This is a more readable description of the set-up in that room:
The room dimensions were:
W = 4:5 m, L = 6 m and H = 2:2 m.
The walls and the ceiling were almost completely absorbent and the only major reflection was due to the floor.
It's by ...Wes Phillips
This week I took a little trip to AT&T's research facility in Florham Park, NJ where my friend Jim Johnston is technology leader of the Speech and Image Processing Center. Johnston (J.J. to those who know him) has been working on Perceptual Soundfield Reconstruction, an all-out attempt to develop a multichannel recording/playback system that preserves a sufficiency of spatial cues to realistically immerse the listener in a simulacrum of the original recording venue.
Forget everything you've heard about multichannel. Compared to PSR, it's all a joke -- from Quad to all the way up to five-channel SACD. Oh, Ambisonics works well enough, if you're willing to put up with a listening room that looks like a tornado hit a music store and a single-person-head-in-a-vice sweet spot. But PSR does all that with a minimal amount of equipment -- and does it better, to boot.
True, there's DVD-A and SACD multichannel, but very few producers seem to "get" it -- Sony's still trying to ambush us with trombone players hiding in the back of the room. Besides, even discrete five-channel surround tends to either bunch up in the center channel or else collapse completely when you turn your head around to look at the rear loudspeakers.
That's the beauty of PSR -- it almost can't be used for the devil's work of secreting brass-players willy-nilly around the soundstage, and it doesn't even waver a jot, no matter how much you move around within its focused area.
Dico, dico, dico
The Florham Park research facility is a quiet place with Stickley Mission-style furniture arranged in conversational groupings all over the joint. "Wow," I remarked to John Atkinson, whom I'd accompanied. "You can't even put your butt down on less than $3k worth of seating in this joint. Imagine what they put into the labs."
When we reached Johnston's lab, we discovered that the actual audio equipment he was using was very firmly in the affordable "prosumer" camp. The amp rack sported a home-built ten-channel balanced passive attenuator (J.J.: "None of the digital units we tried were quiet enough, so I just built something analog") and five Hafler P7000 power amplifiers.
Jim then swung open the heavy, insulated door and ushered us into the extremely quiet, double-walled listening room, where five Snell C5As were arranged in an implied circle around a group of comfy chairs. John Atkinson and I grabbed seats, each slightly off of where the sweet spot would be in a conventional system.
Then Johnston grabbed a keyboard and keyed in a command -- and we suddenly were totally immersed in a fairly large church acoustic, listening to organ music. It wasn't the best organist I've ever heard -- or even the best organ -- but it was scarily real. The sound came from the front, but we were so solidly within a huge room that we could hear certain pipes reflecting off the balcony behind us.
It didn't matter how much we squirmed or moved our heads, the sound was spread evenly across the front of the room, and the spatial cues were coming not from behind us, but from all around us. And, it didn't seem to matter where we were sitting, the soundstage remain centered and huge, even well away from the sweet spot.
And it went on. Johnston played an amateur early music ensemble of the "if we were too good at this, we wouldn't be authentic, now, would we?" variety. As a member of many seriously-flawed amateur orchestras, I was immediately at home. God, it sounded real!
Then Johnston cued up one of the recordings he had made when we were all out in Northfield, MN recording Cantus a few months ago and I gasped. Not only was this the best sounding recording he'd played us, it had details out the wazoo. In two-channel, it sounded pretty good -- the singers were spread between the speakers and I could clearly hear the hall. But in multichannel mode, I was in the acoustic and I could hear how the singers were arrayed in a semicircle across the stage. I could hear the acoustic "sail" above the stage bouncing the sound back down at us. There was a part of my reptile brain that actually wanted to escape. This was heavy mojo.
Actus id verberat
Then Johnson laid a bomb on us. "Listen to this," he said and cued up an excerpt of a pretty good semi-pro orchestra playing Dvorak. I was impressed, noting all the spatial details and U-R-There solidity, when Johnston announced, "This is playing back at 320kbps"
I didn't immediately grasp the implications. "Huh?"
"Let me put it this way, how much would you pay for season tickets to the New York Philharmonic that delivered this sound to your living room every week?" O my god -- I'd cut off my toes for that. Then I got it -- at 320kbps, a broadband connection could deliver five channels of really satisfying sound anywhere. No more hour-long commute to Lincoln Center, except for when I really wanted to make the trip.
We listened to 320kbps and switched back to full-resolution a few times. The difference is noticeable, but 320 is damn fine. The acoustic was solid and unwavering and the soundstaging was a good as any stereo I've ever heard. Any!
Nil volupti, sine lucre
I asked Johnston about the process. Since I'd been with him in Minnesota, I knew his recording rig used seven microphones in a globe-like microphone array about a foot across (informally dibbed "the Deathstar.") Five microphones were aimed in a circular array, spaced 72 degrees apart. In addition, one microphone pointed straight up and one pointed straight down. In Northfield, Johnston had confided that the vertical information was crucial -- and was the final part of the puzzle to fall into place.
"J.J., you record in seven channels and play back in five. That implies that some level of computation is involved in the process."
"Not as much as you'd think. All that's involved is an addition, a multiplication and five additions -- and that's it."
"When will it be ready?" I asked.
It's ready now. We can license the technology for some amount up front and our nickel a disc. At that price, you just get the use of the technology, of course. You won't get me with it, but the beauty of it is how simple the setup is." (Anyone interested in arranging licensing should contact John Rudder at firstname.lastname@example.org.)
"You mean I could license Perceptual Soundfield Reconstruction today?"
Please, somebody -- anybody -- take him up on it! You have only your lousy soundstage to lose!
And the home concert subscription system?
"That's going to be tricky. It's going to upset musicians and unions and getting that sorted out will be time-consuming, but we could move on it pretty fast, if we had anyone interested in developing the idea with us."
Any idea what it would cost?
"I intentionally stay out of all that 'real world' stuff -- I'm happy off in my 'blue sky' territory."
So there you have it. We finally have a sane, logical, intensely musically satisfying multichannel music reproduction system. Now all we need is to start taking advantage of it. I sat on the fence over SACD and DVD-Audio until only recently, but I have no questions about how much I want PSR. I want it and I want it now.
As for the home delivery option, think about it for a second. I know I'd subscribe to several programs if I could have good seats in my own home. I could root for the New York Phil, my home team, and drop in on favorites, such as Vienna and Philadelphia, when they came to visit my favorite venues or at their home concert halls. I could even -- be still my heart -- hear Dylan again in a small club before I die.
Come on, let's make this happen. Someone has finally answered my audiophile prayers with a solution aimed straight at music lovers. I know it's not like the bad old days of The Phone Company's schemes of world domination, but AT&T's still big enough to have some clout, isn't it? Please tell me it's going to happen. It has to happen, doesn't it?