Inside Views: E-mu Systems
Continuing our series that looks at hi-tech companies and design personnel working in the field of music and recording technology, Paul Gilby talks to Marco Alpert, Director of Marketing & Sales for E-mu Systems of America - the company behind the Emulator and Emax samplers.
Continuing our series that looks at hi-tech companies and design personnel working in the field of music and recording technology, Paul Gilby talks to Marco Alpert, Director of Marketing & Sales for E-mu Systems of America - the company behind the Emulator and Emax sampling keyboards.
Tell us how the company began?
"E-mu was founded as a partnership by Dave Rossum and Scott Wedge around 1971 to make modular synthesizers. The reason for its existence goes back to when Dave was studying at the University of California. The Physics Department had bought one of the early Moog synthesizers and nobody around the place could figure out what to do with it. Dave became fascinated by it and that summer he and a bunch of friends decided to try and build their own synthesizer.
Towards the end of that summer an old high school friend, Scott Wedge, called by one day and became immediately fascinated by what they were doing. When college started again, Dave and Scott stuck together and decided to build a better synthesizer than the Moog. As this project continued they decided to make a business out of it and start to manufacture state of the art modular synths."
They obviously must have felt dissatisfied with the performance of the Moog system?
"Yes, they did. It was widely known that there were certain aspects of the Moog that were less than ideal, such as the unstable oscillators and its limited functions. They set about the task of building a system with a lot more functions, rock-steady oscillators, and filters with increased range. Also, from the very beginning they avoided an analogue keyboard and designed a digital one instead; in fact, E-mu have never made an analogue keyboard to date. In those early days they also built the world's first modular digital sequencer."
How did you get involved with the company, Marco?
"I didn't join the company until about 1975 and met Dave and Scott by accident. Around 1973 I went into a music store in San Jose, California, looking for an ARP 2600 synthesizer and one of the guys working there was named Ed Rudnick, who is today actually the Director of Manufacturing for E-mu. He was a salesman in this music store and I went in there asking about the ARP and I guess I must have been one of the first people to ask fairly intelligent questions.
Whilst there, Ed told me he had some friends who were actually making synthesizers and I thought this was unbelievable. I couldn't imagine that private individuals could make such a thing. So he called up his friends, Dave and Scott, who at that time were working out of an apartment. We went over to meet them and found a few people sitting around in the living room of a small apartment, building synthesizers! I was just amazed and immediately recognised what they were building was an incredible instrument, but it was well beyond my financial grasp. Anyhow, I just kind of stayed friends with the E-mu guys for a year or two...
In 1975 I went down to Los Angeles and it just so happened that E-mu were looking for a sales rep in that area, so I got the job. After four years I moved back up to E-mu headquarters in North California where I'm based today."
Was the modular system widely available at the time? Who was using it?
"We had very little international distribution in the early days and the modular system built up its reputation largely through word of mouth. Sales were almost entirely in the States, with very few going out for export. It was mostly rock stars who were buying it in America - Frank Zappa bought a huge one, and so did lots of colleges and universities. There are still some hundreds of them out there! We feel we did fairly well with them at the time but we didn't really know how to advertise and market the product.
At the same time as the modular system was being manufactured and sold, the company was working on a lot of other projects for other people. For example, our monophonic digital keyboard was developed into a polyphonic one and that was licensed to Oberheim and used in their original 4 and 8-voice systems. The microprocessor-based version of that, which I believe was the first commercially available microprocessor product of its type, was the 4060 16-channel polyphonic keyboard and sequencer. That basic technology was licensed to Sequential Circuits for the Prophet 5.
Dave Rossum, our chief engineer, also worked with a company called Solid State Microtechnology to develop what were to be called the 'SSM chips'. Those were the first semi-custom integrated circuits developed specifically for music synthesis."
Did the licensing deal with Oberheim and Sequential help bring in revenue to fund the company's expansion?
"The Sequential one did, but Oberheim wasn't a huge volume sale product and didn't contribute a large amount. The Sequential Prophet 5 was, of course, something completely different! It was a huge success, and during the period when I first came to work for E-mu full-time, it gave us a nice secure financial base which allowed us to work on projects like the 'Audity'. I don't know if you have ever heard of that?"
No, I haven't. What is it?
"Well, E-mu were commissioned by Peter Baumann, who had just left Tangerine Dream, to develop a completely computer-controlled polyphonic synthesizer with voice circuitry that was much more complex than the Prophet 5 approach. We developed a synthesizer on a board about sixteen inches square which contained a couple of oscillators, different filters, VCAs, LFOs and transient generators - just about everything - it was a wonderful voice!
We developed these boards for him and then decided to develop the ultimate computer-controlled synthesizer. It would be called the 'Audity' and cost something like $70,000; we actually built up a prototype. The whole reason why we were able to pursue projects like that was because we had the Sequential royalties supporting us.
It was around that time that we stopped and said to ourselves, 'We have put a lot of other companies on the map, maybe it's about time we got into the market ourselves'."
"In 1980, we came back from the Los Angeles AES show having seen the Fairlight CMI for the first time. We spoke with some people who had used it and what became clear was that although the CMI did a lot of things, most people couldn't figure out how to do any of them except for sampling. I'm not sure it was even called sampling back then. We came back from the show and thought, 'If people are prepared to pay $27,000 for the Fairlight (that's what it cost then), might they be willing to pay $10,000 for just a keyboard that was optimised for sampling?'...
From that initial idea we developed the Emulator I digital sampling keyboard and managed to establish it very quickly. By today's standards it was pitiful, with its 8-bit sound quality, small memory and lack of control over the sampled sounds.
Things have moved very quickly since then and we are not talking about decades, we are only talking about five years. The Emulator I sold from 1981 until 1984 when the Emulator II was introduced. It was a great success and, even though the Emulator I was limited, it was the best available for its time.
During its lifespan we developed close contact with many professional musicians who were using the instrument and it became very clear to us what they would like a sampler to do. You see, when we designed the Emulator I, nobody had any idea what they wanted it to do. So, when the Emulator II came out we had a much better idea of what facilities people actually wanted in the next generation of samplers. History has proven our thoughts right."
Who was using the Emulator I in the early days and what were they telling you?
"Actually, a lot of relatively unfamous studio musicians formed the core of people who passed on remarks about how we could improve the instrument. There were also the rock star owners like Stevie Wonder and Jerry Harrison of Talking Heads. People would just phone us up and say things like, 'Why don't you expand the memory?', 'Put some analogue sound shaping on it' or 'Make it loop better' etc. It's thanks to this type of user feedback that we feel we have been able to develop products that are more closely related to what the musician actually wants."
What factors do you attribute to the success of the Emulator II?
"I think the EII is just self-evidently powerful. It was optimised very well to carry out one particular function rather than a lot of functions in a compromised way. I also think that, as a company, E-mu tend to be very sensitive to the user-interface because the situation has become more and more computer orientated. It's all too easy to make people interface to the instrument as if they were computers - we are not in favour of that approach; we feel that, first and foremost, it has to be a musical instrument because the people who are using them are musicians. Consequently, although our instruments have a great depth of functionality and control, we try to design them so that you can interact with them successfully at any level of operation. If all you want to do is take pre-recorded disks and load them in and play, that's going to be easy and straightforward. If you want to take the next step and modify the sound, there again, we make that level of control straightforward as well. You can go as far as you want, but at each step the user-interface has to be effective."
We have seen other companies come along and develop a variety of peripherals for the Emulator II, like Digidesign's sample editing software which runs on the Apple Macintosh, and the CD ROM from Optical Media. Have you had a close working relationship with these companies?
"Yes, we have been connected with Digidesign since the Drumulator days. They started out doing alternative sound chips for our Drumulator and they built themselves a very successful business out of it. When they came up with the idea for the Sound Designer editing software, we were really enthusiastic. It's a perfect relationship because we don't want to be a software company and they don't want to be a hardware company.
You see, if you are somebody like Synclavier or Fairlight and you really only have one product with a reasonable lifespan, you are required to support it with software. After all, that is what the product really is - once you've built the hardware section, it's a software-based system that continually grows. E-mu, however, are a much more broad-based company, as the EII isn't our only product. We have the SP12 sampling drum machine, Emax, and more things to come. To have the support of Digidesign allows us to simultaneously develop an instrument on both software and hardware fronts, without the need to support an after-market software division.
In the case of the CD (compact disc) ROM Alan Atkins, head of Optical Media who developed the system, just happened to live close to E-mu headquarters and dropped by one. He is one of the foremost experts in the world on the CD process and came to us literally off the street. He said that he had access to a new technology that's so new that there were no commercial applications for it, and that he'd like to make a device for the EII that would be the first application of this technology. We said, 'Great! Why not?' So we started to work together and some months later the first CD ROM system was developed and linked to the EII, giving it some 500 banks of information."
How far do you see that relationship continuing in the future?
"Clearly optical storage is a very exciting tool in a field like sampling where storing large amounts of data is important. We feel that the headstart we have in this area is something that we must exploit. Beyond that I can't really say too much right now."
We've spoken at length about the Emulator, but what of the Drumulator?
"Well, we think that our Drumulator really pioneered the mass user market for sampled sound drum machines. At the time of its release there was only the Linn LM1, but its price placed it out of reach of most musicians.
The Drumulator was actually a sort of stopping place on the way towards developing the EII. We were looking at a new technology for sampling and when we had got to a point where the EII had started to evolve, but was still a year or two away, we realised that in that form it could be used for something like a Linn LM1, but at a third of the price. So that's how the Drumulator emerged.
As it turned out, by the time we had finished the EII, the technology changed yet again but several of the basic elements remained.
The Drumulator was a huge success, competing very well until the very low-cost Roland and Yamaha machines came along. I think it was a very important product in terms of putting us on the map worldwide - more so than the original Emulator. Again, as with the EII, the Drumulator enjoyed the support of a range of peripherals developed by third parties. I mentioned the Digidesign sound chips, there was also a JL Cooper mod that let you install three sets of the Digidesign chips and switch between them. Jim Cooper also developed a sound processing board for the E1 which could be installed in the machine and it let you shape the sound with filters and envelope generators. Although users felt a need for these facilities, we weren't prepared to redesign the E1 at the time. So, Jim came out with this retrofit which had its own set of knobs and a printed label that matched the Emulator front panel - he did a real good job.
When the Drumulator was finally retired we were concentrating our efforts on the EII, so we left drum machines alone for a while. The budget drum machine market had become so cut-throat with the prices dropping to three or four hundred dollars, that we felt we would stay with the pro end of the market and we started to look at designing a companion for the EII.
The SP12 drum machine came out of that idea quite naturally. At the time there wasn't anything out there that had 12-bit resolution, so we developed the SP12 to have the edge over other products by using 12-bit technology and, most importantly, making it a user-sampling machine. This gave us a product which we felt was exactly what the professional wanted. At that time, drum machine use was related directly to pop music and people were getting fed up of hearing the same drum sound, so we gave them the power to sample their own sounds and inject their personality back into the music."
You have now introduced the Emax sampling keyboard into the mass consumer end of the market. Does Emax carry the true pedigree of E-mu or has it been compromised to fit a target market?
"The existence of Emax is largely due to the development of our 'E-chip'. This chip contains the bulk of the circuitry required to build a sampling machine and is really the logical development of the technology we employed in the EII. It's an amazing chip measuring about 4x1 inches and it serves three functions: it generates addresses to the memory, it carries out some digital signal processing, and it provides 2:1 ratio data compression.
The E-chip is capable of giving 16-bit resolution and storing data as 8-bit words. There's a lot of confusion about this area. People know about 8-bit companding systems, which is what the E1 and the Drumulator employed, but it only really gives you about 12-bit resolution and around 4-bits of distortion; so you do get greater dynamic range but at the cost of distortion.
The E-chip uses a technology that gives true data compression and in the case of Emax, we are using 12-bit A/D and D/A convertors, so Emax is essentially a 12-bit machine. Its data compression technique takes the information about each sample and spreads it over a number of adjacent data words.
We have, of course, chosen to eliminate certain functions from Emax. It doesn't have SMPTE, the built-in sequencer is not as complex as on the EII, there's no hard disk option... but it does have all of the most useful functions of a sampling instrument, and it's available in both a keyboard and a rack-mount version."
Could you tell us about the future plans of E-mu?
"I can only say that we are in the process of expanding our presence across the spectrum of the electronic musical instrument market. We have established ourselves successfully in the professional market and the introduction of Emax sees the start of our desire to address the mass consumer market again, as we once did with the Drumulator.
When the Emulator 1 was being produced, E-mu Systems was just a little company of about 9 people based in a Victorian house in downtown Santa Cruz. Today, we have some 120 people working in the company and we are looking at further expansion now that we have Emax."
You've given us a very interesting and revealing insight into the company development and philosophies behind E-mu Systems; could you finally tell us what your approach is to providing factory samples for your instruments?
"We believe that the musician's attitude towards sampling machines is that he perceives the quality of the instrument by the quality of the samples available. This is why we are always working very hard to generate a series of quality factory samples. With the EII, there are literally thousands of samples available and we will probably transfer some of the EII library to Emax.
The quality of samples is very important, because if the recording of the sound you are making is not good enough for mastering, then it's also not good enough for a sample. It's a case of 'garbage in, garbage out'. We take a lot of care when recording our samples. We use a Sony F1 digital recorder and then feed it through our own special computer which has the ability to clean up the sound before transferring it into Emax, say. The idea that you can just plug in a cheap microphone, hold it in front of sound sources and record ready-to-use samples is silly... It's a seductive misconception!"