The utmost concentration is essential when mixing the 3 channels of the original Mercury recording tapes to two channel stereo in order to achieve that perfect stereo-balance and a real to life sound at all instances, for all instruments.

"Real to life" means: dynamics that are detailed and frequencies that are harmonious in all registers - as is the case in the recording of 'Pictures At An Exhibition' played by pianist Byron Janis on Mercury CD 434 346-2.

The grand piano has extraordinary presence and you can clearly imagine that you are almost able to touch the black lid, while seeing the keyboard from the side and part of the bronze frame with the strong, tensed strings, and you hear the wooden construction. As if you were there.

It is also a question of microphone placement, of choosing the right position of the instrument in the studio or concert hall and of using top quality tape recorders, playback amplifiers and monitor loudspeakers. But to mix the three channels down to two-channel-stereo when the signal is transferred to a DAT recorder while preparing the CD reissues, that takes as great a skill as the initial transfer to the lacquer in the nineteen fifties and sixties. Obviously Wilma Cozart Fine has the ability to concentrate in abundance. The Mercury Compact Disc with 'Pictures At An Exhibition' is the eminent proof.

At left: Wilma Cozart Fine at the Western Electric mixing console and surrounded by a host of components as she poses for the camera at the occasion of the release of another batch of CDs containing transfers of legendary Mercury 'Living Presence' tapes. Connoisseurs can easily spot the modular Audio Suite (designed by Mark Levinson) which is one of the few preamplifiers in the world today that can boast of extreme neutrality.

LIVING PRESENCE

The label or term 'Mercury Living Presence' originated from music critic Howard Taubman's review of the orchestral recording of Mussorgsky's 'Pictures At An Exhibition' (orchestrated by Maurice Ravel) made years earlier.

Taubman wrote in the fall of 1951 in The New York Times about that recording, that "one feels oneself in the living presence of the orchestra". From then on "Living Presence" became Mercury's quality slogan that distinguished the label from its competitors and "Living Presence" adorned practically all following Mercury-issues.

MG 50000 was reviewed in High Fidelity Magazine, Volume 2, Number 2, of September-October, 1952. Critic C.G. Burke wrote:

In the summer of 1950 Wilma Cozart accepted a positon with Mercury, as is stated in many articles.
When it became known that the youthful Kubelik was to leave the Chicago Symphony - because of a conflict with the orchestra - the Chicago Symphony would no longer be available for Mercury to make recordings with. It was announced that from the summer of 1953 on Fritz Reiner was going to be the conductor and he was contracted by RCA. After her appointment with Mercury Records, Wilma Cozart immediately proposed that Mercury should hire the Minneapolis Symphony with Antal Dorati who had been recording for RCA Victor in the past. She knew Dorati from the time she was his secretary with the Dallas Symphony and also when she had moved to the orchestra of Minneapolis.

Recording projects with Antal Dorati and the Minneapolis Symphony were immediately undertaken: MG 50004 (Borodin, 2nd Symphony, and Stravinsky, Firebird Suite) released in June, 1952; MG 50005 (Berlioz, Roman Carnaval, and Ravel, Alborada del Gracioso) released in December of that same year; MG 50008 (Tchaikovsky, Symphony No. 5) released in November, 1952. MG 50009 (Rimsky-Korsakov, Scheherazade) released in September, 1952.

At the same time recordings with Kubelik were being made. Kubelik's Brahms for instance was released in December 1952. The last recordings of Kubelik/Chicago for Mercury were made on 3-5 April 1953: Mozart Symphony No. 38 “Prague” (issued on MG-50015) coupled with Symphony No. 34 which was recorded from 4-6 of December, 1952). Schoenberg's Five Pieces op. 16 and Hindemith's Symphonic Metamorphoses on Themes of Weber were issued on a rather late release (MG-50024).
See also the separate page MONO DAYS
and check James Lacy's Complete Living Presence discography.

AUDIOPHILE CD WITH BYRON JANIS

Wilma Cozart Fine: 'Not so long ago a dealer called and told me that he had some trouble in selling a pair of Thiel high-end loudspeakers. The client had been listening to all sorts of music but was unable to decide if he would buy the speakers. Until the dealer played this CD and the client went home with the speakers and the CD.'
The performance of Mussorgsky's original piano version of 'Pictures At An Exhibition' by pianist Byron Janis was recorded decades ago (December 25 - 29, 1961) in New York, in the Ballroom Studio of C. Robert Fine - Bob for insiders - who married Wilma Cozart four years earlier. Today this recording is as modern as it was then, because the sound quality and the total image are a showpiece (in the best sense) of what is possible in recording technique, and will remain so for years to come.

This edited image is from an advertisement of Audio Devices, Inc., promoting their audiotapes, audiodiscs and audiofilm in 1955. The caption:

"Robert C. Fine, President, and Al Mian, Chief Mixer, at master control console of Fine Sound Inc., 711 Fifth Avenue, New York City."

The first paragraph in the ad reads:

"In professional circles Bob Fine is a name to reckon with. His studio, one of the countries largest and best equipped, cuts the masters for over half the records released each year by independent record manufacturers."

NEW MIX

Wilma Cozart Fine has transferred over 200 Mercury tapes that date back from the years when analogue recording was an art as well as a science. She mixed the many legendary tapes to a DAT recorder from which the CD-masters were made.

In order to make this possible it was necessary to restore the original recording and mixing equipment, Ampex 300-3 stereo tape recorders (using 1/2 inch tape) , the 35 mm Westrex machine, and the Western Electric mixing console. All were used at the end of the nineteen fifties and the beginning of the nineteen sixties, the early stereo days.

The 35 mm film with the recording of Rachmaninoff's Piano Concerto No. 3 with Byron Janis and Antal Dorati was lost and a tape copy of this recording was used for the the LP issues and for the transfer to CD done by Wilma Cozart Fine. The first CD release does not have the 35 mm banner either. Apparently the 35 mm film recording was only discovered much later and was the basis for a new transfer to LP by Speakers Corner.

Naturally the recorders and mixing console is valve-equipment. Although the transistor had been invented decades before, this was the only quality equipment available. In the process of restoring, not only the circuitry had to be checked, but the heads of the recorder should have the precise gap and should function with the right bias and equalization in order to read the signal to the max. The restoration of the equipment was not without difficulty. It took about six months to compete. The Ampex machines are special machines with three heads and three channels, with three head amplifiers. They were built specifically by Ampex for Bob Fine.

EARLY EXPERIMENTS

Wilma Cozart Fine: "Bob handed the specifications to Ampex. You know, he was a technician and an inventor. Already in 1955 he experimented and compared the quality of 2 and 3-track stereo. He said that only recordings made with 3 channels could provide a good stereo-image."
Those of you readers who have witnessed the rise (and decline, I should say, in some sections) of the high-fidelity scene through the years - or have read about the often painstaking trials that led to the extraordinary achievements in the sixties when the analogue sound reproduction was at its peak if you owned top quality playback equipment - may know that especially in the beginning of the stereo era, some amateurs used a center-channel, that is to say that a third loudspeaker was placed in between the stereo pair to reproduce a signal that was the subtraction of left and right channels. Others experimented with a 'mid-speaker' that radiated the addition of the left and right signals, but at a precise adjusted level.

STEREO FROM 1956 ON CD

One of the earliest proofs of Bob Fine's stereo technique is on CD as well. It is Mercury 432 005-2 with Zoltán Kodály's 'Hary Janos Suite', 'Dances of Galanta' and 'Maroszek Dances', and Bartok's 'Hungarian Sketches' and 'Romanian Folk Dances', all performed by the Minneapolis Symphony Orchestra conducted by Antal Dorati. The recording date: November 1956!

In November 1953, Robert E. Blake, Don Gabor's recording technician, already recorded in stereo for the Remington label. The orchestra was the Cincinnati Symphony Orchestra directed by conductor Thor Martin Johnson. The orchestra performed with the Helsinki University Chorus. RCA and English Decca (London) followed in 1954.

However, Bob Fine's stereo recording is unique because of the use of three microphones and the specific placement to record the performance on three tracks (channels) in order to fully capture the original orchestral balance.

UNIQUE

So a stereo-recording is only then true to life if the sound stage is picked-up by three microphones. This means in case of an orchestra, one microphone for the left section, one for the right section and one for the players sitting right in the middle in front of you.

DIGITAL RECORDINGS AND 3 MICROPHONES

I read on the internet a discussion between technicians telling that they were using this basic setup but were not pleased with the result. They concluded that the original 3-microphone setup has its flaws. They forget that the digital format of 16 bit and 44.1 kHz. sampling frequency is completely different from analog. Yes, the digital format has a dynamic range of 96 dB and also a signal to noise ratio of 96 dB.

RESOLUTION

Despite the fact that analog formats have a S/N ratio somewhere in the region of 80 dB or even lower, the resolution is much higher if compared to the CD. The digital format of the CD is a linear format where with decreasing recording level the resolution is decreasing as well.
If, when using a digital format, microphones are positioned above the front of the orchestra and 7 feet high, the Bob Fine way, instruments far away in the back of the orchestra get a less precise result when playing softly (low recording level). In that case the addition of dither to the signal will help.

CLOSE-UP

Many modern recordings are made by close miking of all the sections of an orchestra, a band or an ensemble. Many technicians understandably do use more than the three channels used by the Mercury team. Today they are likely to use 24 channels for recordings of large orchestras and will adjust the various channels for obtaining the original orchestral balance and will monitor the balance of the sound meticulously.
This is specifically the case if there is a great difference in level between the loudest and the softest passages in the music. The soft passages should be recorded in close-up.

Check what is the difference between the linear format and a high resolution digital format here: History: 25 Years CD.

There you can also read about the analog approach (miking) in the early days when technicians started recording in the digital format.
And: in the days of analogous recording, valve amplifiers were used in recorders. mixing consoles and steering amplifiers. Naturally the playback system consisted also of valve amplifiers and the big loudspeaker systems had large woofers with a relatively high resonance frequency. If today one wants to make recordings like Mercury did, analogous tape machines have to be used for the recordings. When ready they can be converted to the best digital format.

The diagram shows the microphone setup and the 35 mm film sound recorder with 3 tracks. In the cutting room the various takes were spliced. Each recording had a minimum number of splices which makes the recording all the more natural. Nowadays, with the digital technique up to 400 (or even more) splices are not uncommon in a recorded symphony. (Diagram drawing by R.A.B.)

.

 

 

 

The liner notes of early Command recordings explain Bob Fine's microphone technique:

35 MM SOUND FILM CHARACTERISTIC

That same Byron Janis CD with Pictures At An Exhibition carries two encores. One of them is recorded on the 35 mm film recorder made by Westrex. Westrex had modified the machine mechanically and electronically to Robert C. Fine's specifications, in such a way that three heads were aligned and three channels with electronics were built in. Certainly inspired by the Cinemascope and Widescreen movie technique which for their sound recording used 5 tracks on the film. In the movie theater there were 3 speakers in the front (left, middle and right) and there were two speakers in the back of the theater and some supporting speakers on the side walls. 

Since dialogues in movie theaters have to be understood, originally these film recorders had a frequency-characteristic that could not be identified as high fidelity. So they had to be modified in order to achieve the desired characteristics: an extended frequency range and perfect linearity, with the benefit of an increased signal to noise ratio of the 35 mm sound film.

If this correction is not made, the sound on LP gets that slightly "glassy" character as often heard in Everest recordings.
Also the attention paid to the cutting of the discs and the use of the best cutting-lathe played a mayor role in achieving excellent dynamics and a detailed sound quality, though not every reviewer was always pleased with the clear and often bright results. But those who owned quality-equipment enjoyed practically every first release bought of any Mercury recording. One should not forget that quality cartridges of quality brands like General Electric, Fairchild and Ortofon in the mono days generated a different sound appropriate for the way recordings were made and records were cut. And later the many stereo models with the newly devised elliptical styli, translated the modulated groove in the early stereo years in a completely different way than modern high end cartridges do with their ultra fine line tips often generating a somewhat slender sound.

EVEREST

Everest initially used 35 mm film for recording and explained the advantages on the inner sleeves: thicker tape, less print-through, wider tracks, higher dynamics.
When Everest stopped their high quality productions, the Studio with 35 mm recorders and consoles were sold. Bob Fine acquired the 35 mm recorders and had them brought up to date for high fidelity. How come that there is a distinctive difference in the sound characteristics of a Mercury Living Presence recording (and of early Command recordings as well) if compared to the early Everest stereo releases?

These differences are caused by the choice of microphones and microphone placement. There are also the differences in the electronics of mixers and amplifiers used when cutting a lacquer. These are: frequency characteristics, signal to noise ratio, dynamic capabilities, phase characteristics, and distortion values. Remember: Bob Fine had the recording characteristics of the 35 mm machines changed by Ampex.
Every record label has its own "sound". In RCA and Decca (London) recordings a slight phase shift can be noticed as if the signal is somewhat weaker in the lower-mid/mid section. Everest recordings show a less chiseled transient.

COMMAND

Command recordings however have fantastic sound because of the application of exact the same microphones, the perfect placement of the microphones, and the electronics for microphones, recorder, and cutting lathe.
And C. Robert Fine and George Piros were responsible for each and every recording as is printed on the inside of the early Command Classics gatefold issues.

NOISE REDUCTION

Often these differences are also brought about by the unfortunate application of the Dolby Noise Reduction System. Mercury did not use a noise reduction system.
As said, a 35 mm sound film recorder was originally designed for the recording of speech to be mixed with the music and voice over recordings and finally put on the sound track of the film.
The frequency characteristic and the dynamics were mainly designed to reproduce the dialogues so they could be well understood by the audience even those people sitting in the last row.

C. Robert Fine had the electronics of his 35 mm recorders modified for recording music in high fidelity. To what extend Bert Whyte had already corrected the characteristic is not known to me.

CAMEO 4:35 MM

The use of 35-millimeter film was also chosen by the engineers of the CAMEO record label. But now 4 tracks were used. Above is the explanation taken from a cover of such a 4 track recording.

MORE QUESTIONS

Invited on behalf of Polygram Classics & Jazz by press officer Dorien van Londen I was in the privileged position to meet Wilma Cozart Fine in Holiday Inn Plaza Hotel in the center of Amsterdam. We had been talking already for quite a while. But I wanted to know more about Mercury's past and the transfer of the tapes to CD.

In Mercury's special recording van Robert Eberenz just has put on a new reel of 35 mm tape into the recorder.

RAB: Was this recording technique patented by Mercury? The label Everest also used 35 mm film recording, and Command did the same.

WCF: No, there were no patents. Everest was Bert Whyte's label. He started an audio shop and later on he became a journalist. He died in 1994. My husband and I were good friends of Bert Whyte.

RAB: How about other labels and the competition?

WCF: We came already in the nineteen fifties with 'Living Presence' and only sometime later RCA with 'Living Stereo'. We were so busy with producing that we actually paid hardly any attention to other labels. And there were the big labels: Columbia, RCA, American Decca.

RAB: Did Mercury have their own plant where the records were pressed?

WCF: Yes, we had one for pop. Classical LP's were pressed at RCA's factory. We cut the lacquer and they manufactured the matrix.

RAB: I admire so much that you have mixed the three tracks manually towards two channels during the cutting of the record. I myself made audiovisuals and had to mix the tape with speech and the tape with music manually and synchronize it with the images. I had to start a fade-in at the right moment and finish the transition to the following image on preselected moments in the music, sometimes slowly and at times very quickly. If I made a mistake I had to start the whole program of 20 minutes all over again. I can imagine how intense your work must have been.

WCF: Yes, we never mixed the channels to a master tape, but always directly to the cutting lathe. A tape would introduce extra hiss and we did not want that.

RAB: As the transfer was in progress you could change the stereo-balance. Did you do that?

WCF: I had to prepare myself well and I had to note beforehand which passages needed small adjustments. Too much correction is not good. That would disturb the natural sound balance of the orchestra.

RAB: Was there more than one lacquer cut from the same recording so that two masters were made, possibly with a small difference, and RCA used both of them?

WCF: No, only one lacquer was cut. Of course things did go wrong sometimes. You know, the groove with a high level of dynamics need more space. It can be done by the cutting machine automatically. We did it manually. I would say to the technician who was cutting the lacquer: 'Here comes the loud passage' and he then varied the distance between the grooves. It could happen that we were busy and that I had still three more minutes of music on the tape and he had already reached the final groove. And then everything had to be done from the start. But we also have cut lacquers that contained a maximum of information, almost 30 minutes for one side.
.

When a reel is full, a new reel with unrecorded tape is inserted. To make splicing possible, the performers will start playing several bars earlier in the score. The recordings will overlap.

This same procedure is followed whenever a mistake in the performance has been made: noise, false note, or whatever. The performers will start playing at a point much earlier in the score. Now there is enough overlapping to make a perfect splice.
But...

Just after S1 a mistake in the performance has been made.
So at S1 of take 1 a splice should be made to continue at S2.
But there is a problem. The perforation at S1 of take 1 and S2 of take 2 do not coincide. See the drawing. Unfortunately the splice cannot be made at S1.

However both the music signal and the perforation do synchronize much earlier in the recording, at S3. There the splice will be made in order to avoid an incorrect transition. This means that a new take has to start much earlier than is the case with 1/2" audio tape.

The 35 mm brown recording film was not transparent of course but was visualized for better understanding. (Diagram drawing by R.A.B.)

The Mercury people were always proud of the "simple" way they made their recordings. The liner notes of MG 50038 say:

" This recording was done at Orchestra Hall, Chicago, on January 8 and 9, 1954, using a single Telefunken microphone hung approximately 15 feet above and slightly behind the conductor's podium, the orchestra playing in normal concert set up throughout these recording sessions. Fairchild tape machines were used for the recording sessions; and the processing from tape to disc was done through Fairchild tape machines, McIntosh 50-watt amplifiers and the Miller cutting head, employing the Fine-Fairchild margin control system of variable pitch and variable depth of cut - thus assuring a 100-percent reproduction in disc form of the recording captured originally on magnetic tape."

Note: Zoltan Kodaly composed The Peacock Variations for the 50th anniversary of the Amsterdam (Royal) Concertgebouw Orchestra and the work was first performed in 1939, Willem Mengelberg conducting.

RAB: Did you record more than one take and did you splice them?

WCF: Yes, the disadvantage of the Westrex-recorder that we used on location was that it could handle spools with a maximum of 10 minutes. In case the orchestra made a mistake, we stopped and the recorder was prepared with a new reel with magnetic film. The orchestra had to continue in the same tempo. No difference should be noticed. Sometimes we started all over again if the mistake was made in the beginning of the work. That amounted to a lot of tape. A tape that contained a mistake could not be used a second time.
Or we just continued after a mistake, but then a few bars earlier in the score in order to have enough length to make a perfect splice. You know that 35 mm sound film is transported via sprocket wheels. The music and the position of the sprockets on both takes do not always coincide. That is why we needed more tape in order to be able to make the splice much earlier at a spot where music and perforation were identical.

RAB: I first came in touch with 'Living Presence' many years ago as a student on a summer job when I was traveling to New York. On 1600 Broadway I bought the 'Peacock Variations' by Zoltan Kodály with Antal Dorati as conductor.

WCF: With the Chicago Symphony. That was in mono. We did not do mono-recordings in this series of CD's.

RAB: No stereo-version was made in later years?

WCF: No.

RAB: I thought it was very well done. One could hear a big orchestra in quite a natural way. Everything was there in the right perspective.

(I told Wilma Cozart Fine that I bought several Mercury LPs: Brahms' Second Piano Concerto with Gina Bachauer and Stanislaw Skrowaczewski (SR 90301), Rachmaninoff's 1st and Prokofiev's 3rd with Byron Janis and Kyril Kondrashin (SR 90300), and Prokofiev's Fifth Symphony with Dorati (SR 90258). We talked about how in the 3rd movement of Brahms' Second Concerto you could almost see the strings of the cello, and how beautifully Janis played the second movement in Rachmaninoff and the variations in Prokofiev. Exceptional were the sections of Brass and percussion in Prokofiev's 5th Symphony. But all instruments are all there in the right proportions. Nowhere there is an emphasis on a single instrument in particular or on some specific aspect of an instrument. These works are also available on CD.)

RAB: Not long ago I bought a Nixa LP with Vaughan-Williams' 8th Symphony, out of curiosity and I was amazed by the excellent recording. Now I do read in the notes of Polygram Classics & Jazz that you and the 'Living Presence' team made this recording in England in 1956.

WCF: Nixa was a subsidiary of the Pye factories. We made various recordings for them. Later the contract between Philips and Pye ended and Pye was purchased by EMI. EMI releases the recordings that we made for Nixa.

RAB: Apart from the Sviatoslav Richter-recordings for Philips, did you make recordings for other labels?

WCF: Yes we made a number of recordings for the Italian publisher Ricordi. They too do not belong to Philips. The specialty of the releases of the Mercury-recordings on CD is that we restored the old equipment and used it, I do not know how EMI does it. They do not have our equipment.