| Home | Audio Magazine | Stereo Review magazine | Good Sound | Troubleshooting |
THIS WEEK'S STATE OF THE ART
Anthony H. Cordesman
Tighten your deltoids, loosen your cannon, and flaken up! We're about to talk serious digital. Well, as serious as it gets. No issue will ever be as important as the precise cone shape needed to produce the optimal Tip toe; this is supposed to be a hobby we enjoy.
The Broad State of the Art
Fortunately it is possible to enjoy most of the latest compact disc players. We no longer have to worry about "perfect sound" that actively irritates and costs over $1000. The bulk of the current generation of Philips -- and Sony-process players now sounds very much like decent mid-fi receivers. These players still lack harmonic detail, the transparency of the best phono front ends, accurate deep bass, a smooth top octave, and the ability to reproduce the natural air and sweetness of low-level passages. Yet some models cost as little as $129, and good Philips-process players like the Magnavox 2040 routinely discount for around $200. It's a bit churlish to complain that such players still don't offer better value than the best record-playing systems. They do offer better "tracking/alignment" than the vast majority of phono and tape players, far more accurate timbre and frequency response, and better dynamic range. At worst, such players introduce most buyers to a far higher level of fidelity than any other signal source available at this price point.
If there is any drawback to this sudden proliferation of acceptable players, it lies in the fact that the cheaper ones are beginning to decline in quality.
Many are now designed to cost, to the point where their output circuitry, scanners, and digital circuitry suffer.
The frames are also getting much cheaper, and the expected life lower. This is not true of the medium-priced machines available on the U.S. market-which generally benefit from the latest manufacturing technologies and economies of scale-but the latest Philips machines in Europe definitely seem to constitute a step down, and the flood of things to come from Japan has a strong element of dreck. This may be the time to buy that Magnavox 2040 for $200 . . .
Do All Digital Players Sound Alike?
The vast majority of current players sound alike. Not identical, but very much alike. They differ in bass extension and up per midrange smoothness, and in many minor ways, but the advantages and disadvantages balance out. None really sounds bad. None gets farther than Good to Very Good.
The Philips-process players are still slightly smoother than the latest Yamaha players (whose sound seems to be getting worse, in both brand-name and OEM versions) and most Sony-process players. You can still count on Philips-process machines for long-term listenability, while you never know where a non-Philips machine stands in terms of phase performance and trick filters.
Nevertheless, the latest samples of virtually all CD players outperform many of the high-priced high-end machines of not long ago. A stock Magnavox 2040, for ex ample, has more focus, detail, and harmonic information than a Revox. A stock Hitachi at least equals a high-priced Kyocera. The Meridian MCD and McIntosh MCD7000 compact disc players now sound only slightly smoother than the latest stock Philips players. While such stock Philips players may not equal the Mission 7000R's highs, neither do they have its continuing lack of deep bass, and the regular Meridian MCD is now only a tinge sweeter than the regular Philips-supplied machines.
Many of these lower-priced players also do about as well as the most expensive Sony machines, with some careful reservations. Sony seems to be making some production adjustments in its top-of-the-line players that steadily improve their sound.
After a tip from a reader. I managed to borrow a later model Sony CDP-650ESD. This unquestionably had more transparency and detail in the highs than any player I've yet heard, and a far better overall level of sound quality than previous Sony’s. Nevertheless, I've heard some 520s and 620s that sounded like old-style transistor receivers, complete with grain and hardness. Maybe LA can get some word out of Sony regarding such changes, but in the interim most samples of the top-priced Sony’s do not outperform a stock Philips except in resolving the extreme highs, and do not do as well in the more critical upper midrange.
In short, price is literally no object in choosing between most current CD machines. If one adds the growing problem of quality control and consistency within a given brand and model, there is virtually no correlation between brand name, price, and sound quality.
Bullshit Advertising and Bullshit Physics
The flood of new CD player features now being advertised is largely a flood of hype.
There are models coming out, with better digital filters and better audio sections, but the vast majority of features advertised in magazines or at the Vegas CES are classic examples of "bullshit physics" -implied technical differences unsupported by any respectable theory, technical specifications, or proof of superior performance.
This is particularly true of companies making vague claims regarding a player's ability to read digital information from the disc. A particular new laser assembly either reduces the rate of error, or error correction, when reading a CD, or it does not. It is either measurably superior, or it isn't. A CD stabilizer either reduces error rate, or it serves no purpose. These are phenomena that can be measured and counted. There is no excuse for advertising features without technical proof of their value when there are no subjective factors to influence performance.
Criticizing the advertising of new and better filters is a bit less certain. Filters really are being improved, but not all of the improvement is measurable. Nevertheless, a given filter and digital circuit can be measured for phase and frequency response, which will be either more accurate, or not.
My point is that with digital many things can be measured; claims of technical superiority should be substantiated by such measurements. Incidentally, this is also true with respect to providing decent descriptions of the analog output circuitry. Even if no one can prove what sounds best, it is clear that decent circuits and components are critical.
It is possible to describe these improvements precisely, rather than make vague claims to superiority.
Unfortunately, virtually no CD-player advertising provides meaningful technical detail. Some even omit all useful information, relying purely on deception or obfuscation. For example, McIntosh's advertisements begin by saying that "the McIntosh has the best sound of any compact disc player," without showing any source for the quote. (Presumably, Mr. McIntosh's totally deaf Scots Granny thought it was great.) The pseudo-tech ads are no better. What does Sony's unsupported claim to sell "third-generation digital technology" in its CDP-302 mean? What does Pioneer's resonance data on its disc stabilizer mean in terms of improved reduction of tracking error? What measurable effect does Kyocera's ceramic base have on tracking error? What on God's earth are Denon's "hand-tuned" DA converter and "real-time phase correction circuitry," and how on earth does its "linear phase filtration" result in "perfectly"(perfectly?) flat frequency response? What are JVC's "Y servo system" and "in dependent suspension system"? I suggest you regard all technical claims in CD player ads as lies unless supported by reasonable technical data. I also suggest that you insist on bearing a difference before you pay an extra $100 or more for the digital equivalent of "ring around the collar" (or, actually, its prevention). The same holds true for the rising number of snake oil modification salesmen who claim that their process is proprietary or secret. Respectable firms like The Mod Squad will not tell you every detail of how to do their work, but they'll give you a damn good picture of what they give you for your money. I've had two "modified" machines come in with nothing but a few capacitor changes and very high price tags.
Get a full, written description of any modifications before you buy from a stranger.
Stereo Review's Sound Alike Contest My comments do not mean that it isn't worth hunting out the few CD players that do make a difference, some of which I will describe shortly. At the same time, they may indicate some of the reasons I'm not surprised by the results of the listening survey of CD players in the January, 1986 edition of Stereo Review.
While I approach listening from a totally different perspective, my own listening to over 50 CD players has indicated that the sound of the latest 10-15 players has converged to the point where it is indeed hard to say that their sonic differences matter, at least compared to their common strengths and limitations.
I also think the quality of the tests by David L. Clark, and the reporting by Ian Masters, is far better than Stereo Review's previous work on cables (or Julian Hirsch's ambivalent and uncertain comments on high-end turntables). The CD listening test is an article that every serious audiophile should read.
At the same time, this article should be approached with some important caveats:
• David Clark clearly states that it is not possible to correlate the statistical results to the kind of differences audiophiles listen for in buying equipment (see his comments on page 55 of the article). Ian Master's comments on page 56 do not contain Clark's careful qualifications, and end up essentially saying that the results are all things to all men.
• I wouldn't expect to be able to distinguish subtle sound differences between CD players under the listening conditions used in the Stereo Review tests. Comparatively rapid shifts from player to player, using a mix of test tones and music on an un familiar system, good as this one seems to have been, would confuse the hell out of me, especially given the noise and positioning problems common in most group tests.
• I realize that Stereo Review could not de vote enough space to print a full technical article, but any such experiment is only valid if it has a controlled error budget and proven parameters of validity. I feel it is incumbent on Mr. Clark to show that his ABX comparator methodology can consistently demonstrate that listeners could consistently detect reasonable levels of distortion, differences in dynamics, sensitivity to sound stage data, etc. before going on to apply the methodology to specific equipment.' I Such a demonstration of the ABX's validity may never be offered, for the key reason that the "long- term listening" differences Tony refers to are thought by the ABX people.
In general, to be imagined. The ABX methodology has conclusively been shown to successfully discriminate certain kinds of differences: 1.0 dB frequency response variations, and the like.
-LA
• No one can quarrel with the basic statistical analysis used in the ABX tests; , the problem lies in showing the extent to which the method is valid in detecting the kind of differences that emerge in prolonged listening to music.
• These caveats take on special importance because many of the previous results from AB testing have tended to indicate either that everything sounds the same, or that listeners cannot assign any aesthetic value to differences heard under such test conditions. The statistical results never clearly indicate whether the equipment is being tested or the methodology.
• I would like to see such AB tests validated by controls using other listening methods. They could be preceded and followed by long individual listening sessions on the listener's own systems, to see what differences the listener thought he or she heard before and after the AB tests.
Presumably, a large sample would avoid many aspects of listener preconditioning before the AB test. Results of the kind Mr. Clark reports in SR would actually de-condition listeners from praising the more ex pensive, exotic, and/or feature-ridden equipment.
I don't mean to carp at Mr. Clark's work; properly qualified, it is important and very interesting. The mystique of the high end deserves constant deflating, and hi-fi nuts are just that. I do feel, however, that the limits of AB testing need far more attention. We risk replacing the everything-is measurable versus nothing-is-measurable rift in the audiophile community with a conflict between AB Testers and the Golden Ears.
The Meridian Professional Re-evaluated: $1400
I also find Mr. Clark's results for the Meridian MCD Pro surprising, however. I have already said that I can no longer get very excited about the regular Meridian MCD or Mission 7000R. The competition has nearly caught up, at much lower prices, to the point where only rich audiophiles would pay for the kind of differences involved.
[2. Oh yes they can; Les Leventhal of the University of Manitoba argues in a very interesting paper that the statistical analysis generally used is weighted in favor of finding no difference between devices under test. WC will be presenting a summary of his paper in an upcoming issue.
-LA
------ Meridian Professional
The Meridian MCD Professional, however, is a different kettle of fish. To start with, it measurably performs differently than most players in the deep bass.
On a speaker with good bass extension (the Infinity RS-1B and IRS III, or VMPS Towers), an RIA reveals notably stronger bass dynamics on the MCD Pro, even if it is set for exactly the same average white noise loudness.
I have had the Meridian MCD Pro, the Mod Squad improved version of the MCD, the Mission 7000R, and a Magnavox 3040 set up in my listening room at the same time. Visitor after visitor picked out the MCD Pro's superior detail and musicality during extended listening sessions, without any prompting or hype from me.
My wife and daughter have developed a stony indifference to audio technology and brand names, but they consistently opted to listen to the MCD Pro in spite of its in convenience in loading and programming.
This does not, however, mean the MCD Pro remains this week's top dog. The second version of the PS Audio CD player now offers all the sonic advantages of the MCD Pro, plus better bass and higher resolution, at about half the price. Further, Steve McCormack, of The Mod Squad, has shown that the Meridian Pro can be made embarrassingly better with $300 worth of modifications-to the point where every MCD Pro owner who really wants the best possible CD sound is virtually going to have to pop the money for the Mod Squad conversion. Another engineer, Wayne Emerson, has achieved further benefits by adding his group delay filter for $160. The result is a $1860 "super-Pro" which only the most hard-core high-ender will buy, but which serves as very material proof that better products are possible.
Prolonged use of the stock MCD Pro has also revealed a number of flaws of consider able importance in a 51400 machine. These include (a) the obtrusiveness of the hum field mentioned in my original review. The machine radiates into preamps and phono equipment with a grim persistence no other machine approaches; (b) the over-full bass and lack of bass control mentioned in my previous comments; (c) a tendency to over heat, and for the programming features to freeze unless the machine is turned off and on at the rear; (d) far too loud a "bang" from power on-off transients; and (e) the tendency of the rubber disc stabilizer to hurt as much as help because of the difficulty centering it. Worse, The Mod Squad discovered some serious wiring and assembly errors in the machine I sent for conversion; the ugly spectre of the British Golden Quality Control Rirkey lurks somewhere in the MCD Pro's innards.
The MCD Pro is still one of the few CD players on the market with what I feel to be a truly audible advantage over the cheaper Philips machines. Nevertheless, the state of the art has moved forward to the point where outside modification is needed to get the best from this machine, and the appearance of a sonically superior PS Audio player at about $800 leads me to feel that the MCD Pro is no longer competitive.
The Mod Squad Upgrade of the MCD Pro: $300
The only thing I hate worse than expensive modifications to top-quality high-end gear is being stuck with gear you can't modify or upgrade. Fortunately, you can keep your MCD Pro at the cutting edge of the state of the art, but only at a price.
For $300, Steve McCormack of The Mod Squad will fully check out your Meridian Pro and perform a number of important modifications. These include changing the ICs in the analog circuit to improve the sound, replacing the ceramic integrator caps with top-quality film capacitors, in stalling bypass capacitors in the power supply and several other critical areas, improving the regulator to reduce noise, upgrading the wiring, and putting in top quality RCA jacks.
The result is far better-controlled bass, with no loss of bass depth. You now get both bass power and dynamics, and musical accuracy and control. You also get another important notch of improvement in mid range and treble-resolving power, more convincing string and woodwind tone, and more realistic voice rendition.
The Mod Squad conversion results in a player that outperforms the best sample I've heard of the Sony 650, and slightly surpasses the PS Audio, something the stock MCD Pro definitely does not. Admittedly, the Mod Squad conversion raises the total cost for the player to $1700, which is a lot with such a limited probable time window of superiority and an $800 rival. Nevertheless, I consider the conversion an absolute must for audiophiles who have already sunk $1400 into an MCD Pro; it might even be considered cheap. Call The Mod Squad at 619-436-7666; you won't regret it.
The Deluxe Mod Squad Conversion of the Meridian MCD: $500
I can't be equally positive about the Mod Squad conversion of the regular Meridian MCD, but this is a matter of price, not quality.
The Mod Squad has a whole family of con versions for the better existing CD players, including the Kyocera DA -01, the Magnavoxes, and the Missions. At the lower end of the price scale-$200-$300-you get better ICs in the analog section, better resistors, bypassing of electrolytic caps with film capacitors, upgraded wiring, bypassed power supply caps, and high quality RCA jacks.
These are all modifications many technically inclined audiophiles have made in both the U.S. and the UK; there's no doubt about their merits. You don't solve the basic problems of such CD players in resolution of musical detail, but the midrange and high end seem sweeter and more balanced, and the upper midrange loses the "grain" resulting from second rate analog circuitry. There is less brightness, and more of the missing lower midrange.
For an extra $200-$300 you can go whole hog, getting many of the mods made to the MCD Pro, including film capacitors throughout the signal path (either internally or in an external box). In the version for the Meridian MCD, you also get a new power supply, which justifies the $500 for the deluxe conversion as opposed to the $300 for the standard. Having listened to the deluxe mod of the MCD, I have to agree that it sounds smoother and sweeter, has better bass, and less midrange hardness and irritation. It is a very listenable machine, and one that stands out from the crowd.
At the same time, the Mod Squad conversions, at $200-$500, do not solve any of the basic sonic problems in CD sound. It's pretty hard getting worked up over such conversions, given the probable cost to switch to a PS Audio, including a trade-in on your old machine. Not to mention the rapidity with which other manufacturers are coming out with improved machines.
I'd stick with the lower-priced conversions, or go for a trade-in.
The Wayne Group Delay Filter: $160.
The Wayne Group delay filter is both an alternative to the Mod Squad conversions and a supplement to them. Once again, I suspect any MCD Pro owner totally com mitted to the state of the art will go for the Wayne Group Delay Filter, as well as the Mod Squad modification. Others will probably want to choose between the Mod Squad effort and the Wayne Group Delay Filter, or compromise by combining the lower-cost version of the Mod Squad con version with the Wayne Filter.
Very briefly, the Wayne Group Delay Filter is an active filter in a box, with input and output jacks and a separate power supply. It is intended to correct the group delay induced by the anti-aliasing filter used during the CD recording process. This involves as much as 40° of phase shift, and the Wayne Group Delay Filter corrects this within about $5. At the same time, the Wayne helps correct for internal phase shift in the players. One measurable result is that it removes both the lead spike apparent on most recorded square waves, and about 3-6 DB of the ringing.
Music, however, is at issue; the Wayne has very audible benefits. It cleans up the hardness and lack of detail in a lot of other wise unpleasant early classical CDs--particularly those with hard string tones.
The upper midrange and treble of most discs sounds more natural; the transparency is improved in inner detail, with more air and harmonic detail, and percussion loses some of its flatness or clipped character, sounding more live.
In fact, the Wayne Group Delay Filter provides all the sonic benefits it is advertised to provide-which must be something of a first for a high-end "black box". While I was not able to try it on every CD player in sight, it provided some sonic benefit with every player I tried except the PS Audio, where similar correction seems already to have been applied. While it does not lift the mythical "thousand veils," it does lift a hundred or so. CDs sound consistently better.
The Wayne Filter even gives some potential side benefits. It allows a little more headroom in dubbing CDs to tape: It even improves the sound of some records and moving coils if inserted between your preamp and power amp, although it is not sufficiently neutral to make this a good idea for permanent installation by the owners of really top-drawer electronics. It evidently helps compensate for the cutter roll-off in some records, and the resonance in some moving coils.
In fact, the only problem I can see with the Wayne Group Delay Filter is that its effects are additive to the Mod Squad conversions, not an alternative; I don't see any clear trade-offs between them. There is, however, some hope for a cheaper approach. Steve McCormack and Wayne Emerson are combining forces and putting together a version of the Wayne Group Delay Filter that can fit in many players, thus eliminating the cost of a separate box and power supply. Call the Mod Squad or Wayne Emerson, (619) 443-7229 or 223-8151, for details.
The PS Audio CD Player: Today's Best Stock Player Costs $800 PS Audio seems to be on a roll at the moment. Paul McGowan has taken a regular Philips 2041 CD player and made a major series of changes that result in a CD player that clearly outperforms every unit but the Mod Squad version of the MCD Pro. He corrected a hardness during loud passages that showed up in his first machines, and which robbed those first units of many of the sonic benefits inherent in the technical design.
PS Audio describes the changes to the basic Magnavox 2041 as follows:
• Isolation of the laser servo-mechanism power supply.
• Passive analog prefiltering to eliminate the slewing-induced distortion that PS Audio claims is the cause of hard CD sound.
• A new PC board with discrete analog amplification and filtering.
• A direct-coupled audio output stage with low bass roll-off.
• A discrete high-voltage power supply for the audio amplifier.
• A high-capacitance power supply.
• High quality wires, RCA jacks, and capacitors.
I'm not going to try and evaluate the technical merits of these features, but the net sonic result is a stock CD player with the best apparent resolution of harmonic detail to date, superior low-level resolution, more detailed imaging, superior soundstage width and depth, and excellent long-run listenability.
These "high-end" adjectives mean that the PS Audio player gets a far more natural musicality from massed strings, the upper octaves of the woodwinds, and brass. Percussion has a more natural sound, with less bite and hardness. Voice has more tonal quality and dramatic character, choral music has more depth and coherence, and voices are naturally located without spotlighting.
At the same time, the PS Audio lacks the uneven spectral character of the Mission and Meridian. Its subjective frequency response delivers on the promises of the CD system. The PS Audio pushes the state of the CD art just one notch closer to outperforming the best phono front ends in those areas where they still retain a margin of superiority.
The bass in the PS Audio CD player is as deep as that in the Meridian pro, but much better controlled. It directly rivals the Mod Squad MCD Pro in every respect but the upper midrange, where the tonal balance of the latter seems superior. The PS Audio compensates with superior upper octave sweetness and detail, and just slightly more natural imaging and center-fill than the MCD Pro equipped with the Wayne Group Delay Filter.
This does not mean further progress is not possible. The top-of-the-line Sony players clearly provide more upper octave detail, which may reveal the limits of 14-bit oversampling in the Philips-based machines-or it could be due to Halley's comet. Nevertheless, the Sony’s do better.
Furthermore, PCM tape machines, and analog disc, still have levels of resolution in the upper five octaves missing even from the PS Audio and Mod Squad MCD Pro.
Euphonically or not, the best moving coils still give a better illusion of being near a real soundstage.
That said, the PS Audio is currently the stock machine to beat. It outperforms the Meridian Pro in every respect, at about half the price. It is easy to load, and has good control features-although no remote control or advanced programming features.
Given the fact that the Cambridge keeps disappearing from the market (it was "introduced" at the Chicago CES, but production versions never really showed up in the U.S.), breaking down as a prototype, and escalating in price (a wonderful combination!), the PS Audio is clearly this week's hest buy.
--
[based on a Feb. 1986, Stereophile review article]
Also see:
ADCOM GCD-300 COMPACT DISC PLAYER
Prev. | Next |