A pal discovered and sent me this article, which first appeared in Hi-fi Review magazine in April 1989, floating about on the internet. I apologize for each and every mistake you might find within it. It appears to have come from a scan of the printed article and bears evidence of the OCR software having trouble differentiating between ‘1’s, ‘I’s ‘J’s and with punctuation marks. It was also written early in my career – before I had developed my grammatical pedantry and before my writing style had matured. I reproduce it here, though, because there is some fascinating and insightful content within its 5000 or so words.
Naim supremo Julian Vereker talks with Malcolm Steward about his company’s past, present and future
At various stages in the development of the British hi-fi scene different company names have seemed to push themselves firmly into the forefront. During the past two decades none has had a larger share of the limelight than those of Linn Products and Naim Audio. In fact, for many years those two concerns seemed inexorably linked: both shared much the same fundamental philosophy and approach to manufacture, and these links formed such a bond that the companies actively promoted each other’s products. It was a seemingly perfect set-up bringing mutual benefit to both concerns.
Beneath the surface, however, the relationship was not as idyllic and blissful as it appeared outwardly to be. Many factors eventually contributed to the dissolution of the corporate entente. The most publicised of these was Linn’s decision to manufacture its own amplification. Meanwhile Naim, who could now offer an alternative front end to the Sondek with the NAT 0l FM tuner, were expressing heretical dissatisfaction with the turntable: just prior to the often inaccurately reported split I asked Naim what new developments it had in the pipeline. The reply was that there were plenty of improvements it could make to its amplifiers but that these were having to be held back because they would simply show up what Naim perceived as deficiencies in the LP12′s performance. This was just before the company released the first of its infamous Armageddon modifications to the Sondek.
Now entering its twentieth year of trading, Naim Audio’s position in the UK hi-fi scene has shifted from one of being a renegade upstart kicking against convention to one where it verges on being part of the ‘establishment’: it is not yet perceived with the same reverent conservatism as, say, QUAD, but it is on its way to a far more general and similar acceptance. While this might be deemed as a positive benefit there is an obverse side to the position which is less palatable. The press, speaking in general terms, operates on a cyclic pattern which often sees it playing champion to the underdog, placing it on a pedestal until it becomes ‘accepted’. Once this acceptance is achieved the game changes direction and it becomes fashionable to knock the new king of the castle off his perch. The usual excuse for this behaviour is quoted as being “a desire to appear unbiased.” If one looks at the magazines of the late seventies, every reviewer and his aunt seemed to own and express a sincere respect for Linn/Naim systems, but, almost overnight, this became ‘politically unsound’ and many writers changed tack and started to investigate other combinations in order to preserve their credibility. Some now seem to change their systems with the regularity that most of us reserve for our underwear.
However, none of this has seriously affected Naim Audio. Despite remaining in favour with only a handful of journalists it has progressed very satisfactorily without the capricious enthusiasm of the main body of the press. In January I decided to pay the company’s Managing Director, Julian Vereker a visit at his Salisbury headquarters to discuss Naim’s past, present and future.
I began by asking him the inevitable how’d you-get-started question.
Malcolm Steward: What preceded your entry into hi-fi manufacturing?
Julian Vereker: I earned my living until 1967 as a racing driver racing Minis.
(I interjected that although I knew of his involvement in motor racing I hadn’t previously realised that this had been a professional venture.)
JV: Yes, I earned lots of money doing that: if you won everything it was very profitable.
MS: So how did you progress from tuned-up Minis to hi-fi amplifiers?
JV: I retired from racing when I was twenty two after a very good year – I’d won sixteen out of twenty three races and been placed in the rest and so it was a good time to quit and sell my car. The car was well known and its sale realised about £650, which in those days was a sizable sum – enough to live on for a couple of years. I retired to Salisbury and for some time I didn’t do a great deal – I worked for Downton Engineering and Janspeed for a while but then I became bored with cars so I went to live in a pub.
While there I developed an interest in 8mm film and putting sound with it, and went to see a local company who made a sound system to go with my camera. I bought all the necessary bits and pieces and asked about the tape I would need. The guy told me I’d need magnetic tape with holes in it so I said “Fine, give me ten rolls”. When I found out that it cost two pounds and ten shillings a roll I told him that it was too expensive and that I’d make my own. Suspending his disbelief that I could actually do that he said that if I could make it I could supply him with it. So I went home and modified a sewing machine to do the job. I went back to his office a week later to show him the tape I’d perforated and the owner placed an order for me to provide him with 100 rolls a month. The venture proved so successful that I built an automated machine to do the job.
Around this time I started to come into contact with a lot of musicians through my girlfriend who was at art college. They’d come to visit and I got into recording them, buying tape recorders and microphones. I also manufactured a machine that turned lights on and off in time with music – the kind of thing you see everywhere in discos nowadays. Such machines already existed but they weren’t able to switch much current. My 40-Amp triac based device, the Synchrolight, could switch 30 kiloWatts and I used to take it to parties regularly At one of these I met a film director who hired me to light a film set – the disco scene in ‘Buttercup Chain’ with Jane Asher and Hywel Bennet – for which I got paid £1400, an enormous amount of money. I did another film for another £1000 after which the disco light business expanded and I didn’t feel I wanted to compete on a professional level. Basically I looked upon this sort of work as a profitable and exciting dabble – something to do and enjoy while deciding where my ultimate future lay.
Just before I did the first of those films Naim Audio was registered, actually as Naim AudioVisual, in July 69.
To earn myself a living I sold a few Synchrolight systems, assembled a few hi-fi systems for various people, and worked part-time putting mixer desks together. The Naim amplifier thing started with the mixers. These weren’t my designs and when I looked at what the circuits did on a scope the waveforms didn’t look very good. I thought that if they didn’t look good it had to correlate with something being wrong with the circuit: if you could see that distortion on a screen there had to be a way of eliminating it through better design. So I started reading books about transistor design – I’d had no formal training whatsoever – and it took me about a year to learn enough to design the NAP2OO, which became the NAP25O. It then took me another year learning how to get it to work inside a box, and learning what was really important in amplifier design, which factors were crucial. Despite the product emerging from this fiery baptism, however, the 250 remains virtually unchanged eighteen years later.
MS: You were building for professional users in the early days. In 1973 Naim equipped Capital Radio’s studios with 24 amps and speakers. When did you enter the domestic hi-fl market?
JV: There were two key people involved in that move. The first was a dealer: David Row at Exon Audio in Exeter who I met at an APRS where I was sharing a corner of somebody’s stand (an exhibition for the Association of Professional Recording Studios). He bought an NAP2OO (which is what the 250 was called then) which was the beginning of retail sales. The second person was Ivor Tiefenbrun (MD of Linn Products) to whom I was introduced by KEF’s Roger Walker. We listened to Ivor’s turntable and speakers with an Amcron amplifier which Ivor thought had to be superior to my ‘underpowered’ offering. I thought the system sounded execrable, no merit at all. After a bit I suggested that the amp was no use and that we should substitute my NAP 160 (the smaller amp which followed the 200/250). The sound was transformed and Ivor declared “now that’s what I call an amplifier”. Ivor then played the amplifier to all the dealers he visited with his turntable and told me that he would buy every one I could make. It is worth noting that at this time Naim Audio was virtually a one-man band and monthly production was hardly into double figures.
Then in 1974 I went to the hi-fi show in Harrogate with Ivor, which was an eye-opening experience: journalists were offering to review my amplifiers but only if they could have one for free or if I was prepared to pay them. I was appalled that magazines behaved this way but that, I was told, was the way of the world. I decided that being the case I wasn’t really interested in reviews, but a writer called Paul Rasmussen asked if I would let him review the amplifier and I agreed that he could provided that he bought one. He did, and after that people really started to get interested in who Naim Audio was.
And the rest, as they say, is well documented history. From its humble back-room-in-a-back-street beginnings, where Vereker worked single-handed for much of the time, the company progressed to its present factory, which offers 20,000 square feet of heavily computerised production and office accommodation on Salisbury’s Dolphin Industrial Estate. And a head count of the staff now numbers over 50 people. The company also has its own premises in Chicago, which house the American subsidiary company, Naim Audio North America.
What I wanted to discuss now was Naim’s fundamental philosophies, and get the low-down on forthcoming new products. I also intended to see if any of the myths that surround the company might be debunked in the process. Were any of the apocryphal tales and folklore that traditionally associate themselves with Naim and its equipment actually based on any truths?
MS: What has happened to those products that everyone has long known about but which still haven’t surfaced? I’m thinking particularly of the NAC52 preamp and the FL1 electrostatic speaker.
JV: The 52 is in the final stages of production drawings, so we should be able to show a production prototype at the end of April. Production will then get started soon afterwards. But there are quite a few new things which are currently on the way and these have taken up all our R&D people’s time – the ARO arm and the new NAT 02 tuner, for example, and the NAC A5 cable.
MS: So what’s the story of the new cable?
JV: BICC told us it would no longer be making NAC A4 for us, which meant that we needed to source a speaker cable elsewhere. While this meant that we would be rather short of cable in the interim period it encouraged us to produce an even better cable. To be honest we were quite happy not to deal with BICC any more. It had infringed our copyright on the original cable – NAC A4 was our design – by selling similar cables to other companies. People have literally ripped us off but I’m no longer bothered about it: suing people is really a waste of energy and time, It’s simpler just to design something better. And we knew that we could make something far better given the time and resources to develop it. Which is what we’ve now done with the A5 cable.
MS: So, what specifically does the development of such a speaker cable involve?
JV: If you want to refine the cable you need to make some prototypes. And that’s not the cheapest of exercises: you actually have to shut down a plant that’s manufacturing thousands or hundreds of thousands metres of cable per day and sit and run a hundred metres of one type, a hundred of metres of another, and so on. It’s an incredibly expensive and time-consuming business.’
MS: Isn’t much of the process still empirical?
JV: With NAC A4 it was very much a case of “with this design we think we can expect this and this to happen” and it was made and it performed pretty much as we expected it to. But it was by no means an optimised design. With the A5 a lot more thought went into it and we had many samples made in order to get as close as we could to the desired result.
But to get back to your original point, this and other things have delayed the arrival of the NAC52 and FL 1. The other point is that while it is easy to build a one-off which works really nicely you have to design things such that you can make lots and lots of them and they will all be consistent.
Most of the work we’ve been doing on the NAC52 over the past year has actually been on electromechanical aspects rather than the electronics: the electronics are pretty much finished and decided. It has been getting the mechanical side sorted to our satisfaction which has taken us a very long time. Buying off-the-shelf items is generally unsuccessful as they never seem to do what you want. Conversely, having things built to order gives you probleMS: often you wait eight to ten weeks to get an item which you have specified only to find that it’s not quite right. So you respecify it and wait another ten weeks for a further sample to arrive. This doesn’t have to happen too many times before you discover that a whole year has flown by.
Work on the electrostatics is still progressing quietly but Guy Lamotte, their designer has been heavily involved in the production of the ARO tonearm. We made the first prototypes of the ARO in March 1988 but it took an age to find an engineering company who could fine engineer all the component parts to the tolerances we required. In June that year I thought we had it right but our sales people weren’t happy with the finish. Finally, in February this year, we produced the first batch that everybody deemed satisfactory. If it all goes according to plan, by the time you read this there will be arms in the shops and Guy will be back at work full-time on the FL 1.
MS: Let me change the subject briefly and ask you to clear up one or two misunderstandings about Naim equipment – or rather get authoritative answers to questions in response to which I’m sure many people, will have received conflicting opinions. Firstly, why use BNC connectors on the phono inputs of your pre-amps?
JV: If you look at the characteristic impedance of the majority of arm cables it’s somewhere around 60 – 80 ohms. By using a BNC plug/ socket which has an impedance of 50 or 75 ohms, depending on which one you pick, you bridge the interface to the cable with the correct impedance. An RCA (phono) connector has an impedance around 220 ohms so from the signal’s viewpoint it sees a ‘step’ in its path. This step causes reflections (at electron level) and its effects are very audible.
This is apart from the electromechanical aspects – particularly with the large, heavy RCAs which hang on the back of the preamp sockets, waving in the breeze. These are amazingly sensitive to mechanical feedback: they vibrate all over the place and because the signal is so delicate at that stage, the colorations they impose on the signal are very marked. The difference between the low mass, cheapie, RCAs (which tend to sound better assuming they’re made properly) and the huge ‘audiophile’ types is clearly audible in terms of mechanical feedback, ignoring the quality of the electrical connection.
In the systems of ten years ago you wouldn’t have heard the difference between BNC and RCA. In fact, in those days we did lots of things that I knew would affect the sound quality but you couldn’t hear the effects at the time. The early preamps used DIN phono inputs, which, in theory, should have sounded better but they didn’t And whatever connectors one used the ground connection should be isolated where it entered the amplifier – again I couldn’t hear or measure a difference. So back in the early 1970′s we used the cheapest RCA sockets: there’s no point in using something more elegant if it doesn’t bring any benefit. As systems got better we changed to insulated RCAs. And as they got better still you could hear the problems RCAs caused and so we went over to BNCs.
MS: With regard to your tuners, which in my opinion are unsurpassed, why are they so much better than the competition and what accounts for their higher than average prices?
JV: First of all I don’t consider them to be expensive.’
MS: In comparison to, say, a QUAD FM4 – which many regard as being a ‘good’ tuner – they are.
JV: No, it’s a different ball game altogether. The first thing is why ours sound different to anybody else’s and, in our opinion, very much better: we’ve set out to do something different. Most of the people who’ve designed tuners have started – I assume – from a communications background and so they attempt to get the maximum intelligibility from the maximum number of stations in a given bandwidth. So they don’t worry as much about frequency response, phase response and so on; they just want to get the signal out of the noise floor Whereas we said the signal is very delicate, it has a miserable modulation index – i.e. there’s a lot of important information packed into a rather narrow deviation band – so you have to treat it with a great deal of care and attention. Therefore the front end needs to have a very wide bandwidth. Not so wide you get the station next door in there as well but wide enough so that you don’t distort the phase. That makes it less satisfactory in terms of dredging weak signals out of the noise and also stations which are close together: but the resultant station once you get it properly is much, much better.
MS: But it is a more expensive approach to tuner design. For example, the alignment and peaking up of the front end takes a lot of time: each tuner head takes over an hour to align correctly.
JV: All our tuners, the 01, the 101 – which is now passing away – and the 02, use the same head, which we make ourselves, and which is an expensive piece of equipment. If you buy in a Japanese head, a really good one, if you’re buying a thousand pieces, it will cost you about £8. The really excellent ones from the past used to cost about £15. If we were to sell ours to other manufacturers – even in quantities of 1000 – they would cost at least £70. The gadget we use to align the heads costs £30,000 alone.’
Once you’ve started off with the idea that you’ve got to look after the signal, when you go all the way through the tuner you find many more areas to attend to. It really is a miserable signal and it wanders all over the place in terms of level. Some people have good aerials that feed half a Volt into the tuner: others can only manage a few micro Volts. A tuner needs to work across that band reasonably well. Now it’s obviously not going to work with a few micro Volts because a compromise we’ve made is that there’s no point dredging stuff out of the muck – that’s for DX freaks: it’s not for people who simply want to enjoy the music. The music will be better if you’re dealing with 10, 20, 50 or, preferably, 100 microVolts of signal.
With the 1 uV stuff you’ll never get a good sound no matter what you do.
MS: I suppose that anyone buying a Naim tuner will at least have a half decent aerial.
JV: Not just a half decent aerial. On the ends of our tuners you can hear as clearly as day the difference between one excellent aerial and the next. For instance, we have two aerials here with which we can make direct comparisons – Ron Smith’s G17 and G23 They’re Galaxy aerials and are both excellent. The G17 having quite a lot less gain and less narrow acceptance actually has a flatter gain curve over the whole bandwidth and upsets the phase slightly less and generally sounds better.
MS: Is there an ethos that neatly encapsulates Naim Audio’s approach to manufacturing hi-fi – one all pervading philosophy?
JV: The reason I started Naim Audio was that I wanted a system upon which I could listen to and enjoy music. When I was recording my musician friends at the start of the seventies, I could not understand why the tapes I had made never sounded like the people playing or their instruments. The tapes, I’ve later proved, were OK, it was the replay chain -the amplifiers and speakers – that was inadequate. So I tried to sort it out. I liked the result and so did other people who heard what I’d done. That’s how we got into making things for other people.
But, in reality we’ve spent an awful lot of time in the past twenty years initially just making products for ourselves then discovering that other people wanted them too. Like the NAP 135 power amplifiers: we never thought that anyone would want to spend a thousand pounds per channel for an amplifier. All the things we’ve done, with very few exceptions, have been done for ourselves. If somebody in the company wants something badly enough we’ll probably have a go at making it and it may well turn into a product.
MS: Have any products been purely the result of market demand?
JV: Yes, the NAP9O. Our European distributors asked for an amp between the NAIT and the NAP 140 so we built the NAP9O. We wouldn’t have built it otherwise, but it has, nonetheless, proved very successful.
MS: Going back to the ‘folklore’ for a minute, may I ask about sonic differences between the NAC32 and NAC62 preamps? Some people claim that the NAC32 is vastly superior but I have to admit I’ve never detected any great differences between them.
JV: The preamp story is interesting so I’ll tell you how they came to be. The numbering is simple. The first was the NAC 12 – Naim Audio Control amplifier Number 1 with 2 channels. So the NAC32 was our third, 2-channel preamp and it came about because dealers said it would be nice to have a preamp with two phono inputs for comparing turntables easily and that a mono switch would be handy for setting up speakers. Then people wanted two tapes and so it grew to have five inputs. Right from day one all the preamps have been able to accept over 7 volts RMS through their high-level inputs, so when people say they weren’t designed for digital I tell them that 7 volts is far more than the two and a bit volts maximum that CD players put out. The fact that CD players don’t always sound wonderful through them is not our fault However: things can be masked a little using the special 328 input boards.
Because of system hierarchy we decided that there was no point in economising with the preamp – that it’s best to make it the best one can, and just alter the facilities. Essentially the performance of the current preamplifiers is the same. They do not sound the same, and no two will sound exactly the same, but in terms of absolute performance they’re pretty much on a par. You may prefer one in one situation yet prefer the other in another situation but to isolate which one is better is pretty difficult. lf you are listening to fifty preamps in our dem room and there’s a mix of 62s and 32s you may well find that there are some 62s that work better than 32s and vice versa. You might prefer a whole batch of 62s to another of 32s. Next week it may well be the other way round. The differences between the two preamps are not that large: there are probably bigger variations between a current 62 and one from three months ago. There has to be: we use, for example, three different suppliers for our printed circuit boards. We have to because we use over 80,000 PCBs each year. They do sound slightly different but that’s not to say better or worse.
I’m continually amazed when I read things written by certain people in certain magazines which are based on oft-repeated folklore when it would be so easy just to ring us up and ask us for the truth. One magazine, for example, said that Naim kept no records in the early days. Had the journalist in question spoken to me – the only person guaranteed to know – he would have found out that that wasn’t the case. We do in fact have records of not all the amps but certainly all those that were given serial numbers: and there was only a handful at the very beginning that didn’t have them – two or three maybe out of the fifty thousand plus we’ve built.
MS: Finally I’d like to ask you about the ARO arm – why and how it came about?
JV: Simply because there was someone here who wanted to make an arm and he did. Having made one for himself and one for me, other people in the company said they would like one, as did other people outside the company.
The thing is you can swallow the way that things are talked about by other manufacturers for so long and you can believe what they say if you have confidence in them. And when it was explained to us that what a turntable, arm and cartridge is trying to do is to recreate in an open loop situation when the way in which a record is cut in a closed loop situation, then you take it for granted that this is the understanding and this is how it actually works. After a while you start thinking – because it doesn’t sound the way you think it ought to or measure the way you’d imagine – that there are clearly things wrong with it and how you can fix it.
When you cut a record the lacquer is absolutely microscopically flat. Yet vinyl records are anything but flat. Because of the way conventional arms work with conventional bearings, the stiction, and the way their mass is disposed, they tend to apply a torsional stress to the cartridge cantilever all the time the record is being played. So they tend to mistrack at higher frequencies and have other problems like phase relationships going to pot. Basically they sound bright and splashy. One of the solutions is to make a unipivot arm which is dynamically balanced – correctly – so that it can actually float around at subsonic frequencies, to deal with the warps and inequalities in the vinyl without putting torsional stress onto the cantilever.
There are of course other necessary properties in the arm’s design – like defining the bandwidth precisely and making sure that the phase response – mechanically – of the whole arm is correct. All these little details help and you’ve heard the results. We honestly wouldn’t have bothered to make the ARO unless we felt that it was going to be a lot better than anything anyone else had offered or was likely to offer. It’s a question of why make life difficult for yourself. If it’s easier to employ a unipivot – despite other people’s reservations about the principle – it doesn’t matter. We do it to please ourselves, so that we can continue enjoying our records at least until we can figure out how to make CDs work properly.
MS: Did you just admit to trying actively to improve CD reproduction?
JV: We think we know how to do that, but we’re concentrating on other projects at this particular moment. But we’re pretty sure we can improve CD replay – and that’s not by addressing the areas that other companies seem to focus on. I’m sure that the medium is not inherently flawed – a particular disc or recording may well be and there may be no way of fixing that, but we’re confident that using that format you can make discs that will sound better than anything on vinyl. If it’s done properly you have the inherent advantage of correct timing which with most turntables you don’t – they slow down when asked to work.
MS: But surely as a manufacturer primarily concerned with amplification components you are still dependent to a very large extent upon the turntable: if none that is available meet your criteria that must place you in an invidious situation?
JV: One has to accept the inherent compromises of each design and pick whichever suits one the best. There’s an answer that’s right for you and one that’s right for me. The fact that I would violently disagree with an awful lot of right-for-them answers – the particular thing of soundstage, detail, and depth and imagery, i.e. the presentation being more important than the content, I think is totally wrong. There’s no point in having great presentation unless you have the content. I’d far rather talk to Einstein on the telephone than talk face to face to a doorman in a hotel. One may be ‘live’ and ‘there’ and visual and all the rest, with perfect presentation and fidelity but it’s the content of the conversation that’s important.
MS: A good analogy, which I think neatly sums up the Naim philosophy and one that will nicely terminate this discussion – albeit at the expense of all your potential doorman customers.