The digital meters in today’s DAWs, while arguably not as cool-looking as illuminated analog meters, are a vast improvement in terms of accuracy—although they’re still not perfect, as we’ll discuss later. And they also provide a lot more options, such as . . .
Being able to change meter resolution can be very handy. For example, when tracking, you might want to cover as wide a range as possible (e.g., 90dB so you can see any little glitches or noises). On the other hand, when mixing, you might decide to go for 12dB or thereabouts on the output bus. This lets you monitor the all-important peak values, and gauge the approximate amount of loudness maximization that may be required. For example, if the output meters make it to 0 frequently, but otherwise spend very little time at 12dB, then the song will probably need to be made hotter during the mastering process.
PLACEMENT AND SIZING
Some programs let you change the meter size, as well as choose whether the meter is horizontal or vertical. Generally, the meter will be longer and easier to read in a horizontal view—and offer higher resolution, as well—but it will take up more screen space. With a vertical view, you can try making the meter short enough that it serves more as an activity/clip indicator. This saves a lot of screen space, and, in many cases, tells you all you really need to know (Figure 1).
Note that meters do use up CPU power (albeit not much), so if you really need that last ounce of performance, disable any meters you don’t need. This is because they’re part of the GUI, and need to be updated frequently. With some programs, you can change the refresh rate so that the meters place less of a strain on the CPU when set to a slow refresh.
WHERE DO THEY GO IN THE SIGNAL PATH?
If you can choose whether playback meters are pre- or post-fader, and whether bus meters are pre-fader, post-fader, or pre-fader and post-FX, so much the better. With a mixer screen, I tend to choose post-fader metering, so it’s easy to see at a glance which tracks are contributing the most amount of level. But in a track view, as used when recording, a pre-fader setting lets you monitor track activity so you can check whether a track has signal, regardless of the fader position.
A typical choice of responses includes peak, RMS, or peak+RMS response. Ideally, you’d want to be able to set these characteristics independently for the record, playback, and bus meters.
Peak readings are most useful while recording, so that you can determine if any signal levels are loud enough to cause overload problems. Most meters can also hold any peak overload conditions, so even if your eyes are away from the meter when it happens, you can still know that 0 was exceeded.
RMS provides an average reading, and can be helpful while mixing if you want to see the average amount of energy a track contributes to a song. For example, a percussive sound will typically have a high peak value and a low average value, whereas distorted guitar will have a high average value that almost equals the peaks. In this kind of situation, unless you want to squash lots of dynamics, the peak vs. RMS indications give you a clue about setting the proper levels. You’ll want the guitar, with its high average value, lower than the percussive sound to retain a good balance.
My favorite option, though, is peak+RMS response. In the credit-where-credit-is-due department, the first time I saw this was with Peavey mixer LED meter hardware. The average level was lit solid, and the peak was a single LED, tracking peak levels and “floating” above the average reading. What makes this so useful is that you can get a sense of the relationship between the peak and RMS levels, making it easier to determine if a signal needs limiting, or whether it would be better served by compression.
USING METERS TO CHECK FOR LOOP DISTORTION
Meters aren’t just for measuring record and playback levels—they can provide valuable insights into the behavior of all audio signals.
For example, many programs these days can handle “Acidized” loops, which include metadata for stretching characteristics so that they can match different arbitrary tempos or keys.
However, with these programs, I’ve noticed a strange problem where even though the Acidized clips were normalized to something less than 0 (e.g., –0.1 dB), the track meters sometimes indicated overload. When I played back the samples normally, they clearly didn’t distort. But as I continued to monitor the meters, I would notice occasional flashes of distortion.
After putting in a lot of hours editing a loop library, I finally figured out what’s going on. If a file plays back without distortion, is converted into an Acidized loop, and plays back at its “native” tempo, it will not distort. However, speeding up or slowing down the song may cause slight volume increases that push the overall level past the point of overload. This is because the DSP that does the time stretching relies a lot on crossfading, so if phase-coherent audio sections overlap, the signal peaks could add up and clip.
This effect is not at all predictable. For example, a 120 BPM Acidized loop clip might distort at 80 and 140 BPM, but not at 160 BPM. So, when creating acidized files, my final step is now normalizing them to –3dB. During mixdown, watch the levels of any Acidized files to make sure they’re not distorting when the tempo changes.
IT’S PERFECT BECAUSE IT’S DIGITAL, RIGHT?
Well . . . not necessarily. As most people know, digital audio takes a series of “snapshots” (the sample rate), processes them, then smoothes these stairsteps at the output to produce a continuous, analog-type waveform. However, the meters almost always monitor the waveform before it has been smoothed, not after. As a result, it’s possible that during the process of reconstructing the output signal, the output level may momentarily exceed 0, even if the meter doesn’t show it. This is one reason why many engineers don’t use the full headroom available in a mix bus (hey, there’s usually at least 24 bits to play with anyway—you’re not losing much), and produce final mixes that are down a few dB from 0dB. The mastering engineer can always make sure that the final signal reaches full code.
AND ABOUT THOSE LITTLE NUMBERS . . .
A great feature of many digital meters is that they include a number that indicates the maximum (peak) value obtained during playback. Furthermore, the fader will be calibrated, as well. So for example, if the meter indicates the maximum signal during playback was –3.16dB, then you can move the fader up +3.15dB, and, at least in theory, the signal will never exceed 0.01dB. However, as noted above, you can’t always assume this figure is 100 percent accurate, so you’d more likely raise the fader by +2dB or so to leave a margin of error. Conversely, if the number indicates, +4.2dB, odds are you already heard it was distorting, and fixed it! But if not, you can bring the fader down to –5dB and feel a lot safer.
EQ MAGAZINE SAYS YOU SHOULD MIX WITH YOUR EYES!
No we don’t! Realistically, you should be able to do a pretty good mix without even seeing the meters, and many people (myself included) would advocate not caring too much about what the meters are showing—it’s the sound that matters.
However, meters are like test equipment—they’ll point out problems much faster than you might find them on your own. To illustrate, suppose you have a dense mix with 40 tracks, and you swear you hear some distortion in the chorus. The track meters will show you where the track is clipping (or, at the very least, the bus meters will show the part of the song where clipping occurred), and you can find out which track is causing the problem by muting tracks one at a time, observing the meter number that shows the peak value, and then turning down whichever track pushes the level over the edge. The bottom line is the more you know about metering, the more you’ll know about what’s happening with your audio.
The Ears Have It
Craig Anderton warns against obsessively locking your eyes to those meter levels, and it seems no less a studio legend than ‘50s and ‘60s Columbia Records A&R honcho Mitch Miller (yeah, the Sing Along With Mitch cat) would strongly agree. In an apocryphal tale about a long-forgotten ‘60s session, Miller is said to have been listening to a mix—eyes closed, arms crossed, lost in his evaluative trance.
“Turn up the horn section,” said Miller to the engineer.
“I can’t do that,” Mr. Miller,” answered the console jockey. “The levels are already pegging the meters. See?”
Miller opened his eyes, surveyed the board, and walked out of the control room without saying a word. The next day, every single meter in the studio was hidden under several layers of gaffer’s tape. The record man had made his point. His ears were to be obeyed.