Sampling Ensoniq (Feb 1988, article)

[chrisNova777]

http://www.muzines.co.uk/articles/sampling-ensoniq/3473
Sampling Ensoniq

My first encounter with Ensoniq occurred well before the Mirage was even a dream on the musician's horizon. I was at the January 1984 NAMM show in the States, and being the inveterate badge-looker that I am, noticed Bob Yannes concentrating intently on some new piece of gear. Bob had been a frequent contributor to Polyphony magazine (Electronic Musician's predecessor), and his articles had been uniformly excellent. So, I went up and introduced myself, and complimented him on the work he'd done for the magazine [Ed. - Craig was its Editor]. It was the first time we had met in person and we had a pleasant chat, but as we parted, he turned and said with a sly grin, "Watch for a 10-to-1 price reduction in sampler technology next year."

At that time, the world was used to big-ticket Fairlights, Kurzweils and Emulator IIs, so a 10-to-1 reduction would mean a lot. Normally, I would have dismissed that as the usual kind of hype that's part and parcel of a trade show, but I knew Bob well enough to realise that maybe he was on to something. I also knew that since 1982, Bob had been involved in some kind of company that was doing consulting on custom chip technology, and that prior to that he had been the 'whiz kid' on the Commodore-64 design team who had designed the SID (Sound Interface Device) chip for the C-64, as well as much of the rest of the computer. When this kind of person talks, one pays attention.

At that show, PVI (the name the company went by prior to becoming known as Ensoniq) was demonstrating a little drum machine card for the Apple II, and a prototype of a $200 drum machine with sampled sounds. Nowadays, with devices like the Alesis HR16, Roland TR626, and Kawai R50, that may not sound too impressive. But remember, at that time the $1000 E-mu Drumulator was considered inexpensive, and a $200 drum machine was bound to turn a few heads. Unfortunately, the sound quality was not all that great - nothing over 6kHz - and the Apple card, while useful, had the fatal flaw of not allowing for external synchronisation. I filed PVI under 'Promising, see if they're at the next show.'

Then towards the end of 1984, I started getting the first phone calls. "Have you heard about this $1700 sampler? It's called the Mirage." Although I had seen some ads, I figured that perhaps it was another case of vapourware. Then I remembered Bob's comments about a 10-to-1 reduction in sampler pricing.

At the February '85 NAMM show, the Mirage sampling keyboard was formally introduced to the public. It wasn't vapourware; it was real. The three founders of Ensoniq (Bruce Crockett, Al Charpentier, and Bob Yannes) - all refugees from Commodore - had channeled their knowledge of VLSI (Very Large Scale Integration) technology into producing a musical product with a remarkable degree of power for its price. The result of their VLSI experience, the 'Q-chip', was impressive: it integrated over 20,000 individual components (quite a few for a custom chip of that time) to produce a sampling keyboard with 16 voice pairs, onboard filtering, and reasonably good sound. To date, the Mirage has sold over 25,000 units and according to Bob Yannes, "As far as we've been able to determine, that's more than all other keyboard samplers combined." He's probably right.

When Sound On Sound asked me to do a profile of Ensoniq, I called up their PR firm to set up a good time and date for an interview with the three founders. A day later, I got a call and was told that they really didn't want to do an interview, and that they didn't want to concentrate on personalities. Rather, they felt the instruments should speak for them. This was all done without any arrogance; it was obvious that Ensoniq wanted to focus any attention they would get on the instruments, and perhaps a bit of shyness worked its way in there too. Fair enough. That's why you won't see a lot of comments from Al or Bruce; as a matter of fact, most of the interview came out of a long conversation with Bob Yannes at a later date. But they were more than willing to give me a factory tour.

In February '86, Ensoniq had expanded to a 26,000 square foot production facility in Malvern, Pennsylvania to keep up with the demand for their products. Unlike many American manufacturers, all Ensoniq products are manufactured at their plant (for a while European models were being manufactured in Italy, but this was no longer necessary when the company moved to larger premises). About a year later, Ensoniq moved to a 70,000 square foot facility, and that's where the visit took place.

The tour began with their classroom, where dealers are taught, in small groups, how to use and demonstrate Ensoniq's products. The classroom has about a dozen workstations, sort of like the language labs in some schools, where dealers can get up to speed on Ensoniq's line. "We have a limited number of dealers, and we intend to keep it that way," said Bruce Crockett. "We want to make sure that our dealers can support the product and the customers." The classroom was a reasonably new venture, but it seemed to be working well for the company.

Next came a stop at the in-house VLSI design centre, complete with the requisite posters of VLSI chip innards on the walls. This is the heart of Ensoniq's success. The Q-chip was designed in such a fashion that it not only formed the heart of the Mirage, but was also transmuted into the ESQ-1 synthesizer - whose rate of sales has surpassed the Mirage - and the sound chip used in Apple's IIGS computer. Even the brand new SQ-80 [see review: p8] is based on Q-chip technology (although the EPS sampler uses an entirely new chip - see review: p36). Aren't there dangers in not using off-the-shelf parts, though? Bob replied that "These chips can be made by just about any company. We're not locked into a single source; in fact, several companies are capable of producing these kinds of chips in quantity."

Next came a look at production. "We're not proud of our reliability record for the Mirage," said Bruce with a degree of much-appreciated candour. He then proceeded to show me the investment they had made in automated assembly and testing equipment; circuit boards are extensively tested on the component level, and any deviations from the expected are carefully noted and analysed. According to the charts I saw, the number of failures has indeed been drastically reduced, and Ensoniq seems committed to increasing reliability by whatever means possible. Of course, this is also enlightened self-interest. Reputations die hard in the music biz, and no manufacturer wants to be saddled with a 'nice products but they don't last' reputation. At this point in their career, Ensoniq is in a position to break out into being a major company by any standard. It's important that reliability problems don't tarnish that reputation, which is perhaps why units are not just tested and re-tested, but burned-in for a period of time to weed out any cases of 'infant mortality' (chips that check out fine, but inexplicably fail shortly after installation).


Ensoniq assembly line workers hard at it!


Next stop was the office of the Chief Financial Officer, Malcolm Walter. I fully expected this to be a good time to catch up on my sleep but, actually, this was just as interesting as the rest of the tour. Malcolm explained the ways in which they work with a dealer, even helping individual dealers work out ways to avoid stocking large inventories, and maintaining a profitable turnover of equipment. Seldom has the manufacturer/dealer relationship been explained so well to me, and being somewhat removed from that scene, I learned quite a few things.

Finally came the customer service centre. Ensoniq has a reputation for excellent customer support, and I can see why. The department, managed by Dan Garrett, had about a half-dozen bright-eyed young techno-types manning the phones and answering questions from the mundane to the sublime. One of them said, "Most of the answers involve saying 'have you read page XX of the manual?' and that takes care of it. But sometimes we do get some really unusual applications-oriented questions." Dan also emphasised that they will stay on the phone with a customer until the problem is solved. "The last question we ask is 'do you have any more questions?' If the answer is no, then we're finished; otherwise we keep going."

I couldn't help but notice the atmosphere at the company - surprisingly low-key and quiet. I was assured that when production starts up on a new product things are considerably different, but even then, I suspect Ensoniq's brand of intensity might have a different flavour from most. Different companies have different personalities. Ensoniq lacked the arrogance I've found with some companies that made it to the top (and that arrogance is what sends them to the bottom just as fast), and people seem relaxed and wholly dedicated to the job at hand. I asked Bob what their secret was.

"First of all, there aren't any secrets within the company. Any employee can know how the company's doing, which products are selling the best, pay grades for different jobs, whatever. Of course, most people don't really need to access this information, but the fact that they can get it is good enough.

"We have a lot of discussions with the managers, and the managers talk over any decisions with their team. We also have factory-wide meetings once a month - that's about 180 employees. Communications are kept very open."

I remarked about the openness of that structure, and how much I agreed with those principles.

"When people are a part of something, everything works better. If ESQ-Ms don't sell that month, we all know it. And regarding the arrogance thing you mentioned, I don't think it would happen to us... Bruce wouldn't let it happen."

Although Ensoniq kept their promise not to talk personalities, I did manage to glean a few facts. Part of the company's sensitivity to making assembly lines work as they should is because that's where Bruce Crockett (Ensoniq's President) started out many years ago. Bruce has no musical training, which on one hand is a disadvantage, but the flip side is that he seems extremely responsive to what those outside the company have to say. Most of the time when I visit a company, they want to spend their time telling me how great their product is. Bruce spent his time trying to get me to tell him what I thought about their stuff and, especially, where something could stand improvement.

And speaking of personalities, the Ensoniq Performance Sampler's design features represent a variety of people's talents. Bob Yannes, who remains the enthusiastic fan of musical electronics he has been since his youth (he's now around 30), designed the basic hardware limits.

He described the EPS as "Like a Macintosh, but 30% faster and it can do Direct Memory Access, which the Mac can't." Working within these hardware limits, the software team (led by Alex Limbaris) implemented an unusually comprehensive array of features. I was particularly impressed with the auto-loop implementation and asked Bob how that came about. "It was so hard to get a good loop on the Mirage that we, especially Tom Metcalf [the 'company sampler' - Ed.], learned a lot about what goes into making a good loop. The design team took all that expertise and built it into the software. Basically, it's like having Tom Metcalf in the EPS, except it doesn't burp."

Ensoniq seems to have also developed a bit of a personality among the many users of its products. The Transoniq Hacker, for example, is a magazine devoted totally to Ensoniq and its products - yet has no affiliation with the company whatsoever. Bob talks with enthusiasm about the Hacker, and considers it "A great publication - we don't have any control over them at all, and I wouldn't have it any other way."

Unlike Commodore, where Bob was pretty much the main guy for the Commodore-64, here he is quick to point out that the design work is very much a team effort, from the software to the hardware to the (in my opinion, excellent) user's manuals. He is reluctant to take much credit at all for the EPS and other instruments, preferring to correct interviewers about how much he didn't do on these machines, and to cite the exceptional contribution made by the software designers. How was this different from designing the C-64?

"What happened was that Jack Tramiel (the head of Commodore at the time) came in a couple of weeks before the Consumer Electronics Show and decided he wanted to have a new computer to exhibit there. I basically threw the thing together, and Bob Russell - who did a lot of the VIC-20 software - took the code kernal from a VIC-20, and there we were. What has it sold now - five, six million units? I don't think anyone expected that to happen at all." Or as he said in a letter to Electronic Musician commenting on a C-64 program a reader had submitted, "It's gratifying to know that the computer is still serving a useful purpose for musicians." From what I understand, he never did get a bonus or anything for designing what is surely the best-selling home computer ever.

Ensoniq has now branched into Japan, and is in partnership with Felix Visser (a prominent figure in the Dutch music scene) for Ensoniq Europe. With the introduction of the SQ-80 and EPS, they're poised to go even further. The EPS in particular is a remarkable sampler; it's probably the most cost-effective sampler since... well, the Mirage. And, frankly, it delivers a lot more for £1695 than the Mirage did when it was introduced at £1695.

We'll let Bob Yannes have the last word. As he was concluding a press tour to show off their products, he responded afterwards to some of my comments about how much I liked the EPS. "You know, you can think you have something really great when you're designing something new, but you never really know until it gets out there and people respond. It's really been great showing these instruments around, and knowing that people like them."