Hi All,

For context, I have learnt mixing and mastering from a more traditional background, -9db or -6db headroom in your mix sorta to bring it all up in mastering kind of deal. However, I’m struggling to understand how contemporary styles that feature clipping / heavy distortion maintain loudness on streaming services like Spotify or Apple Music.

Example Track 1

Example Track 2

Example Track 3

I can get fat-sounding mixes when redlining the master and clipping overall (in Ableton, if its relevant), but I’m confused about how this works in mastering. Clipping individual instrument groups (drums, bass, etc.) doesn’t achieve the same ‘crunch,’ and I struggle to maintain this sound in mastering.

I’ve considered that maybe these artists don’t master their tracks, but their waveforms still show more dynamic range and sound louder than mine. Can anyone recommend advice or resources on this type of mixing and mastering? Any help is appreciated.

 

First of all, the tracks contain distortion and contrasting elements (during production/mixing, that don't overlap/mask too much) which ups the percived loudness. Loudness is not just making the "waveform looking like a sausage" but it's also about making things be percieved loud in the areas where we are most sensitive (see Equal Loudness Curve, Robinson-Dadson).
It's also about contrasts - something can't be impactful if it's loud all the time.
The third track is the most distorted of them all.

Secondly, mastering usually contains multi-stage dynamic processing to up the overall loudness (when needed) - not just one limiter doing all the lifting.

Conclusion: I think you are really asking about distortion (which reduces dynamics yet creates overtones/harmonics). Am I correct? Preferably do this in the production/mixing and not in the mastering stage. You get a much better result when you can control distortion on individual channels and busses during mixing, and let important impacts/transients through (which increases impact/meat/"power") with parallel processing for example.

/Baxter, mastering engineer

 

Last edited: Aug 27, 2024

I listened briefly, and I agree with Baxter. This is not about clipping the master to hell, but more about sound design and contrast in mixing. Sounds to me like they distort the hell out of bass/synths but let transients get through, vocals be intelligible etc.

edit : in the third track I hear some sidechaining after having distorted it to hell, so even if the transients are smashed, the rest of it is pumping so they still get through. I might be wrong tho, listening on laptop speakers.

edit 2 : also they all sound super smashed etc. but use very different distortions IMO, hence the sound design aspect

 

Last edited: Aug 27, 2024

I only listened to the first one. that material was already like that when handed off for final mix, mastering, etc. The easiest way to tell this, is by listening to individual elements and seeing if they are all clipped the same way and amount. Separate the elements you are hearing, because a clipper is going to apply whatever outcome to the entire track "globally". Or at least to individual groups as a sum. If you want this amount of controlled distortion, and it isn't already there before you get file to master; that ship has already sailed. Of course, you could still use your clippers anywhere you wanted but that would be for increasing RMS and not increasing peak. Not for generating this amount of controlled distortion.

You should demo the Schwabe Orange Clip plugin anyway. But it is still not going to turn a Niki Minaj song into Skinny Puppy. I would buy that clipper.

 

Last edited: Aug 27, 2024

Well, first off, in the second example only the bass is clipped. You can notice the lack of intermodulation and how harmonic distortion kinda ducks under vocals.
Second off, you're overthinking it. Just mix it as you would and clip the mix bus until the bass is clipping. Make sure to enable oversampling to catch ISP. Then true-peak-limit what's left.
Trust me the producers didn't think too hard about how to clip this.
Third, have you tried dying? There's like 66.6% correlation between your examples and dead artists.

 

Don't actually try dying. I can guarantee you it won't improve your mixes.

 

Thanks for the feedback, I can see it is distortion / clipping on individual elements which would make sense as the vocals can fit over the top without them distorting too. I can try the effects in my own music for better control.

Would you say the same is being applied in tracks like this?

I'm basically working on some tracks that clip a lot in the master (as I really push the 808s to be super loud etc.) and find I am really liking the sound with it - but yeah it does go way into the reds on the master. My only concern is that lets say I put this on streaming services, would it then be brought down to a noticably low level? I have heard streaming services do this, but I'm not too sure it still applies, as apparently 100 gecs newer stuff hits at -1 Integrated LUFS. Mine is currently sitting at around -4 Integrated LUFS. I'm wondering if

1. This will become a problem when distributing through streaming services?
2. (Playing devils advocate) Are there mastering processes that can still liven up a track if it was clipping like this.

Thanks in advance

 

That your music doesn't get turned down but 100 gecs does? If you want quieter mixes, that's a problem yes.

Liven up lol. No you don't need to repair it after you have damaged it, the music sounding damaged is part of the whole idea of the song, of the soundscape. Why would you want to repair it? I thought you want the damaged oversaturated sound?
You know, what you need to learn is how to make aggressive but controlled mixes. When a mix is aggressive but uncontrolled it sounds like something is wrong, but when it's aggressive and the power is controlled it (and it's the right genre) then it sounds really good. I don't think there is anything special going on there processing wise, if you can't achieve the sound work on your basics. Stop worrying about LUFS except if your client says so. I think LUFS are completely meaningless for the average producer and only uneccessarily confuse them like with the Reddit post or your questions, you have other more serious problems than LUFS.
You know, instead of beating around the bush you could tell us:
How were your attempts at making such masters? Show us audio examples.
Did you use any multiband or parallel processing, how did your processing chain look like? Did you just throw a clipper on the master bus and wondered why it doesn't sound as good as the reference music? Lol. I mean we don't know, you should tell us this important information instead of wasting all this time

 

im a huge fan of this stuff. clip everything, all over the place. dont pay attention to numbers, just clip it. K-clip is often used, and is affordable. OTT is also a big part of this sound, as well as izotope trash2, or any saturation, distortion really. Experiment.

As others have noted, a big part of it is the arrangement and sound selection too.

If you want a trick as to how to make this stuff slammed, clipped and squashed, but still have your transients, Clip, slam, saturate etc the master, get the master to where you want it, but leave a little headroom. Then bounce it out, put it into a new session, and grab the kick and snare from your session, line them up with the master, but zoom in and only keep the transient part of the drums. Put a little peak catcher compressor on the master, and clip it again a little bit for good measure follow with saftey limiter.

 

omg chill lmao if its a waste of time dont comment (???) im tryinn

 

Thanks! This is good advice, I feel like they do distort a lot and break the 'rules' which is makes it kinda cool right, but its just about how does this work if i then want to distribute it. I think like I have the actual sound, which is a clipping master track i want I'm just getting bogged down on the specifics of is this gonna be fucked if i upload to streaming etc etc. tbh I'm just probably gonna do whatever and see what works lol, maybe upload on soundcloud privately to see how it responds there at least

 

Yes.

You are doing it wrong. First, learn simple gain-staging. This is very important to learn. Then create a great mix where you have some headroom on the finished mix. Apply subtractive mixing (meaning, turn down things that are too loud rather than turning up things that are too weak). No need to clip the masterbus/mixbus nor "mix/master towards a loudness goal". That's not how you create percieved loudness per se. You have 144dB of dynamic range in 24bit. Use it (let there be some headroom on your unmastered mix) and let the mastering engineer do the final processing needed.

If your mastered track is loud then yes, it will be turned down in volume (to meet the platform's loudness specific, which is usually/default at -14LUFS). That's just how loudness normalization works. Soundcloud and Spotify in your web browser does NOT apply loudness normalization. However, 85+% of the Spotify users use the default loudness setting (-14LUFS) on the Spotify app.

1. No. It's part of what loudness normalization is, which is a tool to minimize the need to reach for the volume button/knob all the time as a user.
2. I don't know what you mean with "liven up". Just make your mix sound as good as possible and the mastering engineer will do his/her thing to make it sound as good as it can on multiple playback systems.

 

i usually go for the "all in the purple", an "all in the red" meter style inspo from the old ogs somewhere. The more it distorts the better they like it.

 

Bro, just create 2 busses right before the Master channel, CLIP bus and CLEAN bus. CLIP bus can act as your "master channel" if you want with all the clipping/saturation processing. CLEAN bus can have just a copy of the kick/snare/vocal/whatever peppered in to your liking. Should save you a lot of work.

 

this can work for a lot of things. I love parallel processing saturation, buttttt its completely different than placing JUST the transients of your main drums back in.

 

So this is what comes from a traditional background now. And i thought -16db headroom is already squashed

 

Honestly, I just distort and then use a transient shaper to bring it back lol but everythings worth trying

 

this can also work! try my method, its crazy the amount of punch you can achieve. Im not sure where i first heard this method, but i wanna say it was from Noisia.

 

how about compresing and then bringing back the transients. Wait what? Im confused...

Edit: Trolling but actually gonna try this.

 

this trick is very good in electronic high LUFS genres, I do it the same way, my chain is:
on both snare and kick Eventide Omnipressor plus a clipper with 1 dB GR

Omnipressor has a couple of presets that sounds fire for transients tratment, just have to be careful with the gain staging, tweaking the right input level is the most important thing. Very rarely I experience phase issues, just be careful and check it.

I also have a chain on my stereo buss to do the same trick without making another project, but straight while mixing the whole session