so if a mastering engineer receives a mixed session peaking at -1db lets say its said that he's more likely send it back and tell them bring it down to -6db to make headroom and to resend it back to him

my question is

why can't the mastering engineer just lower the waveform by -5db and work with that?

 

because the mastering engineer needs some headroom with levels that are original in your master/export, the "loudness" is his enemy.

-6db is quite mild. if he's mastering only with digital.. thats maybe ok,

but a mastering engineer working only on analog hardware i used to work with always asked me to send him masters between -12 and -8db! that would sound surprisingly "low", but those tube compressors , vintage eqs and tape machine likes that better.

 

Last edited: Apr 25, 2021

If he's just talking about lowering the level it doesn't make sense. Neither does the 6dB headroom he's asking for.

 

So he can't pull down the wav to the optimum level himself?

 

Chances are.. they won't honestly redo the entire mix from scratch anyway. Which is why I prefer to lower the input (not output) of the mix myself if it's too loud when I pull it up. I'd have to ask them to lower the input level of all their tracks at the production stage (pre-FX) for it to really count. They would have to redo FX settings, thresholds, etc. because the signal wouldn't hit them as hard and cause them to react the same. The mix may sound more undesirable on their end. And all that might only be helpful if the dynamic range/noise floor on their converters are horrible. The most crucial thing to me is that the mix is not over-compressed/limited from the start. And some single mixes are so spot-on as far as tone and dynamic peaks from the start.. they don't even need to hit a limiter at the mastering stage.. but just rendered through a better clock source at a louder level. Personally, I wouldn't send just 1 lone mix elsewhere to be mastered unless it was part of a soundtrack with various artists. Mastering is suited for a number of mixes combined into one overall cohesive project. That's really what sets it apart.

 

My advice to you:
If the mastering engineer is not able to adjust the peak level himself, run!

 

Hope I can ask a question (I never dared to ask) as a mix-idiot. When/ if I mix for demo purposes I am "targeting" -12 to -14 LUFS, and that way I'm never really hitting anything above -4 dBTP.. BUT when I look at open projects from "PROs", their (processed) kick is hitting -6dB or even higher ...and then they add like 50+ more tracks. I'm not talking about some crazy wannabes, but guys like eg. James Wiltshire of the Freemasons, that are not so young anymore and used to mix on analog gear. And they are advocating dynamic mixes their whole life..

So I always hear "leave the headroom for the mastering eng" on the forum, and OTOH I see (pre master) "pro" mixes that do not follow this rule at all. So which is it..??

I promise, I'll still leave the mix part to the mix engineer because one - I rather spend time making music ad two - because why not. But this puzzles me every time..

 

its a power play, he's trying to make you do a dumb chore you don't understand so when he runs the ozone master assistant you will think he did some pro stuff you don't understand

 

Peak level isn’t important in digital, as long as you’re not clipping.
Dynamics really are.
-1dBTP is alright for me as a mastering engineer as long as the mix has enough dynamic range to breath.

 

probably because you sent him a 24 or 16 bit file?

otherwise it woud be a joke.

 

he is obviously not even able to tell people what he wants prior to receiving it.

 

If your render is 24bit, and it is clipping, it has to be re-rendered, but if it's 32bit float there is no need.

 

I'm a mastering engineer, send it to me and I will profesionally lower it down to -6db

 

That's why "stem mastering" was invented.

 

Honestly it really doesnt matter as long as its not distorted an sounds good...so any good Mastering Engineer wont care about the levels. Sometimes my mixes turn out -9 to -6 LUFS even before mastering them. thats right and peaking -0.1 dbPT . Analog gives you alot of head room. Im hybrid mixing out Black Lion Audio DAC through my console ,good EQs and driving my Bus compressor with 2 matched killer preamps. on the way back in hitting 2 mastering Pres and recording back in on another system with either my Antelope Orion or Black Lion converters. I can easily get mastering levels without even trying is my point ...just mixing through my gear and converting or printing the mix back in. Even ITB only those levels wont matter... the quality of sound and balance of the mix will

 

You may send +20dBFS to mastering engineer, or even louder. Just save as 32/64 bit FLOAT.
It more depends on how you mix, clip/distort/saturate in mix, plugins with calibration like -18dBFS = 0dBVU or something like that, so internal "analog" clipping happens when it hits -18dBfs. All shit may happen only in source recording, mixing, effect plugins, chains etc. Not in DAW output file.

 

Last edited: Apr 26, 2021

This idea that you should leave x dB FS of "headroom" to the final downmix for the master engineer has been around as long as I can remember. That is to say at least from the late 90s.

But why?

1. The way early digital tools worked and how they were used.

You could actually make early DAWs clip easily by going to red at any point. The whole paradigm was very much analog. Similar rules about how you should leave at least x dB FS of "headroom" to every single individual track are still around too. In a way this master track headroom rule could be interpreted as a desperate attempt the lower the odds that the complete idiots often doing the mixing would not clip the audio.

With modern DAWs and their 32bit floating point engines the concept of "headroom" doesn't really even make sense anymore. You can go hundreds of dB's to the red and still not clip the audio.

2. General confusion about how digital audio works.

To this day even many seasoned professionals believe in all kinds of early myths about digital audio. Perhaps the most prevalent is the continuing confusion about the so called "inter-sample peaks". The idea is that if you don't leave enough headroom to the downmix the "inter-sample peaks" will clip and you can only get rid of them at the source ie. not creating them during the downmix process ie. leaving enough headroom. This is the most common reason given for this final downmix headroom requirement these days.

This is false.

In the digital domain there is no data between the samples and therefore no inter-sample peaks. They are only created during the digital-analog conversion. You can simply turn down the audio before it hits the converter.

 

I wondered this myself in the past because 'everyone' related to mastering was talking about the same subject: "Leave at least 6dB headroom!"
No one actually said that it didn't matter at all, because you can always add a gain plug-in at the start of the mastering chain. So I think it's more related to the mixdown and having everything in the right balance. James wiltshire explained once something similar with K-metering:

 

DAW have always been using 32 bits - or overflow bits - to process audio.

and the only moment where you need to leave headroom is when you export a master.

there are only two reasons why you had (and still have) to leave headroom in a master: CD, because cheap consumer players can not properly reproduce fully maxed out audio, and mpeg, because the transform will create values higher than in the input and then the result will be clipped. (in a range much higher than only the inter sample peak phaenomen)

what this guy aks for is that his customer kind of creates a "master" for him, and i´d strongly recommend to look for an alternative offer.