PreSonus Blog

Monthly Archives: September 2018


Friday Tips: What’s a Phase Meter—And Why Should I Care?

I’ve always appreciated Studio One’s analytics—the spectrum analyzer, the dynamic range meter in older versions and the more modern LUFS metering in Studio One 4, the K-Scale meters based on Bob Katz’s research, the strobe tuner, and the ability to stretch the faders in the Mix view when you want to couple high resolution with long fader travel. But I wonder if the Phase Meter and its companion Correlation Meter get the props they deserve, so let’s look at what this combo can do for you.

Phase Meters—Not Just for Mixdowns!

Most people consider a tool like the Phase Meter as being only for checking final mixes. However, one very useful technique is putting it in the master output bus, and soloing one track at a time (remember, you can Alt+click on a track’s Solo button for an “exclusive solo” function). This gives some insights into the phase, level, and stereo spread of individual tracks in a way that’s more revealing than just looking over panpots.

Correlation Meter Basics

In brief, the Correlation meter (the bar graph at the Phase meter’s bottom) indicates a stereo signal’s mono compatibility. This was of crucial importance when mastering for vinyl, because it could indicate if there were out-of-phase audio components in the audio that could possibly cause the stylus to jump out of its groove. These days, it’s largely a stereo world but it’s still important to check for mono compatibility—after all, when listening to speakers, you don’t have perfect stereo separation. You’ll usually monitor correlation in the master bus, but for individual tracks, it can indicate whether (for example) a signal processor is throwing a track’s left and right channels out of phase.

The Correlation meter reading spans the range between -1 (the right and left channels are completely out of phase, with no correlation) and +1 (the right and left channels are identical, and correlate completely). With most mixes, the bar graph will fluctuate between 0 and +1.

Mono Readings

If the Phase meter displays a single vertical line, then the left and right channels are identical, and the track is mono. The Correlation bar graph meter at the bottom confirms this with its reading of 1.00, which means the left and right channels correlate completely—in other words, they aren’t just similar, but identical.

Left and Right Readings

If there’s a single, diagonal line on the L axis, that means that all the signal’s energy is concentrated in the left channel. Similarly if there’s a single, diagonal line on the R axis, then all the signal’s energy is concentrated in the right channel. If you pan a track where the left and right channels are identical (as shown by the Correlation meter displaying 1.00), then the line will move from one channel to the other.

Stereo Signals

With stereo, you’ll see an excellent visual representation of how much the signal extends into the stereo field. The vertical size indicates the level. As you pan the signal left or right, the stereo field will become narrower around the line that moves from left to right until at one extreme or the other, you’ll see only a diagonal line on the L or R axis.

Note the correlation meter is showing +0.47. This means that there’s about an equal amount of similarity between the left and right channels as there are differences, but nothing is out of phase.

Mid-Side Encoded Audio

With Mid-Side encoded audio, you’ll see amplitude around the L and R axes, as well as along the M axis. Because the L signal is the center and the R signal the sides, you’ll see a lot more level along the L axis. Also, note the Correlation meter setting of 0.00—this means that there’s no similarity between the right and left channels, which is what you’d expect with a Mid-Side encoded signal.

Binaural Pan Signal

Studio One’s Binaural Pan processor widens the stereo image so that there’s much more energy in the right and left sides than in the center; this image shows what happens when you set the widening to maximum. Compare this to the reading for stereo signals—you can see that in this case, the energy extends further out to the right and left. Furthermore, the Correlation meter shows that there are no significant similarities between the right and left channels, which is a result of the Binaural Pan processor being based on Mid-Side processing.

Phase Issues

Here, the Correlation meter shows a negative number, which means there are out-of-phase elements within the stereo mix. Occasional negative blips aren’t a problem, but if the Correlation meter spends a substantial amount of time to the left of 0, then there’s a phase issue that will interfere with mono compatibility.

 

Friday Tips: Create the “Barberpole” Audio Illusion

The Shepard Tone (aka Barberpole) is an audio illusion where a tone always seems to keep rising (or falling). You may have heard it before—to build tension in music by Swedish House Mafia, Beatsystem, Data Life, and Franz Ferdinand, as the sound effect for the endless staircase in Super Mario 64, for the sound of constant acceleration for the Batpod in The Dark Knight and The Dark Knight Rises, at the end of Pink Floyd’s “Echoes” from the Meddle album, or in the soundtrack for the film Dunkirk in sections where the goal was to produce a vibe of increasing intensity. Check out the audio example, and you’ll hear how the tone just goes on forever.

 

Thanks to Studio One’s Tone Generator, it’s easy to produce a Shepard Tone loop—just follow the step-by-step instructions, in a song with the tempo set to 120 BPM.

 

  1. Insert the Tone Generator in the Input L+R insert section so that you can record its tone, and edit the Tone Generator generator’s settings as shown in the screen shot. The goal is the longest possible sine wave sweep from 20 Hz to 22 kHz.

 

  1. Start recording, then click the Tone Generator’s On button. After recording the file, trim the beginning and end respectively to just before and just after you can hear a tone, and add a short fade in and fade out.

  1. As shown above, copy the track, and offset each track’s beginning by two measures compared to the track above it. Keep copying and offsetting until the start of the last track is at the same measure as the end of the first track.

  1. Now select all, and drag the entire group to the right so that there’s a bit of an overlap between the two groups. Select everything, then type X to turn the overlap into a crossfade with a linear curve. Next, create a loop that extends from the start of the lowest track to the end of the highest track.

 

  1. Choose Song > Export Mixdown, set the Export Range to Between Loop, and under Options, check Import to Track and Close After Export. Solo the mixed track, and play it—you’ll hear a continuously rising tone. Now we need to turn it into a loop.

 

  1. Follow the instructions in the July 27 Friday Tip of the Week on how to create pads that loop perfectly. The above screen shot shows the basic concept; Track 1 shows the first steps. Copy the clip, move the copy to the right so it overlaps the last four measures of the original clip, and then crossfade the overlap with a linear crossfade.

 

  1. Track 2 in the screen shot shows the next step. Bounce the two clips together, then split at the end of measure 4 to remove the first four measures, and at the end of the crossfade to remove everything after the crossfade. Loop the section that remains, and you have your never-ending upward Shepard Tone, as a glitchless loop. Note that when you bring it into a project, don’t stretch it to conform to tempo—there is no tempo. And if you want it to go on forever…just keep typing D!

 

Download the loop here!

 

Bonus Tips:

 

  • I recommend adding a Pro EQ—reduce the high frequencies somewhat with the HC (High Cut) filter, and boost the low frequencies with a bit of a shelf, to increase the illusion’s effectiveness.
  • The “classic” Shepard Tone requires that the tones be one octave apart. However, offsetting them by 2 measures at 120 BPM seems close enough.

 

 

The Virtual “How Does It Sound in a Car?” Tester

This tip is for those who won’t sign off on a mix until they’ve heard it in a car. There may be a scientific reason why this is beneficial: Noise tends to mask sounds, so if one instrument you want to hear gets lost in the noise and another jumps out, try a mix that raises and lowers those levels, respectively.

The ear doesn’t discriminate level differences as accurately as pitch differences, so without noise masking a sound, the level may seem okay. But as soon as you mix in noise, an important sound may disappear. If you increase the level just a bit so you can hear it, when you remove the noise there’s a very good chance you’ll like the new level setting better. Think of this as doing something similar to compression, but without applying any actual dynamics. You’re just making sure that the levels needing parity, have parity.

Of course this doesn’t mean you want everything jumping out of the noise—those tambourine and shaker parts are probably just fine as they are. The main sounds to listen to here are vocals, leads, drums, and bass, as well as their relationship to each other.

This also doesn’t mean you should mix consistently with noise, as it will bias your hearing (and besides, it’s truly annoying). I add noise in with a mix as a last diagnostic step. If the mix has sounded fine up until then and passes this final test, I consider it ready to master. And I don’t need to go driving anywhere, either.

Setup

Just follow the steps, and you’ll be good to go.

 

  1. Create a stereo audio track, and insert the Tone Generator effect. Turn the track’s fader all the way down.
  2. Choose Pink Noise as the waveform.
  3. Click On to start generating noise.
  4. Turn up the track’s fader to add noise to the mix.

One very cool aspect of the Tone Generator’s noise is that it’s true stereo where the left and right channels don’t correlate, so you don’t get any center channel buildup (as would happen with a mono noise signal).

As to how much noise to add, it’s kind of like maximizing. Set it 6 dB below the mix’s peaks, and you’ll hear what occupies the upper 6 dB of dynamic range. Set it 12 dB below the mix’s peaks, and you’ll hear what’s in the upper 12 dB of dynamic range. This isn’t an exact spec per se, but it provides a rough standard of comparison.

As crazy as this idea sounds, try it sometime and tweak your mix. Then turn off the noise, take a short break so your ears get acclimated back to normal hearing, and then check the mix again. I won’t be surprised if you hear an improvement!

 

Songwriting with Impact XT

This tip is for those of you who didn’t see my Studio One workshop at Sweetwater GearFest 2018, were turned away because of that pesky fire marshal’s rules about crowds, or who didn’t realize Studio One 4 has some pretty advanced looping capabilities—as well as the ability to trigger pitch transpositions for loops.

With Impact XT, you can load loops on pads, and then trigger them (on and off) in real time via MIDI notes. Assign each output from an Impact XT pad to a track input (in the screen shot, Track 5 is recording the output of Impact XT M4), set all the tracks to record, and you can record the results of your improvisations.

The following screen shot shows the results of recording the first part of a potential song. Note how some tracks have sounds that extend the length of the recording, while other tracks had their sounds brought in at specific times by triggering an Impact XT pad.

This by itself is pretty cool, because you can weave loops in and out to create an arrangement. The song goes longer than this, but the above shows what you’re hearing in the following audio example. Granted, it’s not much of a song—it just kinda drones on and on. But keep reading…this is just the start.

The process becomes far more interesting when you bring the chord track into play, because you can transpose the loops to create a chord progression that becomes the basis for a song. All the tracks, even the drums, were set to follow the chord track. Listen to how although some of the original loops added a fourth to the tonic, when this was synched to the chord track, all of the loops followed a tonic-to-fourth chord progression. In other words, it wasn’t just one loop adding a fourth, but the entire song transposing to the fourth. We also gained an intro; here’s the chord progression that was used.

And here’s what the chord progression sounded like after harmonic editing. The major difference is in the intro, and transposing to D to kick off the second half of each verse.

Working this way can be very inspirational because you can create a basic arrangement with loops, and then use the Chord Track to create a chord progression. Although PreSonus is careful to point out that Harmonic Editing is more for “prototyping” songs and they expect that you’ll want to replace the “scratch” parts, I’ve found that many times the scratch parts end up being keepers—and I gotta say, I love what happens when you tell drums to follow the chord track!