February 2023

Limitless Musical Pleasures
Roger Skoff writes about technology and music.
Article By Roger Skoff

  When performers or recording engineers talk about "cutting a track", those words can be the literal truth.

Making a phonograph record – the first kind of sound recording – has always meant using some kind of a sharp object (most recently a heated cutting stylus), to cut a groove into the surface of some suitable material, and to wiggle that groove right, left, up, down, or, for stereo, in all of those ways, diagonally at the same time, for the purpose of recording sound.

The recording material has ranged from Edison's beeswax to a special lacquer, and the finished recording has been everything from a wax cylinder through, currently, a heat-stamped vinyl disc, but the process, itself, has always remained essentially the same.

It's always been to cut a groove into something, to modulate that groove analogously to the changes in air pressure at the recording microphone, and then to use the wiggles in the groove to force a playback stylus to move in that same pattern of changes, either to directly drive a diaphragm to move the air in a listener's room (a la the old Victrola acoustic horns) or to create a voltage or current to drive an amplifier to drive a speaker to do the same thing.

In any of those cases, the amount of excursion of the groove (or the diaphragm) from dead-center reflects the loudness of the signal, with louder (or lower frequency) sounds calling for greater movement and higher frequencies or quieter sounds calling for less.

And that's the first problem. With the grooves (actually one single spiral groove) evenly spaced, if the sound is too loud or too low in frequency (or both), the swing (lateral movement) of the groove can be so great as to cause adjacent groove sections to touch, or even overlap, and that can make for major tracking problems. It can even, if it's sufficiently extreme, cause the playback stylus to jump out of the groove, creating a "skip" or even ending playback entirely.

Part of that potential problem was dealt with, either by design or incidentally, when record cutting/playback equalization (originally varying by company, but now standardized as the RIAA curve) was instituted early in the development of commercial recording. The above graphic given you both the 'recording' and playback RIAA EQ curves.

Early record-cutting equipment operated at a constant velocity – meaning that the amount of stylus travel per unit of time was always the same, regardless of the frequency to be recorded. The result of that was that bass frequencies produced more groove swing – a louder recorded volume – at any cutting level and treble frequencies produces less, which, on playback, both resulted in an unnatural frequency balance and naturally contributed to the possibility of groove overlap. The solution was easy; just reduce ("limit") the amount of bass going to the cutting head and boost (provide pre-emphasis for) the amount of treble. This was known as an "equalization curve", and it served several important purposes at once: By reducing bass groove swing, it helped to solve the overlap problem.  

Second, less groove swing meant that the groove spiral could be tighter (less space between adjacent sections), which meant that more music could be recorded in any given space. Third, boosting the treble in recording helped to "drown-out" "surface noise" (ticks, pops, and "groove hiss") and make for greater listening enjoyment. And, as a fourth sonic benefit, it worked near perfectly with the Rochelle Salts "crystal", or "ceramic" phono cartridges of the day to produce a more natural tonal balance without the need for playback equalization. (The very rolled-off frequency response of the typical crystal cartridge naturally countered the boosted treble of equalized recordings, and, for its time, sounded just fine – without the need for any playback treble de-emphasis at all!)

That worked to a substantial degree, but other things could still make for problems: One was simply the natural dynamic range of music, which can go from dead silence to a full orchestral tutti in an instant, and which – unless the recording gain was constantly "ridden" by hand or until the automatic limiter and "variable-pitch" recording (groove spacing that changed automatically with the volume of the signal to be recorded), were invented – could still result in distortion or groove-skipping.

Another thing was AM radio broadcasting. Even today, most new popular music is first heard by people on the radio and AM radio was the first type of music broadcasting that came along. ABove shows the listening of various types of media back in 2017, which has changed quite a bit in only the past five years.

Unlike FM radio, which works be changing (modulating) the frequency of a carrier signal to transmit the music, AM radio works by changing its power level – modulating the amplitude of the carrier wave. What that means in practical terms is that as the level of the music gets louder and softer, the transmitted power level gets higher and lower, resulting in the station's broadcast coverage area – and the potential number of people who can receive it and hear the music it's playing – getting larger and smaller, as well.

Because the number of people listening is the station's basis for its advertising charges – its primary revenue source – and because the more people listening to a piece of music, the more people are likely to buy it, both AM stations and popular music artists and producers have, for more than the last hundred years wanted their music played as loud as possible.

One obvious way to do that is to increase the total power of the radio station's transmitter, but that's subject to FCC (Federal Communications Commission) rules. Another way is to increase the average power output of the transmitter so that it's operating as close as possible to its maximum output level at all times. To do that, more signal processing was added, and not only were music's loudest sounds compressed toward the average level but its quietest sounds were expanded so that the volume level transmitted could always be as near to the limit as possible, maximum broadcast coverage could be maintained, and the largest potential audience would be assured.

That satisfied nearly everybody – except for music lovers and audiophiles. A very important part of the vividness and realism of a recording lies in its dynamics. The crash of a cymbal and the "kick" of a kickdrum are thrilling parts of the musical experience and are major reasons why so many people love horn loudspeakers.

Now, though, combining digital recording, the computer, and the internet, more than 60,000 radio stations can present your favorite music on-line, around the world, so the old reasons for signal compression or expansion simply no longer apply, and artists, engineers, and producers can concentrate on their art and on making recordings that don't have to compromise their dynamics in order to ensure the largest possible audience. Hopefully, a new kind of radio will result in new, un-limited, higher fidelity, more dynamic recordings, and even more opportunity for us to...  

 

Enjoy the music!

 

Roger Skoff