Superstar recording and mastering engineers gathered with the press in New York to discuss and play examples of their high-resolution audio efforts.
In conjunction with CE Week this year, the Consumer Electronics Association (CEA), the Digital Entertainment Group (DEG), and the Producers and Engineers Wing of the National Academy of Recording Arts and Sciences (NARAS) hosted what was dubbed The High-Resolution Audio Listening Experience at Jungle City Studios in New York City. Located on the 11th floor, the studio offers a commanding view of the Chelsea neighborhood outside and a state-of-the-art recording studio inside, where JBL had installed a pair of its incredible M2 Master Reference Monitors, which reach from 20 Hz to 40 kHz.
Fine nibbles and drinks were on hand at the HRA listening event at Jungle City Studios. (Photo by Mark Henninger)
The event coincided with an announcement that the CEA, DEG, The Recording Academy, Sony Music Entertainment, Universal Music Group, and Warner Music Group had developed a formal definition of high-resolution audio (HRA)—"lossless audio that is capable of reproducing the full range of sound from recordings that have been mastered from better than CD quality music sources." In addition, four "descriptors" were introduced to specify the provenance of HRA or so-called "master-quality" recordings:
From a PCM master source 48 kHz/20-bit or higher (typically 96/24 or 192/24 content)
From an analog master source
From a CD master source (44.1 kHz/16-bit content)
From a DSD/DSF master source (typically 2.8 or 5.6 MHz content)
I see at least one discrepancy in this announcement. The formal definition of high-resolution audio includes the phrase, "recordings that have been mastered from better than CD quality music sources," yet one of the descriptors for a presumably HRA master-quality recording is "from a CD master source (44.1 kHz/16-bit content)." In my view, HRA should conform to the first definition—that is, it should be better than CD quality. Also, analog master sources rarely exceed CD specs in terms of frequency and dynamic range, so I don't know why that is considered high-resolution audio.
Then I realized that the key phrase in the formal definition is "lossless audio." Practically speaking, I believe the formal definition and master-quality descriptors are best suited to distinguish these recordings from lossy MP3s, which they all beat handily in terms of sound quality, at least when the MP3 is at a low bitrate. Now that processors are powerful enough, memory is cheap enough, and online bandwidth is plentiful enough to easily support lossless audio downloads, they can compete with MP3s in the convenience department and deliver a superior music experience at the same time.
L-R: David Chesky, Mark Waldrep, Bob Ludwig, Frank Filipetti, Chuck Ainley. (Photo by Harris Fogel)
During the evening, several world-class recording and mastering engineers talked about and demonstrated their work with high-res audio. First up was mastering engineer Kevin Reeves, who played several clips from recent remastering projects, including "Statesboro Blues" by the Allman Brothers Band from a live show recorded in 1971 (16-track master), "It's All Right With Me" from the 1959 album Oscar Peterson Plays Cole Porter, "Bodhisattva" from Steely Dan's 1973 album Countdown to Ecstasy, "Strutter" from Kiss' eponymous 1974 album, and "Livin' For the City" from Stevie Wonder's 1973 album Innervisions. In most cases, the original analog masters were converted to DSD128 (5.6 MHz) and then to 192 kHz/32-bit for mixing or mastering. The Kiss and Stevie Wonder tracks were digitized directly to 192/32, and "Bodhisattva" was transferred to DSD128 and converted to 192/32 only for playback at this event.
All these clips sounded great—as good as I've ever heard them—but that was on a $25,000 pair of speakers (including the amps) in a professional recording studio. And I seriously doubt any of the original master tapes had any audio information beyond what CD can reproduce. Finally, why digitize with DSD128 and then convert to 192/32? Why not just convert the analog to 192/32? Seems like an unnecessary extra step to me.
Next up was Chuck Ainley, a multiple Grammy award-winning recording engineer based in Nashville who works with a lot of country artists. He played clips from recent recordings he did with Pistol Annies (recorded analog, mixed in 96 kHz/24-bit PCM), Mark Knoffler (recorded analog, digitized at 192/24, mixed at 96/24), Lee Ann Womack (recorded and mixed digitally at 96/24), and Miranda Lambert (recorded and mixed digitally at 96/24). I was quite put off by the harsh sizzle in all of these recordings, which I assume was an artistic choice (and I use that term loosely here).
I got to sit in the sweet spot a couple of times—hey, no one else was brave enough to sit in that chair, so I did! (Photo by Mark Henninger)
Legendary mastering engineer Bob Ludwig talked about his use of a hardware/software system called Plangent Processes
, which eliminates speed variations in analog tape. According to the website, "it begins with an ultra-wideband low-distortion custom reproduce head and subsequent associated hand-wired preamp, followed by proprietary DSP that provides total speed stabilization and wow and flutter correction." Bob is quoted on that site saying, "...the Plangent Process gave all of this music a new width and dimension that make it connect with me emotionally in a more profound way than I had ever experienced before."
He started with Bruce Springsteen's "Born to Run," playing the 1984 CD master followed by a 96/24 remaster, which did sound more stable and open after the Plangent Processes. He also played "Thunder Road" from the same album, before and after it was remastered at 192/24 with the Plangent Processes. There's an audible burble in the speed of the analog tape machine at the very beginning of the track, which most people probably wouldn't notice, but it's definitely there as we heard when he played it several times. After the Plangent Processes, it was gone.
Next, he played a live recording of "Factory" from Band of Horses—first a 256 Kbps MP3 followed by the DSD master, which definitely had a larger sound, especially the sense of the room in which they were playing. After that, it was "Blue Moon" from Beck's new album Morning Phase, first at CD specs and then at 96/24, which had slightly cleaner, clearer reverb. Finally, he played a clip from Jack White's album Lazaretto, which had been recorded entirely in the analog domain (one or two sync'd 2-inch 8-tracks mixed on an analog console to 1-inch stereo and then to LP); we listened to a 192/24 transfer, which sounded quite good, though I wasn't blown away by it.
I spoke with Bob after his presentation and asked him if there is anything beyond CD specs to capture from old analog tapes. "Yes," he said, "depending on how the tape was stored, how the heads were aligned, and the mixing console. There can definitely be frequencies beyond 20 kHz, but probably not dynamic range beyond 90 dB." However, he acknowledged that most of the difference between a CD and a high-res version of the same content is in the remastering, not in content beyond CD specs. He also pointed out that CDs can sound very good, especially if the original recordings are higher resolution and downconverted to 16/44.1 at the end.
Recording engineer Frank Filipetti is a six-time Grammy winner who makes no bones about hating analog recording. "Analog has a massive list of distortions that must be corrected; the medium is not flat," he said during his presentation. "We've defined it as a pleasing sound, and people are used to it, but the best digital is better than the best analog."
Frank had just finished recording Frank Zappa's 200 Motels with the Los Angeles Philharmonic and conductor Esa-Pekka Salonen at Walt Disney Hall to commemorate the venue's tenth anniversary, and we were the first people outside the production crew to hear it. He prefers using a sample rate of 88.2 or 96 kHz, but this recording was done at 48 kHz/32 bits because of the massive number of simultaneous tracks—180, with a separate microphone on each musician—and he didn't want the digital-audio workstation (DAW) to choke. The sound was very clean with each instrument clearly delineated.
Next up was David Chesky of Chesky Records and HDTracks, an online source for high-res audio recordings. David is a big proponent of binaural recording, in which small microphones are placed in the ears of a dummy head. Normally, the only way to experience the benefits of binaural recordings is to listen on headphones, but he is working with physicist Edgar Choureiri at Princeton University to allow the use of directional speakers, DSP, and personalized filter profiles derived from measurements taken in each listener's ears, which brings the full surround effect of binaural recordings out of headphones and into the room.
David Chesky talks about binaural recording. (Photo by Mark Henninger)
He couldn't demonstrate this effect at Jungle City Studios—it didn't have the right equipment—but he did play an excerpt from a children's ballet he composed called The Zephyrtine, which had been recorded at 192/24 with a Neumann KU 100 binaural-head microphone suspended over the conductor. He applied no dynamic compression, and one benefit of high-res audio was immediately apparent in the super-wide dynamic range, from extremely soft to exceedingly loud with no distortion I could hear in the few seconds I had my fingers out of my ears. Obviously, the JBL M2s are up to the challenge of HRA.
Finally, Mark Waldrep of AIX Records took the microphone and talked about his philosophy of high-res audio—recording ensembles playing together in a real concert space, using mostly stereo pairs of mics, capturing/mixing/mastering everything at 96/24 with no dynamic-range compression or artificial reverb, and providing 2-channel and multichannel mixes from the audience and stage perspectives. He also extolled the virtues of the HTC M8 smartphone—as did David Chesky—which can reproduce HRA up to 192/24.
Mark Waldrep ended the evening with some top-notch recordings. (Photo by Mark Henninger)
Mark had prepared a medley of clips from his recordings, including "Mosaic" from Laurence Juber, "Lone Star" from Carl Verheyen, "Let Them In" from John Gorka, "Somewhere Somebody" from Jennifer Warnes, "Lowlands" from Hanna-McEuen, "On the Street Where You Live" from Steve March Torme, "Mujaka" from The Latin Jazz Trio, Mozart's Clarinet Quintet K. 581 from The Old City String Quartet, and "Rainbow Connection" from Willie Nelson and Paul Williams. All the clips sounded exceptional, with clean and present vocals, clearly delineated and detailed instruments, and unfettered dynamics. Everyone in the room was very impressed, as they should be—this is the epitome of high-resolution recording in my opinion.
And what a fine way to end the evening with the lyrics from "Rainbow Connection":
I've heard it too many times to ignore it
It's something that I'm supposed to be
Some day we'll find it
The rainbow connection
The lovers, the dreamers, and me
What could be more appropriate for a room full of dreamers setting the stage for the next generation of high-quality music recordings?
Like AVS Forum on Facebook
Follow AVS Forum on Twitter
+1 AVS Forum on Google+