   
|
||
 December 15, 2007 His Master's Voice I’ve been collecting music in one form or another, whether on vinyl, CD, or tape, for more than 35 years. That means, of course, that I’ve read a lot of recording credits -- even in the tiny print of CD booklets. I’ve also read many articles about music and recording, so I should probably know how the final sound of a CD or LP reaches us. In theory, I do. But whenever I’ve talked to the people whose job it is to record, mix, and master a musical performance, I’ve realized that there were things I had an intuition about or thought I knew that weren’t quite accurate. My conversations with other music lovers, some of them audiophiles, have led me to believe that many of us possess more assumptions than facts about the various stages of recording. With that in mind, I decided to ask some of the people who have given us wonderful-sounding CDs and LPs over the years exactly what goes into creating that magic. Let’s start at the beginning, with the actual recorded performance. As Don Grossinger, whose credits include the LP remasterings of the Rolling Stones’ Decca catalog for ABKCO a few years back, reminded me, "In the old days, a group of musicians -- say, bluegrass players -- would stand around a single mike. Soloists would step forward to get closer." Before the advent of microphones, recordings were made through a purely acoustic process that involved capturing sound with a horn made of wood or stiffened fabric. The vibrations created by the musicians traveled through the horn to a needle that etched the vibrations in wax. In 1926, when electrical microphones were introduced, the sound quality dramatically improved. Still, recordings were done with only a single mike, placed close to the soloist or singer, with a resulting limited dynamic range and detail. And the stylus’s movements were still cut directly into wax. By the late 1940s, magnetic tape was in wide use for music recording, and proved more reliable, accurate, and archival than earlier media. Microphone technology had also improved. Amplification had become part of the audio signal, and mikes were now designed for sensitivity and sound quality as well as for maximum output. Consequently, they could pick up more musical detail, which could then be captured on tape. Room acoustics, microphone placement, and other considerations increasingly came into play, and recording engineers began to explore the possibilities of using multiple mikes, even though the output signal was still monaural. Even so, recording techniques varied from engineer to engineer. C. Robert Fine, who engineered the great Mercury Living Presence LPs, made his legendary mono recordings using a single mike.
When I asked engineers how they preferred to record musicians in the studio, most seemed to like some semblance of a live performance. Steve Hoffman, whose mastering work with DCC and Analogue Productions is legendary, told me in an e-mail, "I group the musicians around the room like they are playing together. I want the sounds to blend together. It sounds natural that way, like it occurs in real life. I hate the sound of four people playing in what seems like four tiny closets." John Herbert, who does wonderfully warm analogue recordings at Lion Studios in Singapore, agrees: "depending on the style of music, but preferably, and certainly for jazz, recording the whole band together, with careful mike placement to minimize spill." Herbert will sometimes use screens to control spill; i.e., when the sound of one instrument leaks into another instrument’s mike. In moderation, spill can help create atmosphere: "I find that some spill can actually help give a sense of spaciousness (and ‘liveness’) to a recording." Too much spill, however, can muddy the sound. While multitrack pop recordings employ many additions and sweetenings, they often begin with a solid foundation laid down live in the studio. Mike Giblin, bassist for Cherry Twister and Jeffrey Gaines and leader of the power-pop band Parallax Project, has recorded a variety of musicians at his Red Chuck Studios, in central Pennsylvania. His approach to getting the best results "depends on the song and the environment. I always try for the purest, most natural performance possible, with the goal of capturing as many of the instruments as I can at once. For a rock band, that usually means all of the backing instruments at once. Whichever way the performer is the most comfortable delivering the song is generally the best way to try to record it." Much of the work in a pop recording -- guitar solos, percussion, background vocals, and so on -- is done after this basic backing track is laid down. A key aspect of recording is microphone placement, which requires both technical skill and intuition. "Microphone placement is the ‘black arts’ of audio recording," Giblin told me. John Herbert’s approach is "pretty standard -- follow the 3-to-1 rule, watch out for nasty spill, watch your phase relationships. Choose the right mike for the instrument, as far as possible." Following the 3-to-1 rule, the engineer places each microphone at least three times the distance from the next mike as it is from the instrument it is recording. In other words, if the mike for the saxophone is 2’ from the instrument, the mike for the piano should be at least 6’ away from the first mike. The rule also applies to reflective surfaces. If a vocalist is 1’ from the mike, the mike itself should be at least 3’ from any wall that isn’t acoustically treated. The 3-to-1 rule also applies to instruments, such as drums, that require multiple mikes. I cite all this technical stuff, in simplified form, as examples of how a recording engineer must know physics, acoustical engineering, and the technical specifications and capabilities of his or her equipment. "Read the manufacturer’s manual," advises Walter Sear, of Sear Studio in New York. On top of all that, good ears are essential. A little smart experimentation helps, too. "The most valuable piece of advice I ever received [about mike placement] was from an accomplished British engineer who had worked with Ron Wood, George Harrison, and countless others," Mike Giblin recounted. "He told me to always approach it from the player’s point of view. He said players play so that it sounds good to them, so why not mike it that way? So he would always place a pair of stereo mikes just above the drummer’s ears, or a mike just out from the acoustic guitarist’s chin. What I learned was that this is where dynamics and articulation seem to be the most dramatic -- because, after all, they are the one who is piloting the ship." When the musicians begin to play, the engineer has to set the recording levels and capture the magic on tape. "The engineer is supposed to control recording levels as you record," Walter Sear told me: "fingers on the faders and eyes on the meters. You are as important a part of the recording and musical process as a conductor, except that you have the added responsibility of keeping the levels [between] the noise floor and the overload point of the recording medium." Most of the engineers I spoke with preferred the sound of analogue, but some liked the ease of digital recording. "Analogue has the edge, sound-wise," John Herbert e-mailed. "Digital has the edge when it comes to manipulation (cut and paste, pitch correction, etc., etc.). My preferences -- for sound -- definitely analogue." Steve Hoffman’s heart also remains loyal to the older technology. "I love analogue," he wrote. "Warm, rich sound, natural, no harshness." Walter Sear is blunt: "Analogue sounds good to the human ear, digital sounds chopped-up and annoying." Mike Giblin, on the other hand, finds that digital works best for him, for now. "As far as a preference, I would have to go with digital, for a couple of reasons. Running a small studio that records a lot of small-market and semiprofessional musicians, the cost of entry for analogue is just too prohibitive. Tape machines and tape are expensive to acquire and maintain, and for what I can offer via the digital medium for the same price, analogue cannot compete. In addition, everything with analogue takes longer, and in my market, efficiency and cost-effectiveness are two primary goals." When the tracks for a record have been laid down, it’s time to mix them. According to Walter Sear, "If you have recorded the basic tracks and you have ridden levels as you recorded, the mix should be two-thirds done before you begin." As Adam Ayan of Gateway Mastering explains, "Mixing is basically the art of balancing a number of tracks in a multitrack recording, with the outcome being a stereo mix (in the case of most consumer formats)." The mixing engineer takes what’s on the tape (or digital files) and prepares it for mastering by setting the levels of the instruments and vocals, placing them in the soundstage, and adding effects, such as reverb, delay, or equalization (EQ), in order to enhance the sound. "The goal of mixing," Mike Giblin says, "is to assemble all of the parts of an individual song into a cohesive whole." To get an idea of how it’s done, let’s look at a simple mixing project. I’ve played on some recordings in studios and sat in on some mixdown sessions, but the project I want to look at is one I did on a 4-track cassette recorder: A friend of mine had asked me to help her record a demo of a song she’d written so she could send it off to a publisher. We kept things pretty simple. She played acoustic rhythm guitar and sang, I played some guitar, and we both did some basic percussion. I recorded each element on a separate channel, beginning with her guitar, then her vocals, then my guitar, then the percussion. We used Shure 57 microphones fed directly into the 4-track. So far, so simple. Each element had its own track. The next step was to send the 4-track tape through a stereo so we could set the levels and place the instruments. I recorded everything dead-center. Experienced engineers have the option of panning during a recording so the instruments are to the left, right, or center, but those positions can always be shifted later. First, I placed my friend’s vocals solidly in the middle of the soundstage, and slid the output level up so they’d be out front a bit. I put her acoustic guitar in the middle as well, but pulled the volume back a hint so it would sound as if placed behind the vocal. I put my guitar in the left channel, and moved the panning knob around a bit to see where it would sound best. I ended up putting it at about 9:30 or 10 o’clock in the soundstage, and set the volume at a place where it enhanced but didn’t overwhelm my friend’s vocals or her own guitar playing. I put the percussion at 2 o’clock, more or less, and set the output level so the percussion was present but still behind the two guitars and vocals. The recording sounded flat, so I punched in some delay. Actually, I plugged an old Ibanez analogue delay foot pedal into the effects loop and fooled with the level on the 4-track until the music sounded as if it was being played in a three-dimensional space. If I’d had a reverb pedal, I would have tried that. I adjusted the EQ so the vocals were focused and present but not sibilant, and used it to dial in the tone on my friend’s guitar so the listener would have a sense of each string ringing out. My friend didn’t sell her song, but the publisher commented on the high quality of our demo. All those years of careful listening paid off, I guess. If you take what I’ve described in the last few paragraphs and multiply it by . . . well, by a lot, you get some small idea of how much work mixing can be. With multitracking, it becomes a fine art. Instruments can be placed precisely in the soundstage, subtle touches -- maybe a bit of percussion, a hint of acoustic guitar -- can be added at just the right volume to create an illusion of depth, reverb and other effects can be added to create ambience, and so on. Gain and equalization can be judiciously applied to give instruments more clarity and immediacy. Mixing isn’t always so arduous. Through the mid-1960s, engineers used their abilities to create great-sounding mono recordings, even when they were using tape machines with two or more channels. Even today, when a recording has fewer elements -- for example, a jazz performance -- the engineer can begin to set levels in real time. John Herbert mixes "as soon as we start recording. I try to record with a picture of the final mix in mind. Of course, this picture will get modified as the recording progresses -- but with care, not too drastically." When the mixing of an album is completed, it’s time to master it. Ron McMaster, who has done such an exemplary job with so many of the Blue Note reissues on CD and vinyl, as well as Mosaic Records’ limited-edition boxed sets, describes the transition: "The mixing engineer will take all of the recorded tracks and mix them down to a two-track stereo final mix or to a 5.1[-channel] surround-sound final mix. Then those mixes are ready for mastering. Mastering is the last stage of the recording process." In mastering, the mix is prepared for pressing in vinyl or replication on CD, DVD-Audio, or SACD. The objective in mastering an album is to bring, in Adam Ayan’s words, "some sonic consistency from mix to mix (song to song), so the album can have a cohesive feel, as opposed to sounding like a bunch of different recordings and mixes that do not live together well. The listener should get a sense of cohesiveness overall (matching levels track to track as well) when experiencing an album as a whole." Mike Giblin likens it to cooking. "Mixing is making a dish," he wrote to me, "and mastering is making a meal. In mixing, you make sure everything works together in the song, whereas in mastering you make sure all of the songs work together in an album." Some of the tools used in mastering are compression, equalization, limiting, and noise reduction. Used judiciously, these can help bring out a recording’s strong points. EQ can bring a little more focus to a ride cymbal or shimmer to an acoustic guitar. Compression can help give some instruments more punch. "You can use compression to increase the impact of a kickdrum without changing the tonality," Don Grossinger points out. "You can change the impact of the bass without changing the sound." Overused, compression can rob a recording of dynamic range and soundstage depth -- successive reissues of some titles, especially in pop music, seem to push more detail to the front. When a mastering engineer applies good taste and judgment, however, compression can overcome limitations, such as a vocal performance’s slight change in volume because the singer moved away from the mike. Mastering is the least subjective stage of the recording process. The recording and mixing engineers, as well as the producer, have spent so much time on the details of the music that they don’t always have the distance required to create a consistent sound and feel for a recording. "Mixing in many ways has somewhat more flexibility but somewhat less perspective," Grossinger notes. "With mastering, you don’t have as many tracks. I basically sort of correct things that weren’t perfected in the mixing stage." For John Herbert, the task is to "[tie] the whole bunch into a consistent whole -- make the album sound as if it was all recorded in the same place, space, and time, so there are no major sonic disturbances from track to track." The mastering engineer’s assets are his or her ears and a good sense of judgment. Ideally, he or she shouldn’t leave a firm sonic footprint on the final product. Grossinger believes that "The overall tonality should be established in the mixing stage and perfected in the mastering stage. The ideal mastering job is that the recording and mixing engineers did such a good job conveying the music that I have little to do."
Remastering a well-known record presents its own challenges. Don Grossinger’s masterings for ABKCO’s vinyl reissues of the Rolling Stones catalog a few years back were compared favorably with the original UK Decca pressings. Grossinger had to balance his respect for the original recordings with his desire to bring out their best. "I was pretty damn reverential, but I did a fair amount of EQ on them. To get them to sound the way they did required some work. Beggars Banquet had to be speed-corrected. Hearing some of these tracks in the original, they sound way better now." Ron McMaster has to maintain that same fine balance when dealing with legendary engineer Rudy Van Gelder’s original recordings. "I always try to maintain the original sound of the mixes and still deliver a cleaner and clearer version than the original. The original Blue Note recordings were very well done, and my approach has always been to maintain that original sound and yet adjust for the current technology so the listener can enjoy those old recordings once again on their new, modern equipment. I do not want to make any adjustments that would take away or distract from the music -- for example, adding a lot more bass onto a track that originally had only a small amount on the original vinyl release. The listener would be very surprised and unhappy to hear his or her old favorite sounding so different from the original." Home recording gear has become affordable and easy to use, but it’s no substitute for a well-designed recording studio and the skills of a trained audio engineer. Walter Sear is alarmed at the increasingly poor quality of recordings: "I have seen my art deteriorate into amateur land, where incompetent people who haven’t paid their dues pass themselves off as ‘engineers.’ There is a reason why the audience doesn’t want to buy the product anymore." (Sear has written a number of articles on the state of the recording art. See http://members.aol.com/searsound/articles2.html.) Think of your favorite records and ask yourself if you can name the craftsmen who made them sound so good -- Geoff Emerick, Roy DuNann, Rudy Van Gelder, Bob Ludwig, George Peckham, Bernie Grundman, Ron McMaster, Don Grossinger, Stan Ricker. Alan Parsons was the engineer for Pink Floyd’s Dark Side of the Moon, Chris Thomas mixed it, and the US 1994 CD was mastered by Doug Sax. My LP copy was mastered by Wally Traugott, but his name isn’t listed on the cover -- it’s in the deadwax on the LP. Fred Plaut was the engineer for Miles Davis’s Kind of Blue and Dave Brubeck’s Time Out. His name wasn’t even on the covers of those two important, best-selling jazz LPs. Musicians are ultimately responsible for what we hear on a record or CD, but at all stages of making a recording, it is committed audio engineers who make sure we hear those musicians in the best possible sound. Ellington Masterpieces is a terrific performance by Duke Ellington and his band, and Fred Plaut’s astonishingly intimate recording puts you in the room with them. Should more people know Plaut’s name? Listen to the few titles I’ve mentioned here. You tell me. …Joseph Taylor
|
||
|
The 1950s seem to have
been the turning point for recording. Musically, Louis Armstrong’s Hot Five and Hot
Seven sides, recorded from the mid- to late 1920s, sound as fresh today as the day they
were recorded. Sonically, they’re ancient history. But Armstrong’s 1950s records
for Decca and Verve sound as immediate today as they did when committed to tape a
half-century ago. While there have been countless innovations in recording since the
’50s, from stereo to multitracking and digital technology, it was then that the
groundwork for high fidelity was laid. The recording, mixing, and mastering engineers I
spoke to continue to use techniques that were established and perfected in what many still
consider the golden age of recording.
While it’s important to listen
to the recording at a slight remove in order to make good sonic decisions, the mastering
engineer also tries to keep the artists’ intentions in mind. "[I] make the
recording sound lovely, and make sure it has enough impact and clarity, and that it sounds
as much like itself as it can, with nothing obscured," Grossinger says. "The
artist put an awful lot of emotion into it, and you want to make sure that all comes
out." As Adam Ayan explained to me, "Every artist’s sound and style speaks
to me a little differently, and the best thing I can do is respect what works best for
their music and recording, understanding that what might work well for one recording or
genre will not necessarily work well for all recordings."