Philips Research

The history of the CD - Technology

The technology of the Compact Disc is very complex, but in general it can be divided into 'digital data processing', 'optics' and ’mechanics. These disciplines allowed a system to be developed more than twenty years ago that was superior to any other concept, and that since then has grown into an enormous family which could never have been imagined at the time of the introduction.

The development of the Compact Disc was first made possible by the invention of the laser diode, which is an essential part of the Compact Disc and all other optical recording systems. The basic principle is that a fine laser beam is focused on a surface that contains digital information in the form of tiny pits. Since the surface of the disc is reflective, the laser beam is reflected with the pattern of the pits to a photodiode, after which the signal can be detected and converted into analogue audio information. This means there is a non contact readout system, which cannot damage the information carrier, so that a Compact Disc in principle has an unlimited lifetime. This form of readout is highly reliable, even if the CD is worn or damaged. The CD also gives excellent reproduction quality, with negligibly low wow and flutter, a high signal-to-noise ratio and a wide dynamic range. In terms of performance the switch from analogue to digital audio processing is a more far reaching one in the history of the gramophone record than that from mechanical to electronic recording and reproduction at the beginning of the last century.

While analogue technology allows a signal-to-noise ratio of only 60 dB or less, and a low channel separation of less than 30 dB, the Compact Disc offers a much higher performance. The digital signal processing means that both the signal-to-noise ratio and channel separation are higher than 90 dB. Because a 1.2 mm thick transparent layer protects the digital infor¬mation on the Compact Disc, damage and dust are not in the focal plane of the laser, which is used to ‘read’ the disc, so that they have little influence on reproduction. And most of the faults that arise can be corrected, thanks to the digital signal processing. This is possible because the information stored on the disc also contains error correction bits. If there are so many errors that correction is no longer possible, these can still be detected and ‘masked’ according to a defined procedure. So the chance that consumers will hear any of the clicks that are so well known from LPs is virtually eliminated.

The high information density means that a playing time of originally around an hour, and now up to 80 minutes, can be achieved on a disc with an outer diameter of only 12 cm. And because the disc itself is so small, the players can also be very compact. An additional feature is that a CD can also contain 'control and display' information in the form of so called C&D bits. These allow user information to be added such as the number of tracks and the playing time, as well as the names of composers and the titles of the tracks (CD Text).

The information density of a Compact Disc is relatively high, and is related to the wavelength of the laser, which is used, as well as other parameters. To allow as much information as possible to be stored, it is therefore important to use a laser with the shortest possible wavelength. When the Compact Disc was developed, the available infra red lasers had a wavelength of 780 nm (nanometer). The red lasers with even shorter wavelengths (650 to 635 nm), like those now used for DVD and other systems, were then not yet available.

The pits in a CD are 0.6 micron wide (1 micron is 1/1000th of a mm), 0.12 micron deep and 0.9 to 3.3 micron long. A disc full of these pits, with a track pitch of 1.6 micron, has a capacity of 650 to 700 MByte. Since the information is recorded on a spiral track, and is read out at a constant speed of 1.25 m/s from the inside to the outside of the disc, the rotational speed decreases as the disc is played from 500 to 200 r.p.m.


Oversampling
With the use of oversampling technology - which allowed the 16-bit Red Book standard to be met with the 14 bit digital-to-analogue converters that were available when the system was introduced - a very high sound quality was obtained. The only difference, in the Philips players, is that this is temporarily multiplied by a factor of four. As well as allowing 16 bit-equivalent performance, this also offers additional benefits. For example it gives a higher signal-to-noise ratio and dynamic range. And instead of requiring a steep analogue output filter, it allows a relatively simple analogue filter to be used to suppress the remaining interference signals. Because even digital filters still allow some interference signals to pass through, which is why every player with digital filters also has an analogue output filter. The higher the degree of oversampling, the more effectively that a digital filter can suppress the interference signals, and the simpler the analogue output filter can be. Which in turn benefits the final reproduction quality. It’s also a positive factor that the use of a digital filter has no audible effects, while a steep analogue filter causes phase changes, which affect the overall reproduction quality.

However Philips’ oversampling technology, originally born out of the necessity to use the early 14 bit D/A converters, and dismissed as a 'technical joke’ by other manufacturers who believed that a true 16 bit D/A converter followed by a steep analogue filter was the only way to go, was quickly embraced by most manufacturers of CD players. Because it meant there was no need to use highly complex analogue filters, while at the same time it allowed the often serious non-linearities of the D/A converters that were available at the time to be concealed.


Conversion
The conversion of the digital ‘zeros’ and ‘ones’ into an analogue signal also proved to be a tougher challenge than was at first thought. And it was also very difficult to keep the conversion process linear at lower signal levels, for example between -60 dB and -100 dB.
At the introduction of the CD player, every player had a so called 'ladder' D/A converter, followed by a steep analogue filter to remove frequencies above 20 kHz. Philips was the only company to use four times oversampling, with a digital filter, right from its first player. Because four times oversampling means that four samples are taken every 1/44,100th of a second instead of just one, this in combination with first-order noise shaping, which Philips was also the first to apply, allowed 16 bit resolution to be achieved with a 14 bit D/A converter.


Bitstream conversion
At the end of 1988 Philips was the first to introduce 'Bitstream conversion', a highly advanced technology that uses 256 times oversampling and avoids the digital distortion such as that arising with ladder converters. The result of this process is a 1-bit data stream, which is then converted into an analogue signal in a 1-bit D/A converter. This consists of a relatively simple network of capacitors. In practice the linearity or accuracy of a 1-bit converter is significantly higher than that of traditional (ladder) 16 bit converters.