The Nightmare of Mixing Audio

In drummer Jacob Slichter’s excellent book, So You Wanna Be a Rock & Roll Star, he examines the music machinery behind the rise and fall of his 90s band Semisonic.  When discussing the release of their ubiquitous hit, “Closing Time,” Slichter reveals how the song went through several mixes and how each was devoted to a different audience:  Bob Clearmountain handled the initial mix, but Jack Puig remixed it to give the song a heavier edge for alternative radio, Don Gehman did a pop mix, and Puig returned for an acoustic mix.  Depending on what radio station you listened to, you might be getting an entirely different sound to your favorite song.

With the advent of streaming services, I have to believe that alternate mixes are employed less often today, but Slichter’s story illustrates just how important – how defining – mixing is to a song, no less than the musical performance.  In the hands of a gifted engineer, mixing can push a decent song into the stratosphere.  In the hands of an amateur, a really good song can end up sounding merely adequate.  I’m more the latter than the former, but I’m getting better.

I’ve sent mixes for my new album, The Human Form Divine (available on streaming services in a month or so) to a mastering engineer in Chicago, and I’m praying that I’m still happy with them when I get the tracks back.  We shall see.  I’m allowing myself a few weeks of distance from what was an arduous process filled with second-guessing, self-flagellation, irritation, bewilderment, resignation and – ultimately – some degree of satisfaction .  I am not a natural mixing engineer, and it shows; I started mixing my new album last October and finally finished in March!  It’s a good thing I don’t do this for a living.

Part of the problem is that I’m dealing with imperfect recordings.  For example, when we recorded drums and bass in the studio, we couldn’t get the snare to stop rattling whenever my son Sam hit his toms, so at home I had to go back and substitute every tom hit with a sample tom hit that we recorded in isolation (thank goodness we did this as a fail-safe tactic) and mute the snare accordingly.  I employed a similar technique on the snare when raising the volume wasn’t possible due to the accompanying hi-hat and cymbal interference. 

In the studio I also didn’t notice that the bass was drastically out of tune on two tracks, mostly because we were only recording drums and bass and there wasn’t a reference track that allowed us to easily recognize off-the-mark pitches.  Oops!  NOTE TO SELF: TUNE YOUR INSTRUMENTS BEFORE EVERY RECORDING! The bass performance on the title track of my new album was so good that I didn’t want to lose it because of tuning issues, so I ended up copying the bass tracks (one amp and one direct) six times, changing the tuning differently on each grouping to end up with one relatively in-tune performance.

But mostly the mixing process is challenging because my ears don’t pick up on subtle distinctions.  I’m good at broad brushstrokes – volume, panning, and basic compression, for example – but the nuances of slight variations of compression or EQ are largely lost on me.   Luckily, with a lot of trial and error, over time I used a few techniques that I was able to employ across the board.

Thanks to my bassist friend Johnny’s suggestion, for the first time I used saturation to help bring out a track rather than using a lot of compression.  This was a tricky balancing act, and one I’m not sure I mastered it, but I was able to get bass and snare to cut through mixes a touch better without having to squash the hell out of them. Even so, on some busses I used parallel compression a bit with good results.

I also utilized sidechaining extensively.  With the help of a couple of tutorials, I ended up doing the following:

1)      Ducking the bass slightly to make room for the kick.
2)      Ducking the overheads slightly to make room for snare and toms.
3)      Ducking guitars and/or synths to make room for vocals or solo instruments.
4)      Ducking a vocal delay bus with the original vocal track so that the delay can only be head at the end of a syllable.

The above helped enormously, as getting the vocal and snare tracks to sit in a mix has always been a challenge for me, as has the kick/bass relationship. 

Also helpful was being aware of accentuating frequencies in the 300 Hz range to allow bass guitar to be heard on smaller speakers.  I’m amazed at how the bass disappears on many professional recordings from long ago when played on tiny speakers such as those on a cellphone.  Today, mixing engineers are more cognizant of this inevitability.

I also used a high-pass filter on multiple tracks, boosted my vocal a touch at around 1800 Hz, and used a high shelf boost on overhead busses for a bit of sheen. Eventually, I created a kind of EQ blueprint that worked for these particular recordings.

Even with all the above and more, I had to go through mix after mix after mix of each song, listening on five different sets of speakers (studio monitors, stereo speakers, car speakers, a decent Bluetooth speaker and my phone) plus a pair of headphones to get a sense of what was and what wasn’t working.  And once mixes were complete, I got additional feedback from my son and two of my musician friends, Johnny and Anthony.  Luckily, by the time I sent them the final mixes, they agreed that the songs were in a good place, requiring just a few minor edits.

I have aspirations of one day hiring all of this out and getting someone who I trust to handle the entire mixing process so that I can free myself of this nightmare.  But then again, what fun would that be?  As much as I say I hate the process, I also love the challenge.

New album forthcoming!