In addition to the recording and production work I am often engaged in, I do quite a bit of mastering other people’s mixes into finished works as well.
This usually entails applying a touch of EQ to sweeten a track’s overall sound; maybe cutting some lumbering lows or adding a pinch of high-end sugar at the uppermost frequencies, or adjusting the overall sonic spectrum so that all of the songs come together as a unified whole.
Mid/Side manipulations can widen or narrow the stereo field, bringing depth to reverbs, delays, and other ambiences, as well as punch and focus to content in the bass and lower midranges.
Some compression may also be applied to fatten a track or to reduce its dynamic range in an effort to subsequently bring its volume level up to a more desirable and/or compatible one.
Perceived loudness can be further enhanced by the tasteful application of limiters or volume maximizers, wringing out every drop of presence in an effort to meet or beat the volume levels of other commercial recordings…
It’s that last consideration that prompted an email from The Eruption, a band I’ve worked with before on their albums Giapetto’s Work Shoppe and Standing in the Shadow of the Scarecrow. They were pretty happy with their current project’s mixes, but they were wondering if I could just make them louder, as they weren’t quite up to the same volume levels as other releases.
I told them I’d try and they uploaded the tracks to me. After listening through, I was able to give the tracks a noticeable boost, using the techniques outlined above. I sent them back and asked if they realized that their stereo tracks had in fact been mixed in mono. Lead singer/bassist Daughn Stombaugh seemed a bit surprised, but did say, “Oh, that’s why the music feels like it’s on top of my head!” He asked what could be done about that and I told him there weren’t a lot of options if all they had were those mono mixes. If they still had the original session files, I might be able to recreate the mixes in stereo. He said he thought they did and I asked what program they used to record them. Pro Tools, he said. I told him that I have Pro Tools and that there was an outside chance I’d be able to do it, if the plugins they used were the same as mine.
The first challenge was my muted enthusiasm in working with Pro Tools. Not a slam, mind you. I’m simply more comfortable (and fast) with PreSonus’ Studio One. Heck, I even used version 3 over the newer 4 for the exact same reasons! I like to make decisions and react intuitively when I’m working, not search for the right tool, try to figure out where or what to click when selecting something, and so forth…that just gets frustrating almost immediately and slows the process immeasurably.
I first tried to export all of the individual tracks from within Pro Tools, but it just made for a lot of unwanted clutter and a little bit too much control over things. I mean, the band had already made a lot of the choices that got them to this point: the sound, balance, and panning of the drums. The tone and level of the bass, guitars, and keys. The relationship between the lead and backing vocals and where they should set within the track and so on.
I went back into Pro Tools and took another look around. I noticed that aside from the two separate bass tracks (one clean and one distorted), everything else had been assigned to a buss, which is essentially a stereo sub mix of a bunch tracks that are routed to a single fader. So, if you’re mixing and someone (probably the drummer) yells out “More drums!” you can push up that one fader and all of the drums will get louder, but with the proportional volume relationships between the kick, snare overheads, and toms intact. Same with the vocals, which I felt were already mixed really well.
Enter the concept of “Stems.” Stems are stereo (or sometimes mono) wave files that represent a chunk of the arrangement. In this instance, each sub mix was exported as a whole to its own .wav file, then placed in context alongside each other in my preferred Studio One DAW.
First I imported the drums in two versions: One with reverb, one without. This allowed me to cut the reverb trails where there were tight stops in the arrangements by cutting and pasting the reverb-less sections into the verbed performance. I added some aggressive color to the drums by running them all through my favorite compressor plug-in, the ACME Opticom XLA-3. This effect has a very cool feature that allows you to get the unique “sound” of its saturation without actually engaging the compression circuit, which brings the drums forward without squashing them, keeping them super punchy. I also added a plug-in called Haiku, to broaden the highs and narrow the lows.
Next came the Basses. I again used the ACME unit to introduce some growl on both the clean and distorted tracks. I added some bumps and dips to the EQ curves of both, to even out the frequency buildups and deficiencies where they occurred. I added the Maag EQ 2 to the distorted one for a little more ‘tude and reigned in its slapped sections with a T-Racks LA/2A limiter.
The multitude of guitars were already close to crushed, so I put them through a T-Racks Fairchild 670 limiter with just 1 dB of compression, which imparted a warm solid glue amongst them with a 1 dB boost at 10k from a Baxter EQ. with a rolloff of -3dB at 166Hz and a hard filter at 24Hz.
I ran the lead/backing vocal stem through the John Joseph Puig/Vocals processor by Waves. This is an incredible plug-in with a lot of functionality. I worked particularly with the De-esser, the “space” parameter to impart some deep, but subtle ambience and worked the “attitude” and “presence” parameters to bring this crucial element into focus without pulling them too far out from the mix.
As I was mixing toward the final master, I didn’t need to do all that much to the mixes themselves, other than make them louder, which is how we here in the first place! I used a T-Racks Stealth Limiter on the master output of each mix at about -3dB of limiting and a Waves L2 Ultramaximizer with about -2 additional limiting on the final master, with an output of -.5 dB to avoid clipping.
I sent Mp3s of the mixes to the band; they made suggestions and I sent back revised versions until we came to a place in the process where it was decided one more session “in person” would probably do it. A three-hour session and an email with two minor tweaks later, we were done!
I was especially happy to have been able to keep the mixes sounding very close to the band’s original vision, while being able to develop each element into the very best version of itself. With results like that, I can look forward to using this stem mixing technique more and finding out just how deep those rabbit holes can run!
Sven-Erik Seaholm is an award-winning independent producer and recording artist (www.kaspro.com)