McDSP - Analog Processing Box

McDSP founder Colin McDowell chats to Sam Inglis about his career at the cutting edge of plug-in development. Colin explains his approach to plug-in design, and talks about the revolutionary Analog Processing Box — a plug-in co-processor that employs configurable analogue circuitry rather than DSP.

Chapters
00:00 - Introduction
00:17 - Getting Started Creating Audio Products
01:08 - The Necessary Training
02:01 - Algorithm Development
05:40 - The User Interface
07:26 - Vintage Emulation
09:30 - 20 Years Of Plug-in Development
11:32 - How Useful Is Machine Learning?
14:10  The APB (Analog Processing Box)
17:07 - Analogue Electronics
18:31 - Component Tolerances
20:28 - Emulating Tubes
21:51 - Combining Signals
23:04 - The Capabilities Of APB
24:58 - Extending The Capabilities Of APB
25:46 - APB For Atmos
26:30 - APB For Mastering
28:11 - Other Analogue Effects
31:02 - Using The APB For Saturation
32:02 - Future Developments At McDSP

McDSP Biog
McDSP is an innovative Emmy award-winning Silicon Valley audio software and hardware company founded in 1998 by Colin McDowell.

McDSP technology can be found in pro audio plug-ins for popular digital audio workstations including Pro Tools, Logic, Cubase, and Avid live sound systems. McDSP audio algorithms are also used by companies such as Audiokinetic, Bioware and Microsoft, and in consumer products like the LouderLogic iOS audio player.

In 2019 McDSP introduced the world's first programmable analogue processor, the APB-16. The McDSP Analog Processing Box (APB) combines the flexibility of software plug-in control with the fidelity of premium analogue processing. Options include compressors, mastering limiters, transient enhancement devices, multi-channel and multi-band applications.

https://www.mcdsp.com/

Sam Inglis Biog
Editor In Chief Sam Inglis has been with Sound On Sound for more than 20 years. He is a recording engineer, producer, songwriter and folk musician who studies the traditional songs of England and Scotland, and the author of Neil Young's Harvest (Bloomsbury, 2003) and Teach Yourself Songwriting (Hodder, 2006).

https://www.soundonsound.com

Catch more shows on our other podcast channels: https://www.soundonsound.com/sos-podcasts

Creators and Guests

Host

Sam Inglis

Editor In Chief Sam Inglis has been with Sound On Sound for more than 20 years. He is a recording engineer, producer, songwriter and folk musician who studies the traditional songs of England and Scotland, and the author of Neil Young's Harvest (Bloomsbury, 2003) and Teach Yourself Songwriting (Hodder, 2006).

What is People & Music Industry?

Welcome to the Sound On Sound People & Music Industry podcast channel. Listen to experts in the field, company founders, equipment designers, engineers, producers and educators.

More information and content can be found at https://www.soundonsound.com/podcasts | Facebook, Twitter and Instagram - @soundonsoundmag | YouTube - https://www.youtube.com/user/soundonsoundvideo

Hello and welcome to the sound on sound people and music industry podcast with me Sam Ingalls. In this episode, I'm delighted to be joined by Colin McDowell of MacDSP. Welcome Colin. Thanks for having me Sam.

MacDSP is a name that if you're a Pro Tools user especially or any kind of computer user in the music business, you'll have known about for a long time because MacDSP were one of the first developers of third party plugins for Pro Tools TDM that I can remember back in the day. How did you get into that business in the first place?

Well, um, like many people who try to make audio products for a living, I'm a terrible musician slash audio person. So I, you know, I always liked music. I always liked engineering. I thought any way I could figure out to combine them would be great. And so I, at some point did land a job at digit design. I used to work on plugins and then, uh, at some point I just wanted to do it on my own.

How should I say it? I'm not the best keyboard or guitar player, but if you want a plugin that makes a certain sound, apparently I have skills to do that. I don't have any other skills, but um, those are the skills I have. But I guess back in the day, there wasn't much training available for people who wanted to get into plugin design.

How did you learn about it? So my undergraduate degree is in, um, I'm a double E electrical engineering with my specialties in signal processing. So I've done a lot of signal processing prior even to doing it for the music industry. So from Pro Tools, which had its, you know, one of the first sort of plug in platforms, I learned about that there's a specialized, software component you can build as a specialized, you know, algorithm to manipulate a signal.

I'm like, that's me. That's what I can do. This is great. So as I got that job at DigiDesign and I got to work in the um, plugins group, I sort of got to see how the plugins are put together and then I just, Boy, this would be a lot of fun to do it on your own and come up with your own algorithms and things like that.

And, uh, eventually I got make DSP started in 1998 and just have been doing it ever since. So in those early days, of course, you were coding for the DSP chips on TDM cards. And then at some point you've moved over to also writing native plugins. How different are the two from your perspective? Firstly, I think, um, any algorithm development should be a platform agnostic.

You should just be coming up with something that you think is fundamentally, sounds good, and is, um, not, uh, overly complicated. I think it's, um, interesting to me, not just in the audio signal processing realm, but in signal processing in general, most of the best algorithms can only be broken down into very simple components.

Um, yeah, maybe there's some math in there somewhere, but ideally it's like it's, it's not convoluted or overly complicated, so that's to start. Then your question, is it like an embedded DSP system or a native system? Those, if you've managed to come up with an algorithm that is, um, straightforward, it's broken down into, you know, easy to understand components, the creation of the algorithm, either it's an embedded DSP, or a native processor.

The workload, I think, is almost equivalent. Um, if it becomes complicated, then you need to back up and look at your algorithm and go, what was I thinking when I said, hey, six feedback loops into this thing and then we'll just like, randomly take an average and there it is. Oh, stop, time out, cut back. So, I think they're almost equal.

Embedded systems like TDM or native, the only thing that I think is different is that in embedded systems, you have to define all your memory space. Thanks. Which is just like it's like it's like cleaning your room or your studio, right? If you do it enough you get really good at it. You know where everything is You don't lose half your stuff every time you clean up.

So Managing your memory in an embedded system is kind of like keeping your studio clean then conversely in native You don't have to do that. So then sometimes you get sloppy and do something really dumb. Like, Hey, I thought I'd put all my mic cables in a piggy bank on the top drawer in the back room.

Isn't that smart? No, we need all the mic cables now. Oh, well, I'll be back. I crammed them out of this piggy bank. I'll have to break it open with this hammer. These are the jokes, people. So this process of coming up with an algorithm for a plugin, is that something you do almost on paper or in your head, or is that something you, do you have tools that help you do that?

Well, because I've been doing it for a long time, my wife would say it's all in his head because it's all he talks about. I think a lot of the times it comes down to, yes, in my head or just on paper. A lot of time on paper, it's like, oh, I wrote down some formulas. It's like some flow diagram, like, oh, the signal is coming in here.

It has this problem, or I want to get this out of it, or make it do the following. Then I might say, well, if I had one of these, and one of these, and the initial flow diagrams, um, you know, they're like, wow, it's really complicated. Surely there's got to be a better way to do this thing I'm looking at. And then we sort of break it down, and it's like a decent flow diagram.

Then we use tools like MATLAB. I'm mostly a MATLAB fan. I just like MATLAB. And we use that to maybe like simulate the flow diagram. I can put test signals into it, take test signals out, and kind of see, yes, that's what I thought it would do. Or, oh, that's terrible. Well, I'm glad I didn't waste a bunch of weeks making a plug in out of that.

Um, but then at some point I go, hey, that's pretty good. And then we take it and start building up a plug in in the UI. And, um, and when I say building up a user interface, I mean just like something that looks like this. Just totally awful, you know, hey, look a green box with just like, you know knobs from their plug in It's like every every prototype plug we make is like oh look I took the background of filter bank with the knobs from the 6060 input module and Just grabbing something just to piece it together and go but then we just play with it at the very end We'll do the user interface.

I mean talking about user interface there. How important is the user interface? Do you think in the Usability of a plug in. I think it's as important as the sound Um, I mean if the plug in sounds great, but it's difficult to use no one's gonna know because they want how to use it Conversely if the you know, user interface is great, but the plug in sounds like no one's gonna like it either You just have to have both.

Um, which I think it's good It's good because it makes people these days at least, you know, think about what am I making how the customer use it? What are the important features I want to make sure are? readily accessible or discoverable and some of the more tweaky things. Yeah, if it's not quite right in your face, that's okay.

But the, you know, more experienced user will discover them and then be able to, you know, exploit them on relatively quickly. So is that something when you're coming up with the initial algorithm, you already have in mind what controls you want to present to the user? I do, at least these days I do.

Generally, as I come up with the algorithm, I already have the controls in mind because, you know, the algorithm is supposed to do X and to accomplish X I need a a threshold and a, I don't know, sensitivity and uh, I don't know, something else, you know, and I'll have those in mind as the controls that's going to be.

And so it all comes together. And that's something else you mentioned about this to controls is sometimes you make an algorithm, you go, this is the algorithm and this is how it's going to sound. Okay. It sounds how I'd like. Great. Here's interface. Great. And someone goes, Oh, I'd really like to adjust this.

What? Yeah. I'd always like to adjust this part. Well, why? Because when I do this, this, and this PS, I'm like, you know, famous mix engineer, film, whatever, you know, TV person. Oh, So you know what you're talking about, and that's a common usage point. I got you. I'll add that control. Can do. Then you change it. So, sometimes there are things like that.

But it's good to have a control set in mind as you go forward with the algorithm. And I'm interested in your, uh, your take on vintage emulation, because a lot of the Mac DSP plugins are capable of recreating the sound of some vintage units, but you never seem to package them in kind of photorealistic vintage GUIs or anything like that.

It's true. I mean, I think, you know, like, uh, our most recent release, we definitely have tried to make some user interfaces that look kind of real, but they have lots of, you know, functionality you expect in software. That's, I think, kind of our, we're trying to find our balance point. But as far as like emulating vintage gear, I mean, Don't get me wrong.

I've spent a lot of time looking at a lot of, you know, other, you know, pieces of equipment that people say, Oh, this is the holy grail of EQ or compression or whatever. But I'm not sitting there going, Yeah, they use the 1024 or 586 diode with the resistor bridge, blah blah blah. I'm like, I don't care. I just want to know how does it sound?

Why does it sound good? What does the, what are the shapes of curves and things look like? And then just kind of come up with my own way of doing it. Which I think is way more fun. I don't know how to say it. And I think then that's, um, so I guess. I think if people want to emulate vintage stuff, I think it's great.

If they want to make something new, I think it's great. I just like making things that sound good. So I don't really care if it was made yesterday or, you know, 60 years ago. If it has some said people, oh, I love the sound of that. I just want to know why. Why does it have a good sound? I don't care like, you know, it's made out of tubes or solid state or I don't know.

What's up? Rat running around in a cage powering the thing. I don't care. I just want to know why does this sound good? Why does it work for people who make this kind of music or whatever? And I just want to know that because I think that's fascinating. I just want the sound. If I get the sound, great. It can look like whatever you want.

So in as much as you refer to this vintage gear, it's like, it's kind of a black box for you. It's like, as long as you get the transfer function in and out, then you don't really care what's going on inside. And in my prior life career, whatever, yes, black box design is exactly what I specialized in. I didn't really care what was inside of it.

And sometimes I didn't notice it. Yep. Just, you know, make it do what that does. Okay. That's kind of creepy, but I'll do it. And so that's what I would do. So how do you think plugins have improved since those early days? I mean, we're talking about well over 20 years of plugin development here. How have they got better in that time?

Um. I think they've gotten better because the, you know, customer's expectation has, um, has, has the bar has been raised. You know, I think that a lot of people are good at making good quality audio productions. And so they're, they're paying attention to like, Oh, what does this EQ or compressor sound like?

And they really actually kind of, no, at least to my mind. And so I think that the game has been upped for everybody, which I think is really good. It just means better products come out and that's fine. You know computers have gotten a lot faster. So I think people are are more I don't want to say it Earlier in this podcast I mentioned don't make things overly complicated But you know now you can make some pretty complicated algorithms.

So I think that some people are exploring more I won't say sophisticated, but you know Okay, more sophisticated ways of coming up with a way to process sound that's very useful to the the end user I think that's exciting. I I just and I think just uh, um I think there's enough people doing it now where it's like, Oh, like, that's like a, a thing people do as opposed to like, As you mentioned when the DSP first started I'd spend half the time explaining.

This is a plug in It's a separate piece of software that did not come with your DAW. You have to buy it from us No, we're not DigiDesign or Apple or Steinberg or whatever. Oh, yes, make DSP. Yes, that's us. Mm hmm. Yes back to the point Okay, so we make a plug in. It's a separate piece of software. You have to buy it So that's all done.

Here we go. Oh, yeah plug in I get it. Oh, oh great Well, fine. So, you know, that, that they, you know, just like there was an acceptance transition of doing audio production in a computer. There's also a, oh, I understand what a plugin is and why I might want some extra ones that did not come with my DAW that might do some of the same things, but differently or have a different sound.

In that way, it's a very mature market. So plugins can be better. So you were talking about how processor power now means we can have more complicated plugins for good or for bad. One buzz phrase that crops up a lot in terms of that kind of plugin design is machine learning, which a lot of companies claim to be using.

Is that something you've explored at all? Yes, we have. Um, I think the only thing that end users see these days is we have a module called shine. which is just like a high shelf, but like, it's a linear phase, high shelf, nice sounding EQ. Um, actually my, my daughter who, um, was in college, she's studying engineering like her dad.

And she said, I want to do some AI dad. I said, Oh, you do, huh? Well, here you go. Train this algorithm to learn what this kind of EQ shape would be and just keep training it until you get something that sounds good. She's like, yeah, I got this. It's like at the end of the summer, she's like, you know, I was like months later, she's like, dad.

Why is the computer so stupid? I'm like, it's called artificial intelligence, kiddo. She's like, yeah, okay, I get that Yeah, okay. It makes a lot of sense to me now. I understand. Got it. Um, Anyway, Sean also stands for secretly Helena's intelligently networked equalizer s h i n e Um, so yeah, so we we have explored stuff with it Um, but mostly it's to just see If we can come up with like different prototypes of things And then just look at it and see what the end result might be and if that's something useful to us And that's about it I think that the use of AI is great.

It's very interesting for like, you know, trying to train the computer to like, I don't know, master all your mixes the same way or come up with an intelligent EQ curve to suit a certain application. I think the only thing about AI that I don't like is it usually to have a system that's trainable and, and, you know, Smart enough in quotes to do those things.

It becomes a pretty complicated system. And if at some point in the training process, it's trained incorrectly, it's really hard to train something out of an AI. I mean, you can, it's just kind of a weird thing. It's like, you know, teaching someone how to ice skate. If you only teach them how to skate backwards, they'll never skate forwards.

Yeah. And one of the classic problems with AI is that thing called the black box problem, where you can. You can train an AI to do something, but then you've actually got no idea how it's doing it. So if you want to modify its behavior, what do you do? That is correct. That is in fact, you know, a huge conundrum, right?

You can train your Google, whatever the heck. Oh, it's Tuesday. Sam, don't you need to buy some new microphone stands? You know, the short kind, because that's what you need, right? Oh, oh, how does it know? And then you go do it. But you know, how does it know? Yeah, I know what it knows. It's there's some parameterization there and stuff.

But yeah, it's kind of like, ah, Yes, you are correct, sir. So, you have been exploring a different technology though, because back in 2019, you introduced a thing called the APB, which stands for Analog Plug in Box. And this is, for anyone who's not seen it, it's actually a digitally controlled analog device.

It looks to your computer like a DSP box. Plugin board that you've load plugins into protals or logic or whatever But in fact the processing is taking place in the analog domain. It's super clever. Sounds amazing How did you come up with that idea? Well, like most people with obsessive compulsive disorder if you think about something long enough at some point you go wait we could do this Really you think ah, what do you know?

You know, what we make on our talked about the algorithm, all our plugins have literally the piece of code is called a brain and the brain is like, that's what the algorithm does. And, you know, we made it so that we could have, you know, the brain can sit in an AU or a VST 3 or an AX plugin embedded or native.

Blah, blah, blah. At some point, someone went, maybe his name was Colin, I'm not sure. Wait a minute, this is like a self contained piece of code that has inputs and outputs and controls. You know, I don't even know what's inside. In theory, so couldn't I just take the guts out of what's in the brain and put it like, back into analog?

I mean, could that work? That would be possibly really cool, and really expensive, but why not? So, uh, yeah, we tried some of that, and some of it didn't work. It didn't work. I was like, oh, okay, that doesn't work. That doesn't quite translate. But the parts really did. And, um, Then we're like, well, wait, because if like, you know, imagine like that brain piece of code is like the same thing.

We're just talking to the same thing. We just happen to, the guts of the brain are just in an analog board somewhere, you know, maybe a little more complicated than that, you know, sample accurate synchronization and automatable and configurable and yada, yada, yada. But um, um, Yeah, I guess that you mentioned vintage gear.

People have like, you know, this religion about, you know, vintage versus new, digital versus analog. And like I sort of tried to state, I don't really care what it is. I just want something that sounds good. And wouldn't it be interesting if you took like, here's a compressor we've made before, but what if we took it and put it in the analog domain?

And we'll choose like, you know, some of the most expensive parts we can find, and just put together a signal path and go, how would that sound? And At some point, we're like, wow, that sounds really good. You know, to the point where you even take like, you used to have these silver APBs, that you'd not paint it or anything, and they were hot as heck.

And we'd take them, and people would go, hey, does this sound good to you? Well, it sounds really good, can I keep it? No, no, no, no, just a prototype, nothing to hear, sign this NDA, thanks for coming, see you later. And so, very long windedly, um, it just seemed like an interesting project to explore. It turned out really good.

I'm very appreciative of the people who've taken the time to check it out, because I do, in my completely biased opinion, think it's really cool. So have you worked on analog electronics before, or was this an entirely new development for you? The most analog electronics I've ever done was, you know, still when I was in college, which was, you know, okay, I'll admit it, back in the late 80s, early 90s, you know, before the internet, when dinosaurs roamed the earth, yeah.

But, you know, I like that stuff. My other hobby, some behind me, not in video, but I do have a lot of analog synthesizers. I grew up in Buffalo, New York, so my sort of knowledge of what is a good sounding analog device really sort of originated from analog synths, and why an analog synth sometimes is like, wow!

And a digital synth is kind of like, eh, hmm, not quite what I want, but it's cool. What's different about it? Gee. Um, but we did hire a few additional engineers who had a lot of hardware experience. Um, and, uh, and they've done a really good job and they're kind of like me, you know, when I told them the project is basically we're going to make an analog device.

It's configurable, controlled the software, and it's basically no budget. Just make it as cool as possible. Really? There's no, like, I have to skimp on this or that. No, because if we skimp on anything, they're going to just like, you know, Crucify us on the internet. So you just, you know, can we use 32 bit converters?

Sure can. Can we use gold plated Wemacap blah blah? It's like, yeah, I don't care. I mean, whatever, whatever dude. Yeah, I just just so that also was fun for them because it was like a I don't know. We just I'm just trying to make the projects fun here I guess when you're developing digital plugins, one difference is you don't have any issue with tolerances, component tolerances, for example, in digital.

In analog, that must be quite a major concern, I guess. It is. Um, I think that, um, we seem to have come up with some ways to mitigate that. Actually, interesting to what the tolerance is. So the APB itself, um, is self tuning. So you need to turn it on. You know, it has a bunch of calibration points all across its analog board and elsewhere, and it's just listening to itself all the time.

So, if it thinks it needs to, you know, adjust itself, it just does it. Because even when we, when we manufacture them, like the, the, the manufacturer right here in Northern California, Um, it was funny because the manufacturer's like, okay, we see what it is, it looks kind of complicated, how do we test it? Oh, we have this automated application, it'll test all the channels for you.

And you're like, oh, really? Yeah, you just. Click run test and it'll do it and then at the end it'll tell you if any of the channels have like some tolerance issue And you have to like, you know fuss with that. Yes your question. Yeah, the tolerance stuff is difficult Um, but I think we've chosen parts that are very reliable, but the apb units also are somewhat self tuning And it's worked out pretty good.

So it's, it's analog, but it's, it's not your granddad's analog. No, I mean, I mean, some people say, Hey, there are tubes in there. I'm like, no, there are no, there are no tubes in an APB because that would be breakable. Um, you know, or get really, really, really hot, like 16 channels of that. I think there's like 124 op amps, like in an APB 16 or something like almost dumb where you're like, okay, well that's, um, a lot of heat.

Yeah, it is. Well, how do you dissipate that? Well, you know, Just happened to have a dad who's into this kind of stuff, and he has a friend who's into physics, and he suggested, you know, emissivity, you can look that up later, and just understand what that is, and paint the inside a dark color, and make sure it has lots of surface area, make it a material, aluminum, that will, you know, absorb and dissipate the heat more quickly.

It's fun. But talking of tubes, it doesn't have tubes in, but some of the plugins you've made for AVB, Do emulate tubes. What's that's right. What's the process of, of emulating analog using analog. How has, how does that work? How is it different from doing it digitally? It's a new world. I'll tell you that sometimes it goes well.

And sometimes some of the occasional like, Oh, sorry, I have to rip my headphones off because something didn't feedback the way I thought it was supposed to. Um, so the idea is, is that when we have. Again, that brain that can do a certain sound. And we said, hey brain, we're going to make you analog now.

Brain says, fine. And then we asked the brain, you know, please, there's this model that we can put into the brain. This, you're going to be like a tube based compressor. And the brain's like, I got you. And then it'll configure the analog system to do that, in as much as it can. And then we're like, okay, cool.

Um, and um, there are, you know, it's not, software, which don't have like a blank slate, but we do have enough configurability in the analog chain that we can come up with a lot of sounds. And so we just sort of know what those ranges are, and we just try to massage them into, yup, this is about the tube model, or this is kind of a new tube model, because we had to modify this or that.

We put it all together, and um, what you get is what you see with the APB plugins we keep coming out with for the APB 16 So when you have, uh, say a compressor model, as it were running on the APP, is the whole thing running in the analog domain or is it just the gain reduction element that's analog and the side chain detection circuit is digital?

Good question. Um, I wouldn't call it the side chain circuit, but I'd call it that when the APB operates any APB plugin, there are two signals that go down to the APB box. There's the audio. Then there's like this control signal, which is like, okay. Yep. It might have like a side chain in it It'll have other things in it too, but they are married together So the sample accurate control over that converted analog signal as you go into the box the control signal goes off It says hey, we're doing this today.

Thanks for coming. And then the signal is going through the analog components You know, maybe there's clearly like, you know, maybe there's like a setup point Like when you instantiate the plugin and we set the configurable analog up But then after that the control signal being married to the audio signal We'll just, um, do what we want it to do.

And it does it in a way that's, you know, not just real time, but, you know, synchronized perfectly. So you come out the other end, there's no like a, Oh, I missed that peak or I didn't catch that. It's all in a line. So, I mean, what can you do with APB that you feel that you, you weren't able to do with, with a conventional digital plugin?

Hmm. Um, there was one interesting point to start with is that, um, Distortion. I know a lot of people get really excited about distortion or saturation or whatever that is. That's great. Um, I think it's a good part of what audio production should be. There should be some exploitation of that type of a processing element.

Um, but in the APB, It's all kind of like built in. It's like, yeah, I'm an analog signal path. I can only go this loud. If you go louder than this, I'm not gonna like clip digitally. I'm gonna saturate with some like probably pleasing curve that some hardware engineer named Glenn busted with for a year to say, it's gonna look like this.

Oh, that looks really good. Check it out in the scope. So it has some of those built in elements. So when you're making something in it, you can kind of go, huh? What if the user cranked up the makeup gain in the compressor? So what? It doesn't matter. It's going to saturate in some one of three saturation circuits, calibrated perfectly to the analog digital converter, so when I get it back into the DAW, they'll just have, you know, the analog, you know, distortion, and none of the whatever digital mayhem that may have occurred.

So, um, you asked me what the difference is. One difference is just that there's this saturation element of the APB circuit that's just kind of built in. Whereas in digital, you're like, ah, I can get as loud as I want. You can see on Power BI, it's like, it's like a pen, right? Some people like, a ballpoint pen, some people like a felt tip, you know, whatever it is.

It's just that, ah, it's different. Oh, I like this. Oh, I hate this. And so it's interesting just to take a different tool set, analog versus software, and go, well, what can we do with this? And then finding, you know, strengths and weaknesses and go, well, these are the strengths, let's see what we can do with these strengths.

And then just, you know, keep pounding that, you know, idea. I'm coming up with, you know, new compressors or different types of way to compliment a digital equalizer with a custom analog output stage, you know, just those types of things. It's fun. At the moment, all the plugins that are available for APB are fundamentally, mostly they're compressors.

What, what can and can't you do with APB? Could you, for instance, extend it, the concept to mic preamps or guitar amps? Ooh, um, mic preamps or guitar amps. Um, that is a possibility. Um, I think in the current configuration Um, that is designed for line level inputs, so we'd have to have a different flavor of some kind of input circuit to do that.

But don't think that we're not thinking about things just like what you said. So, um, hmm. Anyway, um, so there's that. Um, um, yeah, but mostly it's, you know, it's dynamic range control or saturation stuff. Um, and, and those are sort of the wheel houses of what the APB design that we've come up with does. I guess one other thing that you could in theory do with APB that you can't do with most analog devices is beyond stereo.

You could go to multi channel compression if you wanted to. Is there any call for that? Have you explored that? Uh, yes. There's also this thing called Dolby Atmos. Apparently it's kind of popular. I don't know, you know, more channels than your brain can keep track of in a mix. But yes, um, there are, you know, Several clients right now using the APB, they use it in their Atmos stuff because they want to have analog, but you know, configured across many channels to do something, and apparently it seems to fit that market well.

So yeah, we actually do have some multi channel requests. And I guess another obvious use case for it would be mastering, because mastering engineers like analog gear, but they also like complete recall, which is what you get with APB. Yes, um, there definitely is a mastering engineer, uh, interest in the product, um, and for exactly the reason you just said, or for most engineers in general.

Oh, I can recall it. Wait, what? It's analog and it's recallable. Don't forget all the presets and automation and the sample accurate control of all the analog. Yeah, right. Okay, that's um, those are all the problems in the analog and you solved them all. Yes, we did. Wait, what about delay compensation? Yeah, I solved that too.

Oh, I can't hate this thing. God dang it, I'll have to buy six. Fine. Um, yeah, um, yeah, recallability is uh, is huge. Um, in fact, actually, if my own personal self, you know, just um, A long time ago, maybe when I played in a band or two, we'd go into the studio and I'd, Oh, all this equipment is great. And we're doing our stuff.

And I had to watch somebody patch in some outboard gear for a thing. Oh, all those drums sound great or whatever. And I just would panic. Cause I'd be like, okay, I had the electrical engineering student in me is looking at what we've done. And I see how it's kind of repeatable, but not really recallable instantly.

So like, if we don't finish this, like today, We have today to get this done because if we come back tomorrow It might not be exactly the way it is and I'm not even sure I've kept track of everything that we've just patched together in the right order even though I'm trying to and that's You know, look the nervous twitch is coming back So, um, yeah Recallability is um That's a good point, Stan.

But I guess also, I mean, people love, you've chosen to focus on saturation and compression and limiting with APBs and stands. There are other things that people love in Analog 2, Analog EQ, for example, Analog effects and delays. I guess, could you apply the same concept in those fields? Ah, yes, you could. Um, and again, no confirming or denying what we might be tinkering with.

Um, but we'll see. On the topic of EQ, however, or delays, I'll point out these two things with the, uh, delay and EQ types of effects in the APB. First one, on the EQ, any curve that you want from any vintage equalizer you can pretty much nail in a software model. That's not the, what's in question when it comes to the sound, quote unquote.

It's the, what was the input stage Either prior to, during, or after that EQ curve has been applied to the signal. That is, you know, that's, to me anyway, that is like where the, oh, that's why it's different. It's not that I can get the, I don't know, scoop of a Pultec or the shape of a parametric from whatever.

It's that at some point it goes to something else. There's something else it goes to, right? That analog system that's around that equalizer that has a fixed range, you know, that if you get up too close, it's going to. Not clip, it's going to saturate in some way. What is that system? And so in those ways the APB is well configured to do that.

If someone really really really wants to have, you know, the op amps and the feedback loops to do the the EQ stuff. Yeah, you can do that. But then you get stuck into these fixed topologies where oh, it's only going to be a parametric. Oh, it's only going to be a shelf. Then like yeah, and also you're going to use up so many op amps.

On one channel if you put like, you know, eight of those together side by side, that's gonna be like the sun. So I'm, not sure still thinking about though. Um, the delay line stuff Um, you have stuff like, you know, your bucket grade analog delay or things like that. That's interesting But also even just if you have like a delay and imagine though, what is it about the analog delay you like?

Well, there's like the delay time part and that's sort of the slew of the delay time being longer or shorter Okay, we can kind of approximate those I think in digital, you know pretty well You But again, it comes back to what is the output stage? What happens in that feedback loop? Where in that system is the part where the analog component goes?

Ah, This is my jam. I'm stepping in And uh, yeah, so like maybe like in the feedback loop of a delay line. Oh, what if it went through analog? It actually would be really cool. Yeah, so if I cranked up the feedback and now the delay was going, you know bonkers But it kept going through an analog system as it fed back and went bonkers You That might be really cool.

Not that I would know anything about the delay for the APB that would have such a circuit. So for you, what people love about analog. EQ and delays and so on is generally the the analog enos pretty much all comes from Saturation and that's what you need something like the apb to do you can actually Emulate the other parts digitally and it's fine Yes, that's that's what I believe and not everyone has to agree with me and that's okay I mean, I think that you know, we're just exploring something that's kind of new and different and it's fun and people that like it Knock yourself out.

But yes, that is my Reasonably strong belief because I've invested a lot of time money and make DSP is self funded folks. Yes So, um, yeah, I I think it's it's it's a fascinating way to apply or combine analog and digital processing kind of a hybrid system The configurability of the analog system we've come up with is I think is really exceeded my expectations.

I'm very pleased with it and Yeah, we're just gonna keep making more Stuff that sounds like a good mission statement Well, we've had a few hints then about what might be coming down the line any any further things you can tell us about the future What's what's what's coming? I think that right now we're still rounding out some of the like your bread and butter functionality for the apb You know, I think we got A good pile of compressors, a couple of limiters.

We have one equalizer. I'd like to add a few more of those, cause that first one on the Royal Q, it turned out really good. So that was very exciting to work on. And, um, so we'd like to do a few more, bread and butter tools like that. And then from there, yeah, we'll go on exploratory, like the, Hey, how about like a delay system that has the you know, analog feedback path and things like that Um, some people have asked us about transient designers, which I don't think is unreasonable It's haven't quite had time to experiment with that.

Also on what's coming next to the apb Maybe you've heard. Um Climate change. Oh, also, um, global supply chain, crazy issues. So I, I'll, I will certainly say to anyone who asked, yes, of course, we've been affected by that because I hoard gear like for, you know, a hobby. I'm also been good at very good at hoarding parts for the APB, you know, but we, we've been sold out since October of 2021 and, uh, we're still back ordered, but, um, we have more on the way.

So I think some people say, hey, what's the holdup? I'm like, well, you know, when you say I want the best 32 bit, you know, converters on the planet, well, sometimes you've got to wait for them. Some developments with APB have not been as quick as I would like, and some of that I would put a little bit of blame on the global supply chain stuff, which is annoying.

But also, um, if you haven't heard, Apple came out with a new hardware paradigm. Right about the time we shipped the APB, Hey, the M1, it's gonna be great. Whole new hardware processor, you'll love it. Okay, so in the immediate future at Make. DSP, um, we'll be releasing, uh, native M1 support for the APB plugins, um, which is, I think, kind of cool, um, and was some work, uh, and then after that, I think we'll get back into the wheelhouse of making more APB plugins that will round out the feature set of what I think should be delivered with a purchase of an APB unit.

Well, if you've put down your deposit and you're waiting for an APB to arrive, I can promise you it's worth the wait. It really is a stunning achievement. Thanks, Sam. In the meantime, thanks very much to our guest today, Colin McDowell. Thanks for having me. You've been listening to the Sound on Sound People and Music Industry Podcast with me, Sam Engles.

Thank you for listening, and be sure to check out the show notes page for this episode, where you'll find further information along with web links and details of all the other episodes. And just before you go, let me point you to the soundonsound. com forward slash podcasts website page, where you can explore what's playing on our other channels.