Everything Still Matters

Soon after the previous entry in this blog—Everything Matters—was posted, I heard from a friend who recently purchased a 24-bit, 192 kHz, high-resolution download of a classic album.  Like many of us, he sought an even better “view” of the recording than is offered by the CD version he already owns.  To his surprise, he prefers listening to the CD version, and finds the high res download as sounding “a bit bright.”

The authors of some recent tech website articles denigrating high resolution might see my friend’s comments as vindication.  In my view, this says more about the authors than it does about the audible reality.  Why these websites didn’t choose authors more experienced with systems for music playback, and more interested in sound quality, remains a mystery.  (Vide John Atkinson’s very well considered Access Journalism vs Accountability Journalism.)

In order to determine whether high resolution is the source of the problem (any problem), it must be compared with its standard resolution equivalent.  This means for a valid comparison of delivery formats the only difference must be the delivery format.  Both versions must be created at the same mastering session, by the same engineer, using the same channel (signal path).  There, as the man once said, is the rub.  In most cases the two items being compared were created at different mastering sessions, often by different engineers, in completely different mastering studios.  Right away any sort of equivalence is out the window.

Different mastering engineers have different ears, different sensibilities, different approaches, different talents, and different weaknesses.  Even the same engineer might take a diverging tack when remastering something they’ve mastered in the past.  When the two versions are done by different engineers the likelihood of variance in their methods is pretty much a sure thing.  This is expectable since they don’t share a common set of ears, and no two engineers I know of will do things the same way.  With regard to new masterings, in Everything Matters I said, “Sometimes the results are improved sonics, with newly revealed nuances from the original recording that were lost in the original mastering.  Other times, and sadly all too often, the remastering is simply a louder, brighter rendering.”

There is also a very good to excellent chance the signal path for the two versions differed.  Even in the same studio, things tend to change and evolve over time.  For an album like the one my friend purchased, which was originally recorded on analog tape, the A-D converter used in mastering can have a profound effect on the results.  This is particularly true at higher resolutions, where I have found many converters are stretched beyond their capabilities.  To wit, a lot of converters specified for 24/192 actually perform worse at this rate than they do at lower rates.  This I attribute to the significantly increased demands on clocking accuracy and on analog stage performance at the wider bandwidths.  It would seem to be easy to put 24/192 on a spec sheet but not so easy to design a device that can perform to the potential of the format.  And the converter is just one of several components comprising the signal path, each of which will have its own sonic consequences.

All of the above assumes the same source tape was used for the different versions.  This is a big assumption, even when “original” is claimed.  I’ve experienced a number of instances where, having handled the tapes myself, I knew the subsequent claims from some quarters of “original” were at best mistaken. Whether original or not, if different source tapes were used, the outcome could be acutely altered.

The bottom line here should not be surprising: A carefully made CD (or CD resolution file) will easily outperform a not-so-carefully made 24/192 file.  This has to do with how effectively the capabilities of each delivery format are realized—or not realized, as this case illustrates.

I concluded Everything Matters by saying “Everything after the microphones (i.e., mic cables, AC mains power, AC mains cables, mic preamps, recording format, recording device , mix, if any, mastering, playback format, playback device, interconnecting cables, amplification, speaker cables, speakers, speaker positioning, vibration isolation, room acoustics, etc., etc.) merely determines how much of what was captured the listener gets to hear.”  In my experience, when everything in the production of an album is the same except for the delivery format, a 24/192 file should reveal so much more of the source as to make the 16/44 (CD) version sound coarse, ill-defined, airless, and broken by comparison.  So either my friend’s 24/192 file was created from an inferior source, or the mastering was just not up to that achieved for the CD.

To my ears, properly done digital audio at 24/192 fulfills the promise digital made back in 1982 when the sonically hamstrung CD format made its first appearance.  I have said elsewhere that 24/192 is the first format I’ve ever heard where I have not yet been able to distinguish the output from the input—the first format I know of that is capable of giving us a virtually perfect rendition of the source.  In view of this, I must admit to being somewhat astonished at the negativity from some quarters of the tech web and tech press.  Nevertheless, if music lovers are to receive the benefits of this wonderful fruit of technological progress, the folks creating it must tend their crop more carefully.

Everything Matters

In my earliest days as an assistant in a recording studio, I learned the ropes as basic tracks were laid down on tape, overdubs and “punch-ins” were added, stereo mixes were created, and the final, edited mixes were mastered.  From those early days onward, I frequently heard what I came to refer to as The Three Most Heard Phrases in Record Making:

“We’ll fix it in the mix” (when something in the recording was not quite right)

“We’ll catch it in the mastering” (when the mix didn’t exactly fix it)

“They’ll never hear it at home” (when it still wasn’t nailed in the mastering)

Rumor has it that Frank Zappa once said “We’ll fix it in the shrink wrap.”

There has been some discussion on the Internet recently about which aspect of record making has the greatest influence on the sound of the finished product.  Some say the space in which the original recording is made makes the greatest difference.  Others say “it’s all in the mastering.”  Asking three people will elicit at least four different answers.

The space where the recorded performance occurs will impact the musicians, influencing the sound they hear from the instruments and from the environment, thus influencing how they play.  But with typical recordings, where the microphones are placed close to the instruments, the sound of the environment is often minimized.  Alternatively, some recordings are made using very few microphones, which are placed to deliberately pick up the environment as well as the players.  Here, the space becomes in integral part of the final sound, as it is when a listener attends a live performance.  Still, this does not suggest the space has the greatest influence on the sound.  One might be in a fabulous room but the wrong microphone selection or the wrong microphone technique can very easily override the sound of the space.

Typical major label recordings involve three main production phases: recording, mixing, and mastering.  There are those who feel the mix—where a multitrack original is turned into a stereo (or in some instances, surround) version—is the biggest factor in shaping the final sound.  Instrumental and vocal level balances are determined during the mix.  Tonality is often adjusted, placement of instruments and vocalists on the “stage” is determined, and special effects might be added.  There is no doubt about the magnitude of influence the mix has.  Yet the final result will always be influenced by the ingredients used to create it.

With the advent of the remastering phenomenon, labels have given new life to older releases.  Sometimes the results are improved sonics, with newly revealed nuances from the original recording that were lost in the original mastering.  Other times, and sadly all too often, the remastering is simply a louder, brighter rendering.  Either way, through these re-releases, many music aficionados have discovered the influence mastering has on the sonics of a recording.  As the last stage of production and the first stage of manufacturing, mastering is the last opportunity to make small—or large—changes to the sound.  While the degree of influence on the final sound can sometimes be quite surprising, here too, the final outcome is always fundamentally impacted by the source material.  A pachyderm’s aural appurtenance does not a silken wallet make.

To my ears, an mp3 of a Keith Johnson recording, listened to in the car, on the highway (with the windows open!), can reveal more Life than many a typical studio master played on the finest, most carefully set up systems in the most optimally treated rooms.

In my experience, once the signals are leaving the microphones, 90-95% or more of the recording’s ultimate quality ceiling has already been determined.  Everything after the microphones (i.e., mic cables, AC mains power, AC mains cables, mic preamps, recording format, recording device , mix, if any, mastering, playback format, playback device, interconnecting cables, amplification, speaker cables, speakers, speaker positioning, vibration isolation, room acoustics, etc., etc.) merely determines how much of what was captured the listener gets to hear.

Two worlds, heroes and real stereo

In the September 26, 2013 entry entitled Why doesn’t it sound (in here) like it sounds out there? I mentioned a growing awareness in my earliest days as an engineer that what I’d previously thought of as the world of audio was in fact two different worlds.  On the one hand there was the audiophile world that suggested recordings could be made and played back with the aim of recreating the sound of real musicians, playing real instruments in real spaces.  On the other hand was the professional world in which I’d been working, where the idea of sounding real did not seem to be a frequent consideration.  Some of the audiophile gear, particularly among the loudspeakers, was fantastic at approaching the sound of real music.  In contrast, the studio monitors were capable of playing at extraordinary volume levels.  Many of the recordings popular among audiophiles excelled at capturing a sense of real life, including not only the sounds of the instruments but the spaces in which the recorded performances occurred.  In contrast, the studio recordings, many of which captured their own magic, never allowed the listener to suspend disbelief.  They sounded like recordings where some audiophile recordings sounded instead like the music itself.

As my skills and experience grew, I came to feel each of these worlds could learn something from the other, to the mutual benefit of all concerned and especially the listener.  Yet how often I was reminded of how little interaction there tends to be between pro and audiophile worlds.  To this day, it seems to me that all too many pros often miss opportunities for much higher quality audio, while at the same time, all too many audiophiles lack an understanding of how their records are made.

One of the best examples of the latter is the oft mentioned desirability in the audiophile Internet fora of “flat transfers”, the idea being that the finished masters used as sources for replication of the finished product should have no equalization or other processing applied.  (The term “flat” suggests no frequency equalization is applied – nothing to “tilt” the frequency response – but the term can also imply no other processing as well.)  In my early days as an engineer, I felt the same way.  Then I had the opportunity to hear how most master tapes actually sound.  When one considers the types and number of microphones, where they are typically placed, the signal path in most studios and the monitoring (of which I have already spoken in earlier entries), it should come as no surprise that most master recordings need help.  Put another way, if a recording was made with sufficient treble energy to bring on a headache in the listener and if I can make that recording hurt less by the judicious application of frequency equalization, I would think EQ is in fact, a good thing.  Indeed, a tiny minority of very well made recordings are best served with no EQ at all but most studio recordings will benefit quite significantly from some well considered equalization.

Now I can understand that sometimes, the reason a recording might hurt so much is precisely because of EQ – bad EQ, perhaps because the engineer was trying to make bad monitoring sound right.  This goes back to The Questions I mentioned in the previous entry.  Is the EQ being applied to address a flaw in the recording or is it mistakenly being used to address bad monitoring?  This question should have been preceded by “How trustworthy is the monitoring in this studio?”  Whatever the reason, when heard on a system capable of getting out of the way sufficiently to allow one to hear the recording itself, careful application of EQ can be used to repair at least a good part of the damage done the first time.  If the recording can be made more listenable with EQ, should EQ be avoided simply because it has been misused elsewhere?  I’d first ask “How artifact free is this equalizer and the settings I intend to apply?”  (That last question, like most of the others, turns out to be critical for those interested in making high quality recordings.  A big part of the reason many have come to think of EQ as bad is that among those who use EQ, the question is almost never asked.  Even on the best equalizers, the wrong settings can cause sonic problems.)  With a positive answer, I would elect to apply the EQ.  My experience has supported this approach over the course of hundreds if not thousands of recordings.

I have always very much enjoyed it when an audiophile sensibility entered into the pro audio world and pointed to what could be.  Some of my fondest memories of my years at Atlantic involve the time I spent with the great mastering engineer George Piros.  George would often tell me of his early days working with Bob Fine and Wilma Cozart on their recordings for Mercury Living Presence.  I had not heard of these recordings before and when I followed up to find some of them, I found a great many joys both musical and sonic.  C. Robert (“Bob”) Fine was to become one of my engineering heroes and one of my inspirations insomuch as he made his stereo recordings with only three microphones.  George was to become one of my mastering heroes for his preservation of musical dynamics.  He remains one of the tiny handful of mastering engineers I can name who did not routinely apply dynamic compression to his signal path.  Everything George mastered, from Bob Fine’s classics to much of Atlantic’s classic jazz, rhythm and blues, and rock recordings has a sense of Life to it.  One of my favorite memories was made one day when walking past the outside of his mastering room.  I heard loud music through the heavily padded, “soundproof”, double-door “airlock”.  I decided to visit and upon entering the room, saw George leaning over the microscope of his lathe, examining the fresh groove he was cutting into the lacquer disc, while AC/DC’s music virtually peeled the paint from the mastering room walls.  For me, it remains one of the great moments in rock.

Not that George was an audiophile.  He was just one of those folks who could accurately intuit what a recording needed and apply it to get the results he wanted.  He did not mince words with regard to the program material or some of the tools popular in audio engineering circles.  George was famous among those who knew him for his “Piros-isms”, his unabashed commentary that would sometimes include language that would make a marine drill sergeant blanch.  But he was more famous for the same honesty he brought to his work and that honesty brought some audiophile “names” up to his mastering room.  Through George, I was introduced to Bert Whyte, whose monthly “Behind the Scenes” column was a favorite of mine in all the years I’d been reading Audio magazine.  Bert was the engineer on the great recordings for the Everest label.  (Like Bob Fine, he too used only three microphones and created fabulous results.)  I got to take home a test cut of a Whyte recording of Stravinsky’s “L’Histoire du Soldat”, a personal favorite, though I’d never heard it sound so good before.  On another occasion, another “name” from the audiophile world dropped by to work with George: Joseph Grado, inventor of the stereo moving coil phono cartridge, creator of the moving magnet cartridge I was using at home at the time and, as I learned, an operatic tenor too!  I remember being in the mastering room with both discussing monitors.  Joe said “You see Barry, George uses these studio monitors with no complaints” to which George responded “You mean those pieces of #$^%?”

I feel more than fortunate to stand astride both audio worlds and to have learned a great deal from each.  In an effort to find a synthesis of both worlds, to find the underlying unity which I felt to be larger than either, in 1987 I decided to become an independent engineer and formed Barry Diament Audio.  The learning opportunities once more expanded geometrically.

As I did not yet have my own studio and mastering work was starting to come in, I sought out studios where I could rent time, my prime criterion being monitoring I could trust.  As may be concluded from what I’ve said in this and previous entries, this was not an easy task.  In the end, I found a small number that were willing to accommodate my request for certain monitoring arrangements.  They would have to do until the time came when I designed my own room.

In keeping up with contacts in both the pro and audiophile worlds, an opportunity arose to visit with the editor of one of the audiophile publications I was reading.  I knew his reference playback system was reputed to be among the best.  What I was not at all prepared for was the fact that after being an avid music lover and audio enthusiast since childhood, after having done pro audio work in a number of studios and after having read all the books and journals on the subject I could find, I was going to hear stereo for the first time.

My conception of stereo before that evening was probably a lot like that of other folks, based on what we’d been “taught” over the years and what we’d heard on the old stereo demonstration recordings.  There might be a piano on the left and a guitar and bass in the center and drums on the right.  There might be a marching band proceeding across the room from one speaker to the other.  Most of the time whether in folks’ homes or in audio dealerships, I’d seen stereo speakers placed as far apart as a room would allow, often in the corners.  I thought I’d made great progress when I found that moving the speakers out of the corners and away from the walls resulted in much improved sonics.  What I didn’t realize is that I was still listening to a pair of what might effectively be mono sources, playing together.  There was “sound from the left speaker” and “sound from the right speaker” and some sound in between.  It was nice and it was fun but as I came to learn, it wasn’t stereo.

Among the first things I noticed when I visited that evening, aside from the jaw-dropping gear I’d only seen in magazine photographs, was that the speakers were, to my mind, “in the middle of the room”.  They weren’t just out of the corners and off the wall, they were well out into the space.  There was lots of room all around them.  I’d never seen a setup done this way.  One of the first records played that evening (we were listening to vinyl) was an old, well worn Leonard Cohen album.  The track featured his voice, accompanied by acoustic guitar.  This was not a super record, just an ordinary studio production and an old copy of the record too.  What I heard was something entirely unexpected on my part and something entirely new to me.

First, there was no “sound from the left speaker” and “sound from the right speaker” and some sound in between.  To my ears and brain, there were no speakers at all.  I had a scarily distinct sense of the artist and the air in the studio around him.  It was almost as if I could see him sitting on a tall stool in a large room with the lighting turned down.  (Of course, I have no idea whether he was sitting or standing and what level the lights in the room were set to but this was the impression created in my mind by the sound alone.)  Now some of that experience must be attributed to the gear itself.  The loudspeakers were outstanding at “disappearing” for sure, as they rightly should have been for their six-figure price and the commensurate associated gear to which they were connected.  But as I was to learn, their placement played a commanding role in allowing the system to achieve its potential.  And applying what I learned that evening to other, less extravagant speaker designs would similarly unleash their potential in ways that were new to me.

Stereo by definition implies solidity and hence three dimensions.  Properly done, the listener does not hear sound from the speakers.  On the contrary, the speakers seem to disappear and the entire part of the room in which they reside comes alive with the audio equivalent of a hologram.  The sound occurs on a stage (a soundstage) and the images upon that stage too, occur in three dimensions.  On the finest recordings containing such information, the listener can perceive a sense of depth, with for example, the instruments in the back row of an orchestra seeming to emanate from well behind the wall behind the speakers.  (In order for the listener to perceive them from a good pair of properly placed loudspeakers, these spatial cues must of course be captured in the recording.)

While this was all news to me, I later found the idea for proper placement of stereo loudspeakers dated back to the 1950’s in an article Peter Walker wrote for the English journal Wireless World.  Many know Walker as the designer of amplifiers and electrostatic loudspeakers marketed under the Quad name.

Monitoring was something I’d long recognized as critical in any recording or playback situation, yet this recognition existed for many years before I had the opportunity to hear real stereo for the first time.  Now I was starting to apply this newfound knowledge in my own listening room and in the studios I worked in.  (I wrote a bit more about speaker placement in an article called Setting up your monitoring environment.)  This was also making me think anew about how to capture real stereo in recordings.