Hello everyone, I am a 90’s producer returning back to make some new stuff. Things are very different now !
I am making some material to go on Spotify and similar.
I understand the final output integrated loudness should ideally be -14 LUFS
Individual component audio clips which I have recorded before the final output stage, show a warning in the statistics page in Cubase Pro 12, with an integrated loudness level of -15.6 and a true peak of -1.97dbtp
Am I correct in thinking that so long as the final mix is -14 LUFS then the loudness of the clips within does not matter?
The Cubase manual appears to be telling me to record at -23
Bit confused and would like to get this straight in my head, all help appreciated !
Forget about these rules. Modern music often has to be from -7 to -10 LUFS depending on the genre to get a certain sound.
Try to make your mixes sound as good as possible, no matter the loudness. Thats the nice thing about it, you don’t have to care anymore about volume guidelines.
You’re right - the LUFS and peak values of the individual clips is not important.
The LUFS level you aim for is up to you, and as TJ99 says, some genres are often much higher than -14 LUFS integrated and in fact heavy limiting is part of the sound people are aiming for.
You need to understand the impact of that though - when played back through many normalised streaming services, your music will be turned down until it averages -14 LUFS. It will sounder quieter and less dynamic than other tracks with lower integrated LUFS.
I’d say -14 LUFS is a good start point if your genre does not need a heavily limited sound.
You need to be handling both. Cubase is working in floating point so any LUFS will not clip. However you can get clip when it is converted to fixpoint. If you dont get the LUFS right the distribution channel will mess with your audio and that is usually very bad for the quality.
What is your source for this statement?
Streaming services measure loudness on the audio material they host and apply a loudness penalty if considered too loud. This “penalty” is just the streaming service turning down the volume on the on tracks exceeding the streaming service’s loudness limit. This great feature can be turned off in the streaming apps and does not alter the source audio in any way.
@scottthegreat This is normally not a subject a producer and/or mix engineer worry themselves with (too much). It is normally the mastering engineer who ultimately sets the final loudness.
If you are mastering your own material, which is something I fervently discourage, I would recommend you use reference tracks and pick your target loudness based on similar productions.
Finally, as @Tj99 stated above, with loudness penalties in place on most streaming platforms, which is a good thing, aiming at a specific loudness does not hold as much value anymore.
That has nothing to do with what we are discussing.
You claim that streaming services “mess” with the audio if the loudness of it is outside of their guidelines? What does “mess with” mean in this context?
Yes, there are some weird posts on the web talking about the streaming services limiting or compressing your track, but it does not.
Streaming services only normalize the level, it only reduces, or increases the gain.
If your tracks lose 4 dB when normalized, then just turn up the volume in your car, headphones, etc.
LUFS is about the “density” of the sound.
A classical piece will have much lower LUFS than heavy dubstep, even if they both have a max level of -2 dBFS.
Most streaming services recommend a -14 LUFS because at some point, audio engineers found out that modern music was enjoyable at this particular density or loudness.
This is more like a common agreement than a real standard, for this reason, some would recommend -15 or -16 LUFS.
When this “perceived” loudness is too high, the streaming service will only turn down the volume accordingly. This is just an “ear” thing.
OP, the level at which you record is very different from the level of the final mastered track.
If you’re recording at at least 24 bit, which you should, the noise floor is so low that you no longer need to try to record at close to peak values.
In fact, you should leave plenty of headroom when recording individual tracks so that you leave plenty of space for unexpected peaks.
Aim for something like -24 to -18dB average dBFS on your channel meters.
If you’re not recording but using samples or VST instruments, you don’t have to do this as you’re recording the midi instructions not the audio. But if you’re mixing midi and audio tracks you will need to to get consistent levels.
If you’re using effects plugins that emulate real devices, you may need to feed them signals at -18dB for them to work as expected.