The loudness wars have been raging in audio since the late 1990s but now, after two decades, there seems to be the possibility that they may at last be over.
Whereas until recently it was a major function of mastering to make a track seem as loud as possible, soon there may be little point in doing that.
Where do people listen to music? Spotify, Apple Music, iTunes, YouTube, television and radio. Oh, and CD too (and even vinyl, but for technical reasons things are a lot different on that medium).
But Spotify, Apple Music, iTunes and YouTube all either control subjective loudness levels or offer listeners the ability to set all songs to around the same subjective level. In television, loudness control is a legal requirement both in the USA and EU. Radio hasn't quite got its act together yet, but it will come.
So if a mastering engineer really maxes out a song so that it will be the loudest ever on CD, it won't be any louder on loudness-controlled services. And it will sound dull and lifeless compared to a song that has been allowed to retain its dynamic range.
Will producers and mastering engineers conspire to get around the new loudness rules? Here is an example where two songs that have been normalized to the same loudness seem to sound slightly different. Loudness of course is a subjective phenomenon so the comparison will sound different to different people, or maybe the same near enough. Perhaps there are better songs to use as examples? Or maybe you can find a way of bucking the system.
Here's the video…
Transcript of video…
I'd like to talk about what could potentially be the new battlefield in the loudness war.
You'll need to understand what the loudness war is for this video to make sense. You'll also need to know about Loudness Units relative to Full Scale, commonly known as LUFS or you can just say LUFS if you like.
Here I have two tracks – Hallowed Be They Name by Iron Maiden and Uptown Funk by Mark Ronson and Bruno Mars. I'm going to measure their LUFS using the Waves Loudness Meter Plus, or WLM Plus for short.
LUFS measurements should be taken over the full duration of the track, so I'm going to play them all the way through. To make things faster, I can measure both tracks at the same time.
What I can do now is normalize both tracks to a target value of -16 LUFS, which is a commonly-used standard. I'll also set the true peak limiter to restrict the true peak level to -1 dBTP. This will help to prevent any clips due to intersample peaks.
So now if I play them both all the way through again, they should measure the same LUFS level.
So far so good, but have I achieved the desired objective of making both tracks equally subjectively loud?
Hmm, what I hear is that Uptown Funk is louder than Hallowed Be Thy Name, even at the same LUFS level. Not by much, but to my ears there is a definite difference. It may be that since heavy metal music is expected to be loud, then the fact that it isn't may seem to make it quieter in comparison. Or it may be that the more gappy arrangement of Uptown Funk allows the rhythm track to be louder to compensate.
Either way, I suspect that producers and mastering engineers around the world are seeking ways to get around the LUFS barrier and once again make their tracks louder than everyone else's. I hope not. I'd like the loudness war to be well and truly over. Let's bring dynamics back into music again.
I'm David Mellor, Course Director of Audio Masterclass. Thank you for listening.