How To Mix A Song

Beginning of the Mix

It’s no secret that there is an inherent conflict between art and commerce,
and it is within that conflict that a challenging divide in record making
falls. A mixer’s goal is to excel and execute all that is needed creatively,
yet the end goal is also to produce an album that sells. Anyone can trim
extraneous time spent on a mix so that there will be money saved for you
and the band. One area that can burn time away unnecessarily is getting
started. For example, when you receive a session and can’t start immediately, it can be a creative damper. In my time with Jack, I learned the
importance of sitting down and going from zero to sixty mph in a couple of
seconds. In other words, hitting the spacebar and unmuting and bypassing

some channels and plug-ins to see the tonal and sonic possibilities of the
mix come to life. This ability comes from one of the benefits of the DAW
world. Starting a mix again (prep work of routing, choosing plugins, figuring out naming, etc.) can take from thirty minutes to a couple hours. That
extra time, however, adds up over the course of a couple weeks, months or
years. And one must ask, “who’s paying for that and how it will it affect
the ultimate course of the record?” Therefore, I offer some insight on how
to expedite and streamline this process to allow you to dive into what you
really want to be doing, which is the creative side of mixing.
During the first hour of any project, I like to find my bearings by starting
to address points noted: the main areas of timing and tuning to the specifics of tones along with any bad recording anomalies. These undesired
artifacts can be clipping, distortion, edits without fades or improper fades.
If there is an excess of timing or tuning issues, I’ll get in touch with the
band/producer. If it’s a small thing, I’ll make the edit and try to accommodate and work around the issue. It’s ideal for a mixer to receive sessions
in which all tracks are cleaned up with completed edits, fades applied and
consolidated clips, and that playlists are cleaned out unless there is need
for an alternate take. Markers should be located in the correct spot, and all
tracks should be labeled with the name of the instrument. The comments
section is a great spot for information such as the mic, pre and part that the
instrument is used in. I can’t stress that enough. Somebody’s name is not
an instrument, for example. Lastly, there should not be any unused audio
tracks, and if any virtual instrument tracks have been used those should
be printed as audio tracks. Sessions sent having the playlist and clip list
cleaned out helps with file size in that you are not sending any unused
parts, which allows for faster uploading and downloading time. Plus, it
allows for lower chances of confusion in regards to any unnecessary clips
or tracks which may be questioned for the intended use in the mix.
Other important considerations are needed when beginning a mix. If you
have the luxury, decide whether you will mix completely in the box or use
a console/summing mixer. My preference is a hybrid of the two: part in the
box and part on the desk. This allows some variation in the tone as well as
the tactile and ergonomic variety of not looking at a screen. This hybrid
approach is how I mixed Buck 65’s last record, Never love. I discuss this
record more at the end of the chapter. One reason that I like to mix on a
desk, and without total recall, is that when the day or session is finished, the
mixes were either right or not. I like that challenge—it forces me to make
decisions in the moment. Buck’s Never love was mixed on a desk without
recall, as was what I mixed for Beck. There’s a certain amount of fear that
pushes me in that type of situation by knowing that I’ll never have that time
again and I can’t just go home and open the session to make changes. Thus,
the safety net of recall creates sterility and mediocrity. There are times,
however, when I don’t have the luxury of mixing on a desk and will use an
old Yamaha summing mixer from the 1970s. It is a simple six-channel mixer
with mic/line selection and three band EQs. Usually, I’ll send the drums
and bass to channels 1 and 2, the remaining music to channels 3 and 4 and
all the vocals to channels 5 and 6. This is a basic type of stem processing;

it glues the major sections together. Next, I may use a compressor on the
output before I print back into Pro Tools. My go-to bus compressors are the
SSL, SmartC2 and the Neve 33609. With this setup, I like to have a solid
set of analog-to-digital convertors going back in, because if the convertors
are average, it seems to cancel out the benefits of summing through a mixer
altogether. The D/A A/D conversion process with poor electronics is far
more degrading than the benefits of analog summing.
Over time, I’ve created mix template sessions, which are sessions only
populated with various auxiliary tracks that contain plug-in chains on the
inserts. The use of template sessions is another concept that I picked up
and developed from my time with Jack. These not only save time but also
accelerate the mixing pace. These include dynamic manipulation: parallel compressors, gates for kicks, snares and bass; harmonic variance/
tone control: saturation to distortion; width: stereo image exaggeration; as
well as pitch, reverbs and delays. If starting the first song of a full record
or series of songs, I may import some auxiliary channels or these chains
directly from the template sessions to the tracks. When I import them, they
are already named and a bus is selected; thus, I may only have to add a
send to the channel. This allows me to simply unmute several possibilities
in a short amount of time and gain vast perspectives quickly, as a result.
This approach prevents me from having to change settings and go through
a bunch of pages of the plugin. Example compressor options in these template sessions offer a variety of colors and ADSR shaping. The tones are
largely determined by their electronic design and range from various tubes,
transformers and transistors. These options will have various settings for
attack and release time along with ratios for the ability to increase attack
for punchiness or create a longer release for sustain, thus helping to bring
out the body of a snare. I’ll use high ratios to limiting (10–20–1000 to 1)
for putting a part front and center. In the reverb world, they vary in types
again based on personality and tone: from classic metallic and shimmering
plates and chambers to the odd irregularities of a spring. I continue these
options with delays (tape based to digital), harmonic effects (tape and tube
emulation to guitar amp simulators) and stereo image effects. I really love
the versatility of the Sound toys EchoBoy.
Another time-saving strategy when moving between songs is an
approach I learned during my days working in a largely analog world.
The goal is to not to zero out the console, but to leave the outboard gear
patched in. Generally, a band or artist records an album in chunks or in
one fell swoop of tracking with a similar setup of instruments and mics,
thus creating the possibility for a very consistent sound for the record.
Chances are that you can leave all of the outboard gear patched in and,
similarly with DAWs, route or import all the of same plugin chains on top
of the corresponding tracks of the next song. Then you can simply toggle
between the insert points being engaged and out to hear EQ or dynamic
changes. These are presets in a sense, ones that you’ve created, and more
than a single effect but chains of processing. Now, you may have to do
some slight adjustments but you are not beginning from scratch, which
can save half an hour or so because it eliminates the need to redo all the

routing and plug-in selection. In the analog world, this patching can eat up
hours, along with valuable time in DAWs as well. In Pro Tools, you can do
this by setting up a mixing template or using import session data to copy
inserts from your previous session to the current session’s inserts of the
respective channels. Chances are the EQ and dynamic adjustment should
be close to what you attained on the previous song. When you import from
a template session, you bring in a few choices per key insert points on the
main instrument tracks. The result is easily toggling the bypass in and out
to offer a variety of sonic possibilities. Add to this a couple back buses
scenarios and you can go from zero to sixty in a few minutes. This allows
you to save time and to progress creatively in the mix exponentially. For
every record I mix, after I’m done with the mix I typically will make a
template session or export key plugin chains to these mix template sessions. To clarify, I’ll have several back bused (parallel) chains for main
elements: kick, snare, overheads, toms, bass, guitars, vocals, etc. Then,
there a variety of reverbs and delays. This works in line with the concept
of using presets of plugins as a starting point. All of these chains usually
have some tonal commonalities: for example, an EMI compressor to an
EMI EQ. As I stated at the beginning of this section, the way I have these
organized falls basically into two categories: colorful to transparent. This
is usually determined by whether they have an origin ranging from tube to
transistor. Other specs considered are discrete, transformer coupled or not,
and the implementation on an integrated circuit. Possibly, the more metal
involved equals more color. Knowing these specs can help to eliminate the
guesswork when trying to capture the desired tonal shape and color. With
mixing, you are trying to capture a moment in time, and that requires the
ability to work with efficiency and effectiveness. One should work to know
which pencil, crayon or knife to grab while it does take time to experiment. This will help in your speed and the ability to execute the sonic
landscape that you or the artist is searching for in the mix. Along with the
color aspect of the plug-in, I also consider how it allows me to detail and
control the ADSR of a signal. Broadly speaking, I decide whether I need
to retain the attack, add attack, truncate the sustain or exaggerate the body.
I may also consider whether I want to knock down and smooth out the
transients or add some punch. Knowing if the plugin is based on a tube
or transistor will offer insight into how fast the compressor will respond.
Generally speaking, the beauty of the transistors over time are their speed
of capturing transients, thus allowing for more attack. The downside to
them is the apparent loss of color. Let’s check out a couple of examples.
A typical chain for the kick could be a couple of different mics: in, out and
sub. I may give each a touch of individual EQ’ing and then sum down to
an auxiliary track for global control. For the most part, the strategy for this
setup is as follows: the inside kick captures the attack/beater for some nice
punch (the 120 Hz area) or sub depending on the mics used, the goal of
the outside mic on the resonant head is to complement the inside mic with
sub or punch, and the sub-mic like the converted NS-10 or the Yamaha
sub mic is just that sub, 20 Hz–50 Hz. What you have in this scenario
of the low-end world is the kick in segments with some overlaps, along

with what I call nesting qualities like 20 Hz–40 Hz from the sub mic, 40
Hz–100 Hz and then 100 Hz and above. This is very close to how the kick
was treated on Buck’s ‘Danger and Play’. This is a broad breakdown, but
provides one with many options for a kick sound. What to expect from
the mics is an important consideration. The mics are funneled down this
aux, which allows you to carve further to taste with respect to the song.
In regards to the kick chain, I may use some kick back busing options of
compressors broken down in this way: one for punch, one for sub and one
for attack. That’s it. I essentially imagine the ADSR chain for each signal,
which enables me to break it down and separate it. Basically, this is what
you can do with multiband compressors, but for my simple mind I like to
break it down into these planes or snapshots of time. Wrapping up using
templates, a pitfall may be that it pushes you to familiar territory, so if the
band, artist or song is requiring something fresh, don’t go that route.
Now that we’ve talked a bit about the prepping process, let’s move on to
the beginning stages of the mix.
As I mentioned earlier, my goal as a mixer is to serve the song. So,
where does one start after the prep stage? My strategy is to break down the
mix into two broad parts: production and engineering. Some things I’m
listening for from a production standpoint are the arrangement, the lead
and ancillary parts, instrumentation, fills and transitions (handshakes), the
execution of the performances in the areas of timing and tuning, phrasing
and the dramatic arc of song: the use of tension and release, how it ebbs
and flows, where the climax may be, the melodic and harmonic structure and
overall feel and mood. From a technical standpoint, I’m listening for clarity, definition, the overall balance from low to high and possible holes
or frequencies that are missing or the buildup of frequencies. I consider
whether the overall balance is muddy or too bright. Other sonic considerations are tone and timbre of the instruments: is there an overall warm and
fuzzy feel, or does it feel clinical and sterile? Also, I determine if there has
been any processing in the area of compression—too much, too little? In
the area of ambience, I consider if the mix is overall too dry, or is it more
of an established ambient setting? Is it too wet or too dry? Are there too
many effects, etc.? I’ll listen to the whole song from top to bottom in these
respective areas along with soloing certain elements to see how they sound
on their own to see if there are any anomalies or artifacts. Some things
can sound awful soloed, but in the mix is what matters with the relative
relationships. As you listen, take notes, not just mentally, but actually write
them down and make a checklist. Have a keen ear to the lyrics and note
key words or phrases that should have automation pushes (rides), as well
as effects, delays or reverb tails. What I hope to obtain by this approach
is a sense of any major concerns, pros and cons, whether I need to cover
up a poorly played part that cannot be fixed with editing or try to provide
clarity for the vocals. If the song sounds like it’s in a good spot regarding
production and engineering, I typically just take the approach of buffing,
highlighting all the beautiful moments and building off of what’s already
there. Mixing is this process of relationships and relativity. If you adjust
one element, you should consider the effect on the surrounding parts, not

just what is happening in the immediate vicinity (section), but down the
road in the following sections of the song. Analysis to me is what’s working within the mix and what seems unrealized or what’s not working to
its fullest potential. It all goes back to the song. As a mixer, I believe you
toggle the hat of a producer and engineer. On the production side, I ask,
“What does the song need?” Sometimes I believe that songs, while technically complete, may feel like something is missing, and that is something
you can provide during mixing. You may have to fly a part around, mute
something that is arranging, or rely on some bells and whistles of ‘tricks’,
weird delays, reverbs or effects. It can be like a sleight-of-hand card trick.
Mixing is guiding the listener through an aural landscape of a sonic story.
Some artists seem surprised or caught off-guard by this approach when
they realize that a mix is not just simply a process in sonics. It very much
has to do with a holistic understanding of the song and why it might then
be necessary to mute a part or fly something around. I can operate from
a strictly sonic/engineering angle and work with EQ, compression, panning, etc., but to me it’s the full picture from a production and engineering
standpoint