Normalisation - The end of the loudness war ?
Over the last two decades there has been a growing debate among audio engineers that music levels are becoming exceedingly loud. Sometimes to the point where it is creating a negative impact on the product.
A working example of the increase in loudness and the decrease in dynamic range can be seen below.
The above picture is a waveform of a Bryan Adams track recorded in 1983. As you can see there is a large amount of headroom between the programme and 0 DBFS (decibels relative to full scale). The peak of this programme occurs at -2.5dB, meaning that this track can be increased by the listener by 2.5 db before any loss in audio quality can be perceived.
Fast forward to the year 2000 and here is a waveform from an album by the pop band Westlife. It is immediately obvious that this is significantly more compressed and louder than the song by Bryan Adams, and illustrates clearly the rise in audio levels from the 80′s up until the 20th century. Another item to consider is that the waveform above contains evidence of clipping when you zoom in on the peaks. With this in mind it seems as though there has been a trade off between audio quality in order to raise the volume of the track.
Why is it that music has risen to such a loud level?
During the 90′s there was competition between record labels and music producers in order to create the loudest sounding music. The reasoning behind this can be attributed to promotional purposes. If a song appeared on the radio that was louder than another, the chances are it would stand out to the listener and if the assumption is that ‘louder is better’ the other songs played on the radio would sound inferior to the louder track. Take a look at the graph below showing the gradual rise of loudness over the years.
The above video details a number of mastering and mix engineers views on the subject.
What’s being done to ease the trend of rising music levels?
Bob Katz, author of ‘Mastering Audio: The art and the science’ recently stated on his website that ‘the loudness wars’, as it has been dubbed, is over. This is down to the new algorithm being implemented by iTunes Radio that effectively levels all of the music being played so that it plays back at the same volume. This ‘leveller’ is something called Soundcheck. And it has been around for a number of years in iTunes as a tick box that the user can enable. However, with iTunes Radio, the user has no option as the the algorithm is merged into the radio signal. This means that the listener can change between each radio station and various genre’s of music and the loudness level will appear constant. It also seems that youtube has implemented a similar normalisation technique as described by mastering engineer Ian Shepherd in an articles linked below.
http://productionadvice.co.uk/youtube-loudness/
http://productionadvice.co.uk/youtube-loudness-normalisation-details/
Below is a video showing the effects of youtube loudness normalisation on an audio track.
So with arguably the two biggest players in terms of music distribution both using a similar method in order to control music loudness, it seems there may no longer be an advantage to compressing and limiting your music so that it reaches such a loud volume level.