How would you raise your LUFS (when volume is already maxed)?

[NewUser]

Hi,

I've got the volume maxed. So, I'd have to lower things to raise things... I currently am at -12 LUFS and I'm trying to reach -10 LUFS with -2.5 max true peak to match my reference for a cover song.

I am thinking I need to either add another instrument or some how lower/raise EQ levels... not sure.

What would your methods to raise LUFS or any tricks you know?

[ggrey]

Are you limiting or compressing the maxed volume on your master or buss? You may have peaks that are maxing the volume level that do not strongly affect LUFS number.

[serr]

By "maxed" do you mean some peaks already full scale?
It's common to use a limiter to reduce peaks to then allow you to boost a mix into volume war loud territory.

[bFooz]

Limit/clip drums separately before summing into LR.

But going for -10 LUFS and -2.5dB TP gives you just 7.5dB range. If we are talking about integrated LUFS, that's too low. I'd set just -1dB TP.

[Lynx_TWO]

-1dB True Peak is enough to account for inter-sample peaks when converted to a lossy codec. There’s zero point in going louder than -14 LUFS since the streaming service will just turn it down when you upload, making your mix sound thinner and less dynamic than everyone else, instead of louder. Apple Music actually normalizes louder tracks to -16 LUFS. You shouldn’t be looking at loudness for your reference track; all you need to listen for is balance and EQ. I’d actually recommend normalizing all your reference tracks to -14 LUFS for better accuracy and to avoid falling for the natural inclination of the brain to think that louder sounds better. By keeping your reference tracks to their integrated loudness without normalizing to -14 LUFS, you’re actually doing it backwards.

[NewUser]

Originally Posted by Lynx_TWO

-1dB True Peak is enough to account for inter-sample peaks when converted to a lossy codec. There’s zero point in going louder than -14 LUFS since the streaming service will just turn it down when you upload, making your mix sound thinner and less dynamic than everyone else, instead of louder. Apple Music actually normalizes louder tracks to -16 LUFS. You shouldn’t be looking at loudness for your reference track; all you need to listen for is balance and EQ. I’d actually recommend normalizing all your reference tracks to -14 LUFS for better accuracy and to avoid falling for the natural inclination of the brain to think that louder sounds better. By keeping your reference tracks to their integrated loudness without normalizing to -14 LUFS, you’re actually doing it backwards.

If you turn off the normalization in Spotify you'll hear how the track was actually supposed to sound like.

I'm doing a cover song and matching it's LUFS. If someone turns off normalization, I want it to be very similar in loudness to the cover track.

[Lynx_TWO]

Originally Posted by NewUser

If you turn off the normalization in Spotify you'll hear how the track was actually supposed to sound like.

I'm doing a cover song and matching its LUFS. If someone turns off normalization, I want it to be very similar in loudness to the cover track.

Ahhh… ok I understand why you are doing that now. And yes, Volume normalization (turning down the song) doesn’t change the sound of the song at all. It’s simply like turning down the volume on your radio, except even more transparent since it’s a digital volume reduction. For example; reference level for mixing is 85 dB SPL (although a lot of us mix at a way lower volume then that, it’s a good technique) so if you were to measure SPL on speakers and turn the volume control on your amp up and down per song to match that reference level of 85 dB SPL depending on the song, you’d get the exact same effect. If a song sounds thinner when turned down it’s simply because you are hearing the real lack of dynamics and over-compression in the song.

Now, when it comes to Spotify, the only time Spotify will turn UP and mess with your track is if someone sets the volume normalization to Loud and your song is lower than -11 LUFS… so if you’re really worried about that (I’m not, most people never touch the defaults and just, ya know, set the volume level in their car to one setting) then the loudest you should go to avoid potential screwage is -11 LUFS with -1dB True Peak, or -10 LUFS with a True Peak of 0dB

I do get what you’re saying though in your particular case; if someone DID turn off volume normalization and listened to the cover and the original back-to back, then they would notice a difference in volume. However, considering that things get “remastered” all the time (although I’m not sure if Spotify allows you to replace your existing albums or songs? I could see that being a potential issue with payments…) then that reference level might change

Now, since the resulting LUFS are going to be the same AND the reference has a True Peak is at -2.5, you DO have an opportunity here to use an extra 1.5 dB of dynamic headroom for the same loudness, which could potentially make your cover sound better and more dynamic than the original at the exact same perceived volume! . Another way you could potentially make it a more-emotional connection is through changes in volume of different tracks throughout the song through manual automation by a couple dB… just enough to be subconsciously different. That way, while your LUFS will be the same and the overall track will have the same loudness, your LRA will have more dynamic range.

I actually did that here when I mixed this track today (I’m not sure I like the bass levels yet; original tracks are pretty flat and lifeless…. But this also took longer than I like to mix so might have suffered from some hearing fatigue as well…)

https://youtu.be/q4kz3zvfRT4

[NewUser]

Lynx_TWO, Thank you for your information.

Out of curiosity, I did ask ChatGPT this same question and it's response follows:


There are several ways to increase the perceived loudness of a song without increasing the volume, some of which include:

Compression: Compressing the audio signal can increase the overall loudness of a song by reducing the dynamic range (the difference between the loudest and quietest parts of the audio). This can be achieved using a compressor effect, which reduces the volume of the loudest parts of the audio signal while leaving the quieter parts unchanged.

Equalization: Adjusting the balance of different frequencies in the audio signal can also affect the perceived loudness. Boosting certain frequencies or cutting others can change the way the audio sounds and make it seem louder or quieter.

Stereo Imaging: Adjusting the stereo image of the audio can also affect the perceived loudness. For example, making the audio wider or more focused can alter the way it sounds and make it seem louder or quieter.

Limiting: Using a limiter effect can help to increase the overall loudness of a song by preventing the volume from exceeding a certain level. This can help to increase the perceived loudness of the audio without actually increasing the volume.

It's worth noting that increasing the perceived loudness of a song too much can result in audio distortion, which can negatively affect the quality of the audio. It's important to find a balance that sounds good and avoids distortion.

[NewUser]

Originally Posted by ggrey

Are you limiting or compressing the maxed volume on your master or buss? You may have peaks that are maxing the volume level that do not strongly affect LUFS number.

I'm using compression, but I will take a look into using limiting.

Originally Posted by bFooz

Limit/clip drums separately before summing into LR.

I'll give this a try.

[Lynx_TWO]

I’ve used limiters on drum tracks before, but in comparison I’ve always found it’s just more dynamic sounding if you reserve the limiters for the mastering stage…

[ggrey]

Originally Posted by Lynx_TWO

I’ve used limiters on drum tracks before, but in comparison I’ve always found it’s just more dynamic sounding if you reserve the limiters for the mastering stage…

Dittos. My master FX includes in this order: Compression/Limiting/SPAN (spectrum)/LUFS(meter). Works great (IMHO) and as Lynx says, used appropriately, improves dynamics.

[Lynx_TWO]

What limiter do you use? I’m using Fabfilter’s L2, set it to 4x oversampling, run one pass, calculate the difference in LUFS (so for example if the mastered track comes out to -15.7 LUFS, -15.7 - -14 = I just set the threshold at 1.7 dB with TP at -1dB) and second pass crank it to 32x oversampling if it’s knocking off more than 1dB or so… default settings are pretty transparent even if it’s knocking off 4dB… but there’s different modes to color the sound if you want.

[antiClick]

Originally Posted by Lynx_TWO

-1dB True Peak is enough to account for inter-sample peaks when converted to a lossy codec. There’s zero point in going louder than -14 LUFS since the streaming service will just turn it down when you upload, making your mix sound thinner and less dynamic than everyone else, instead of louder. Apple Music actually normalizes louder tracks to -16 LUFS. You shouldn’t be looking at loudness for your reference track; all you need to listen for is balance and EQ. I’d actually recommend normalizing all your reference tracks to -14 LUFS for better accuracy and to avoid falling for the natural inclination of the brain to think that louder sounds better. By keeping your reference tracks to their integrated loudness without normalizing to -14 LUFS, you’re actually doing it backwards.

Almost 100% of professional clients expect a -10 to -6 LUFS file depending on the genre. I also give them -14 files, but they don't publish that. Don't ask me why.

If you want it louder without much distortion try a clipper before the final limiter. Or try clipping the drum bus, then a limiter on the master. How much loud it can get also depends heavily on the frequency content.

Just don't overdo it. If you don't have the experience I'd just obey the standards and leave it at -14.... but it's not mandatory. It depends heavily on the style.

[antiClick]

Originally Posted by NewUser

Hi,

I've got the volume maxed. So, I'd have to lower things to raise things... I currently am at -12 LUFS and I'm trying to reach -10 LUFS with -2.5 max true peak to match my reference for a cover song.

I am thinking I need to either add another instrument or some how lower/raise EQ levels... not sure.

What would your methods to raise LUFS or any tricks you know?

If I understood you correctly...
Your master volumen should always stay at 0. You can insert a JS volumen plugin to
rise/lower your levels before the mastering chain. Then calibrate the thresholds according to your LUFS target.

Usually a song with fewer instruments will be able to reach higher loudness before distorting. Equalization also plays a key role, and some people mix againt a limiter to automatically adjust for that.

When you start mixing try to start with the rithm part so that it reads about -20 RMS on the master track, that way you will have plenty of headroom for the rest of the instruments before master peaks. Then, when you finish your mix, the mastering chain will bring your bus to the desired volume.

[ggrey]

Originally Posted by Lynx_TWO

What limiter do you use? I’m using Fabfilter’s L2, set it to 4x oversampling, run one pass, calculate the difference in LUFS (so for example if the mastered track comes out to -15.7 LUFS, -15.7 - -14 = I just set the threshold at 1.7 dB with TP at -1dB) and second pass crank it to 32x oversampling if it’s knocking off more than 1dB or so… default settings are pretty transparent even if it’s knocking off 4dB… but there’s different modes to color the sound if you want.

Compression: Klanghelm MJUC (paid)
Limiting: Reaper's own ReaLimit (Free)
Spectrum: SPAN by Vexengo (Free)
LUFS: Reaper's Loudness meter (Free)

Set up in that order. I tried the free Klanghelm MJUCjr for free and after a few weeks loved it so much I paid for the full version.

[Lynx_TWO]

Originally Posted by antiClick

Almost 100% of professional clients expect a -10 to -6 LUFS file depending on the genre. I also give them -14 files, but they don't publish that. Don't ask me why.

If you want it louder without much distortion try a clipper before the final limiter. Or try clipping the drum bus, then a limiter on the master. How much loud it can get also depends heavily on the frequency content.

Just don't overdo it. If you don't have the experience I'd just obey the standards and leave it at -14.... but it's not mandatory. It depends heavily on the style.

Yep, unfortunately part of our job now is to educate clients. What I’ve done is crush a master to, say, -8 LUFS, then normalize it to the same level as the -14 LUFS master and ask them which one they prefer. I don’t think I’ve had one client pick the crushed one yet… but if they did then of course, that’s what they’d get and I’d keep the “better” master for my own listening . Unfortunately, clients usually don’t understand that their music will likely be turned down to the same level as everyone else’s (they just play back the two on headphones or speakers and pick the one that’s louder since they are falling for the ‘louder must be better’ trick our brain plays) so louder isn’t more competitive like it used to be. Except for maybe CDs, but in that case the DJ will likely normalize them down anyway since having drastic volume differences in a venue is pretty annoying… and with CDs, hell, we should be able to have MORE dynamic range than streaming since it’s supposed to be “higher fidelity”. But, yes, I get that we don’t live in a perfect world lol

In addition, having a master that’s already crushed really sux if it gets played on the radio, since most stations are going to crush it down even farther for playback. Gotta educate the clients! . Most people don’t listen to whole albums these days and don’t want to keep changing volume controls for every single song…

For example, Here's a track crushed to -6 LUFS and then normalized to -14LUFS:
https://drive.google.com/file/d/1CFk...usp=share_link

And here's the same track mastered at -14 LUFS:
https://drive.google.com/file/d/1fQ9...usp=share_link

You'll notice that the -6LUFS sounds louder at first, but then when the dynamics actually start, it's really bad. Sounds like Apashe, actually

[Lynx_TWO]

Originally Posted by ggrey

Compression: Klanghelm MJUC (paid)
Limiting: Reaper's own ReaLimit (Free)
Spectrum: SPAN by Vexengo (Free)
LUFS: Reaper's Loudness meter (Free)

Set up in that order. I tried the free Klanghelm MJUCjr for free and after a few weeks loved it so much I paid for the full version.

Nice! Will definitely have to check out that compressor. Currently, I pretty much only use POWAIR by SoundRadix for compression since it's totally transparent and has some unique features such as being able to monitor the loudness and apply compression the same across loud and soft parts of a track, but always on the lookout for better/unique tools

[splitting_ears]

I wouldn't think too much about LUFS (neither confuse this scale with actual loudness) but instead focus on making the music sound good.

If the OP is concerned about a lack of density with his/her track, there are many techniques that will help getting there from the mix stage. This will lead to tremendously better results than simply pushing a mix into a limiter -no matter how sophisticated.

By the way, -14LUFS is in no way a target to aim for. If you are still confused about this topic, this thread might contain valuable information for you: Targeting Mastering Loudness for Streaming (LUFS, Spotify, YouTube)-Why NOT to do it.

Hope that helps!

[norbury brook]

Originally Posted by splitting_ears

I wouldn't think too much about LUFS (neither confuse this scale with actual loudness) but instead focus on making the music sound good.

If the OP is concerned about a lack of density with his/her track, there are many techniques that will help getting there from the mix stage. This will lead to tremendously better results than simply pushing a mix into a limiter -no matter how sophisticated.

By the way, -14LUFS is in no way a target to aim for. If you are still confused about this topic, this thread might contain valuable information for you: Targeting Mastering Loudness for Streaming (LUFS, Spotify, YouTube)-Why NOT to do it.

Hope that helps!

THIS ^^^^^^^^^^^^


M

[Lynx_TWO]

Originally Posted by splitting_ears

I wouldn't think too much about LUFS (neither confuse this scale with actual loudness) but instead focus on making the music sound good.

If the OP is concerned about a lack of density with his/her track, there are many techniques that will help getting there from the mix stage. This will lead to tremendously better results than simply pushing a mix into a limiter -no matter how sophisticated.

By the way, -14LUFS is in no way a target to aim for. If you are still confused about this topic, this thread might contain valuable information for you: Targeting Mastering Loudness for Streaming (LUFS, Spotify, YouTube)-Why NOT to do it.

Hope that helps!

Oh, I never mix to a target; I just tend to mix more dynamically, and then the master just gets limited to -14 LUFS with a -1dB TP. Some mixes I’ve done didn’t need the limiter to normalize to that target, but most of mine do. If you master at a lower loudness, say -18 LUFS, then streaming services like Spotify won’t turn it up and your music will then be noticeably lower in volume than everyone else’s. Also, while LUFS is a pretty decent measurement for loudness, it’s not perfect. For example, if you export a song played with a cello, viola, and violin at -14 LUFS using Reaper’s render dialog to set l-LUFS to -14 and the limiter to -1dB True Peak, it’s going to be noticeably louder than, say, a metal mix due to the frequency distribution. We’d have to map the entire range of human hearing and sensitivity to frequencies to get a better measurement of loudness, instead of the two filters used in the current standard. Which I feel like would be totally possible these days…. Perhaps ITU BS.1771-5 will tackle this.

[Daodan]

Originally Posted by splitting_ears

I wouldn't think too much about LUFS (neither confuse this scale with actual loudness) but instead focus on making the music sound good.

If the OP is concerned about a lack of density with his/her track, there are many techniques that will help getting there from the mix stage. This will lead to tremendously better results than simply pushing a mix into a limiter -no matter how sophisticated.

By the way, -14LUFS is in no way a target to aim for. If you are still confused about this topic, this thread might contain valuable information for you: Targeting Mastering Loudness for Streaming (LUFS, Spotify, YouTube)-Why NOT to do it.

Hope that helps!

+1! Great thread.
I'll add this video https://www.youtube.com/watch?v=pPS7g0dhEHY. Well explained too.

[PitchSlap]

Clip individual drum tracks and limit individual musical tracks to increase headroom.
I'd recommend Stillwell's Event Horizon as a light weight drum clipper, and it also has a JS version.

IMO, the idea to to control dynamics/peaks a little bit over many stages rather than doing too much on the master bus.

Also, LUFS is perceived loudness which will be frequency dependent. Made sure the mid-range is right and the rest becomes easier.

[Lynx_TWO]

Great points, and great videos!

The video @Daodan shared; the guy talks about a master he did which is here:

https://m.youtube.com/watch?v=3M37f7l-1QA

Not only is the song amazing and well produced (hugely important) his master sounds more dynamic than most stuff I’ve heard, and certainly more dynamic than a lot of stuff I’ve done. Definitely learned some things here that I’ll be learning to apply for future. I’ll download and analyze that song later since now I’m really curious to see what the LUFS, LRA, and frequency distribution is.

[serr]

The trick all the kids are doing with the streaming services nowadays is to bludgeon the mix up to -8 or -7 LUFS with brick wall limiting volume war style and then turn it down until the meter hits -12 or -14 LUFS depending on the target streaming site. You end up with 2 - 6db louder at the end than if you mastered properly to -14. This is as audibly damaging as you might guess but no one listening to pop music on some shit device will hear it. I don't recommend doing this!

This is literally what ended up inadvertently happening with a lot of streaming site sources being volume war edition CDs. Someone apparently noticed they could get more volume doing it intentionally.

Do what you think you need to do. This kind of thing leads to multiple versions of releases. Volume war version for "radio" vs proper version for discriminating downloaders. That often plays out as only the volume war edition ever getting released. Back in the day with FM radio the sound was substandard with the radio compression. But that was considered a broadcast limitation and not baked into the album masters. You buy the album, bring it home and get the full sound. Bake the destruction in for cheapness and now no one gets to hear the full production.

[Lynx_TWO]

Yea man, I’d pay money to get a hold of all the multitracks by Apashe and remix and remaster them. They are between-3LUFS to -5LUFS. With a TON of audible, nasty distortion. Can’t tell me that that’s a “creative decision”. I used Stereo Tool to declip and then natural dynamics to restore punch, then used Gullfoss to reEQ them, Soothe2 to get rid of resonance, and turned up the vocals by 3 dB in RX Advanced Music reBalance module to account for the resultant loudness increase of the instruments. The results were FAR less fatiguing to listen to and even a friend of mine who I did it for, who’s in no way into mixing, said he could hear a big difference.

And yea, Radio stations are already going to mangle your master so you might as well give them a mix with more dynamic range so it’ll have less damage.

[serr]

Haha yeah, when someone can use restoration tools like blunt declipping and improve the sound of a full program master, you should feel very bad and know your work is absolutely the worst!


We're all trying to bring clear and focused sound. Raising the volume far into distortion and bionic treble boosts do kind of ring your bell with that! Gets fatiguing though and reduces your audience to only those in your genre. Our ears are still tuned to midrange too. Anyone remember midrange?

Clickbait: There's this one trick mastering engineers hate to have you know. The volume control! You can make your music louder by turning this knob clockwise.

[Lynx_TWO]

Originally Posted by serr

Clickbait: There's this one trick mastering engineers hate to have you know. The volume control! You can make your music louder by turning this knob clockwise.

Ha! I’m totally gonna make a YouTube video “Learn this one neat trick to increase volume on your mixes!” And then it’ll just be my hand turning up the volume control on various amplifiers

[NewUser]

So, I ended up EQing the vocals more, adding dynamic volumes for chorus and break, adjusting the loudness of the bass and kick.

The final touch, I removed Waves Audio WLM Plus and replaced it with SONIBLE Smart:limit which I feel made a huge difference.

Reaper says the master came out at -10.5 LUFS with 2.7 LRA ... While SONIBLE said -11.1 LUFS with 3.5 LU. Both agreed on -2.5 true peak.

Sonible settings for Spotify (LOUD) is -11 LUFS with -2.5 true peak. So, instead of trying for -10 LUFS.. I accepted -11 LUFS.

[Lynx_TWO]

Originally Posted by NewUser

So, I ended up EQing the vocals more, adding dynamic volumes for chorus and break, adjusting the loudness of the bass and kick.

The final touch, I removed Waves Audio WLM Plus and replaced it with SONIBLE Smart:limit which I feel made a huge difference.

Reaper says the master came out at -10.5 LUFS with 2.7 LRA ... While SONIBLE said -11.1 LUFS with 3.5 LU. Both agreed on -2.5 true peak.

Sonible settings for Spotify (LOUD) is -11 LUFS with -2.5 true peak. So, instead of trying for -10 LUFS.. I accepted -11 LUFS.

That’ll work!

[serr]

Originally Posted by NewUser

Sonible settings for Spotify (LOUD) is -11 LUFS with -2.5 true peak. So, instead of trying for -10 LUFS.. I accepted -11 LUFS.

So around -8 LUFS and then turned down 2.5db until the meter hit only -11 LUFS. I guess, do what you think is working. You might want to normalize the smashed master to compare it with your original just to hear the difference and see if you still think this is a good idea.

Volume war technique is still alive and kicking, folks! It's evolving!

This is only going to lead back to dual inventory. FM radio used to be a lo-fi thing with heavy compression but you knew when you bought the album and brought it home it would sound normal. Making masters with loudness baked in instead of doing that on playback where appropriate leads to dual inventory.

[Lynx_TWO]

Originally Posted by serr

So around -8 LUFS and then turned down 2.5db until the meter hit only -11 LUFS. I guess, do what you think is working. You might want to normalize the smashed master to compare it with your original just to hear the difference and see if you still think this is a good idea.

Volume war technique is still alive and kicking, folks! It's evolving!

This is only going to lead back to dual inventory. FM radio used to be a lo-fi thing with heavy compression but you knew when you bought the album and brought it home it would sound normal. Making masters with loudness baked in instead of doing that on playback where appropriate leads to dual inventory.

From what I gathered, the OP put out a cover of another track that was smashed to -10 LUFS, and on the off chance a user had disabled volume normalization in Spotify, he didn’t want the cover to be noticeably quieter than the original. And yes, Radio is STILL a low fi thing that still smashes even loud masters (maybe digital radio is different?) so I always encourage people to just master what sounds good and then just export at -14LUFS with TP of -1dB in Reapers Render Dialog to make sure it won’t be noticeably quieter when played back on streaming services.