July 1, 2010

Audio Evolution: 1950-2010

The trigger

Recently, while reorganizing my library, I perused an article in the September 1983 issue of The Abso!ute Sound (Vol.8 No.31), "The Threat of the Compact Disc to the Sound of Music," by renowned mastering engineer Douglas Sax, of Sheffield Lab Recordings. Overwhelmed by curiosity and nostalgia, I resurrected a blue T-shirt I’d bought at the 1984 Winter Consumer Electronics Show, in Las Vegas. Sax’s commendable crusade and its catchy slogan, "Stop Digital Madness," had received sustained support from skeptical audiophiles.

In concluding his piece, Sax wrote, "For me, all digital attempts thus far have been a failure. I simply cannot enjoy music that has been digitally processed, and the enjoyment of music in the home is the sole reason we have a high-fidelity industry. I support analog recording because it works.

"It is a time-proven process that contains musical information which is accessible to all and which has a resolution that allows the listener to continually discover hidden nuances as he improves the abilities of his home playback system."

The words thus far in that first sentence, reinforced by his description of analog as being "a time-proven process," prompted me to conclude that Sax, like most audiophiles, would eagerly anticipate future research and development into the optimization of digital sound, as audiophiles continued the quest to hear more and more, until the resolving capabilities of home playback systems approached its horizontal asymptote of live musicians performing in real spaces.

Sax’s article was published at a point in audio’s history halfway between the birth of stereophonic high fidelity and the current level of refinement of analog-to-digital conversion techniques. In the 27 years since that article’s publication, what sort of evolution has taken place? How much more are we hearing today at home, and how far away does utopia remain?

But first: Where and when did stereo begin?

Alan Dower Blumlein

British electronics engineer Alan Dower Blumlein was born June 29, 1903. In 1931, while working for EMI, he invented stereo, a discovery regarded as being so revolutionary and ahead of its time that it was shelved. Blumlein then focused on television, and was on a design team which produced the first modern TV system, adopted by the British Broadcasting Corporation in 1937.

At the beginning of World War II, Blumlein switched to telecommunications, and by 1942 was a key member of a top-secret team designing a revolutionary radar system that would enable the Royal Air Force to bomb from above the clouds. However, the flight of a Halifax bomber outfitted with the system ended in a crash in Wales on June 7, 1942, killing Blumlein and ten others.

According to his widow and son, Blumlein struggled with spelling, and could not read until age 12. But he was clearly a genius who, according to one distinguished colleague, "had he lived would have been considered the Faraday of our age." Blumlein is remembered for his technique of recordings with a coincident pair of microphones, to overcome some of the inherent deficiencies of spaced pairs of mikes.

The digital blueprint

We are all aware that a plethora of high-resolution formats exists today. However, in 1983, the limitations of early digital sound frustrated such engineering gurus as Doug Sax, Keith O. Johnson, Jan-Eric Persson, Jack Renner, John Eargle, and their colleagues worldwide. In that year a unique technology had emerged, and was by then about one year old. In it, a laser beam of high-intensity coherent red light scanned under the surface of a shiny disc of polycarbonate. This 120mm-diameter compact disc (CD) rotated at a speed that continuously varied from 200 to 500rpm in playback devices known simply as compact disc (CD) players.

In contrast to an LP, the laser would begin its tracking of the data spiral from the center of the disc, moving spirally outward from track to track, toward the disc’s outer edge. A first-generation CD accommodated up to 74 minutes of playing time, almost twice what was stored in the two grooves of the average LP.

The laser made sense of the data by discriminating between the pits and land encrypted on the medium. A resultant bitstream of pulse-code modulated (PCM) digital data, converted into electrical signals, was amplified and reproduced as sound by a pair of loudspeakers. It was as simple as that. Or was it?

Late analog vs. early digital

Few doubted that digital formats offered potentially significant advantages over their analog counterparts. Everyone wanted to hear pristine recordings with lower noise floors, wider dynamic range, higher fidelity, and no surface noise.

In addition, mastering engineers preferred the many obvious conveniences of digital editing on computers. This technique was considerably friendlier and far more precise than the razor-blade process used in the editing of analog tapes.

The digital restoration and archiving of analog recordings would require substantially less storage space. Prolonged shelf life, coupled with ease of maneuverability and retrieval, would be guaranteed. However, the (in)famous marketing slogan of Philips and Sony, coinventors of the CD -- "pure, perfect sound forever" -- was certainly inappropriate.

In common parlance, the problem with first-generation digital sound relates to the phenomenon of number crunching. Since digital is essentially a statistical process, it will always be an approximation compared to analog. Evidently, it was not commonly realized or accepted in those early days that a sampling rate of 44.1kHz at a resolution of 16 bits was woefully inadequate for optimal conversion of sound waves to binary code.

At that time, it was impossible to digitally encode all or even a substantial portion of the information contained in analog signals, especially the signals of audiophile recordings. The latter sounded special because the best of them were recorded in real time, direct to two tracks, and pressed on virgin vinyl from half-speed masters.

High-resolution recordings include enough information to reproduce, on playback, subtly defined soundstages. Ambient information gives spatial cues to listeners by revealing nuances that make them sound eerily similar to live microphone feeds. Such ambience is usually several decibels lower in volume than the music, and must be eked from the recorded source using audiophile playback systems.

Psychoacoustic determination of size of space, depth of image, timbre, placement of instruments and musicians, is essential for good listening. Such information must be accurately captured on whatever storage medium is used. True fidelity of reproduction would be obtained only when digital formats could sample the live sound more often; in other words, when they could record at a higher level of resolution.

Those with golden ears were obviously alarmed by digital’s very sterile sound. This was coupled with unpleasant residual artifacts and distortions superimposed on music. Their verdict: Digital had been insufficiently field-tested.

The vital ambient information described above was stored in the least significant bits of the analog-to-digital conversion process. However, this data was inadvertently truncated, similar to how numerical values are rounded off in mathematics.

As a result, the "perfect sound forever" campaign set off loud alarm bells and unprecedented pother. It may never be known if Sony and Philips ever actually believed in their slogan.

Audio evolution and revolution

After World War II, several innovations created favorable conditions for substantial improvements in audio recording and playback quality. Technology developed for the military and aerospace industry gradually filtered down into commercial products. This was an era in which vacuum-tube electronics, long-playing microgroove vinyl records, open-reel tape recorders, and multiway loudspeaker systems reigned supreme.

The term high fidelity, or hi-fi for short, was universally used to describe equipment, records, and tapes intended to provide faithful sound reproduction in the home. Even in those early days, knowledgeable consumers paid detailed attention to technical specifications. Some preferred to assemble sound systems comprising individual components. Hence markets were created for separate turntables, tuners, preamplifiers, and power amplifiers to drive separate loudspeakers. All-in-one systems secured an even larger market share.

Names such as Marantz, McIntosh, Bozak, Thorens, and Wharfedale became well known. The evolution of the transistor into an electronic device for use in hi-fi gear instigated heated debate because it sounded and measured different. It seems that the respective intrinsic characteristics of tubes vs. solid-state (i.e., transistor) devices will forever remain an emotionally subjective and divisive issue.

Audio magazines published feature articles about these developments, and reviews of new products, and consumers began to depend heavily on them for guidance in making purchases. Well-known publications such as Wireless World, High Fidelity, Stereo Review, and Audio were eventually superseded at the very pinnacle of high-end audio by newer competitors both in print and online. Audio shows, such as the summer and winter CESes, and specialized events like Munich’s High End and Canada’s Festival Son & Image grew in popularity.

Magazines flourished through advertising and online sales. The birth and growth of the Internet brought forth tremendous innovation in publishing online, and publications such as those produced by the SoundStage! Network began to prosper. Consumers now expect new products from manufacturers at least once per year. This expectation has propelled the industry.

Meanwhile, engineers, scientists, and marketing executives were feverishly at work. More precise high-resolution digital formats evolved, all of them based on higher sampling rates and longer word lengths to capture or interpolate and store all of the missing information that first-generation digital converters misplaced.

Everyone was excited, and there were encouraging prospects for a bright future. Optimization of digital knowhow would mean goodbye forever to analog’s intrinsic problems of restricted dynamic range, surface noise, pops, wow, and flutter.

The future

Analog-to-digital conversion systems such as Direct Stream Digital (DSD), High Definition Compatible Digital (HDCD), and Greater Ambience Information Network (GAIN) have gradually gained niche status. They are now readily available if you know where to look, to the delight of discerning audiophiles. But audiophiles continue to be restless.

HRx is a trade name for high-resolution audio WAV files on DVD-R data discs. Developed by Keith Johnson for the independent label Reference Recordings, it has been commercially available since 2007. HRx delivers exact bit-for-bit copies of Johnson’s HDCD master recordings, made at a resolution of 24-bit word length at a sampling rate of 176.4kHz. It is already being acclaimed by many as the ultimate in fidelity for two-channel sound.

Audio’s digital revolution seems to be finally finding a sound footing. In addition, significant advancements in signal processing have made it possible to synthesize a reasonably good approximation of the sound of a concert hall at home. Room-correction techniques have become commonplace and reasonably priced.

Ultramodern recordings and playback systems can deliver the full 96dB of dynamic range of a symphony orchestra to your sweet spot. Thousands of audio-related patents have been granted to engineers, scientists, inventors, and innovators all over the world. The industry is sustained by pioneers driven relentlessly by passion, unselfish dedication, and a burning desire to hear more. They have made sterling contributions to the fabulous and fiercely competitive domain affectionately and reverently known as high-end audio.

I am honored to have met and talked with the likes of Peter Perreaux, Keith Johnson, Bob Carver, Ralf Ballmann, Marcel Riendeau, Michael Pflaumer, Noel Lee, and Andrew Payor. They all have one thing in common: They are driven, not by the almighty dollar, but by their love of music and pursuit of excellence.

However, analog is still very much alive, as evidenced in the recent and continuing resurgence of LPs, and the turntables, tonearms, and cartridges to play them with. Very soon, Andrew Payor intends to launch his Rockport Technologies System V Sirius turntable, with its patented tangential-tracking, air-bearing tonearm and colossal price tag. One thing is certain: Many members of the high end’s so-called lunatic fringe anxiously await the arrival of this engineering/architectural masterpiece.

On the digital side, Berkeley Audio Design’s Alpha DAC is one of the latest toys to have attracted the attention of reviewer and consumer alike. Members of its design team, the inventors of HDCD, claim that "Unequaled interpolation technology upsamples 44.1kHz CDs to produce almost 176.4kHz quality and produces superb fidelity at all sampling rates from 32kHz to 192kHz."

Conclusion

The output quality of high-end audio recordings is approaching a threshold. The industry has evolved from its preferred format of retrieving information stored in the grooves of long-playing vinyl records using a phonographic pickup stylus. It is now beginning to move on from music stored in the pits and land of polycarbonate CDs retrieved using a laser beam. Playback of high-resolution encoded data files from hard disks running on computerized music servers is now commonplace. What else will avid listeners request? Time will tell. Perhaps Doug Sax’s famous slogan should be modified: "There’s No Stopping Digital Madness."

Dedicated to my grandson, Matthew Sandiford.

. . . Simeon Louis Sandiford
simeons@soundstagenetwork.com