Recording - June 2016 - 57

But how did I program those crossfades? Let's take another short side trip
into the past.
At the Automat
Back in the late 1970s, as 24-track
recording became more common,
recording engineers faced a problem:
they didn't have enough hands to ride
all the faders for a complex mix, particularly if there were changes in EQ, compression, etc. as the song progressed. A
mixdown session became something of
a performance piece, with studio assistants, stray bass players, and even the
house accountant pressed into service to
push one fader or button. (I remember
one session where my sole task was to
boost the level of one particular drum
fill, lasting three beats, by 3 dB.) And
everyone had to do everything right, or
start the mix over again.
This wasn't a tenable model for the longterm, and console manufacturers began
introducing console automation, which
memorized the fader moves and allowed
engineers to repeat them ad infinitum,
revising and updating as needed.
When DAWs came along, one way in
which they mimicked consoles was with
automation. Take a look at Figure 3,
which shows two channels of the
Ottoman Underground recording; the
upper channel (Will's banjo) has a wiggly line at the top. That line automates
the motions of the fader. At 1:48 it
drops from full scale to a little lower, at
1:53 it drops a little lower as the clarinet in the adjacent track comes in, and
a little after 1:55 it jumps back up.
Tweaking the fader in this way to change
volumes is known as gain-riding, and the
ability to ride gain well on a recording or
mixdown was one mark of a skilled recording engineer. Automation made the job a
lot easier; if you made a momentary bad
move, you could go back later and fix it
without having to redo the whole mix.
Getting fancy
Automation on hardware consoles
was pretty limited. On most of them, you
could program fader moves and muting
on the individual mixer channels, but
that was it; you couldn't automate fancier functions like EQ and compression.
Clever studio engineers, however, figured out a way around that.
I've included an example as Figure 4;
in it, you'll see two stereo WAV files with
mix-automation lines superimposed on
them. The lines mirror one another; as
one goes down the other goes up, making a crossfade. The mix begins entirely
with the top version, but by the time the
song is about 75% complete, the DAW
has faded over to the bottom version.
Why did I do that? As it turns out, I needed to change the EQ as the file progressed,

going from flat EQ at the beginning to a
+4 dB high-frequency shelf at the end.
Some DAWs nowadays can automate
plug-in settings like that EQ shelf, but some
can't; mine couldn't, so I had to get tricky.
Back in the day, an engineer would
have routed the signal from the tape track
that needed fixing into two channels of
the console, rather than one (this is
known as multing the track to those two
channels). The engineer would dial the
desired shelving EQ into one of the channels, leaving the other set flat. He or she
would then set the automation to begin
with the flat channel all the way up and
the EQed channel all the way down. The
automation would then slowly fade the
flat channel down while fading the EQed
channel up to full level. The audible
result? The mix would slowly morph from
flat to high-boosted as it played.
That's what I did in the Figure 4 mix,
more or less. Instead of multing a recorded track to two physical mixer channels,
I made an additional copy of the track
with the word HIBOOST in its title, performed the shelving EQ operation on the
file, and saved it. I loaded the flat track
into Channel 1 of the DAW, loaded the
HIBOOST version into Channel 2, and
drew in a crossfade, going from all-flat
to all-HIBOOST at suitable points in the
song. I told the DAW to generate a mixdown, and the resulting file slowly morphed from flat to EQed.
You can use this technique to morph
from an un-messed-with track to one with
compression, or just about any effect
that a DAW will do. If you use two files
(one pristine, one messed-with), you can
even open the track in an external program and play with it, then save the
result and use the DAW's crossfade function to morph at will.
Moving right along
The last two articles describe bare
basics of what a DAW can do-I
haven't even mentioned MIDI sequencing, and every DAW has its own fancy
functions, though they all fall more or
less within the
"recorder/console+effects/word processor" paradigm. For a look at some of
these possibilities, read Recording's
reviews of DAW software and plug-ins.
But we need to keep going. Next I'm
going to talk about a practice that has
become nearly ubiquitous in modern
recording, though I didn't actually use it
on the Ottoman Underground sessions.
We'll generate some creative fiction
about what I might have done next
month-stay tuned!
Paul J. Stamler (stamler@recordingmag
.com) is a recording musician, engineer,
educator, and collector of bizarre vintage
recordings, living in St. Louis.
RECORDING JUNE 2016

http://www.placidaudio.com

Table of Contents for the Digital Edition of Recording - June 2016