Recording - May 2016 - 8

Note to faithful writers: When sending your letters by email,
please remember to include your name and where you hail from.
Filtering from top to bottom
Dear Paul Stamler: I continue to look forward to
each new installment of the First Steps series. It's like
getting a course in recording in a way that refocuses
every step in the process. What a great contribution!
I enjoyed your refreshingly light touch approach to
EQ in the December 2015 issue. Describing EQ as
controlling atmosphere was an enlightening and helpful
metaphor. But it did raise one question. I recall reading
other articles-in Recording, I think-urging readers to
put a high pass filter and a low pass filter on every
track. The idea, as I remember it, is to eliminate sonic
energy or bits of data that we don't really hear.
I think the argument went that low frequency content, at least, can build up as multiple tracks are combined, so that filtering it out should become a
recordist's standard practice. Do you buy this idea,
Paul? Thanks,
Tom Bowman

Paul J. Stamler replies:
Hello Tom, and thanks for the compliments. To
answer your question, I don't routinely add filtering to
recordings, for a variety of reasons. I'll talk about the
low frequencies first.
I find that I seldom see excessive low-frequency content
in the recordings I make, for a couple of reasons. First,
I'm lucky enough to record in a studio with exceptionally
good isolation from external noise, including lowfrequency noise. Second, I usually record using cardioid
or hypercardioid microphones, which don't have the
super-extended low-frequency response of omni mics such
as the DPA 4011. In addition, some of the mics I use (like
the Neumann KM 84 pencil condensers) are transformercoupled, which further limits low-frequency response.
On the high end, I've mentioned that I often record
at a sampling rate of 44,100 samples/second. An
A/D converter at that rate must incorporate steep antialiasing filtering at about 21kHz-it's not optional, it's
a built-in feature. That filtering is drastic enough to
wipe out ultrasonic content; I don't need to add to it.
So perhaps the clearest answer I can give is that I
don't routinely add additional filtering to what's inherent
in the system, and I don't advocate that others do so.
I'm old-school enough to follow the principle of doing
as little as possible to the signal, or maybe I should say
"as little as necessary." Most filters add phase-shift to
the signal (the steeper the filter, the more phase-shift),
and I don't like to do that unless there's something to be
gained; for example, a 100 Hz highpass filter used to
compensate for a directional mic's proximity effect is
often a worthwhile trade-off. Peace-PJS

Tom, allow me to add a bit of possible clarification from the
Editor's chair. Unless I miss my guess, you are remembering
some advice given by Scott Dorsey on cleaning up noise buildup
by removing frequencies that are less important on those tracks
that contain no useful musical information in those frequency
ranges. One example would be rolling off the highs of a bass
track (realizing that important information exists well above 1
kHz due to pick noise and string harmonics); another would be
rolling off the lows of a ride or crash cymbal recording to reduce
bleed from the kick drum.
Paul's advice to do as little as possible to the signal is a good
one, and how much you try to carve away stuff that doesn't matter must ultimately be a musical choice that you make on a trackby-track basis. Thanks for your letter.-MM
Where'd the lows go?
Dear Paul Vnuk: I read your review in the March 2016 issue
about the PSI A14-M monitors. The frequency range is 56 Hz on the
low end. Does this mean that the E-string fundamental on a bass guitar (41 Hz) cannot be heard through these speakers? I have always
wondered about this regarding smaller studio monitors. Thanks,
Steve Harrow

Hi Steve: Paul is hip deep in sessions this month and asked me
to field this one.
You're correct... if you were to play back a sine wave at 41 Hz
on those speakers, you would only hear it very faintly in comparison
to, say, a sine wave at 60 Hz. It'd still be there, but much quieter.
Fortunately for us, we don't make music with sine waves!
A bass guitar relies heavily on the harmonic structure of the
vibrating string; the fundamental may be almost inaudible, but
the first harmonic is at 82 Hz, easy to hear on speakers that are
3 dB down at 56 Hz. You lose the "oomph" of moving air at
those low frequencies, but your brain is capable of reconstructing the fundamental from the upper harmonics.
Be careful to realize that it's not a question of needing a
speaker of a certain size to produce any audio of a certain long
wavelength (low frequency)... it's a matter of needing a speaker
of a certain size to throw sufficient low-frequency audio any sort
of practical distance in comparison to higher frequencies. Sound
disperses according to the inverse-square law; twice as far
away, one quarter as loud.... three times as far away, one ninth
as loud... etc. That means (in very oversimplified terms) that it
takes a lot of power to throw sound a long way, and the efficiency of the process at low wavelengths decreases as the speaker
gets smaller.
Flipping that around, that's why you can get great bass
response from the tiny balanced armature drivers inside in-ear
monitors... they're inside your ear canals and only firing audio a
half inch or so down a sealed tube, rather than throwing it three
feet or more across a room. It doesn't take much to do the job in
that case! Thanks for reading-MM

Write to us at: Talkback, Recording Magazine, 5408 Idylwild Trail, Boulder, CO 80301
Or save stamps and send email to talkback@recordingmag.com
8

RECORDING MAY 2016

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