yes, but who cares. if you mix music properly you can not win the loudness war against people who design synthesizer sounds in order to get the most out of a certain LUFS and Peak by taking advantage of the weaknesses of the LUFS calculation.

if you are destroying your music yourself using a limiter you could as well leave that same process to youtube.

unless your LRA is not below the lowest standard of a platform*), just do not use a limiter.

use stuff only when required. mastering is not a creativity experiment.

*) youtube and some others do not even have a minimum at all.

 

Last edited: May 21, 2026 at 12:41 AM

It's a little difficult to explain without bringing other ten things into play here, the best way you can understand that is that the volume you add to the track reflects into your speakers/headphones as more volume, but if you turn your speakers down, the actual loudness of the file hasn't changed.

That's exactly what I've been saying since my first post. If you want to compete, you need to get as loud as possible without compromising the sound. There's no reason to target exactly -14 if you can cleanly go beyond that, unless you really want that dynamic range, and even that is genre-dependent. If you make a metal track at -14 exactly and put it on Spotify, you're going to be quieter than every other metal track, even though it's the same integrated LUFS at platform level.

In the end, you should just get the most of your tools without compromising the vision of the song. No one wants to hear an acoustic guitar crackling because someone pushed it too hard to achieve a -6 reading.

 

The best transient response i even listened in my whole life was made using a total of zero limiters and compressors on the Stereo Buss. Thank you sir George Massenburg for the huge legacy on how real expensive audio work should sound!

 

No disrespect, but this is a weird post. What are you talking about, bro? Is this real?

 

It is a pretty typical example of low-effort content farming. I didn't even notice the Safari Audio Meaw plugin mention, I thought he was trying to use some of the usual AI/ML stuff that might be expected, like Sonible plugins. Who uses "guitar pedal-style" plugins for this sort of stuff? If the idea is to just cut some corners, stick that Musik Hack Master Plan plugin on the 2-bus and call it a day.

 

it is not (only) his fault that he got misleaded by all the nonsense you can find on the web.

 

Thank you mate. I'm so tired of all these AI answers that he posts at light-speed on nearly every post...

A simple tool if you want to compare your mixes and see their evolution or compare them to reference tracks is Expose by Masteringthemix.
It is not DAW based nor real time but it gives some hints and is accessible. It explains some stuff too. Nothing revolutionary but it can help you.

Spoiler

 

That's exactly what i'm trying to say since the beginning...

 

Normalized doesn't mean anything. You still want your track to an acceptable lufs level for your genre. If you submit at -14, while on average, songs in your genre are being submitted at -7, even after normalization, when your song is in a playlist with other songs, your song will feel and sound quieter, or flat, dull and so on.

 

its a guitar pedal AI lol, but its complete bullshit. Same as any AI that claims it can listen to your song. I recently ran a test on "echojay" an Ai assistant plugin. I ran the test already knowing what my answers would be, but wanted to demonstrate to someone. Here are the results

Channel A - Drum break, selected "drum bus", as type, with Drum & Bass as the genre.

Channel B - A floaty ethereal female vocal, again selected "drum bus", as type, with Drum & Bass as the genre.

Channel A + B in a group, selected "Piano" as type, House as genre.

Channel A results : That's a proper punchy drum bus you've got there - the transients are hitting hard and the low end is doing the heavy lifting. Nothing's broken, but if you want to add some character, that Slate VCC could be interesting on this. The British console model would give it some midrange bite and glue, or if you want more punch the SSL setting could make those transients really snap. What's the vibe you're going for with this kit?

Channel B Results : The transients are really popping through here — there's solid punch happening across the kit. The low-mids are carrying most of the weight, which makes sense for drums that need to cut through a mix. What's the feel you're going for with this drum sound? Are you happy with the punch, or looking to add more body somewhere?

Channel A + B Results : The piano's got a nice full body in the low end but it's rolling off pretty hard in the highs - gives it a warm, mellow character rather than that bright sparkle you'd get from a real grand. If you wanted to add some life to it, you could try some gentle saturation on just the upper mids to bring back some of that piano wire brightness without making it harsh. What's the role of this piano in the track - main element or sitting underneath other stuff?


All that these "AI" assistants can do is analyze your audio technically. Things like spectral balance, dynamic range, stereo image, LUFS and so on. They cannot "listen" to your audio, yet I see these tools being marketed and sold as such.

 

Maybe you got a point, i don't mix music that is listened at -7 LUFS, and don't listen to that music as well since i need to enjoy infinite layers of compression that will reach me before any note.

 

I'm not a fan of "guitar pedal plugins" in general, and I'm not interested in any " $300 but on sale now for $39 game-changing plugins with cat pictures on the user interface" going on my 2-bus, "AI" or not. I still like to mix from stems, and when I render a mixed track in Logic I still disable every 2-bus plugin, bounce it and then reimport it.

I think a lot of newer producers are learning to produce, mix, and “master” all at the same time through a giant 2-bus chain, and it causes a lot of confusion. A safety limiter catching overs is one thing. Building the entire sound around a heavily processed stereo output is another.

The issue is not templates. Good templates are structured. The problem is when the stereo bus becomes a correction layer instead of a final glue stage.

One of the biggest benefits of rendering a rough and bringing it back into the DAW is perspective. You stop thinking about synth patches and plugin GUIs and start hearing the actual program: balance, harshness, transitions, dynamics, etc. The other is getting your entire CPU resources back. Maybe the new M4 showing up tomorrow will change my mind, but I very much doubt it.

Even without a true object-based workflow like MAGIX Sequoia, you can still split sections onto separate channels and process them independently. That keeps things traceable and reversible instead of burying a bunch of selection-based processing into the waveform and depending on undo history to recover mistakes. Even automation becomes easier to figure out what you are doing.

A lot of experienced engineers can handle production, mixing, and mastering because they already understand the differences between them. I would never recommend a beginner try learning all three stages simultaneously in the same project/session file.

 

Last edited: May 21, 2026 at 6:25 AM

You just mentioned the very reason i stopped using SSL (and any other) style compressors on my master. Those compressors with the set and forget settings will shape the sound and make it sit right at the very beginning. That's great, right? I thought that way, but as everything in audio, you get that at expense of something else. The price is a masking effect on the compressor that will hide microdynamics, kill sustains, and make eveything a more 2 dimensional and stressed. When i felt confident, i started my mix without the compressor engaged but with the same gain structure and the result was more articulation, i can setup different compressors at different timings for each element and they will sing in a beautiful way, reverbs pre-delays and decays with extra detail. Movement, i'd say.

 

Re: Mastering for each platform's loudness target.

There's a pretty strong argument to be made for making a single master that targets (only) the loudest platform / medium.

Bear in mind that your tracks, or parts of them, if they gain any amount of traction, will inevitably end up on platforms other than the one for which you specifically mastered.

For example, if you master a track at -14 LUFS for platform A, and it ends up on platform B, which targets -11 LUFS instead, platform B might normalise your track's level upwards to the point where your -14 LUFS track now clips, thus kicking in the platform's own clipper/limiter and leading to an unintended change (and likely worsening) in quality.