How did you start recording musical artists?
We began 20 years ago with a solo piano album recorded in Zipper Hall, which is a beautiful 430 seat concert hall right across the street from what became Walt Disney Concert Hall [in L.A.]. We had wonderful microphones, two [Neumann] U 47s, I believe. We recorded with about a 10-hour microphone and hall set up. The techniques have remained fairly much the same since then. We usually use classic, vacuum tube microphones. Those U 47s, I think, were from the early 1970s. The microphone I use the most now is an AKG C24 stereo mic from 1969, I believe, which used to belong to Frank Sinatra. From the microphones, we then go straight to our recording media. No mixer. We are also now using SonoruS Audio's Holographic Imaging to capture and incorporate two rear channels so that we can also release surround sound for NativeDSD [Music, audio download site]. Multi-channel is now our norm, as well as the stereo album. We go from our microphone through a Yarlung microphone preamplifier – designed and made by Elliot Midwood for us – and then from there we go straight to the analog tape deck, our SonoruS ATR12. And to record high resolution, we use the Merging Technologies Hapi recording at 256 fs DSD. We also use the SonoruS ADC to capture high resolution PCM. Essentially three different originals split off of the same signal coming from the microphones and the microphone preamps. And that's it. The challenge with this – as you know perfectly well from setting up in very difficult situations outdoors – is that once we have the recording, that's the recording. The mixing and mastering, so to speak, happens onstage with the musicians. If we want to change the EQ, we do that by changing the microphone position or adjusting the acoustics in the hall beforehand. If we want to change the depth of the recording, then we do so in-person, physically onstage. It’s such that once we have the recording captured in its three different media, that's what gets released as analog tape, CD, high res downloads, or MP3s. That's what the end-user hears. It's very simple and it's also excruciating. [laughs] The musicians know this in advance, and I prepare them so that our first day is essentially a setup day. I've learned that on the setup day we don't have the musicians perform any of the music that we're planning to record, because I don't want the musicians to have already worn it out when we do the recording itself. Instead, those moments can then be the fresh takes, after I'm not making any more changes with the microphone setup, with the hall, or anything. For the practice day, we use other material that has similar peaks and loud passages while we get everything set right technically. But avoiding the core material during the setup process helps the musicians to not already get bored with the repertoire that we’re all there to capture.
When you say changing the acoustics of the room, do you mean by utilizing the baffling in some of these halls to soften the sound?
Yes. Zipper Hall, for example, has six different curtains that we can move, located up in the ceiling. The Samueli Theater in Costa Mesa [California] has curtains that can be pulled to any location along the sidewalls in the balcony and behind, so that can affect how much natural decay there is. Since we aren't adding reverb in the post-production process – actually, we don't really have a post-production process – we then have to make sure that the natural reverb and the natural decay is what we want in the concert hall. That's what takes a lot of time figuring out. We used to think we could do this in headphones, but I'm not good enough to be able to guess about what I can hear in headphones and know how that will translate to the end listener. I do use headphones during the recording sessions, but not for the soundstage and not for the ambience in the hall. For that, we set up in a separate space, usually a different room – a playback room – which we laboriously adjust with speakers and compensations in the room to make sure that the sound is as close as what we can get, in terms of the soundstage. We then take a thumb drive from the recording in the concert hall, go to the listening room, and play the recording there with the musicians to figure out what changes we want to the sound. What changes we might want to make during that setup day. We only listen back the first day, when dialing in the technical elements.
I feel similarly that recording and listening are two different activities. When musicians stop to listen it can kill the momentum of the creative process itself, and often they become judgmental, self-conscious, or get stuck in their heads.
The listening room also gives us an opportunity to gently re-coach the musicians as to how they're supposed to approach a particular repertoire. One of the things that is often an issue is that we are usually working with musicians who are used to performing in large spaces, like opera houses and 2,000 seat concert halls for orchestral music or chamber music. Their habits are big gestures. They are used to projecting the emotional content of the music so that it will be obvious to someone sitting in the middle of the concert hall or the last row. Part of what we do with the musicians is tell them about this and help them listen to our approach, which is extremely intimate. The microphones are not very far away. I don't usually have to say too much. When musicians are listening with us in the playback room, they can hear the scale of the emotional content of their music, and they recognize the need to adjust it. Usually musicians pick up on it immediately when I convey the intimacy of our recording sessions and I don't have to say anything. But if they're not getting it, I will say something like, “Now, think of this as performing this piece or singing this piece in the shower. Where you hear all the details, whether you like them or not. That is the way we record. We're in the shower!” [laughter] Huge changes in dynamics are less valuable. Instead, we work with the musicians on making adjustments in color. For those emotional changes in the music, we don't have such extremes of very quiet passages, and then thunderous passages. Loud passages don’t pick up as well when microphones are fairly close. It's a collaborative process with the musicians, and at the talent level that we're fortunate enough to work with, the musicians usually get the concept quickly.
It's like the difference between film acting and stage acting.
That's a good description.
What speakers do you have set up in the listening room?
These are small speakers, designed and made by Arian Jansen [of SonoruS]. They don’t have a brand name or serial number. Arian is the one who also designed and built our tape deck, as well as the Holographic Imaging processor that we use for the surround sound. The speakers are single drivers, so there's no crossover. The purpose for these speakers is to provide accuracy of soundstage. These essentially have no phase errors, which is unusual. We don't want any phase errors, because we have to be able to tell exactly where the musicians are going to be in the soundstage. These speakers don't play very loudly, but that's alright. I'm not listening for that. I'm listening for soundstage and timbre. They work beautifully for that. Arian powers them with an early prototype of tube monoblocks [amplifiers] that he designed for his own studio work elsewhere. We use a simple computer and play our test files through JRiver [Media Center app] through a small DAC. It's not that elaborate, but the key is that it's accurate. We never hear the full glory of what the musicians are able to do in the listening room. That's not what that setup is for.
You said that the microphones are not far away. What distance from the musicians are the microphones?
In the case of the Quartet Integra recording, the group were sitting in a semicircle on stage. The microphone, the C24, was probably four feet from them.
And how are the microphones mounted? Ground level, up in the air, above the artists, or even with them?
We've done different things. In a recent session, I had this massive speaker tripod stand and we have a little gizmo that attaches the microphone to it. It was higher than I’d anticipated. For our first tests, we started out quite low, as we had done in the same concert hall early on – 18 years ago. But that placement wasn't right for this particular group. We wound up raising the microphone. It probably ended up four-and-a-half-feet high off of the ground. The AKG comes with a suspension system, and we’ll sometimes put damping material under the microphone stand itself. For the Sasha Cooke vocal album with full orchestra [If You Love for Beauty], that was two U 47 microphones and Sasha did not have her own vocal microphone. We put her in the center of the orchestra; second row. The conductor graciously moved over to the side, which was a first for him. We had two U 47s set up, one on top of another. And they were about 14-feet away from Sasha, but they were also picking up the full orchestra. We needed to pick up 90 musicians! This was [Gustav] Mahler, John Adams, and [George Frideric] Handel. It worked beautifully.
You must have more mics for the surround sound?
We have additional microphones halfway back in the hall, also rather carefully set up, but that's not so specific – those don’t have to be to the inch. They are omnidirectional microphones on each side of the hall, and they each face toward the center of the concert hall so they're at a 90-degree angle to the stage.
What microphones do you use for that?
Earlier in our days, we used a pair of Schoeps M 222 vacuum tube microphones with specially designed omnidirectional capsules. For the last eight or nine years, we have been using small, omnidirectional microphones for the rear channels, built for me by a friend in Sweden. They are extremely quiet and work very well. They also do not have a brand name!
And this is 1/2-inch stereo or 1/4-inch stereo tape? You record at 15 IPS, correct?
We used to record 1/2-inch tape, using two tracks. Bernie Grundman is our vinyl mastering engineer, and he has a 1/2-inch tape machine. We have a special head that we had built in San Francisco to record two tracks on 1/2-inch tape, and Bernie has the same head. But at one point he said, “Bob, why are you using 1/2-inch tape?” I said, “Because it's ‘better.’” He said, “Bob, it isn’t. We've had the opportunity here to make comparisons, and at 15 inches per second, 1/4-inch tape is always better.” I asked, “Why?” He said, “There are several factors involved. One of them is the biasing. Another is head alignment on a 1/4-inch tape head. With 1/4-inch, the alignment can be absolutely perfect. With a 1/2-inch deck, it's harder to get it exactly right. There can be slight cancellations and phase errors.” He pointed out to me that the slightly richer, thicker sound of 1/2-inch tape does not compensate for the lack of precision in the tape electronics. And, as he pointed out and demonstrated for me, 1/4-inch tape has a better ability to capture the air in the concert hall. He didn't have a theory for why that is, but he said it is true and demonstrated it. Since then, we’ve exclusively used 1/4-inch tape.
And you don't do any editing after you’ve recorded a session?
This is the other thing that usually chases musicians away if they’re feeling apprehensive. We feel that it's important for the musicians to arrive extremely well prepared. In the 78 rpm vinyl era, musicians who came to make recordings were so well-rehearsed. They had to be, because they only had one take. If they made a mistake, then the lacquer had to be thrown out and the whole process started again. Those errors were incredibly costly. This is one of the reasons why the performances from the 78 era tend to be so fine, so well-rehearsed, so in tune, and so exciting, because the musicians had really prepared for the recording sessions. Those were very different days than a rock group today that will book three and a half months in a recording studio without the vaguest idea of what they're planning to do, and then work together with the producer to write, create, and record the music as they go. I prefer working with musicians who are extremely well-rehearsed – just like the engineers in the 78 era had the privilege of working with. The musicians we work with know in advance that we will be recording full movements in one take. There are lots of record labels around who will work with musicians differently and edit the music to death. There's a famous recording – that I will choose not to name – and one of the movements is all pizzicato. Rumor has it that there are 115 edits just in that short movement. Yes, the technical result is perfect, but that's not the way we do it. It's also not the way I want to spend my life! But, in addition, there’s a musical reason. I want the musical emotional line to be what the musicians created. If I – as a producer or engineer – am editing the piece after the fact, my own musical tastes are going to inform those decisions. In a way, it will be me playing the string quartet, as well as the four musicians. That's not what I want. I want it to be their musical trajectory, their musical arc, and to the best of their abilities. Yes, our albums have mistakes on them, and that's okay.
Mistakes are often where the humanity enters.
