LUFS confusion... any help?

I'm working on this track and the stereo output is peaking at around 11dB (not -11dB). Since there's a lot of automation and other factors, it's not an easy task to lower the individual volume faders so I just put a Gain plugin with the gain at -12dB. Now the stereo output is peaking at around -1dB. After that, I've put the Loudness Meter.

Thing is, all 3 meters (M, S and I) are peaking at around -12. Shouldn't they be peaking at the same value as the stereo output, since it's after the Gain plugin?

For example, if I bypass the Gain plugin, the stereo output distorts (of course, because it's at 11dB and then the M, S and I peak at around 0dB or -1dB

Am I missing something here?

Yes. Peak is something *entirely* different than Loudness.

Yes, I know that. That's not what I'm trying to figure out .

My confusion is: why is my stereo output peaking at -1dB and yet the LUFS plugin is showing -12dB?

The M is for peaks, that's not loudness, so it would make sense the M would peak at -1dB.

For example, if you start a new project and you insert a Test Oscillator with a sine wave peaking at -12dB (the default) and then you insert the LUFS plugin, all M, S and I will peak at -12dB.

So my confusion here is not the difference between peak, rms and loudness (which of course, I know the difference )

I'm just confused about what I mentioned before. Why is it peaking at -12dB when the stereo output is peaking at -1dB?

The M in a Loudness meter stands for Momentary Loudness. It shows the *Loudness* value (not *Peak* value), integrated over a sliding time window of 400ms (as opposed to 3sec for S and the entire length of the piece for I).

Ok then whoever made a video about the plugin, explained it the wrong way.

I thought (according to that video):

M = peaks

S = RMS

I = loudness

In that case, it makes more sense. Got ya

Thanks for clarifying...!

@fuzzfilth do you have any guidelines when it comes to analyzing a mix using the LUFS? For example, on that video the guy was saying that if the track is peaking at around -1dB and the Integrated value was around -10, that's a good starting point. If it's lower that than, it means that the track is heavy on transients. Of course each genre is different, but do you have any guidelines that "kinda" apply to any genre?

I tend to bounce my tracks while I'm mixing and I open them in an audio editor so I can visually see how it's behaving in terms of peaks and other problems, but in terms of loudness, a waveform could be inaccurate. Any advice?

Others may chime in with different opinions, but I'm an avid opponent to the "make it so a meets b and x meets y and you're fine"- school of thought. I know, it's just a starting point, but as you said, every genre is different. Girl From Ipanema would require very different dynamics than Smells Like Teen Spirit or Firth Of Fifths or the latest Moss album.

Check out the free Youlean Loudness plugin. It shows you a history graph of your Loudness M, S, I and True Peak (yes, this time it's Peak, actually it's better than that) values over time. Feed it your favourite music tracks and look how they do. Listen closely and you will learn what works and what doesn't.

Others may chime in with different opinions, but I'm an avid opponent to the "make it so a meets b and x meets y and you're fine"- school of thought. I know, it's just a starting point, but as you said, every genre is different. Girl From Ipanema would require very different dynamics than Smells Like Teen Spirit or Firth Of Fifths or the latest Moss album.

Check out the free Youlean Loudness plugin. It shows you a history graph of your Loudness M, S, I and True Peak (yes, this time it's Peak, actually it's better than that) values over time. Feed it your favourite music tracks and look how they do. Listen closely and you will learn what works and what doesn't.

The "issue" with analyzing other tracks is that I will be analyzing mastered tracks. My goal is to analyze mixed tracks before mastering.

So for example, as I mentioned when I talked about the other guy, he said that a mixed track (in this case he produces electronic music) that's peaking at 0dB and has an Integrated value of around -10dB or -12dB is a loud enough mix (good starting point for mastering), meaning when I master the track I won't be dealing with too many transients and stuff that will force me to do a lot of "fixing" that could have been done while mixing. Does it make sense?

So I'm not sure if there's a "standard" that can somehow translate across most genres. Maybe not... Analyzing my mix right now (and looking at it), I can see that there's no unnecessary peaks/transient stuff going on and it's actually peaking at around 0 dB, while the integrated is around -12dB, so it kinda matches what he said, even though I didn't mix to match those levels, it just happened that way:

For many the current thinking is that LUFS should be around -14. Remember that this is an average of the whole track/album. Many commercial releases are pulling back from the loudness wars and some newer cds may sound quieter than what we’ve had for a while. It’s nice to have some dynamics back in our music.

Dynamics really depend on the genre. Classical music and House music are supposed to have different dynamics. Dubstep, for example, is not based on dynamics. It needs to be loud and in your face all the time (I'm not talking about music that just sounds distorted in a bad way and poorly compressed where it's pumping in a bad way, of course)

Also, you can have a super compressed song and still be at -14. Here's why:

The -14 you mention is a relationship between 0 and that same number (-14) so for example, if you have a song in Logic and you start lowering down the volume (considering you are using a Gain plugin before the Loudness Meter, of course), the Integrated value will eventually show -14, regardless of the dynamics of the song. Actually the dynamics are metered via the LU range and not the Integrated (which is the -14 you're talking about). Both songs A and B can be showing -14 on the Integrated, one is probably peaking at -1dB and the other one at -5dB and yet, song B is show 1 as LU Range (heavily compressed) while song A has an LU range of 5. So with the new -14 standard, you're not fixing the loudness war at all Look at the image below

The -14 is more about the final listening experience (Spotify, Radio, TV, etc) and not at the mixing or mastering stage. Basically the -14 is just to allow all music to be perceived at the same loudness level. Maybe song 1 will have to be peaking at -3dB, while another one will be peaking at -10dB and yet, they still sound as if they have the same loudness. That doesn't change the dynamics at all

Hope it makes sense.

with the new -14 standard, you're not fixing the loudness war at all...Basically the -14 is just to allow all music to be perceived at the same loudness level.

"Perceived Loudness Level" is *exactly* what the loudness war was about and no, it wasn't about the aesthetics of overly compressed and clipped music.

Also, working with Loudness normalisation does not *force* you to have dynamics. It has become an artistic choice. You can overcompress if you like that sound and still not punish the listener with a jump scare every time your song rolls around in the playlist.