About Re-Mastering (by Harvey Reid)

About Re-Mastering

Harvey Reid has played and taught guitar for 50 years, was a former national Fingerpicking Guitar Champion, and has released 32 highly-acclaimed solo recordings of original, traditional, and contemporary acoustic music. In 1980 he wrote the first college textbook for folk guitar. He lives in Southern Maine.

The Beatles catalog was remastered in 2009, and since I have been remastering a number of titles from my old back catalog I thought it might be a good idea to explain a little bit of what that means. It is extremely important, yet subtle, and very few people understand what is going on. It's quite interesting how much is going on "under the hood" even in a simple solo guitar recording.

[NOTE- before I get started... ] The other word that applies to cleaning up and "modernizing" old recordings is "remixing," which unfortunately has been hijacked by the hip-hop music people and generally is seen to refer to incorporating all or part of an old recording in a dance-mix digital collage. Most recording since the 1960's has been done with multi-track machines, and the word "mixing" refers to adjusting the relative volumes, tone and stereo positioning of each of the tracks. In analog days this used to be done by hopefully alert engineers who moved the knobs while the song was playing to do things like make the harmony vocal go up and down or to boost the saxophone during its solo. "Automated mixing" consoles showed up in the 1980's that allowed users to do this and save the changes so that they could then modify and fine-tune the mix and do very minute and precise things. For nearly a decade now the majority of mixing has been done digitally, so that physical moving of knobs by engineers as the song rolls has largely disappeared, though a number of engineers still insist on doing their mixing through a console and not a computer. The mixing process allows each of the tracks to be placed anywhere in the left to right stereo spectrum (or even to move around in some cases) and each of the tracks can be adjusted for tone (usually called EQ) and volume (often called GAIN.) Re-mixing can only happen when you have access to the multi-track raw recordings, and can thus run each of them through a channel in the mixing board, and make a new set of adjustments to them. Some of the processes associated with mastering and re-mastering are also involved in mixing and re-mixing, such as EQ and reverb. In mastering, the changes can only be applied to the mixed track, whole song or whole album. In mixing, you can change things on one particular track within a multi-track recording.

Mastering is a vital but little-understood process that happens after each of the songs is mixed, to prepare the finished album for duplication and distribution. It involves putting the songs in order, adjusting the fades at the end of songs, the length of the spaces between them, and the very-important issue of absolute volume and relative volumes of the songs. EQ adjustments are often made during mastering, and sometimes reverb and other effects are added. A properly mastered album flows nicely from song to song, and when listened to in various situations: high or low volumes, in a car, in a resonant room or a "dead" room, on big or small speakers or headphones-- needs to sound the best it can. You don't want the listener constantly reaching for the volume knob to turn parts of it up or down. (In an ideal world, you might master an album for small speakers or big speakers, but this is not an ideal world.) If a song is too quiet, during a background-level listening, it can disappear entirely, and likewise a song that is too loud compared to the others can "jump" out and be annoying. Trying to listen to a symphony at dinner is an example of this. The quiet movements are inaudible and the loud movements drown out dinner conversations. Different instruments, because of their tonal properties have a different "presence" at high or low volumes. My guitar can sound really loud in a nice studio, but practically disappear on a car speaker on the road. My autoharp, though, does not sound that loud in the studio, yet it cuts really well and jumps out of the tiny speakers in a gift shop. A melody played on low end of the guitar might vanish in the same setting. When I record with Joyce Andersen, her voice and fiddle behave very differently in different listening environments than do my voice and guitar, and the mixing and mastering processes try to create an optimized sound that addresses all these concerns.

An odd thing has happened in the world of recorded music over the last 20 years or so. Though the ability of recording equipment to capture all the nuances of sound has increased, this increase has not showed up in the finished recordings. For decades, in analog recording, there was a delicate balance involved, where you would set the mike levels carefully so that the loud parts of the music would not "pin the needles" and cause distortion from being too loud. Likewise, if the music was very quiet in parts, you would hear the hiss of the tape or background noise in the room. When digital recorders showed up in the mid-1980's they seemed to offer a quantum level better "signal-to-noise" performance, and for the first time, it was possible to capture a dramatically wider spectrum of dynamic range in recorded music. Theoretically, digital music has about 30 db more "headroom" or dynamic range than analog. CD's also have a lot more dynamic range (96db) than vinyl (68 db), and feeding too hot a signal or one with too much bass into a record needle can make it literally jump out of the groove. Classical and jazz music fans applauded when digital arrived in the mid 1980's, and it looked as though recordings were going to be able to sound more like "real life" than ever, and allow the musicians to play and sing dynamically and expressively into the mikes and be captured. I got into this realm myself, and my early digital recordings, starting with "Of Wind & Water" in 1988 (#104) were done "direct to digital" and thus captured the dynamic reality of my music better than previous technologies. With CD players not adding their own "surface noise" it seemed that music recording had indeed gone to a new level. However, another more powerful and essentially opposite force soon entered the arena: compression and limiting. (I wrote a whole essay about compression that covers a lot of this same turf.)

Compression refers to a "squeezing" of the dynamic range of the music by lowering the loud parts and also boosting the quiet sections, and "limiting" refers to only chopping off the peaks of the loud parts, though quite often both compression and limiting are referred to as compression. Record labels and radio stations figured out quickly that if you could lower the volume of the loudest notes in a recording (the so-called "spikes" because of the shape of their sound waves) a few decibels (db's as they are called in audio) then you could raise the overall level of the song that much higher. The more you remove the peaks of the music, the louder the whole song can be. Engineers had always done a little "spot compression" or modest limiting to make music more manageable and "civilized" but once a lot of records started being released that were heavily compressed, they overwhelmed all the uncompressed music out there by being significantly louder.

This is why various CD's are various volumes when you put them into an iTunes playlist or on shuffle in a multi-disc changer, and artists like me who had thought that people might like to hear the music as "life-like" as possible found that their CD's were not being heard, especially in places like restaurants where the music was playing at background levels. (It became sort of like nuclear deterrent logic, or the sport of wrestling, where you had to starve yourself down to a much lower weight class simply because everyone else was doing it. If you did not, you would face a much larger opponent who would mop the mat with you.) The result of the so-called "loudness wars" is that a commercial CD, which theoretically has 96db of dynamic range, only uses about 6db of that between its loudest and softest parts. The equipment to "squash" the music has gotten better and cheaper, and record labels even have employed the technique of putting a big record out to "bid" to a number of mastering engineers, and whoever made it the loudest got paid to do the job. This encouraged enginners to compress the music much more than they even wanted to, and has resulted in a landscape of pop music that essentially has had one crucial element, namely dynamic range, removed entirely. It causes music to "jump" out and be noticed, but it also causes listener fatigue very quickly, and many tests have found that music listeners always prefer music with dynamic range. Many audio researchers believe that over-compression of pop music is as much as fault for the decline in record sales as is the internet and other better-known factors. People simply can't help get tired of listening to modern pop recordings after a few songs.

The compression process not only removes the dynamic range of the music, but it also changes the tone of voices and instruments noticeably. As a musician whose whole life has been about producing and recording a strong, rich sound in the acoustic stringed instruments I play, I am very sensitive to this. I can't stand to listen to the "O Brother Where Art Thou" soundtrack, for example, which has been heavily compressed to the point that the mandolins and guitars sound awful to me. I can't stand to listen to satellite radio play acoustic guitar music either because the tone is painfully anemic, and I usually end up listening to honky-tonk country or 60's rock instead because it doesn't sound as bad. It reminds me of what I heard someone say about a bad-sounding acoustic guitar pickup many year ago: "If that is what an acoustic guitar sounded like, I never would have started playing in the first place."

Digital technology has allowed compression to become smarter and faster and less noticeable, and the compressed music has now completely taken over in all forms of recording, as well as in many stage performances. Interestingly, radio stations have for many years compressed their signals heavily in a similar kind of volume war with other radio stations, and because so many listeners listen in cars and in the background. Modern pop music is generally so compressed now that whispering and shouting are the same volume. When a singer belts out a huge note it is no longer louder than the rest of the song. Like most things, compression has advantages and disadvantages, and though a lot of the "life" and excitement is being systematically squeezed out of music, it also allows the music to be more "user friendly" in situations like listening in a car or at a dinner party. A friend of mine played in the Les Miserables orchestra for more than 10 years on tour, and I went to see the show one night. I noticed immediately how great the music and the singers were, but mentioned to him that if they had turned off the compressors I would have enjoyed it a lot more. He said "I remember to the day when they first insisted on introducing the compression in the live show. From that day on the audience stopped leaping to their feet instantly at the end for a standing ovation." The audience had no idea what was being done to them, but they were less moved by the compressed music, but the show producers had controlled and "tamed" the singers and the band.

The issues and choices are very sinister-- heavily compressed music really jumps out and is very "up front" and "present" and initially seems to be more noticeable, but after extended periods of listening it becomes tedious and one-dimensional. In truth, the music has had one of its dimensions (dynamics) removed, and the puzzle is that at first it can sound really great. When you listen to something like "pop" flavored acoustic music like Alison Krauss, Taylor Swift or Mark Knopfler (which are heavily compressed) along side a more "traditional" folk or bluegrass CD, the compressed music is much louder, and you can hear the individual elements quite clearly. The soundtrack to the "O Brother" movie is a great example of how good bluegrass can be squashed flat for mass consumption yet sound horrible upon close listening. Compressed music seems deceptively more appealing, like it is easier to listen to, since the lead vocals seem to be right in your ear, and the harmony vocals and solos are also easily audible. But something very critical has been removed, and you don't realize at first that it is missing. It's a trade-off-- and by removing the spikes and the ability of the music to startle or annoy you with its "punch" the mastering engineers have also removed a vital part of the music's ability to move us emotionally. It is a "de-clawing" and "civilizing" process, and it unfortunately removes a vital element of the music's ability to affect us, and at the same time it becomes louder and more noticeable than music that has not been "de-fanged."

I was listening recently to Taylor Swift's "Speak Now" CD, which is a terrific example of what has gone wrong. Especially the acoustic songs like "Never Grow Up" sound fine as long as you keep them way in the background, but when you turn up louder they sound almost freakish and inhuman. Every emotion and possible increase in volume is squashed flat as a pancake, and if you try to listen to the human who is singing and playing the guitar, it sounds almost alien. I can only hope that some kind of pendulum has swung as far as it can go, and that listeners will somehow demand that they get the dynamics back in their music. Consumers can't possibly understand what has been done to them. Emotion is a big part of music, and volume/dynamics is a vital element of that emotional element.

The other side of this odd coin is that uncompressed music in an environment like a car, where there is a constant drone of background noise, can be annoying in a different way, since loud parts jump out and quiet parts disappear altogether. I have concluded that a modest amount of compression often improves the average listening experience, but that the excessive compression that has become essentially mandatory in modern pop music becomes intolerable. It removes a vital part of the human art and "essence" that is supposed to be the driving force behind the music. It also bothers me greatly that so many artists who play dynamic, acoustic, non-pop music feel the need to copy the pop producers, and vote to squash their music also, and remove a vitally important component that might actually draw discriminating listeners to their music.

The most unfortunate thing about this near-takover of the music by compression is that ultimately, it should be an end-user decision. All the problems could be solved if the end-user simply turned up the volume on quiet songs and turned it down on the loud ones. We should all have a compressor knob on our listening equipment, and we should decide how much to squash the music we listen to. It can be useful and it's not all bad. But compressors are still very expensive, so we don't get to make that decision, and we also don't get to choose to buy the uncompressed version of the CD we want-- they only offer one, and there is no way to uncompress it at the user-end of the music. Dolby was cheap enough that manufacturers of all tape decks were willing to pay a few dollars to Dolby Labs and have it built into their tape decks. If only compression could have worked something like that, and listeners could have been taught to choose for themselves.

Uncompressed music is best when you are sitting in front of the speakers or listening on headphones, and trying to feel and enjoy the emotion in the music. When the guitarist slams a chord really hard, you can feel it, as opposed to a heavily compressed recording where it sounds like all the other chords. The issues are very subtle and somewhat sinister, and compressed music can be very seductive, and at other times can be extremely boring, since so much of the impact of music involves use of dynamics. It's sort of a paradox to a musician who is recording, because as soon as you really sing a loud note or bash a powerful chord, you now understand that that will have to be removed electronically in order that the song won't end up too low in level to be heard in the world at large.

So why did I re-master my recordings? A number of reasons, the primary ones are that they really do sound better in "average listening environments" and they are no longer dramatically quieter than anybody else's. When I started doing digital recording, it was sort of a sealed box, and it was not easy for anyone to tinker with the waveforms, and the tools were few and clumsy. We sort of take it for granted that digital means you can zoom in and tinker with it, but in the early days its appeal was the cheap tape cost, low noise level, and the fact that there was no quality loss when it was copied or duplicated. In the last 20 years the field of digital editing has exploded, and now there is a huge pile of amazing digital audio tools in everyone's toolbox. With very subtle tweaking of the tone, spatial positioning, and even the phase of the stereo waveforms, they can be made to sound much more musical. There isn't a formula that is applied, but when you put each song under the "microscope," in most cases there is something that can be done to improve it considerably, almost like focusing a pair of binoculars. I have been pleasantly surprised to find that even my minimalist, 2-track recordings have been yielding many surprises, and now sound much more life-like and "present." The re-mastered recordings sound like a piece of cloth has been taken off my ears, and they fill the room better and are easier to listen to. On one track on my Guitar Voyages CD, moving the right channel 5 milliseconds in time made a dramatic improvement in the lifelike sound image. Who knows why? I'm just glad I got to listen for myself and make the artistic choices while I was alive, and it's a nice consequence of being an indie artist who owns my own catalog. I wonder what Paul McCartney thinks of the remastered Beatles catalog, and how much of a vote he wanted or got in it all.

I thought that some of the songs I had been hearing for 20 years would sound wrong to me once I re-mastered them, but it has been quite the opposite. I can't believe how much better they sound, and though this is pretty technical stuff, I am really excited about being able to make these decisions myself, and to improve them another quantum level. I succeeded in my first goal of capturing a lot of real performances on tape, when I was playing really well and in tune, and though I do some things better than I did 20 years ago, I can't play with as much energy and sizzle as I did back then. Some of the hot guitar pieces I am working on in this current re-mastering sound as good to me as I ever played.

I did not do anything dramatic with compression, and the compression I use is to my ears actually helpful because I often play too dynamically after years of street music and trying to get people's attention in noisy environments. The end result is that my remastered CD's are about 7-8 decibels louder than the old ones, and do not "vanish" in multi-disc changers. I intended to compress them a modest amount to make them "in line" with other acoustic music being recorded today, and also to clean up some small noises and glitches. Since my old recordings were uncompressed and dynamic, they sound embarrassingly quiet by modern standards. I am actually much happier with the re-mastered sound, and I will admit that it is nice to "civilize" the music a little bit, so that the "spikes" are less jarring and the overall recording is to my ears easier and more pleasant to listen to. I hope you agree, and I don't think any listeners will be appalled at what I have done.

The other advantage of remastering older recordings, especially for a guy like me, is that the noise-removal software is dramatically better than it used to be. I have always done my recordings live, and there have always been a number of chair and finger squeaks, bone cracks and other extraneous sounds that can pretty easily be identified and cleaned out. I know some people think they are "charming" but a click in the middle of a long fading guitar chord at the end of a thoughtful song does not help, and I have taken the liberty of removing a few of them on older recordings. There are still plenty of breaths and rustling clothing sounds you can hear if you listen closely, so those of you who like that stuff will not be disappointed.

I saved a couple boxes of my old uncompressed CD's, and there are thousands of them out there, so if any of you are audio freaks and want to hear the music more the way it originally was, with no compression and all the original noises, feel free to buy one from me while they last. There are not many CD's out there that are any more "purist" than my early digital CD's. They just aren't very loud and they don't work as well for background music as my newer ones, or as most recorded music today.