The 6 dB of Headroom for Mastering Myth Explained

Have an urge to learn, or a calling to teach? Want to share some useful Youtube videos? Do it here!
Post Reply
WongoTheSane
Moderator
Posts: 1851
Joined: 14 Sep 2015
Location: Paris, France

12 Jul 2018

Nice article from a mastering engineer, casting light on peak limiting before sending to master:

https://theproaudiofiles.com/6-db-headr ... explained/

User avatar
normen
Posts: 3431
Joined: 16 Jan 2015

12 Jul 2018

Hehehe, great article. Things can be that simple, right? Theres a simple reason for something and at some point it becomes a myth and people make up all kinds of reasons.

WongoTheSane
Moderator
Posts: 1851
Joined: 14 Sep 2015
Location: Paris, France

12 Jul 2018

normen wrote:
12 Jul 2018
Hehehe, great article. Things can be that simple, right? Theres a simple reason for something and at some point it becomes a myth and people make up all kinds of reasons.
You'll like this one, if you don't know it already...

A young woman is preparing her first thanksgiving dinner. As she gets everything ready for thanksgiving day, she very sternly reminds herself to let the turkey finish thawing in the sink overnight. She puts it in and places the dishrack over the top of the bird. Her husband walks into the kitchen and sees this. “Why are you doing that?” he asks.
“My mom always did that to help the turkey thaw” she told him.

The next day Mom calls to see how everything is going. “Fine, Ma. I have everything ready to go in the oven. I even remembered to put the rack over the turkey last night.”

This seemed to confuse her mother a bit. “What are you talking about?” she asked.

“Oh, I remember you always put the dish rack over the turkey when it was thawing in the sink,” she said.

There was a pause on the end of the line. “Yes, but honey, we had cats!”


(this version is from https://www.snopes.com/fact-check/grand ... ing-secret)

User avatar
FLVZ
Posts: 519
Joined: 17 Aug 2016
Location: ZW | GB

26 Jul 2018

If only we had 32bit export :oops:

User avatar
QVprod
Moderator
Posts: 3488
Joined: 15 Jan 2015
Contact:

27 Jul 2018

Flavolous wrote:
26 Jul 2018
If only we had 32bit export :oops:
There's a whole thread on this already, but as stated in that article, just don't clip the master. Otherwise you're not missing anything.

User avatar
FLVZ
Posts: 519
Joined: 17 Aug 2016
Location: ZW | GB

27 Jul 2018

QVprod wrote:
27 Jul 2018
Flavolous wrote:
26 Jul 2018
If only we had 32bit export :oops:
There's a whole thread on this already, but as stated in that article, just don't clip the master. Otherwise you're not missing anything.
I wasn't trying to start the fire again. In all honesty, its just inconvenient. In a world where you can mix freely and export in 32bit then just reduce the volume we're forced to work like we're running on analog "just dont clip the master" of course, but thats going to cost me a few hours just to readjust the mix when i could have exported in 32bit and just lowered the gain. I think you would welcome 32bit export too, also there are certain mixing and mastering tricks opened up by 32bit export, so yeah. We are missing out on a few things.

User avatar
QVprod
Moderator
Posts: 3488
Joined: 15 Jan 2015
Contact:

27 Jul 2018

Flavolous wrote:
27 Jul 2018
QVprod wrote:
27 Jul 2018


There's a whole thread on this already, but as stated in that article, just don't clip the master. Otherwise you're not missing anything.
I wasn't trying to start the fire again. In all honesty, its just inconvenient. In a world where you can mix freely and export in 32bit then just reduce the volume we're forced to work like we're running on analog "just dont clip the master" of course, but thats going to cost me a few hours just to readjust the mix when i could have exported in 32bit and just lowered the gain. I think you would welcome 32bit export too, also there are certain mixing and mastering tricks opened up by 32bit export, so yeah. We are missing out on a few things.
Should the mix be clipping, the volume has to be reduced either way though doesn't it? And while 32bit has more dynamic range, it won't have any affect on the dynamics of the mix. Also, unless you're using fx post master fader, you can just turn the maser fader down to prevent clipping. I'm not speaking against a 32 bit export option, but the article really does present 2 key things (common digital audio rules) for exporting a mix for mastering. Don't clip, and don't use a brickwall limiter. The mastering engineer's converters will be 24bit.

I'd genuinely love to hear what tricks mixing and mastering tricks you can achieve with a 32bit audio file of a mix that isn't clipping (peaks above 0dBfs) though.

User avatar
FLVZ
Posts: 519
Joined: 17 Aug 2016
Location: ZW | GB

27 Jul 2018

QVprod wrote:
27 Jul 2018
Flavolous wrote:
27 Jul 2018


I wasn't trying to start the fire again. In all honesty, its just inconvenient. In a world where you can mix freely and export in 32bit then just reduce the volume we're forced to work like we're running on analog "just dont clip the master" of course, but thats going to cost me a few hours just to readjust the mix when i could have exported in 32bit and just lowered the gain. I think you would welcome 32bit export too, also there are certain mixing and mastering tricks opened up by 32bit export, so yeah. We are missing out on a few things.
Should the mix be clipping, the volume has to be reduced either way though doesn't it? And while 32bit has more dynamic range, it won't have any affect on the dynamics of the mix. Also, unless you're using fx post master fader, you can just turn the maser fader down to prevent clipping. I'm not speaking against a 32 bit export option, but the article really does present 2 key things (common digital audio rules) for exporting a mix for mastering. Don't clip, and don't use a brickwall limiter. The mastering engineer's converters will be 24bit.

I'd genuinely love to hear what tricks mixing and mastering tricks you can achieve with a 32bit audio file of a mix that isn't clipping (peaks above 0dBfs) though.
Yea its a great read! I am just ignorant to understanding why we don't have 32bit export, if anyone knows I'd love an answer. :lol:

dustmoses
Posts: 197
Joined: 04 Oct 2015

31 Jul 2018

If Reason allowed you to select and drag multiple faders and drag not clipping would be trivial.

I love reason does any other DAW have this terrible flaw though?

User avatar
QVprod
Moderator
Posts: 3488
Joined: 15 Jan 2015
Contact:

31 Jul 2018

dustmoses wrote:
31 Jul 2018
If Reason allowed you to select and drag multiple faders and drag not clipping would be trivial.

I love reason does any other DAW have this terrible flaw though?
Studio One used to have the same issue before version 3 . Turning down the Master fader is a good solution too as long as you don’t have Any processing after the Master fader which in most cases you probably won’t.

EdGrip
Posts: 2343
Joined: 03 Jun 2016

01 Aug 2018

...and if you have processing after the Master fader, just bung a Selig Gain in there last thing before the output.
You haven't got Selig Gain? It's really useful!
But if you haven't got it, just bung the free kHs Gain RE in there instead.
If you are RE-averse, just bung a 6:1 line mixer in there and use it as a gain knob.

Another thing Selig Gain is handy for is for automating volume changes (because we don't want to start automating mixer faders, do we? Because that's entering a world of pain, Donnie.)
You can automate the output volume knob on whatever the last device in the chain is (if it has an output volume knob - Saturation Knob, I'm looking at you), but then you have to think about it again if you re-order or delete FX.

But yes, groupable faders would be very handy.

User avatar
FLVZ
Posts: 519
Joined: 17 Aug 2016
Location: ZW | GB

03 Aug 2018

QVprod wrote:
31 Jul 2018
dustmoses wrote:
31 Jul 2018
If Reason allowed you to select and drag multiple faders and drag not clipping would be trivial.

I love reason does any other DAW have this terrible flaw though?
Studio One used to have the same issue before version 3 . Turning down the Master fader is a good solution too as long as you don’t have Any processing after the Master fader which in most cases you probably won’t.
But then are you not vulnerable to clip at channel level ? You turn down the master fader and end up masking the clipping on some channels ? I thought this was why messing with the master fader is not recommended?

User avatar
jam-s
Posts: 3035
Joined: 17 Apr 2015
Location: Aachen, Germany
Contact:

03 Aug 2018

Flavolous wrote:
03 Aug 2018
But then are you not vulnerable to clip at channel level ? You turn down the master fader and end up masking the clipping on some channels ? I thought this was why messing with the master fader is not recommended?
As the internal processing is done in 32 bit (and 64 bit summing) you have to try incredibly hard to clip any of the mix channels (seeing red there (in the virtual ssl) does not mean jack, unless it is the actual clip indicator).

User avatar
FLVZ
Posts: 519
Joined: 17 Aug 2016
Location: ZW | GB

03 Aug 2018

jam-s wrote:
03 Aug 2018
Flavolous wrote:
03 Aug 2018
But then are you not vulnerable to clip at channel level ? You turn down the master fader and end up masking the clipping on some channels ? I thought this was why messing with the master fader is not recommended?
As the internal processing is done in 32 bit (and 64 bit summing) you have to try incredibly hard to clip any of the mix channels (seeing red there (in the virtual ssl) does not mean jack, unless it is the actual clip indicator).
I guess thats another myth then :D

User avatar
QVprod
Moderator
Posts: 3488
Joined: 15 Jan 2015
Contact:

03 Aug 2018

Flavolous wrote:
03 Aug 2018
jam-s wrote:
03 Aug 2018


As the internal processing is done in 32 bit (and 64 bit summing) you have to try incredibly hard to clip any of the mix channels (seeing red there (in the virtual ssl) does not mean jack, unless it is the actual clip indicator).
I guess thats another myth then :D
Yeah this is a myth that comes from Analog. Also my suggestion of turning the Master fader down is after the mix is basically done. Not at the beginning of the mix. Even if you were mixing analog, you would notice channel clipping long before you got to that point.

User avatar
Biolumin3sc3nt
Posts: 662
Joined: 16 Jan 2015

05 Aug 2018

What's funny is I remember meeting Justin in the early 2000's in Madison Wi. I believe it was a backyard party where we ended up discussing drum Micing techniques, before I was to venture off to Full Sail in 6 months. When I went to see the source of the article, i noticed the face and the name. Glad He's doing well and yeah good old Smart Studios.

Such a random flashback!!

User avatar
Boombastix
Competition Winner
Posts: 1929
Joined: 18 May 2018
Location: Bay Area, CA

06 Aug 2018

Funny to read about the myths. The engineer says two very easy to understand things:
1. Don't clip your digital audio file (ugly distortion if you do, and it cannot be fixed)
2. I have a volume knob (i.e. he doesn't care about head room, he will turn up the volume if needs to, just don't clip the signal (see §1)).
Lol
10% off at Waves with link: https://www.waves.com/r/6gh2b0
Disclaimer - I get 10% as well.

User avatar
Biolumin3sc3nt
Posts: 662
Joined: 16 Jan 2015

06 Aug 2018

Boombastix wrote:
06 Aug 2018
Funny to read about the myths. The engineer says two very easy to understand things:
1. Don't clip your digital audio file (ugly distortion if you do, and it cannot be fixed)
2. I have a volume knob (i.e. he doesn't care about head room, he will turn up the volume if needs to, just don't clip the signal (see §1)).
Lol
I'm not sure anything You just said is a Myth!!!

Clipping Signals is just that! Do all of Your mixes look like a brickwall square wave??

User avatar
Boombastix
Competition Winner
Posts: 1929
Joined: 18 May 2018
Location: Bay Area, CA

07 Aug 2018

Biolumin3sc3nt wrote:
06 Aug 2018
I'm not sure anything You just said is a Myth!!!
[/quote]

No, I was just referring to the article and the "myth" that you should leave xx dB of headroom. :lol:
10% off at Waves with link: https://www.waves.com/r/6gh2b0
Disclaimer - I get 10% as well.

User avatar
selig
RE Developer
Posts: 11685
Joined: 15 Jan 2015
Location: The NorthWoods, CT, USA

07 Aug 2018

Boombastix wrote: No, I was just referring to the article and the "myth" that you should leave xx dB of headroom. [emoji1]
There ARE good reasons IMO to leave some headroom when exporting a mix for mastering - not sure how any of this is a “myth” unless one believes it somehow sounds better with more headroom than less (it doesn’t).


Sent from some crappy device using Tapatalk
Selig Audio, LLC

User avatar
normen
Posts: 3431
Joined: 16 Jan 2015

07 Aug 2018

selig wrote:
07 Aug 2018
Boombastix wrote: No, I was just referring to the article and the "myth" that you should leave xx dB of headroom. [emoji1]
There ARE good reasons IMO to leave some headroom when exporting a mix for mastering - not sure how any of this is a “myth” unless one believes it somehow sounds better with more headroom than less (it doesn’t).


Sent from some crappy device using Tapatalk
I guess the myth part is where somebody says "No dude, you gotta leave 6dB, not 3dB, you can't send that to a mastering studio" and then in the worst case somebody sends the mix file through Skype to get a 3dB boost. ;) I mean theres people putting emphasis on the "6" there, not the "headroom", I see that too.

It's like Wongos story - you gotta put the dish rack on - if you have cats. But otherwise protecting your turkey is still a good idea, that can't really become a myth :lol:

User avatar
selig
RE Developer
Posts: 11685
Joined: 15 Jan 2015
Location: The NorthWoods, CT, USA

07 Aug 2018

normen wrote:
07 Aug 2018
selig wrote:
07 Aug 2018


There ARE good reasons IMO to leave some headroom when exporting a mix for mastering - not sure how any of this is a “myth” unless one believes it somehow sounds better with more headroom than less (it doesn’t).


Sent from some crappy device using Tapatalk
I guess the myth part is where somebody says "No dude, you gotta leave 6dB, not 3dB, you can't send that to a mastering studio" and then in the worst case somebody sends the mix file through Skype to get a 3dB boost. [emoji6] I mean theres people putting emphasis on the "6" there, not the "headroom", I see that too.

It's like Wongos story - you gotta put the dish rack on - if you have cats. But otherwise protecting your turkey is still a good idea, that can't really become a myth [emoji1]
Agreed, not so much a myth but instead it's potentially useful information passed along without any context.
Selig Audio, LLC

User avatar
Boombastix
Competition Winner
Posts: 1929
Joined: 18 May 2018
Location: Bay Area, CA

09 Aug 2018

selig wrote:
07 Aug 2018
Boombastix wrote: No, I was just referring to the article and the "myth" that you should leave xx dB of headroom. [emoji1]
There ARE good reasons IMO to leave some headroom when exporting a mix for mastering - not sure how any of this is a “myth” unless one believes it somehow sounds better with more headroom than less (it doesn’t).


Sent from some crappy device using Tapatalk
Back in the days when we put final mixes on DAT we had to avoid harsh digital distortion when recording to DAT tape. So mastering guys said, leave xx dB of head room, but it was mainly to avoid over-driving the DAT tape into digital clipping/distortion. 6dB was the "safe" standard often used.
But nowadays:
My understanding is that mastering studios have quality equipment and the can change the volume in the digital domain to whatever you they want before they apply EQ and so on. For a finished track (digital) that will be converted to analog signal by the user/listener, you may want a small head room for inter sample peaks (ISP), so a -1.5 to -2dB peak can be smart. -2dB is about 0.22V below 0dB just in case the DA converters are of low quality and cannot convert ISP above 0dB. MP3 conversion also needs some head room, about the same. Am I missing something?
In either case, if you mix your master to 24-bit, the noise floor is so low that some head room won't matter, it will sound the same after mastering. I would just keep a 1.5 to 2dB head room as a standard, pre-master mix, or finished master.
10% off at Waves with link: https://www.waves.com/r/6gh2b0
Disclaimer - I get 10% as well.

User avatar
selig
RE Developer
Posts: 11685
Joined: 15 Jan 2015
Location: The NorthWoods, CT, USA

10 Aug 2018

Boombastix wrote:
selig wrote:
07 Aug 2018
There ARE good reasons IMO to leave some headroom when exporting a mix for mastering - not sure how any of this is a “myth” unless one believes it somehow sounds better with more headroom than less (it doesn’t).


Sent from some crappy device using Tapatalk
Back in the days when we put final mixes on DAT we had to avoid harsh digital distortion when recording to DAT tape. So mastering guys said, leave xx dB of head room, but it was mainly to avoid over-driving the DAT tape into digital clipping/distortion. 6dB was the "safe" standard often used.
But nowadays:
My understanding is that mastering studios have quality equipment and the can change the volume in the digital domain to whatever you they want before they apply EQ and so on. For a finished track (digital) that will be converted to analog signal by the user/listener, you may want a small head room for inter sample peaks (ISP), so a -1.5 to -2dB peak can be smart. -2dB is about 0.22V below 0dB just in case the DA converters are of low quality and cannot convert ISP above 0dB. MP3 conversion also needs some head room, about the same. Am I missing something?
In either case, if you mix your master to 24-bit, the noise floor is so low that some head room won't matter, it will sound the same after mastering. I would just keep a 1.5 to 2dB head room as a standard, pre-master mix, or finished master.
My experience:

Going back further in time, mixing to analog meant you CAN go over 0 dB since headroom was built into the metering system. But when moving to digital, there was no built in headroom with the metering system, and we had to totally re-think levels.

Besides the headroom issue, moving from analog to digital meant we were exclusively using peak metering to set mix levels. With analog, we had peak LEDs on some meters to show peaks, but metering was largely VU based.

This move from analog to digital necessitated the workflow change of building in our own headroom when mixing to digital sources. At that early point, it had little to do with mastering engineers IIRC, though I’m sure they were seeing the same thing we were seeing. It was simply obvious we needed keep the concept of headroom from the analog world as we moved into the digital world.

16 bit DAT mixing had it’s own issues, one being the analog stage of cheeper DAT recorders potentially distorting BEFORE the signals even reached digital clipping. In the digital domain, all clipping is pretty much the same - 16 bit clipping is no different from 24 bit clipping. This had nothing to do with “overdriving the DAT tape into clipping/distortion, since tape doesn’t distort digital signals! It’s the A/D we’re talking about here, same as with any digital recording medium from tape to optical disc to magnetic or solid state disc.

Finally, there’s a difference between a pre-master mix and a final master, with regards to headroom. The topic here pertains to pre-master levels, so I’ll stick to that. With 16 bit sources one could argue the need to “kiss the reds” to retain dynamic range. But as you note, with 24 bit systems, there’s no need to do so.

And as I’ve mentioned elsewhere, in some cases your mix levels are not 100% repeatable from pass to pass, considering random elements such as algorithmic reverbs and non-synced LFO FX. The levels may not vary by much from pass to pass, but if you’re trying to avoid all clipping while mixing to 24 bit, and we’re talking about pre-master levels, there’s no advantage to spending the extra time (and risk of clipping) to push for that extra dB or two.

I instead just relax and enjoy mixing, and am happy when my levels are anywhere above -6 dBFS but below -2 to -3 dBFS. Which happens to be where my mastering engineer still requests my mix levels to sit, FWIW.


Sent from some crappy device using Tapatalk
Selig Audio, LLC

Post Reply
  • Information
  • Who is online

    Users browsing this forum: No registered users and 6 guests