ORTF 💫

 

Today engineer Albert Laracine, the inventor of the ORTF technique in the 1960s, passed away. 

As a tribute, I will share some information I have about its history.

[Quotations from Albert Laracine, Marcel Barbin, Alain Gerbe]

This microphone technique was popularized in the 1960s and was named after the French National Broadcasting Agency (developed by Albert Laracine under the direction of R. Condamines at the RTF. Albert Laracine was one of the pioneers in France in high-quality stereophonic recording in the 1960s), consisting of an arrangement of two cardioid microphones with a capsule spacing of 17 cm (similar to the average spacing of a pair of human ears) and a physical opening angle of 110°. Angular distortion (within the SRA) is around 5°, which is not bad, but slightly higher than other similar techniques. 

Although the most precise positioning (lower distortion in the stereo image of the instruments) is achieved by placing the microphones at the exact distance of the recording angle (in this case between 67° and 101°), this distance can often result in an unbalanced relationship between direct and reverberant sound. 

The practice of using ORTF (or “le couple ORTF” in French) closer to the source is common, although this causes a “flattening” of all instruments located outside the 101° SRA field. In other words, if we have, for example, four rows of violins outside the recording angle, the four rows will sound as though they come from the same position (the far left of the stereo image, for instance). It is up to you to determine your priorities and whether this stereo-image distortion is desirable/acceptable (or whether another microphone configuration should be used). 

One possible approach, therefore, is to begin determining the ORTF location based on the balance between direct sound and reverberation, even if this results in some stereo-image distortion. The author considers this preferable to an excessively reverberant sound, especially in problematic recording rooms. 

The resulting image is usually very natural for instruments located within the SRA (recording angle), and complementing the cardioid pair with two omnidirectional or subcardioid outriggers can be an excellent combination when recording larger ensembles. 

In the book “A Practical Guide in the Decca Tradition” (recommended reading) by Caroline Haigh, John Dunkerley and Mark Rogers, the authors compare positioning between the Decca Tree and ORTF techniques. If the main pickup uses directional microphones that naturally capture less reverberant sound (such as an ORTF pair), it will be mounted higher and possibly farther back than the Decca Tree. 

The microphones may be placed at such a height and distance that the distance to the front strings and to the winds is not very different (although keeping some extra distance from the winds is desirable to create an appropriate sense of depth). After finding the reverberation balance, the next step is angling the microphones to obtain the desired balance between sound sources. This was addressed earlier, but let us recap using the chart beside it. If possible, when using other stereo microphone techniques or spots, consider employing combinations that result in similar positioning on the stereo image. This will help create a natural and well-defined image even when mixing different microphones. 

The angular-distortion charts created by the author help form an overview of the distortion of each technique, but with so many variables it is preferable to use sites or apps that perform the calculations in real time for stereo-image visualization. You may therefore vary source distance, microphone spacing and angles to generate the desired result. 

Microphones 

When comparing the microphones used in Europe with those used in the United States, what stands out is the very large number of electrostatics used in Europe. The reasons for this are multiple, the first perhaps being that most electrostatic manufacturers are European, and the second their high transient quality, or perhaps the reverse. The exact stock of electrostatics at ORTF (in 1974) is unknown to the author, but is probably around 3,000 at minimum. 

Their main use was for studio work, but they were also used for outside broadcasts of musical programs and even variety shows. This shows how electrostatics had become stable and immune to difficult working conditions, especially after developments related to working conditions in TV studios. For example, television encouraged the use of diaphragms that withstand temperatures up to 80°C with virtually no change in response curve or directivity. Television also promoted the use of cardioid, hypercardioid or supercardioid systems. Decent dynamic types must now have internal damping to avoid effects of mechanical shocks and friction; this also applies to lavalier microphones frequently used in round-table debates and variety programs. There is no standardization of microphone types at ORTF (in 1974), except for the general use of electrostatics and the specified 200-ohm impedance with a special socket. 

The modern electrostatics of the time were powered by 9 volts (phantom power). This value was chosen for the first transistorized types because of the possibility of using an inexpensive 9-volt battery for field work. Trends in other European studios pointed toward a 48-volt system and a 12-volt system. The 48-volt system required either a converter outdoors or bulky batteries. 

Contribution of Schoeps 

According to Schoeps, around 1962: The “ORTF” stereo recording method was developed using tube microphones from Schoeps and custom mounting accessories. The essential setup was a pair of cardioids spaced 17 cm apart with an angle of 110° between them. 

The author believes that the microphone used was likely the M 221 B. 

1974: The ORTF method was specifically supported at the beginning of the Colette series, using a pair of Colette cardioid microphones, two KC active cables and the special STC stereo bar, so that the correct spacing and angle between capsules were already predetermined. 

1976: MSTC 3/3p/4/5: The MSTC is the first microphone specifically designed for this recording technique, with quick setup and simple cabling. The four different amplifier types supported the four predominant microphone-powering methods, all offering identical performance. 

1992: MSTC 6: As with the CMC 6 the previous year, standard phantom power of 12 and 48 volts was now available in a single amplifier that switched modes automatically. (Parallel phantom power and 12-volt “negative” powering had virtually disappeared by this time.) 

2021: MSTC 7: The latest version uses the updated circuitry of the CMC 1 amplifier, with lower operating current, equal or greater headroom, better interference protection and a new “look”. 

Albert Laracine 

It is not possible to properly portray the ORTF technique and the institution from which it came without speaking about Albert Laracine. There is little information about this chapter in audio history, and the author attempts here to make a historical record not available in other books. The following are internal ORTF documents, interviews and publications by Albert himself. 

INTERVIEW: 

It is unlikely that the name Albert Laracine is familiar to you. He is one of those who, initially in the shadow of ORTF’s “giraffes” and now the INA, teach, research and ultimately make radio. 

Shepherd of microphones, guardian angel of sound engineers, he is for many the symbol of a certain idea of this profession. ... Well, I feel that we never recreate reality, but manufacture a product that on one hand does not resemble the original (which, I admit, is not necessarily a bad thing: it is debatable) and, on the other hand, in my opinion, does not serve music. 

For me, music is an art of communication between people in every sense. When there are several musicians, music is not just in the notes each one plays, but also in the interactions between the different timbres. When one does spot-microphone analysis, it is impossible to obtain or reproduce this subtle interplay of interferences. In short, one does not obtain what seems essential: the blending of sounds. I believe, moreover, that what most listeners expect goes in this direction: to hear a flute, a harp, an oboe—they do not care! What matters to them is the overall musical atmosphere. 

As soon as sound is fragmented in recording, part of it dies. Not everything, of course, but a part. And not an insignificant part... ... The ORTF pair was born from imitation of nature, quite simply: the artificial head with two microphones in place of the ears. This concept was not revolutionary, since the first tests in this field must have occurred at the end of the last century, perhaps in 1887, after the last war. 

Michel Philippot, by placing two microphones where the ears would be (he did not have a proper artificial head when working at the Théâtre des Champs Elysées), and because he himself had excellent natural ears, achieved admirable sound recordings that are still considered models today. The Acoustics Laboratory at the time was tasked with studying this experiment to determine the parameters that would yield the best possible results from the process. In short, it was a matter of rationalizing what had been obtained by trial and error. 

The laboratory began experimenting with all existing systems, including MS, multi-microphones... At that time, I was only a listener whose opinion they sought on what I heard. One among others, since the Research Laboratory aimed to collect a statistical opinion. Through successive approximations, the MS system was eliminated, as was the 90° bidirectional microphone (Blumlein) and several others. Finally, the system of two cardioids used by Philippot was chosen, but the exact positioning was still unknown. 

Tests were multiplied in Bordeaux (we had no free studios in Paris) on theater plays, music, presenter interventions... I will spare the details. In any case, it was from this study that the parameters we use today emerged. Many researchers also studied the problem using purely mathematical methods (Karlceon must be mentioned for his remarkable work). All arrived at the conclusion that 17 cm and 110° did not constitute the ideal solution, but represented at least the least bad of all those proposed up to then. 

For us, in any case, it was the one that provided the best spatial distribution in width and depth: the one that distorted space the least. It must be made clear, however, that this is valid only when considering loudspeaker listening with the listener well positioned, forming the three vertices of an equilateral triangle. 

The system was developed specifically for use in this situation. I think that if headphone listening had been chosen as the basis, we would have arrived at different results. That said, this is only one stage: now we must go further... ...The MS pair was eliminated because the impression of space—and we are increasingly convinced of this as our research progresses—is given by differences in intensity and time. 

It is the brain that decodes this information to create the impression of space. In the MS system, there are no time differences. Why? It was the consequence of a deliberate intention. At the time, we worked mainly for listeners who listened in mono, and the problem of compatibility arose. It is evident that the physical compatibility of two ORTF-style microphones is not good. Acoustically, however, it is sufficient: in any case, for years and years we never had complaints. But the Germans, and others far more organized and precise than we were, said: no, this is not possible, you must use a recording method that does not introduce phase difference and therefore no time difference. 

To meet this expectation, it is undeniable that the MS pair is by far the best possible solution. But you have understood why, on our side, we eliminated it immediately... 

DEVELOPMENT OF STEREOPHONY IN FRANCE ACCOUNTS OF A. LARACINE: 

In 1945, in the experimental studio directed by Pierre Schaeffer, José Bernard and Jean Wilfrid Garrett revived the old research from the 1930s on the use of an artificial head to reproduce a spatial sound image over headphones, using as test subjects the students of the sound-capture class of which I was part (we may also mention Clément Ader’s demonstrations in 1881, who carried out the first public wired transmissions, based, according to him, on stereoscopic effects). 

During this work, the first radio transmission took place, with the two channels switched on two AM transmitters (National chain and Parisian chain), the only ones operated by France at the time for national broadcasting. It was also at this moment that the first loudspeaker auditions were attempted and the equilateral triangle was outlined. A broadcast requiring two transmission chains was unrealistic at the time, and this is probably why there was a long period of inactivity, during which only a few broadcasts, such as the famous “Larme du Diable”, were carried out, but they had only a distant relation with stereophony. It was a show that used two sound channels without any relationship between them. 

With the introduction of frequency modulation and the construction by Mr. Herbaut of a transmission-reception system allowing the simultaneous transmission of two very high-quality sound channels, beginning in 1956 and the following years, research was conducted at the ORTF Acoustics Laboratory under the direction of Monsieur Chatenay and his deputy Mr. Condamines. All existing solutions and systems were thoroughly tested and compared during blind auditions by various groups: musicians, actors, assorted listeners and, of course, chief sound operators. Many artificial heads were built in the Laboratory to achieve what we considered the best compromise: the ORTF Pair, initially built by hand and later developed by the SCHOEPS company. 

Every year, during the Festival du Son in Paris, under the initiative of Michel Philippot, friendly comparisons brought together all sound engineers and producers from international broadcasting organizations. Each presented their best productions, explained their working methods, and general discussions allowed for comparison of viewpoints. 

These meetings were very fruitful. The ORTF Pair, although highly appreciated by all professionals for its qualities, did not enjoy the immediate success expected, since it is less monophonically compatible than coincident systems, which, at a time when the vast majority of listeners listened in this way, was an inconvenient and even prohibitive flaw for some. 

Unfortunately, now that this problem no longer exists, since mono listeners have become a small minority, the French method struggles to establish itself, and almost no one knows why XY, MS or multi-microphone systems are preferred. But through habit, it is still little used. 

Evolution of audio equipment at ORTF 

At the end of the 1939–1945 war, ORTF (RTF) was left with a considerable number of burned transmitters and worn-out audio equipment. The studios had only portable mixers with two to four inputs; only two new consoles were built during 1944 with considerable hardware difficulties. The new equipment studied before 1945 was quickly put into service from 1946 to 1947. Special features included the use of balancing faders between microphone inputs, between the four disc inputs, and between all microphones and all discs. This special arrangement was created mainly because of the importance of drama programming and worked well for classical-music recordings. It was insufficient for variety programs and had to be complemented in those cases with additional mixers. 

Since 1938, and until the arrival of the tape recorder, program recording was done with disc-cutting lathes, usually two per studio with a (manual) synchronization device allowing continuous recording. Naturally, all this equipment used tubes and was transformer-coupled between elements. 

Around 1960, a new generation of audio consoles appeared, characterized by the use of linear-travel faders in all main positions. There were 8 microphone inputs on these sound consoles, plus 4 machine, tape and disc inputs, one or two main outputs, plus echo and monitoring outputs. The monitoring and intercom system was built in. The nominal microphone input level was minus 40 dB, and in an external rack operators could configure a preamp for ribbon microphones or dynamic microphones with 20 dB gain, or a power source for electrostatics. 

Some correction devices were added to the console. In 1967, the first transistorized sound console went on air in TV studio 15. Silicon was used. The mechanical layout was similar to that of the 1960 consoles, except that the number of inputs increased considerably, especially for TV (18 microphones plus machines, 2 VTR inputs and 2 telecine inputs). More filters and equalizers were added, and 4 main outputs were available, as well as secondary outputs. Since this equipment was considered not very flexible and also resulted in very large consoles, a new type was launched in 1969, smaller but with larger patch panels. Small consoles with 12 inputs, 4 mix outputs, controlled by direct current, were introduced. The nominal gain of the transistors was up to 10 dB higher than in previous models to allow adequate pickup of low-level sounds in TV. All channels in these consoles had insertion points, providing the possibility of adding special effects such as compressors, special filters, etc. In 1972, as costs increased, it was decided to add more flexibility to the equipment so that any type of console could be built, from a simple 4-input mixer to larger types. With the help of industry, a modular type now emerged, each function—attenuation, filtering, switching, etc.—assembled in independent modules in light metal cases. DC control was provided for faders with a view toward future automation applications. 

The first generation of this is currently being tested. Regarding automation in audio consoles, the opinion of ORTF working group S7 is that such an application could be of interest for mixing studios, whether audio or film, where sound is recorded on multiple tracks. “Auto-memory” would essentially be a memorization of fader positions both during mixing and playback, as well as the recording of filter positions. This would facilitate repetition and readjustment of the mix. Another form of automation was introduced at ORTF with one of the automatic-sequence programs (3rd program). The system automatically sets the appropriate configurations for VTRs or telecine, as well as the appropriate switching for studios. Sound fades or mixes are performed between programs under predetermined conditions for fade time and fade amount. In case of special or urgent announcements, the system can be temporarily switched to manual control. Another application of “automation” exists in standard EBU commentary units. In these units, the commentator’s voice level is controlled automatically, and the voice-background mix is done automatically using special compression devices. This may lead to a semi-automatic news studio. It should also be said that the new switching equipment used by ORTF is entirely solid-state, except for the telephone for now, due to high cost. Problems of stereophony All sound consoles built since 1960 can work in stereophonic mode, adapting when necessary pan-pots or a special unit combining pan-pot and base control. An A + B signal is produced using pairs of electrostatic cardioid microphones at 110° with membrane axes 17 cm apart. The console faders are then mechanically coupled by rails. Stereo program material, as broadcast, is composed partly of discs and partly of ORTF productions. 

From time to time, international stereo concerts are broadcast, mainly under EBU supervision. There have been experimental stereophonic TV broadcasts at ORTF. The first, around 1964, was done with a front-back sound effect in the play “Les Perses”. The sound part was recorded on 16 mm magnetic tape using the standard center track for the front sound and a side track for rear effects. Front sound was transmitted on the normal TV sound channel and background effects via AM transmitters. Three other productions of about 20 minutes were recorded since then with the normal left-right system. The A + B result was used as mono-compatible sound in the normal stereo system, and A and B in the normal stereo program. Two musical pieces and a drama were recorded and broadcast. One of the goals discussed was that it would be perfectly possible to use a normal stereo listening setup as long as sound coming from “off-screen” elements was compatible with actor movements, camera panning, etc. Technically, this means an extensive rewriting of the script and the possibility of switching “cutting” with changes in the camera position of the microphones (left-right to right-left). The issue of stereo sound for TV raises the question of two-channel sound with images. An international standard must also be defined for this purpose. Reports on this were submitted by the EBU to CCIR Committee 10. This would also leave the possibility of transmitting in two languages and would mean, if not used for stereo, the possibility of dual information in TV programs. CCIR refers to the International Radio Consultative Committee, which defined standards for radio communications and, by extension, broadcasting. UER is the abbreviation of “Union Européenne de Radio-Télévision”, known in English as the European Broadcasting Union (EBU).

~~~~~~~~~

Thanks to Julian J. Ludwig for the above essay.

RIP to Albert Laracine and thanks for his contributions to music 💫