A standard studio-recorded track will go through a number of processes before it’s ready to be released.
First, the artists will work with a producer, whose job it is to make sure the artists write the best song they can write, arrange it in the best possible way, and record it professionally (with the help of studio engineers).
This team will usually then hand their multi-track recording over to a professional mixing engineer, who will make each part sound as good as they can get it, and they will blend it together to make a coherent mix that has the right tone and the right energy.
And then it will go to the mastering engineer.
The mix should already sound great before it gets mastered, so why the need to do any more to it?
The answer is because although the mix may sound great by itself, through the speakers of a professional recording studio, it won’t necessarily sound great on every sound system, from your car stereo to your phone. It’s also likely that it might have a noticeably different tone and average loudness level from other tracks on that are going on the same album, and it might sound very different to tracks by other artists in a similar genre.
The purpose of mastering is to make recordings sound great compared to all the other music out there. And it is one of those processes that you shouldn’t even notice unless it’s done wrong.
What does this actually involve?
There is a fairly long list of boxes that mastering engineers need to check in order to get a track or album ready to go out into the world.
(Why doesn’t the mixing engineer take care of these in the mix? A few do, but it’s normal for a mix engineer to spend between 8 to 12 hours on track, and after that most would much rather give it to someone with a fresh pair of ears.)
One of the first considerations is error correction. Even the best of us can make mistakes, and sometimes it takes that second pair of ears to notice them.
For example, certain frequency spikes that are easy to miss in the mixing stage can mean that your track starts to distort in mastering when you raise the average loudness to professional levels. Or, to take another example, it can be easy to miss when certain frequencies on the left side are out of phase with those on the right, and this can mean that when you play the track in mono (e.g. on your phone or a single Bluetooth speaker) an instrument can suddenly disappear from the mix.
Another consideration is that the mix should sound right when listened to before or after other commercially available music, and particularly with any other tracks in the same album or project. It certainly shouldn’t sound distractingly different.
A crucial aspect of this involves balancing the tone of the mix – in other words, making sure that the lower bass frequencies, the mid frequencies and the high frequencies all average within a consistent range. So, for example, a track that’s recorded to sound great in a club doesn’t sound thin and harsh compared to the other tracks that get played.
Not everyone wants to fit into these frequency ranges that the music business tends to prefer. And fashions change, of course: sometimes bass is in, sometimes clear and punchy midrange is popular, and sometimes a detailed top end is what everyone aspires to. But the mastering stage is when, if you’re choosing an unconventional tonal balance, you can make sure that it actually sounds better than the conventional approach.
Perhaps the first process that many people think of when they think of mastering, and certainly the most talked-about, is the average loudness of the track. ‘Loudness’ may seem like a subjective thing, but in sound recording it is actually measured very specifically in Loudness Units Full Scale (LUFS).
Commercially available music has historically had an average loudness that is as high as possible without distorting or degrading the sound quality, because music that is louder generally sounds better-balanced (as discovered by a famous academic paper from 1933).
And even though new recent international broadcasting standards combined with the rise of streaming have seen that ‘race to the top’ ease back somewhat, there are still plenty in the mastering world who talk about the ‘Loudness Wars’: a competitive tendency in the industry to try to produce louder music than anyone else.
Personally, I think that Spotify, YouTube, Apple Music and the rest have changed the game by automatically adjusting the level of each track — so even if you succeed in making your track louder than everyone else’s, these streaming services will just turn it down more.
That said, I also believe that very squashed tracks can actually sound good in some circumstances (particularly in certain kinds of dance music), although they tend to be the exception, rather than the rule.
Which leads neatly on to…
The first casualty of the Loudness Wars has been dynamic range: in other words, the distance between the loudest point in the track and the average loudness of the track.
Tracks that have a very high average loudness have that much less headroom for variation between loud bits and quiet bits. It’s all pretty loud, basically — and that works for some styles of music, but if the track hasn’t been specifically written and recorded to sound like that then it can end up sounding unnaturally squashed.
Good mastering is about finding the dynamic range sweet spot, where it has just enough loudness and energy without sounding over-processed.
Once the master files have been completed, they are ready for distribution. This might mean being uploaded to a streaming service such as Spotify or Apple Music, or it might mean getting a CD, or even a vinyl LP, printed.
Traditionally, all of what is described above as mastering was technically known as pre-mastering, since the term mastering was used specifically to describe the physical process of creating the master recording (e.g. glass master CD or acetate master disk). But now that more and more people are choosing not to release their music in physical formats, it could be argued that it’s all mastering now.
If you are getting a CD made then your mastering engineer should provide you with a DDP (Disk Description Protocol) file that you can send directly to the CD manufacturing plant. DDP is a standardised computer file type that ensures the manufacturers will get all the crucial information that they need, as well as the actual audio files, to print accurate copies of CD.
If you would like to get vinyl printed, that is a much more specialised process, and I would strongly recommend getting the track mastered by someone provides that specific service. Vinyl is a great format for recorded music, but it does come with certain limitations in the manufacturing process – for example, tracks at the beginning of a side generally sound clearer than those at the end, and the louder and more bass-heavy each track is, the less total minutes you can record onto on each side.