Recording - November 2017 - 25

tends to ripple out equally in all directions, like a threedimensional pebble in a pond. Above the Schroeder Frequency, the sound will "beam" more, traveling in a straight
line until it reflects, gets absorbed or diffused, or simply travels far enough to run out of energy.
In these higher tones above the Schroeder Frequency, we
are more concerned with reflections. Early reflections, which
arrive sooner than about 25 ms in time, are generally not
heard as distinct echoes, but rather are perceived by how
they affect the original sound. In terms of frequency, these
reflections interfere with one another and create comb filtering
(narrow and steep peaks and nulls). They confuse our psychoacoustic sense of directionality and stereo imaging, since we
are hearing basically the same sound coming from several
places. We can test both of these with frequency response
and waterfall graphs, as well as impulse response, which we
will look at in more detail next month in Part 2.

SO HOW SHOULD MY ROOM SOUND?

With all these concepts in mind, how should our rooms
sound? For an accurate listening/monitoring environment,
we generally aim for an average RT60 between about 0.4
seconds up to maybe 0.8 seconds or even 1 second if it's a
larger room. For recording rooms this can be longer, 1-1.5
seconds or so. More importantly, we want RT60 times consistent
across the frequency spectrum.
We also want to minimize early reflections arriving sooner
than about 25 ms. In tracking, these come from sounds from a
source bouncing off walls before reaching the mic. In mixing,
they come from sound leaving the speakers and bouncing off
walls, ceiling, floor, or mixer before reaching our ears. These
are revealed with an impulse response graph. We want to
aim early reflections away from our ears, or get rid of them
with absorption and maybe diffusion placed at reflection
points to create a "sweet spot" for monitoring or recording.
Treatments like this also begin to reduce the overall RT60 time,
as well as flattening the response out by reducing/eliminating
comb filtering.
Diffusion gives us an effective tool to preserve high frequencies in a room. If we rely solely on absorption, it can result in
a room that sounds too dead at high
frequencies. Diffusion is very effective
at leveling out the RT60 times across
the frequency spectrum without reducing
the higher frequencies too much. In
addition, diffusion just sounds good,
especially for rooms recording
acoustic instruments.
We also want an accurate bass
response, which is achieved with
plenty of bass trapping and intelligent
placement of speakers and the listening
position. Bass traps and broadband
absorbers-large and thick enough to
work at bass frequencies-will absorb some of the excess bass moving

around the room creating the interference, thereby flattening
the overall response and also reducing ringing and resonances
that occur at modal frequencies.

THE SUBJECTIVE TEST:
LISTEN TO SOMETHING FAMILIAR

After the Clap Test, the next test is to listen to the sound of
something familiar in the room, usually an instrument or a
set of speakers. It's a very useful test, because it can quickly
convey the sound of the space, but human hearing is fraught
with complexities that apply to all sensory experiences. For
instance, all A/B comparisons we listen to must be level
matched, or the test isn't really fair. We almost always
perceive the louder sound as "better", even if it's the exact
same thing just a few dB louder. And our ears are more or less
sensitive to some frequencies than others, and this changes at
different volumes; look up Equal Loudness Contours for an
explanation of this, or read "Listening Levels: Set And Forget!"
from our March 2015 issue.
So whenever we conduct informal listening tests, whether
listening to music over speakers or playing an acoustic instrument
in a room, it's best to use sounds we are familiar with. In
a recording space, this could be a favorite acoustic guitar
whose tone is very familiar. In a mixing space, it could be a
high-quality favorite recording that we know intimately. The
tonal familiarity will help us identify strange acoustic anomalies
that jump out at us in the unfamiliar room.
We know we must get the sound in the room right first. We
usually think of this in terms of the instruments themselves (new
strings, a superior rented instrument, etc.) or the players
(well-rehearsed and adequately trained and warmed up for peak
performance). However, it also includes the acoustics of the room,
which we can affect by positioning both the instruments themselves as well as the microphones. The widest variety of recording
options come when the sound of the performance is filling the
room, which allows the recordist to capture the performance in
a number of ways using close mics, spot mics, and room mics.
I'm a guitarist, so I like to use an acoustic guitar-preferably my jumbo-body instrument that produces hefty low

RECORDING November 2017

Table of Contents for the Digital Edition of Recording - November 2017