The Fifth Element

There are two compelling and (one not so compelling) reasons to record this way, instead of just using two actual rather than virtual angled X-Y directional microphones. First, unlike X-Y pairs, in M-S technique the Mid mike is completely on-axis to the sound source. Obviously, this would make less of a difference for an orchestra than for a point source. But for natural recordings of voice, guitar, harpsichord, piano, or any other small source, it should make a worthwhile difference.

Second, by using M-S technique, very fine adjustments of the balance of direct and ambient sounds are possible far more easily than by moving the microphones' physical locations, by merely turning the Width knob on the M-S matrix.

The not-so-compelling reason to use M-S technique is actually the reason for which Alan Blumlein's invention of it was first commercially implemented, by Holger Lauridsen of Danish State Radio. M-S allows stereo broadcasts to be auditioned in mono with little or no degradation. When the Left and Right channels are combined, the ambient information, being phase and antiphase from a common source, cancels itself out, leaving a clear mono signal.

M-S's combination of stereo image flexibility and mono compatibility make it a natural for film-sound use. Indeed, Jerry Bruck was Stereo Sound Consultant for Fame, believed to be the first use of M-S microphone technique for a motion-picture soundtrack. Since then, M-S technique has become well established in the film-sound industry.

Practical M-S
Unless you've had the experience of hearing the width of the stereo soundstage and associated room sound go from "not enough" to "too much," then back to "just right," all at the turn of a knob, you can't understand how empowering M-S technique is—or how much good clean fun it can be. So here's what I will do. I will make a recording of a point source in which the stereo width goes from all-direct sound to all-ambient sound, then back to what I think is an optimum blend . (You can download these examples, burn them to a CDR, and play it on your stereo—or even just route it or Airport it over there, if you're set up for that.)

To use the True Systems P2analog in M-S mode, you connect the Mid mike to Channel 1, the Side mike to Channel 2, and engage the M-S button on the front panel. The Channel 1 gain knob controls the volume of the Mid mike, while the Channel 2 gain knob controls stereo width. The P2's analog outputs then put out conventional Left and Right stereo signals. Of course, with the M-S function not engaged, the P2 functions as a conventional 2-channel stereo microphone preamplifier.

As if M-S capability weren't enough, the P2 also boasts a stereo image (phase) indicator consisting of a row of colored lights that run, left to right, from green to yellow to red. The left, green end indicates that both channels are largely in phase, which means that you are, intentionally or not, recording in mono. The right, red end indicates that both channels are largely out of phase, which suggests that you might have something connected wrong, or that you really like the sound of an M-S array with all S and no M. Centered in the middle of yellow, with excursions into red and green on both sides, means that you're in the stereo ballpark. The stereo display is not only a confidence enhancer, it's addicting to watch. I was happiest with the image when the indicators were more green than red, but I hate the hole-in-the-middle effect.

The P2 has an aluminized front panel of a distinctive fire-engine red. Its industrial design and fit and finish are excellent, as is the owner's manual.

Into the sanctuary
Steve Dreyfuss, an affable local recording engineer, allowed me to tag along on some of his projects so I could play with my new toys. We first made about two dozen mike-position and pattern tests in the Meeting House (1774–75) of the First Baptist Church in America, recording the 1884 Hook-Foley/Baker organ (which includes pipes from the original organ of 1834), preparatory to recording it for the Rhode Island chapter of the American Guild of Organists' planned CD documenting significant pipe organs of Rhode Island. Test-to-test repeatability was ensured by organist Steve Martorella's having recorded a one-minute soft-to-loud flourish onto the organ's hard-disk drive (a "player" system somewhat akin to Yamaha's Disklavier), so it could be replayed at the touch of a button.

Not to take anything away from the True Systems P2analog, but if you're willing to spend about the same money and do without the P2's direct inputs, M-S matrix, and stereo indicator, there is one microphone preamplifier that I found sounded even more liquid, coherent, and involving: the Grace Design Model 201 ($1995). This is a stripped-down hot-rod of a purist mike preamp: for each channel there is a single indicator light for signal presence (green) and the approach of overload (it changes to red). The construction quality is deluxe. Thoughtful touches abound, such as small ovals made of dry-erase marker-board material for each channel, useful for making notes about settings.

Because the Grace 201 uses stepped resistor arrays to set Gain, it also provides rotary-potentiometer Trim controls (with 10dB ranges) to set fine level adjustments. In normal practice, the Trim pots are left turned fully clockwise; fine adjustments are accomplished only by reducing the recording level, not increasing it. The Trim controls are good to have, because no matter how deft the hand that switches the Gain settings, there will be an audible discontinuity in the recorded sound as one set of resistors is switched out and another is switched in.

For both the True and the Grace, balanced analog outputs were carried to the digital recorders on Wireworld Eclipse 5 interconnects ($345/0.5m balanced pair). I found the construction quality superb; the XLR connectors were a joy to use. Their sound was sufficiently engaging that when, to break in the Eclipse 5s, I used them to run the darTZeel nhb-108 power amp, it was no letdown at all.

In two words, the True Systems P2analog is more analytical, the Grace Design 201 more lyrical. I think that both are cases of getting what you pay for. Were the final microphone choice to be an M-S array, the True Systems would get my nod; were it to be a more conventional array, I would choose the Grace Design. No tears and no hearts breakin', no remorse.

The envelope, please
A few days after the recording tests at the Meeting House, Steve Dreyfuss, Steve Martorella, and I gathered to hear the results on the ESP Concert Grand SI loudspeakers, darTZeel amplifier, and EMM Labs source components. The differences were intriguing.

My favorite array was M-S with figure-8 microphones for both Mid and Side (in this case, AKG 414 dual-capsule, variable-pattern condenser mikes).

I felt that this gave the most detailed center image, as well as conveying information about the room's shape and not just its size. Figure-8 mikes, however, roll off in the bass. That is a tradeoff I might be willing to make for my own listening, but in the real world, most people judge an organ recording by the bass pedals and little else.

A compromise I could live with was M-S mode, figure-8 for Side (of course) and wide cardioid for Mid.

Interestingly enough, a third Steve, Stereophile's Stephen Mejias, with whom I shared the test recordings, preferred wide cardioids in X-Y mode angled at 120°.

However, nothing gives an impression of size like an A-B spaced-microphone array. As of this writing we are still of two minds (at least) between spaced wide cardioids.

Steve Martorella and I think this has the best overall tonal balance. However, Steve Dreyfuss' preference is for spaced omnidirectionals, which gives the most bass.

I like to benchmark my own work against commercially successful recordings, so one of the tracks I thereafter played was "Jul," from the audiophile staple Cantate Domino (Proprius CDP 7762). Steve Martorella did a bit of a double take and asked to see the CD booklet. He read it, looked up, and said, "I've played this organ!" Seeing as the organ in question is in Stockholm, all I can say is: schmall verld! I think our work held its own very well against Cantate Domino.

A new organ recording I've been listening to a lot is In Spiritum: Olivier Latry Plays César Franck, recorded at Notre Dame de Paris (SACD, Deutsche Grammophon 00289 477 5418). The playing and sound are beyond superb; the delicacy of Latry's playing in the Op.18 Prélude is especially gratifying. Highly recommended.

DSD on the cheap!
Back in 2002, I wrote: "DAT is, as far as I am concerned, the most practical and intuitive approach to affordable live digital recording." Belay that. Digital audio tape, sad to say, is on its way out. Although DAT machines will continue to be made for the foreseeable future, the handwriting is on the wall: as time goes on, fewer and fewer DAT machines will be supported by replacement parts.

But despair not. The TASCAM DV-RA1000 digital recorder sounds phenomenal and is a screaming bargain ($1500, frequently discounted; www.tascam.com). The DV-RA1000 is a component-width, two-rack-unit–height professional recorder that records "Red Book" 16-bit/44.1kHz CD data to conventional CDRs, or—this is the rather mind-boggling part—records hi-rez audio directly to DVD+RW blanks, all the way from 24/88.2 PCM up to Direct Stream Digital (DSD), the SACD format (footnote 4). (The DV-R1000 records DSD as DSDIFF files, the raw data format used to create an SACD file.)

Two good microphones (the AKG 414 has a street price under $1000), some cables and stands, a good mike preamp, and the DV-RA1000 will, all told, set you back less than $6000, and the results are tremendous. We recorded the 24 microphone-placement and pattern tests to "Red Book" standards for transportability. However, after we were covered on that, I asked Steve Martorella to repeat the test piece and also to play live, so we could have DSD tracks for comparison. I don't think anyone will be surprised to learn that the DSD versions blew the "Red Book" tracks into the weeds. Forget the music, even—DSD's superiority was plain from the room ambience and my speaking voice while slating the takes.

So, have we immanentized the eschaton? Not quite, sorry to say. Once you've recorded something in DSD onto a DVD+RW, apart from listening to it on the DV-RA1000, playing it for your friends, or lending it to another DV-RA1000 owner, there isn't much you can do with it. Even basic editing of DSD is a legendary chore. As far as I know, the cheapest rig that would let you go from DSD tracks on a DVD+RW to an authored project ready to send to the SACD pressing plant are $25,000 Sadie or Pyramix workstations.

My guesses are that dilettantes—I use the word in its good sense, of an admirer or lover of the arts—will have fun with the DV-RA1000's DSD capabilities, while professional users will use it to capture higher- but not highest-rez PCM, then bounce those tracks over to a more affordable PC- or Mac-based PCM music-production program. One could even take a thwack at the piñata by the Quixotic gesture of using a DVD-Audio production program such as Minnetonka Discwelder (from $99, footnote 5) to create regular DVD-Video discs, the default stereo soundtracks of which would be 24/96 PCM, and replicate those videoless DVD-Vs on a DVD burner. Perhaps that might work for live-band tapers. Indeed, 24/96 hi-rez stereo on DVD-V was the hi-rez format that should have been, as it is playable on the untold millions of standard DVD-Video–only players. It was doomed by its lack of copy protection and surround-sound capability.

Questions or comments.

---

Footnote 4: Answers to FAQs on TASCAM's website claim that the DV-RA1000 is a true DSD recorder and that it does not reformat hi-rez PCM to get DSD.

Footnote 5: By the time you read this, TASCAM may have released a special version of Discwelder Bronze, available free to registered DV-RA1000 owners in the US, that will import (but not export) DSD files.