Brian and I drove to Philly on Sunday, December 3, 2017, to attend the first of two world premier performances and recording sessions for Symphony for a Broken Orchestra by composer David Lang, a luminary in the "new music" genre of classical music. I preceded this with studio interviews in Brooklyn with producer/engineers Arun Pandian and Devin Greenwood, the folks charged with recording and mixing stereo, surround, and 3D audio releases of this performance. Brian and I revisited Arun and Devin later for a photo shoot and follow up discussion about the mixing process. Symphony for a Broken Orchestra was commissioned by Temple Contemporary, Tyler School of Art at Temple University. Over 1000 physically broken classical and band instruments were collected from the attics and basements of various (many underserved) public high schools throughout Philadelphia, and a website was set up to raise awareness and solicit sponsors. Private and corporate sponsorship of individual instruments will pay for repair and reintegration of the instruments into the Philadelphia public school system's music programs. The Philadelphia Public School System's 2007 instrumental music budget was $1.3 million, but that same budget in 2015 had been reduced to $50,000. Fundraising efforts will expand with future commissions and performances in other major US cities with similar public school music budget cuts.
Before the Event
How did the two of you meet?
Arun Pandian: We met because a good friend of mine and Devin's is a songwriter named McGowan. [To Devin] How did you meet McGowan?
Devin Greenwood: Basically through John Legend's old Philly crew. He was writing songs with an Australian guy, who goes by the name of Old Man River, and he wanted both Arun and I to play on the project. We ended up forming a band, but this guy lives in Australia – halfway around the world! We were meeting and commiserating a lot, and eventually we became good friends.
AP: I'm from Texas. I originally went to school for finance. I landed the Miseducation tour with Lauryn Hill. I did that for a few years, and I got to play on both the record and the tour. Through that, I played with Santana, Mary J. Blige, and Nas. After that, I was in New York for a while. I did some off-Broadway performing with Savion Glover and Jared "Chocolate" Crawford, who I ended up producing in Philly for a record label, with Chris Schwartz of Ruffhouse Records, which The Fugees came out on. I was producing Chocolate on that label in Philly and then came back here to New York. After 9/11, I was at a crossroads. I asked the question, "Do I want to be a guitar player anymore, to tour with people and be on the road?" I decided that I didn't, that I'd rather control my own career. I started getting into sound, audio production, and engineering. I taught myself. I bought a piece of gear, learned how to use it, sold it, and got another piece of gear. I mentored with Henry Hirsch [Tape Op #56] at Waterfront Studios upstate, working with him as a producer. I also learned a lot by doing my own music. Since then, I've worked with artists like Norah Jones, Mumford & Sons, and Gabriel Garzón-Montano.
DG: Mine's a similar arc. I played classical piano as a kid, but ended up studying film in college, which is a real passion of mine. After film school though, I got a chance to tour with [vocalist] Donna Godchaux from The Grateful Dead. I did that for a couple of years when I was 22. Later on, I played with Lo Faber from God Street Wine, became Amos Lee's first music director, and did several years internationally with Melody Gardot, singing and playing [Hammond] B3. I was songwriting inbetween all that – Steve Forbert actually covered one of my songs – and I put out a record back in 2006 called Eye in the Cloud. Like so many singer-songwriters, I felt like I was on track to get signed, but at the same time I was starting to get into studio craft as well. Just outside of Philly there was a place where Shelly Yakus [Tape Op #31] used to work, called Scullville Studios. Shelly did all of the early Tom Petty, John Lennon, The Band, Dire Straits, U2, and so many famous songs that you know, like [Van Morrison's] "Moondance." He's a really great dude. I made my record there with Shelly and Rachel Alina. I learned about studio engineering and recording to tape, and I was seduced by this whole world. When I started producing local artists, and one of them got signed, people wanted me to make records for them. All of a sudden I was making money doing that. Around that same time, Norah Jones had me play on her third studio record [Not Too Late] and we were recording Amos Lee's first album as well at Sear Sound [#41] and the old Magic Shop [#66]. Before long I'd moved to New York and opened my own studio, The Honey Jar. Since then I've worked with people like Sufjan Stevens [#70], who was in the neighborhood, and Steve Reich [#15] – which is how I got into the new music world.
AP: I'm currently teaching audio engineering at Long Island University. A good friend of mine, from my Lauryn Hill days, is head of the recording technology program over there, and he approached me about teaching. It's right down the street from where we are now. My first inclination was to recoil back – I'm actually doing this for real. I don't want to go teach in a classroom. He said, "Oh, come on. You're really into this and you've got a studio. It'd be great for you to come talk to these kids." I did it, and I was surprised by how much I actually enjoyed it. After teaching at LIU, I started to think about doing a Master's in Music Technology or something like that. I ended up at NYU, studying in the Tonmeister program, where I got to learn about classical recording techniques, as well as 3D audio. Devin knew that I was doing this, so he told me about the broken instruments project, and he suggested we team up and record it. They weren't planning to record the concert. We partnered with Sennheiser. Speaking of, we'd like to thank David Missall and Axel Brisard for donating the microphones to record the event. We also have a GoPro Fusion camera that I purchased. I'm going to shoot 360 video of the performances as well, for the sound installation.
AP: There will be a vinyl 2-track stereo release of the 3D audio recording.
DG: We'll mix the 2-track record either here at the Honey Jar, or in Arun's studio, Signal Corps Recording. There will also be a sample bank, with avant-garde electronic music being made from those same sounds.
AP: And there will be the 3D sound installation, which I'll be mixing at NYU. They have the lab and the speaker setups that we need.
Let's talk a bit about how each of you transitioned from being touring musicians to producer/engineers.
AP: I'd been talking with Henry Hirsch; he knows an enormous amount about gear. He's been through it all, especially with Lenny Kravitz and the budgets they've had to buy and sell gear. Just talking to them about their vision for what type of gear to buy – the ideal type of gear that was contrary to what you see in most studios. Generally it boils down to one rule: the fewer electronics there are in the signal path, the better the audio. A lot of the older gear fits that mold. The more you put in the [signal chain], the more it stops sounding like what you were trying to capture in the first place. So, Henry gave me suggestions on what gear to buy. He said, "If you ever see a WSW, the Austrian branch of Siemens, you might want to grab it, because a lot of people don't know about it." That started a search. I had originally opened a studio where all I had were mic pres going straight into a Pro Tools system. I thought, "Oh, I don't need a console." Then I bought a summing mixer and there was a pretty drastic change to me in the depth of field. Then I figured, "Oh, great. Now I need to get a console." I remembered from my experience of going into studios as a producer before I knew what I was doing, the engineer would just spread out the music on the console. For the type of the music I loved, I needed to experience that sound.
DG: Once you experience it, it's hard to go back. It's very alluring. I learned about breaking out the tracks in a studio where we had this [Pro Tools] Mix Plus system. I was in there for the first time working on my own, and for whatever reason I said, "Let me break out the bass to its own channel." I'd been mixing and listening on two channels. I could hear it better and clearer. That was a unique, seminal experience, where it was so crystal clear what a specific piece of gear does.
AP: Henry suggested I find a console like this one. I saw it on eBay for $3,000. The console was in parts. The guy hadn't put it together; he just shipped it to me. I was talking to techs, and they were quoting me these enormous prices to put it together. No one had the schematics, and nobody had any of the parts or the frame. I was on Gearslutz, and I found this guy named Damir Rogina from Croatia. He goes by the name Roginator – he's head of the MCI forum. He also knows everything there is to know about Siemens and WSW. He has all the schematics and all the extra parts. He even had an original frame. I flew him and his partner down, and they put the whole thing together, refurbished it, recapped it, and did everything in like two weeks.
What year is this console?
AP: Early '60s. Slowly, I realized how good it sounded and how well it's built. It's built beyond what anybody would normally need. It's basically [built for] Austrian government money for broadcast and military purposes.
It looks like a broadcast console.
AP: Yeah. It's a pretty heavy-duty build, and it's super simple. There's not a lot going on. It's just EQ, mic/line channels, five buses, three auxes, and that's about it. Once I had the console, I kept looking for other Siemens and WSW gear, and slowly I collected and refurbished some. They're built to a different rack standard from what we use in the US. I ended up finding Siemens racks, which I had customized by my friend, Steve Masucci [Tape Op #124]. There's another studio in Austria called Casino Baumgarten. They have a WSW tube console, which is where my tube pres came from. I slowly started to buy the gear and refine it down. It comes out of my desire for the studio to have a specific sound and direction. When people come here, they come here for that purpose.
DG: Arun turned me on to the idea of sticking to a brand and the way to use artistic limits to make a piece of art have more of a signature, by doing that with a studio. Instead of having all the cool gear, just curate it to one brand.
AP: It comes from the need to have your own sound. When we were artists, we were doing that. Why not do it on the engineering side of things? It doesn't make sense to cater to everybody's needs all the time. I think we have too many choices now with digital. Somebody has to make a decision somewhere. Might as well start with the gear.
How would you describe the sound that you've achieved?
AP: It goes along with what I was saying earlier – the simpler, the better. Not having too many choices. The console doesn't have the ability to re-route itself in 100 different ways. There's no recall; no fancy features. This keeps everything very simple. A lot of the gear back then wasn't made to necessarily have a vibe. It was made to translate in a euphoric way, to be very clean, and give you what you put into it as much as possible. That's kind of how I try to do things, to have some clarity and then have the depth.
DG: I was looking for the same thing, just a Class A discrete console. I actually had the same guy that Arun hired, Damir, come over and rebuild mine. I used to have the very first MCI 500 series – I had the prototype, and nobody had schematics. He rebuilt that, but it ended up being a bit of a nightmare. I wanted to downsize to something more manageable, so this is a Studer 189 MkII. It's a mobile recording console, originally used to record classical music from a truck. It's fully modular, so even while it's turned on and running, you can open it up and swap channels out.
DG: Here's the fader cassette [clicks out part of a channel strip while the console is on and playing background music in the control room]. You can pop the channel out, just like this.
And it doesn't affect whatever is running at the time?
DG: It doesn't affect what's going on. And it's double-sided, so there's actually [components] on each side of the board. They packed everything into a really small package in a way that only the Swiss know. The reverb input modules were all built for plate reverbs, so you've got send, as well as a low end rolloff for your send, and then you've got an EQ on your return. These change the actual distance on the plate. There's a proprietary connection on the back that lets you plug into the motor on the plate, so that you can change the reverb time right here.
Do you have that hooked up to your EMT plate?
DG: Not yet, but I will eventually. Obviously I'm not going to have four plates. The others will become stereo returns for something else. The other cool thing about this is that it interfaces with a Studer tape machine. I bought the tape machine – this 8-track deck. I can control the transport, arm the channels; and here are my eight returns.
What year is this console?
DG: 1974. Even the patchbay – everything is right here on the back. I'm a minimalist with the amount of gear that I want to use. I have that [Teletronix] LA-2A, which is a Rev 1 from the '60s, the very first edition of that model. There are four compressors in the console, as well as a plate reverb. That's my setup. My own musical style is cinematic and spacious, but also minimalist.
Let's talk about the Broken Orchestra project. I'm not that familiar with 3D sound myself.
AP: We've all experienced it, right? It's just spatial.
DG: In real life, it's 3D. If you're in the woods and hear the birds, it's 3D.
AP: On the recording/production side, we've mainly experienced sound in terms of left/right and front/back. Now we're adding the height channel to that.
By height channel, you mean up and down?
AP: Yeah. It's something that's always captured in the mics. An omni mic has sound that's captured in all directions, but it hasn't been the focus of recording production or the way speakers have been set up. The 5.1 is just the main layer around your ears, in the horizontal plane.
DG: In movie theaters, you've experienced surround sound, but someday it'll be in 3D.
Center channels lining the ceiling.
AP: Which they're already doing with Dolby Atmos.
The setup for the performance is in the round, but reversed. The audience is in the center, and the orchestra is the equivalent of stadium seating where the crowd would usually sit for theater in the round.
AP: Yes – the conductor is going to be dead center.
DG: [The conductor], Jayce Ogren, is a protege of the composer, David Lang. David will be there, but he's not conducting. Jayce is also a player, in that he's signaling how long the movements are going to last, how many hits there are going to be. He's bringing groups in and out the entire time.
Funding will be used to restore whichever instruments can be restored, so that they can go back into the Philadelphia Public School system?
DG: My friend, Danielle [Birrittella, a singer known as DIA], coined a phrase yesterday – she calls it innovative philanthropy. It's about being creative at something that America has been pretty good at, which is bringing private money into the social system to do things that we care about. It's about bringing creativity into that process. I'd say that this kind of music in this scene relates to classical in the same way that Kandinsky or Jackson Pollock relates to Renaissance painting. It's [a postmodern] version of that music. It's a very different experience than hearing a Vivaldi piece. It's kind of psychedelic, actually.
AP: It's very 21st century, which is why 3D plays into this. It's sonic experimentation, using classical and jazz band instruments.
Some of the fundraising and promotional videos for the event show guys using mallets on broken cellos.
DG: When you take the strings off a broken cello, it becomes something else...
AP: A percussion instrument.
DG: Those guys are from Found Sound Nation, which is an artist collective I'm a part of. It does really great work all over the globe. FSN used to be the social arm of Bang On A Can, which was formed by David Lang, Julia Wolfe, and Michael Gordon to promote this kind of music.
Arun was telling me that there are going to be multimedia installations at performance spaces, or MOMA-type museums. This really is innovative philanthropy.
AP: That's why we got so excited about the project. It's a way to create art and help people at the same time.
DG: There will also be members of the Philadelphia Orchestra playing. The performances will be at the 23rd Street Armory in Philadelphia. The idea is you play the note where it would sound normally on the instrument, and if it's wrong, that becomes part of the piece. That's in the tradition of John Cage, who brought indeterminacy into music composition with rolling dice or using the I Ching to decide where notes go. That was a Modernist innovation, and it is very much a part of this music tradition which, again, is a very elitist tradition.
AP: There will be ten ensembles of musicians around the perimeter of the venue. Each ensemble is going to be a mixture of broken instruments. It won't just be violins or horns. From what I've been hearing in rehearsals, the piece jumps around between the different ensembles, doing all kinds of things. It's playing with the physical space.
How many people will be in the audience?
AP: 400 to 600. There will be 400 performers in the orchestra.
So the performers will be spread out more than the listeners, with physical space between the ensembles around the perimeter of the venue?
AP: Each ensemble will have its own group leader, and sometimes there will be call and response between the group leader and the ensemble. The group leader will be instructed to do something and then the ensemble will repeat it. Each group leader will play an instrument and will be directed by the master conductor.
Will the group leaders be children or professional musicians?
AP: Most likely professional musicians and teachers.
DG: The student players are in high school. The oldest adult players will be in their 80s and 90s.
Let's talk about the mic setup that you will be using to record the concert.
AP: Basically, [3D audio recording techniques] start with general stereo setups, like Blumlein, NOS, ORTF, or an X/Y spaced pair. People have taken those systems and turned them on their sides. What we're going to be using for this setup is an M/S setup that's been turned on its side, so it's now an M/Z setup.
DG: It's mid-side recording, turned perpendicularly.
AP: What you're getting when you're concerned with stereo playback is an inter-aural difference. You're concerned with the time differences between the left speaker and the right speaker, as well as the left ear and the right ear. When you're dealing with the up and down Z plane, it's mostly a spectral difference. It's the difference between us being closer to the floor, and the way sounds hit the floor and come back to us – and then certain sounds being above us. And not only the way they hit the pinna of our ears and bounce off our shoulders into our ears, but the spectral difference of the floor and the distance to the ceiling. That's what we're trying to capture with that Z-plane.
There is going to be a high ceiling at the armory. The rafters are going to create a reverb time that wouldn't happen in a studio or another smaller place.
AP: Exactly. We're just trying to capture that Z-plane and then translate it into the reproduction system.
DG: So, the omnis and cardioids are your left and right, which is all surround, and then you have your figure-8s facing up and down for the vertical plane.
AP: For us, it'll be cardioids on each of the ten ensembles, because we want to reject the audience [sound] from the back. Then it's going to be a figure-of-8 setup behind that, which will be facing up and down. We took the side and flipped it up.
Who determines what the front of the mid-side is?
AP: The mid. The mid is the front.
But if it's a 3D...
DG: There is no front. It depends on the orientation of the viewer once they're in the playback system for the installation.
Every person who sees the live performance will have a different listening experience, a totally different blend of the ten ensembles, with a slightly different experience of the reverb of the room.
AP: Absolutely. It's going to be our challenge then to break it down to stereo in some way that works for the vinyl release. The 3D installation will almost be easy, because we'll just set up speakers mimicking the mic pairs on each instrument. We're using a coincident 3D setup. People have also experimented with spaced setups. The setup I'm using was pioneered by Paul Geluso. There's another setup by David Bowles, called a Bowles array, and that's basically an NOS/ORTF setup turned on its side. Then Gregor Zielinsky of Sennheiser does a spaced-pair setup turned on its side. I'm giving you a breakdown essentially of where these ideas are coming from. When you're doing 3D audio recording, the thing you're challenged with is the direct sound you're getting from in front of you, as well as however you're getting the height channel. It's the correlation between those two channels, how they interfere with each other, and trying to reject the front signal and get as much reflection on the top and bottom as you can without smearing the two.
Does the playback format compensate for these variables and make some kind of "mix" decision? Did someone create a software crossover?
AP: No. Right now those three guys and their setups are the ways they've come up with to best deal with these problems. It's something that people are still researching and experimenting with. We are recording 3D in a huge, immersive setup. I don't know if there are a lot of those that have been done so far.
After the Event
Let's discuss the recordings that will be released of the concert. Will decisions be made to take a section from the first performance and a section from the second performance?
AP: I think we'll definitely consult with David Lang. He'll probably be there when we do the final mix. But it's not even for certain that we will comp between the two – that's just what we've been doing so far. One of the things that got him excited is the talk of doing an immersive installation experience.
There will be multiple mixes...
AP: There will definitely be a stereo mix. There will be some sort of surround mix, depending on how we want to do it. It's either going to be 9.1 or 7.1. And then there will be the installation, which is a whole separate thing.
The installation will be the 3D audio?
Because of the limitations of most playback systems, even high end audiophile systems, I guess that most people will hear the final release as a surround mix?
AP: The surround mix will be a 3D audio surround mix. It'll be like an Auro-3D mix. It'll have a 5.1 main layer, and then you will have four height channels. We'll probably put it in some format like that. There are multiple formats that you can use that are equivalent to that. A lot of hi-fi people actually have the capability of splitting it out in such a way. The new Kraftwerk material was done in 9.1. I believe it was 3D, so it wasn't just a main layer. The format is usually on Blu-ray and there are two main proprietary systems for processing: Dolby Atmos and Auro 3D.
So, the decoding process will be where the choice of playback channels happens, per the speaker configuration of each listener's music system?
AP: Exactly. That's the point. If you only have 5.1, the mix will be decoded as 5.1.
A portion of the mix is happening at NYU's lab, where you're doing the Tonmeister program.
AP: They have a 20-speaker setup in the research lab, and a 10.2 set-up in Dolan, which is their main studio. It can get confusing, the way these speaker configurations are described. Like if you had a 9.1 system, it could be described as a 5.1 main layer with four up top, so it's 5.1.4. But some people call that 9.1
9.1 sounds cooler.
AP: But 9.1 could technically be a main layer with no 3D at all [all of the audio channels happening in the horizontal plane].
Now that the performance has taken place, what did you learn from the experience?
AP: On the recording side, it was shocking how smoothly things went, logistically. I'd say that the recording techniques we used were a good balance of not only the budget, but just logistically, setting everything up in the time that we had. It could have gotten really complicated with that many performers.
As an informed outsider, I thought it went about as smoothly as it could have possibly gone.
AP: It did. And after the fact, we haven't encountered anything that we regret or wish we'd done differently. The thing about doing it in the future is that if it's a new composer and setup, it's going to be totally different, so we'd approach it completely differently. This time, we curated it for the way that this specific event was set up. We were out of the way of the musicians and most of the audience. I think we were pretty inconspicuous.
DG: We had the overall technique planned out, but then we ended up putting spot mics here and there. I was walking around, looking for people with a combination of apparent competence, as well as a somewhat functional instrument, with a particular emphasis on low end. There wasn't a lot of low end in the recording, so we close mic'd some basses and bassoons. That came out really well and brought some bass into it. Those kinds of decisions were spontaneous.
AP: There were also dress rehearsals a couple of days before, so we had the benefit of modifying mic heights and placements.
Since it went so well, has there been any interest yet in doing it again someplace else?
AP: Yes! We're forming a non-profit organization to commission more compositions for broken school instruments in different cities. We hope to work with a whole range of New Music composers to shine a light on the current funding crisis for public school music departments, and to establish funds to keep working instruments in those schools.