??...how do you mean?

I'd say it quite often is overused.

1) Compression is commonly used on a mix to make it "louder".
(and "louder" is thought to be "better"). In situations where
the listener controls the volume to their preference this would seem to be pointless.

2) People trying to mix a "hit record" often compare their mix
to the sound of a popular radio station. The radio station uses compression (usually 5 band). So the track being mixed gets compressed to match the radio. When that track itself gets airplay it will be compressed again by the station, so it gets double treatment. If the new track is playing when someone is mastering their latest creation, and they happen to use that station for comparison...............

3) As I actually enjoy the expressive variations in volume which occur when someone plays an instrument, I don't usually hear compression as an improvement.

andy butler

Frequently, yes, compression is overused.

Sometime in the last year or so, an article came up on Slashdot.org about a recent Rush album being overcompressed, or limited, actually, such that the drum peaks were missing the "oomph" that they would have had were they left natural.

Not a new phenomenon, either. I am a fan of 80's music (mostly because I came of age in that decade) and have found that many of my albums (yes, I mean vinyl) sound a LOT better if played through an expander. It doesn't have to be a wild, off-the wall expander, just the playback section of a standalone Dolby B processor will do it.

Interestingly enough, the CD re-releases of many of these sound fine, but this is not universal.

As for broadcasting, good compression and limitation is kind of necessary there. You are working with source material that has a much greater dynamic range than the medium into which you are trying to shoehorn it, and you need to make the most of it. By using 5-band compression, the tonal inequalities from recording to recording go away.

For AM stations, especially talk stations, I am willing to forgive a great deal of excess compression, because there is a direct correlation between the RMS of the input audio and the envelope of the RF output of the transmitter. In other words, more audio power means more RF power. This can be even further enhanced by judicious use of negative peak limiting.

However, I think that compression is less important for FM than AM stations, yet many FM stations just *WAY* overdo it. FM is a near high fidelity signal format. It boasts double to triple the frequency response and typically a 15-20dB better S/N ratio than AM. As such, the compression could be peeled back a bit.

Only four FM radio stations in my region seem to understand that: WRPI, a college station (engineering college, go figure), WAMC, an NPR station, WMHT, a classical music station, and WEQX, an independent rock station.