One of the complaints about electronic music instruments and controllers is that they lack the expressiveness of acoustic instruments. Although future instruments will take advantage of MIDI 2.0’s enhanced expressiveness, two options are available right now: polyphonic pressure, and MPE (MIDI Polyphonic Expression). Studio One 5 can record/edit both, and ATOM SQ generates polyphonic pressure…so let’s dig deeper.
First, there’s some confusion because people call the same function by different names. Channel Aftertouch = Channel Pressure = Mono Aftertouch = Mono Pressure. Polyphonic Aftertouch = Polyphonic Pressure = Poly AT = Poly Aftertouch = Poly Pressure. Okay! Now we’ve cleared that up.
Aftertouch generates a control signal when you press down on a keyboard key after it’s down, or continue pressing on a percussion pad after striking it. Aftertouch is a variable message, like a mod wheel or footpedal—not a switch. A typical application would be changing filter cutoff, adding modulation, or doing guitar-like pitch bends by pressing on a key.
There are two aftertouch flavors. Mono pressure has been around since the days of the Yamaha DX7, and sends the highest controller value of all keys that are currently being pressed. Polyphonic pressure sends individual pressure messages for each key. For example, when holding down a chord for a brass section, by assigning poly pressure to filter cutoff, you can make just one note brighter by pressing down on its associated key. The other chord notes remain unaffected unless they’re also pressed.
Controllers with polyphonic aftertouch used to be fairly expensive and rare, but that’s changing—as evidenced by ATOM SQ.
As expected, you need a synth that responds to poly pressure. Many hardware synths respond to it, even if they don’t generate it. As to soft synths, although I haven’t tested all of the following, they reportedly support poly pressure: several Korg Collection synths, Kontakt, Reaktor, all Arturia instruments, all U-He instruments, XILS-Lab synths, TAL-Sampler, AAS synths, Albino 3, impOSCar2, Mach5, and Omnisphere. If you know of others, feel free to mention them in the comments section below. (Currently, Studio One’s bundled instruments don’t respond to polyphonic aftertouch.)
Figure 1: ATOM SQ being set up to generate Poly Pressure messages.
With ATOM SQ, press the Setup button. Hit the lower-left “pressure” button below the display, then spin the dial to choose Poly (Fig. 1). Note that if ATOM SQ outputs poly pressure, most instruments that respond only to channel (mono) aftertouch will ignore these messages.
Record poly pressure in Studio One 5 as you would any MIDI controller. To edit pressure messages, use the Edit window’s Note Controller tab. Select Pressure for the Type, and then the Pitch of the note you want to edit. Or, click on a note to select its corresponding note Pitch automatically. You can then edit that note’s poly pressure controller as you would any other controller (Fig. 2).
Figure 2: The selected Note’s data is white; unselected notes of the same pitch are blue. The gray lines in the background show the poly pressure controller messages for notes with other pitches.
It may seem that editing data for individual notes would be tedious, and it can be. However, because poly pressure allows for more expressive real-time playing, you might not feel the need to do as much editing anyway—you won’t need to use editing to add expressiveness that you couldn’t add while playing.
A fine point is that it’s currently not possible to copy Note Controller data from one note, then paste it to a note of a different pitch (probably because the whole point of poly AT is for different notes to have different controller data). However, if you copy the note itself to a different pitch, the Note Controller data will go along with it.
Although ATOM SQ can adopt a layout that resembles a keyboard, it would be a mistake to see it as a stripped-down version of a standard keyboard. Controllers with polyphonic pressure tend to think outside the usual keyboard box, by incorporating pads or other transducers that are designed for predictable pressure sensitivity. Poly pressure has been around for a while, but a new generation of MIDI controllers (like ATOM SQ) are making the technology—and the resulting expressiveness—far more accessible for those who want to wring more soul out of their synths.
It’s not surprising a lot of Studio One users also have Ableton Live, because they’re quite different. I’ve always felt Studio One is a pro recording studio (with a helluva backline) disguised as software, while Ableton is a live performance instrument disguised as software.
Fortunately, if you like working simultaneously with Live’s loops and scenes and Studio One’s rich feature set, Studio One can host Live as a ReWire client. Even better, ATOM SQ can provide full native integration with Ableton Live when it’s ReWired as a client—once you know how to set up the MIDI ins and outs for both programs.
Now ATOM SQ will act as an integrated controller with Ableton Live while it’s ReWired into Studio One. Cool, eh?
To return control to Studio One, reverse the process—in Live, set Control Surface to None, and toggle the MIDI Ports that relate to ATOM SQ from On to Off. In Studio One’s Options > External Devices, For ATOM SQ, reconnect ATOM SQ to Receive From and Send To.
Note that with ATOM SQ controlling Studio One, the Transport function still controls both Live and Studio One. Also, if Live has the focus, any QWERTY keyboard assignments for triggering Clips and Scenes remain valid. So even while using ATOM SQ in the native mode for Studio One, you can still trigger different Clip and Scenes in Live. If you switch the focus back to Studio One, then any QWERTY keyboard shortcuts will trigger their assigned Studio One shortcuts.
Note: When switching back and forth between Live and Studio One, and enabling/disabling Studio One and Ableton Live modes for ATOM SQ, to return to Live you may need to “refresh” Live’s Preferences settings. Choose None for the Control Surface and then re-select ATOM SQ. Next, turn the various MIDI Port options off and on again.
Vocoders are processors that use the audio from vocals (called the modulation source, or modulator) to modulate another sound, like a synthesizer pad (called the carrier). However, no law says you have to use vocals as a modulator, and I often use drums to modulate pads, power chords, and more. While Studio One’s toolset doesn’t have enough resolution for super-intelligible vocoding with voice, it’s perfect for drumcoding, which actually benefits from the lower resolution.
This tip is for advanced users and requires a fairly complex setup. Rather than go into too much detail about how it works, simply download the Drumcoder.song file, linked below, which has a complete drumcoding setup. Load Drumcoder.song into Studio One 5, press play, and you’ll hear what drumcoding is all about. (Note that the file format isn’t compatible with previous Studio One versions. However, based on the description in this tip, you should be able to “roll your own” drumcoding setup in previous Studio One versions.)
Let’s check out an audio demo. The first half has the drumcoded sound only, while the second half mixes in the drum (modulator) sound.
But wait—there’s more! Although the drumcoder isn’t designed to be the greatest vocoder in the world (and it isn’t), you can still get some decent results. Here, the voice is saying “Even do some kinds of vocal effects with the PreSonus drumcoder—have fun!’
Next, we’ll explore how it works…or if you’re impatient, just reverse-engineer the song.
Vocoding splits the modulator (like voice or drums) into multiple frequency bands. In a traditional vocoder, each band produces a control voltage that corresponds to the audio’s level in each band. Similarly, the carrier splits into the same frequency bands. A VCA follows each carrier band, and the VCAs are fed by the modulator’s control voltages. So, if there’s midrange energy in the modulator, it opens the VCA for the carrier’s midrange audio. If there’s bass energy in the modulator, it opens the VCA for the carrier’s bass audio. With a vocoder, as different energy occurs in different bands that cover a vocal’s frequency range, the carrier mimics that same distribution of energy in its own bands. This is what generates talking instrument effects.
Vocoders typically need at least eight frequency bands to make voices sound intelligible. Studio One’s Splitter can divide incoming audio into five bands, which is enough resolution for drumcoding. Fig. 1 (which takes some graphic liberties with Studio One’s UI), shows the signal flow.
Figure 1: Drumcoder signal flow.
The Drums track provides the modulator signal, and the Mai Tai synthesizer provides the carrier. The Drums track has five pre-fader sends to distribute the drum sound to five buses. As shown in Fig. 2, each of the five buses has a Splitter (but no other effects) set to Frequency Split mode, with splits at 200, 400, 800, and 1600 Hz. The < 200 Hz bus mutes all Splits except for 1, the 200-400 bus mutes all Splits except for 2, the 400-800 bus mutes all splits except for 3, the 800 – 1.6 kHz mutes all splits except for 4, and the > 1.6 kHz bus mutes all splits except for 5. Now each bus output covers one of the five bands.
Figure 2: Splitter settings for the five buses.
The Mai Tai carrier has a splitter set to the same frequencies. Each split goes to an Expander, which basically acts like a VCA; see Fig. 3. We don’t need to break out the Splitter outputs, because you can access the sidechain for the Expanders located within the Splitter. (A Mixtool follows each Expander, but it’s there solely to provide a volume control for each of the carrier’s bands in the control panel.)
Figure 3: Effects used for the Mai Tai synthesizer carrier track.
As to the bus outputs, the < 200 Hz bus has a send that goes to the sidechain of the Expander in the carrier’s < 200 Hz split. The 200-400 Hz bus has a send to the sidechain of the Expander in the carrier’s 200-400 Hz split. The 400-800 Hz bus has a send to the sidechain of the Expander in the carrier’s 400-800 Hz split…you get the idea. Basically, each bus provides a “control voltage” for the corresponding “VCA” (Expander) that controls the level of the carrier’s five bands.
Fig. 4 shows the Control panel.
Figure 4: Drumcoder macro controls.
Threshold, Ratio, and Range cover the full range of Expander controls. They affect how tightly the Expander follows the modulator, which controls the effect’s percussive nature. Just play around with them until you get the sound you want. The Expander Envelope settings aren’t particularly crucial, but I find 0.10 ms Attack and 128.0 ms Release work well. Of course, you also need to enable the sidechain for each Expander, and make sure it’s listening to the bus that corresponds to the correct band.
The five knobs toward the right control the level of the individual bands by altering the Gain of the Mixtool that follows each band’s Expander. The five associated buttons enable or bypass the Expander for a particular band, which can give some really cool effects. For example, turn off the Expander on the Mid band, and with the Song’s Mai Tai preset, it almost sounds like a choir is singing along with the drumcoded drums.
Although the Drumcoder isn’t really designed for vocal effects, it still can be fun. The key is to bring up the > 1.6 kHz Bus slider, as this mixes in some of the voice’s “s” sounds, which give intelligibility. Experiment with the Expander controls to find what works well. If you really want to dig into vocal applications, edit the Splitter frequencies to optimize them for the vocal range instead of drums…or leave a comment asking me to pursue this further.
Due to the complexity, if I think I’m going to use the Drumcoder, I’ll just treat this song like a template and build the rest of the song from there. But once you understand the principle of operation, you can always add the effect in to an existing song as needed. I have to say this is one of my favorite Friday tips ever… I hope you enjoy playing with the Drumcoder!
Although Studio One 5 doesn’t have a tape emulator plug-in per se, it can emulate some of the most important characteristics that people associate with “the tape sound.” Truly emulating tape can go down a serious rabbit hole because tape is a complicated signal processor; no two vintage tape recorders sounded the same because they required alignment (influenced by the engineer’s preferences), used different tape formulations, and were in various states of maintenance. However, emulating three important characteristics provides what most people want from tape emulation.
Check out the audio example to hear what this FX Chain can do. The first part is unprocessed, while the second part uses the default FX Chain control settings with a little underbiasing and head bump. The difference is subtle, but it adds that extra “something.”
This FX Chain starts with a Splitter, which creates three signal paths: one for saturation, one for hiss, and one for hum (Fig. 1).
Figure 1: FX Chain block diagram.
After auditioning all available Studio One 5 saturation options, I liked the TriComp best for this application. The Pro EQ stage preceding the TriComp provides the head bump EQ and has a control to emulate the effect of underbiasing tape (more highs, which pushes more high-frequency level into the TriComp and therefore increases distortion in that range) or overbiasing (less highs, less distortion).
At first, I wasn’t going to include tape hiss and hum, but if someone needs to use this FX Chain for sound design (i.e., an actor starts a tape in a theatrical production), then including hiss and hum sounds more authentic. An additional knob chooses 50 or 60 Hz hum, which represents the power standards in different countries. (Note that the closest you can get to these frequencies is 50.4 and 59.1 Hz, but that’s good enough). However, I draw the line at including wow and flutter! Good riddance to both of them.
Because creating three splits reduces each split’s level, the TriComp Gain control provides makeup gain.
Turning Bump on adds a boost at the specified frequency, but also adds a 48 dB low-cut filter around 23 Hz to emulate the loss of very low frequencies due to the head bump. As a result, depending on the program material, adding the bump may increase or decrease the total apparent bass response. For additional flexibility, if you turn Bump Amount down all the way, the Bump On/Off switch enables or disables only the 48 dB/octave low-cut filter.
Fig. 2 shows some typical spectra from using the FX Chain.
Figure 2: The top curve shows the head bump enabled, with underbiasing. The lower curve shows minimal added bump, but with the ultra-low cut filter enabled, and overbiasing.
The controls default to rational settings (Fig. 3), which are used in the audio example. But as usual with my FX chains, the settings can go beyond the realm of good taste if needed.
Figure 3: Control panel for the Tape Emulator.
For example, I rarely go over 2-3% saturation, but I know some of you are itching to kick it up to 10%. Ditto tape hiss, in case you want to emulate recording on an ancient Radio Shack cassette recorder—with Radio Shack tape. Just remember that the Bias control is clockwise to overbias (less highs), and counter-clockwise to underbias (more highs).
There’s a lot of mythology around tape emulations, and you can find some very good plug-ins that nail the sound of tape. But try this FX Chain—it may give you exactly what you want. Best of all, I promise you’ll never have to clean or demagnetize its tape heads.
With physical audio media in its twilight, streaming has become the primary way to distribute music. A wonderful side effect has been the end of the loudness wars, because streaming services like Spotify turn levels up or down as needed to attain a specific, consistent perceived level—squashing a master won’t make it sound any louder.
However, the “garbage in, garbage out” law remains in effect, so you need to submit music that meets a streaming service’s specs. For example, Spotify prefers files with an LUFS of -14.0 (according to the EBU R128 standard), and a True Peak reading of -1.0 or lower. This avoids adding distortion when transcoding to lossy formats. If the LUFS reading is above -14.0, then Spotify wants a True Peak value under -2.0.
Fortunately, when you Detect Loudness for a track on the mastering page, you’ll see a readout of the LUFS and LRA (a measure of overall dynamic range), as well as the True Peak, RMS (average signal level), and DC offset for the left and right channels. Fig. 1 shows an example of the specs generated by detecting loudness.
Note that this hits Spotify’s desired LUFS, but the left channel’s True Peak value is higher than what’s ideal. This readout also shows that the average RMS levels for each channel are somewhat different—the left channel is 1.2 dB louder than the right one, which also accounts for the higher True Peak value. This may be the way the artist wants the mix to sound, but it could also indicate a potential problem with the mix, where the overall sound isn’t properly centered.
A simple fix is to insert a Dual Pan into the Inserts section. Use the Input Balance control to “weight” the stereo image more to one side for a better balance. After doing so and readjusting the LUFS, we can now give Spotify exactly what it wants (Fig. 2). Also note that the left and right channels are perfectly balanced.
A Crucial Consideration!
You don’t want to mix or master based on numbers, but on what you hear. If you set up Dual Pan to balance the channels, make sure that you enable/bypass the plug-in and compare the two options. You might find that balancing the left and right channels not only accommodates Spotify’s requirements, but improves the mix’s overall balance. If it doesn’t, then leave the balance alone, and lower the track’s overall output level so that True Peak is under -1.0 for both channels (or under -2.0 for LUFS values above ‑14.0). This will likely lower the LUFS reading, but don’t worry about it: Spotify will turn up the track anyway to reach -14.0 LUFS.
Coda: I always thought that squashing dynamic range to try and win the loudness wars made listening to music a less pleasant experience, and that’s one of the reasons CD sales kept declining. Does the end of the loudness wars correspond to the current music industry rebound from streaming? I don’t know… but it wouldn’t surprise me.
My mastering specialty is salvage jobs, which has become easier to do with Studio One. But this gig was something else.
Martha Davis’s last solo album (I Have My Standards, whose mastering challenges were covered in this blog post) has done really well. Since the pandemic has sidelined her from touring as Martha Davis and the Motels or going into the studio, she’s releasing a new song every month online. These involve excellent, but unreleased, material.
That’s THE good news. The bad news is that her latest song choice, “In the Meantime,” had the drum machine kick mixed so loud the song should have been credited as “Solo Kick Drum with Vocal Accompaniment.” With a vocalist like Martha (listen to any of her many hits from the 80s), that’s a crime. She was hoping I could fix it.
Don’t tune out, EDM/hip-hop fans. What about those TR-808 “toms” that are always mixed way too high? When I was given a Boy George song to remix, those toms were like sonic kryptonite before I figured out how to deal with them. And let’s not get into those clichéd 808 claps, okay? But we have a solution.
I tried everything to deal with the kick, including EQ, iZotope RX7 spectral reduction, mid-side processing using the Mixtool, and more. The mix was mostly mono, and the kick was full-frequency—from low-frequency boom to a nasty click that was louder than the lead vocal. Multiband dynamics didn’t work because the kick covered too wide a frequency range.
In desperation, I thought maybe I could find an isolated kick sound, throw it out of phase, and cancel the kick wherever it appeared in the song. Very fortunately, the song intro had a kick sound that could be isolated as an individual sample. So instead of going directly to Studio One’s mastering page, I went into the Song page, imported the stereo mix into one track, created a second track for only the kick, and dragged the copied kick to match up with every kick instance in the song (yes, this did take some time…). It wasn’t difficult to line up the copied kicks with sample- (or at least near-sample) accuracy (Fig. 1).
Figure 1: The top track is from the original song, while the lower track is an isolated kick. After lining the sounds up with respect to timing, flipping the kick track phase removed the kick sound from the mixed tracks.
The payoff was inserting Mixtool in the kicks-only track and flipping its phase 180 degrees. It canceled the kick! Wow—this physics stuff actually works.
But now there was no kick. So, I added the Waves LinEQ Broadband linear-phase equalizer (a non-linear-phase EQ can’t work in this context) in the kick drum track. This filtered out some of the kick drum’s lower frequencies so there was less cancellation while leaving the highs intact so they would still cancel as much as possible. Adjusting the shelving frequency and attenuation let in just enough of the original kick, without overwhelming the track. Even better, because the kick level was lower, I could bring up the low end to resurrect the bass part that had been overshadowed by the kick.
The mix traveled to the mastering page for a little more processing (Studio One’s Pro EQ and Binaural Pan, IK Multimedia’s Stealth maximizer, and Studio One’s metering). After hitting the desired readings of -13.0 LUFS with -0.2 True Peak readings, the mastering was done. Sure, I would much rather have had the individual tracks to do a remix, but it was what it was—a 28-year-old two-track mix.
To hear how this ended up, the audio example first plays an excerpt from the mastered version. Then there’s a brief pause, followed by the same section with the original file. I’m sure you’ll hear the difference in the kick drum.
Listen to an audio example from In the Meantime here:
Although it’s always better to fix issues at the source, here’s a tip to help repair recorded vocals during the mixing phase. The technique (which is featured in the new book How to Record and Mix Great Vocals in Studio One – 2nd Edition) combines multiband dynamics processing with equalization to both de-ess and reduce plosives. Although the screen shot shows the Multiband Dynamics processor in Studio One 5, this technique will work with previous Studio One versions if you duplicate the settings.
In the screen shot, the Multiband Dynamics’ Low band settings are outlined in red, and the High band settings are outlined in blue. (Note this is not the actual interface; the high band panel is pasted into the image from a different screen shot so you can see both the Low and High band settings simultaneously.)
The High band acts as a de-esser, because it applies compression to only the high frequencies. This helps tame sibilance. The Low band compresses only the low frequencies, which reduces pops. However, this preset also takes advantage of the way Multiband Dynamics combines equalization with dynamics control. Turning down the Low stage Gain all the way further reduces the low frequencies, where pops like to hang out and cause trouble.
For the High band, vary compression to taste. The compression settings are less critical for the Low band if you turn the Gain down all the way, but in either case, you’ll need to tweak the settings for your particular vocal track.
And that’s all there is to it. When a loud pop or sibilant sound hits the Multiband Dynamics, it’s compressed to be less annoying, while leaving the rest of the vocal intact. Vocal repaired!
Like being able to change what happens when one Event overlaps (covers over) a different Event.
Prior to Version 5, overlapped Events were treated the same. The overlapping Event became translucent, so you could see the waveform or note data of the Event underneath it. This is ideal for making audio crossfades, which is one of the main reasons for overlapping audio Events. To create a crossfade, type X, and optionally, click and drag up/down at the crossfade junction to shape the crossfade curve. Then you can shift+click on the overlapped Event, type ctrl+B, and combine them into a single Event. With note data, overlapping Events is helpful when combining, for example, the main snare hits on one track with alternate snare hits on a different track.
Another option after overlapping Events is mixing them together. Shift+click on the overlapped Event to include it with the overlapping Event on top. Then type Ctrl+B to mix audio, or G to merGe note data.
However, if you don’t crossfade or mix, then the region below the overlapping Event is still there. The overlapping Event is grayed, which can get confusing if you have a lot of muted sections; and if you remove the overlapping Event because you want to replace it with something else, it’s not obvious where the overlap occurred.
Some programs default to deleting, not just covering over, a section that’s being overlapped by another clip. This is useful when you’re doing lots of edits, because you’re not left with vestigial pieces of regions that still exist, but don’t do anything. To accommodate this type of workflow, Studio One 5 now offers a “no overlap” mode for Events. There are three ways to access this (Fig. 1).
Figure 1: In addition to using a keyboard shortcut, Studio One can default to “No overlap when editing events,” as chosen in the Arranger view or under Options.
Selecting “No overlap when editing events” deletes the overlapped part of an Event, and the replaced section looks like it’s part of the track (i.e., not grayed). However, if you later decide you didn’t really want to delete the overlapped region, then just remove the section that overlapped it. Now you can slip-edit the edge of the underlying Event back to where it was.
(Note that if you enabled Play Overlaps in a track’s Inspector, or chose “Enable Play Overdubs for New Audio Tracks” in Options/Advanced/Audio, so that you could overdub over an existing track and hear both the original track and the overdub on playback, enabling “No overlap when editing events” overrides this setting.)
Granted, this may seem like a small change, but it accommodates more workflow possibilities—especially if you learn the keyboard shortcut, and choose the right option at the right time.
Full disclosure: I’m not a big fan of chorusing. In general, I think it’s best relegated to wherever snares with gated reverbs, orchestral hits, DX7 bass presets, Fairlight pan pipes, and other 80s artifacts go to reminisce about the good old days.
But sometimes it’s great to be wrong, and multiband chorusing has changed my mind. This FX Chain (which works in Studio One Version 4 as well as Version 5) takes advantage of the Splitter, three Chorus plug-ins, Binaural panning, and a bit of limiting to produce a chorus effect that covers the range from subtle and shimmering, to rich and creamy.
There’s a downloadable .multipreset file, so feel free to download it, click on this window’s close button, bring the FX Chain into Studio One, and start playing. (Just remember to set the channel mode for guitar tracks to stereo, even with a mono guitar track.) However, it’s best to read the following on what the controls do, so you can take full advantage of the Multiband Chorus’s talents.
The Splitter creates three splits based on frequency, which in this case, are optimized for guitar with humbucking pickups. These frequencies work fine with other instruments, but tweak as needed. The first band covers up to 700 Hz, the second from 700 Hz to 1.36 kHz, and the third band, from 1.36 kHz on up (Fig. 1).
Figure 1. FX Chain block diagram and Macro Controls panel for the Multiband Chorus.
Each split goes to a Chorus. The mixed output from the three splits goes to a Binaural Pan to enhance the stereo imaging, and a Limiter to make the signal “pop” a little more.
Regarding the control panel, the Delay, Depth, LFO Width, and 1/2 Voices controls affect all three Choruses. Each Chorus also has its own on/off switch (C1, C2, and C3), Chorus/Double button (turning on the button enables the Double mode), and LFO Speed control. You’ll also find on/off buttons for the Binaural Pan and Limiter, as well as a Width control for the Binaural Pan. Fig. 2 shows the initial Chorus settings when you call up the FX Chain.
Figure 2. Initial FX Chain Chorus settings.
Because chorusing occurs in different frequency bands, the sound is more even and has a lusher sound than conventional chorusing. Furthermore, setting asynchronous LFO Speeds for the three bands can give a more randomized effect (at least until there’s an option for smoothed, randomized waveform shapes in Studio One).
A major multiband advantage comes into play when you set one of the bands to Doubler mode instead of Chorus. You may need to readjust the Delay and Width controls, but using Doubler mode in the mid- or high-frequency band, and chorusing for the other bands, gives a unique sound you won’t find anywhere else. Give it a try, and you’ll hear why it’s worth resurrecting the chorus effect—but with a multiband twist.
At first, the changes to the effects in Version 5 seem mostly cosmetic. But dig deeper, and you’ll find there’s more to the story—so let’s find out what’s new with Limiter2 (Fig. 1).
Figure 1: Limiter2 has had several design changes for Version 5.
The control parameters are more logical, and easier to adjust. Prior to V5, there was an unusual, by-design interaction with the Ceiling and Threshold controls; setting Ceiling below Threshold gave the limiter a softer knee. However, the tradeoff was difficulty in obtaining predictable results. Besides, if the soft knee aspect is important to you for dynamics control, just use the Compressor with a really high ratio.
In Limiter2, the Threshold is relative to the Ceiling—the Ceiling sets Limiter2’s absolute maximum level, while Threshold sets where limiting begins below the Ceiling, based on the Threshold parameter value. For example, if Ceiling is 0.00 and Threshold is -6.00, then the limiter’s threshold is ‑6.00 dB. But if the Ceiling is ‑3.00 dB and the Threshold is -6.00, then the limiter’s Threshold is -9.00 dB. Makeup gain occurs automatically so that as you lower the Threshold parameter, the output level increases as needed to meet the Ceiling’s target output level.
Modes and Attacks
There are now two Modes, A and B, and three Attack time settings. The pre-V5 Limiter had less flexible attack options, which mostly impacted how it responded to low-frequency audio; the waveform could have some visible distortion when first clamped, but the distortion would disappear after the attack time completed.
I’ll spare you the hours I spent listening and nerding around with the (highly underrated) Tone Generator and Scope plug-ins to analyze how the new options affect the sound, so here’s the bottom line.
In applications where you want to apply something like 6 dB of peak reduction to make a track or mix “pop,” the Limiter2 performance in Mode A is essentially perfect. Unless you’re into extreme amounts of limiting or material with lots of low frequencies, Mode A should cover what you need 95% of the time (and it also outperforms the pre-V5 limiter).
If you’re using Limiter2 as a brickwall limiter to keep transients from spilling over into subsequent stages, then use Mode A/Fast attack for the highest fidelity and give up a tiny bit of headroom, or Mode B/Fast Attack for absolute clamping.
Fig. 2 shows how the fast and slow times compare. The following were all set for 50 ms release times, 1 kHz sine wave input, and -20 dB Threshold—so Limiter2 was being hit pretty hard.
Figure 2: Fast and Slow attacks compared for Modes A and B, cropped to 150 ms duration. Top to bottom: Mode A/Fast, Mode A/Slow, Mode B/Fast, Mode B/Slow.
The visuals are helpful, but on signals with fast transients, you may hear more of a difference with the different attack times than these images might indicate. Nonetheless, it’s clear that Mode B/Fast is super-fast. If you look carefully at Mode A/Slow, you’ll see a very tiny downward blip on the first cycle of the attack (it’s less visible on Mode B/Slow). Mode A takes about 20 ms to settle down to its final level.
For more background on the nuts and bolts of how this works, the tradeoff for Mode B’s higher speed mostly involves very low frequencies (under 80 Hz or so, and especially under 50 Hz). With a 50 Hz sawtooth wave, 100 ms Release, and a significant amount of limiting, Mode B/Slow gives some visible overshoot and distortion. Mode B/Fast reduces the overshoot but increases distortion. Mode A does less of both—with Slow, there’s less overshoot, and with Fast, there’s less distortion. Note that any distortion or overshoot occurs only when pushing Limiter2 to extremes: very low-frequency waveforms, with steep rise/fall times, short release times, and lots of limiting. However, this is mostly of academic interest with waveforms that have lots of harmonics, like sawtooth and square. The level of the harmonics is high enough to mask any low-level harmonics generated by distortion.
I also tested with a sine wave, which gives an indication of what to expect with audio like a kick drum (e.g., 40-60 Hz fundamental) or low bass notes. Mode B/Fast has less distortion than Mode B/Slow, while Mode A, fast or slow, flattens peaks almost indiscernibly (Fig. 3).
Figure 3: A 30 Hz sine wave with about 15 dB of limiting. Top: A Mode. Middle: B Mode/Fast. Bottom: B Mode/Slow.
In this situation, Mode A would likely be my choice, but as always, use your ears—the light distortion from Mode B can actually enhance kick drum and bass tracks. Also note that which mode to use depends on the release time. For example, with a short (35 ms) release, B/Slow had the most audible distortion, B/Fast was next, and B/Normal had no audible distortion.
While I was in testing mode, I decided to check out some third-party limiters (Fig. 4) with a different program. These are all marketed as “vintage” emulations, and set to the fastest possible attack time.
Figure 4: The results of testing some other limiters.
In case you wondered why some people say these vintage limiters have “punch”…now you know why! The time required to settle down to the final level is pretty short (except for the bottom one), but the limiter doesn’t catch the initial peaks. This is why you can insert one of these kinds of limiters, think you’re limiting the signal, but the downstream overload indicators still light on transients. Incidentally, the Fat Channel’s Tube limiter has this kind of “vintage punch” response in the Limit mode, while the Fat Channel’s one-knob, final limiter stage—although basic—is highly effective at trapping transients.