Digital Domain (Dec. 1985, By Ken Pohlmann)

Home | Audio Magazine | Stereo Review magazine | Good Sound | Troubleshooting


Departments | Features | ADs | Equipment | Music/Recordings | History




SHOP TALK


Do hi-fi salesmen give you a pain in the neck? They rank second on my list, close runners-up to used-car salesmen. (After 16 weeks of looking, I still haven't found a good used Corvette-anybody out there know of one?) Anyway, stereo salesmen give me a pain. There are just two types--the dud who spends 15 minutes looking for the on/off switch, and Mr. Techno, the walking glossary of audio gibberish. With the advent of digital audio, the situation has gotten a lot worse. Everywhere you shop for a CD player, you have to endure non-information (e.g., "Wait and ask Biff when he gets back from his lunch hour") or frankly wrong information (e.g., "They're the same as LPs, only the grooves are 60 times smaller").

With Christmas almost upon us, and CD players at the top of practically everybody's list, the situation in the showroom is serious, indeed.

Thus I have compiled a list of effective maneuvers for smart CD shop ping-considerations that will separate the great players from the merely good and, more importantly, communicate to the salesman that you need neither his help nor his hindrance. Don't worry-this text has been specially encrypted so you and I can communicate freely without stereo salesmen listening in. They think they're reading a column on their favorite topic--sales commissions.

Are all CD players alike? The answer is definitely no. First, some players are a lot more expensive than others. Second, some sound better than others.

And the correlation is sometimes the inverse. I recently auditioned a $300 player and a $2,000 player with a group of recording engineers, and we agreed that the less expensive one sounded better. So, exactly how do you shop for a CD player? There are seven important questions to ask

1. Who is the manufacturer? This is the best overall measure of quality in CD players. In my evaluations of players, I have found that certain manufacturers deliver consistent quality, whereas others apparently haven't got the hang of it yet. Do your homework, and ascertain exactly which companies have the required expertise; it isn't always obvious. There are some sleepers out there. Sanyo, for example, is a major manufacturer of Compact Discs and has thoroughly researched CD technology, both in terms of disc manufacture and player design. Their bud get Model CP660 player has sound that rivals the best.

2. How good is the error handling? Discs aren't indestructible, especially as they enter the car- and portable-player environments, and they also get dirty. Unfortunately, consistent specifications for error-handling performance are still not available. Even without a special test disc, you can gauge a player's ability to play through errors without mistracking. Take a strip of opaque, black tape, perhaps 1 mm wide, and lay it radially across a disc. A good player should play through it.

Now try two strips, on opposite sides of the center hole, and see if the disc plays. Keep loading the tape on until the player capitulates. An excellent player will have no trouble with six strips of tape.

Of course, even a normal disc contains errors, and the player's error-correction and error-concealment circuitry must prevent their audibility. Correction strategies yield absolutely correct data, so errors become nonexistent and inaudible. Error concealment simply synthesizes data to the best of the circuit's ability. Because not all optic/servo systems are equal, players have differing success in reading disc data accurately; the poorer the data read from the disc, the more concealment must be used. And concealment circuits aren't equal, either. The bottom line is audible errors. On some players you hear something funny once in a while-not a mute or a click, just badly concealed data, and it sounds weird for an instant. On good players, you don't hear this very often.

3. Does the player have dual D/A converters? When the output data is multiplexed through a single D/A, the audio signal must be alternately switched between the stereo channels-and that switch (another analog stage) can cause distortion. Moreover, an 11-uS difference between left and right channels will be created. Al though this can be compensated for, it is difficult to provide linear compensation across the frequency band; the result is often an interchannel phase difference at high frequencies. Dual D/A converters guarantee zero phase error between channels.

Unfortunately, not everything is sunny in dual D/A land. A pair of converters might mistrack due to fluctuations in their reference voltages or to anomalies in the chips themselves; the result could, again, be phase error. Sooner or later, manufacturers will solve both problems by offering dual D/A converters on a single chip. Until then, the evidence weighs in favor of dual D/A converters.

4. Does the player have digital out put filtering (oversampling)? In theory, there should be no audible difference between analog and digital output filters. In practice, the audible difference is slight, but significant. Primarily be cause it is much easier to design a good digital filter than a good analog filter, the former technique tends to de liver consistently good sound. In addition, a digital filter is contained on a chip, whereas an analog filter is comprised of many discrete components.

Thus, the digital design yields greater reliability and long-term stability, and lower power consumption.

Also, oversampling is considerably cheaper to produce. That means the manufacturer can invest some of the savings in other facets of the player, beefing up subsystems such as the audio output circuitry, including the gentle low-pass filter necessary to re move extreme high-frequency components present at the output of a digital filter. Of course, it also means the manufacturer might have simply beefed up his profit margin.

5. How does the THD measure up? Specified THD and measured THD sometimes are worlds apart. The problem is supersonic artifacts (at least above 21 kHz). "Beat" frequency components caused by clock interactions, and other high-frequency junk, can in crease in-band measurements of THD.

Therefore, some manufacturers measure THD with a 21-kHz "brick-wall" filter between the player's output and the distortion analyzer's input. This guarantees that the supersonic junk does not raise THD.

But is that really kosher? If you're measuring harmonic distortion at 20 kHz, and you put in a 21-kHz filter, of course you're going to get a great reading. Besides, some players don't have any of that junk; is the use of filters fair to them? For example, with a low-pass filter inserted, I recently measured a THD of 0.0022% at 1 kHz and 0 dB on a Sony CDP-102, and THD of 0.0030% on another big-name player.

Without the filter, the Sony measured 0.0029% and the other player measured 0.35%. Sure, the junk is super sonic, but I'd prefer it to be absent.

That's where Audio comes in handy: when Len Feldman measures THD, he is careful to state the conditions of the measurement and the presence of any supersonic components.

When evaluating Compact Disc players, look for a single master clock, as opposed to several unsynchronized clocks; this tends to reduce supersonic junk and provide good THD readings-as measured the old-fashioned, unfiltered way.

6. Are there vibration problems? CD listeners tend to play their discs at louder levels than others, and there is augmented bass response. This can lead to airborne and structure-borne vibrations from loudspeakers, which tend to cause vibration in discs which are not tightly clamped. As the disc flutters, the pickup servos cannot keep track, the error-concealment circuits cannot cope and the result is added distortion at loud listening levels.

(That's why users of some players have learned to put in two discs simultaneously for better gripping.) Checking for this problem in stereo showrooms is one of my favorite pas times. If you crank up the amplifier level, the problem may occur, but it might not audibly manifest itself be cause of amplifier or loudspeaker (or room) overload masking the problem.

You have to crank up the level, then listen with closed-ear headphones.

This is guaranteed to drive salesmen crazy. But listen closely: On some players you'll hear a deterioration of sound quality as the error-concealment circuits go into overdrive. It's difficult to describe the effect, but you'll know it when you hear it, if you keep an open mind-remembering that it's something you've never heard before in analog equipment.

7. Does the player sound good? This question is tougher to answer than you might think. In general, all CD players sound okay. A showroom has always been a tough place to listen to anything (except the salesman's pitch). With digital audio, it's even tougher. The trick is to concentrate on one aspect of sound quality at a time.

Listen for badly corrected errors. Listen for stable localization. Listen for high-volume disc instability and for full ambience. Listen for naturally warm sound. It isn't easy, and it takes time.

Sometimes it takes me days of com parison before I can reliably choose the better player. Of course, you might have a better pair of ears. And as all of us become more familiar with what we're listening for, we'll get better at it.

Okay, I'm already over my word quota. Go out there and hit those show rooms, holiday shoppers. And watch out, you salesmen!

(adapted from Audio magazine, Dec. 1985)

= = = =

Prev. | Next

Top of Page    Home

Updated: Friday, 2019-05-10 16:42 PST