It seems that most discussions about loudness management tend to
focus on distribution as opposed to the content creation side of the chain. In
fact, loudness management works best when content is properly managed from the
beginning rather than somewhere downstream.
The document at the heart of the CALM Act, ATSC A/85, is very
comprehensive and does its best to address loudness management at every step.
There are many areas of the document that are of particular interest to
television audio mixers. We’ll touch on a few of those areas here as we look at
how loudness management plays out in the real world of television audio mixing.
One of things that surprised me when I first read through A/85 was
how much space the document gives to setting up the control room environment.
Rooms are divided into five categories, from high-end mix rooms in category
one, to category five, where spaces are so difficult to work in that the
engineer is advised to mix using headphones.
The document goes on to discuss proper loudspeaker location and
calibration, controlling acoustics issues and even loudspeaker selection.
This eventually leads to another recommendation that might be a new
concept to some, which is to calibrate the control-room monitor volume control
and leave it alone. This recommendation most likely originated in the film-mixing
world, where the control room has excellent acoustics, a very low noise floor
and where the monitor volume is calibrated to –85 dB SPL, and then practically
Of course, most television mix rooms are the chaotic antithesis of
this kind of environment, with intercoms blaring, equipment with fans nearby
and constant shifts of focus from the engineer as he prepares for the next
event coming his way.
The idea behind the A/85 control room recommendations is that the
mix engineer should have a trusted, stable, calibrated mix environment that
will allow him/her to rely on their ears rather than staring at a loudness
meter for the duration of the program.
The distractions and noise ever-present in television audio mix
rooms may cause control-room mix volumes to vary somewhat from A/85
recommendations, but the important idea here is that the audio engineer is able
to mix with their ears rather than focusing on the meter.
LOUDNESS AND TRUE PEAK
Although some loudness meters display additional values on screen,
there are two key measurements that the mixer must pay attention to: the
Loudness value (LKFS), and the True Peak (TP) value. Loudness is the audio
loudness of the mix, while true peak is the peak level of the mix. True Peak,
like LKFS, is another new measurement and uses four times oversampling to catch
peaks that older peak meters might miss.
All broadcasters have target loudness and maximum true peak values
that they expect all content to meet, so the mixer will need to know what those
values are and tailor the mix to them. If that information is unavailable, it’s
a safe bet to mix to –24 LKFS and –2 dBTP, which are the recommended A/85
values for content without metadata.
Actually mixing with a loudness meter can seem a bit odd at first
because most of them simply display a number on the screen that represents the
loudness value of the mix. Some meters have other types of displays, but the
loudness value is always a number.
There should also be a true peak reading somewhere on the display
so it can be monitored as well. As for what should be measured, according to
A/85, the anchor element should be isolated from the rest of the mix and
measured. For most programming, the anchor element will be speech. If the
anchor element can’t be isolated, then all mix channels, except for the LFE
channel, should be sent together into the meter and measured.
Anyone mixing commercials or shortform content that is under three
minutes is expected to measure the five primary surround channels, again minus
the LFE, and not worry about the anchor element. It’s extremely important to
mix to the actual target value rather than to either side of it. A/85 gives
mixers 2 dB of breathing room on either side of the target, but go beyond it in
either direction and the mix is suddenly out of compliance.
I mentioned earlier that the preferred method of loudness
management mixing is for the engineer to mix using their ears rather than
staring at the meter, but there’s another reason not to watch the loudness display
too closely. Loudness measurement is a measure that takes shape over time and
the instantaneous readout on the meter won’t necessarily help with the
long-term measure of the mix. Loudness meters typically measure in momentary
(400ms), short-term (3s) and long-term-integrated measurements.
For live and long-form content production, the long-term integrated
measurement should be used and the content should be measured from beginning to
end if possible. There are certain places in the program where the loudness of
the mix will be most scrutinized, so particular attention should be paid to
these spots. The beginning of the program, in and out of commercial breaks and
the end of the program are all places where the program will play against other
content. If the loudness levels of the program and the adjacent content don’t
match, it could generate complaints from listeners.
Despite having new measurements, meters and specifications to pay
attention to, audio engineers are not really called to make significant changes
to their workflow in order to properly manage loudness. They should mix as they
always have while adapting to a fixed monitor volume, letting their ears guide
them to proper loudness levels in their mixes and occasionally checking the
loudness meter for reference.
The payoff for the audio engineer is that it significantly
increases the likelihood that the mix they hear in the control room will be the
same mix the viewer hears at home.
Jay Yeary is a broadcast audio engineer and television consultant
based in Atlanta, Ga. His days are spent working alongside the engineering
teams of a large media company.