Matt Phillips talks with us about his background in acoustics, his career of “ghost producing” plugins, and his new plugin company Kalifornia Dynamic.
So you essentially “ghost produce” entire plugins for other companies?
Yeah!
Which is both good and bad for this interview because a lot of the time they make me sign NDAs and documents where I’m not allowed to say what I’ve done for them. There’s one past project in particular I wish I could tell you about!
There are a few people like me. We’re contractors essentially. We go where we’re needed, all kinds of different projects.
Alkane is actually a personal project. That’s not something I did for another company.
Ah, so Kalifonia Dynamic is your own brand?
Yes, that’s right. It’s fairly new. Alkane is the first product.
Alkane is a harmonic processor?
It’s four flavors of saturation and distortion in one package. They’ve all got their own kind of unique sound. I’ve always been interested in nonlinear modeling, and I love trying all the different brands and styles of saturation plugins out there.
My biggest problem with almost all of them is the transient power that gets lost in the process. So Alkane was designed with that in mind, to cover that common shortfall. A lot of pro users love Alkane for that exact reason. Joseph Trapanese for example, who was kind enough to test out my early beta versions and give me great feedback, uses Alkane in all of his sessions now. It’s his go-to saturator.
It keeps the transients more crisp and clean, lets them pass through?
Exactly. There’s a lot going on in Alkane to make sure those transients are preserved, but to give you just one example: you’ll often encounter overly aggressive automatic gain control in other saturation and distortion plugins, and a lot of the time that will kill your transients, or at least bring them down to a level that’s not creatively desirable. The automatic gain control in Alkane is subtle. It’s under the hood doing its thing, but I designed it to be more natural than intrusive.
I’m building an additive synth, so wanted to selfishly ask if you had any perspectives around an “additive purist” approach to saturation.
Are you looking for avenues to go down in terms of how pleasant or unpleasant your saturation sounds? Or do you just want to go the whole way and boil it over?
I’ll take whatever tips you can give me! I’ve looked into what my favorite plugins do to the spectrum and got a bit stuck because difference tones show up, there’s a lot of non-harmonic content, etc. I just thought you could maybe throw me a bone here since you’re the pro saturator. [laughs]
Yeah! I’m trying to think because you’re coming at it from a completely purist point of view. With Alkane, I do it in a very different way.
If you’re looking to give them a bite or saturate knob, I would stick to the fundamentals to begin with. Just look at your trig functions and kind of build things out that way. I’m not suggesting you should just drop in a tanh and call it a day, but start with the basics and go from there. You could also experiment further by adding in bias adjustments.
Maybe a better question: When you think of saturation, how do you think about it? What’s your mental model? Without using words like “warmth!” [laughs]
It depends what you’re using it for. There’s creative distortion where you just want to drive everything to the max and just break it up completely. Then there’s the more mastering grade saturation where you’re just introducing enough chaos to smooth out the experience.
I do a lot of tuning by ear. So when I think of saturation, I immediately think of the problems. Take the tube saturation model in Alkane, for example. If that was just a straight up textbook model of tube-like behavior, then you would likely get that horrible sounding low-end breakup when you cranked it. So I’m considering these kinds of things first, and how to handle them in my own models.
And of course, I’m always thinking about how saturation can make something sound pleasant. Not just how to implement those qualities but also how to enhance them, which is what the Thud and Shine controls on Alkane allow you to do. That’s the hardest part, versus designing something that’s just driving audio into oblivion. Anyone can do that really. I don’t want to say it’s art, that sounds ridiculous —
It’s at least sculpture!
Yeah. So I always go into saturation thinking, if I use this function or this model I come up with, what am I going to need to do to make that actually acceptable and usable in this product? And that might involve introducing or further manipulating certain harmonics, which you would perhaps not ordinarily expect from a particular type of component, for example. That’s when tuning by ear really helps.
I’ve been thinking about psychoacoustics and how beating turns into roughness. Like, if you are playing an electric guitar through heavy distortion, even playing consonant intervals you’ll get a rough sound, the tones are modulating, you end up with a lot of frequencies clustered together.
A lot of what we do comes down to psychoacoustics. It’s ironic that within the digital environment we occupy now, with all of these incredibly powerful hosts and plugins, we’re still striving to make signals objectively poorer from an analyzer point of view, because they sound more pleasant to us as humans. There’s this push to kind of go backwards and introduce the chaos and imperfections of analogue.
So, how did you fall into “ghost producing?” Did you start off by augmenting teams? Delivering entire products requires a lot of trust on the company side.
I’ve worked in audio for probably 18 years now in one form or another. When I first moved from the UK to the US, I worked for a company, a hardware DSP design company, working on some of their new products. I developed a good understanding of what it took to see a commercial product through from start to finish.
I’ve always been interested in making plugins and audio tools from the really early days of the Apple Core Audio API. Back in 2005, when I was still in the UK and at college – or university as we would call it – my final dissertation was based around the design and creation of Audio Unit plugins — long before we had the luxury of JUCE. I always kept that interest going over the years. It got to the point where friends and contacts in the music industry started using my experimental creations in their professional lives, so I knew I could deliver plugins that would work commercially.
Was college your first foray into “real programming” and C++?
I did a lot of C++ during college. I remember thinking, “I have a working Audio Unit. This is really cool. If only there was a framework that could really help me out when it comes to needing the VST version. And then the AVID version.’” Obviously JUCE ended up solving that problem for many of us.
I went on to complete a post-graduate certificate in intelligent systems, which involved a lot of advanced C++ and that really refined things on my end. I also used C++ throughout my master’s degree in acoustics. It just ended up being a kind of constant throughout my career. I was so heavily immersed in audio projects for permanent employers, I eventually thought, “Well, it’d be really cool to branch out and maybe do some contracting work for DSP and plugin design, perhaps some acoustic software development as well.” I combined all of that, and ultimately Phillips Audio Designs was born.
Without JUCE, I wouldn’t have a company. It’s not a shortcut per se, but there is no way on earth, in my humble opinion, that anyone could start a viable audio plugin company without it. If JUCE didn’t exist, I think you would see far fewer examples of innovative startups in our industry.
Without JUCE, I wouldn’t have a company. It’s not a shortcut per se, but there is no way on earth, in my humble opinion, that anyone could start a viable audio plugin company without it.
–Matt Phillips, Kalifornia Dynamic
I checked out your ADC mentorship interview and it looked like that went well?
Yes, the first mentee — I can’t remember when I signed up for that, it was last year — but we’re still in touch! We actually met in person a few times.
I was so lucky because he’s so enthusiastic.
I don’t know how much you can divulge, but is the “ghost producing” done on a flat-rate or monthly basis, or is it more of a licensing kind of thing? What kinds of setups do you prefer?
A lot of the full-cycle development projects, start-to-finish plugin designs, will ultimately involve a royalty share or license.
I’ll often say producing the plugin or the software or the product — that’s the easy part. You’ve then got to maintain it over time. So, I don’t just make products for these companies and then leave them in the lurch. Royalty splits might sometimes include provisions to cover the maintenance side of things, within reason. There might be additional fees up front or throughout, depending on the size and complexity of the project.
I don’t think I’ve ever had two clients with the exact same pricing structure in place. It very much depends on what the design is and what the scope is.
I would love more details there, but I understand if —
No, it’s fine! Anything you want to know.
Royalty splits always come up re: what’s fair. Many newer devs tend to think that they’re doing most of the work and should therefore get most of the split. But having an established audience, knowing the market and all these things are pretty damn important — 100% of zero is still zero.
Exactly. That’s a great point. A lot of the time the way you structure a royalty split, depends on the presence that brand has, and maybe their track record. So if you’re, for example, designing a plugin for a startup company, then you have to be much more wary because it’s on them to get those numbers and get those sales.
And like you said, 50% of zero is zero. I just dropped a number there when I probably shouldn’t have.
The ideal split is of course 50-50, but I assume—
Only if the products sell. When in doubt, don’t be afraid to secure your time and effort with those upfront or ongoing fees.
Just to be clear: 50% of revenue?
It’s usually product revenue. You should always negotiate safety nets in the event that, for example, there’s a $100 plugin and someone decides that they’re going to sell it for $3 on Cyber Monday. There are clauses you can build into your agreement to make sure you’re fairly compensated for that kind of scenario. Similarly you should also consider how to manage royalties if your product gets rolled into a bundle or subscription package.
Let’s pivot and talk acoustics. Your first product in your new brand is a creative DSP tool, which is pretty different from your background in various subfields of acoustics. So, on one side you have this highly technical “corrective” stuff, on the other, this creative tool for musicians. I wonder how you see those things.
There is a lot of overlap. I’ve always been fascinated by sound and music. My grandmother was a very well known concert pianist from a London conservatoire who taught me piano via classical training. My father was a successful jazz pianist.
I was always more interested in the science of audio and behavior of sound in spaces, though. Whether it be processed audio through DSP or room acoustics, how sound behaves in a space, that kind of thing. Some of the tools I’ve made for clients in the past are things like kind of noise measurements in spaces, studies on performance of fan noise in enclosures within audio products specifically.
When you study room acoustics, you kind of have an appreciation for the space you’re working in. How the sound of hardware you’re analyzing or the performance of the audio processing is behaving in the space. Personally, having the acoustics background gives me a bigger appreciation of everything that’s going on in a space and how that affects products.
With acoustics and room correction on your resume, does that automatically mean you’re a reverb nerd?
I like reverbs. I’m kind of a reverb nerd. I was at one point heavily into 3D sounds, ambisonics, binaural recordings, that kind of thing. When I was in the UK I was out there taking a lot of measurements of big spaces for reverbs and for IRs and that kind of thing. And using sound field microphones to capture spaces for ambisonics and B-Format and above, Higher Order.
More of a room acoustics nerd, rather than just reverb, if that makes sense.
I saw offshore acoustics, hydrophone seismic stuff in the description of that YouTube video. Were you in the water with hydrophones or was it more on the DSP side? What was the deal?
The biggest part of the deal is I needed to pay off my student loan and working offshore is a great way to do that! [laughs]
It was oil exploration. Seismic survey, that kind of thing. Offshore in the North Sea in winter, which was an experience. I was on the technical side of the hydrophone arrays. I was working with the data that came back in from them. Setting up the filtering. It’s all about impedance matching. They make these huge explosions on top of the water and then measure the reflection from the surface with the hydrophones. It was my first real world hands-on engineering job and it was an eye opener, but it was great.
Even though I was on the technical side, there were still plenty of ropes and cranes and ladders and stuff that I don’t do anymore.
[laughs] Don’t do ladders anymore, that’s great!
Don’t do ladders anymore.
Matt was interviewed by Sudara.
Have comments on this interview? Ideas for who we should talk to next? Let us know on the JUCE forum.