By Paul Vnuk Jr.
Twenty years after its creation, the Empirical Labs EL8 Distressor is a modern classic. It just hit serial number 30,000, and its most common descriptor over the years has been “the Swiss Army knife of compressors.”
It’s curious, then, that despite rumors and “coming soons” from other companies, there still is not an actual 1:1 Distressor plug-in. Empirical Labs does have a software model of its EL7 Fatso Jr in collaboration with Universal Audio for the UAD-2 Powered Plug-in platform, but why no Distressor plug-in from Empirical Labs? The answer is because Empirical founder and President Dave Derr decided to release the Arousor compressor plug-in instead.
Is the Arousor a Distressor? It starts out that way, but thanks to the open-ended world of digital modeling, it’s expanded beyond the original in ways that may be impossible (or at least cost-prohibitive) in the real world.
Like the Distressor, the Arousor is a classic feedback-style compressor. Its user interface launches as an instantly-recognizable Empirical piece, thanks to the four large white knobs and an 18-segment LED meter showing 0 to –30 dB of multi-colored reduction. There’s also the famous extra BAD! LED, found on many EL devices, that lets you know when your signal is within 0.5 dB of hard clipping on the output.
- DECEMBER 2017: Presonus Studiolive Ar12 USB Mixer/Interface
- NOVEMBER 2017: Focal Shape 40 and Shape 65 Studio Monitors
- OCTOBER 2017: Mojave Audio MA-1000 Condenser Mic
- SEPTEMBER 2017: Empirical Labs Arousor
- AUGUST 2017: Mackie DL32R 32-Channel Digital Mixer
- JULY 2017: API TranZformer GT
- JUNE 2017: Reviewed & Revisited: Steinberg Cubase Pro 9
- MAY 2017: LEWITT LCT 440 PURE
- APRIL 2017: Reviewed & Revisited – Acoustica Mixcraft 8
- MARCH 2017: Prism Sound CALLIA USB Audio DAC/Preamplifier
- FEBRUARY 2017: Universal Audio Manley VOXBOX Channel Strip Plug-in For UAD-2
- JANUARY 2017: SPL Transient Designer TDx
The knobs, marked 1–10, are: In, Attack, Release, and Output. Input drives the compressor, and takes the place of the Threshold control that most compressors have; Output lets you adjust makeup gain (and turn off the BAD! light). Attack is variable from 50 microseconds to 40 milliseconds, and Release can be 50 ms to 3 seconds. Each knob contains the first of many plug-in-only features: a blue numeric value representing each parameter’s current setting. You can double click and enter an exact value as well.
A second set of ten LEDs indicates the Compression Ratio. There is no button for ratio selection; you just click on the ratio LED you want. The Arousor offers two more ratios than the Distressor, with a choice of 1:1, 1.5:1, 2:1, 3:1, 4:1, 6:1, 8:1,10:1, 20:1, and Rivet (“Nuke” on the EL8). Currently 10:1 is not an Opto mode circuit as per the Distressor, although we are assured that a true Opto emulation will be coming in a future software update (see the interview with Dave Derr later on in this piece). The “all-buttons in” Brit mode from the Distressor is also still to come.
The really new stuff
So far the above controls are very much those of a slightly modified Distressor, but on the lower half of the plug-in, things get a little wild. First up is Attack Mod, with a 3-stage AtMod LED and a small black knob where you can adjust the curve of the attack. Note this is not the same as adjusting the knee; turning the control adjusts the attack to a softer and slower slope which changes the initial attack of strong transients. It’s almost a “reverse or anti-knee”.
One of the biggest features of the Distressor is how it can saturate, distort, and clip a signal. This is retained and expanded on the Arousor in the Soft Clipping section. Where the Distressor offered preset choices for second-order or third-order harmonic distortion, the Arousor is fully variable from 0.5% up to 14% Total Harmonic Distortion (THD), complete with 4-stage LED indication.
The last expanded set of controls is all about the sidechain detector. The Distressor’s feature set has highpass sidechaining and band-selectable sidechaining via four preset button choices. However, the Arousor takes off from there. First you get a fully variable highpass sidechain filter with a gentle 6 dB/octave slope and a corner frequency from Off up to 1 kHz. Next is a fully parametric sidechain EQ with a throw of 40 Hz up to 10 kHz with ±30 dB of gain or attenuation and an adjustable Q. Note that both of these detectors are not audio-path filters; they simply tell the compressor what to ignore or what to highlight, and both can be used in tandem. This brings up one feature found on the Distressor, but not the Arousor: a highpass filter which ispart of the audio path, useful for cleaning up low rumble out of the sound pre-compression. We’re not sure if this will be in a future update.
Finally, there’s the Blend control, a wet/dry mix allowing for parallel compression inside the unit.
Like all plug-ins, you can store presets and the Arousor comes with a nice selection of starting points. Also note that around the Arousor interface you will notice blue dots. These indicate the Distressor’s base settings, so you can match the units.
Distressed and Aroused
Of course the first thing I wanted to try with the Arousor was to compare it with my trusty EL8x. I defaulted all of the Arousor’s secondary settings to their Distressor defaults; essentially, all of the non-Distressor features set to flat or turned off. Cutting to the chase, the words “darn near spot on” come to mind. I have used many plug-ins that get this close on attack, release, and even saturation matching, but where many plug-ins struggle is in the translation of digital input and output levels compared to analog. I was really impressed with how close the levels between the hardware and plug-in came. Dave Derr told me he worked hard at getting this correlation correct, and his diligence shows.
I have seen and read some other reviews that claim the units could not be matched knob-for-knob and sounded slightly off. I could not disagree more! In fact, one of my coolest 2017 Winter NAMM Show moments came when famed engineer Ross Hogarth stopped by the Empirical Labs booth and gave Dave and me a tutorial of some of his favored Distressor settings. Ross then proceeded to match those settings, and more importantly, the control and sound, on the Arousor. What an eye-opener!
Speaking of eyes being open: to be fair to the naysayers, one thing I did notice initially was that while my ears told me the two units were working similarly, the metering on the Arousor was a touch more sluggish and not hitting as deep into the reduction range, so to my eyes the units were mismatched. According to Dave, this is/was a known iLok-induced bug in the first few revs of the software. This has been corrected in the latest software update, which should be out by the time you read this. I tried the new update and the meter is indeed more accurate, snappy, and Distressor-like, so your ears and eyes will no longer be at odds.
The Arousor, like the Distressor, is a Swiss Army knife compressor. It works great on vocals, electric guitar, bass, snare drum, and more. Thanks to its more tweakable detector features, it’s also great on delicate sources like acoustic guitar, mandolin, and violin—not instruments where I would usually reach for my Distressor.
In the same vein, I have never been a huge fan of stereo Distressors on the 2-bus, and on the drum bus I always prefer to set up parallel compression routing. With the variable throw of the highpass detector and the built-in Blend control, the Arousor is instantly better suited to 2-bus work, and it’s just as good as the Distressor on the drum bus… just easier to set up and tweak.
The Arousor’s price tag is on the high side for single plug-ins. Empirical Labs is unapologetic about this, considering how much time, money, and design work went into this plug-in. Empirical Labs is going to expand its plug-in offerings in the future and is 100% committed to the Arousor for the long haul, with regular updates and new features added regularly for registered users—all with no update fees through the year 2020.
As mentioned, I played around with the forthcoming Release 2.0. In addition to the bargraph meter behavior being improved, the look of the plug-in is more 3D and lifelike, you can now share presets, and the big news is that the first of many planned Expert features (in the form of a small blue “E” button) has been included.
This feature in this current revision opens up the Blend section with an extra dry-signal trim control. This may seem strange, since turning the signal more wet yields less dry signal, but what it does is allows you to tweak the overall dry-wet level so the output volume remains consistent and matched when sweeping the Blend control between 0 (dry) and 10 (compressed). Further “coming soon” features include the Opto mode, more under-the-hood parameters, and more little blue dots to indicate settings for Distressor presets.
My quick and simple wrapup: the Arousor is like its hardware ancestor—highly versatile, simple to set up and use, and hard to sonically mess up. It’s a great investment that you can and will use on anything and everything, because above all else it sounds fantastic.
Delivery: Web download
Formats: Windows 7+ and Mac OS X 10.7+; AU (Mac only), AAX, AAX-DSP, VST2, and VST3
Copy Protection: PACE iLok 2 or iLok 3 USB key
License: single user/two activations provided (requires two iLok keys)
Documentation: PDF download
More from: Empirical Labs, www.empiricallabs.com/arousor.html
The Ideas Behind The Arousor
Interview with Dave Derr of Empirical Labs
Let’s start with a little bit of background. Prior to starting Empirical Labs, you were at Eventide?
Dave Derr: I had been in a band until I was 33, an attempted full-time musician. But I’d always had an interest in electronics, and when you’re in a band, you either learn to repair your own gear or you’d better be rich! So I already had basic electronics under my belt. I have two brothers who are both engineers, so there’s a lot of technical background and a love of science there already.
Thanks to a chance meeting at a Radio Shack one day, I got a job working as a technician for a company that built medical electronics. I got very lucky; the head engineer, Jim Bryan, also loved audio. He and I bonded over music and he was so kind to me. I would do little projects of my own after work, paying for the parts I used, but he’d say, “You don’t have to pay for parts… what you do after hours helps you learn to be a better employee here.” I now do the same thing here at Empirical Labs.
I had friends at Eventide, although I’d never been there. I saw an ad that Eventide was looking for an engineer. Even though I didn’t have an engineering degree, I had taught myself a bit… I interviewed, got lucky again, and was hired as an engineer. That was the training ground for Empirical Labs. I worked with Ken Bogdanowicz and Bob Belcher, who now run Soundtoys, and those guys were my mentors.
On the H3000, I started out designing PC boards and working on the analog side. They were software guys; once they had the algorithms and DSP code working, I got involved on the software side because they needed presets that would be useful in a studio setting.
How did Empirical Labs come about?
Around 1988, I had bought a 3M 2” 16-track machine, which ended up turning into a recording studio, and did some consulting on the side… so I founded a corporation called Empirical Labs, which technically owned all the studio equipment. Eventually I built up enough business so that I could pay myself, barely, and I tidied up the work I was doing at Eventide (on the DSP4000 at the time) and left. It was a very amicable parting; I’ve stayed good friends with Richard, the owner of Eventide, and I count his friendship as a vital driving factor in the development of Empirical Labs.
Within a few months, I was working fulltime in the studio, and because of my love of electronics, I was always building things: a custom switcher box, or a submixer, stuff like that. I was always fascinated with compression, so I started building compressors… before the Distressor, I’d probably built four or five different types.
At the studio, my favorites were the LA-2A, the 1176, and the original Valley People Gain Brain (the Gain Brain II is a very different design). As with anyone who’s worked mainly with consumer-level gear and suddenly gets a piece or two at that level, I had that “Ah…!” moment when I heard the difference, and then I asked myself, “Why do these particular pieces do the magic?”
I started narrowing down the field, learning how to get that sound without having to rely on really old gear. It was a selfish thing at first. I built a couple of prototypes, then a refined model, and over two or three years, it became the Distressor!
The Distressor is interesting because it was an analog compressor, but there were digital elements as well. That wasn’t common, was it?
Not at all. Pretty much everything was either all-analog or all-digital. The thing that was obvious to me was that my two 1176s were very different in response, and so were my two Gain Brains. The same settings on each wouldn’t give the same results. From my experience at Eventide, I remembered putting digital controls on the analog signal path. Looking at compressors, I wondered if I could get rid of those inconsistencies that way, and from the very first prototype, I was like, “Oh my gosh… this is so easy to do!” [laughs]
Just as an example, I could substitute digitally-controlled analog switches for traditional switches or relays in the control path, and easily get them clean enough to work well (since the user never hears the audio in the control path). Very quickly, I was able to do things that had never been done before, while maintaining a feedback compressor’s characteristic sound.
Just as with the Fatso plug-in, where you can open up the hood and get at functions that weren’t available on the hardware, the Arousor is sort of like “Distressor 2.0” in that it looks and functions very much like a Distressor until you pop the hood and have access to extra functions a Distressor can’t do. What was it like, designing your first plug-in? Was it fun to go back and revisit your design and help it to evolve?
It was awesome! Eventide, as you know, was primarily a digital company, and I was very excited about getting back into the digital domain. Of course, look at what’s happened in the last ten years—from studios to live sound, everything is digital! It was expensive to develop… everything always costs more than you think it’s going to, and that’s stressful. But actually working with it? That was a blast, it was so much fun.
As you said, there was so much stuff that we could do very easily that had been impossible before. That’s actually one reason why we didn’t call it the “Distressor plug-in”… with all it could do, we didn’t want it to be thought of as just a software Distressor. Making sure that new stuff won’t break existing projects is a huge, terrible job… we’ve probably put a year, all told, into making sure that the software could evolve without breaking anything. We designed the Arousor to be, if this is a word, ‘evolutional’.
There are at least 16 features hidden inside that we will gradually reveal and make editable as the software matures. We’re in this for the long term. Over the next few years, users are going to see and get stuff that no one else has ever offered. Just as one example, the Arousor has no Opto setting like the Distressor’s… but it will get one. It’s planned for.
Most hardware companies partner with software companies to create plug-in versions of their gear, in a sort of “co-branding”. Why did you decide to develop the Arousor entirely in-house with your own team?
That’s a good question, and there are three parts to the answer. The first is, as I said, we’re in this for the long run. We see ourselves as becoming more and more of a plug-in company. Second, the situation where you have another company developing the software takes off the burden of development and marketing… but that company has to be compensated for doing most of the hard work, so it will take 70% or more of what you sell the product for.
Third and most importantly, if another company’s writing your code, you have less control. We worked previously with a world-class plug-in developer, and it was fantastic—they did a great job and their code team was a well-oiled machine. But if I wanted to devote three months or a year of a code team’s work time to do things that someone else considers small or even trivial… I have to have control in order to make those decisions. Remember I said it took a year to get the Arousor code ‘evolutional’? I had the option to do that and have it done to my satisfaction. I’m not usually an OCD person, but with products… that’s my art, and I get picky. I wanted to be able to get as picky as I wanted with it. Ask my engineers! [laughs]
Historically, the first compressors didn’t have sidechains; you’d have to build your own using external EQ. Then they all had single-button highpass filters for internal sidechains. The Arousor includes a variable highpass filter and a full-on EQ to tweak the sidechain response. How did you arrive at that?
On the Distressor, because we were using digital control on the sidechain, it became very controllable; we could switch different EQ circuits in and out. Those had to be fixed—you can’t tweak frequencies or boosts or cuts. The Arousor basically has the same signal flow diagram, but adding an adjustable EQ costs no more DSP than a fixed one! What would have taken capacitors and extra components in the Distressor, you can do just with code in the Arousor.
In the digital domain, it’s easy to create a sidechain using another DAW channel or plug-in if the compressor is a feed-forward design, but a true feedback compressor is much more limited in that respect. So we put the full parametric EQ and variable filter right into the plug-in. It cost us very little and got around the traditional limitations.
Anything you can tell us about future plug-in designs?
Well, the obvious thing is an EQ, but we’re asking ourselves if the world needs another EQ plug-in [laughs]. Of course, there are things one can do that haven’t been done yet… or at least haven’t been done properly. If we do an EQ, it’s safe to say that while we won’t try to simply reproduce the Lil FrEQ, the same way the Arousor wasn’t limited to just reproducing the Distressor. But whatever we do will be influenced by the Lil FrEQ.
The other thing that’s kind of obvious is a delay. There are a million delays out there, but there are features that I’ve never really seen implemented right. Remember how you looked at the variable sidechain filter on the Arousor, and said to yourself, “Well, that’s an obvious thing to do”? The delay we’re planning will have a couple of features that will have people saying, “Well, that’s kind of obvious… how come nobody’s done it before?” [laughs]