I would like to have your opinion on the volume of the tracks that should be sent for a release on Spotify, Amazon, Apple etc.
Spotify and others have a recommendation (-14 dB LUFS) that I followed for my first release. This was a mistake, because I read later many articles saying that no one (especially the majors) follow this rule. Pro tracks often peak positively (in dB or dBTP).
For my next release, I have a WAV track that peaks at -0.2 dB, but if I convert it to MP3 (what Amazon will do for example) it peaks at +0.8 dBTP. I hear no artifacts though.
hi @EMBmusic , first of all let’s say that there is a little bit of the common confusion between loudness and volume here. The peak volume (which in fact and after mastering should not exceed -1dB and after mixing possibly be around -6dB to leave the mastering engineer enough headroom to work with. But a track that peaks at -1db can be perceived not as loud as a track that has the peak at let’s say -3dB. This depends on loudness and here it’s the LUFS that give the scale.
-14 LUFS means that the overall PERCEIVED volume is higher than that of a track with let’s say -21 LUFS and lower than a track with -9LUFS despite their peak value in absolut dB that could be highest on a track with less LUFS.
It maybe sounds still a bit confusing but don’t worry, the practical side is that you master your tracks with a peak of -1dB (artifacts can emerge when converting the digital track to an analog audio track and not be perceived on every music rack the same way) and bring the loudness to the value that sounds best to you, knowing that for example spotify will lower the volume and or loudness of tracks with more then -14 LUFS to not be perceived as disturbing loud when played between other songs (bringing for example the loudness down to -14 LUFS).
But louder (in the sense of loudness) goes to the expense of dynamic range so louder is not always better although this thought led to the famous Loudness War. So my advice is just to try to find the best compromise for your track between loudness and dynamic range and to keep in mind that the algorythm that will bring your track back to the loudness value of the stores is beyond your control, so maybe you want to not make it overdo on a track of yours with to high a loudness.
I hope this could help a bit
Everything you say is completely true, but the problem is that the major companies don’t care at all. Here is a very interesting article:
You can read here that 1 track from Imagine Dragons peaks at +2.5 dBTP! I think that, coming from iMusician, such a track would be rejected by Spotify, Amazon, etc. So we can’t fight against the major companies…
For my first release, I followed Spotify’s recommendations and I find that in some contexts, my music sounds very low compared to known recordings. Especially in Spotify for browser (Edge, Chrome…), which does not normalize music (and I read somewhere that a significant percentage of listeners do not use the Spotify app, or turn off normalization in the app).
So for my upcoming release, I tried to go higher in volume. I did some tests with the Spotify app (I make a playlist between Spotify files that sound very loud and my local files) and I find my music less inaudible compared to big studios
To summarize your recommendations, you advise me to make a music that sounds good to me, while being careful not to exceed -1 dB, if I understood correctly?
thanks for the article, it is really well done.
in the end, it is not really the peak that makes a song sound louder but the loudness. The issue with the -1 dB is just a precaution to not run into unwanted distortions while playling back and I would recomand to stick to this. For what regards loudness and dynamic range I would recomand to keep it balanced to the style of your music. you wouldn’t want to compress a piano song for example, but a dance song can use a lot. If you overcompress you just risk that a lot of people won’t hear the song you mastered but an artificially expanded one. If you overcompress a song it will sound worse than an undercompressed one.
So if you should get a grip on an audiophile version of these majors you’d notice that they peak without distortion and also keep the right dynamic range for their song while for a CD this doesn’t make that difference and they are generally more compressed.
That said, just make your music sound the best way and don’t care too much about the loudness if more loundness results in changing the sound of your song.
