Compression: Essential Audio Tool or Deadly Sin?

Those of you who do not work in the audio field probably suspect correctly that many things are done behind the scenes in the studio to the recorded music we hear. People are aware that music is layered and multi-tracked, and that some pretty fancy editing and splicing are going on, much as they do in video. However, there is another type of audio manipulation known as compression that is unknown to the public, that is becoming rampant, almost reaching a point now where it is mandatory for something to be compressed to be considered "commercial" or even "professional."

Compression is the process of electronically narrowing the dynamic range, or volume, of music once it has been converted into an electrical signal. (A related process known as "limiting" only compresses the upper end to prevent loud peaks. Compression also boosts the quite parts.) Some types of audio equipment, such as guitar pickups, telephones, and automatically do this by nature of their inability to transmit a very wide dynamic range. I call this involuntary compression. The history of the recording and audio industry has been a steady widening of the dynamic range of speakers, microphones and recording devices, supposedly striving to capture real live music in all its dynamic glory. Now, largely as a by-product of the commercialism of recorded music, it is seen as desirable to reverse this process, and to make the music as compressed as possible. Voluntary compression.

Why? If you make a tape of a song, you have to set the levels on the tape recorder based on the loudest part of the song. If there is one part that is really loud, then the overall level of the whole song must be reduced to keep the loud part from red-lining and distorting. If the song's dynamics are squashed so there are no loud parts, then the whole song can be loud. In the earlier days of the vinyl industry, the challenge was to find a way for one record to be louder then another on a jukebox, so it would get attention and sell more. Incidentally, as a result of some of the "my record is louder than yours" wars of the past, all radio stations are now required by law in the U.S. to compress everything they broadcast, as a way to prevent some music from sounding louder than others. For this reason, listening to classical music or jazz on the radio is sort of pointless, since that is the music that is supposed to have some excitement and dynamic range. It is also part of the reason why radio stations have never played many cassettes, because the tape hiss in the quiet parts and between the songs gets boosted.
Compression makes music more palatable, like white bread. It is less likely to startle anyone if it makes a steady drone like a television (which is also very compressed, by the way) and it makes for better audio wallpaper. The most sought-after producers and mastering engineers use compression endlessly in what they do, and when I read the audio magazines, it is THE major topic of discussion. It seems to me that it is pointless to use the finest mikes and recording gear in the world to record real, living music, and then squash it flat as a pancake before it is released to the public. There is a growing battle between the commercial music establishment and the so called "audiophile" market, which consists of those who have bought stereo systems and speakers that are capable of reproducing a wide dynamic range, and who now want to hear some music that showcases what their gear can handle.

As a musician, there is a non-stop pressure on me to compress my recordings. Though it is an artificial and unnatural alteration of the music, it does have its advantages. If you listen to music in the background at a low level, then when it is compressed, there will not be loud parts jumping out and startling you, nor will there be quiet parts that fall out and become inaudible. If you are driving in a car, you usually have a more limited dynamic range and a lot of background road noise, and in order to hear the quiet parts you have to risk having the loud parts distort your speakers. If you make mix tapes where you combine commercial songs with things like folk or bluegrass, you will notice that the "commercial" cuts are louder. If you play these mix tapes in a place with ambient noise, like a restaurant, the louder, compressed songs will jump out and be noticed. Play a sympnony in a restaurant and you will not even hear the quiet parts, and every few minutes there will be a huge volume swell that will jump out and then vanish when it quiets down. This is why things are compressed. If you are listening on headphones or in a situation where you are really focused on the music, the compression makes it less interesting and exciting, and you want the dynamic range.

It is almost necessary for recording to be released in 2 versions, compressed and un-compressed, letting the listener can decide if they want the real thing or not. This will of course never happen. Compressors are quite expensive, or else it would be easy to just install one in your audio system at home if you want and solve the whole problem of compression. It would be great if every restaurant had a compressor running so that the CD's they have in shuffle-play mode will all be the same volume.
I can't advise anyone of how to deal with the issue-- it is a tough personal decision, and I only write this to help people be aware of the issues and the choices. In my own recordings, I usually do "spot compression" where the volume of the loudest peaks of the music is lowered (just for the duration of the note or word) in the mastering process to allow the overall level of the song to be a bit higher, and to eliminate excessive volume bumps. I find this to be a compromise, and allows the music to have its normal dynamic life. But I do not mix my recordings for commercial airplay, and I have this luxury.

READ THE ARTICLE ON REMASTERING WHICH IS RELATED