Last time I talked about why records made in the seventies sounded so goooooood. Punchy. Danceable. And also, why the music started to suck. One factor… the widespread use of ‘compression’.
But there has been a second compression scandal which began in the early 90’s and continues to this day. This is commonly known as The Loudness Wars.
Now when CDs first came out, they were mastered a lot like vinyl. Sure the dynamic range is much greater, but the actual perceived volume of the CDs was quite similar to good ol’ records.
A funny thing started happening though. Like most drug habits, it was subtle at first, but over time, it turned into a monster that ruins lives. Little by little, CDs (the new ‘records’) became louder and louder.
Now technically, they didn’t get ‘louder’. They can’t really get louder. The strongest signal a CD can handle is exactly the same loudness as the strongest signal a record player can. But what can and does happen with digital tools is that the dynamic range can far more easily get squeezed. Once you can get a signal into a computer there are all sorts of things one can do to make the difference between the softest and LOUDEST signal smaller and smaller.
For a silly example, try doing a ‘stage whisper’, ie. you’re whispering, but really as loudly as if you were speaking in full voice. Psychologically we still perceive this as ‘speaking softly’—even though we’re not. But what does happen if someone keeps speaking in a stage whisper for more than a few seconds? We get tired of listening to it!
What compressors can do is squeeze a signal so hard that even the softest signals have almost the same true loudness as the loudest. With enough compression, that lone flute looks as loud on an oscilloscope as the screaming electric guitar. For short periods, we perceive the flute as ‘soft’ and the guitar as ‘loud’. But very quickly our ears become fatigued… something (we can’t quite put our finger on it) tells us that this is not pleasant–and not very natural.
Now, when you’re out on the dance floor? Who cares? The louder the better, right? So dance records were the first to succumb to super-compression. After all, not too many flute solos at the disco, right? Heavy metal? Also a likely candidate! OK, so certain genres were easily swept up. But then another funny human phenomenon came into play.
For short periods, humans will almost always perceive a louder version of a song as ‘better’. In blind listening tests, if you give people two mixes of a song and change nothing except that one is 1db louder than the other, almost everyone will prefer the louder mix.
OK, you’re in your car and you listen to a Metallica record and it’s super-compressed and it’s LOUD coming out of your speakers. Then your mood changes and you put on a jazz record. If it’s older, and was mastered before The Loudness Wars, you will almost invariably have to reach for the volume knob as it will sound much too quiet after the twin-guitar attack. Remember that’s not because the instruments on the jazz CD are ‘quieter’. It’s mainly because the dynamic range on the jazz CD is so much greater. Same with older rock records… they will sound much too quiet when compared with a ‘modern’ record.
Now take two records by similar artists in any genre. Remember what we just learned about psycho-acoustics; people prefer the louder of two similar signals. Armed with that information, I as a record executive can give my ‘product’ a competitive edge by mastering it as loudly as the Metallica record. I can compress the snot out of the alt-acoustic-indie-singer-songwriter so that the apparent volume of her voice is as loud as James Hetfield. And I guarantee that, all things being equal, you’ll prefer her sound (at least for a short period) to her less processed peer.
So what’s the problem? Fatigue. After a very short time, the ear gets physically tired with the constant level. It starts to realise that it’s being tricked and tends to not process the music on as deep a level. Basically, the brain tunes out much of the ‘content’ in favour of just ‘the mad bass’ (the beat.)