Under the frequently asked questions section of Spotify for Artists, you can find information on mastering and loudness on Spotify. The streaming service makes it clear that they use loudness normalisation at -14 LUFS. Moreover, engineers can get somewhat confused with how the information is presented. The common misconception is that engineers should master your music at -14 LUFS.
What is LUFS?
You may be wondering, “what is LUFS?”. Firstly, the decibel (dB) is a logarithmic unit that describes how loud a sound source is. Within your digital audio workstation, dB is often expressed as dB FS, or decibels relative to full scale. LUFS on the other hand expands on this. Originally, LUFS was known as LKFS, or ‘Loudness, K-Weighted, Relative to Full Scale’. This was standardised in the ITU-R BS.1770 standard. Later on, LKFS became LUFS. Simply put, this was only a name change. LUFS is an acronym for ‘Loudness Units Relative to Full Scale’. As previously defined by the ITU-R BS.1770 standard, LUFS is a unit of loudness measurement used for the purposes of loudness normalisation.
Why Loudness Normalisation?
Spotify uses loudness normalisation on their platform at -14 LUFS. The purpose of this is to make sure that when listeners have a playlist of multiple songs, each song appears to be the same level of perceived loudness. The sole reason behind this is user experience and to create a smooth listening experience when going from song to song.
This is not a recommendation for the mastering process. Spotify sharing their loudness normalisation process, simply lets us know that the loudness war is over. All audio across Spotify is therefore perceived as the same level of loudness. This is to say that whether your new EDM banger is at -5 LUFS, or pop hit is at -10 LUFS, when played back-to-back, each piece of audio material will sound almost identical in perceived average loudness.
It is also worth noting that across the hundreds of streaming platforms, each platform uses their own loudness normalisation ‘target’.
Okay, What Does This All Mean?
Overall, this means that when mastering, you should make your track as loud as where it naturally wants to be. As an example, it’s common for electronic dance music records to be as loud as -5 LUFS. Whilst film score can be as quiet as -22 LUFS. There is an argument that the limiting process where loudness is achieved is a key component to giving ‘louder’ genres its distinctive level of ‘energy’.
Lastly, the real enemy when mastering for distribution is data conversion. For example, when going from .WAV to compressed .MP3, it’s common for your track to naturally increase by up to 3 dBFS ‘peak loudness’ as a result of the data conversion process. Therefore, the only advice Spotify provides to mastering engineers is to ensure that when mastering, for engineers to leave a 2 dBFS ceiling so to accommodate the data conversion processes after sending to distribution.
At Vasonic we take great care when it comes to dynamics processing, loudness and data conversion upon mastering client work. If you have any questions with regards to our mastering processes, feel free to get in touch.