Have we truly seen the death of dynamic range?
Over the past probably two decades there has been a radical shift in the way records sound when they are finally released. Part of this is attributable to changes in technology. Cassettes faced off with the CD and died. CD’s are facing off against MP3’s and, although currently engaged in a protracted guerilla campaign, are certain to be defeated (for better or for worse). In terms of a change in sound, the first conflict represents the demise of high frequency hiss and frequency ranges of 100Hz to 12000Hz (at best) followed by the rise of theoretical digital black silence and a strict 1 Hz to 22050Hz frequency range and a maximum dynamic range of 96dB. In theory, CD audio technology should have allowed for far greater dynamic range than was ever possible with vinyl records or cassettes. However, since CD’s became commonplace we’ve actually seen a general trend towards less and less dynamic range and commercial releases that sound louder and louder.
What do I mean by all this? Well, firstly, you need to understand the idea and issues of dynamic range. In music, dynamic range refers to the dB difference between the softest audible sound (or silence) and the loudest sound in a recording. As I said earlier, silence on a CD is infact -96dB whilst the loudest sound is 0dB. For analogue recordings the dynamic range was theoretically infinite (depending on the medium) but severely limited by the signal to noise ratio (a measurement in dB of the volume of background noise compared to the volume of the recorded signal), wear on the tape heads, degradation of the recording tape etc. The result was that professional grade recording equipment could produce recordings that were marginally more ‘dynamic’ than CD’s and far more capable of reproducing high frequencies (pro analogue gear even outperforms the 96000Hz sample rate of DVD’s but can’t match their 24bit or 144dB dynamic range). The noise floor of analogue recordings generally sat at -80 to -55dB. The result was that any signal which fell below this threshold would be reproduced at a lower volume than the natural hiss on the tape. Vinyl was even more limiting.
So, how did recording engineers deal with the limitations of the signal to noise ratios whilst maintaining dynamic range? Well, effectively, they cheated. Professional tape has an incredible capacity to record “hot” or overdriven signals. You can drive the volume of a signal going to tape as much as 6dB over the maximum volume that the equipment or tape is created to record without creating excessively unpleasant sounds in the recording – in fact, it’s this very practice which creates the sound that many people call ‘analogue warmth’. Such a practice is not possible with digital audio where 0dB represents a hard limit and any signal above this barrier will be utterly destroyed and result in horrible clipping. Analogue engineers pushed their signal levels discretely to generate a sense of loudness above the actual dynamic range of their medium. This was necessary due to the fact that they were raising the volume of soft sound at the other end of the spectrum so that they could be clearly heard above the background noise (hiss) of the tape. To do this, they used a combination of “fader riding” and compression.
There was a side effect to all this compensating for poor quality recording mediums. Mixes began to grow louder. As recording technology improved in the analogue domain, the capacity to record soft sounds clearly also increased. However, “good” recordings were already characterised as loud recordings for a number of reasons and, instead of increasing the dynamic range, engineers began using their advanced mediums to further push the signal. This was a response to the popularity and growth of radio, particular in cars where background noise (engines etc) would drown out quiet sections of music. Also there was the eventual advent of the cassette and the Walkman (both of which produced poor quality sound that needed to be drowned out by loud music). The earlier flawed medium of vinyl had produced less upward pressure on volume due to the fact that it was physically limited in the amount of signal that it could carry (that is the depth, width and corrugations of the grooves).
The rest is cultural. “Loud” music became “good” music. The average music consumer’s ears became trained to appreciate heavily compressed, limited and overdriven music. Music that allowed significant space in the final mix, with high volume sections and quiet sections, simply didn’t sound “full” or complete. This was reinforced in clubs, pubs and automobiles – each of which adds another round of limiting and compressing to the already mastered sounds.
But what have we lost?
My last tip was about adding space to a mix. I talked about reverb and stereo micing and delays. All of these techniques make use of dynamic range. An echo is an echo because it grows fainter with each repetition, for example. Today, mixing engineers use progressively diminishing cutoff settings to remove the presence of echoes because relying on diminishing volume makes delays difficult to sit in the very hot and driven mixes that they construct. Dance music producers have replaced quiet sections that build to crescendo with filter sweeps – again replacing dynamic movement with frequency manipulation.
Unfortunately, to the trained ear (and, believe me, casual listeners have been trained to pick up on particular audio cues just as much as engineers) it sound good. It has energy and movement – the essential components of good music. What it lacks, though, is subtlety.
And that’s a little sad.