In the music industry, there are two types of people: those who cannot wait for every new software update and aspire to own every piece cutting edge technology, and those who prefer the tried and trusted, only accepting 'new' when it has really proved itself as 'better'. Tim de Paravicini most definitely falls into the latter category. But he is no Luddite: he believes in the steady improvement of recording and playback technologies. "With an analogue system, you can go on and on improving, just like the motor car industry. Cars today are better in most respects than their 50s and 60s counterparts. New models are a logical development of their predecessors." This, de Paravicini believes, is not the case with digital music technology. He blames virulent marketing: "Because music is such a continuous event, it is difficult to do with digital, because the brain is very sensitive to glitches and minute errors in the high frequency range. If you scratch CDs they won't play, unlike vinyl. When they demonstrated the robustness of CD on Tomorrows World in 1981 they wiped strawberry jam on, but it was a con - they put it on the opposite side to the data. The major CD promoters destroyed heaps of vinyl in order to advance the CD with its larger profit margin. Vinyl disc has the instant access of hard-disk recording, and I've got vinyl pressed in the 50's that is still acceptably quiet. Whatsmore, a modern cassette machine damn nearly gives a DAT machine a good run for its money. Standards should be carefully thought about and implemented". He is also well-known for his preference of valve technology: "In the early days of recording, valves were expensive, therefore designs tended to feature less of them. This discipline led to cleaner signal paths. When cheap transistor technology came along, this simplicity went out of the window."
Following the introduction of the new Zonal 999 tape (previewed in the August SS) and approving comments from de Paravicini, a demonstration of his top-flight 1-inch analogue stereo recorder using the new tape was arranged. Present were Adam Francis, Consultant for Professional Audio to Zonal and Sam Hann, Zonal's Sales Director. The setting was Dave Gilmour's Astoria Studio (courtesy of studio manager Phil Taylor), appropriate not only for its wonderfully accurate monitoring, but also the huge amount of EAR/de Paravicini equipment, including eight compressors and a de Paravicini-modified mix buss on the Neve VR Legend desk.
Description
The 1-inch format mastering machine is built using a Studer C37 as a donor. This workhorse º" machine was popular with broadcasters in previous years. Redundant machines are therefore readily available, stripped and rebuilt "to the last nut and bolt" to de Paravicini's demanding specifications. This is not just a transport modification: all electronics are rebuilt to the highest standards. de Paravicini has also rebuilt C37s as ½" machines due to customer demand. But since 1986, de Paravicini's entire 1-inch machine output comprises a mere eight machines. This looks likely to change, with three mastering houses in the US acquiring machines within a year, giving more people confidence to use the format. Famous users include Bill Bottrell (for work with Sheryl Crow and Michael Jackson). More recently Aerosmith mastered their double-live album to 1-inch. Producer Jack Douglas, on auditioning the format, reportedly exclaimed, "It sounds huge, man!" Perhaps unsurprisingly, competitors have sprung up in the USA pinching the format idea. de Paravicini dismisses their machines' electronics as inferior to his, although their use of alternative 'donor' machines interests him. "If starting from scratch now I would use a more modern transport", he says. Indeed, there is a possibility of a future collaboration with these 'upstart' competitors who will rebuild Ampex ATR transports while de Paravicini takes care of the electronics.
In Use
Tim sets his own standards when it comes to line-up. Despite using high level tape, he sets 0VU at 200NWb/m. "I use 999 for the cohesiveness of the sound, because with the low noise floor of the 1-inch, the headroom is relatively irrelevant. It gives a safety margin during live recording and reduces print-through." I was initially concerned about hiss we could hear, but this was soon traced to our input source rather than coming from the 1-inch tape, which was very quiet. de Paravicini typically recommends over-biasing by nearly 1dB less than is conventionally recommended, giving improved HF distortion figures.
We listened to a range of source material, switching between input and off-tape monitoring. The sound was undeniably huge and vibrant. One all-digital recording seemed subtly enhanced off the 1-inch tape, and superb dynamics and frequency definition were witnessed by all present.
Design Philosophy: 1-inch
So how much does the machine cost? de Paravicini: "In the order of £9,000 ready to go. If you spent a gazillion dollars on it I could do better still, but the only thing that will equal this sonically would be a 24 bit 400kHz true linear digital system: both systems will then satisfy the hearing mechanism." de Paravicini points out that he wrote in 1982 of the need for 400kHz 24 bit and yet today the big players are still discussing 96kHz and perhaps 192kHz as possible formats. He points out that with the rapidly falling prices of digital storage media, multiplying to a 384kHz sampling rate is not unreasonable and absolutely ideal. Until that happens, however, there may well be a market for machines such as these. Current de Paravicini-modified C37s run at 15 and 30 IPS. However, one future improvement he proposes is the seemingly oddball change to 18 IPS, which he is proposing in an AES white paper. "I have agreement with all the owners of my 1-inch machines," he says. Apart from possibly frightening potential purchasers, this has yet to be implemented due to the notional compatibility with other 1-inch formats for emergency editing or playback. Presently, de Paravicini recommends 15IPS with bass response down to 7Hz within 3dB for minimum phase distortion in 40-100Hz region, important, he says, for rhythmic elements. de Paravicini is a great believer in the importance of frequencies above and below the conventional 20Hz-20kHz range.
He points out that one can sense the standing wave of a large cathedral which is below 5Hz. The brain knows this as our body senses sound through other parts of the body than the ears. "This traditional notion of the hearing stopping below 20Hz is absolutely rubbish. We detect sound down to the resonant frequency of the body, about 3Hz", argues de Paravicini. Deaf people certainly seem to "hear" through the body. "Likewise, the high frequency notion of 20kHz is also rubbish. In essence, we detect audio up to about 45kHz. We can't say we hear it as a tone, but we are certain that something is going on. My method of demonstrating this was several years doing work on ultrasonic bath cleaners. Everybody in the room suffered the after-effects of tinnitus. You are aware of this excruciating feeling going on. My speakers go up to 40kHz. If I put a 20kHz tone from an oscillator through them, most people in the room are in discomfort. With a 24kHz tone the younger people in the room are in discomfort and discerning it quite obviously. So traditional myths have to be thrown out the window. I wanted a tape machine that roughly embodies what the hearing mechanism is about: 15IPS enables a range of 7Hz to roughly 40kHz within 3dB with a good running time. At 30IPS it goes from 14Hz up to 80kHz theoretical, 60kHz in practice, but performance at 15IPS is already better than most digital systems and perfectly adequate for human hearing."
His proposed 18IPS standard degrades the bass by a negligible amount but gives 45-50k top end and is arguably the best compromise. I asked Tim about the improvement over ½-inch. Surprisingly, he explained the difference in terminology more commonly related to digital equipment: "The number of magnetic particles on the tape with ½-inch is roughly equivalent to 23-24 bit while 1-inch is 24-25 bit. This is necessary as the ear can discern distortions as much as 80dB down. Also, modulation noise is lower than ½-inch, with better bottom end solidity." Although emphasising the importance of large playback heads for extended bass response, de Paravicini accuses major multitrack manufacturers of the crime of "cost-unbalanced engineering". He maintains that corners are often cut with playback electronics. "Many 1950's recordings have response up to 40kHz - recording these frequencies is relatively easy compared to playing them back." de Paravicini has designed playback equipment for Mobile Fidelity to enable remastering to the highest standards. Poor playback electronics overload the high frequencies and cause what he refers to (onomatopoeically) as "spitching", with LF and HF artefacts. "When mic noise is the noisiest thing then that's fine - mics inherently have at least 6dB more noise than human ear. But in most cases the mic is not the weakest link - it is more likely the mic amp, console or monitoring."
Zonal 999
Naturally, the best tape possible is important to the success of the 1" analogue format. de Paravicini is particularly impressed with Zonal 999's packing, which improves HF phase and quality. This tape holds together well with no deposit on the heads visible during our tests. de Paravicini points to historic shedding problems of other tape manufacturers. And this tape is much more fluid and pleasant to handle than a certain competitor's product. Rather alarmingly, de Paravicini suddenly grasped and crumpled a foot or so of tape with his hands, to demonstrate another advantage of the 1-inch format. It is twice as difficult to stretch compared to ½-inch. To coin a phrase: "If you even touch DAT tape you're up the swannee without a creek!" Interestingly, Zonal does not make any digital tape. Sam Hann explains: "We decided that with our experience, we are so committed to analogue we will probably stay to the bitter end - if there ever comes one. Anybody can produce digital, there is no art to it. Analogue, there is actually an art to it. You can vary the performance of the tape by using different oxide. What you've got to get right is the actual formulation. If you don't get the formulation right you are likely to have shedding, flaking, etc. That's the art of making tape. Digital is a different ball game altogether. And with analogue, you can do so much with it. This is why I am inviting Tim to help us. If he were to make a suggestion there are ways of tweaking it, with digital you can't." Hann says that the only customer who makes suggestions for specifications are the BBC, so perhaps with de Paravicini's increasingly accepted standing in the industry he is now in a position to influence future tape formulations. de Paravicini welcomed the invitation, immediately musing on metal formulations, and promised to give the matter some thought.
As to the future, de Paravicini is working on a 35mm audio format with advantages of easy mechanical lock to film, and space alongside the sprocket holes for timecode. Featuring a roughly similar track width to 1-inch he claims the quality of reproduction will be equal. He also has ideas for a multitrack format with increased track width.
I brought up the much-discussed current hot topic of surround-sound. de Paravicini is not a fan. "Good true stereophony should give you every dimension on playback if the speakers were perfect. You don't need more than two channels since God only gave us two ears." Our discussion even extended to the advantages of the old analogue mobile phone network compared to the new digital systems. However, for de Paravicini the future looks bright, the future looks like brown tape…
Thanks to Astoria's Cyn Taylor & Damon Iddins for their hospitality.
Reproduced with kind permission from www.georgeshilling.com. Copyright ©1997-2001