IN YOUR EAR
The Sound of Cushions The first stereo headphones were circumaural types, which covered and surrounded the ear and were sealed against the side of the head by cushions. Portable tape players popularized supraaural types, which rested on the outside of the ear. Now, insert earphones, which plug right into the ear, are becoming popular too.
Which of these three types can produce the highest fidelity? A recent letter to the Journal of the Acoustical Society of America (Vol. 83. No. 4, April 1988) offers some suggestions.
The six authors (Jozef Zwislocki, Barbara Kruger, James D. Miller, Arthur F. Niemoeller, Edgar A. Shaw, and Gerald Studebaker) were actually testing to determine which earphone classes could be recommended as standard for audiometric testing. They point out that for all 'phones, sound pressure at lower frequencies "is essentially the same at all places within the ear as long as the entire system is airtight. At higher frequencies, air leaks are less important, but the sound pressure produced at the ear drum is critically dependent on the wave properties of the earphone and the external ear.
The geometry of the cavities coupling the earphone to the ear drum, which is affected by such factors as the positioning of the earphone on the head of the listener, becomes the most critical issue. These wave effects become important ... at frequencies above about 2,000 Hz." Traditional, circumaural phones were less leaky, and therefore more reliable at low frequencies than supraaural phones. They also minimized noise from both physiological and ambient sources.
The authors go .on to point out, however, that at high frequencies, wave effects in the ear cavity make sound pressure at the ear drum "critically dependent on the geometry of the earphone and ear and the exact positioning of the earphone." Supraaural phones are less reliable at low frequencies because the coupling between them and the ear is variable and unstable. This causes unpredictable amounts of loss below about 500 Hz, plus small, variable sound-pressure enhancements between 500 Hz and 1 kHz. As with circumaural types, high-frequency response will vary with positioning.
Insert earphones are the least leaky, and therefore the most reliable at low frequencies. They also minimize some of the high-frequency wave effects, and some physiological noise. (In my experience, they also minimize ambient noise fairly well.) Exact performance, however, will vary considerably with the shape of the individual ear and the insertion depth.
These feelings apply only to types of 'phones. A good model of any given type should outperform a poorer model of another type, and any given model may still perform better on your ears than on someone else's.
The Wishing Hour
Now is the time to start thinking about what tomorrow's preamps, with full digital control and signal processing, might be able to do for us. If we air our ideas now, they might influence first-generation designs; wait until later, and designers' ideas might become circumscribed by prematurely accepted conventions.
Write and let me know your ideas on the subject, and we'll publish some.
Meanwhile, here are a few of my own first thoughts, to get things started.
Digital preamps should have MIDI control interfaces and serial interfaces for control by home computers. A control-program language should be written which all preamps could use, as an aid to users and designers alike. I have no idea what these interfaces will allow us to do, but I'm sure that other users will think of interesting applications for them.
Tone and EQ controls might be tied in to a graphic display and a light pen, so that one could draw or edit control curves on the screen. (I saw that implemented in Technics' Osaka labs about 10 years ago.) It should be possible to tie in these control circuits with acoustic analyzers, to set up accurate curves for given rooms and listening positions.
I'd like my digital preamp to have a phono stage with just such advanced EQ facilities. By playing a test record, one could calibrate a basic EQ curve for every cartridge in the house. Then one could modify that curve to personal taste, and call up further modifications to match equalizations used on records made before the RIAA curve became standard. In my system, it would be nice if calling up the 78-rpm curve automatically switched the input to my Dual CS 5000, the one turntable I have which can still play the darned things.
I'll print more such blue-sky ideas when I come up with them--or when you do.
Reinforcing Prejudices
My prejudice for classical concerts over pop, nowadays, has less to do with my taste in music than with my taste in sound. Pop concerts use sound-reinforcement systems;
classical concerts use acoustics.
In a hall with good acoustics, while the sound does change from place to place (e.g.; sharper and better focused in the front of the orchestra, more resonant and blended in the upper balconies), the basic sound is recognizably the same no matter where you sit. Carnegie Hall is always Carnegie Hall, and never Alice Tully Hall or Philadelphia's Academy of Music. Acoustics color the sound (which is why some halls sound warmer or colder than others, among other things), but the coloration of a decent hall is fairly subtle. Further, acoustic listening is free of distortion, until the sound gets so loud that your ears overload. (If you don't think acoustic instruments can get that loud, listen to an unamplified big jazz band!) Sound systems, all too often, are expensive ways to simulate bad acoustics. It's hard, perhaps impossible, to create a speaker array that covers a wide frequency range, has smooth and even frequency response, and delivers roughly the same amplitude and frequency response at every seat in a vast area.
The people who install and run such systems try to make the sound as intrinsically smooth and natural as possible, then they use equalizers to get it still more natural and smoother.
The problem with this approach is that equalizing for perfect response at one point does not make the sound perfect everywhere--or, sometimes, anywhere--else.
In the past few months, I've run into two examples of this: Linda Ronstadt's appearance at New York's City Center, and Harry Belafonte's appearance at the Garden State Arts Center, in Holmdel, New Jersey. City Center is an auditorium with pretty good acoustics (the New York City Opera used to perform there); Garden State is an amphitheater which seats more than 5,000 under its roof, and can hold about 5,000 more on its sloping green lawn in good weather.
In both cases, the sound in my seats was shrill with a hole in the midrange. Ronstadt's sound man told me the system had been equalized (and it sounded good, indeed, at the sound booth), but he altered the EQ to get more midrange and less treble elsewhere in the hall. At the Belafonte concert, I could see the sound man's real-time analyzer displaying a frequency response which would do 'any high-end audio system pretty proud. But where I sat, about 30 seats to the left and 30 rows further up, the sound was screechy, and sibilants were overemphasized. I stood by another sound booth, a few rows ahead of me and about 50 seats further left, and the sound was clear and natural.
Alas, these concerts probably could not have existed without sound systems. Ronstadt might have been able to fill City Center without it (other singers have), but I'm sure her tour took her to places where the acoustics weren't as good-and even at City Center, night after night of loud singing wouldn't be good for the voice. (In opera houses, casts or whole operas rotate, so the principal singers can get some rest.) I doubt that any singer could fill Garden State without amps.
Dampfinger
Trying to protect musical royalties by holding back DAT is a lot like trying to block a rising sea by plugging a leaky dike with one's finger. Proscribing DAT by name won't prevent the arrival of other, equally good digital recording systems. Proscribing digital recording systems per se would prove difficult because such systems are already here, in the form of 8-mm VCRs with digital sound, telephone answering machines that record your outgoing message on a digital chip, and so on.
Their quality may be lower than DAT's, but they are still digital.
Proscribing the use or sale of recording systems that exceed a given quality level would cause severe public-relations problems for the recording industry.
Quotes from Keith Audio engineers rarely wax poetic, but Keith Johnson did when introducing the speakers he designed for Precise Acoustical Laboratories. I thought these two quotes were worth passing on:
"One starts with the brutality of science, and ends up with art.
"Once you have discovered what's wrong with a driver, you can never listen to it again."
(adapted from Audio magazine, oct. 1988 )
Also see:
= = = =