Mastering - Loudness war still going strong on Beatport and Soundcloud?

I’ve finally started falling back into a groove of actually making music regularly. I’ve been out of it for a while and started actively looking at what has been going on in the world of mastering. I tend to do my own mastering and was happily surprised at how nice and easy it is to make good sounding masters for Apple Music, Spotify, Youtube, etc… Maybe pushing it up to have it turned down by maybe -1db to -2db (for those services that don’t turn it up at all) and it still sounds great at those levels.

Then I started looking at tracks on Beatport and Soundcloud (I make Dance music so I have to master for those services as well). I found lots of tracks with integrated LUFS of -8 and even momentary up to -6 LUFS!?

I’m having to run a master compressor into a mastering limiter and maybe even a 3rd limiter to reach those levels without distorting (sharing the load so no particular limiter has to do all the work). Then I’m fighting with losing some low end and trying not to let anything get too harsh :confused:

I was feeling all happy about the current state of the Loudness wars until this :face_with_thermometer:

I want things to be more dynamic, but I get that it might just be still pushing things to the limit(er) and how make things sound better while doing that. I wish it wasn’t so but unfortunately it is.

Anybody have thoughts on this? Even if you aren’t doing your own masters, what are your mastering engineers giving back to you for your tracks heading to Beatport or Soundcloud?

1 Like

Just heading to bed so apologies if these videos don’t help. I seem to recall Dan explaining loudness, in a modern streaming context, quite well in these two videos:

Also:

As a resource for current loudness levels for various platforms.

6 Likes

I was about to post this as well hahaha.

1 Like

Thanks for those (they might help others get up to date on that stuff).
I have no issues with the regular streaming services and understand the loudness penalties and even really like what’s going on there. I’ve even already came up with a way of mastering for those services I that I like to use.

It’s Beatport and Soundcloud specifically who don’t have any LUFS standards and so everybody is still fighting to be the loudest. Definitely in the Dance music and Pop music worlds at least.

Haha, awesome! I love it when I’m on the same wave as someone at the same time :wink:

Didn’t @sl1200mk2 start a mastering service a while back? Maybe I should pull you into this conversation? :stuck_out_tongue:

I guess I want to hear some thoughts on how mastering engineers are approaching Beatport and Soundcloud releases. I get that it’s a balance of pushing it as hard as you can while trying not to destroy the track. But maybe a fresh or experienced perspective might help for me to feel solid in whatever I decide to do for those services (besides cry and whine about why they haven’t introduced LUFS standards too).

When I see a discussion about mastering, I get out the smoking jacket and pour some Cognac. They always bring some insight and intellect. Not like those uncultured swine talking about a synth that hasn’t even been heard or seen yet. I digress.

7 Likes

:sweat_smile:

…don’t relate to soundcloud…that’s full blown up with reremastered rerecorded dj sets and and heaps of hobby if it’s louder, it must be better attitude…u can drop all comparison to that…no need…
and beatport…is no streamingfarm…it’s a dj next tune please marketplace for the hot shit from last week that’ll be gone tomorrow but ooops what will i play tonight…

loudness war ain’t over…but it got way better recently…
be a bit louder than asked for by spotify and apple music…but u really don’t have to crush and squeeze it anymore…

if u got damned good gear with dead solid oversampling rates, u can of course still go even beyond all limits…but those results again, are only for those who think louder sounds better, which is a neverending story, we’ll never ever gonna get rid off in total…
while hey…sometimes it happens…but it’s a rare case, where overblown actually really worx well on a track…but that’s nothing u should aim for…

master for a clean, warm and fad lowest octave that carries the mix as a solid foundation, for quite some overall transparence, accurate spikes in transients and a crispy but never too harsh top end…and u’ll be fine…

and well, most of all this is not to achieve by mastering…it’s all in the arrangement and the mix in first and second place…that leaves enough headroom for a good final master treatment…and to leave THAT to some second, fresh ears is always a good idea to consider…even in right into ur face dance music of all kind…

Thats kind of what I figured out by mastering and matching to some stuff I liked. I went up a couple of db higher than what they specify and was able to match my master pretty well to things I like on the bigger streaming services.

1:1 comparison (volume) of what I had to do post mix for mastering showed no noticeable changes in the dynamics of the track (and no distortion). I’m also doing careful lookahead, attack, release adjustments to keep the dynamics and not gain any distortion. Very easy for Apple Music, Spotify, etc…

Agreed and something that should always be considered when you can go back and do something in the mix (such as my case).

I can see that being beneficial. Though I still have to decide if I like the master and what needs to be changed if I don’t (even if I’m not the one doing it). I also consider my reference tracks as a “second set of fresh ears” too.

I may end up having some stuff mastered but I plan to get on a pretty consistent schedule (life permitting) and could see it costing quite a bit each month. So I will probably save those for the ones I don’t feel like I can quite get there.

Thanks for chiming in, everything you said is solid and makes sense :slight_smile:

Did a lot of sharing about my thoughts about this. Bottom line I don’t go along. I want my tracks to sound as I intended and not squeezed to one crappy waveform so it sounds like them bloody commercials between tv programs. If someone wants to hear it loud they can use their volume knob. I ask a mastering engineer to back down, they like that because they care about quality. If you find one that don’t he/she lacks passion for the job. We all should rise and start the dynamics war. The good always wins from the bad :grin:.

4 Likes

I’m a mastering tech by the way, that’s what I do as a main job so feel free to contact me if you need any guidance.

About that loudness war thing, I’ve stopped caring a while back, but even more now with all those streaming regulations. When I know something will by played mostly online I make the best sounding around -14 LUFS, -9 for Bandcamp releases, Cds and vinyls, uneless i’m asked otherwise by clients.

Soundcloud is probably the WORST sounding platform of all so you can avoid taking it as a reference.

8 Likes

Yes to this a thousand times hahaha

1 Like

Thanks for doing your part <3

Thank you :slight_smile:

I think I can probably handle -9 even though I’d prefer like -12 to -10 tops. I have a few (software) mastering compressors that have been doing a great job at getting me up to -9 without any significant or detrimental changes to the music. Thank you for your input, It makes me feel better about reaching realistic levels without having to sacrifice a ton.

Edit: I know someone is going to ask (or at least wonder) about the gear (it’s Elektronauts) :upside_down_face: : SPL Iron and more often the Shadow Hills (red) mastering compressor into Fabfilter L2 (32x oversampling). A lot happens before that though, it’s just the loudness part that those have been helping with.

@Stazma : Are you leaving final masters at -1 dBtp for Bandcamp, CD, Vinyl or still something like -0.3?

Usualy I leave those at -0.7, wich I think is a fine in between.

1 Like

The venerable Craig Anderton explains it pretty well here:

4 Likes

Haha… happy to contribute. My preferences and what I’ll do for the artist are generally two different things. First and foremost, I’ll always try to do what’s best for the material. My personal preference for dance tracks is around -9 LUFS for the most part. I really just try to give the track ‘what it needs’ and the combination of gain staging my analog chain and then my commonly used plugins in the DAW with some limiting (generally 1-2dB and no more than 3dB) that’s generally where I land. I’m not just looking at values though and trying to reach or not reach a certain number… I’m listening to the dynamics of the track and what’s going on. It also has to retain a feel and groove.

That said, when it comes down to it I’m going to do what the client wants. Thankfully, most of the time I’m able to do what I think is best and the people I work with trust me for that. However, I do work some that want it a certain overall level and that’s what I’ll get them (as cleanly and best that I can). Ultimately, it’s not my art/music and not about me… so, while I’ll absolutely make suggestions (and sometimes even give two versions) if an artist wants ‘X’ then that’s their call.

Just about any decent tools ITB can get you -9 LUFS and it’ll sound good. We’re spoiled for amazing tools these days. Ideally, I’ll get a final mix coming to me around -20-22dB FS. I prefer to get a much gain (incl. color/flavor) in the analog domain as I can, but it’s just preference at this point and not absolute requirement. Once back ITB, I’m doing any last compression (if needed) and EQ’ing, then maximizing and/or limiting to reach the desired result.

I hate to say it, but if you really want to do it right, there’s generally not just any one set of tools you can use for every track. You’ll still get good/great results most of the time, but if you’re chasing that last 3-5% (whether that’s quality or level) the final tools such as choice of limiter are a big deal. They all have their own sound and are very material dependant. At this point, I’m almost never just using one limiter (as in brand) – it generally changes from track to track.

If I could only have one limiter it would be FabFilter Pro L2, but I go between all of these:

  • FabFilter Pro L2
  • Ozone 9 Maximizer
  • Voxengo Elephant
  • NewFangle Audio Elevate
  • Sonnox Oxford Limiter

The DMG Audio stuff is excellent (among the best) in terms of quality, but for whatever reason I just can’t get along with the workflow and get lost with it, so I don’t use their tools.

If you want to get ‘loud’ or ‘stupid loud’ then a clipper is an absolute requirement. I settled on Sir Audio StandardClip a while ago, but also use Voxengo OVC-128 quite a bit too. Most clippers are similar and many of them will null with each other. As long as you’re using something with decent oversampling, then you’re good to go. Generally, I’m just clipping the absolute peaks (usually the snare hits), so the limiter is reacting less overall to those.

If you want stupid loud (-6 LUFS or lower) and even attempt to retain some level of quality, then you absolutely have to clip (2-3dB or more) and use multiple limiters with each handling 3dB of reduction or more. IMO, there’s just no way around it. Trying to do it all with just a clipper and single limiter it’ll sound pretty bad.

I could go on and on, but this post is getting long enough. If anyone wants to ask me more or about other tools I’m happy to reply.

Last bit… I love Dan Worrall’s videos, so here’s a recent one that’s relevant. He’s a bit harsh in this one – not entirely wrong, but… harsh. :grimacing: :sweat_smile:

10 Likes

I like these videos but he needs to stop gating his voice like Jason Voorhees found a brand new machete and a dozen teenagers necks. While not as damaging as “idiot djs”, this is really unpleasant to hear.

1 Like

Haha, Dan Worrall not pulling punches.

It is true though, too many idiots with no knowledge out there perpetuate this kind of thing.

1 Like

I end up with around -10 to -8 LUFS for DJ material. It’s a sweet spot derived from vinyl actually. The loud 45rpm cuts.

2 Likes

Could someone confirm to a newb like me, generally when people talk about LUFS, they are referring to Integrated Loudness, i.e. measured over the entire track, correct?