Tutorial in RMS, LUFS I/M/S, EBU R128, K-Weight etc.?

[Lophophora]

Originally Posted by Pashkuli

Just because some people have established a "standard", does not mean it should be standard for everyone.

Not sure what standard you're referring to. LUFS is a measurement tool, not a standard. Setting a uniform target value would be a standard, but it's not the case: at the moment various streaming platforms are using their own targets, not all the same. And fortunately there is no target for artists, producers and engineers... we are still free to produce our music at whatever loudness/dynamic range we think is best.

There definitely is a trend toward a uniform integrated LUFS target value on streaming platforms though, and the soon-to-be-released revision of the TD1004 AES technical recommendation is probably going to reinforce that trend. But that is only for streaming platforms, not for mastering engineers or producers.

Anyway, my point was that LUFS has a very specific weighting, which is documented. Of course you could choose a different weighting, but then you're not measuring LUFS anymore, you're measuring your own custom weighted scale or whatever you want to call it.

Loudness is a completely subjective thing.

It is true that two audio sources that have a different frequency content while being close in loudness might be perceived differently by different people, especially if these persons are at a very different age. But that is not what I would call "completely subjective", it is just a minor perception difference. In general we all perceive loud sounds louder than quieter ones, it is far from being random or completely subjective.

As you rightly pointed out, all of us don't have an equal sense of hearing. But LUFS is the best measurement tool we have so far since it is the closest to what we generally perceive. It is a noticeable improvement from RMS.

The thing is that people tend to focus on integrated LUFS, which is what is used by streaming platform to normalize. Of course when you average loudness over such a long time period (the whole length of a song or an album) it doesn't necessarily relate to what we're hearing at the moment, and when a song with a loud ending finishes as the next one starts with a very soft intro, we will perceive a great loudness difference even though both songs might have the same integrated LUFS values. But when you compare the same time windows (short term 3s or momentary 400ms), LUFS is really matching what we hear closely.

Mid age people tend to have much more sensitivity to Loudness (again subjectively).
For example I can not listen to my favourite metal (extremely heavy genres of it) music on what is now "too loud" for me, that was "quite normal" loudness level when I was 14 to 25.

I see your point, although I don't suppose you have actually measured precisely the SPL level when you were that young so you can't be sure you are actually listening less loud now. That is what I would call subjective.

The standard has been made in order to establish an armistice during the so called Loudness war.

Again, not a standard. But I know what you mean and I agree: normalization is slowly but surely helping with the loudness war.

There is not included a very important aspect of it: genre, style, timbers (of arranged instruments).

I'm curious: why (and how) would you want to include genre in loudness measurement?

[Pashkuli]

Every unit is a form of standard and they change over time.

Even the tone standard of 440Hz is a standard pitch. But yes the unit is Hz, which depends on the unit of a second, which is or has been accepted as a standard.

Do we have to accept it as the final "true measurement"? No.

LUFS are not an exception. They are subjective and relative.

Sound pressure level depends on so many factors, it is not possible to make and compare objectively two measurements in two different times.

(just for reference: I used to get asleep in my teen years with headphones on blasting death metal: Cannibal Corpse, Napalm Death, etc.; today I listen them on pretty low level and definitely not as lullabies).

[maxdembo]

Here we go again.

[Pashkuli]

Originally Posted by maxdembo

Here we go again.

People getting too much obsessed with precision advices... in something creative such as Audio (mainly songs and music) mixing.

There are few technical aspects, no doubt, but they are more for preparation.

That is why Bob Katz made his "standards" quite flexible.

Tell people to cut lows below 200Hz... and they will cut it on everything. And that is why "my god\religion is better than yours", if you know what I mean.

[Pashkuli]

Originally Posted by Lophophora

Again, not a standard. But I know what you mean and I agree: normalization is slowly but surely helping with the loudness war.

I'm curious: why (and how) would you want to include genre in loudness measurement?

It is a standard of loudness measurement and some referenced "units" called LUFS.

Well, of course genre, arrangement is important.

A dense (symphonic metal rock) mix will always sound louder than a modern pop synth-beat with autotuned vocal on top even if the latter by measurement shows relatively higher RMS or LUFS... (for the most people).

Loudness is quite subjective and people (listeners) do not relate to it by a measurement, rather by perception (highly subjective).

The "objectified" standard for LUFS (with all its variables) is a recommendation for creators and broadcasters, based on scientific, yet purely statistical data.

Here is the best "tutorial":
An Integrated Approach to Metering, Monitoring, and Levelling Practices

[Pink Wool]

So, it's "do what ever you like"? No standards? But don't use you ears, trust the meters as you can't hear it anyway?

[Pashkuli]

Originally Posted by Pink Wool

So, it's "do what ever you like"? No standards? But don't use you ears, trust the meters as you can't hear it anyway?

No, it is not.

[Pink Wool]

Originally Posted by Pashkuli

No, it is not.

So WHAT is the advice your giving then?

[Pashkuli]

Originally Posted by Pink Wool

So WHAT is the advice your giving then?

About what? What advice? I do not understand.

[Pink Wool]

Originally Posted by Pashkuli

About what? What advice? I do not understand.

Okie dokie. Never mind then.

[mozart999uk]

Originally Posted by Pink Wool

Okie dokie. Never mind then.

:-) :-) :-)

[xpander]

Originally Posted by Pashkuli

Just because some people have established a "standard", does not mean it should be standard for everyone.

Negative. Standards are definitely for everybody to be used, otherwise they wouldn't really be standards to begin with.

If you need to follow a specific standard is a whole other thing though. You can ignore specific standards all you want, and personally you might be fine doing that. But there are places/cases where you are required to meet values specified by certain standards. Otherwise your work or product will not be accepted or be considered valid.

That the standard requirements are changed and updated through times is yet another topic. None of that removes the real value of the standards. To offer everybody a common ground and official terms/measures to be used.

[mozart999uk]

Originally Posted by xpander

Negative. Standards are definitely for everybody to be used, otherwise they wouldn't really be standards to begin with.

If you need to follow a specific standard is a whole other thing though. You can ignore specific standards all you want, and personally you might be fine doing that. But there are places/cases where you are required to meet values specified by certain standards. Otherwise your work or product will not be accepted or be considered valid.

That the standard requirements are changed and updated through times is yet another topic. None of that removes the real value of the standards. To offer everybody a common ground and official terms/measures to be used.

Well said!

[mozart999uk]

Originally Posted by Abraham Liftin'

I've never seen anyone pollute a site with so much garbage, while being so arrogant doing it. This is bordering on a mental problem.

Who are you talking about?

[alanofoz]

Originally Posted by Pashkuli

yeah, "conversion factors" still ≈

Maybe, but still accurate to many decimal places. At least in my country, not sure about yours.


This could go on ad infinitum, so I'm hopping off the train.

[Pashkuli]

Originally Posted by xpander

Standards are definitely for everybody to be used, otherwise they wouldn't really be standards to begin with.

I do not know who said it or when but...
If you follow standards (stepping on\following other people's steps) you will never leave a trace of your own.

There are so many different standards for the same thing.
I am quite sure the one that all have agreed upon for the time being is the unit of time. Other than that you've got plenty of different standard units for space, volume... etc. If we were to look at it in historical context, things could get pretty crowded.

Anyway. Those are more related to Physics.

If we look at Music and Art in general, even the technical aspects of those arts... talking about standards and "how we should conform with those" is cringe. It is detrimental to creativity and pushing things forward to change, to be new.

There have been so many conventions made during the centuries as "standard", which... are just pathetic (in Music fundamental development that is).

[mikusan]

This is not even funny anymore. Can I mute this ***hole?

EDIT: nevermind, found the 'ignore' button - haven't had to use it for a while. Now if only we could do this in 'real life'...

[Lynx_TWO]

Originally Posted by Pashkuli

(just for reference: I used to get asleep in my teen years with headphones on blasting death metal: Cannibal Corpse, Napalm Death, etc.; today I listen them on pretty low level and definitely not as lullabies).

Well… I mean… not to be THAT guy… but we actually DO have completely objective hearing tests through audiologists working in hospital audiology departments who can measure your hearing. And honestly, if this was your childhood experience (my own father listened to classical and jazz and was extremely careful with his hearing in the Navy, which he instilled in all six of us kids and to which I am thankful for to this day) then you should actually get your hearing professionally checked so you know exactly where your own deficits (or lack thereof) lie.

And again, not to be THAT guy but if we are going to talk about 440 hz and tunings based on seconds, then we might as well discuss how 1 second is defined officially as the 9,192,631,770 cycles of the radiation that gets an atom of the element called cesium to vibrate between two energy states. Fun fact, scientists are considering using strontium instead of cesium for all atomic clocks, since it’s about 430 trillion times per second which would incidentally bring the accuracy of GPS to about 1 cm in accuracy vs a couple feet currently.

Point being, while all of that is actually very cool and relatively fascinating in its own right, it’s not super relavant to audio…

Back to LUFS as a standard: look, standards improve as time goes on and more data is collected. You can build a tool of measurement for LUFS based on ITU-R BS.1770 revision 1, 2, 3, or 4 of the spec. The big key is being transparent to the user as to which revision of the ITU-R BS.1770 standard you are using.

Who knows, with ITU-R BS.1770 revision 5 (ITU-R BS.1770-5) we may get another weighting which indeed does even-better-mimic human hearing and finally reach the pinnacle of loudness matching to the point where you’d technically be able to fully-automate the mixing process to where mixing becomes the user telling which instruments (you know what? Not even instruments, which SOUNDS) he/she wants the user to hear at which times in the mix.

This would actually be extremely cool as it would free up the technical side for the purely-human creative side. As a mostly-left brain person myself, for me this would be the dream allowing me to better use the right side without the left brain getting caught up in details and blocking the right side. Until then though, Deep Breaths everyone

EDIT: and if it helps, think of standards as Best Practices

[Pashkuli]

Originally Posted by Lynx_TWO

Well… I mean… not to be THAT guy… but we actually DO have completely objective hearing tests through audiologists...

And again, not to be THAT guy but if we are going to talk about 440 hz and tunings based on seconds, then we might as well discuss how 1 second is defined officially as the 9,192,631,770 cycles of the radiation that gets an atom of the element called cesium...

Back to LUFS as a standard: look, standards improve as time goes on and more data is collected.

I don't know who "THAT" guy is (maybe an expression, with which I am not familiar).
Nothing serious with my hearing. I can still hear those fuскing mosquitos getting in through the opened window at summer nights.

The point is not 440Hz or A, B, C, D...

The point is those changed over the centuries. 440Hz is actually quite new standard (from the late 30s, meaning it is less than 100 yo).

And it is good that it changes. As you say we find new means of redefining things and that is for good!

Just way till you see the number π getting redefined.

Calculating loudness (statistics) and experiencing loudness (individual) are different things and for different purposes.

That is also a point for discussion in the other thread: Normalize (by Loudness LUFS) vs Manual Gain Staging

[Cosmic Pig]

I'm kinda baffled. This tune is running -9.6 on the lufs I. Am I screwing this up or this another rabbit hole where I should ignore you hair splitting bastards and just listen?

I hope you guys know I meant lol with the bastards remark. Y'all rock hard in my books... except when your sucking me into vortexes of alternate universes chasing bafflement.

But really, is that too high, or is the violin screwing the numbers? Usually my stuff lufs out to -11 or -12.



https://soundcloud.com/user-44639725.../s-JfQW3zC9V7G

[kirk1701]

How is this thread still going?

[Cosmic Pig]

Originally Posted by kirk1701

How is this thread still going?

See, when you add a post it goes to the top. Hencely, still going.

Hope this helps.

[Lynx_TWO]

Originally Posted by Cosmic Pig

I'm kinda baffled. This tune is running -9.6 on the lufs I. Am I screwing this up or this another rabbit hole where I should ignore you hair splitting bastards and just listen?

This. Just listen. -9.6 LUFS may be entirely appropriate for your music. Just know that streaming services will turn your track down by 4.4dB, so make sure you are listening to it at that level and it’s not sounding too compressed. You can use Foobar2000 and install the ABX comparison tool to see if you can hear the difference between a more dynamic and less dynamic version of your mix:

https://www.foobar2000.org/
https://www.foobar2000.org/components/view/foo_abx

Off topic, the BPM analyzer is great for clients who send you tracks with no reference BPM

https://www.foobar2000.org/components/view/foo_bpm

[Cosmic Pig]

Awesome thanks Lynx. It's great to get confirmation of my suspicions. I don't think its overcompressed, more just dense with instruments. On the other hand compression is cumulative so I'll go in and make sure.

Thanks for the foobar links, those look really handy.

[bucca]

mbe late to the party
i agree with what xpander 've been sat about the meaning of standards.
if in need of using 'standards for measuring' e.g., they should be able as accurate and as quick as possible to read out. that's one point i see room for improvments in many places.

[bolgwrad]

Originally Posted by Lynx_TWO

Correct, and, actually, back at @bolgwrad, I probably oversimplified the explanation of -18dBFS. Here's something a bit more in-depth: https://www.audiorecording.me/what-i...ecordings.html

Note this: https://www.audiorecording.me/wp-con...conversion.jpg

Thanks again

[Lynx_TWO]

Originally Posted by Cosmic Pig

Awesome thanks Lynx. It's great to get confirmation of my suspicions. I don't think its overcompressed, more just dense with instruments. On the other hand compression is cumulative so I'll go in and make sure.

Thanks for the foobar links, those look really handy.

Speaking of compression, check this out when you get a chance. The only k-weighted compressor I know of that’s designed for digital audio from the ground up. It’s replaced all my compression tools and several other tools as well. Super intuitive and straightforward to use too. The only thing really missing is a display like Fabfilter’s C2, but using a monitoring plugin on the Monitor FX bus solves this issue.

https://www.soundradix.com/products/powair/

[Lynx_TWO]

Originally Posted by bolgwrad

Thanks again

No problem! Always cool to read through the history of this stuff. Helps put things in perspective for me