Compression is glue

My students are afraid of using compression as a recording tool. They’ll use every plug-in under the sun when mixing—and get good results too—but they won’t touch the real thing.

I put this down to one of two things:

  1. Extreme caution: if we mess up while recording, it will be harder to fix; or
  2. We’ll fix it in the mix.

The first is certainly true. But aren’t you using your ears and listening to how it sounds while you are recording?

The second might be true, but only if the performance is not messed up, the mic is reasonable and positioned properly, and the rest of the signal chain is OK too. (And you don’t need a Wagner U47 and a Forsell mic pre!)

Compression is like glue. It sticks sound to the listeners’ ears, whether the sound is a kick drum anchoring the beat or an intimate vocal. In modern—and not so modern—pop music, you’ve got to use it. It’s part of the sound.

The history of recording is the history of compression. In the early days, we fought against the limitations of the medium. Too low a recording level and the signal would be swimming in a sea of noise made worse by our having to turn it up to hear it. Too high a recording level and distortion would result. Sometimes that distortion might add a nice edge but if the signal got too loud it’d turn nasty.

Or, if the recording was transferred to a new medium, compression was required to fit it to a new set of limitations. Vinyl is the perfect example: if the signal is too big, the groove swings to wide and the needle skates across the disc.

In the old days, we had a duty not to scratch our customers’ records.

Even radio needed compression, to smooth voices and raise the sound above the interference so easily coming between the transmitter and the receiver. (Have you listened to AM radio lately? It still sounds crap.)

For this reason, some of the best compressors are old broadcast units… Pye, AWA, Federal, Gates and that really rare stuff built for the U.S. military. That’s right… while the troops were marching their radios were pushing sound into the ears of the cowed civilians.

It wasn’t too long before engineers realised compression could be used as an effect on individual sounds, especially as rock became more common. The drums had to punch through, the bass chug along and the guitars blast. And while all that was going, the vocals had to be squeezed into the small space left at the top of the mix.

Our ears don’t respond to peaks, but averages. They’re like VU meters. So a lot of the time you want to set your compression to leave the peaks alone (a longer attack time preserves the brightness of that transient too) and squash the body of the sound. Boost that and it’ll stick to the ear, just the way you want it to. When you come to mix, everything will sit in a tighter dynamic and you won’t have to work so hard to make that kick drum slam and those vocals heard.

(OK, so there’s a little more to it than that, but you get the general idea.)

Some compressors aren’t hard to use at all. Want a kick sound? If the drum is not too badly tuned, dial in 3-4 dB of gain reduction on a dbx 160 and feel it! Tickle the vocals with an 1176.

It’s possible to take it too far of course, especially in mastering. As long as we have FM sound, we’ll have broadcast limiters that will destroy your sound if it threatens to overmodulate the carrier. And even with digital, how’s your mix going to play with other material when the dialog normalisation level for TV is -20 dBfs or lower?

Incidentally, the use of such limiters in the broadcast signal chain and the very nature of our hearing (averages, not peaks) render this report from A Current Affair completely bogus.

So the moral of the story is: don’t be afraid of compression. Make it your friend.

0 Responses to Compression is glue

  1. There are currently no comments.

Leave a Reply


A vanity publishing venture of David Rodger, sound production teacher and wannabe PHP developer