Hart, Mickey: Mickey Hart's Mystery Box (with Debbie Charles, Julie Isaac, Jeff Sterling, Habib Faye, and others). Recorded ca. 1995, at Studio X, Sonoma County, California. Engineers: Mickey Hart and Tom Flye. 54+ minutes. Ryko 10338.

Hart, Mickey: Mickey Hart's Mystery Box (with Debbie Charles, Julie Isaac, Jeff Sterling, Habib Faye, and others). Recorded ca. 1995, at Studio X, Sonoma County, California. Engineers: Mickey Hart and Tom Flye. 54+ minutes. Ryko 10338.

As I have noted in a previous review or two, small recording companies sometimes have to come up with something distinctive to attract serious audiophiles -- audiophiles who would otherwise ignore what the company was doing if the latter did not offer some kind of "special" technology. Some of these special things (128x oversampling, 20-bit recording systems, etc.) are interesting and most are basically harmless. They offer no subjective improvement over the conventional, 16-bit, compact-disc system, but they do not screw things up, either. However, a few can have the potential to cause problems and tax the resiliency of the PCM-digital recording medium to the limit. By now, the HDCD (High Definition Compatible Digital) recording/playback system, developed by Pacific Microsonics and initiated into the recording program of Reference Recordings some time back, is fairly well known. Indeed, its operation has been analyzed widely, and either praised or damned by assorted critics. Whatever one thinks of it (and nobody, save die-hard record-review critics who could care less about technology, appear able to ignore it), the recording/playback process has at least probably been a financial success -- although I have no way of telling just how successful. In any case, a number of high-end hardware companies now market decoders and the procedure has been adopted by a few other recording companies for some of their material. One of these is Rykodisc, who has interacted with Reference Recordings on a number of recordings.

This release is a good example of what the process can do and before discussing it perhaps a brief overview of HDCD technology is in order, particularly when we consider that the process shows signs of becoming even more popular with both high-end and more mainstream record producers and listeners. The HDCD process supposedly eliminates many of the so-called "distortions" that supposedly plague all competitive digital-recording systems. Fully functional HDCD appears to be a digital compression/expansion and noise-shaping process that requires a separate decoding chip to realize its full potential. As far as I can see, the full system in many ways subjectively mimics the behavior of some of the analog-circuitry, dynamic-range expanding devices which were offered by dbx, RG, Phase Linear, and Pioneer for a number of years, even though it works in the digital domain. Without decoding, recordings made with the system will have the upper 9 dB or so from the loudest recorded passages compressed to about a 3-dB spread and will also have material at the lowest signal levels compressed to the extent that they may have a slight tendency to be pumped up and down in loudness. In addition, there may be other audible artifacts, such as an amplified fade to-silence and highlighted background reverb and an overall higher average signal level. For some individuals, an undecoded recording, because the background reverb will be more audible than usual, may be more pleasant sounding than a decoded one. However, I should note -- and this is very important -- that neither it nor a decoded version will be as absolutely accurate as a properly dithered, old-fashioned, 16-bit version.

From what I can ascertain, although the initial recording technology is 24-bit, the final compact-disc product has the expansion instructions under the command of a code embedded in the LSB of the recording -- reducing the part that handles the music to 15-bit resolution during passages where dynamic range must be expanded. Pacific Microsonics claims this anomaly is sonically inconsequential. Although the HDCD concept is an interesting exercise in digital signal manipulation (after considerable delay, its technology was finally described in a paper presented by Keith Johnson and Michael Pflaumer at the 101st AES convention in November, 1996), as I indicated above, I cannot help but believe that it was primarily designed to be a promotional item. Exotic processes such as this are a way for smaller recording companies, operating in a world where digital systems all sound remarkably alike, to somehow set themselves apart from the competition. Indeed, one reviewer has likened HDCD to a solution looking for a non-existent problem to solve. Another far less generous critic has stated that the system essentially creates low-level distortion artifacts that require the decoder to remove -- and that the decoder still does not remove all of them.

The net result is subjective digital performance that is no better -- and possibly worse -- than what can be provided by a good 16-bit system. That Reference Recordings, Rykodisc, and other recording companies who have adopted HDCD have material that sounds as good as it does is a credit to a PCM-digital recording technology that is already so low in distortion that it can tolerate such weird manipulations. This recording is a case in point. Hart is known for producing material with showpiece dynamics and those who like what he has done in the past will like what he has done here. In this respect, he is much like Keith Johnson, who has a knack for placing microphones so that they get all the dynamic impact (and maybe a bit more!) that percussion and low-bass-generating instruments are capable of delivering. The material is clean and clear, with lots of electrical manipulations that embellish, but do not overwhelm, the musical content. Music of this kind would be very difficult for even a fairly out-of-whack digital encoding system to screw up in an obnoxious manner, for there is a considerable amount of electronic horsing around throughout the program. Although this is a heavily manipulated recording, the combination of showpiece dynamics, well-defined instrumental separation, and frequency balance is demo grade -- and will entice many an enthusiast to cue up the disc when discriminating guests want to hear their system will do. As a further enticement, bass enthusiasts will appreciate the very powerful and deep tones that appear throughout. As with so many other recordings that have been electro-digitally manipulated, DPL decoding, with the center steering engaged, ruins the show. There is an almost total collapse into the center and I can see no reason to not opt for the phantom-center mode if the user desires to employ Dolby decoding to put a little low-cost ambience into some surround speakers. Some kinds of center steering will work, however, and both the Yamaha A1000 processor in my small system and the A3090 version in my larger one did well with their Classical/Opera modes engaged. (Classical/Opera is just a file name assigned to the program by Yamaha; the processing will work surprisingly well with less seriously pompous material). Best of all were the jazz-club simulations that both processors offer. Those were a hoot to listen to, dovetailing successfully with the recorded material, to say the least. Indeed, once engaged, I could not get myself to turn them off -- a bad habit for a critic to get into. Vocals were bloated somewhat but the instrumental impact was spectacular.

-- HF