I hate music technology! – Part 1 – Compression

Okay, that’s not strictly true, but there are elements of it that drive me up the wall. Music tech has made music what it is today; vibrant, varied, high quality and accessible, and for that I am grateful. However, it has lead to several practices within music – especially mainstream pop that make me want to seal up my ears with concrete.

Let’s get the obvious one out of the way first; compression. For those who aren’t familiar with compression, it is a process that alters the dynamics, or loudness, of a recording by reducing the volume when the level goes above a certain point, turning a recording with lots of volume changes into one that is a lot smoother.

Sounds useful, right? Well it can be, and indeed in many cases it is essential. It is often required to even out the levels of vocal recordings, and when used correctly on a bass guitar you get that fantastic thunky bass sound. Even compression and limiting at mastering can be useful – but in moderation!

Looking at the amplitude envelope (a visual representation of the amplitude of a waveform over time) of a couple of songs that I happen to have a couple of versions of it is easy to spot the increase in volume. The first song (How Soon is Now by The Smiths) is noticeably louder in the second instance, and if you look at the right channel in the second instance (lower waveform) you can see that most of the dynamic range has been squeezed out of the track. It is also worth noting that the envelope of the remastered recording is like a big block, all of the peaks have been trimmed down to maximise the overall volume without causing distortion.

The Smiths - How Soon is Now (Amplitude Envelope Comparison)

The second song (Tom Sawyer by Rush) doesn’t suffer this loss of dynamics because it was fairly constant to begin with, but notice again that the peaks are being cut in order to squeeze the track under the 0dB threshold.*

Rush - Tom Sawyer (Amplitude Envelope Comparison)

So what’s the problem? Louder is better, right?

By compressing the music this much you no longer get the contrast between loud and quiet, a musical device that composers have used for centuries in order to give life to a piece. By removing those dynamics the loud sections lose their power, because they’re not much louder than the quiet sections. Granted, in electronica and hiphop overcompression is a big part of the sound, but this culture of remastering albums and sucking the dynamic range out of them just to boost the volume seems a tad unnecessary to me. Also, after listening to heavily compressed music for a while your ears begin to get tired. This will happen anyway, but it sets in a lot faster with heavily compressed music.

Compression is an important tool for producers, but it’s a tool that has been abused. I don’t hate compression, but the way that older recordings are being remastered and given the brick wall treatment makes me think twice about buying a remastered recording. Compression may solve many problems encountered in the studio, but the way that recordings are being squashed and squeezed has sucked the soul out of some of my favourite recordings, and that makes me hate it.

Today (25th March 2011) is the second annual Dynamic Range Day, set up to highlight the effects of the loudness war, and to try to educate musicians and engineers about the dangers of overcompression. Their message is simple - music with dynamic range sounds better. The website has plenty of information about the loudness war so it's well worth checking out.

Next time I’ll be looking at what I call preset culture, and how the sound of pop music has been affected by the advances in synthesizer technology.

*In digital recording the highest level you can record before encountering clipping is 0dB. The minimum depends on the bit depth of the recording, but for CDs the standard is approximately -96 dB, and on commercial recording systems using a sample size of 24 bit the range is roughly -144 dB. The higher the bit depth the greater the dynamic range.