Audio Etc. (Feb. 1987)

Home | Audio Magazine | Stereo Review magazine | Good Sound | Troubleshooting


Departments | Features | ADs | Equipment | Music/Recordings | History




MIKING A POINT


The development of Denon's new super-one-point stereo digital recording technique (using many mikes), and of the prototype Denon Digital Time Alignment Con sole, had directly to do with the music of Gustav Mahler, The technique was shaped for Mahler during the year and some months it took to record his nine-plus giant symphonies, with the Frankfurt Radio Symphony Orchestra under Eliahu Inbal, in Frankfurt's Alte Oper concert hall.

The symphonies do differ. And so the technique was worked out gradually according to need, reaching its final definition, appropriately enough, in the huge Symphony No. 8 with some 700 performers, as I described in this space last month.

Out of the six-language control-room briefing for journalists held at the time of that recording session, and from some helpful Denon fact sheets (in English), I have learned enough about the procedures to fill out the picture for you. Yes, all these recordings, including No. 8 with some 24 mikes, are in the true sense "ore-point" jobs, picked up by a single pair of stereo mikes at a single optimum location in the hall the sort of recording style that became famous in the 1950s with Mercury's Olympic series of mono LPs. That magic spot, not always easy to pin point in actual practice, was definitely located in the Alte Oper just before the recording of Mahler's "littlest" symphony, No. 4, says Denon. This work, featuring a soprano soloist, was recorded entirely without extra mikes, from a single point only. Not only that: The two mike signals were fed directly into the digital recorder, not even going into a mixing console. How's that for purity? As per last month's account, No. 5 was almost as pure, with only a very spare, occasional use of "assistant mike" coloration. But there were still the biggest symphonies to come, and they were a problem, obviously, since they required vast assemblages, immense spatial layouts, acres of musicians. Wisely, these symphonies were recorded more or less in order of size and complexity, through the big No. 8.

That way, the developing time-delay technique, beginning with Nos. 6 and 7, recorded last spring, could increasingly be used to "conquer space," so to speak, compensating for all those vast distances from close-up mikes to the main stereo mikes. With time delay, the entire orchestra, chorus and soloists could be brought within a few feet of the central spot-and yet simultaneously remain in their proper perspective out in the hall.

In No. 6 and No. 7, for instance, there is a cowbell, a snapped whip, a mandolin, a guitar, a tinkly celesta, and even a hammer (hammering on some thing that sounds like a good two-by four). In recording, all of these needed some extra definition, and they got it, time delayed. So did the big percussion, far to the rear of the huge stage.

Denon is a bit confusing in its use of singular and plural, but a moment of study makes all clear: There were of course two omni mikes in stereo, up front, mounted a couple of feet apart for more realistic ambience. In a huge hall, that's a single point. On the other hand, the assistant mikes, which I now find were indeed B & K cardioid proto types first used in these sessions, were not stereo-each was a single pickup and thus directionless. Mixed into the two front stereo channels--with digital delay--they merely added color, definition, roundness, presence, without affecting the stereo spread. Nor, thanks to the well-known cardioid "sonic spotlight," did they appreciably add to the recorded ambience. Not even with the large choral forces in No. 8; each choir sang within the spot light of its own assistant mike.

These B & K cardioids, by the way, apparently have unusually clean "fringe" response off their central axis, which is why they blend well with the B & K omnis used for the main stereo.

Now for the follow-up. Denon's PCM digital recorder has only four tracks. Enough for 700 performers? With ingenuity, yes.

The main stereo channels took priority, each mike going to its own separate track, unmixed. These two channels are the vital core of the recordings. For most of the Mahler, the other two tracks took the combined mix from the assistant mikes, whether few or many. For these, obviously, some mix-down was necessary at the recording session, done in the normal manner. But the result was kept strictly separate from the main channels.

On special occasions this distribution may be changed. For possible future surround sound, to take an example, at the enormous final climax of Symphony No. 8, with all 700 performers going full blast and a brass choir far up in the rear balcony, the brass was given its own track (presumably the assistant mikes were lumped together; it wouldn't matter much in that roar of sound!). Later on, that brass can be "placed" wherever desirable for a surround effect.

I did not get to learn which assistant mikes were mixed into which track with which others. Too many people, too many languages. But this is not important, since the mix would be different for each work of music, in any case.

What matters is that ingenious time de lay to bring all those mikes, no matter how many, into phase with the front stereo mike pickup. For the entire 700 performers (more or less, mind you--I didn't count them), Denon got away with only three groups, organized ac cording to distance, using only three of the nine available delay modules in the console. That's ingenuity.

The Frankfurt Radio Symphony does indeed go on the air, in addition to giving regular concerts on a par with those of our best orchestras. There fore, the Mahler performances were re corded for broadcast as well as for Denon's CDs. From something said by Peter Willemoes, Denon's Danish associate engineer, I gather that at first the Mahler concerts were miked separately for broadcast by the Hessischer Rundfunk, the local public radio. Later, it seems, they switched to a mixdown taken from Denon's tapes, which conveniently have the pure stereo one-point pickup-ideal for broadcast-on separate tracks, and the assistant material available to be added according to the Rundfunk's taste. Their radio must be mono-compatible, and no doubt compressed to bring Mahler to radio size--a very different mix from the final Denon product, yet out of the same tapes.

The outward recording schedule for all this Mahler is interesting. It was set up to allow an interchange, for repairs, between concert recordings at the superb Alte Oper and no-audience sessions at the same place. That is vitally helpful in case of disaster, major or minor. Considering its vast space, the hall has a relatively small number of seats, and every one of them is padded with a species of plush or pile fabric which simulates the sound-swallowing effect of an average human body. Thus the reverberation changes minimally with an audience on hand, and a mix can be made, for repairs, with only a small adjustment of the ambience. A mistake or an unwanted noise in a concert recording? It can be removed with a segment of non-concert rehearsal, or even with a special retake if feasible.

Thus Denon got down as many as five complete performances of the Mahler symphonies, several concerts plus a rehearsal/run-through session ahead of time and even a retake session as an extra. There were two concerts of the big Symphony No. 8 on successive nights while I was there, and both were recorded. There was also an earlier run-through, with starts and stops, and no audience-I heard someone on that tape say "take 24," in English. It is dangerous to depend entirely on concert performances, where stops are impossible. The same mis take can happen twice-the same bus can roar past at exactly the same moment. With interchangeable stop-and-go studio-style sessions, there is ample insurance.

I should note again that the German audience was astonishingly quiet and cooperative. That helps enormously, as every recordist knows.

Ah, that hall! As I've said, it was recently built inside the bombed-out shell of the old opera house, which had been standing in picturesque ruins for more than 40 years, ever since WWII.

The new concert hall opened only in 1981, long after-thank heavens-the disastrous postwar period when "modern" halls, one after the other, proved to be acoustic monstrosities in spite of all the glass and metal, the geometrical curves and the flat planes, the bizarre acoustic "island" reflectors hung in space. This time out, the designers knew what they were doing, even in a space that might be the biggest ever.

The hall is built from what we would call the third floor upward. Underneath is a fabulous low assembly room, with a mirrored ceiling lit by thousands of tiny star-like bulbs and a mirrored divider halfway down to double the whole. If you fumble long enough against that mirror, you discover the restrooms behind. Opposite, a coat check half a block wide. Big thinking.

Inside, the hall seems very wide, and it is very high. A vast, deep balcony structure takes up half of what is a semi-rectangle, though the main hall seems squarish. I soon noticed that there are no parallel walls; the sides converge, and the ends do not match.

Balcony extensions curve out from each side and all the way around the front, beyond the stage on the floor.

Nothing is geometrical; the two side walls are covered with various levels of box-like balconies and enclosed glass booths in front (not used by Denon). All this is built out of expensive wood deep russet-colored mahogany. Superb for sound.

Moreover, those two slightly con verging side walls, I gradually realized, are completely asymmetrical. Not a single shape on one side is matched by the same on the other. Complementary for the eye-but different for reflected sound. The high ceiling appeared to be hung, sloping forward (?), with rows of small reflecting planes at various angles. Down below were those plush seats, a dull greenish color to balance the russet mahogany. All this is astonishingly harmonious, a prize example of order within deliberately calculated chaos, and one wonders why it hasn't been done this way before. When you come down to it, that postwar architectural style didn't allow for good acoustics. Now there are new styles of building, and we can apply our acoustical knowledge to good effect at last.

All this may help explain the extraordinary sonic effect of this big space. Directionality is perfect, never con fused; you see, and you hear what you see. Everything is ultra-clear and sharp, never dry and dead, always with good presence. But what surprised me most, as I've already said, is that the hall sounded like the Denon recordings, not the other way around.

To illustrate: From my seat in row 8 on the floor, the seven solo singers were lined up across the front of the stage, each with a close-up solo assistant mike, from maybe 25 to more than 50 feet away from me. But every one of them, in that purely acoustic transmission, sounded crisp and clear, their words sharply defined-as though those individual mikes were shaping a close-up sound. Indeed they were, but only for the recording! We at the con cert heard 100% acoustic sound. Nevertheless, the effect was exactly like a good recording of solo voices. I could hear other solo sounds, far in the orchestra rear, even segments of chorus still more distant, that had exactly the same "assistant mike" clarity, as heard in the best recording. But this was strictly the hall itself! In such a place, how can you go wrong?

(by: EDWARD TATNALL CANBY; adapted from Audio magazine, Feb. 1987)

= = = =

Prev. | Next

Top of Page    Home

Updated: Sunday, 2019-05-05 10:26 PST