With the advent of the CD and general digitization mastering engineers began to raise the loudness of music in the 1990s, hoping that their CD would sound louder than the competition. A foolish idea. With the boosting of the sound signal nothing is basically wrong, every musician sometimes uses a compressor or a limiter. But boosting the signal only to make it sound as loud as possible is a stupid idea. Why? Because it sounds terrible! Music must contain dynamics to keep it lively sounding. If not, it becomes boring and terribly painful for the ears.
Fortunately, measures have now been taken to combat the Loudness War, which has plagued the music world for two decades. With the advent of the EBU R128 Peak to Loudness Ratio, it has gone hard. Virtually all online services are now equipped with an algorithm that ensures that all music is set to the same sound volume. This includes YouTube, Spotify, Apple Music and TIDAL. TIDAL has recently had a detailed investigation carried out into this matter. TIDAL wants to offer the highest possible quality and was created to deliver even higher sound quality than Apple Music and Spotify (non-lossy). But is it then desirable to adjust the volume between songs or albums?
The Dutchman Eelco Grimm, who was also involved in the development of the European standard EBU R128, conducted a study for Tidal. He wanted to measure the loudness of music using the so-called BS1770-4 method. This method not only takes into account the peaks in the signal (since peaks are temporary) but also takes into account the peaks and the average dynamics in the signal. The latter makes it unique because the method takes into account human hearing (certain frequencies we experience as louder even though they are not louder in loudness) and moreover, the dynamics can be measured over a long period of time as well as over a short period of time. This gives us perfect “insight” into the loudness of the signal.
In collaboration with TIDAL, Eelco has examined 4.2 million albums (!) for loudness. This report can be read/downloaded online (PDF). The most important question he asked is: should TIDAL apply volume compensation per song, or should this be applied to an album as a whole? If the latter is the case, an album will always retain the differences in loudness between the individual songs.
Research has shown that the test panel prefers to maintain differences in loudness per album. The album is then compensated in comparison to other albums, but the mutual tracks are not.
The standard Eelco came up with is -14 LUFS. Simply put simply, this means that the average dynamics (LUFS) of an album may lie at -14 dB below 0 (the digital boundary, after which clipping occurs).
One of the important recommendations Eelco TIDAL gave was:
To avoid clipping, only attenuate tracks, never apply positive gain. If the loudest track of an album is softer than target level, all tracks of the album will play soft.
Albums that are therefore loudly masted are significantly reduced in volume. This applies to the vast majority of the 4.2 million albums that were investigated.
Eelco recently spoke extensively about the EBU standardisation and his research for TIDAL with Ian Shepherd in The Mastering Show:
An interesting story Ian brings up herein. Taylor Swift’s latest album is too loudly mastered, just like her previous albums. But since YouTube does not increase the volume just like TIDAL, compensating for the volume will make the softer songs on that album to sound even softer. The same applies to other streaming services.
What about podcasts?
The standardization of music is simply a breath of fresh air. However, there is a problem: you cannot simply apply this standardisation to podcasts, to spoken audio. Speech requires more dynamics than music. Which standard should we therefore apply to podcasts? I expect this to be around the -20 LUFS, but research and international consensus on this subject must finally determine it. This will really happen in the years to come, because the Loudness War has no chance of survival. Whether Taylor Swift wants it or not.
Finally, a picture of how Mr Michael Jackson started to sound harder and harder over the years. It is now history. Loudness is now finally standardised.
Sometimes, playing by the rules is not such a bad idea …
