Mini Masterclass: Loudness Targets for Mastering (...It's not what the streaming services tell you) 

Thanks so much for checking out the channel! Glad to have you here :-) -Justin

Absolutely! Spot on. Thanks for sharing your track and your take.

Well said. Often the low end takes up so much energy and “space” without contributing towards perceived loudness

Just don't pan it in mastering :D

Very true!!

Absolutely. Mastering can make it louder, but the best and loudest-sounding clean masters come from arrangement, recordings and mixes that were crafted with that in mind.

Loudness is also in the production, arrangement, sound choices etc....

So awesome to hear, thanks Francesco! -Justin

Awesome to hear! So happy to be useful. -Justin

Why thank you brother, good to have you watching! This video must really be getting out there :-)

Thanks! Great to hear. So glad to be helpful!

Awesome to hear, thanks for tuning in! -Justin

Great idea. Will see of I can hit that sometime. The short answer is that once you’ve mastered one track you can now throw away all your other references, and just use that first mastered track as your primary reference for loudness and tone, mostly by ear. (Although you also want to reference adjacent tracks against eachother.) The long answer would discuss a lot of the issues you may run into trying to do this, particularly when tracks have wildly different arrangements, instrumentation and production standards. Sometimes, you’ll have to avoid pushing tracks that are easy to make loud as loud as they could go so that they will sit properly next to tracks that won’t. Sometimes you have to make the most bombastic tracks a little less bombastic so you don’t overwork the less bombastic ones to much. Hope that helps! -Justin

@@SonicScoop Thanks Justin, that does help to think of it that way. Cheers!

The best way I’ve been able to reference is to yes buy from iTunes or Amazon, then import the track directly into the mastering session. So that you hear both on the exact same speaker setup. Unfortunately you may have to manually record the song into your session. But I find having both (or more) songs in the session easy and straightforward referencing 💯

True story! You are not alone. You don’t always need to go that loud, though it certainly can work when done well. But there is definitely good reason to go louder than -14 in most genres.

@@SonicScoop one hundred percent 💯👌

Exactly! Mastering directly for -14 lufs never sounded equal to another song that was mastered louder and then turned down to -14 lufs. Dont follow these "temporary streaming services" because when streaming is out and a new way to listen music comes about, your song is gonna be compromised. Learn your own music genre and master for what the norm is considered in that music genre. If your genre calls for -16 lufs, master for that. If your genre calls for -5 lufs, master for that.

Last track I mastered was in fact around -5..so for my taste normally a bit too much. BUT it sounded good in wav or 320 MP3 - as soon uploaded to YT sounds shitty :( The track however was not mixed flawless and only 16bit, I guess with 24bit, better mix and more headroom I could be much better.. also the YT algorythm seems to be quite sucky - I mostly perceive Soundcloud as better (?). Spotify "preview" version also seems worse..

@@SonicScoop I also uploaded with "tunes2tube" is it possible this degrades audio even if they say it does not?

If you’re going to a mastering engineer shooting for -14 average level or lower is great, unless you’re loving the sound of going hotter. Anywhere from -22 to -18 average level for mixing is very safe if you want to leave loudness and most all the bus processing to a mastering engineer. But you don’t have to go that quiet if you’re getting great results at a hotter level. EDM mixes might be louder right out the gate, sometimes more like -12 to -8 even before mastering. Sometimes hip hop can be on the louder side at the mix stage as well. Honestly, the peak level doesn’t matter TOO much because the mastering engineer can just turn it down. Some people will give specific targets for that but I don’t. 0 is fine for peaks as long as it’s not more crushed than it should be.. -1 is fine. -3 is fine and starts to make it unlikely that I have to turn it down to improve my gain staging which is nice. -6 is fine but a bit conservative. Ultimately, the output of the final master wants to be lower than 0 on a True Peak meter. If you’re not using a True Peak limiter, this could require shooting for a lower peak level on a traditional limiter. Some streaming services recommend going for a True Peak level of -1, (or as low as -2 for louder masters) but I’ve found that this is pretty conservative and you can get distortion free results out of the codecs at higher levels than this. Hope that helps, Justin

I know this is an old comment but I've recently had the same issue (as most other beginners). I made sure I hit what was the sweet spot (-14) and my track was much lower and I noticed the dynamics were out. As you say, I realised it was about relationships rather than a fix all lufs number. I'm learning a lot. Important points, thanks!

Haha! It’s funny but things actually seem to be moving in this direction 🙈

I would say it depends. Whichever is the best set of speakers. The word "monitor" is more about marketing 2-way powered speakers to the studio market - it shouldn't be assumed to mean "better".

Do you mean ADPTR Audio Streamliner? It’s a good one!

@@SonicScoop yes that's the one. 😝

I love dynamic EQ. I’m more likely to use it in place of traditional EQ or in place of a multiband compressor or de-esser. I’m looking for something different out of a traditional compressor in mastering. That’s more about overall envelope shaping. Hope that makes sense! -Justin

You can turn off the volume normalization on services like Spotify and Apple Music. You can also play downloaded WAV or MP3 files in your DAW or music player app. It’s good to level match so you’re not swayed by loudness alone. Hope that helps! -Justin

@@SonicScoop level matching always 👍

I habe both and tbh ProQ3 is just the best surgical EQ ever made. And i got L2 to just have more possibilities than in the Ozone Maximizer. But stuff like Ozone Imager 2 and Exciter I also couldn‘t live without while mastering so yea...get both😂

They’re both great. Where they have comparable tools, it’s going to come down to personal preference. I think the strengths of each are that Fabfilter has the visual interfaces down pat, with super responsive, attractive and fast GUI designs, often with a lower CPU load than iZotope, while iZotope has even more mastering-focused tools, a subscription option and its own standalone apps as well. iZotope’s Ozone has the imager that FabFilter doesn’t have, and RX has all the noise reduction stuff that doesn’t exist in the Fabfilter world yet. But then again, you can can get the iZotope imager from them for free these days :-) If you’re going beyond mastering, FabFilter has its own standalone delay, that iZotope doesn’t have yet, as well as dedicated filter plugins. They both have synths with totally different vibes and iZotope has Vocal Synth and Break Tweaker as well. If you were to buy EVERYTHING, iZotope has more stuff (minus the dedicated delay) so it’s going to be more expensive if you want every last plugin instead of a smaller bundle. But they both have smaller bundles too, though iZotope has subscription options that FabFilter doesn’t have yet, so its hard to say there as well. Ultimately, they are both great, and different people will prefer different brands. This is my first time using FabFilter and I’m blown away by how good the GUIs are and really impressed by how great they sound. I’ve been using iZotope stuff for a decade or more and really love them too, and don’t plan to stop. But I’m DEFINITELY adding these FabFilter tools into the rotation. As far as features and results there are things I like better about each of their compressors, limiters, EQs and saturators, which are the tools I use every day, so it’s really hard to say! I just can’t give a clear winner for any of those tools. They’re both just some of the best in the business. Ask me again in a year and I’ll have even more to say I’m sure :-) Hope that helps! -Justin

@@SonicScoop thanks for the thoughtful reply,..... to your point, there is something about the izotope gui that just feels heavy and clutter and fabfilter that feels snappy in comparison. i have the Izotope music production suite, and it is good, but i find myself only using certain apps like ozone. i never got used to neutron and just go with ableton stock eq, and imager is great, but somehow the other apps didnt "stick". i also find there is a lot of gimmicky things in izotope. i will give fabfilter a try, and ..... we know how the movie ends.... will have both :)

Listener's perspective: if a song is too quiet on average, you can turn the knob and enjoy. If the song's brickwalled, you can't do anything about it, it will sound choked. Yes, louder sounds better, but as far as Spotify etc. punishes the excesive loudness for everyone it's a fair game. I believe the experienced engineers can adapt quite fine.

Yes, I think it's a positive good. It removes the incentive to push things just a little bit louder to beat the next guy at the detriment to your own sound. And, it makes it easier to listen to very dynamic music along side music that isn't. It's just an imperfect fix. There will always be ways to "game" loudness, and we can't ignore that there may be some benefits to going louder than the recommendations for some music. I think it's working. It's not working "perfectly". But we can't expect it to. It's still better than what we had, I think! Hope that makes sense, Justin

Classical music is an example of a genre that doesnt require loudness. Certain music genres are built on loudness because that loudness is a sound has been for the last 50 years. Thats what a volume knob is for.

@@16bitworld2 I was pointing out so that people understand what LUFS integrated is. To exagerate and for them to hear what it really is. These target people are saying of best way to get loudness is not true, because it wont be true everytime. They keep saying do this and that as if they are still peak metering, when its like EQ "Program/Material dependent. That it always have different results based on the program/track at any target. a song more dynamic will have higher Peaks than A squashed one. Hence I pointed out here also that Just go for the sound you want.If that heavily limited sound, then go for it the tone matter than the loudness because you wont get the same results anyway once its normalized. LUFS is not what most people think as if its one target will result the same for everything. Thats why in film this is just our target or at the back of our head, while mixing the final output as we ride those faders, not after the fact. It gets tricky for music. Plus I dont know if Spotify really use the proper ITU R bs 1770 for LUFS. Its wrong to say that -10 LUFS integrated as the target will give you a louder master than -14. Cause again it depends on the program itself. if your sound is squashed already mixing, just dont bother with LUFS.

Yes you should! Especially if you are not using a True Peak limiter. But in this case, I knew that I wasn't going to play back the audio samples normalized to 0dbFS so it didn't really matter :-) I believe the files are turned down by a couple dB in the video editing session, and then the final output of the video is like -1 or something, so it's not a problem we run into here. But good point, and something I probably should have explained.

The levels do jump around in this video, but that’s by design. The idea is to hear what these different target levels sound like, without normalization as well as when level matched. The first clip is the loudest one. If we matched the dialogue to that one, by the time we got to the quietest one, the voice would be much too loud relative to the music. Apologies for the jump in the first master but we do give fair warning that it’s coming :-) Hope that makes sense, -Justin

@@SonicScoop Understood. Thank you Justin for taking the time to respond.

Yup, the bassier the track, the harder it is to make it be loud and sound loud without making it sound worse. A lot of mixes may have more sub energy, low energy or low midrange build up than the references without the mixer knowing it because their system is compromised, AND because it’s hard to listen to a song at an appropriate level of brightness for as long as it takes to mix without fatiguing. This is probably the case in 70-80% of mixes that come in for mastering, so often, EQ is a big part of getting it loud and clean, like you heard in the first track especially. Things like using sidechain EQ on your compressors or limiters can help with this too. That said, you dint necessarily have to “beat” all your references these days. That’s something that streaming services are absolutely good for! But within reason. If you master a techno track to -14 you’re probably doing it wrong. However if you master it to -5 and it’s sounded way better at -7, you’re probably also doing it wrong :-) Hope that makes sense! -Justin

@@SonicScoop Ending up at -5 seems to be impossible since my speakers would explode because of the distortion the low freqs cause. But yeah, using appropriate EQ and compressing already when mixing is definitely a good foundation for a louder master.

Yeah, you don’t necessarily have to go to -5 to have a great sounding master. Sometimes going louder is counterproductive. There is definitely a point of diminishing returns. If you’re getting the sound you want at a lower level than you are in good shape! Though, with the right strategies, it is possible to get cleaner and louder than you might expect. But that might not be necessary. If you ever wanted to hear where another mastering engineer might take it, you’re always welcome to send me tracks to work on :-) Have a good one, Justin

@@SonicScoop Thanks for the offer. Actually i'm not a mastering engineer, but i mix and master all my productions by myself, and am therefore very interested in all these topics.

@@SonicScoop How much would it cost for a 1 track "try out" master to compare yours to mine ? I also make more "technoish" music (mixed with oldschool trance / industrial I would say)

It’s tricky to say because these things are so subjective. One man’s overcompressed travesty of a record is another man’s badass bombastic impressive beast of a record. (Even after volume normalization.) But yes, there is definitely a point for any record where additional loudness will just make it sound worse. Fortunately, the streaming services make it so this is no longer necessary. If record sounds great at -12 LUFS and annoying at -9 LUFS, or if one sounds impressive at -10 but worse at -8, or awesome at -7 but distorted at -6, then there’s no reason to push it to the higher value anymore, just to make it loud. So streaming services HAVE succeeded at that. I think that’s awesome. People ARE backing off on loudness now. Just not all the way. But there may still be good reasons to master a rock record to say, -12 or -10 or -9 instead of -14 in going for a sound you actually love. If the artist actually LIKES a more limited sound, even when level matched, there may be reason to go louder than that. But if they’re pushing it just for loudness alone, it’s a mistake. I agree. But whatever they choose, it’s almost inevitable that at least SOME of their fans will disagree :-) So there’s always that! I hope that makes sense. -Justin

@@SonicScoop It does make sense. What sounds good is subjective. We're trained listeners to think that an over-compressed crappy mix is "bombastic." The vast majority of people don't listen to music on a good system. Often these days, people are listening on a mono bluetooth speaker. The horrors of it all! Maybe good mixing will have a small comeback the same way vinyl did. At least the streaming services are trying to help.

Who are you to tell an artist how they should make their own art? If kings of leon wanted their album at -7, thats their artistic choice. If you disagree, then you go make your own album and make it as quiet and as dynamic as you want.

If it sounds good, it is good. If it doesn’t, first do no harm. But if needed, operate! -Justin

Absolutely. But how much is enough and how much is too much? And how can you get things to sound louder while staying dynamic? We go deep on some of that here. Loudness isn’t only about limiting. It’s about EQ balance and other factors as well. Hope it’s helpful!

Haha, yes, the levels do jump around in this video, but that’s by design. The idea is to hear what these different target levels sound like, without normalization as well as when level matched. The first clip is the loudest one. If we marched the dialogue to that one, by the time we got to the quietest one, the voice would be much too loud relative to the music. Apologies for the jump in the first master but we do give fair warning that it’s coming :-) Hope that makes sense, -Justin

imho EDM is great for various purposes, but it lacks a human connection because lyrics are minimal or non-existant, so it's kind of like watching a movie without a protagonist.