Tuesday, October 23, 2007

Considerable Sounds: Analog Vs. Digital Circuitry - Which Is Better For Music?

Digg This!Netscape BookmarkReddit BookmarkDel.icio.us Bookmark

By DC Music Editor Benjamin New

There are 10 kinds of people in the world.

those who read binary, and those who don't.

The encoding of sound into digits has changed the media through which people experience music forever. Proponents of analog sound have much in common with the Spartan’s suicidal last stand. In the digital age the inevitable demise of analog technology is certain. Or is it? Let’s have a look at the characteristics of both the analog and digital worlds and see if we can make any qualitative decisions. Analog sound is continuous. Recordings made using analog circuitry have no missing information.

But digital recordings are like film in the sense that “snapshots” of sounds are played back in a succession that “fools” the ear into hearing them as moving. This is much like the way a film is a series of still frames fooling the eye into percieving motion. The drawback is that audio information is missing in between the “frames”.
Analog as mentioned is far superior on this particular count. No missing information whatsoever.

This is the reason that older albums if not remixed especially for digital sound are actually inferior on CDs when compared to their original analog release. This loss of information manifests itself noticeably in the upper midrange frequencies. How many times have you listened to a fantastic sounding LP, bought the CD version later and noticed the harshness of certain sounds? (Woodwinds and violins seemed to suffer horribly from this malady in particular). The missing audio between the frames of digital sound are the culprit.

Look at an analog soundwave compared to a digital soundwave.

Well of course things have and continue to improve. Producers and engineers are aware of the problem and compensate for it on most releases today. CD audio files have a sampling rate of 44.1 k. The sampling rate defines the number of samples per second taken from a continuous signal to make a discrete signal. ( analog to digital conversion).

For time-domain signals, it can be measured in hertz (Hz).

The inverse of the sampling frequency is the sampling period or sampling interval,
which is the time between samples. (The holes in the sound).

So in other words to continue our movie analogy,
CD audio is like a movie for your ears with 44,100 frames per second flashing across the screen.

Here is the lowdown on sample rates encountered in popular media.

Samples/sec Description
5500 One fourth of the Mac sampling rate
7333 One third of the Mac sampling rate
8000 Telephony standard.
11025 a quarter of the CD sampling rate,
or half the Mac
16000 Used by the G.722 compression standard.
18.9 k CD-ROM/XA standard.

22 k (22050) half the CD sampling rate,
32000 Used in digital radio,
TV work, also long play DAT and Japanese HDTV.
37.8 k CD-ROM/XA standard for higher quality.
44056 Professional audio equipment to fit an integral number of samples in a video frame. 44100 The CD sampling rate.
48000 The DAT (Digital Audio Tape) sampling rate.
96000 Professional Studio Recording Software

Another weakness of digital audio is that physical sound must be converted to digits and then converted back into sound again so that the listener can indeed hear it. These DACS (digital/analog converters) are most often the weakest link in the chain. A quality Dac runs about $2,000.00 U.S. so you can safely assume you don’t have any in your 99.00 player.

Inexpensive DACS create noise, distortion, and other degrading artifacts. The deficiencies we observe in digital sound may not really be the fault of the technology itself but in the conversion of that technology back into sound waves. In the last few years Professional sound studios have been using the newest versions of software such as Pro Tools and Steinburg Cubase that allow for recording audio in a 24 bit 96 kHz sampling rate (with 32 bit internal processing) ! This is 96,000 “frames” of audio delivered to the ear in a second. ( CD audio as mentioned is 44.1 kHz, and is 16 bit. Generally speaking the more bit’s, the greater the amount of audio information theoretically encoded in a given sample). This goes a long way to bridge the quality gap between continuous and discreet sound. Does it sound good? You bet! Unfortunately at the moment there are no devices available for consumers to hear this more pristine CD audio sound.

It seems consumers are happy enough with CD sound and no one is gambling on a new medium or player in the consumer market at the moment. (Manufacturers are still trying to get their act together on DVD Audio and SACD which we will discuss in a moment.) Meanwhile consumers have embraced MP3 players. Ipods and the like have blossomed, the fact that one’s entire music collection can be archived into a small player and easily carried around is indeed marvelous. However the sound is compromised by the inherent data compression used to shrink the audio files. On any serious playback system the degradation is obvious. Mp3 audio is not for an audiophile experience though it certainly has it’s place.

Sony and Phillips who invented the CD offer the only consumer high end audio devices using high sample rates. These are their SACD (Super Audio Compact Disks). This and DVD Audio which is similar, use an entirely different scheme to create and playback music. One of the few industry moguls to raise his voice against the digital format in its early days was the late Maurice Oberstein, an American who was head of the Polygram UK (and later Universal) label. "Do you realize we are giving away our master tapes here?" he asked at an industry event. At the time, everybody was far too busy counting the cash coming in from CD sales to listen. But as the advent of the recordable CD dawned, and a black economy in counterfeit and illegal trade flourished, Oberstein was proven right. Regardless of how we may feel about the music industry’s gluttony and poor management in recent years, the end result of file sharing and piracy is that a recorded song has less value financially today than any other time since the birth of the industry. This hurts artists as well as the businessmen.

Ideally music business and music artists partner to create product and both benefit. The implementation of this partnership worked well in many cases and not so well in others. The beef many have had with the business is that artists often did not benefit in proportion to the business folks and so many great artists were “taken to the cleaners“ by less than scrupulous record company executives. However, the golden age of rock music for instance was ushered in by cigar chomping “old guys” who did not particularly understand the music but were willing to take chances and see what happened. They hired the next generation of executives believing that they understood the new music better and would serve to continue if not improve market positions. The opposite ended up being the case. The new breed never took chances and acted as if their primary concern was supporting their drug habits. The new business model favored providing product that was cheap to manufacture and produce that would yield maximum profits with little investment.

The advent of the Compact Disk brought with it a tidal wave of profits for the industry. Why? There are 2 reasons primarily. First and foremost CDs were and are much cheaper to manufacture and package than LPs were, yet the public was paying considerably more for them . And secondly the size of the marketplace increased exponentially as people who had already purchased music in older formats purchased it again in the newer format. Record execs did not have to raise a finger to have their companies be highly profitable. When this gravy train came to a halt there was no one left in the business who had the vision or fortitude to deal with the problem in any reasonable way. They continue to stick their fingers in the holes of the Dyke as leaks and cracks appear everywhere.

Back to SACDs and DVD audio. Many have suggested that these new formats are merely an attempt to reclaim the copy protections the music industry has lost control of. This may be true to a degree. I think the music industry does want to control duplication of it’s products and has a right to do so to a degree. The fact that standard 44.1 kHz audio can be accessed and copied from these disks but not the high definition material seems quite fair to me. Ultimately if music isn’t profitable fewer and fewer people will bother to create it. And that would be disastrous. I see serious damage already as talented young people turn away from higher education in the arts in favor of more marketable professions, closing the door on perhaps the next John Lennon, Kate Bush, or Randy Newman. SACD and DVD audio do use high sample rates, but there are only 2 factories that are allowed to create the disks. The HD layer of an SACD disc cannot be played back on computer CD/DVD drives, nor can SACDs be created except by the 2 licensed disc facilities in Shizuoka ( for Sony) and Salzburg (Phillips).

The vast majority of SACDs are hybrid disks. That is to say they have 2 layers. One layer has conventional 44.1 kHz CD audio embedded in it while a second layer has the higher resolution/multi channel information. The disks are backwards compatible with existing players and drives which will read the CD audio on the lower level while a special player is required to read the high resolution audio. These players are also forward compatible in that if the High resolution layer is not present on a disk, the laser will read the conventional layer, so older CDs will indeed play on these machines. In fact most any type of disk is playable. Photo disks MP3 disks, etc.

Conventional lasers see "through" the HD layer in a SACD.

The first SACD players on the market cost in excess of $5,000.00 U.S. but now a player can be had for $849.00 and these players also play High Definition Video Disks. My guess is that when the price drops a little more, consumers may embrace this technology if just for the fact that they can play all types of media in addition to the High Resolution audio and video on a single machine. Multi channel audio has been tried before but was never more than a novelty for consumers. However today most people actually have the ability to take advantage of multi channel recordings already set up in their homes as home theatre systems. I think the future looks bright for the Sony and Philips SACD technology. The competing platform - DVD audio is not backwards or forwards compatible so I suspect it will go the way of the “beta” video tape. SACD players are not permitted to digitally output an unencrypted stream of DSD ( Direct Stream Digital) making it impossible to duplicate (at least at the moment, we all know for every lock there’s a lock picker). SACD audio is stored in a format called Direct Stream Digital (DSD), which differs from the conventional PCM used by the compact disc or conventional computer audio systems. DSD is 1-bit, has a sampling rate of 2.8224 MHz, and makes use of noise shaping quantization techniques in order to push 1-bit quantization noise up into inaudible ultrasonic frequencies. This gives the format a greater dynamic range and wider frequency response than the CD.

Promotional materials about SACD supplied by Philips and Sony suggest that the system is capable of delivering a dynamic range of 120 dB from 20 Hz to 20 kHz and an extended frequency response up to 100 kHz, although most currently available players list an upper limit of 80–90 kHz. Conventional wisdom is that most human hearing recognizes 20-20 kHz. So why does anyone care about sound waves at 70 Khz ? Well every sound ( except sine waves) possesses partials and harmonics. I believe that harmonics and overtones exist well into the frequencies you don’t actually cognitively hear but still perceive in some sense. And if they are missing you are subtly aware of that too. The process of creating a Direct Stream Digital signal is conceptually similar to taking a 1-bit delta-sigma analog-to-digital (A/D) converter and removing the decimator which converts the 1-bit bitstream into multibit PCM ( Pulse Code Modulation, the type of sound encoding used in Pro Tools and virtually all digital audio recording software) . Instead, the 1-bit signal is recorded directly and in theory only requires a lowpass filter to reconstruct the original analog waveform. In reality it is a little more complex, and the analogy is incomplete in that 1-bit sigma-delta converters are these days rather unusual, one reason being that a 1-bit signal cannot be dithered properly: Most modern sigma-delta converters are multibit. Because of the nature of sigma-delta converters, one cannot make a direct comparison between DSD and PCM.

An approximation is possible, though, and would place DSD in some aspects comparable to a PCM format that has a bit depth of 20 bits and a sampling frequency of 192 kHz. PCM sampled at 24 bits provides a (theoretical) additional 24 dB of dynamic range. Due to the effects of quantization noise, the usable bandwidth of the SACD format is approximately 100 kHz, which is similar to 192 kHz PCM. There are three types of SACDs: Hybrid: The most popular of the three types, hybrid discs include a "Red Book" layer compatible with most legacy Compact Disc players, dubbed the "CD layer," and a 4.7 GB SACD layer, dubbed the "HD layer." It is not uncommon for hybrid discs to carry the "Compact Disc Digital Audio" logo to show that the disc is CDDA-compliant. Single-layer: Physically a DVD-5 DVD, a single-layer SACD includes a 4.7 GB HD layer with no CD layer. Dual-layer: Physically a DVD-9 DVD, a dual-layer SACD includes two HD layers totalling 8.5 GB, with no CD layer. This type is rarely used. It enables nearly twice as much data to be stored, but eliminates CD player compatibility. Hybrid SACDs are the most popular type, they can be played on all existing drives but when played in a SACD player deliver the high resolution SACD sound. They are also multi channel and most SACD recordings are released in a multi channel mix. Does it sound good? Yes. I think most anyone would hear a difference through decent speakers.

As of April 2007, there have been over 4,500 SACD releases, a little over half of which appear to be classical. Notable artists who have released some or all of their back catalog include Aerosmith, Peter Gabriel, Genesis, Depeche Mode,Bob Dylan, Elton John, The Moody Blues, The Rolling Stones, and Nine Inch Nails. Pink Floyd's seminal album The Dark Side of the Moon (the 30th anniversary edition of 2003), The Who's seminal album Tommy (the 34th anniversary edition of 2003) and Roxy Music's Avalon (the 21st anniversary edition, 2003) were released on SACD to take advantage of the format's multi-channel capability. All three were remixed in 5.1 surround, and released as Hybrid SACDs with a stereo mix on the standard CD layer. Many people purchase SACDs even if they don’t have a player yet because the remix on the standard part of the disk is usually superior to previous releases anyway. Plus they figure at some point they will purchase a player. The other side of the coin is compressed audio. We love our mp3 players because we can fit so much music in such a small space. And take it all with us. As far as these compressed files are concerned the earlier formats used logarithmic encodings to squeeze more dynamic range out of fewer bits for each sample, like the u-law or a-law encoding in the Sun AU format. Modern compressed audio files use sophisticated psychoacoustics algorithms to represent the essential frequencies of the audio signal in far less space. Examples include MP3 (MPEG I, layer 3), Ogg Vorbis, and WMA (Windows Media Audio). Now I think these files sound pretty good but they are shy of being high fidelity. They refer to themselves as “near CD quality audio“ and that‘s right.

As mentioned before CD audio is 44.1 kHz. MP3 audio is theoretically capable of encoding sample rates from 1 to 200 kHz. However to my ear even when the sampling rates are higher, because of the data compression, I believe they sound noticeably inferior to standard CD audio. Even on computer speakers I think you hear the difference between a .wav file ( which is uncompressed) and the same sound as an MP3 file ( data compressed). But in terms of fitting a lot of decent sounding music in a small space, these files rule and do an admirable job. They are not however, audiophile grade material . Now DAT ( Digital Audio Tape) uses a slightly higher sampling rate than Cds (48,000 samples per second) and I thought the discerning ear could detect the difference. In terms of consumer audio, Dats never caught on even though they were arguably a superior medium and pretty much a standard in pro audio studios. Historically Nashville was the first to begin doing digital masters. L.A. and New York resisted either because they had so much invested in their analog ways or didn’t see the merits of digital recording.

In the mid 1980s my own band Macchu Picchu recorded one of the first non-Nashville digitally mastered LPs. Released on Europadisk, you might find it in a “cut out” bin somewhere and it‘s actually a very good album. However I remember the attraction to the new technology. There was no signal degradation through transfers, a lower noise floor, and the highs were more present. The downside was the highs may actually be unrealistically harsh. The warmth of analog sound was gone. Anyone who has had the pleasure of listening to the playback of a high-end analog magnetic tape running at high speed with Dolby Spectral knows digital audio isn’t up to this lofty audio standard yet. But exquisite analog systems are very pricey. For instance, a nice high end analog amplifier like the Quest 300B Monoblock pictured here will set you back about four grand.

High quality analog power amplifiers.

Of course you’ll need two for stereo. And these babys are the “budget” versions of their top models such as the $60,000 Ongaku (Japanese for music or sound & joy) and the $95,000 Gaku-On (more music!). You’ll need a good pre-amp too. $35,000 for the M8 pre-amplifier. Hey where are you going? We haven’t even shopped for speakers yet! Why the cost? Well the majority of components are hand made and hand wired for one thing. The entire assembly except for the tubes themselves are made in house one at a time (no mass production). The materials used are expensive and the labor is long and requires great skill. There are very few people who know how to make this all happen. For example all circuits and connections are coated with silver or gold ( the 2 most conductive metals… also 2 of the most expensive.)

The application of silver to equipment must be done strategically, because carelessly silver-wiring parts of a machine will only expose limitations of remaining non-silver components. Silver lacing an entire device will certainly bring about a wholesale improvement but will also skyrocket the cost.
If this were a car, it would be a Ferrari or Lamborghini not a Hyundi. What do components like this provide? In the words of Peter Qvortrup who owns the company, as stated on their website… “an attainment of specific fidelity, low level detail, dynamic contrast, bass control with pressure expansion, intensity of piano attack, timbral separation between instruments, general presence, immediacy and presentation”. In other words you can hear the dandruff falling off Ringo’s hair. Detail.

If you are interested there are several essays about the mission of Peter’s company here. Analog offers warmth and musicality along with a natural sound "Analog" is "analogous" to real life. In real life everything is "continuous". Take the temperature of a place. The temperature of a place varies over time - say yesterday it was 25 degrees and today it is 20 degrees. This change of temperature does not take place all of a sudden - the temperature goes through all of the temperatures in between 20 and 25 degrees when it changes. There are an infinite number of temperatures between 20 and 25 degrees. At one point of time the temperature was 20.1 degrees, at another point of time the temp was 20.5 degrees, at some other point of time the temp was 20.578 degrees, and at some other point of time the temp was 20.78900748 degrees. In general, between any two temperatures, however close they maybe, there are an infinite number of temperatures. This is the idea of "continuity". Analog offers a continuity in the physics of sound .

But what digital audio offers is a longer life expectancy ( digital recordings do not degrade over time as analog recordings do), a far lower cost, and a wider frequency spectrum with more dynamic range. "digital" systems cannot understand continuous things. they only understand only "discrete" signals - e.g '1', '2', '3' etc. It cannot understand anything between 1 and 2 - not even 1.5 . In electronics it is convenient to represent numbers like 1, 2, 3 etc using the binary system because '1' and '0' can be conveniently related to "ON" and "OFF". (In normal day to day life we use the decimal system. The use of the decimal system can be traced back to the presence of ten fingers on our hands - which obviously proved to be very convenient also.) In the binary system one is 1, two is 10, three is 11, four is 100, five is 101, six is 110, seven is 111, eight is 1000 - and so on. (The idea is to try to represent a number using only 1s and 0s.) All digital equipment - microwave ovens, washing machines, traffic lights to calculators and computers use the binary system. Discrete audio signals are inherently flawed. That being said, the closer they can emulate analog in terms of continuous variability the better they sound. ( After all sound itself is analog in nature!) This can only be achieved by higher sample rates.

We must remember that analog is a mature technology while digital is a toddler. And that some of the absolute magic that vacuum tubes and tape saturation offer in terms of sound quality are colorations that we find pleasant, not truly analogous with physical sound. Of course there’s no going back to analog I suppose, and in reality recordings today are hybrids for the most part, as we love the old Neve recording boards, tube preamps , and so many of the outboard signal chains include analog and tube circuits. Let alone that most musicians would not consider firing up an amplifier to play through in the studio unless it had tubes. Both technologies have their pros and cons. But digital offers us better editing capabilities. Higher sampling rates may get us back on the path to true audiophile bliss and go a long way towards fulfilling the burgeoning promise of a luxurious digital future. I for one truly hope so.

Duly Consider and Considerable Sounds are TM of this publication and are subject to liabilities thereof


St Dark said...

Great article overall!
Do realize, though, that Denon released PCM recordings as early as 1972, and those didn't come from Nashville. The Philips/Sony "compact" disc (called that since it was smaller than the currently existing 8" commerical optical disc) took Denon's PCM work (which leapfrogged off of Bell and NHK's work) and packaged it in a format small enough to fit in a dashboard radio.

Also, older recordings transferred to CD in the earlier days often sounded bad as they were not remastered to eliminate the preemphasis, or compensatory EQ applied to get around the higher frequency issues with vinyl and most cassette. Uncompressed digital won't roll off highs, it will just lose detail.
Don't get me wrong, analog is great- I still fire up the Thorens 'table when I can!

Anonymous said...

Great piece! As I was quoted saying when the CD was introduced "Digital preserves music the way formaldehyde preserves frogs: you kill it, and it lasts forever!" Vinyl still rules!

-Michael Fremer
senior contributing editor, Stereophile
editor, www.musicangle.com

Related Posts with Thumbnails