47 Laboratory's 4713 Flatfish CD Transport/Player and 4705 Progression DAC: A Background
Commentary
Constantine Soo
21 April 2002

April 2002 marks the twenty-first anniversary of the CD, and in that time the format has gone through two decades of refinement in the forms of oversampling, jitter reduction, upsampling and so forth. In our continuing understanding of the CD playback process, each innovation resulted in higher levels of sophistication, propelling CD playback to an unprecedented summit. The most noted member born from a crystallization of all these efforts is possibly the $20,000 Linn CD-12 as reviewed by Jim Merod.

In today's marketplace, where advanced and market-dominating mainstream CD players are featuring variations of the aforementioned techniques, 47 Laboratory introduces its statement DAC based on a "Non-oversampling, Digital-filter-less DAC Concept."

47 Laboratory

After careers at prestigious high-end audio equipment manufacturers Luxman and Kyocera, Mr. Junji Kimura founded 47 Laboratory in 1992. In naming his new company, Kimura took the colors of yellow (ki-iroi) and purple (mura-zaki), which contain the syllables of his family name, then coded the colors using resistor color-coding standard of 4 for yellow and 7 for purple. Thus was born 47 Laboratory.

Mr. Yoshi Segoshi of Sakura Systems, 47 Lab's U.S. exclusive distributor, hosted an Interview with 47 Lab, during which 47 Lab's president and chief designer Kimura, and his marketing director Mr. Koji Teramura, iterated 47 Lab's commitment in refining Redbook CD playback. Convinced of the Redbook CD's engineering validity and unharnessed playback potential, 47 Lab considers the audio industry's latest venture into high-capacity digital audio formats as irresponsible, and in 2001 released its statement product, the $25,000 4704 PiTracer CD transport.

The aforementioned "Non-oversampling, Digital-filter-less DAC Concept" is authored by Mr. Ryohei Kusunoki, of whose theory Kimura is an advocate. Published in the November 1996 to December 1997 issues of the Japanese MJ Audio Technology magazine, Kusunoki theorizes on an advanced mode of CD playback technique without resorting to oversampling and digital filtering:

"The difference between the non-oversampling [digital-filter-less] DAC and the conventional DAC with the digital filter lies whether you attach importance on the accuracy in the time domain or in the frequency domain. In other words, whether you choose the musical performance or the quality of a sound. This trade-off line defines the boundary of the current digital audio format."

The time and frequency domains Kusunoki was referring to are the CD standard's 44.1 kHz time domain and 16-bit frequency (or amplitude) domain. Convinced of the Redbook CD's engineering validity and potential, Kusunoki opined on the digital audio progress in the following excerpt:

"…in the next generation digital format offered today, the selling points for better sound are quantizing bit numbers and sampling frequency rates. It only means lowering of distortions and extension of frequency range. The appearance of CD was an epoch-making event as a new format to follow LP. It delivered the sound of the master tape to our listening room. It was a crystallization of efforts of the engineers of that time."

In regard to the digital filter, Kusunoki believes it is an unnecessary evil. He claims that in both recording and playback, a typical digital filter collects data in series, creating a resultant cumulative delay so large that it would become detectable by the human ear ultimately.

Citing the workings of the 8x-oversampling high performance digital filter "SM5842" as example, Kusunoki claims a normally insignificant 0.22-millisecond inherent delay in each sampling process will be aggregated into a highly audible 2.13-millescond total delay time in the SM5842's 8x-oversampling. Claiming a further inevitability of "diffusion of sound coherence" from such serially collected data with the assumption that the signals are intervening with each other, Kusunoki concludes that the less oversampling, the better. He cites examples from prestigious firms such as Wadia and Luxman with their 13 and 3 tap achievements respectively - a tap being the waiting interval for signals in a sampling process - as he believes the smaller the number of taps would control such diffusion more effectively, providing the inherent advantage of his non-oversampling DAC. Furthermore, Kusunoki considers noise reduction characteristics of digital filters as detrimental given the adversely generated frequency and phase distortions.

In regard to today's sophisticated designs in oversampling, Kusunoki also reiterates that oversampling is simply a technique enabling use of gentler analog filter without generating additional information. He also states that in creating more samples, oversampling proportionately creates more errors as well. Therefore, in reiterating the theoretically superior accuracy of 16-bit non-oversampling versus 8x-oversampling/20-bit, Kusunoki summarizes that:

"A natural, stress-free sound that communicates the musicians' intention directly to you…is the sound of non-oversampling DAC. The feel of this sound is closer to that of analog reproduction."

On signal jitter, Kusunoki believes that a controlled, strategic distribution of jitter will represent a cost-effective and realistic approach in yielding an audibly more refined CD sound versus an across-the-band reduction technique. In conclusion, he states that the 44.1 kHz time-domain properties are the determinants of musicality in the D/A conversion process, whereas frequency-domain issues, such as jitter and noise, are secondary and their roles have been overemphasized by the industry.

The 4705 Progression "non-oversampling, digital-filter-less" DAC

Kusunoki's concept is incorporated into the current statement DAC product of Kimura's 47 Lab, the 4705 Progression, which features one surface-mounted 75-ohm digital RCA input and one pair of analog RCA outputs. Design highlights are: absence of both analog and digital filters, non-oversampling, passive I/V conversion, claimed shortest signal path of 35 mm among all DACs, and claimed smallest number of parts used in any DAC, namely 20 parts in total. The unit automatically adjusts to incoming sampling frequencies of 32 kHz, 44 kHz and 48 kHz and supports 24-bit 96k Hz signal conversion.

Miniature in dimensions, the 4705 Progression is unexpectedly heavy at 10 lbs. Supposedly subduing digital noise dispersion, two pieces of solid aluminum block anodized in black on the upper half and in brushed steel on the bottom half constitute the casing of the Progression. 47 Lab claims significant chassis stress reduction with the use of only 3 screws, resulting in "quick transient response, totally stable imaging and…flat energy balance of the sound." 47Lab also claims the DAC's concave top and bottom serve to "function as spike feet to release the vibrations smoothly and stabilize itself without any insulators underneath."

With no frequency filtering in the Progression, 47 Laboratory urges user discretion in amplifier-speaker matching as energy output equivalent to one-third of the music signal at 22k Hz will occur. Sakura Systems stresses the importance of matching ribbon or piezzo speakers to amplifiers of recommended power ratings. In addition to my observation to follow, refer to the Sakura Systems website for more information.

4713 Flatfish CD Player/Transport

Interestingly resembling the Doctor McCoy away-mission portable scanner from the 60's Star Trek, the Flatfish is the smallest single-box CD player I've seen for high-end audio applications. Flamboyant or impressive-looking it certainly is not. According to the Owner's Manual, the Flatfish's specific mass and weight are engineered to release disruptive vibrations. 47 Lab believes that damping a chassis will create delays and modulations in the flow of current, therefore, by constructing a compact and rigid chassis, a quick and proper channeling of mechanical resonance, rather than an over-damping of it, will produce sonically superior results.

In strict adherence to this design philosophy, elevated by a few centimeters and resting loosely on three tubular stainless steel feet in triangular formation, the Flatfish is at the mercy of user discreetness in remaining stable. To avoid toppling it, extra care must be observed in applying force to inserting and pulling cables from the unit's rear.

Said to be almost completely free of any construction stress, the 150g, 0.67inch-thick machined aluminum board doubles its role as the platform and casing of the Flatfish, as well as mounting board for all the driving, pick-up mechanisms and circuits. 47 Lab claims that the unit's lightness and the slightly off-centered spindle summarily diffuse vibrations, which enables minimal reliance on servo-controlled correction system that is sonically detrimental. The company also claims to have achieved quick transient response, thanks to the compactness of the platform and small surface area that limits stored vibration energy to a minimum.

With the exception of the remote sensor that faces the listener, all functions are accessible only from the top, lending a quaint look befitting the humorous Flatfish designation. 47 Lab employs four small surface-mounted toggle switches to control the functions of power on/off, Stop/TOC, Music Skip, Pause/Play. Located on the rear are digital coaxial outputs 1 & 2, and a pair of RCA analog outputs. The most unique top-loading design I've seen, there is neither a disc chamber nor cover. With the spindle being the only breaking point above the surface, the spindle base is at the same level of the unit's top plate, utterly exposing the laser and pickup mechanisms when not covered by either a mounted CD or the included white acrylic lens protector.

After being lowered through the spindle, a CD's inner ring rests tentatively above the spindle base. A very small, acrylic center cap screws down on the spindle, holding the disc firmly in place. As the disc-loading process is entirely manual, the Flatfish won't spin or access the Table of Contents until the user flips the TOC toggle switch. Afterwards, the sky-facing LED window will display total time and number of tracks. Again, from the listening position, track information becomes guesswork as the display is out of sight.

The Flatfish's center cap is a far cry from the flattening and stabilizing functionality of my CEC TL1's substantial full-disc stabilizer. In fact, as my Salamander Synergy 20's top shelf on which the Flatfish rested was lower than thigh height, I constantly witnessed the alarming and discomforting sight of CDs spinning in open air. Despite 47 Lab's claimed minimization of construction stress and disruptive vibrations as advantages of the design, it is intriguing that 47 Lab would decide on a transport mechanism that is not shielded from environmental intrusions. It therefore served as little comfort from the fact that the laser assembly read from underneath the disc and not above, exposing the transport innards and the CD's label to dust collection.

4799 Power Dumpty power supply

Neither the Flatfish nor the Progression could be self-powered and the standalone Power Dumpty acted as a shared power supply for both units. Via exclusive power line cables, the Power Dumpty featured two rear power outputs designed specifically for powering up both Flatfish and Progression. The Power Dumpty itself drew power via a generic IEC AC cord.

Epilogue

Although "Only the Simplest can Accommodate the Most Complex" is Mr. Kimura's design philosophy, his products share no utter inference with a crude, elementary design. While the appearances of both the Flatfish and Progression are admittedly bare, they are the fruition of advanced understanding in related fields of digital audio and the laws of physics. His approach to simplicity is by far the most comprehensive I've encountered.

The Progression is a product of a taxing evolutionary process. Its high-performance DAC chip, a vital building block in the Progression's performance, is an embodiment of the latest in circuit miniaturization. In addition, Kimura's unique accomplishment of extreme signal path minimization was undoubtedly a distillation of decades of experiences and experimentation from his former endeavors. Therefore, the collaboration of Kimura and Kusunoki, leading up to the noted refinement in theory and practice, deserves admiration and congratulation.

I will mention here that the sound of the Flatfish and Progression versus that of the CEC TL1 and Wadia 27 represented for me not disparity, but rather a testimonial to the venues advanced and adopted by brilliant designers in the maximization of CD format's potential. The inevitable and shocking advancement of CD playback quality as personified by the 47 Lab gear both dismays and excites me. I wholeheartedly consider my five-year ownership of the CEC/Wadia system an indispensable and rewarding experience because, as if in preparation for this review, the unique perspective the ownership accorded me expelled all groundless prejudices toward the 47 Laboratory Flatfish transport and Progression DAC in their laboratory-style encapsulation. See you next installment with my subjective evaluations.