POSITIVE FEEDBACK ONLINE - ISSUE 28
as reviewed by Arnis Balgalvis
Even though the Compact Disc has been around for decades, no revolutionary techniques for improving the CD playback process have materialized. By this I mean "something from the ground up." The impetus for improvement has always been there—CD sound quality has been criticized mercilessly and has been taking it on the chin since day one. But in retrospect, attempts to improve CD playback have been slow to come.
To be fair, some progress has been made by upsampling and digital filter techniques. We have also seen some very sophisticated CD transports appear on the scene over the years, with the Esoteric VRDS transport mechanism leading the pack. But let's face it, the thrust was always concentrated along the lines of incremental improvements of the basic approach.
A more stable clock here, an improved drive there, throw in some upsampling, or try a different algorithm, or maybe a belt drive. But in all this time nobody ventured outside the box. No one stepped back and reevaluated the present state of CD playback as a whole.
Not until now, that is.
It was Mark Porzilli, he of Melos and Pipedreams fame, who decided to take an outright fresh look at the existing challenges of extracting and handling the data imbedded in the spirals of the CD tracks. He spent two years evaluating the existing process and its potential playback pitfalls. The result of his insight is a completely new approach for retrieving the CD data and a corresponding, vastly improved playback process of this newly refined data.
A Brave New World
As far as Mark's basic concept is concerned, it's a very straightforward three-part procedure. First, get the complete data related to music off the CD; next get rid of the Reed-Solomon Error Correction Code (RS-ECC); then deposit the "clean" bits in banks of electronic memory—and not a hard drive—and retrieve them from there for playback. Very simple, right? As you will see, not really. Concepts are easy. Implementing them practically is a whole ‘nother world.
To get at all the data residing on a CD, Mark turned to a process he calls "Read Until Right," or RUR. In this approach the CD data is reread many times from a disc spinning at very high rate until all the data contained on the CD has been accumulated. This process is derived from a technique used by studios in the recording industry to ascertain that the optical CD master disc data matches the master tape.
The secret words here are "reread" and "many times". By spinning the CD at a very high speed, and with the aid of a custom laser pickup process, the data is read and reread over the course of many passes, until all the data residing in the pits on the CD has been amassed. Since a CD has 98 blocks of information subdivided into sectors, it can be determined beforehand just how much data is supposed to be on a particular CD.
During this reiterative RUR reading process the angle of the laser pickup is adjusted with each new pass to allow it to read as much of the imbedded data as possible. Whenever a data discrepancy is determined, the laser is commanded to return to that exact place and view that point from a different angle.
As implemented, RUR is an intelligent process and it knows what and where the missed bits are and it returns to ONLY to those bits. RUR stops when a preset goal has been met. For practical reasons, the MP is set to read until 99% of the CD data has been gathered. Without setting a threshold, RUR would read practically indefinitely, attempting to search out the last missing bit of data.
Mark claims that the problem with just about any conventional transport is that during the single pass of the conventional playback procedure, some of the data is bound to be missed. Remember, a CD has a bit rate of 4.3218 Million bits/second. A disc containing 74 minutes and 33 seconds of music contains 6.3 billion bits. Of all those bits only 33% are audio bits, with the rest being overhead. Getting all this data right in one pass is a lot to ask for.
One of the most serious, and also very common missteps, is not starting to read the data at the very beginning of the first block. Once that opportunity to hit the mark has been missed, critical data discrepancies compound, usually materializing in the form of increased jitter.
Mark's next step is to get rid of the data that have to do with the Reed-Solomon Error Correction Code (RS-ECC). But isn't the RS-ECC supposed to be a very crucial aspect that makes it possible for the CD to be played back? To be sure, the Reed Solomon error correction code can be considered as the single most significant contribution during the development of the CD. It basically solved the initial CD playback puzzle, and made the CD playable under real-world conditions.
The RS-ECC is what Sony brought to the table when Philips was readying to take the CD to market. The RS-ECC helped significantly at that time to overcome the myriad of stumbling blocks that raised their ugly head during practical playback. The short list includes dropped bits, jitter, Gaussian Noise, logic inversions, as well as all the possible permutations of interactions of these individual problems. In short, RS-ECC turned the CD into a commercially viable product.
But just like the "perfect sound forever" business, RS-ECC, according to Mark, falls short of perfection. And, he goes on, because RS-ECC is not perfect, it is at the top of the list of reasons why the quality of CD sound is constantly being called into question. The basic problem, he claims, is that in many cases RS-ECC creates uncorrelated bits of data when it finds missing bits that cannot be derived from the existing or redundant data. Mark insists that it is this uncorrelated data that come across as a layer of irritating sounds during normal CD playback. We are talking here about a subtle, almost subliminal, synthetic veiling of the sound. While not something overt, it is a tenuous, drone-like contamination always prevailing in the background. It gets to some people more than others, but it is always there.
Over the years many audiophile friends and I have been wondering just why it is that different transports sound different, and that even changing the support of the transport changes the sound. Furthermore, why is it that CD anti-static procedures make a difference, that sprays and polishes applied to the CD surface made a difference, that high intensity light-exposure made a difference, that demagnetization devices made a difference? All these changes are usually for the better. Why is it that a CD-R copy can sound better than the original? It does make me wonder. If the R-S ECC is supposed to be so powerful, why is it that all these "tweaks" often make a difference? I feel that Mark Porzilli, at the very least, raises good questions when he holds the RS RS-ECC suspect.
The last piece of the MP process is the flash memory. As data, using the RUR process, is retrieved it is stored in an electronic, or flash, memory and not on a hard drive. In his investigations, Mark discovered that data read from a hard drive directly will exhibit higher levels of jitter. If however, the data is retrieved from a flash memory, it is far less tainted and practically jitter-free.
The Memory Player is the culmination of all of the above considerations as well as some others. I consider the MP to be a minor miracle. When two years ago Mark first discussed the basic principles of what is now the MP, I have to admit that I was very skeptical. I was overwhelmed by the complexities of the proposition. That an individual could even consider that a real-world product could be realized from the vast challenges and complexities facing that prospect, was just about inconceivable. At least it seemed so to me. Of course, I kept that skepticism to myself.
Fortunately Mark benefited from the encouragement of more optimistic friends and associates. He forged ahead and made the Memory Player a reality.
A major part of that process was the formation of a new company, Nova Physics where Mark was joined by George Bischoff and Rod Hundley. George and Mark are old friends and have been working together while the memory player was designed by Mark and developed into a viable product by George. For them the MP project was déjà vu all over again—George and Mark go back together for decades. Their first venture was Melos Audio, and after that it was the Pipedreams. At Nova Physics, Rod Huntley is the third partner who is involved backing this venture financially.
The Memory Player Proper
The unit being reviewed here is the MP Transport. It consists only of the extraction and playback sections of the complete MP Transport/Processor unit. But that's not all! It also contains a CD Burner. And, as some of you have guessed, it is also a Music Server! The complete MP Transport/Processor will also contain a very sophisticated professional 32 bit DAC, a tube audio output stage, and an analog output level control. That lists out at $15,400 and includes a 300 CD library. The MP Transport with an 80 CD library and the CD burner goes for $9950.
I chose to get the MP Transport because it was all that was available 6 months ago. And because the innovative heart and soul of the Nova Physics MP is all contained in this piece. The RUR, the software to deal with the RS ECC issues and the flash memory is all here. The MP Transport I have puts out 44.1/16 WAV files on an XLR connector for the AES/EBU format or a RCA connector for SPDIF. A small toggle switches between the two outputs.
After all is said and done, all I need is a high quality D/A and away I go. The MP Transport is ready to accept external clocking on a rear-mounted BNC connector.
The MP Transport is delivered as a two-piece system: the main chassis and a laptop. While the MP Transport chassis is self-sufficient as a stand-alone unit, practical considerations mandate that it be a two-piece tandem. You see, the very sophisticated nature of the data retrieval and playback process, negates, at this time, a simple remote control. Yes, there is the main transport chassis that contains a CD drive/burner, flash memory along with a hard drive, and the memory for the complex operating system software. A touch screen on the main chassis is provided and all functionality resides in this main unit. But to remotely control the whole process, a laptop is provided. At this stage of the development process Nova Physics chose the most practical approach and relegated the complex remote control functionality to a laptop. A custom remote control would eat up too much R&D time and money.
Living with the MP
The dimensions of the main chassis are 17" W by 18 ¾" D by 6 7/8" H. The MP tips the balance beam at a substantial 25 pounds. The front panel features a 6" by 3 5/8" color touch screen display on the left side, a CD drive/burner drawer on the top right, and several pushbutton switches arranged along the bottom of the front panel. A small recessed area on the right below the CD drawer is provided for the controls of the complete MP.
A large illuminated circular switch powers the MP on and off. A push-button on the bottom far left side activates the touch-screen display. The rest of the switches along the bottom of the front panel are not intended for the user; they have been implemented to be used by the factory during initial configuration of the unit.
All functionality for the MP is provided by the touch-screen display, with everything needed to operate the MP residing there. I have seen the MP in action where only the touch screen was used. But in reality, the screen is quite small and the legibility of the text is a bit difficult to decipher. Besides, it is very unlikely that the MP would be located in a spot that would make it convenient to operate from the listening seat.
Hence the laptop. Using a wi-fi connection, the laptop mimics the touch-screen and thus provides all the functionality that is available on the MP's touch screen. And that functionality, as it stands today, is, quite frankly, a handful.
I'll briefly take you through a typical playback scenario. After turning on the MP a minute or so passes, the MP initializes, and a virtual link-up with the laptop is established. Now select the Memory Player icon, and a new screen will appear on the laptop displaying a number of other icons specific to the operation of the MP. You now have choice to play selections from the previously stored data on the hard drive, or you can "extract" the data from a newly selected CD.
For extraction, the CD is placed in the tray and the icon for the Extraction screen is selected. After commanding the extraction process to commence, the info from the CD is saved to the flash memory. The extraction process takes roughly a minute or so for a pop music track and about 12 minutes for a whole CD. Please note: Playing any previously saved selection can take place while the extraction process is taking place in the background.
When extraction has been completed, the CD drawer opens. You now select a playback screen displaying the newly extracted tracks. Choose a track and move it into the playlist screen. Initiate playback by selecting the Play button from the standard Play, Stop, Pause, FF forward, or FF back icon buttons. Playback progress is conveyed by a dynamic horizontal bar graph and by a numerical elapsed time display within the playlist screen.
After playing the desired tracks, the extracted info can be moved from the flash memory to the hard drive. And yes, there is a provision where the selection can be named so that it can be identified as it is stored.
The extraction process can be repeated as many times as it is necessary to build up a library of music. The size of the hard drive in the MP transport that I have is 80 Gigs. That is enough to capture up to approximately 80 CDs. The flash memory is 1 Gig and it will typically hold one CD's worth of data. But if larger storage capacities are desired, there is no problem with increasing the size of the hard drive.
I must confess that it took some time for me to get used to the transition of using a laptop for music playback. While I am reasonably used to dealing with my regular PC, my laptop skills leave a lot to be desired. So right away I was behind the eight ball. To my surprise, I caught on quickly with some practice. But even now, I'll admit that I quite frequently reach for the remote if I want to Pause or Stop a track. It's still not completely intuitive to pick up the laptop and to move the cursor to accomplish the desired command on the screen.
By the way, Mark Porzilli tells me that the usual concerns about component placement do not apply to the MP. Why so? RUR, that's why! Remember, reading the CD is no longer a one shot deal. The MP will RUR the data. On top of that, the data is played back from flash memory. So for the MP, all those concerns about fancy means of suspending products are a thing of the past. Ditto for those CD sprays and polishes. All they can do is to possibly cut down on the extraction time, but it will not affect the final data.
Like everything in audio, sonic differences are subject to perception and interpretation–and therefore subject to discussion. I have to point out that I have experienced the MP transport only with very high resolution systems. Based on that, I will be describing what I heard and what can be expected of the MP sonically. I will bring up several areas of its performance that really caught my attention. These areas of performance are prime examples what I feel best convey the overall capability of the MP. (I should also point out that the MP can only help to deal with what has been recorded, what already resides on the CD. It only works its magic from the realm of the finished Compact Disc side. The MP cannot reach back into the mastering, manufacturing, or recording phases.)
My experience has reached the point at which, when I want to really get the most out my favorite CDs, the MP has become my choice. I find that I reluctantly listen to my system fed by a standard transport. Mind you, I have a Teac Esoteric P-03 on hand—well, at least while I am reviewing it—a fabulous product that just happens to be the best mechanical transport that I have ever used. Why might I use something other than the MP? Well, if I have the time and am looking for the best possible reproduction of my CDs, I go to the MP Transport. If, however, I want to hear something right away, I will succumb to the instant gratification that the P-03 or the Sony SCD-1 that I own provides. That would be especially when I am not in the mood to put up with the longer procedure of extracting a CD into the MP.
Think of photography. Use a digital camera and your photo is available for viewing immediately. If you want all the subtlety that is available on film, well, you have to be patient and wait until the processing lab does its thing.
The whole picture of urgency and impatience changes, however, once a substantial number of CDs reside in the MP memory. Now you have a Music Server. But not just an ordinary Music Server that stores compressed music files on a hard drive. All the music stored in the MP is available in a flash, completely uncompressed, ready for playback. This was never more convenient than during my recent recovery from a knee operation. I could sit in one place, pull up selection from the hard drive into the flash memory, and listen to many of my favorite selections for hours.
The Music Server aspect brings the MP right into the forefront of the current mass music scene. But it does so with the added advantage of providing uncompromised sound quality. With the MP you can still have all your multitudes of desirable music stored, but all of it is ready to be played back at the best possible sonic level.
Getting Down to Business
In order to get the most out of a component, I try to hear it under as many different conditions as possible. It's part of my reviewing process. I have a few audiophile friends who are willing to have me come over with a piece being reviewed. Because I am very familiar with their systems, having heard them many times, they are a great help for me. The fact that the listening settings are as different from one to the other as is possible to imagine is of great help, for obvious reasons.
The MP situation was easier to handle, however. It turns out that two of the systems I rely on already include the MP Transport, with one of them sporting the full-blown MP. Consequently, my MP Transport had to take only one trip. So, counting my own set-up, I have heard the MP under four very different circumstances. In all four systems there was no commonality at all in any of the components, nor, of course, room conditions. Speakers ranged from my own Avalon Diamonds, to the Dali Megalines, to the Model 21 Pipe Dreams, to the mbl 101Es. One system even had tubed power amps.
All this listening was intended to firm up my impression of the MP under various listening conditions. In my own system I used two power amps and three DACs. The power amps I had were the Marantz MA-9S1 monoblocks and the ML 332, with the Marantz being the more transparent, more dynamic and detailed by quite a margin. The DACs on hand consisted of the Esoteric D-03, the MSB Platinum DAC III with an analog volume control and the Nagra DAC. To be sure, all three sounded different. And all three sounded great. The D-03 did manage to impress me more than the other two. Too bad for me—I own the MSB unit. But then again, since I can't afford the Teac Esoteric line, I am ahead of the game.
It was also fortunate that I got my hands on the TARA Labs The Zero AES/EBU and SPDIF digital cables. What stunning products! These components also made their rounds to two of the other systems and everyone who heard them agreed to their superior performance. These cables made my evaluation of the MP much more productive. They allowed me to hear more into the music and unravel its complexities. Thank you, Matthew Bond.
The Sound of Music
One listen to the MP and all little operational misgivings evanesce. They just evaporated. Because what I hear is relaxed and open and pure. And dynamic. I mean, how many ways are there to say "marvelous!" All I can tell you is that from the very first time I heard the MP—not in my own, but in a friend's system—I have been completely taken by its sonic charm.
Those first sounds still waft in my memory. It was pure drama to hear just how much more information was revealed when the switch was made, in this case, from a Wadia 270 SE transport to the MP transport. Suddenly more sound space materialized, the attacks had more life, and the sound took on an effortless demeanor.
This friend happened to be the very first to receive a Memory Player Transport. This was a huge event not only for him, but for all his audio buddies as well. Many exciting listening sessions were set in motion. And a remarkable thing happened. Of the more than a dozen sound enthusiasts who heard the MP there, a vast majority immediately placed orders for their own units. I'm talking about something like 9 out of 10 people.
I have never ever witnessed such overwhelming rush for a product. And, yes, I was one of those who ordered one right away.
The Nova Physics Memory Player is the best sounding source component that I have ever heard. For me, it elevated the listening experience to another level. I found the MP to be very authoritative and at the same time very relaxed. The music has this free and easy flow to it. And I do not mean rounded off or stilted. It is about as wideband as I can imagine. It is quick to respond, the brass has the right bite and the right blat, the cymbals shimmer loud and clear, and the bass goes way down very vigorously.
Because the MP Transport is exceptionally transparent it revealed nuances in a new light. They could be dynamic contrasts, harmonic complexities, or soundstage subtleties. When taken as a whole these now more sophisticated shadings contributed to better dynamics, better resolution and better timbral presentation. As a result, the musical presentation was far more vivid and, consequently, far more involving.
One of the most appealing aspects of this effect was the manifestation of a rapt connection between the listener and the singer. We all know how difficult it is to reproduce the human voice. Well, with the MP in the system, a sort of beguiling, almost soulful communication was created between the vocalist and the listener. It was a mood that quickly established itself where only the performance mattered. It was remarkable; for me, no other source component has been able to create such positive involvement to such a deep degree.
While the MP did have a way with voice, this was not the only area where rapt musical communication was established.
For one thing, the MP had the ability to reveal a new-found dimensionality in most CDs. This was more than just more depth or better layering. It was a wistful expansiveness that brought life to the soundstage. The space between and behind the speakers was alive with sonic activity that not only fleshed out the space around each of the performers, but it also combined them into a continuous sound space. This whole expanse was imbued with a dynamic profusion of detail accenting the ebbs and swells of the musical excursions. The more vigorous surges of a full orchestra took on a forcefulness that propelled the dynamics with an added measure of fervor all across the whole sonic spectrum.
The MP was truly magnificent in the bass region. It rendered the bass quite dynamically, making it much more potent, much more powerful. The fidelity in the bass region worked to provide the music with an exceptionally solid foundation. It was the sort of might that was never overbearing, yet was instrumental in conveying a sense of abundant energy.
All I can tell you is that the MP repeatedly established a very welcome sense of comfort, one that hinted at some of that zest of live music which we take for granted in a live experience. Time and again, I found myself being drawn into the performance with an unprecedented level of involvement. This was unlike attempting to listen intently when trying to evaluate something. No, the MP captured my attention in an almost subliminal manner, a spellbinding state that rebuked any attempts to be dissected or analyzed. At times this was so much so, that when I realized just how involved I had become, it was not unlike being startled out of a daydream. The MP has the ability to draw you into the music and take you for a ride on a magical musical carpet.
The above might lead one to suspect that imaging might be dreamy or diffused. In fact, quite the opposite was true. Even though the images of the performers were dimensional, their presence was confirmed with great poignancy within the soundstage. Instead of pointing to a very specific point, the attention was now focused on a position. This laterally spread out stage now had a more active and a more alive feel without sacrificing any pinpoint specificity. I became aware of the fact that as more and more CDs were played back, I was realizing to a new degree just how different each CD really was. Of course, we all know that each recording has a basic signature. It's the recorded venue, manipulations, certain attributes, some small faults and similar problems that contribute to make it so. Sometimes picking a recording apart is part of the fun—or at least it seems to be for audiophiles. At other times it is just plain annoying. But with the MP you get the awareness without the temptation to dissection. What a relief!
I have saved the best for last. If there's one thing that the MP has going for it, it is an unprecedented sense of ease. It has a soothing effect on the musical calamity. No matter the harmonic complexities, no matter the dynamic demands and excursions, the MP digs into the music in an unprecedented manner to not only reveal new details, but to present them in a relaxing, free-flowing manner. Every note appears to be in place and every instrument appears adorned with credibly palpable timbres. Call it "sonic serenity," if you will. What it certainly is not, is a case of euphonic embellishment.
Under these circumstances of calm, listening sessions can go on for hours, without any stress in sight. But most importantly, the musical presentation, in this serene state, is by far more enjoyable. The music doesn't just blossom, it is in full bloom.
The Memory Player is a significant advancement in the evolution of compact disc playback technology. It is a very meaningful step in the quest to advance the playback capability of the CD.
The MP is not only about a better soundstage, better details, better dynamic shadings. It can do all that and more, rather admirably, thank you. No, the MP is about a better connection to the music, about increased listening enjoyment. Because the extracted data is more complete, it can communicate the recorded musical values more thoroughly and bring the performance closer to the listener.
For a first cut at a product concept of mind-boggling complexity, I find the MP to be a resounding success. This technical marvel is not a development waiting to happen. It is ready for the here and now.
It has my Highest Recommendation. Arnis Balgalvis