Records & Recording (Vol.2, No.1: Winter/Spring 1979)

Editor's Note: This is the Max Wilcox article that barely missed getting into the last issue; the recording sessions it focuses on are now ancient (i.e., 1978) history, but the moral for the big record companies and their producers remains as timely as ever. Note also that the article is actually a sequel to Max's digression, two issues ago, from his original series, which is now beginning to look like his Unfinished Symphony. But then editorial rigidity isn't what you read The Audio Critic for.

Even Less Is Even More

By Max Wilcox

No, I haven't weakened. Those simplified microphone techniques using superior-quality microphones have continued to produce excel lent studio results. In this article I'd like to take you behind the scenes of a recent recording project where further refinements were made in the “less is more” approach discussed in Volume 1, Number 5.

For the last few years, Peter Serkin has been devoting his musical energies to Tashi, a chamber music group he formed with violinist Ida Kavafian, clarinetist Richard Stoltzman and cellist Fred Sherry. Peter's last solo re cording was a monumental performance of Olivier Messiaen's “Vingt Regards sur I'Enfant Jesus” which he recorded in 1973. Last season Peter gradually resumed his solo career, and his recitals were made up entirely of works of Chopin. For some months he and I worked on plans for RCA recording sessions of this Chopin repertoire, and the sessions were finally scheduled for the middle of July, 1978. In June, Peter and I got together to make final arrangements.

The first decision to be made was the choice of a recording hall. Since RCA's Studio A was fully booked for the July dates Peter was avail able, I invited him to visit the hall that had been used for the very first recording I ever produced. The recording was the Brahms F minor Piano Sonata played by Artur Rubinstein, and the hall was the concert hall of the American Academy of Arts and Letters at 155th Street and Broadway in New York City. It is a beautifully resonant hall seating about 800 people and is far inside the building, away from the traffic noise that plagues so many New York recording locations.

Peter went to the hall with me a few weeks before the scheduled sessions and immediately fell in love with the place. It is one of the most beautiful 19th-century style halls anywhere in the world, and since the Academy does not have a concert series, it is rarely used. The decay time of the hall's reverberation is nearly two seconds, and since it is all wood and plaster, the resulting ambience is warm and lustrous.

Peter felt it would be ideal for Chopin.

So, the hall was booked. Now we needed a piano. We visited the Steinway basement a few days later, and, after trying many familiar and unfamiliar instruments, Peter chose a CDA492, a brand-new piano with a particularly rich, warm tone. The action was still new and stiff, but Peter felt any problems it might cause would be more than compensated for by the beautiful sonority. Steinway and I then arranged for Steve Borell to be the tuner in attendance at the sessions. Steve is one of the half dozen great piano technicians in the world, and we felt he could remove the “newness” kinks from the action during the recording sessions.

I then presented Peter with my technical proposals for the session. Peter and I have been re cording together since his first solo records in 1965, so we are long-time friends and collaborators. We work very closely together at the technical and musical aspects of a recording.

Our Tashi recording sessions are noted for their lengthy and painstaking microphone setup tests. Everyone listens very carefully and participates enthusiastically in whatever experiments with microphones and instrumental set ups we devise. It's all aimed at achieving a natural and well-balanced sound in the studio before proceeding with a recording session.

Peter is also used to the luxury of multi track tape recording, with its capability for subtle remixing during the transfer to the final two-track master. The remixing on Tashi records has been minimal, since we devote so much time to achieving the correct balance and instrumental perspective at the session. Still, Peter always felt comfortable knowing we could rebalance a section if some instrument seemed momentarily obscured. His last piano solo recordings had been multitrack, and he had taken advantage of this to change discreetly the micro phone balance during various sections of a composition such as the Messiaen “Vingt Regards”.

I was therefore a little apprehensive about his reaction to what I was going to propose.

I hoped I could convince him of the superior quality these proposals could give the final disc. Since the last Tashi sessions of Mozart and Beethoven Trios had been done with a greatly reduced complement of microphones, I knew Peter was convinced of the scientific and artistic validity of that approach. Still, they were done multitrack, which meant they could be remixed during the final mastering to two track tape. Since piano solo recording quite obviously does not involve balance with other instruments, it seemed the proper repertoire to begin Phase Two of my new (though for some people very tried and true) technical approach.

I therefore made the following proposal to Peter: I wanted to record the piano with two Schoeps/Studer MK?2 omnidirectional micro phones connected to a custom-made '“'mini mum electronics” console (about the size of a large briefcase), which would be connected to an Ampex ATR-100 two-track tape recorder operating at 30 inches per second using Ampex 456 tape with no Dolby. The object: carry the response of the microphone capsule through the absolute minimum of electronics to a high quality tape and tape machine. The original session tape would then be edited for musical purposes, and would become the two-track master tape. This master would then be cut at half speed on a newly designed RCA lacquer channel, and the pressings-well, we would face that problem when the time came.

To my delight Peter's reaction was very enthusiastic, and I began to make the necessary arrangements for equipment. Ray Hall was to be the recording engineer and Ray Rayburn was in charge of the technical setup. As simple as it was, it all needed to be in perfect operating condition to guarantee the full potential of the approach. The day before the session was devoted to draping certain hard surfaces of the American Academy listening room, placing the monitors in proper relation to the console and producer listening area, and tuning the response of the monitor playback system to give the flattest possible playback response at the chosen listening area. This is done by placing a calibration microphone at the chosen spot and adjusting the response of each speaker while feeding it pink noise. The calibration microphone registers the frequency curve of the monitor system and the response of each monitor channel is then equalized throughout the frequency spectrum until the response is as flat as possible at the chosen listening site.

(This doesn't, however, guarantee correct time domain response, but Rome wasn't built in a day.-Ed.) Since the listening room at the Academy has a large bass peak all too familiar from my Rubinstein sessions in days gone by, the monitor tuning made a reasonable control room out of a room that is otherwise so weighted at the bass end that critical evaluations are quite impossible. (This was the reason we discontinued the Rubinstein recordings there in the early 1960's. The resulting records were fine, but the listening was very difficult, and monitor tuning was not a well-known art in those days.) Our first Serkin session was scheduled to begin at noon. That gave us the morning for last-minute technical adjustments and gave Steve Borell ample time to begin the fine-trim on the action of the CD 492. Peter walked in at 12 o'clock and suggested that we should all have lunch while he would get used to the piano. He then went on stage and, since these were Chopin sessions, started playing a Bach Invention of course. As soon as he started to play I asked Ray Rayburn to push the record button. Since he thought we had left, Peter played for a few minutes and then walked back into the control room. I said, “Would you like to hear that?” After letting him re cover from his surprise that his Bach had had an audience, I seated him at the calibrated listening position and played back the Bach Invention. After about a minute of listening he said, “That's the best recorded piano sound I've ever heard. Don't change anything!” I then joked that this was a world record for us in briefness of microphone tests and went buoyantly off to a Westphalian ham sandwich.

Peter happily recorded for five whole days, and Ray Rayburn changed tapes like mad (since at 30 IPS you can record a total of only 16 minutes per reel). We never moved a microphone an inch because the sound seemed to fit all the repertoire, and at the end Peter said it was the best series of recording sessions he'd ever had.

Now, of course, there is absolutely nothing revolutionary about this approach. Gordon Parry and John Culshaw recorded Solti's entire historic Wagner “Ring” cycle on two-track tape, and it is still general practice at English Decca (London Records). Many quality-conscious independent recording engineers have been using this approach for years. Why did the major companies gradually desert two track recording? Well, let me give a brief history of the development of the number of tracks on tape machines.

Until about 1956, all tape machines were two-track and used quarter-inch tape at 15 or 30 IPS. Then the three-track machine, using half-inch tape, was developed. This meant the session tapes could be edited, and then re mixed to the final two-track master. Various instruments, like the soloist in a concerto or the woodwinds in an orchestra, could be put on the center track, and the balance could be altered in the final mix. It also began the subtle but real degradation of the quality of the original tape. Besides going through the electronics of the console of the original session, the tape was then played back through a remix console and copied to a two-track tape.

Each time a tape is copied the crispness of the transients and the general cleanness of the sound are slightly compromised. Nothing drastic, but it's not quite as good as the original. Three-track tape also began the use of more microphones. If we have more tracks, we can use more microphones, right? Artists quickly learned that the balances they heard at the session could be changed in the mix.

Conductors were often surprised at the ultimate balance of a concerto recording when the record was issued. All was well balanced at the session, but when the record was issued, the soloist had often grown to larger-than-life proportions. Soloists would often gently but firmly influence their producers to alter the final mix in their favor. And so it all began.

Four-track, eight-track, sixteen-track, twenty four-track. More and more microphones fed bigger and more complex consoles with more electronics between the microphones and the tape.

As more tracks permitted more micro phones, the concept of an overall sound on an instrumental group faded away. It was re placed by separate microphone setups on each instrument or group of instruments. Since these separate pickups were meant to control the balance of each group, the microphones were almost always used in their cardioid (directional) configuration. Cardioid microphones give lots of control. They also give a rather peaky, beamy sound with a large roll-off in the bass beginning at 50 Hz. What they do give is lots of control. Multitrack tape machines and cardioid microphones have made control the most important word in the recording studio. Pop records are usually based on musical arrangements that were never intended to be heard balancing themselves in a natural acoustical environment. They are written for instruments separated by acoustical panels playing in a dead room into microphones a few inches away from the sound source. It is all mixed, equalized and echoed both then and later, and it makes a lot of money for a lot of people.

The trouble begins when you apply this approach to classical music. As multitrack recording began to dominate the classical re cording scene, the philosophy of where a recording was actually created began to change. Many engineers and producers began to consider the actual session as no more than the beginning of a long process. It became “Get the sound on the tracks and we'll balance and equalize it in the mixing room.” No longer was a conductor really in charge of creating and approving the sound and balance of his orchestra at a session. The producers could say that they would correct any balance problems in the mixing room. All very convenient and economical, but the artistic control was taking a subtle and questionable shift. If a horn solo is covered in a recording, it is usually because the accompanying strings are playing too loud.

The correct solution would be to ask the conductor to replay the passage with the proper balance. That would produce a natural sound that could not really be achieved by lowering the string tracks and raising the horn in the remix. It would also be subject to the musical taste of the conductor, not that of the producer.

The actual sound of the ensemble is also not sacred at multitrack sessions. What the conductor hears in the playback room at the session is usually a flat playback of the tape.

When the multitrack tape is remixed by the producers, many of them add dramatic equalization changes which quite alter the texture and sonority of the ensemble. If the strings didn't sound bright enough at the session, wouldn't it seem more logical to change micro phones and microphone positions until the conductor and producer were both happy with a result that would then be preserved until the final record?

In far too many cases, the multitrack, multi mike world has become a place where producers and engineers treat the players and conductors as tools to be used as they create “their” performance and balance. Listen to Giulini's recording of “Pictures at an Exhibition” with the Chicago Symphony on DGG and hear the percussion section move forward and back, like a dancer on a runway, as the music progresses. Did the producer and engineer know more about orchestration and orchestral balance than Maurice Ravel? Did the Chicago Symphony sound like that when they played at the session? It's all contrived to “wow” the listener as he turns on the hi-fi rig and is knocked out by the impact of the cymbals.

Ravel's masterly orchestration of Moussorgsky's music becomes a playground for electronic razzle-dazzle. Of course, the most dazzling thing of all would be a properly balanced, full-dynamic-range, honestly recorded performance of what actually goes on during a good performance of “Pictures at an Exhibition”.

Thank goodness the pendulum is swinging back. The current audiophile recording scene is based on a return to the reality of the actual sound of music. Superb electronics capturing the sound of a few, properly placed, calibration-flat microphones is infinitely more exciting and, yes, more “commercial” than all of the tortuously remixed, re-echoed, re-equalized constructions that emerge from multitrack mixing rooms. When you hear a great symphony orchestra in Carnegie Hall, do you wish you could put in a 6 dB boost at 8 kHz so the strings would sound more brilliant? Well, neither do I, so why should it be done on a record? Just because a producer likes “bright” string sound? We will look back on all of this as the Dark Ages of electronic meddling, and I'm quite confident that natural sound will prevail once again. Does the true sound of a Guarnerius violin need improving? What it presents is a challenge. To capture the actual sound of such an instrument requires precision recording equipment used in a beautiful acoustical environment. Equalized cardioid microphones will never capture that sound, and neither will the people who operate them. “Control” of the sound has never been the object of true audiophile recording. A rock band playing parts that were never designed to balance each other without electronic help needs “control.” Maurice Ravel's orchestration was designed to work. It needs to be captured.

A whole new generation of recording people are going back to the basics and coming up with superb results. Lincoln Mayorga and Doug Sax returned to the basic principles of 78 RPM disc recording when they initiated the “direct to disc” techniques at Sheffield Labs.

That began the whole current resurgence of audiophile recording, and the classical music record collector will be the ultimate beneficiary. Flat, natural-sounding equipment needs flat, natural-sounding program sources to demonstrate what it really can reproduce. Lots of us are trying to give it to you. The independent labels are at the head of the pack at the moment. Let's hope the majors, with their great artists and resources, will not be far behind.

---------

[adapted from TAC]

---------

Also see:

Even Less Is Even More, by Max Wilcox--A Discography for the Audio Purist: Part II

Various audio and high-fidelity magazines