Paul Northfield has been engineering hard rock and progressive rock since the early ‘70s. Paul has engineered or produced 247 albums and his credits include many of the pivotal recordings from bands including Gentle Giant, Rush, Asia, Queensrÿche, Suicidal Tendencies, Infectious Grooves, Porcupine Tree and Dream Theater to name a few.
Derek and Paul met when they were both working at Advision Studio in London’s West End in the mid-seventies. One of London’s leading recording studios at that time, clients included Emerson Lake and Palmer, Yes, Slade, Paul McCartney, Kate Bush and many others. Derek left Advision in 1976 to form Survival Projects, based in a workshop in the Yes equipment warehouse in Notting Hill. He originally designed and built audio effects for Yes, and later the customer list expanded to include Queen, Pink Floyd, AC/DC, ELP, and many other bands, as well as the BBC Radiophonic Workshop and recording studios in London, Paris, New York and Montreal. In the early 80’s, Derek was the lead design engineer for the Avolites QM500 computerised lighting console which had its debut on the 1984 Michael Jackson tour, and was a mainstay of rock concert lighting throughout the 80s and 90s.

Paul emigrated to Canada in 1978, initially working as a resident engineer at Le Studio, Morin Heights, and subsequently pursuing a successful career as an independent engineer and producer, which continues to this day.


Listen to them discussing the thinking behind the Anatron here:

Okay, so, what is the Anatron?

Derek: It is an analog audio effects processing unit which is digitally controlled.

There are a lot of devices on the market, so how is the Anatron different from other effects units?

Paul: So, when we say, ‘What is the Anatron?’ (the name is from Analog Electronics) it’s basically a good, high quality stereo filter, and Derek will probably tell you a bit more about the technology behind it. The fact of the matter is, it’s controlled completely by a digital LFO and by digitally generated control voltages.

What that means is that you have the classic analog sound with extraordinary control in the digital domain. And with the new technology, that means being able to run it from your iPhone or from an iPad and having much more control than you could with any other system. So, that introduces the ability to use the accelerometer on your iPad to control it, and to be able to control a lot more parameters with one hand. You can essentially, with two X Y coordinates and foot pedal, you can control five things simultaneously with one hand and your foot which is really not easy to do with a traditional analog effects unit.

So it’s not so very different as far as the basic idea is concerned, in as much as it’s based on filtering which has been around for a very long time, but that’s a well-loved kind of effect. It’s perhaps the soul of a lot of the traditional subtractive synthesisers because without the filters, the tones that any of these synthesisers make are uninteresting for the most part and extremely limited.

After not seeing him for a very long time since I moved away from England, when we finally did get together we had a spontaneous conversation about the whole idea of some of the work that he had done when he was around Advision and when he was working with Alan White, and made him a drum synthesizer. He mentioned that of all the things he worked on, his filter – which he designed as part of the synthesiser – was one of the things he wished he had done something more with. So, when he said that, I immediately said to him that I thought that was a very interesting proposition because today we use a lot of virtual synthesis, stuff that’s done inside computers. Modeling and virtual synthesis is spectacularly successful in one respect and gets better every day, but there are certain areas where it really struggles. The finer points of things like resonant filters being some of those.

A good example is that you can buy a virtual MiniMoog that will run on your iPhone for about 14 bucks and it’s remarkably good, but when you compare the nature of the filters with a real MiniMoog, one of which I own, you find that, as good as they are, there’s something about the originals… The term “organic” is often used – and often over-used… but it’s probably to do with the random nature of things when they are in an unstable kind of points in the extremes, which resonant filters often are. So, consequently, I’ve always felt that particularly in the case of a filter, the an analog ones still have a huge amount to offer, but at the same time those old filters have very rudimentary control functions, and now we have moved into an area, into a world where we have extraordinary computer control over things, and the idea of combining computer control with something as simple as a classic analog filter becomes very interesting because you can – aside from being able to implement a lot more control functions – you can constantly upgrade them and improve them, and tweak them with software while still keeping a fully analog signal path.

A lot of the things that the Anatron can do, you can do with other equipment, but invariably it involves a computer, work station software, and a lot of communication software. For the most part, anyone that is doing it that way will be hiding behind a screen playing with a computer. And one of my feelings – when I talked to Derek originally about the project, when he became enthusiastic about actually doing this and turning his old filter into something more contemporary – was the fact that in music now, one of things that really separates people who play around with music (which you can do to an extraordinary degree now with loops and garage band and autotune), where you can make very punchy powerful music, but not necessarily be much of a performer, but performance is really the thing that separates the men from the boys… When you have a communication with your audience.

When you use things like filters and electronic effects, quite often traditionally a lot of electronica has been done with people standing behind walls of synthesisers or computers. For the most part they could easily just be running a tape and you wouldn’t know, whereas when you’re playing a guitar or something like that there is a very visual connection with the audience and so having some radical control over something as fundamental as a filter means that you can interact with the audience and have real amazing control over the sound of your instrument at the same time. That’s the whole idea behind it. For me, I was sort of egging Derek on from the beginning to say, ‘Think about performance, think about performance,’ because you can do a lot of stuff on a computer, but this means that you can do it all live and produce effects that would be very, very difficult otherwise. So, that’s, maybe a long-winded summary of my point of view of what the Anatron is. Maybe I should bang it back to Derek.

Derek: Okay, well thanks for that Paul. That was a pretty good exposition in a lot of ways. You know, I guess in about the late 1990s, the whole conversation out there was that all of this analog stuff is obsolete and no longer necessary. It was much easier to handle and much cheaper to just do it all in software and do it with digital processing. So, I kind of put the whole thing to one side and that point, and then around about that time when you came back to England and we got in touch about two and a half years ago, I was beginning to think that actually there is a bit of a ground-swell towards the idea that modeling things digitally is all very well, but at the end of the day, no matter how good the model is, it’s still only a model. And there are people searching for authenticity, and part of that is – however good the digital emulation of an analogue process – it still is missing two things. One is that horizontally in the time domain, an analog signal is actually continuous. The other thing is, that vertically – in the amplitude domain, the signal level domain – again it’s continuous. Whereas in both those domains, digitally it’s chopped up into slices and you just don’t get the same effect.

Now, I’m not out to convince anybody of that – people have their own views, but there are definitely grounds to support the idea that the analog way of doing things has a completely different character. And I’m personally convinced of that, or else I wouldn’t be doing this in the first place if I just thought it was all smoke and mirrors. On the other hand, the downside of the analog domain is that essentially, most of the time it’s down to getting the settings you want by turning a bunch of knobs, which some people like, as a thing in itself.

The thing about turning knobs is that when you get a sound that you like – if you want to come back to that exact same effect tomorrow, or next week or next year, you’ve got to get those knobs back to the same position, and you might have even just one knob off by a millimetre and it’s not going to be the same sound. The other thing is, like Paul said, if you’re a performer, you do one song, you have one bunch of settings for that… you can’t then take a two or three-minute time out while you re-adjust all the knobs on your console to set it up for the next one. So, what I felt was that it would be great if the signal path was analog, but it could be controlled digitally. First of all so it could be really precise; and secondly so you could store it and recall it at any time; and then thirdly, the other thing is that we use the standard of the MIDI system so that it’s completely versatile for what you control it from. Someone could control it from the knobs and faders on their master keyboard or a MIDI controller panel as long as they set them up for the appropriate channels; or it could be done from software on a custom application; or it could be done from a midi track on something like Logic Audio; or we’ve now developed a control surface that runs on an iPad. We used a fantastic app called MIDI Designer and I’ve set up a whole control surface which controls all the parameters so people can do it that way and that’s really accessible, and that’s something that a whole bunch of people would like. One other aspect is that we could kind of intermodulate and matrix all of the different parameters so that they can be synchronised with one another… but I think that’s about enough for now. Anything you’ve got arising out of that?

Well, first of all, I can hear there’s a huge bit of experience behind this product and the performance aspect that Paul was talking about sounds really exciting and Derek, what you were saying about the accuracy, so you have that creative side, but also the digital side helping you be very, very precise. It just sounds like a really cleaver combination. Could you say a little more about it, what does it look like? What’s the physical format of the unit?

Paul: Well, at the moment it’s built in a 1U 19 inch rack mount unit. The idea is to produce it as a 500 series rack module which is a smaller format that is commonly available in studios. But at the end of the day we could probably get the size down to a smaller format that would be like a large pedal if necessary, but that would depend mainly on the kind of demand or interest in it, because you can even control it with the right adaptors over Bluetooth or Wi-Fi. One of the things that’s nice is that you could have it as a rack mount in a studio environment or in a PA environment and have it controlled by a performer or dancer, probably at this point with an Apple Watch, or with a motion sense accelerometer on it which makes sounds change or be manipulated just by motion. So that’s kind of interesting and it works particularly well with guitar players – for instance where you don’t necessarily want to be tied down to one spot on the stage and you could – just by moving the guitar, rotating it or something like that – you’d be able to trigger the filtering in a way that’s very expressive.

The filtering on the Anatron is much more refined than you get in something as simplistic as a Wah-wah pedal – effective as that is, because it’s probably one of the most used effects of any for the guitar, and has historically been from as far back as Hendrix and it’s been used in many ways. The Wah-wah is actually a very simple form of filter pedal, and has been a fundamental part of music for a long time, but the Anatron pushes it to a whole other level, what with the combination of being able to take advantage of all the new technology that is around that allows you to be disconnected so you can run around on stage and have a sonic impact with what you’re doing without having to touch anything.

I mean, in talking with Derek, we’ve had the fantasy moment of where we have this running Wi-Fi and Bluetooth – which is merely at this point just connecting up the right boxes that already exist rather than inventing something new – but I would love to see an ice-skater going into a spin while she’s controlling the filter because you’d get a very strong sense of ‘wow, this could be amazing for visual dynamics and communicating with the audience.’

The other thing with this box is that it’s fully stereo and besides being able to be used by a guitarist as a sophisticated Wah-wah pedal… for dance music and rave environments, having a fully high-quality stereo filter across the entire mix allows you do to dramatic effects that could also yet again be done with a lot of visual components on stage. So, it’s a stereo unit that is completely versatile: it’s stereo insofar as the two channels can be independently controlled, left and right – or completely synchronized, left and right. You could feed a full mix of stereo into a PA and have it do some radical processing on it, all in real time with no time lag or anything, as is often the case when you’re dealing with a computer environment and stuff like that. Needless to say, it’s a really fun thing to play with. Back to you, Derek.

Derek: Yeah, I think that covers a lot of things I would say in answer to that question. One of the things that was just occurring to me, although we’ve focused so far on the filter – which can be high pass, low pass, band pass, or notch filter, or an all pass filter, which reverses the phase around the filter turnover frequency – there’s another whole stage on each channel before you get into the filter. We put the signal path through a valve – or “tube” as they call it on the other side of the Atlantic – and so straightaway we’ve immediately got the characteristic “tube warmth effect” which is a kind of very subtle low-level octave resonant harmony of everything else that is in the signal, which a lot of people really appreciate in itself.

And then from that basic clean but warm sound you can take it up all the way, to the kind of extreme tube overload sound. Once again that’s programmable over MIDI, and we can modulate that with the envelope of the signal, or envelopes generated from MIDI note messages or from the low frequency oscillator. And because the signal is then going into the filter – if you’ve got some type of clean signal like a flute sound or a very clear voice, if you put that into the filter it doesn’t have a lot of harmonics to work with, but by putting it through the valve first and varying the amount of overdrive you’re giving to the valve, then you can vary what the filter has to work with. Is there anything else you’d like to hear about that or anything that I didn’t cover that, you know, you’d like to explore more?

Yeah, I’d love to hear more of your plans. Like, what’s next in expanding the unit? It already sounds amazing. Anything else in the plans?

Derek: As far as the hardware side of it, I think for this particular iteration (and I have a lot more ideas going into the future), what we’ve got in the hardware side of it is what we are going to put into the production version. There’s a couple of things on this module which are still controlled from knobs on the front panel that I’m intending to actually bring through the digital control domain so that we’ve got a knob-free unit. One of them is a knob which controls the mix of the wet and dry signals so you can take it all the way from not affecting the signals at all through to being purely what’s come out of all the processing or any mixture of the two, in between. We are intending to make that digitally controllable.

Another thing is that we’ve got the ability to put a key audio signal in and use that audio signal to modulate various aspects most specifically the filter frequencies. So, you could put another signal from, maybe another track within the the mix, and then that will actually modulate how the filter is working. Again, we are controlling that with a couple of knobs on the front panel and we are going to make that controllable through the MIDI system on the final product. Apart from those two, that covers it from the hardware point of view.

Obviously we have a microcontroller in there which is processing the MIDI input data stream and turning that into the control voltages which regulate the analogue side of it. Now, I am committed to making the source code for the firmware available under an open source licence which will mean, for one thing, if there are any bugs in the system you don’t have to wait for me to issue the next release to fix that, there will be a community of programmers who can take a look at it and can take it on. More to the point, if you have some ideas of ways that you would like to expand the capability of it, that’s something that you can take on. If you happen to be a coder, you can take it on yourself. A lot of musicians are not necessarily software coders, but they have someone within their community or circle of friends who are programmers who they could turn to. Even if they don’t, I’m envisioning that we are going to build up a community around this. I’m planning to release the source code and have a repository on GitHub so that enhancements can be made and if people are entering into the spirit of this, those enhancements are going to be made publicly available.

You don’t worry about people ripping off your ideas?

Derek: Well, people are going to rip of my ideas no matter what I do. I mean, what do you think about that Paul?

Paul: Well, it’s not easy, it’s a multidisciplinary environment. You designed the hardware and the software, and it’s certainly possible that someone could go to a lot of trouble and copy the ideas, but it takes a lot of effort and a lot of work, and you’ve already got a finished product. I think that there’s always a possibility that people will do that. In the meantime, you’re going to get yours built and hopefully a lot of people will have it and be playing around with it. More power to anyone that wants to take it on, secondhand if you like.

Another thing with this box is that you’re intending to bring out all the six control voltages that govern the main parameters, so that musicians can use some of these to drive traditional modular synthesisers and such like, which is actually very handy because people that have modular synthesisers, they may find it gives them a whole new generation of modulation capability which is no small thing in itself. I just think this needs to exist, out there!

Derek: Yeah.

And talking about people that are going to be using it, as you’ve both been describing it, I can imagine it not just musicians, producers, but also artists, people who work in theatre. Who do you think is going to be your core demographics? What’s your target user for this?

Derek: Right, okay. Several things. One is, we’ve talked about performing musicians and we’ve talked specifically about guitarists, but I see it as also something that would add to the repertoire of keyboard guys, although they probably already have a palette to work with which goes a long way – but I think this could expand their capabilities. Definitively drummers, possibly vocalists, certainly I’m up for looking to see what it could do for other acoustic musicians. So that’s it for the live performance side. I think it has a lot to offer to people, whatever they’re playing.

And then, there’s the whole electronic dance music production scene.. I’ve had a lot of positive feedback in that area as something that can both give a bit more bite to the kind of things they’re doing already, and can kind of expand the sonic palette that they’ve got to work with and can expand the degree of controllability that they have available to them. You know, improvising with it in a real time way. The same thing goes for DJ’s.

Then there is the whole recording studio environment as well which I think it definitely has a lot to offer for. Going right from what we’ve been talking about so far with extreme effects through the fact that you can also use it in a much more subtle way so perhaps you could say a bit of how it looks to you, Paul?

Paul: Yeah, well I mean, because it’s a high quality unit to begin with, so it’s got a nice natural sonic character. The fact that you can run it driven from a workstation means that you can do many of the effects that you could probably have done in software, as we go back to the original argument about the value of a piece of analog gear doing it in a traditional analog way sometimes still has an edge. So, you can use it—like I remember when you were here and we were trying it out – feeding a kick drum through it and envelope triggering the filter to sort of give me a synthetic low end to the kick drum because the resonance of it goes so low you can actually bring in resonances that don’t exist in the original sample.

So you can achieve some substantial effects with it in that manner and also, or you can use it in a very soft and subtle way which gives you a possibility of doing some resonant equalization that’s relatively subtle, yet distinct. There’s not so many actual resonant equalizers out there that are used in the recording process. Most of the digital equalizers out there are very clean and often with an emphasis on precision, accuracy, and low phase shift, whereas this is not that kind of device. This is intentionally a sonic bending device, so the whole idea is that you can use a highly resonant equalization boost that is then blended in very subtly from the blend of the input and the output.

It gives you a lot of different options and the fact that it’s dynamic and it can respond to the input from an envelope point of view or from an oscillator point of view, and the fact that we also have the square wave and the sample and hold functions on it now which gives it a very musical rhythmic quality that opens up new opportunities. While you can do these things often within the computer environment, sometimes it takes quite a lot of work to set it up. So, to have one box that you can feed stuff through and play around with until you get some of those extraordinary inspiring qualities. It’s a fun toy to play with, that’s for sure. So, I can see it in the studio as being a very useful box that you can turn to when you want to give something a little more mystery, or when you’ve got something that’s a little flat sonically that you want to have more movement and more harmonic texture to it. So, it’s very useful in that way.

It sounds like an absolutely brilliant blend of traditional and cutting edge technology. Let’s just dig a little bit deeper. We’ve got a little more technical questions that are coming out to you from a producer.

– I think they’ve pretty much answered them… But I had one other question which is: I was wondering where these units are going to be built, or how are they going to be put together?

Derek: Well basically what I want to do, I’m looking into the logistics of all of that. I just want to make sure that I can produce them so that they are as reliable and as affordable as I can. So, I’m looking around various options for manufacturing around the world. At the moment we’ve got two functioning demonstration prototypes and as Paul said, these are in 1U 19” rank cases; he’s got one and I’ve got the other. I’ve had the thing reasonably well engineered, but I constructed them myself. What we are looking to do is to get to a point where we can do a mass production run and as I say, go over to the format of a 500 series rack unit which is going to be more compact, and more affordable to manufacture.

Great, so, to wrap up guys, could you just give us a summary, top three each in bullet point form of what you personally feel most excited about with this unit?

Derek: What I feel excited about is that all of the feedback that I get from everybody that I demonstrate it to. First off, they say it sounds really good, secondly it provides effects which are really subtle and really controllable and thirdly, it’s something that they can endlessly experiment with.

Sounds amazing. And for you, Paul?

Paul: One of the things that I think is fun with it is that we’ve taken the soul of a synthesiser, if you’d like, and put it in a very simple box and you can feed any audio into it and you can manipulate it in quite a sophisticated way that is immediate and very tactile with all the combinations of the control that you have from across MIDI and particularly when you use an iPad and MIDI Designer. You can literally with one hand probably control six of seven parameters simultaneously using an iPad and a foot pedal and so you can achieve real time performance effects that are really fun and dynamic. So I think that this aspect of it is fun.

The fact that it’s a resonant filter… and one of the things that is really enjoyable about playing with an old Moog is feeding noise through it and resonating the filter and then modulating the filter with noise and with other things to generate some really extraordinary effects that are almost water-like sometimes… It’s really just a filter running at the edge of stability which is an area that the modeling doesn’t really touch and so I find it a really interesting and fun tool to play with like that. One of the great things is that it’s great to put your guitar through and to put a mix through. Taking a really crude keyboard sound and giving it a lot more depth and texture and you can do it subtly or radically. You can also take a mono-structure and give it a substantial stereo effect. So, it gives you a lot of things to play with. You can take really rudimentary keyboards, cheap keyboards, and feed them through this and get substantially dramatic improvements over their sonic possibility with very simple input to it. So, that’s what I find fun about it. You know, anytime you get a box that you just kind of get lost in, in a good way, it’s fun, it’s inspiring.

Well, if you’re a creative person, the opportunity for endless possibility and fun, I don’t think it gets any better than that. It sounds absolutely amazing guys and all the best with it.

Derek: Thank you.

Paul: Thank you.