Audioclinic (Jul. 1980)

RIAA Specifications

Q. For the past 1 1/2 years, I have noticed increasingly "tight" RIAA equalization figures on phono sections of preamplifiers (e.g. ±1/4, 1/3 dB, etc). My preamp is ±1 dB. Can these differences really be heard? At what point does RIAA equalization become audible?

-Bob Hoffman, Worth, Ill.

A. I would have to say that under some circumstances you are not likely to hear much difference between a phonograph having a "tight" RIAA spec and one having a "looser" spec, where the so-called loose specification is ±1 dB.

The question really comes down to how much deviation from flatness of response can be detected by the ear.

This is the issue if we are discussing RIAA or any other parameter in a system. Further, it depends on the frequency range where the curve departs from flatness. If it occurs at the high end, where there is very little recorded energy, we will not notice much of a difference with a total shift of 2 dB-which is what ±1 dB can bring about. However, if the most serious departure of the curve takes place at perhaps 3 kHz, even 1 dB can make a consider able difference. This portion of the frequency spectrum is most sensitive to human hearing.

If the departure from flatness occurs at the very low end of the audio spectrum, again a 2-dB change probably won't be noticed--partly because of the lack of hearing sensitivity. If the frequency at which the 2 dB is measured is really low, it may well be that this shift won't matter since most recordings don't contain any significant bass energy at this point.

If you have a graphic equalizer which can be switched in and out, set it for a flat response and then intro duce some small changes in response, at various portions of the spectrum.

For simplicity, introduce just one boost or cut at a time. Note the audible differences when the equalizer is in or out, and, of course, use really wide-range material for the program source.

Editor's Note:

Two dedicated audiophiles associated with the Univ. of Waterloo in Canada, Stanley Lipshitz and John Vanderkooy, recently made strenuous tests under double-blind conditions, and they conclude that as little as 0.1 dB RIAA deviation can be heard reliably. It is well known that detection of such deviations is much easier with steady-state tones than with music, and some tests indicate that at least 3 dB of deviation on music is necessary before it can be detected reliably. What this matter comes down to then, is whether you, the listener, can hear such a difference and, per haps more importantly, whether it bothers you.

Masking of Music

Q. Do you think that most musical information on recordings is masked? Some instruments get covered up unless one listens very carefully with a good pair of headphones.

Should we not put all this "masked" music to good use?

-Bob Hoffman, Worth, Ill.

A. To some extent musical information is always masked. Sit in a live con cert and note the lack of absolute localization of some of the instruments. Some of the notes played may be blurred because of room reverberation; the faster the musical passages being played, the greater this kind of blurring will be.The blurring is very evident when listening to music in large cathedrals.

Pains were always taken by composers whose works were primarily intended for performance in such environments. They had to write so that the listeners would hear a minimum of masking and blurring.

In popular music this masking effect is introduced by the artist or the producer through special effects. These devices are used to heighten the effect, or mood, of the music being re corded and have proved to be beneficial in terms of increased sales of such recordings. A study of a number of albums will disclose that, at times, there is no masking, that all musical instruments do stand out. Indeed, entire al bums are sometimes made in this manner. Sometimes, however, the so called masking is alternately used and then removed, perhaps in the same musical selection, providing yet another means of contrast.

(Audio magazine, Jul. 1980; Joseph Giovanelli )

= = = =