Is this an accurate LUFS cheat sheet when mastering?

Dusty B

ILLIEN
Battle Points: 37
I know every song will be different. But I came across this online, though it wasn't genre specific. Bass seems a little low for modern hip-hop but wanted to share and also see if anyone has any other LUFS reference they use. Been trying to focus a bit more on LUFS vs. DB readings lately since DBs can be deceiving.

Screenshot 2023-10-24 155002.png
 

Ys Man

ILLIEN
Battle Points: 21
Mastering is on a stereo track so this looks more like a mix sheet but some of these would be clippping on positive true peak?
 

Iron Keys

ILLIEN MBAPPÉ
ill o.g.
Battle Points: 544
Use your ears. Not maths.

However ... there are actually some actual maths.

It's more to do with loudness range and tone curves of top 25 spotify tracks. Their sound ranges all land very close with eachother.


I would really really urge you not to use gimmick based mix approaches (such as each instrument by lufs, pink noise mixing etc etc)

Literally use your ears and listen. Ask yourself if it sounds right, if not, why, then make thar change and ask again.

Every top mix engineer has their own way of mixing. You need to find your truth. But going just by numbers isn't the best bet.

Only on regards to overall loudness and vocal.
 

Ys Man

ILLIEN
Battle Points: 21
Plenty commercial records have TP values of +1.5 so don't sweat TPs too much.

However, be certain here you're talking about true peak and not just normal peak level.
While true there's some things to think about with that
1. They are using high end analog gear clipping for saturation.
2. Generally you are listening to the mp3 which generate intersample peaks, the lossless audio probably doesn't clip.
3. The mix sounds like crap and squashed lacking dynamics.

But I agree your ears are the best judge, the only meters you should be looking at are vu meters on analog emulation plugins and lufs maybe on the master overall, just to make sure your are in the range you expect after deciding it sounds good as your monitoring volume can sometimes give a different perception of loudness
 
Last edited:

Iron Keys

ILLIEN MBAPPÉ
ill o.g.
Battle Points: 544
1. They are using high end analog gear clipping for saturation.
So?
You can get better sounding clipping with analogue gear mostly, but that's not what we're talking about.

2. Generally you are listening to the mp3 which generate intersample peaks, the lossless audio probably doesn't clip.

Intersample peaks are there before. Mp3 conversion can generate clipping from it but also maybe not. Could also generate.

Lossless audio has intersample peaks too.

3. The mix sounds like crap and squashed lacking dynamics.

Only if you've squashed your mix. A mix doesn't have to be smooshed to have intersample peaks. Like I said plenty commercial well produced records with peak values greater than 0dB


Intersample peaks are largely about the playback device...

Yeah mp3 can create clipping by trying to assume and reconstruct the peak.
But what happens with intersample peaks is that thay *TRUE* PEAK value is higher than "0dBFS" so when your audio plays back, it's actually playing back with louder peaks than you realise it has... this can cause distortion in the playback device if it can't handle those values.
 

Ys Man

ILLIEN
Battle Points: 21
So?
You can get better sounding clipping with analogue gear mostly, but that's not what we're talking about.



Intersample peaks are there before. Mp3 conversion can generate clipping from it but also maybe not. Could also generate.

Lossless audio has intersample peaks too.



Only if you've squashed your mix. A mix doesn't have to be smooshed to have intersample peaks. Like I said plenty commercial well produced records with peak values greater than 0dB


Intersample peaks are largely about the playback device...

Yeah mp3 can create clipping by trying to assume and reconstruct the peak.
But what happens with intersample peaks is that thay *TRUE* PEAK value is higher than "0dBFS" so when your audio plays back, it's actually playing back with louder peaks than you realise it has... this can cause distortion in the playback device if it can't handle those values.
Yes i realised the comment about analog isnt really appropriate in the mastering sense but i was more referencing the image for those levels which would make sense on a vu meter i guess. but who would ever mix at those levels digitally? as far as i know most mastering engineers use a software true peak limiter on the end of the chain specifically to reduce this as much as possible.

from my experience mp3 conversion always raises the peaks it just depends what level youre at before i never said lossless doesnt have intersample peaks, i should have said mp3s generate higher peaks.

like i said the commercial stuff that does have values over 0db sound terrible and no "good" music does that just mainstream or electronic. A well mixed track will still sound good clipping but would sound better on good equiptment not clipping, the idea is to translate the loudness to inferior systems thats why they do it. Youre not going to get good sounding music peaking over 0db that isnt intersample peaks. people are even trying to remove those these days.
 
The best setting would be from -9 to -13 LUFS with the dynamic range reading on levels not passing 8DR. It also depends also what type of music you're mastering. Mainly just use your ears
 

members online

Top