Sponsored by Empirical Labs.

BIG FrEQ Intro Offer: click here to save $20!

Design Spotlight: Empirical Labs' BIG FrEQ Plug-In-empirical-labs-big-freq.jpg

Empirical Labs’ highly anticipated new BIG FrEQ EQ plug-in is rooted in the company’s iconic Lil FrEQ hardware equalizer, augmented with inventive features and processes that take full advantage of the digital realm.

Empirical Labs describes BIG FrEQ as “tone shaping with an analog twist”: The plug-in has 12 processing sections, with 6 parametric EQ sections with tone-shaping parameters that let users create modern “flat-top” parametric shapes, plus a unique FINISHER output section that adds organic harmonic distortion. The plug-in’s analog-style workflow offers innovative functions including superior fine-tuning controls for ultra-precise tone shaping, Slope and Width controls, band-specific soloing, and highpass and lowpass filters with adjustable Q and selectable orders from 6 to 96 dB/octave.

Gearslutz sat down with Empirical Labs’ resident mad scientist, Dave Derr, the mastermind behind studio classics like the Distressor, FATSO, and the Arousor, to learn how BIG FrEQ came to be and find out what’s under the hood.

Let’s start by talking about the design philosophy at Empirical Labs.
We really try to not throw a lot of products out there and see what sticks. So, one of our mottos is, make a few great products, where we not only design stuff but are ready to throw it out if it doesn't end up seeming right. And we've done that. We've got four or five products that we designed all the way to the point where they could be put into production and then killed them because either the testing didn't turn out like we wanted, lacked reproducibility, or the product was too expensive, or it just seemed like there was no market for it.

For example, the Distressor was the first product that we released. But the FATSO, which was our second product and our second-biggest seller, almost didn't make the cut because we didn't know how to define it.

People didn't get it?
I was ready to cut it because it just looked too expensive, and it was hard to define what it actually did. The first people who got it were like, "Oh, what do I do with this?" But now it's our second-biggest-selling product.

The idea for this plug-in first came up in 2014, correct?
Yeah, probably. Maybe even before.

Did you have it on the back burner or were you working on it all that time?
I've been thinking about it forever, really. Back when I was at Eventide, I worked with a great team, and we built some EQs there. And I was playing with things there that we now are using in the BIG FrEQ. But to me, the real point of when a software project starts is when I do my first GUI drawing. That's when you get serious about how it's going to end up.

Did you always envision the BIG FrEQ plug-in as an evolution of Lil FrEQ?
The hardware is often a starting point, but we don't want to just copy it; I have no interest in that. I've come to blows a couple times with marketing people because they're like, "Just give us the Distressor. That's what people want." The fact is, other people do that. For instance, Universal Audio, they put out a great Distressor plug-in. It's really amazing. But our interest is always something new. Take advantage of the medium you're working in.

Design Spotlight: Empirical Labs' BIG FrEQ Plug-In-lil-freq.jpg

BIG FrEQ is a descendent of the Lil FrEQ hardware processor.

How did the design evolve along the way?
It changed a lot. We started back in 2014, thinking about a plug-in EQ. We had all these features that we thought would be really unusual and set us apart. But before we even started coding, a lot of those features were put out by other people.

I thought, "Well, we'll keep that feature, but it's been done so now we have to come up with something new. Because otherwise, it's just going to be another EQ plug-in without any advantage." So, yeah, we adapted the whole time and we're still adapting.

Do you often engage the user community during the design process? You get a lot of feedback here on Gearslutz, for example. Do you consider it all?
We take feedback very, very seriously. And we are always really big on beta testing. “Empirical” means “fashion from experience,” or, in the cheap sense, "trial and error." You make sure stuff works in the real world. That's a big part of what the beta testers help us do. As you can see from the Gearslutz forums, some things we can't do, or won't do, or will put off. But we listen to what people say and we take the hints and suggestions seriously.

Gearslutz feedback really has a big effect and I love it. These guys are customers and they're telling you what they want. But we don't just put stuff in there on a whim. Everything is agonized over. There's always a reason why we design something. So, until they actually start playing with it, they won't know.

Some people call critics the greatest optimists, because they ultimately just want the best for something.
Yes, exactly. One of my theories is that it’s the negative critiques that are most valuable and moves you forward. Listen to them, because it’s the people saying "Why'd you do this? Why didn't you do that?" who will often make your product better.

It's an interesting a dance, because so much audio product marketing, especially software, focuses on endless feature lists. How do you balance packing in the features with quality and workflow?
That's a really good question because you just can't offer every feature. Sometimes we try to add everything. But the fact is, it's got to be really useful and it can't be confusing. So it's a constant funnelling of what to do and what not to do. Like when people say, “everybody else does this. Why aren't you doing this?" You have to have a certain amount of self-confidence.

There's also the Steve Jobs thing. We often do market research quite informally, like I'll say, "We're thinking about doing an EQ plug-in. What are your favorite EQs?" Steve Jobs, though, when he was working on the iPhone, his team said, "Let's do some marketing research on this, Steve. We'd get a lot of feedback." He said, "No." They asked, "Why not?" He said, "People don't know what they want until we show them." He was a real innovator and he was always thinking outside the box.

Did you go into this wanting to categorize this as say, a character EQ, or a Swiss Army knife EQ?
Both! Ha. Swiss Army knife and character. Our Lil FrEQ was kind of a Swiss Army knife. We tried to make it a one-stop EQ. But the truth is, no matter how much you try to do that, it can't be the Swiss Army knife for everybody. The Lil FrEQ, we took that starting point. But at the same time, we add color with the FINISHER feature, which is saturation, basically.

What makes these EQs different?
A few things separate us. The parametric section, we built a whole new control called Slope. I don't think anyone else has treated your standard parametric the same way, adding a fourth infinitely variable control. Most of them control gain, frequency, and width of boost and cut; three things.

Focus was something I've been thinking about for a while. In a hardware EQ, you have to have your low-frequency band, you have to have your mid-low, you have to have your mid, you have to have your high-mid, you have to have your high. You can't have every single frequency band go the whole way, 20 Hz to 20 kHz. Not so in digital; it’s almost easier.

The downside is, with that wide range, adjustments get very touchy. The FOCUS control is your way out of that. You can turn a band into a very narrow mid band or a very narrow low band, etc. Tune into a general frequency area you're going to be messing with. When you hit FOCUS, it takes wherever that frequency was and puts it right up top, at 12:00. Then it narrows the range to one octave down and one octave up. And all of a sudden tuning the frequencies for that band becomes really easy because that band is no longer going between extremes, from below what you can hear to above what you can hear.

Design Spotlight: Empirical Labs' BIG FrEQ Plug-In-precision.gif

The Slope, Focus and Solo controls in action

The Solo function is unique; tell me how that came about.
A lot of EQs let you hold the Solo button down and hear the frequencies you're boosting or cutting. Then when you let the button go, you're back to hearing everything. That's a bit of a hassle when you're trying to pick a clear frequency and you're using your mouse. In the BIG FrEQ, you can click the Solo on, and it stays on until you click it off. So now you can fine-tune without having to hold a button down.

The other unique feature is, you can solo more than one band at a time because of that auto on and off. Generally, that's not that useful, being able to solo two bands, because usually you're focusing on one or the other. But what we've already found is people are actually using that. Soloing two bands and getting really weird sounds.

The way Solo works in conjunction with Focus... when I was playing with vocals, the Solo button worked just fine. But quickly, I found that before we had Focus, trying to get that exact frequency was exasperating. You would get it, and then you'd let go the mouse it would move just a little bit. By narrowing the range with Focus and combining that with Solo, finding problem points or weak points is a lot easier. So making what was often a tedious chore hopefully became a lot easier and more fun.

Let’s touch on the interface. I think people sometimes wrestle with their tools constraining the way they work versus facilitating creativity. And they can get into that trap where they work visually.
Yeah, that's a good point. Another one of our design criteria is fun. It's got to be fun to use. When you turn knobs, you hear things.

We agonize over ease of use and speed, and a lot of the features we built changed as we used them. For example, in the original prototype, the gain range was plus or minus 30 dB. Not many people will ever use that, and the control was very touchy. We now have a switch that narrows the range down to plus or minus 15. All of a sudden it becomes easier to use.

When you have a lot in your GUI, you have to take some shortcuts. But you want a shortcut that after using once, you won't forget how to use it. On the highpass filter, we have a little button where you can change what the HP knob does, turning it into a variable Q control, selecting how sharp it is. We tried to put it right in your face. But if we add a knob and switch for everything, it gets even more crowded. So, we took some shortcuts to make it look a little simpler.

How did the FINISHER saturation function come to be?
The FINISHER is a story unto itself. I would say we did seven or eight different FINISHER circuits before we came up with the one we're using now. Our DSP guys would spend a few days on something, and then we'd play with it, and at the end I would be going, "Oh, I just don't think this is useful enough." You've got to be ruthless to a degree… and it hurts to throw away all that work. I give our engineers a lot of credit because they had a LOT of patience, and a pursuit for excellence. Of course we looked at competitor plug-in saturators—and there are some great ones—but it helped us avoid doing the same tricks.

After a three months R&D on the FINISHER, we came back around to a very organic type of saturation, with all the characteristics and pitfalls of analog saturation. In the Arousor, we use a soft-clipping algorithm that wonderfully "scrunches" the peaks, but it is nothing one could do in the Analog circuit world. It looks and sounds so smooth, yet produces almost no aliasing, or un-musical artifacts, even at amazingly high amounts of soft clipping. 

The FINISHER has an organic and more real-world analog saturation, and theoretically, the harmonics generated go on forever, so we had to provide Oversampling. This eliminates the ugly “alias tones” you would get when using 8% or more saturation. Now we know that it’s a different type of saturation that works synergistically with the Arousors. The two saturator shapes are not additive. The soft clipping is further differentiated with filtering around the FINISHER, as it does a low-frequency thing.

Will BIG FrEQ remain native?
For now, yes. But we're always open to licensing things. And we have some great partners. So, absolutely, we'll leave that open.

People in the forums have been asking about Mid/Side processing.
Yeah. The good news is, Mid/Side processing is adding and subtracting, very easy to do. However, making it fun to use and easy, that's going to be the hard part.

Design Spotlight: Empirical Labs' BIG FrEQ Plug-In-dave-big-freq.jpg

Derr at work on the BIG FrEQ.

We can all relate to the difficulty of calling something done. How do you approach revisions?
You're asking the wrong person because to me, nothing's ever done, nothing's ever perfect. I mean, we still have a major Rev of our Arousor coming out after four years.

Every rev is a moving target. But there's some stuff that is black and white. If Pro Tools is crashing, and it quits crashing, then you're done with that. But as far as features, we're always changing what's next.

We are influenced by Agile, or Scrum development, a process that can allow for speed in getting products to market. The key part is flexibility and not locking things in too far ahead. And, lots of communication. Not that we use Scrum in a rigid way, but we allow flexibility and the ability to change on the fly. I mean, look at this market! It's so fast. If I had locked in the features from four years ago, we wouldn't have a chance. As we see what other people come out with, we keep adjusting what we think is important.

It sounds very dynamic, with more opportunities to learn from the process.
Absolutely. And because we haven't locked the features down, even at this point, there's not a lot of things we've ruled out as being possible. During our two months on Gearslutz since we first made BIG FrEQ available, we've done more than 11 revisions.

Were those in response to the feedback that you got?
Absolutely. Some of them were just bug fixes and stuff, some of them never saw the light of day. But we would be, like you said, trying to figure out when something was good enough and done. So, the flexibility of being able to adjust on the fly is really important.

What are you working on for future versions?
Our Lil FrEQ has a dynamic EQ section in it. We don't have that right now in the BIG FrEQ, but we plan on implementing some form of dynamic equalization. We may someday have a click-and-drag mode of EQ adjustment. We may have spectrum analysis. We're going to have Mid/Side processing. We're looking into surround sound capability, multichannel linking capability. And we may have a Compact mode where some stuff is hidden. We probably have on and off another year of development. So we're a long way from done.

I can’t stress enough how instrumental the Gearslutz community has been. I can't imagine actually being where we are without Gearslutz. They’ve been a big part of the development process.

Plus, they're fun. Every now and then I will challenge someone who challenges me. But generally, I listen to them. Your critics are your best friends in development. Listen to those people, because they're out there creating, they're your next customers, you know? They’re telling you exactly what they want. How much easier could it be to define a product?

Visit wwww.empiricallabs.com to download a free 15-day trial version of BIG FrEQ.