components of a song, shouldn’t we also think of them that way when we mix?
One fantastic way I found to force myself to stay focused on the parts rather than the tracks is to use groups when mixing. Some DAWs refer to them as “groups;” others call them “auxes,” “busses” and other things, but they are all essentially the same and serve the same purpose: to group (or route) a collection of tracks into a single point where they can be manipulated as one.
So using our example above, we might send all 8 drums tracks to a “drum group” or the stereo tracks of a single guitar part to a “guitar group.” When I mix, once the basic balance of relative volume and panning between the different tracks going to a group is dialed in, from that point on, I use the group to adjust the part’s place in the mix, and very rarely ever go back and make changes to the individual tracks. That forces me to work (and to think) at the part level: if there are 17 musical parts in the song, I’ll likely have 17 groups that I use to complete the mix.
And I might even have more groups, because I’ll often group musical sections into groups as well. Again, using our example above, I might have a horns group, a strings group, and backing vocals group, and perhaps even a guitars group (e.g, acoustic + electric guitars) and a keyboards group (e.g., piano + organ) where I’ll combine related parts together and, once their relative roles are dialed in, make final adjustments on the section groups, rather than on the part groups.
A wonderful side-benefit of using groups for parts and sections is that our mix is now structured in terms of the song’s musical arrangement, so it complements our thought process from a production viewpoint as well. So, when we think about when we want certain parts to play or not play, or play louder or softer, we only need to manipulate the mutes and levels on the groups, not on all the individual tracks, which, besides keeping us focused on the music, saves us a ton of time.
Another wonderful side-benefit of using groups for sections – especially if you provide mixes for film/TV libraries – is that those groups will often correspond directly to the stems (stereo sub-mixes) that the libraries often ask you for. So, if you need to run off a keyboards stem mix, there’s no extra soloing or muting work to do: simply export the “keyboards” group. No fuss, no muss!
Here’s an example of what I’m talking about… Let’s assume we’re recording a band song. The tracks might be as follows:
- Bass: 1 mic on amp, 1 DI -> 2 mono tracks
- Drums: 8 mics (kick, snare, hat, overhead left and right, 3 toms) -> 8 mono tracks
- Acoustic guitar: 2 mics (body and neck) -> 2 mono tracks
- Electric guitar: 2 mics (speaker cone and room) -> 2 mono tracks
- Piano: left and right outputs -> 2 mono tracks
- Organ: left and right outputs -> 2 mono tracks
- Lead vocal: 1 mic -> 1 mono track
- Lead vocal double: 1 mic -> 1 mono track
- 2 backing vocals: 2 mics -> 2 mono tracks
That’s a total of 22 tracks of audio. Now, our group assignments/routings would look like this:
- Bass: 2 mono tracks -> 1 mono “bass” group
- Drums: 8 mono tracks -> 1 stereo “drums” group
- Acoustic guitar: 2 mono tracks -> 1 stereo “acoustic guitar” group
- Electric guitar: 2 mono tracks -> 1 stereo “electric guitar” group
- Piano: 2 mono tracks -> 1 stereo “piano” group
- Organ: 2 mono tracks -> 1 stereo “organ” group
- Lead vocal + double vocal: 1 mono track -> 1 stereo “lead vocals” group
- 2 backing vocals: 2 mono tracks -> 1 stereo “backing vocals” group
That’s a total of 8 groups of parts. Wow, that’s much more manageable, and everything is expressed in terms of instruments and vocals, not tracks.
Just for grins, these are optional, but let’s add a couple of section groups, assuming we’re going to have to export stems later:
- Bass + drums: 2 groups -> 1 additional stereo “rhythm section” group
- Acoustic guitar + electric guitar: 2 groups -> 1 additional stereo “all guitars” group
- Piano + organ: 2 groups -> 1 additional stereo “all keyboards” group
- Lead vocal + backing vocals: 2 groups -> 1 additional stereo “all vocals” group
That’s a total of 12 groups. Still a lot less than the 22 tracks of audio we started with, and still totally music-focused.
Here’s what it all looks like in a diagram:
Now, you might be thinking, “wow, that seems like a lot of setup work!” But in reality, it’s not. That’s one of the many fantastic benefits of today’s DAW technology: you only need to set up this structure once, save it as a project preset or template, and use it over and over again. That’s exactly how I do it at Azalea Studios, and it has probably save me hundreds of hours over the years – not to mention potential mistakes and headaches.
How do you lay out your mixes, and just as important, what do you do to maintain focus on the music (rather than the technology) when you mix? Let me know in the comments below.
If you liked this article, you might also like:
Learn insider mixing tips, techniques and secrets from a seasoned pro. Learn how to think differently about how you approach a mix, and how to get that elusive “broadcast quality” sound from your own mixes.