Currently doing some mastering and the producer/mixer wants me to push the masters 2dB louder.

Im currently using Klanghelm VUMT VU meter with "0" VU set to -6dBFS (So 0 VU is 6dB below absolute digital clipping), and I'm regularly hitting +2dB to +3dB VU on transients.

I mean, sure, I can make your music so fatiguing to listen to that it becomes unbearable to listen to,

and I'm sure Rick Rubin would be proud of that,

But I'd rather not do that.

 

I'll express my feelings here about this, just know this is my 35 years of experience opinion only and I'm NOT judging any others opinion on this subject.

I started recording music in the late 80's on 8 and 16 track tape, then went digital ADAT in the mid 90's, jumped to Ensoniq PARIS late 90's, was full on Pro Tools on SSL's in the late 2000's in Atl and LA. Today I'm hybrid preferring mixing into my API Box but mostly mastering ITB. When I started nobody discussed loudness, it was all about controlling the noise floor. Over the years it changed to how loud can we make it and working digital... here are my observations.

1. When mixing loud I find -10 LUFS the sweet spot for loud but still sounding like music.

2. My digital production sounds musically better loud if the tracks are treated with a tracking stage. I run everything digital through hardware at Access Analog and in my testing it makes a noticeable difference when pushing the LUFS of songs. Mixing with plugins is faster and better sounding to my ears also.

3. I recently mixed a song for a friend that included a mix that was going on 7" vinyl, the digital streaming mix was between -9 and -10 LUFS but the 7" format required an old school approach come mix time to meet the 7" requirements and the mix was around -20 LUFS, then went to mastering. We got the test press back so I got to compare both mixes. The streaming mix was louder on the iPhone but played on monitors and volume adjusted the 7" vinyl mix sounded great and very musical.

4. Digital rules the mid range but IMHO the low end and top end shine with hardware. My testing over the years has shown me that and why I work hybrid, utilizing the best tools from both worlds to make the best sounding music I'm able to.

5. Louder does not mean better, it just means more distortion for the listener.

IMHO we music creators bought magic (low paying streaming) beans as we've sacrificed sound quality for loudness and it's very apparent in direct comparisons. Unfortunately today good is the enemy of great, MP3's sound good but DSD sounds great. Soon AI music will sound good but we humans will have to sound great to stand out.

Hopefully my experience and thoughts help somebody out there before the bots take over the internet.

 

I appreciate your insight and information about your experiences.

I don't have access to analog equipment for tracking, but I mainly do mastering nowadays and have a decent little hybrid setup.

I have a Midas 500 series rack with 2 sets of analog comps and EQs, then going in to a HHB Fatman stereo tube compressor where I just "kiss" the needle if I want to add a bit of thickness and saturation after the 500 series rack.

I agree that the sub lows and top end freq just sound a tiny bit better through the analog gear, but I also really like using analog compression to shave off like 1-2dB of the mix before hitting digital limiters.

Depending on how well controlled the dynamics of the transients in the mix are, I may end up doing like 8dB of gain reduction thru compression (maybe MB) and limiting to get the track to the loudness some people request.

I would never really want to do any more than that in any circumstances. The music just loses all musicality at vibe at some point.

 

Does VUMT regularly drop below -3dB VU on loud parts?

What kind of music is this? Do the references in his style of music sound louder/2dB louder?

Have you ever tried to put clippers before each compressor and/or limiter in your mastering chain in order to try to get these 2 dB in several small transparent steps?

 

Last edited: Feb 4, 2026 at 11:40 PM

You can go very loud. I regularly hit -4 lufs, usually im hitting -6/7. It depends on genre of course. You likely won't get away with -4 on an acoustic singer songwriter song. What sort of song are you working on ? What genre?

 

The quiet parts don't drop below -3dB for any meaningful amount of time. Maybe only intro/outro and intentional transitions.

I used to automatate quiet parts of this artists tracks to be louder, but he didn't seem to notice any difference and his mixes are good enough that I respect the mix automation. I will occasionally have to automate EQ cuts and boosts for certain sections if there is a beat switch up or really harsh section.

The genre is mainly trip-hop, but a hybrid of "bass" genres and a little trap influence. It's a loud genre, but not as loud as mainstream EDM or some metalcore/djent.

I am mostly against using clipping during mastering unless I have to, as (knowing this artist well) he clips the shit out of his drums, so I would be clipping them again haha. Basically I would be turning all the peaks into harmonic distortion at that point.

His dynamic control of the drums within the mixes is a bit underbaked compared to how I would do it. The drums are the first thing to get turned down by final limiting if they are not controlled (or vol leveled) correctly in the mix. This makes mastering more difficult as bringing the loudness of the track up tends to overly process the transient material, and can degrade the transients.

His mixes have improved a ton, but they can reaaaally vary in terms of how loud the drums are between tracks, especially with the hats and pecs sounds. These areas become quite harsh and stick out even more aggressively as I bring the levels up. Some minor dynamic resonance suppression can help, but it also tends to impact other instruments that exist in the same freq range.

I have matched the volume and dynamic range of the reference tracks he provided (comparing louder sections of each track), but he still wants them louder

I'm already within 3dB of hitting 0dBFS at max peaks within the track.

I did convince him to only go 1dB louder tho

 

Modern trip hop / Bass music can be very loud. -6 lufs or so. While clipping may seem unnatural or against your intuition its often times a part of the sound. Of course the song really dictates what level is appropriate. Theres many sub-genres within bass music, and can really sit anywhere within -14 to -6. The more edm adjacent / trap sub-genres I would definitely lean louder. Songs like the artist Ivy Lab. With modern tools and techniques, the whole "mix is fatiguing" thing doesnt have to be true. Multi stage and multi-band clipping, compressing, limiting, and saturation can get you there. I would say its more-so done in the mixing stage, or with stem mastering, but it can be done with just the stereo track as well, assuming its mixed decently at least. It can still sound musical, not a harsh and brutal mix. Just make sure to give -1 to 2 tp, if you go to those absurd levels.

 

How does it sound? Most of what has been spoken about here are techniques and numbers.

Do you think it can be made louder by another 1 or 2 dB without sacrificing sound quality? If it can't be done, I'd forget about all of this and try to convince the producer that what he wants is a bad idea. He sounds inexperienced if he wants it louder "just because." It might be worth attempting to educate him a little bit.

You could always give him two versions. One version as he wants it and one version as you see fit. Just make the loudest version equal in LUFS-I to the quieter version post-limiting/dithering, then ask him to tell you which one sounds best. Figure out where to go from there. He might trip himself up by choosing the quieter master. It might give you both more to discuss before a final decision is made.

 

Watch the most recent MWTM with Jaycen Johua working on Ed Sheeran's Take Me Back to London. First ten minutes.

He starts with an already very completed rough they sent him, and the "numeric dillema" of clients wanting -7 lufs output is almost identical to your scenario, including the numbers. It's also the same cause you mention in the third post. The quietest parts are barely showing -3dB, yet he wants it louder. The drums you mention having dynamic issues are likely the same as well. It is transient control allowing for more/less perceived loudness which is behind his mix not being ready to live at -7lufs and still sound good. He knows he wants it to sound louder, but that is where you are listening with your ears and he is trying to fix what his are hearing. It's not the numbers, but that is what he is probably looking at. That's a pretty natural thing to do when you are convinced your mix should still sound good at that target.

 

I believe there's a hidden loudness in all music.
Even if a mixer sends me a track with a final limiter, I can still add 5dB more.

 

If you're doing something like jazz or classical you basically want -3dB RMS and something like Metal or Hip Hop -25dB RMS.

 

hybrid master, clipping my lynx hilo2 converter...sounds amazing but its loud

 

Skip the VU meter. What is your current LUFSi?

 

This perfectly describes the scenario. Its not that I can't push the master an additional 1- 2dB, its that the mix wasn't prepared for me to go as loud as he wants me to go.

I'm already shaving around 8dB off of the kick and Snare exclusively, as they were (in my opinion) not properly controlled in thr mixing stage.

I've already made some sonic sacrifices to get the mix this loud in the first place, and to push it an additional 2dB would mean sacrifice more transient punch and clarity to bring up the backing elements of the mix.

People mention clipping, but clipping an entire master has always sounded like shit to me. The degradation on the transients caused by the limiting of the transients is favorable to turning the transient peaks into harmonic distortion AND potentially damaging other high freq material subsequently with a clipper.

I will watch that MWTM if I can find a link to it!

I personally feel that the additional boost to the background elements of around 2dB is not worth the sacrifice of 2dB more off the transients.

Good call tho, as I will send him multiple versions of the final masters, one slightly louder than the others.

My only gripe is that using my analog hybrid setup requires me to export the masters in real time, and I'm already offering this client way more than I should for the $15 per master I'm charging him. It just takes so much more time doing analog recall and additional exports

My work must be good for the price tho, as he continues to send tracks my way lol. (I undercut myself when working for friends admittedly)

Thats funny, as he has sent me Ivy Lab tracks as references before, but those tracks are mixed quite a bit better than his, and as a result they translate to being better sounding LOUD masters.

A little bit more should have been done in the mixing stage, but people love to make their drums as loud as possible, not realizing that they will the be first thing that gets turned down with limiting.

One thing that drives me crazy about him is that he never reference mixes. This is noticeable as his sub basses are 4-6dB quieter than the reference tracks he sent me.

The whole mix is fatiguing thing doesn't have to be true, but it becomes much more apparent when listening on headphones.

Also this guy performs live sets so he could literally just turn his tracks up 1-2dB with monitor controls and wouldn't have to kill the dynamics of the masters

 

I will check the levels when I'm back home working tonight. I can tell you that they are directly in line with the reference material, both peak and RMS.

I would also love to share the final results with all of you when the masters are done and ready to be released.

Then we can truly judge whether or not they actually benefit from being pushed so close to digital clip!

 

I'm not talking about peak or RMS. I want to know percieved loudness (LUFS, integrated). It totally different than peak or what you are calling "digital clip". You can peak at -1dBFS and still be at -8LUFSi.

 

haha nice. when i read your description of the song, Ivy Lab was what immediately came to mind. Their songs are usually mixed great, so I understand your predicament of not being able to get in the ballpark. One thing you can try, is a bit of black magic. Go ahead and clip, limit the shit to hell and back to get the backing elements up, but add the transients back in after. There are a few ways to go about this. You can try something like oeksound spiff, or any transient shaper type plugin in delta, on a parallel track. Mix the transients back in against the clipped to shit main track post clipping / limiting. You can also do this manually, zooming in to the waveform, and extracting just the transients, pasting them on their own channel. Once the transient channel is mixed back in with the main channel post clipping, you can continue your processing as normal. Even if the transients are mixed in late in the chain, its usually a good idea to hit everything with another db or so of compression, limiting to glue them back in. Experiment with treating the transient channel in different ways as well. Eq if needed, try different phase etc.
This doesnt always work, and it can take some experimenting, but when it does work its awesome.

 

Yeah, more like get your client to watch it and his next track will hit -7lufs more easily. Still, why?! But not the point here.

Clipping keeps coming up because workflows like Baphometrix’s Clip-To-Zero strategy and documentation/videos didn’t exactly invent anything new, he formalized and made some guidelines what engineers have been doing informally for years. The real takeaway isn’t “clip everything to get louder,” it’s using clipping early as signal conditioning. Delta listening makes this obvious: you’re shaving fast, high-frequency distortion products and intermodulation junk that you barely perceive, but which still hit your compressor detectors. That material is often too fast to be meaningfully shaped by compressor time constants, yet it inflates detector readings and causes overcompression. It leads to people doing it anyway. Baphometrix documentation comments about that, saying people "overwork their compressor". I'm trying to avoid my usual little shorthand stuff here that you will know, but avoid any pedantic correctors here about micro subjects.

This is why Jaycen Joshua’s point matters: hitting −7 LUFS successfully isn’t a mastering trick, it’s a production decision. His mixes are built to live at −7. When a client mix falls apart at that level, it’s not because it’s “too loud,” it’s because the mix dynamics weren’t designed for it. If your quiet sections are already sitting around −3 dBFS, or only a few dB below the loudest moments, this guys mix has left no vertical room for transient material to move upward and actually punch. The adaptive limiter does what it’s supposed to do: it catches those transients and smashes attempted punch into density. Of course, he frames this as using The God Particle correctly to hawk his plugin, like always. But it's also not a stretch of the imagination to guess your trap/triphop guy is not using it too. You might even tell him to lay off the OTT until he figures this part out.

So the issue isn’t actually volume, it’s mix architecture. Chasing a LUFS number without preserving transient to remainder of your mix ratio just reduces crest factor, worsens effective signal to noise ratio, and you get a flatter, mushier result. Clippers can help alot when they’re used to clean the signal before dynamics, not as a blunt instrument for “louder = better.” I'm surprised no-one has already mentioned Goldclip yet. The iLok demo is fully functional for a couple of weeks. You're already doing the guy a favor just powering up any gear for what you are charging, so you might as well just show him it.

This is the CTZ document: https://docs.google.com/document/d/1Ogxa5-X_QdbtfLLQ_2mDEgPgHxNRLebQ7pps3rXewPM/edit?usp=sharing
"

 

Last edited: Feb 6, 2026 at 1:09 AM