Session Log, The Ultimate Piano Recording

A great recording has three components—a virtuoso performance, a memorable sonic event with carefully selected instruments, and a skillfully arranged recording setup.
Author:
Publish date:
Image placeholder title

By Cookie Marenco

Grammy-nominated jazz pianist Vijay Iyer.

A great recording has three components--a virtuoso performance, a memorable sonic event with carefully selected instruments, and a skillfully arranged recording setup. Last summer, when the wonderfully talented pianist Vijay Iyer asked me to record a solo piano release at my Bay Area facility, OTR Studios, the first two elements were dropped in my lap. My challenge was to create the third.

The Performer, the Piano, and the Formats

Vijay appreciates great recording techniques and an excellent piano. Light or heavy action, key grip, pedal noise—name it, and it varies between pianos, making it a challenge for any pianist to perform at their best.

Made in 1885 and rebuilt by renowned Steinway specialist Sheldon Smith, my seven-foot, 85-key Steinway B has served my studio for more than 25 years. Aft er several albums recorded on this Steinway, Vijay’s comfort with my 100-plus-year-old piano was part of his decision to come to OTR.

Vijay wanted to future-proof his music by recording with the highest fidelity audio gear and format available, so we used the ESE (Extended Sound Environment) techniques devised for our own Blue Coast Records projects. We simultaneously recorded tape to a two-inch, 24-track Otari machine, DSD to a Sonoma DAW, and CD-quality audio at 24 bits and 44.1kHz.

There’s nothing like a well-calibrated tape machine to record acoustic instruments. With tape above $350 per reel, we ran parallel tracks on our Otari to conserve tape. Dolby SR noise reduction let us run at 15 inches per second, extending the recording time of a reel to 30 minutes. Six tracks per recording pass gave us four passes per reel—two hours of recording time in total.

Direct Stream Digital

DSD is a one-bit format, which is a wholly different paradigm than conventional PCM digital recording. Instead of logging a 16- or 24-bit number so many times per second (44,100 for CD-quality audio) to record an audio signal, it samples at approximately 2.8 million times per second, registering a 1 if the amplitude increases compared to the previous sample, or a zero if it decreases. This “wordless” approach is initially hard to wrap your head around, but the results are analogous to looking through a perfectly clean, clear glass window as opposed to a screen—you can see the view through both, but there’s more dimension and depth through the glass.

DSD’s major advantage is that the quiet sections of audio retain the detail of the louder sections—unlike PCM digital audio, where at lower sample rates, the sound quality falls apart by losing bit depth as it gets quieter. Because the “sample rate” from your ears to your brain is much higher than any technology can produce, the comparatively high sample rate of DSD is far closer to reality. DSD dynamics seem to fall faster and more naturally, while PCM sometimes feels puffy in the decay.

Engineers debate DSD versus the current “high end” of PCM recording: 24 bits at 192kHz. Our blind listening tests confirmed DSD as our choice for archiving, mixdown, and multitrack recording.

Image placeholder title

Sony invented DSD in the mid-1990s. Used as the format for Super Audio Compact Disc (SACD), it was intended to replace the CD as a physical format. Sadly, DSD nearly became extinct when MP3 took over, as consumers preferred the convenience and mobility of downloading and storing hundreds of songs on a pocketsized music player. Looking at that trend, Sony lost interest in DSD. Today, the pendulum is swinging the other way: Music enthusiasts are increasingly disenchanted with lo-fi MP3s, and there’s a renewed interest in high-resolution recording—and distribution— of files worldwide.

At the forefront of this movement is Gus Skinas. Part of the original Sony DSD and Sonoma Recording system team, Gus now operates the Super Audio Center in Boulder, Colorado, where he services, sells, rents, and masters for DSD recordists. Find out more at superaudiocenter.com.

Vijay’s Recording Chain

For the Vijay Iyer project, we chose Meitner (emmlabs.com) DSD converters upstream of the Sonoma DAW. Before these, we inserted Bybee Bullets, which are compact, inline XLR devices that reduce the small high-frequency distortion that occurs in recording acoustic instruments, mostly noticeable in harp or piano.

Image placeholder title

Typically, I use a stereo pair of B&K 4012 mics, placed about eight inches apart at a 60-degree angle, straddling the middle of the piano, and not directly over the sound holes or any part of the metal frame. We place piano mics about halfway from the lid to the strings, with the piano lid open full stick. Slight variations in mic position depend on how hard the performer plays the keys.

In Vijay’s recordings, we used a six-mic setup, adding a Neumann U67 and U87 three feet from the piano, and a pair of Neumann KM184s placed 24 inches behind the B&K mics, halfway between the lid and edge of the piano.

We connected all the mics to the preamps with silver/copper alloy cables that we built in-house using a patented custom design. Originally, this cable was designed for the French aerospace industry, and it has incredible ability to reduce RF and extraneous noise. It’s expensive but well worth the price—on remote recording gigs, we can do 100-foot cable runs without loss of quality.

Image placeholder title

We used Millennia HV-3D mic preamps. They’re clean, uncolored, and can handle acoustic piano transients without distortion. With eight channels in a compact unit, they’re also great for remote recording. The more you use them, the better they sound. Finally, we didn’t use any EQ or compression in the recording or mixdown of the Vijay Iyer solo project.

The Future of DSD

Every artist wants to preserve their work at the highest quality, and wants their fans to experience what they hear in the studio. It’s now possible. Limited only by the bandwidth of your Internet connection, delivery of large files is here. DSD (and high sample rate PCM) files are large—around 40 times the size of MP3s given the same running time.

While recording Vijay’s projects, our in-house programmers developed a way to distribute DSD recordings over the Internet. Vijay’s adventurous spirit—along with the willingness of his record label ACT Music—made him one of the first artists to release his music as a DSD download.

With Vijay’s recording, you can compare formats from CD quality to DSD at actmusic.downloadsnow.net/solo. If you’re hooked on DSD piano recordings, you can also get Art Lande’s While She Sleeps and a duo recording with top keyboardist Matt Rollings and Jenna Mammina at bluecoastrecords.com.

To listen in true DSD, you need a device with actual DSD converters. The Korg MR-2, MR-1000 (reviewed July ’07), or MR-2000S are good options. For listening only, Korg’s AudioGate software (a free download at their site or at audiogate.bluecoastrecords.com) converts DSD files to be playable on your best “regular” audio hardware, and even if you’re listening at 24-bit/96kHz (or even 16-bit/44.1kHz), music that was originally recorded in DSD will still sound better.

As evidenced by the eight-track DAW prototype Korg unveiled at a private showing during 2010’s AES convention, DSD is undergoing a rebirth. Its exquisite audio fidelity, ease of use, and convertability make it the format of the future.