Name Behind the Name: Al Smart – Smart Research

GET SMART

Andy [McLaren] Stewart: So Al, where do these famous compressors of yours get made?

Al Smart: Well, there’s a workshop in England and one here in Melbourne. The Australian manufacturing is done at Assemblease in Richmond, and I and one other person test that equipment in batches once it arrives here from the factory. We’re not a major volume outfit: for example in Australia we do batches of about 25 C2s every few months, which is great for me because, at that quantity, I’ve got the time to personally guarantee what goes in the box – I don’t have to delegate the quality control to someone else. At this scale, I can guarantee what I’m manufacturing and still deal directly with customers. I don’t want to compete with the large volume manufacturers like Focusrite or Manley, but I like to think Smart Research offers something that doesn’t really exist elsewhere in the industry.

AMS: The company’s obviously not motivated by money then, if that’s the case?

AS: I hope not. The company certainly wasn’t borne out of a desire to make money. I’ve been playing with electronics since I was seven and that’s what I’m still doing really. I know a lot of the ‘big guns’ in the audio industry started out the same way – and in many ways it’s a clichéd story – but I still love making this stuff, even though the workload becomes very intense at times. The phone rings all day and night too, and I can hardly bear to look at my email inbox sometimes. But the truth is, my love of audio and designing new stuff is what drives me, and it hasn’t got on top of me yet!

AMS: Apart from when you were seven, pulling apart electronics, when did you get drawn to the audio industry? The impression I get is that you worked for SSL for a long time. Is that actually myth or truth?

AS: No, no, that’s true. I suppose it all started back in the very late ’70s at the local eight-track studio when I was 16, and from that point on, school and study was completely displaced by the lure of studios and electronics. Soon after that, I was employed by Wave Studios in London where I was engineering and also doing maintenance. From there I moved over to an electronics company called Digital Audio Systems that was doing everything from rebuilding old Neve consoles to hiring out digital Sony equipment like PCM encoders and 5630s. I was involved in general electronics within the company; so for instance, I was taking apart second-hand Neves and resizing them, building custom monitoring panels, things like that.

AMS: So when did SSL enter the frame?

AS: When I was about 20. SSL was taking off at that point, with maybe 20 employees and the orders rolling in. By some fluke or other I met Colin Sanders – who had basically designed the original SSL on the back of napkins and fag packets at two in the morning during sessions in the little studio called Huge!, which he owned at the time. I got involved in one such session as an engineer, despite the fact that I’d never worked on an SSL before. I just pretended I knew exactly what I was doing and bluffed my way in! I then proceeded to overdrive every signal path I could find, as I fumbled my way around the SSL, repeatedly insulting the console for good measure to disguise my incompetence, not realising that its designer, Colin Sanders, was sitting next to me!

AMS: You didn’t know who he was at that point?

AS: No! He just sat there and very politely took the insults, knowing full well I had no idea what I was doing. But, somehow, as a result of that session I was offered a job with SSL commissioning [installing] new consoles. I think I’d impressed Colin by doing a head geometry line up on both their tape machines before starting the session, and demonstrating the kind of pedantic but technical view they needed for the console’s acceptance. 

At that stage, as an unknown manufacturer, SSL needed to satisfy very pedantic customers – like me – with their new console. During that session, I was very demanding of the studio, and inadvertently showed that I was electronically capable enough to fix any complaints I might have myself. He was also interested at that point in developing a multitrack recorder (we were using one of the few UK-made machines for the session, which was an A80 transport with electronics made by Colin Broad). At the time I was building a one-inch eight-track recorder, using new electronics and an Otari half-inch transport that I had machined new guides and bearings for, so I guess this was of mild interest to him also, although it never came to fruition. This was right when the company started to grow, and the E–Series was still really only a prototype, so it was essential to test them hypercritically in their final environment. When I commissioned the first 6K at Townhouse for example, I fixed and fed back over 200 faults on it that had been passed through testing at the factory. In the two years that followed I commissioned around 40 consoles including Townhouse, Abbey Road, Air Studios, SARM and George Benson’s studio in Hawaii. The job took me all around the world, which was fabulous.

AMS: You were supervising the installs, getting the consoles up and running?

AS: Yeah. So when you bought a console – and when the crates arrived – I would arrive soon after and attempt to make the thing go… properly!

AMS: And you had a few strong hands to lift the consoles into place I hope?!

AS: Yeah, well they were usually already in the studio by the time I arrived[!], although there were a few occasions when I got to see them dangling from cranes, that sort of thing. I remember wedging one in a stairwell at Good Earth in London once… the console had to take two turns down the stairs into a basement. In their enthusiasm to get the console in, the guys that were lifting it kept saying to me, “Mate, we’ll get it in there, don’t you worry.” After they stumbled down the stairwell and wedged it in tight, with no hope of retreat, we had to get a Kango hammer to demolish the building around it so they could move it in, which they duly did. That was a memorable install. Tony Visconti who owned the studio at that time was very impressed with the encounter!

THE BIRTH OF SMART RESEARCH

AMS: Can you tell me about the birth of the C1 compressor? The myth has always been that the SSL bus compressor and the Smart C1 are identical, and in my experience, the C1 is often called an SSL compressor. Did you in fact, as legend would have it, design the SSL quad compressor?

AS: No, no. What happened was that after I left SSL I went to work as chief engineer at Eddy Grant’s Blue Wave Studios in Barbados. While I was there I decided to build an outboard unit, but without the instability and output drive issues associated with those early SSL consoles. I loved the SSL Quad bus compressor, so I built myself a couple of them, but from then on someone else was always borrowing them – I could never get my hands on one, even for my own sessions!

AMS: That sounds remarkably like what still happens today…

AS: …so I just had to keep building more and more of them in the hope that one day I might be able to use one myself!

AMS: So you clearly knew how the SSL bus compressor was constructed, how it functioned and the details of its circuit then?

AS: That’s true. Obviously as an SSL engineer I knew all the existing circuitry. The C1 was based entirely on the SSL compressor and that was how the myth that I’d designed the Quad compressor developed. If you compare the C1 circuit to the SSL circuit, probably 50% of the components are the same. It’s the sidechain that’s the key to the way it sounds. The dynamic process is the same, which gives it its essential flavour, but the C1 had added sidechain components; different VCAs; and outputs. One other area I wanted to change was that of stereo linking, to make a wider image possible. There are different schools of thought as to how best to handle image shift in stereo compressors electronically, but by auditioning the different possibilities it was a clear choice to use a different method to SSL.

Then later came the C2, which was designed in Australia and based around an entirely different signal path to the C1. For instance, the C2 is completely balanced internally. I went right back to fundamentals with the design and also dealt with one of my favourite thorny issues: output drive. One of the things I wanted to do in the C2 was to further improve the output, as a powerful, discrete transistor output stage helps guard against loading problems; like long cables affecting the tone of the signal. Slew rate, headroom, and distortion are commonly far worse in real situations as a result of op-amp output stages than ever imagined in R&D or test departments. So I designed the C2’s output such that, for instance, live customers could drive a multicore without fearing what might arrive at the stage 300 metres away.

AMS: Presumably the C1’s design remains unchanged though? Or did you retrospectively ‘hot it up’ as you developed the C2?

AS: No, I left the C1 alone because it had an established reputation. That’s essentially why the C2 came about: having all these new ideas meant that rather than ‘improve’ the design of the C1, it was time to design a second compressor. The C2 is definitely a sonic improvement on the C1, although the C1 is still used and preferred by many of my customers.

CRUSH

AMS: Conceptually, can you tell me where the C2’s ‘Crush’ feature came from?

AS: The idea behind ‘Crush’ [a feature specific to the C2 compressor] was to extend the characteristics of the C1 into a more aggressive area. A lot of people were using the C1 to slam drum overheads and, in some cases, whole mixes – and they still do – and I wanted to offer more extreme possibilities in that direction while also increasing purity and resilience at the heart of the design. Fisher Lane Farm had asked me to investigate an obscure FET compressor they had; people were using frequency selective compression in modified Dolby A cards; talkback compressors for recording; and the areas of FET distortion and ‘stressing’ were just natural extensions of that trend. The rest of the C2’s design was aimed more at mastering clients who have a different set of criteria to recording and mix engineers. So noise floor, headroom, distortion, and a lower compression ratio of 1.5:1 were also on the ‘improvements’ list (the C1’s lowest ratio is 2:1). I also wanted to add ‘power fail bypass’, and the ability to meter in bypass for live and broadcast clients, as well as other things. Thus the C2 was born.

AMS: And presumably the C2 was trying to retain some of the characteristic tone the C1 had quickly become famous for? Or am I warping history here?

AS: Recording studios were routinely using the C1 as a mix-bus compressor, but they were also using it in extreme situations – and the C1 was great at that – basically it just added lots of ‘musical’ compression to things like drums and guitars. It quickly became renowned for its bombastic sound and I wanted to include that flavour in the C2 design, and take it further by adding the ‘Crush’ switch. What I’m dying to do now is cater to all those customers that want a bit more control over the ‘Crush’ flavour. In the C2, the ‘Crush’ control is engaged via a single switch: so it’s either in or it’s out – although you can obviously moderate it by the side chain settings you use. But there are other aspects of this type of ‘treatment’ compressor that I’m keen to look into further. That’s the plan anyway.

SMART TONE

AMS: How would you characterise the C1 and the C2 in terms of the ‘tone’ they impart? Are they ‘midrange’ compressors as some people describe them: by this I mean, do they appear to exaggerate the midrange the harder they compress in your opinion?

AS: Yeah, well this is a thorny issue in many ways. Neither one is electrically frequency-conscious as such: if you feed 100Hz or 1kHz into them they’ll compress those frequencies by the same amount, but I think the issue of ‘low-end tone’ is mainly determined by when you start compressing slow waveforms with a fixed side chain speed. At 100kHz the sidechain is nowhere near reacting within the length of a single waveform because the cycles are so short. But as you come down the spectrum you eventually reach the territory where the compressor’s reacting within a single cycle of a signal. When this happens, you’re beginning to generate distortion. If you slow down the speed of this reaction time to accommodate low frequencies, you won’t produce as many artefacts, but your compressor will become so slow that it will effectively be useless. The fact is you have to dive in and get your hands dirty some time, while manipulating these artefacts for the best sounding result.

AMS: So clearly it’s better for a compressor to act on one or more cycles of a given frequency rather than a fraction of a single frequency.

AS: Exactly, but the trouble is, you’re trying to impart a ‘one size fits all’ response across the frequency spectrum, which is a physical impossibility. To put this in perspective, at 50Hz one cycle is 20 milliseconds, and the slowest attack time we have is 30 milliseconds, so for most real world situations there will be significant harmonics being added below midrange frequencies.

IN THE WINGS

AMS: Can you tell me about any new gear you have planned – you’ve already alluded to a third compressor I do believe?

AS: Oh, well, I can, but it’s dangerous in a way, to sit here and mouth off about things I’d love to do because they might just be wishful thinking. But certainly I need to explore the compression issue more. There’s definitely at least one more compressor to make.

AMS: Can you give me a ‘potted version’ of the design philosophy behind this proposed third compressor as you envisage it now? Is it based on the shortcomings of the C1 and C2, or is it inspired by what they have already achieved?

AS: I guess, as I indicated earlier, I’m hoping to expand on the ‘Crush’ feature. There’s been very positive reaction to the C2 since it was released and the area is wide open to investigation. I’d really like to explore the grungy region of compression more – the strident kinds of gain reduction.

AMS: It’s certainly true that as the world moves inexorably into the digital domain, analogue compression is one of the crucial flavours that remain highly prized.

AS: And yet everyone’s different. I’m still confounded by who buys the C1 and the C2: both compressors are used in a variety of applications and the overlap is enormous, which surprises me. For instance they both get used in the ‘invisible compression’ arena – like mastering – so I guess even in that area the musical flavours they impart are beneficial to the program information.

AMS: No question about it. And probably the reason why you find it hard to distinguish between C1 and C2 owners is that most people require both at some point, depending on what they’re doing – different aspects of a mix require different styles of compression… and that’s true of mastering as well.

AS: That’s true. And as an engineer, what you probably can’t afford to do is end up getting typecast for the style of compression you apply. Variety is important.

RARE ARTEFACTS

AMS: How do you judge what’s good and bad about an artefact when you design your equipment? Obviously no test measurement is going to actually say, ‘This is good for rock ‘n’ roll, that’s bad for rock ‘n’ roll’. Is it once again down to that old chestnut of using your ears?

AS: In many ways it is, certainly. Building equipment is no different to mixing a record in many respects. You have your preferences and tricks you’ve picked up along the way. You put in stuff you like to hear, you tune out stuff you don’t. But there are two different areas to consider here. I can always say that lower noise, lower distortion and better headroom are improvements as a backdrop to other processes that you then use to superimpose character. I like most equipment to give me a choice between passing exactly what went in, or altering it along the way: particularly consoles. For instance, a large valve console seems an insane idea to me because, as the main infrastructure of your studio, you can never escape the constraints of its sound. But don’t get me wrong; I love the flavour of valve gear, I just don’t like the idea of it being an inescapable flavour in a console.

AMS: So, given that ‘flexibility’ seems to be high on your list of design criteria, will your third compressor design be more versatile than the others?

AS: Actually, I’m probably going to contradict myself here. In the area of outboard gear I think I’d like to take the shackles off and go wholly for a particular sound. I see lots of exploration possible in that area, because analogue flavours seem to be more important than ever. Even in mastering, people are using analogue devices more and more to add ‘flavour’ to recordings that have none.

AMS: And flavoursome analogue gear is what strikes a chord with people these days, perhaps more so than ever before. I guess, ‘technically perfect’ gear doesn’t always equate to ‘musically perfect’ in that sense?

AS: That’s right. The imperfections are often what give equipment their flavour. This point came up in a conversation I had last week with Mark Stent at Olympic Studios in London who’s thinking about a custom made console so he can pre-mix and run projects in parallel to his existing G series SSL. The G’s a crucial part of Mark’s sound: he likes to overdrive it because he loves the musical distortion it produces – if you listen to a Bjork or a Massive Attack record, the sound is astounding and that’s one of the things he does to get that. So to that end the new SSL AWS900s or Sony Oxfords can’t really help him, as it’s the inherent G crosstalk and breakthrough that’s a blessing for him in that scenario. Again, actually lots of FET distortion… I’d say compressor ‘number three’ will work along those lines in all likelihood… but I’m not promising anything!