Digital Audio Discs
How the LP went in for a transplant and came back a DAD
An explanation of the technology behind the new sound recording revolution: how the LP went in for a transplant and came back a DAD!
Once upon a time there was a circular piece of plastic spinning around quite happily at 78 revolutions per minute. He didn't mind that the speed wasn't always constant, or that people stuck pins into his grooves and even went so far as to turn him into flower pots when they got fed up with his musical repertoire. After all, he knew he was just about the most exciting thing since sliced bread. But, over the years, he found himself spinning slower and slower, getting thinner and thinner, becoming noisier and noisier, and, to cap it all, found that his skin was developing crinkles, which necessitated painting himself with bright colours to disguise his poor condition. So, realising that his end was nigh, he buried himself away in a laboratory for five years and a day, cast the magic spell, pulled the power switch, and lo and behold: the son of senile vinyl was born. This bumptious youngster rotated at 1.25 m/sec, never had pins stuck in him, and couldn't be made into flower pots. But, seeing that he was set to take over the world, his creator put him in wraps under a cloud of speculation. Then, one day in the autumn of '82, he was released on the poor, unsuspecting world...
In reality, the digital audio disc (DAD) seems certain to need a fairly generous helping of public support if he's to achieve his desired aim in life, and, given the current economic condition, he probably couldn't have chosen a worst time for his parturition. Fortunately for us, it seems unlikely that he'll have much in the way of competition from rival siblings, as, for once, the battle of the technological giants seems to have occurred long before the expected delivery date. This still leaves a fundamental question to be answered: will he deliver all that his genes seem to suggest?
By the early 70s, several major companies were working on various methods of recording the wide bandwidth of video signals on to disc and tape in formats supposedly convenient for public assimilation. In Europe, Philips had shown the first optical video disc as early as 1969, but others were not far behind. The general target was to produce consumer video disc and tape recording systems commercially comparable with the LP and compact cassette. By the mid 70s, several companies had started work on the concept of the digital audio disc, but much of what followed proved more frustrating than fermentive.
The Telefunken Teldec system resolutely stuck to keeping a stylus in contact with the disc surface. In this case, the 'stylus' was a wedge-shaped plunger attached to a piezoelectric transducer generating a voltage in response to digital hills (1s) and dales (Os) on the disc. The JVC variable capacitance system looked ingenious, but, in practice, there were many problems involved in controlling the stylus tracking and ensuring conductive discs of adequate pressing quality. Philips instead went for a then revolutionary design using an AlGaAs laser that was able to track pits etched on to a disc a fraction of the size of a normal disc. The success of this idea lay with some extremely sophisticated servo mechanisms, and, at that time, it wasn't possible to pack all the necessary electronics into a convenient LSI package.
Sony were also engaged on their own digital research, but, rather than going their own way, decided to team up with Philips in 1979. In 1980, the two giants introduced the format of the Compact Disc, a mere 12cm in diameter and capable of storing more than an hour of stereo on each side. Some of the main changes to the original Philips design that resulted from this merge of minds were a subtly altered sampling rate (44.1kHz rather than 44.3kHz), an increased bit code length (16-bit rather than 14-bit), improved error correction, and an optical readout system using an advanced semiconductor laser (see Figure 1).
Much of the hard engineering work for a compact disc player had already been carried out in bringing video systems to fruition, but this meant there was a considerable danger that some manufacturers might be tempted to use their own knowhow and jump the Philips-Sony gun with their own systems. To prevent this from happening, 49 companies from Japan and elsewhere set up a special DAD committee in late 1978. Between then and December 1980, the following aspects for getting a system off the ground were considered: basis of signal extraction, including optical, capacitance and mechanical methods; signal specification, including sampling frequency, number of bits, redundancy and modulation system; functional features, including the number of audio channels, access methods, playing time and disc diameter. By April 1981, the die had been cast and the final recommendations were made to the Japanese standard bodies; the Compact Digital Audio Disc was duly elected.
The clouds of speculation surrounding the DAD were at least partially dissipated by the specifications laid out for the system (Table 1). In the absence of playing with the real thing, they certainly provide plenty of food for thought! The hours-worth of digital information per single-sided disc is stored as 5 x 1012 bits on a spiral track, which, if unwound into a straight line, would extend for something like 2¼ miles! In addition to the bits used for the coding of the sound signal, another chunk is added to cope with error correction and system control. Each bit is represented, on the disc, either by a flat surface representing a '1', or a microscopic pit, representing a '0' (so that's what McEnroe's "pits of the world" outburst referred to: being zero in his digital estimation!). The 16-bit PCM (Pulse Code Modulation) code that these bits represent allow 216 or 65,536 different sample (analogue) values to be provided for — a far cry from the paltry 28 or 256 steps of resolution in use with most digital synthesisers. Even though quantisation at the 16-bit level results in superb accuracy in converting audio signals to digital code and back, the increased number of bits per word necessitates a very efficient means of shifting all this data. It was fortunate for Philips and Sony that the semiconductor laser was going through a parallel development at the same time as the initial DAD research, and provided an ideal though problematic solution to the system requirements for rapid data transfer.
As the disc rotates, it is scanned from the underside, starting at the centre and moving to the outside, by a spot of laser light kept in place by a leading guide track. This beam detects the sequence of pits and flats at a rate of approximately 4.3 million bits per second (4.3 Mb/sec). The pits themselves are just 0.6um wide, 0.2um deep and between 1 and 10um long; compare this with the 50um width for the average LP groove or human hair! The output from the optical pick-up is in the form of the 16-bit PCM code. A digital-to-analogue (D/A) converter decodes the data stream word by word and synthesises them into a conventional audio stereo signal. To ensure that defects in the disc or in the player do not affect the quality of the signal, considerable protection is built in via the coding scheme. One technique of correcting errors used in the system is the so-called 'interpolation code', by which redundant information is introduced into the music signal code and then used for making corrections if decoding 'checks' indicate that some bits have got lost on the way. In fact, this technique has some parallel with the way in which genes operate in ourselves and everything else that's going about the business of life. Here, though, the code (a 4-digit one) is found in the stuff of genes, DNA, and stretches of this code include redundant or repeated sequences to protect against the havoc caused by a bolt of gamma rays poking a hole in one's genetic blueprints. And, talking of poking holes in things, that's precisely what the DAD error correction is capable of coping with; not just a gamma ray sized-hole but one that's been caused by drilling a 2.4mm hole through the disc! I can just see all the eager beaver Sony demonstrators going around with power drills and generously carving up every Compact Disc that comes in reach...
All this technology is impressive but one thing that's still under further development is an integrated D/A converter. So far, at the various demonstrations of the DAD player around the world, an additional box of more or less discrete circuitry has been added to the system. Sony U.K. inform me that the first totally integrated versions of the converter and servo circuitry are practically complete and should be at a pre-production stage by the end of the year.
It is rather curious talking about something that's only as tangible as a photo of a mock-up and a wadge of specifications. I suppose patience will be rewarded in the long run, but it's very difficult to predict what the public and record industry will make of the Compact Disc. A conservative estimate of the time for take-over in this country is put at about six years. Back in Japan, though, Sony have been planning their campaign for at least this amount of time, as is evident from the progress report below:
October 1976: Digital audio processor for recording and playback on VTR (12-bit, 2-channel).
September 1977: PCM-1 digital audio processor for recording and playback on VCR (13-bit, 2-channel); digital disc system using direct PCM-encoded signal (900 rpm, 60-minute play, optical laser pick-up).
May 1978: X-22DTC ¼-inch, stationary head, digital audio recorder (38 cm/s, 12-bit, 2-channel); digital FM broadcasts in Tokyo.
October 1978: Long-playing PCM disc (450 rpm, 180-minute play); PCM 3224 stationary head professional 24-channel PCM recorder (1-inch tape); DMX-800 professional 8-channel digital audio mixer.
May 1979: PCM-10 domestic digital audio processor and PCM-100 professional digital audio processor; DEC-1000 digital audio editor for use with audio processors.
October 1979: PCM-3324 professional stationary head 24-channel audio recorder (½-inch tape); PCM-3204 stationary head 4-channel audio recorder (¼-inch tape).
May 1980: Studer and Sony agree on standard format for stationary head digital audio recorders.
June 1980: Philips and Sony embark on joint development of the Compact Disc.
October 1980: Prototype of Compact Disc system shown at Japan Audio Fair.
February 1981: Production begun of Compact Disc mastering system, including PCM-1610 professional audio processor, DAE-1100 digital audio editor and DRE-2000 digital reverberation.
As things stand at present, about eleven different hardware manufacturers are gearing up to producing Compact Disc players, including: Matsushita, Sharp, Sanyo, Nakamichi, Onkyo, Philips, Saba, Rotel, Revox and Pioneer. None of this would make much sense unless discs themselves are made available. The CBS/Sony plant in Japan aims for 100 titles by the launch date in the Autumn of 1982. These should then be followed by titles from the Pioneer plant in Japan and the Polygram plant in Europe. Where disc production goes after this is anybody's guess, and I wonder whether the ailing record industry here will do anything other than the usual British thing of burying their collective heads in the sand. The one glimmer of hope is the VHD disc pressing plant that Thorn-EMI are starting up in January 1982, and with a bit of luck may be adaptable to pressing Compact DADs — if adequate pressing quality can be assured.
Finally, the cost. Well, fortunately, it should be somewhat less than the complex electronics might suggest. Sony won't be led into giving a 'yes' or 'no' to the various figures that are floating around, so their quote of "the price of a top-class record deck" is as far as we're going to get for the moment. Mind you, system development won't stop with the standard domestic Compact disc player. The miniscule size of the disc, coupled with the dogged determination of the laser to keep on tracking come what may, makes it a very practical and attractive proposition to develop a car and Walkman version of the player. This is Sony's next line of approach: total aural domination. Tomorrow, our ears; next year, our minds?
Feature by David Ellis
Previous article in this issue:
Next article in this issue:
mu:zines is the result of thousands of hours of effort, and will require many thousands more going forward to reach our goals of getting all this content online.
If you value this resource, you can support this project - it really helps!
Please note: Our yearly hosting fees are due every March, so monetary donations are especially appreciated to help meet this cost. Thank you for your support!