This article is a follow up to our previous article Mixing with Audio Construction Kits, and part of the How to Mix and Master a Song series.
In it, we look into the most important topics concerning the process of mastering a track and provide some useful tips and best practices to keep in mind and try out.
For the demonstration, I’ll be using the track that I mixed in my previous article from the construction kit and Ableton Live template Global Tech House Vol 1 by Producer Loops.
Preliminaries and clarifications
I want to start off by clearing some popular misconceptions about the process of audio mastering and loudness in particular.
Loudness isn’t an absolute property of sound, but rather a subjectively perceived one; that’s why we need to distinguish between perceived loudness, which is the subjective way in which the brain interprets the loudness of a sound, and measured loudness, which is the mathematical measurement of sound volume in absolute and objective terms.
Contrary to popular belief, mastering isn’t the only process that affects the perceived loudness of a track. Factors such as the harmonic and frequency content of the audio, as well as the quality of the transient peaks achieved through previous stages of music production, such as performance, recording and mixing, along with the gear and skills employed, can also be crucial to determine how loud a track comes across, even though they may not affect the measurement of its loudness as much.
There are several reasons as to why two tracks with the same measured loudness may not be perceived as equally loud; an important one being the uneven response of meters to different frequencies, a phenomenon that doesn’t necessarily translate to our own perception.
Low frequencies have a greater impact on meters than higher ones, but not on perceived loudness, which is more susceptible to other factors previously mentioned.
What all this means in practical terms is that mastering can enhance a mix, but it’s not enough to make a poorly recorded and mixed track sound as well or as loud as another track that’s been expertly produced from start to finish, no matter how far we push it and regardless of metering readings.
As I stated in my previous article on mixing, having sufficient headroom to work with is a crucial requirement for mixing, and this applies as well for mastering, as it’ll give us the space needed to accommodate the changes that will occur in the upper dynamic range of the file as we process it. Moreover, a mix that has been already limited, and had its transient peaks clipped, will also yield poor results upon mastering, no matter how much headroom is left after the clipping, because the clipped peaks of the sound will be lost and damaged beyond repair, and no amount of work or skills will be able to bring them back.
The optimal state for a track to be mastered is an audio file without any limiting applied to it, and whose peak values don’t go beyond -6 dbFS.
The resulting mastered file must also include some headroom in order to prevent clipping when the file is converted to compressed formats or played back on cheap sound systems. This headroom may oscillate between -0.3 and -1 dbFS, a value that can be set on the ceiling control of the limiter.
Compression in mastering is very subtle, and its main purpose is that of giving some cohesion to the track.
The typical settings for mastering compression are a threshold that barely dips into the waveform, and a ratio of no more than 2 dbFS; this way we can simply tame a little the unruly transients that may have eluded the compressor during the mixing stage and add some glue to the track, without impacting it in a negative way.
The attack and release parameters must be adjusted accordingly to the rhythmical content of the track, making sure that the gain reduction meter is moving in sync with the music.
There’s no need to compensate for the gain reduction with the compressor, since that’ll be done further down the line with limiting.
An alternative or complement to regular compression in mastering is multi-band compression, which allows for a more precise and fine tuned control of the effect, by making it possible to apply different settings to separate frequency areas. This type of compression is well suited to shape the tone of the track and process the sound in a more selective and precise way, as it allows to attenuate and emphasize different frequency ranges, or simply apply different compression settings to each band.
When it comes to mastering, Eq must be applied with extreme care and subtlety in order to be more beneficial than damaging, and to avoid unnatural results.
The main goal of mastering EQ is to shape the tone of the track avoiding drastic cuts and boosts, so as a general rule, we’ll favor wide bell and shelf filters with only a few dbFS of gain increase or reduction.
In this instance, I will apply a slight boost to the low and high ends and a very subtle and wide cut to the mid range. This is a very common and effective eq curve for mastering that declutters the busiest area of the track and creates contrast between the extremes of the frequency spectrum.
A very useful feature of Live’s Eq Eight is the ability to solo each band by clicking and holding its corresponding node. This allows us to listen to the frequency range that we’re affecting in isolation mode.
To take advantage of this feature, we need to enable the Audition Mode switch and then clicking and holding the nodes we wish to audition.
Limiting is where the real magic of mastering happens, as it is the stage that deals most significantly with the perceived loudness of the track.
The purpose of limiting is to increase the loudness of the track by bringing the levels of its lower peaks up while limiting its higher ones.
This is done in detriment of the dynamic range of the song, so a balance between loudness and dynamic range must be struck in order to avoid sucking all the life out of the music.
Also, there's a loudness ceiling in the digital domain that can’t be exceeded, and trying to push past it will only add artifacts and distortion to the audio.
To avoid artifacts in compressed formats, as well as distortion in playback systems, we’ll set the ceiling of the limiter somewhere between -0.3 and -1 dbFS. This will go a long way to preserve the integrity of the track, no matter what format or device it’s played in.
Then it’s just a matter of increasing the gain of the limiter in order to boost the loudness of the song.
It’s no secret that many modern producers are obsessed with loudness, making the central goal of mastering that of pushing every track to its breaking point in order to make it sound louder.
As stated before, this is a terrible idea, because there are ways to increase the volume of a track when it’s played back, but there’s no way to bring back the dynamic range or the integrity of an audio file once it's been lost.
It’s far more important to deliver a consistent experience that doesn’t force the listener having to worry about the volume management of a playlist because all the tracks fall at different spots on the loudness race.
A good producer doesn’t approach loudness as a competition, but rather as just another quality of sound, capable of providing a unifying character between all the songs of an album or compilation, or a distinctive feature to the particular requirements of a project.
This competitive environment in music production with regards to loudness has forced streaming services to normalize the loudness levels of their content in order to provide a better user experience, averaging loudness standards for different formats such as -14 LUFS for streaming and -9 LUFS for CD.
When it comes to club music, though, these safety standards don’t seem to have caught up yet, and there’s still a lot of irregularity in loudness levels between different productions, with values ranging between -12 to -6 LUFS.
Nonetheless, you should aim for at least -14 LUFS, no matter the genre.
In order to comply with loudness standards you’ll need a loudness meter capable of measuring loudness in LUFS. You can download one called Youlean Loudness Meter for free and adjust the limiter gain until the desired target is reached in the Integrated LUFS section of the meter.
As a side note, there’s an online app called loudness penalty that will tell you how much your tracks will be reduced or increased in loudness across different streaming services, which can aid you to make better choices with regards to loudness levels.
Mid/Side processing is an advanced type of processing that makes it possible to separate stereo sources into the central and peripheral areas of their stereo field and affect each of them separately. This ability makes this type of processing very useful for mastering, since it provides a more accurate and precise control over sources featuring complex stereo content such as full mixes, in a similar way that multiband processing does.
In practical terms, it allows for a better control of the dynamics and tone of the stereo field of a track using common tools like equalizers and compressors.
All we have to do in order to take advantage of this technique is to select M/S from Eq Eight’s global mode menu. Once in M/S mode, we can toggle between the mid and side channels with the edit switch.
When we switch to Mid/Side mode, the eq curve from stereo mode becomes the eq curve of the mid channel, but we get an extra eq curve that allows us to eq the outer side of the stereo field separately.
As we did in stereo mode, we can solo each node of the spectrum in mid/side mode in order to isolate the frequencies that we’re processing by clicking and holding each node.
These are the main concepts, concerns and common techniques regarding the process of mastering.
As with every other aspect of music production, the key is to get familiar with the concepts and ideas behind it, the potential problems that need to be addressed and the most useful and versatile tools for the job.
Instead of looking for the perfect recipe and mindlessly repeat each step of an ordered instructions list, it’s best to understand the type of audio material that we’re dealing with when we approach a master, and know how this type of material reacts to the various tools and techniques at our disposal.
This mindset, coupled with a dauntless attitude geared towards curiosity and experimentation will give us the best results and will make us more capable of facing a broader range of projects.