LUFS -14, is it a mandatory rule?

[jcea]

Hello everyone, I have been following this great forum for many years but I never dare to speak or ask anything because I consider myself a complete novice and I know I would mess up. However, I have been thinking about this for days and I would like to know your opinion .

First of all I apologize for my level of English :S and I hope that everything I put is understood. Second, sorry this "big" text :S but I prefer to detail as much as possible my doubt.

I've been preparing an album for some time now, an electronic one (a kind of mix between Jarre/Vangelis/similar artists and Synthwave music).

I'm not a professional, everything I do is for hobby and to learn everything I can, I always learn new things from the synthesizers, from the DAW itself, etc.
In this album, I have set out to learn a little bit about mastering (not as a professional obviously but at least to make the mix sound a little better hehe).
And well, I'm learning to use which plugins are necessary (eqs, compressors, limiters, etc.) in what order and more or less when they would be necessary to use them.
I got to master one of my songs and, both personally and the opinion of a relative of mine (he is a professional but playing the piano, he has no knowledge of mastering), we think it sounds pretty good (or at least a little better than the original mixed song).

However, since some time ago I have seen both tutorials in Youtube and in other websites that you have to be careful with the LUFS and that should not exceed -14.
I have been investigating a little and more or less I understand the idea (the song should not bet too noisy).
Even I have read news of about Spotify that if your song surpassed -14 (or did not arrive) automatically it adjusted to -14.

Using the Youlean Loudmeter VST plugin I discovered that my mastered song was set to -9! and I thought it sounded good, so I decided to give it a second turn and improve the master.

However, I decided to compare songs from other artists (one of the advices I saw was to use a song from another artist to compare the master) and I can see that, at least, the artists I listen to, do not follow this rule :S.

For example, Carpenter Brut, I consider that the songs are very well mastered. I opened an empty project in my DAW, I put one of his songs (Turbo Killer), I analyzed it with Youlen Loudmeter and I got a value between -3 or -4 LUFS!

Later I checked if it was listened equally strong in Spotify (thinking that perhaps it would sound lower when surpassing these LUFS) and it sounds exactly as strong!

I tried with another Synthwave artists: Le Matos, Power Glove and Waveshaper... same results :S (a bit lowered, about -9 LUFS).

So, my question is, are the LUFS really necessary? Why these artists can overcome the LUFS and nothing happens to them (apparently, maybe I'm totally wrong or maybe I'm not analyzing the song properly). What do you advise me to know what volume is appropriate for the final mix?

Although you don't answer me, thank you very much, at least, for reading this booklet and I'm very sorry :S.

If you agree, when I have the most advanced album I would like to share with you one of the songs, unmastered and mastered, and I would like to know your opinion .

Cheers and thanks again!

[anoise]

-14 LUFS could be too quiet, depends on the material. For synthwave -12 to -9 is fine, I wouldn't go above that (-9, -8, -7...).
The artists you mentioned with -3 LUFS tracks, simply don't care about dynamic range and sound quality. It is fatiguing to listen to some of these albums/tracks from these artists. Don't do that!
For what loudness should you aim?
Read this: https://productionadvice.co.uk/how-loud/

[Kwurqx]

So, my question is, are the LUFS really necessary? Why these artists can overcome the LUFS and nothing happens to them (apparently, maybe I'm totally wrong or maybe I'm not analyzing the song properly). What do you advise me to know what volume is appropriate for the final mix?

The short answer is: for music there are no "rules" for Loudness and/or related Dynamic Range. There has been a tendency over many years to create ever "louder" music. Often at the cost of dynamic range. Mostly to get the listeners attention, and because (in way too simple terms) louder often sounds better.

Lately music streaming services have started to adjust the loudness towards a reference. Often in LUFS since this in meant to measure perceived Loudness as opposed to objective Volume. The idea is to achieve a common Loudness for all material. Meaning that they will either turn up or down the volume for the whole "song" if below or above the target LUFS. Reference values may differ (e.g. Apple -16 LUFS or Youtube -14 LUFS), but will be consistent accross the platform.

For a way more detailed information check out "The Loudness War". Plenty of articles and videos on the subject. Her's one from Sound On Sound.
https://www.soundonsound.com/techniques ... udness-war

[BertKoor]

Read this: https://productionadvice.co.uk/how-loud/

+1

[Kwurqx]

For what loudness should you aim?
Read this: https://productionadvice.co.uk/how-loud/

Ian Shepherd is an excellent source for information and advise on Loudness. He is a pioneer and activist on the subject.

[jcea]

Wow, thank you so much for your answers! I thought that when people saw so much text they would run away hehe.

Well I am more calm, -9 LUFS was the value that I obtained when mastering my first subject and, personally, it sounded well. Anyway I will read the article to understand everything better (thanks a lot for the recommendation! ).

Now the doubt that I have is, what happens if I upload a song with -9 of LUFS to Spotify or Youtube? they lower the volume until arriving at -14 LUFS?

I believe that they do nothing because, returning to the example of Carpenter Brut, the songs in Spotify sound like in other portals (for example, in BandCamp).

Returning to the song Turbo Killer, you download both songs from Spotify and BandCamp (only to study/analyze! no piracy!) and you can check that it gives you -3 LUFS.

I would put a link to the songs from Spotify and BandCamp but I don't know if it would break any rule of the forum :S (anyway it's easy to find this song).

Thank you so much again!

EDIT: I wanted to clarify that until now (before knowing the LUFS) my way of mastering and mixing was to try not to exceed 0dbs and that the song (after eqs, limiters, etc.) sounds good (for me is when I can hear the different sounds separately)

[Compyfox]

Ian Shepherd is an excellent source for information and advise on Loudness. He is a pioneer and activist on the subject.

So are many others, including myself. Shepherd is only very vocal about this - if we ignore his plugin range (MeterPlugs), and his yearly "Dynamic Range Day" promo (for a lot of us engineers these days, every day is "Dynamic Range Day", just saying)

Now the doubt that I have is, what happens if I upload a song with -9 of LUFS to Spotify or Youtube? they lower the volume until arriving at -14 LUFS?

I believe that they do nothing because, returning to the example of Carpenter Brut, the songs in Spotify sound like in other portals (for example, in BandCamp).

Spotify and Youtube will Loudness Normalize.

For Youtube, that was exclusive to big channels until about earlier this year - now it applies everywhere. Youtube will actually (as of the moment of this post) pull down to -14 LUFS ILk

Spotify offers two options in the app: Spotify and Spotify Loud. The former goes to -14 LUFS ILk, the latter to -11 LUFS ILk

Bandcamp doesn't use Loudness Normalization (yet), neither does Beatport.

Returning to the song Turbo Killer, you download both songs from Spotify and BandCamp (only to study/analyze! no piracy!) and you can check that it gives you -3 LUFS.

The question here is... Are we talking Integrated Loudness over time (ILk), Short Term Loudness maximum (SLk) or Momentary Loudness (MLk). It would be nice if we could all agree to add the appropriate shorthand symbol to declare which ballistics/settings we talk about.

EDIT: I wanted to clarify that until now (before knowing the LUFS) my way of mastering and mixing was to try not to exceed 0dbs and that the song (after eqs, limiters, etc.) sounds good (for me is when I can hear the different sounds separately)

That is called "Peak Limiting", and that concept has been frowned upon for over a decade at this point. Most notably with the introduction (and slow adaption) of the K-System v1 (Bob Katz), which asked musicians and audio techs to pull down the loudness to a more reasonable level like -20dB RMS avg (full dynamic), -14dB RMS avg (good middleground) and -12dB RMS avg (absolute max), measured with a Dorrough 40A meter (which the K-System meter uses as backbone).




To answer your question: "LUFS -14, is it a mandatory rule?"

I would say... these days, yes. -14LUFS ILk also opens up a lot of doors for you if you ever(!) decide to publish on more than just streaming platforms. It is easier to transition to Vinyl and Tape (both back on the rise), but also HD audio (Blu-Ray, Streaming) and/or even integration into advertising or movies.

The current trend is to finally pull down the loudness. Most music streaming platforms thankfully actually agreed on a value between -14 LUFS ILk and -16 LUFS ILk. If you master louder than that (let's say -9 LUFS ILk), just because you can and you think "this genre needs it", then your material will be pulled down on the fly while playback (non-destructive). Then in comparison to other productions that were properly mastered at -14 LUFS ILk, your creation will have lost out on transients and therefore sound softer and less "dynamic"/punchy.

The best and shortest visualization I can offer you, is a remake of a nearly 10 year old video by Matt Mayfield. This video was posted about a year prior to AES 142 in Berlin, Germany (2017). The year where Tidal actually promised to drop down to -16LUFS just like Apple, but they never really did.

https://youtu.be/yB7W5Gin9v0


Now before you say "that will be too quiet on portable devices" - more and more players these days include "normalization on the fly", which can also "pull up" quiet tracks and suitably limit them so that you can enjoy your music even in noisy environments.

Another not-to-underestimate aspect is "more reasonable loud material" forwarded to radio stations. I've once forwarded material to an US radio station in the second half of 2000s. That station had some very strong multi-stage compression and limiting before the feed got distributed through FM. And compared to super strong compressed content, which sounded like garbage after running through various forms of compression (think the now infamous incidents of: Metallica - Death Magnetic), my K-20 material was sounding just as loud as the next mix in the cue, but way more clean.




"I hear you. But still - my music won't sound aggressive and punchy anymore / EDM NEEDS THIS!!!!"

Have you ever heard of the concept of "artistic summing compression"?
Or do you really think Club Music form the 1990s/early 2000s was less good/punchy because of the lower loudness?


Food for thought.

[Hermetech Mastering]

No, it's not prescriptive. It's a streaming level recommendation, not a mastering target. Make the music sound great at whatever level the music sounds best at. Maybe that acoustic Jazz album sounds great at around -18 LUFS Integrated for each track, while your EDM client's album sounds fab at -8 LUFS Integrated for each track. Nothing wrong with either of those.

Let the constantly moving loudness targets used by the streaming giants put it where they think best. The whole reason for having Loudness Normalisation is that we DON'T have to think about it, and instead can concentrate on making the music sound the best it possibly can.

If you send me a track, I'll master it for free, and put it where I think its "loudness potential" is, i.e. I'll try to get it as loud as I can before much audible limiting distortion. That should play back with very nice sound quality on all systems, although it might not be as loud as you want it. I could provide a loud version too if you liked, if you wanted to specify a LUFS Integrated figure for the track. It might be nice for you to compare with your own master.

[Compyfox]

I am sorry, but I highly disagree.

If you purposely master loud, just because "it is what the music needs" and also say "nothing is wrong with that", you will ultimately lose out to the competition that goes back to more reasonable techniques from the late 1990s/early 2000s. Their material will sound better at every volume compared to yours. This is a proven fact.


If anything, it is not the quote-unquote "streaming giants" that dictate the game (in fact, for once they finally improve on things for the better). It is still the fault of the labels and especially music magazines for still riding on the age old topic "make it loud and impactful". Again, as if music from the mid 1960s to mid 1990s on Vinyl and Tape was not like that (loud and impactful - there was a Loudness War even back then! But because we had no unified Loudness Normalization). Same goes for early releases on CD.


But maybe our focus these days is just wrong in general - and I can already hear the heated arguments from a mile away (oh boy, if this would be GearSlutz... ). But I am of the strong opinion, that we should finally re-learn how to mix properly, and with reference levels again, rather than just say "do what you do". It is so easy to integrate into your workflow. Then we might not even need to have these debates ever so often "what is the right way to do...".


Apologies if this comes along strong, but this is exactly the reason why I highly dislike these types of arguments (and in fact, one of the main reasons I stopped posting in general). Too many people bogged down in their own set parameters, saying "lemme handle this - I can make things loud and punchy - how I think it should sound like, let the other figure out the rest". This is exactly why we (plural, audio techs all around the world) have been walking in circles for 20 years, and constantly talking around each other.

But hey... nice hustle (final paragraph).

[Hermetech Mastering]

I'm not gonna waste my energy arguing with you. I serve the emotion of the music, not "loudness targets", for which my clients over the last eleven years have been very happy, and continue to come back for more. See my Discogs page for a small percentage of the music I have worked on, from avant-Jazz, to Folk, to Hip Hop, to harsh Industrial Noise, and pretty much everything in between. It's all over the loudness scale, which is as it should be.

https://www.discogs.com/artist/1575649- ... layout=med

Being slagged by the self-designated KVR loudness police for offering to master a track FOR FREE for someone who might be curious and learn a bit from the process, is hardly hustling, and IMHO you have struck well below the belt there my friend.

OK, you're back on ignore.

[4damind]

Yes, -14 LUFS is the recommendation of many online stream providers. For film and TV the recommendations are even lower, down to -28 LUFS (EBU R128).

As you have noticed yourself, current productions are almost always louder.
In my experience many are between -8 to -10 LUFS (for music with little dynamic like EDM, pop etc). So it depends a lot on the style. -14 LUFS on a trance or EDM track would be too quiet.

I would always listen to some current productions in the same genre and then orientate on these values.

[kevvvvv]

cool thread. Helpful

[Compyfox]

I'm not gonna waste my energy arguing with you. I serve the emotion of the music, not "loudness targets", for which my clients over the last eleven years have been very happy, and continue to come back for more.

The same goes for me. And I set a personal absolute upper limit of -12 LUFS ILk and give my clients a valid explanations as to "why" -- always with an eye and ear on the most recent trends and possible future proof releases. It doesn't matter if it's classic, rock, pop, EDM. Doing that pays off in the long run as well - I had clients come back to me several times to ask "can we re-release at lower loudness?" - to which I can always say "sure", because no sound changed. It sounded good no matter the volume on the amp.

It's all over the loudness scale, which is as it should be.

My finalized(!!!) mixdowns these days are between -19 LUFS ILk and -15 LUFS ILk absolute max - if it doesn't sound good at these values (or any amp volume for that matter), it definitely won't sound better at a forced -6 LUFS ILk because "the genre demands it".

The current future trend is an absolute -12 LUFS ILk (thankfully! Facebook finally adapted), with the majority clocking in at -14 LUFS ILk (Tidal, Spotify, Youtube, Amazon Alexa/Amazon Music, now also SoundCloud) to -16 LUFS ILk (iTunes, AES Streaming), with some strays dropping to -18 LUFS ILk (Sony Entertainment) or even -27 LUFS ILk (well, technically -24 LUFS ILk still, but Netflix focuses on "Dialog Measurement").

One should always focus on having a good and finished mix without exceeding a certain loudness limit (say -16 LUFS ILk +-1LU). You can still compress the sum as you seem fit (as in: the glue effect is still possible - in this case it is not used for "limiting" but for artistic effect even).

Heck, be artistic, go to town! Loudness raising and "Limiting" is still a completely different topic.

This... whole thing... has been possible for productions over 40 years between 1960s and early 2000s. Why is it suddenly "not important anymore" in the 2020s?! Why keep on saying "let's make things how you seem fit, and let the others fix possible issues" (in this case - Loudness Normalization by the service you upload to) and/or the classic "exceeding a certain signal strength while mixing is not important - just let the master never clip" (tell that to plugins with saturators that have fixed reference levels, unless you want that over-saturated sound)?

IMO the completely wrong concept of thinking.
But to each their own.

Being slagged by the self-designated KVR loudness police...

...

OK, you're back on ignore.

Thank you for putting a smile on my face with that just absolutely ludicrous comment of yours.

We might have the usual classic clash of opinions and egos going on here. And ego's are so easily and fast hurt these days, it's not even funny anymore.

It is always the same on big audio platforms like KVR and GS - ask one question, get 50 different answers, and by page 2 we're in a full-on comment war, policing as to "what is ultimately correct" and not allowing other and maybe even possible better opinions (because: time tested experience. That comment just proved that.

But hey... you do you. You've offered free help. If you have the capacity, all the more power to you. It was still a "smooth move" regardless.



However - the best and most mature response to just back out, and no need to put anybody on "ignore" (plus even publicly mention it that you did). I've hoped that @jcea learned something regardless. From comments by 4damind and kevvvvv, that certainly seems to be the case. This is what we should all strive for (broaden our horizon, learning something new ever so often).

Have a nice remaining day.

[jcea]

Wow!! So many replies, thank you very much. All your answers are very interesting and help me a lot, really. Some of them I need to study a little hehe. Compyfox, thank you very much for your big explanation, I have to study the ILk, SLk and MLk As I said, I'm not a professional and only want to learn a little about mastering, so this thread has become very interesting.

And please, I don't pretend that there were discussions or anything like that, I would like there to be a good atmosphere.

Compyfox, to summarize a little, my doubts were more focused by comparing my master with other artists of the same musical style (in this case Synthwave).

Basically I didn't understand why the songs of these artists sound louder (surpassing the LUFS). If I adapt my theme to -14 LUFS, when I export the song to WAV and open it with an editor (for example Audacity) I can see that the wave is very small compared with other songs.

Look, I put you a visual example. The two first tracks correspond to my song, the first one without mastering and the second one mastering (with -14 of LUFS). In this example, the third track is a song with -3 of LUFS (Carpenter Brut).



Another example, this time, with another artist (Power Glove) with -9 of LUFS.



Yes, I know, my song is too long

I did not understand why they ask me to establish -14 LUFS when the artists do not do it and I had the doubt of what I had to do, what is the best (in my case).

My goal in mind, when I was mastering my tracks was that if (let's suppose an ideal situation hehe) in Spotify existed a Synthwave playlist, I'm listening to several artists of this style and, in the middle of the playlist, my song appears, it seemed one more in the playlist. I know that it is impossible in my case (I would need better knowledge and better professional tools ... or maybe to pay to a sound engineer), but at least improve the original song a little bit.

Thank you very much to all of you for your contributions

Cheers!

EDIT:

No, it's not prescriptive. It's a streaming level recommendation, not a mastering target. Make the music sound great at whatever level the music sounds best at. Maybe that acoustic Jazz album sounds great at around -18 LUFS Integrated for each track, while your EDM client's album sounds fab at -8 LUFS Integrated for each track. Nothing wrong with either of those.

Let the constantly moving loudness targets used by the streaming giants put it where they think best. The whole reason for having Loudness Normalisation is that we DON'T have to think about it, and instead can concentrate on making the music sound the best it possibly can.

If you send me a track, I'll master it for free, and put it where I think its "loudness potential" is, i.e. I'll try to get it as loud as I can before much audible limiting distortion. That should play back with very nice sound quality on all systems, although it might not be as loud as you want it. I could provide a loud version too if you liked, if you wanted to specify a LUFS Integrated figure for the track. It might be nice for you to compare with your own master.

Hermetech Mastering thank you very much for your offer, I sincerely do not want to bother you and I am aware that mastering is just another professional job so I do not think it is fair to make you work for free. I just wanted to learn a little bit about this world of mastering but without abusing the professionals .

[Hermetech Mastering]

No worries, offer still stands if you change your mind!