The No.1 Website for Pro Audio
 All  This Thread  Reviews  Gear Database  Gear for sale     Latest  Trending
Mastering Headroom Question
Old 20th October 2013
  #1
Gear Nut
 

Mastering Headroom Question

I'm trying to figure out how to deliver a track with headroom for mastering. I'm using cubase 5. ME asked for -6 db's of headroom. It's a house tune.

I've learned to start my projects at lower levels to leave headroom, specifically the kick, I try to start around -8 db...however, this project has gone on for months now and has dozens of tracks/group channels ext....and somehow I'm finding that my Kick(loudest track) is now close to 0 db.

I want to create headroom so my first idea was to turn each track down evenly starting with the kick -6 db's, but as I proceeded to do this, I realized that I have a ton of level automation and I can't select all, and turn down.

My next idea is to create a new group channel and route everything to that channel and turn it down there before going to bus.

- My question: Is this a bad idea? Also, is there something I'm missing here? The kick track is the one giving me problems(the only one close to peaking), and when I turn it down, it kills the vibe of the track. The only other solution I can imagine is to turn each track down one by one, including the ones with automation which is not easy to do precisely.
Old 20th October 2013
  #2
Deleted 691ca21
Guest
Any ME who tells you you HAVE to leave 6dB of headroom is just making your life unnecessarily difficult. Talk to them. It should be possible for them to work with pretty much any level, as long as it's in 24 bit, and there is no digital clipping at 0dBFS.
Old 20th October 2013
  #3
Lives for gear
 
MASSIVE Master's Avatar
 

Verified Member
1) I agree with the whole "any headroom is good headroom" thing... As long as a mix naturally (no limiting, no excessive buss compression) doesn't clip, it's dandy.

2) NORMAL levels -- Not "lower" levels. Sorry, just semantics, but there's nothing at all "normal" about the levels and voltages we're dealing with by the time something is on the shelf. To that end, I'd heartily recommend setting your kick considerably lower than -8dBFS if you're starting with the kick. Just a couple more things hitting -8dBFS at the same time as the kick and your headroom is gone. Calibrating your monitoring chain to a more normal level (I'm a fan of -20dBRMS = ~83dBSPL) goes a long way.

Blah, blah, blah, you're in a floating point system where you should just be able to take your master fader down and get where you need to be. "Good form?" No. "It'll do?" Usually, yes.
Old 20th October 2013
  #4
Lives for gear
 
FabienTDR's Avatar
 

Verified Member
Yeah, don't confuse headroom with dynamic range.

Analogue gear has headroom, music cannot have it! Using the term headroom in this context is even quite stupid and shows a fundamental misunderstanding of all these terms. Digital audio has either infinite headroom (floating point) or no headroom at all (fixed point). There is nothing to discuss about headroom in digital work-flows.

You are looking for dynamic range, which is something completely different.


If a mastering engineer is too stupid to use an amplifier (to drive his analogue chain properly), he should look for a different job anyway.

To be clear, avoid mastering engineers who ask you for headroom in "files": They have clearly no idea what they're doing. Look for someone else.
Old 20th October 2013
  #5
Lives for gear
 

This.
Old 20th October 2013
  #6
Lives for gear
 
karumba's Avatar
Quote:
Originally Posted by FabienTDR View Post
Yeah, don't confuse headroom with dynamic range.

Analogue gear has headroom, music cannot have it! Using the term headroom in this context is even quite stupid and shows a fundamental misunderstanding of all these terms. Digital audio has either infinite headroom (floating point) or no headroom at all (fixed point).
with fixed point, not the format itself has headroom, but the audio data can have. the word & meaning of "headroom" is common understanding for that & also for digital fixed point audio, as e.g. specified by the EBU:

"In digital audio, headroom is defined as the amount by which digital full scale (FS) exceeds the permitted maximum level (PML) in dB (decibels). The European Broadcasting Union (EBU) specifies a PML of 9 dB below 0 dBFS (-9 dBFS), thus giving 9 dB of headroom."

http://en.wikipedia.org/wiki/Headroo...nal_processing)

anyway, this 6dB headroom thing is just to have a rule of thumb. it is very unlikely, that a ME would request that because he his incapable of doing proper gain staging.

i tell my clients this:
* 24bit files
* highest peak within -20 and -0.1dBFS is fine (not wasting resolution, but also no clips)

Quote:
If a mastering engineer is too stupid to use a amplifier (to drive his analogue chain properly), he should look for a different job anyway.

To be clear, avoid mastering engineers who ask you for headroom in "files": They have clearly no idea what they're doing. Look for someone else.
to be honest, if find such statements disrespectful, especially as you don't know the ME or the intention of his request.
Old 20th October 2013
  #7
Gear Nut
 

Thanks for the feedback....So, from what I can tell, I have a dynamic range in my track with just the kick getting close to 0 db. Everything else is sitting well below. From the responses, am I to assume this is fine to send like this or should my highest peak still be somewhat below 0 db?

Also, I saw someone mention that simply turning down the master will suffice.


Thank you,
Old 20th October 2013
  #8
Lives for gear
 
mdoelger's Avatar
Either send it as it is, as long as it's not clipping or lower the master fader, if the ME really needs it 6dB quieter.

I don't understand all that -6dBFS thing either. I think it is just for the ME to make sure, that whatever newbies they are dealing with won't clip or limit the 2buss or something. I don't know. How much will the mastering cost? and why don't you communicate with the ME and clear things up?
Old 20th October 2013
  #9
Motown legend
 
Bob Olhsson's Avatar
 

Verified Member
The reason I leave headroom is because all but the most expensive converters don't sound very good as they approach clipping. I don't want their shortcomings influencing my mix decisions and locking me into something I didn't intend.
Old 20th October 2013
  #10
Lives for gear
 
mdoelger's Avatar
true, but the guy is already done with his mix and obviously satisfied with his own work.

Maybe next time he will find out by himself, that leaving more headroom might make his mixes even better. Until then, wouldn't you as an ME just lower the output of the track inside your DAW and then send it out to your analog chain?

How is that different for you as a ME compared to him leaving 6db headroom or whatever during his mix?
Old 21st October 2013
  #11
Motown legend
 
Bob Olhsson's Avatar
 

Verified Member
I've had the experience of turning it down and finding I wanted to make some changes. That's why I started paying close attention to headroom. I turn mastering projects down all the time, that isn't an issue but mix monitoring can be.
Old 21st October 2013
  #12
Gear Nut
 

thanks fellas....mastering is going to be about $180....now that I'm gonna spend some money it's making rethink my mix. I hate mixing bass, it never sounds perfect.
Old 21st October 2013
  #13
Lives for gear
 
polybonk's Avatar
Or you could bounce it at 32bit and not worry if it goes over 0. As 0 is now arbitrary.

Depends if your DAW can output 32 bit tho.

I get my ableton clients who mix(smash) into compressors and limiters to do this all the time after turning off the plugins on the master buss. Its idiot proof.

Then I just turn the 32 bit file down and master as normal.
Old 21st October 2013
  #14
Deleted 691ca21
Guest
Should clarify you are talking about 32 bit floating point, not 32 bit fixed. Big difference.
Old 21st October 2013
  #15
Lives for gear
 
polybonk's Avatar
True. Fixed would be floating pointlessness. : )

Ableton only outputs 32 bit floating point files.

Is there a DAW that gives a choice for fixed point output? That would be good to know.
Old 21st October 2013
  #16
Lives for gear
 
mdoelger's Avatar
I think sonar has that option. Not sure however.
Old 22nd October 2013
  #17
Lives for gear
 
FabienTDR's Avatar
 

Verified Member
Quote:
Originally Posted by polybonk View Post
Ableton only outputs 32 bit floating point files.

Is there a DAW that gives a choice for fixed point output? That would be good to know.
Of course, try Save as.. 24bit wav (I guess you knew it already).

The fixed point equivalent of 32bit floating point is 23bit (+1 sign bit). The extra 8 bits are (indirectly) used to store the scaling information. This is what gets removed (or added) during the conversion process. This also explains why one should actually normalize floating point files to the full target (fixed point) scale to avoid losses (and maybe little pre-dithering as well since such a conversion introduces minimal losses).

I'm not aware of any audio rate/precision AD converter able to sample 32bit numbers. The demands on the analogue circuitry would be far beyond anything imaginable today (~190dB SNR).
Old 22nd October 2013
  #18
Lives for gear
 
12ax7's Avatar
 

Verified Member
.
It really doesn't matter where your peaks hit if you have already robbed your mix of headroom by smashing/brickwalling everything at the mix stage.

If you have done THAT, you have also robbed the mastering engineer of one of the most powerful tools at their disposal.
.
Old 22nd October 2013
  #19
Gear Addict
 

Quote:
Originally Posted by polybonk View Post
True. Fixed would be floating pointlessness. : )

Ableton only outputs 32 bit floating point files.

Is there a DAW that gives a choice for fixed point output? That would be good to know.
Samplitude does.
Old 22nd October 2013
  #20
Here for the gear
 
sketchdashaman's Avatar
 

Hes talking about the Level on the master fader, before you hit tape (master file).
Dont mess with your mix if it sounds good.
-12 thru -6 are good levels on the master fader= what your VU's say.
Old 22nd October 2013
  #21
Lives for gear
 
polybonk's Avatar
Quote:
Originally Posted by blindjoni View Post
Samplitude does.
Good to know. Thanks.
Old 22nd October 2013
  #22
Lives for gear
 
polybonk's Avatar
Quote:
Originally Posted by FabienTDR View Post
Of course, try Save as.. 24bit wav (I guess you knew it already).
Cheeky.
Old 22nd October 2013
  #23
What is the point in normalizing 32bit FP files? Floating point values actually loose precision when storing larger values.
Old 22nd October 2013
  #24
Lives for gear
 
FabienTDR's Avatar
 

Verified Member
Quote:
Originally Posted by Orson Maxwell View Post
What is the point in normalizing 32bit FP files? Floating point values actually loose precision when storing larger values.
As mentioned above, to avoid losses in the conversion process. Normalizing a floating point file for "fun" is pointless of course.
Old 22nd October 2013
  #25
Ok I see that clause now
Old 23rd October 2013
  #26
Lives for gear
 
polybonk's Avatar
I found truncating a 32 bit FP file to 24 fixed and the same unclipped audio session bounced at 24 bit fixed will null.
Regardless of differences in the 32 bit bounce level wise when its pulled back to the same unclipped level as the 24 bit bounce. Unless you exceed 70dB over 0 on the fp file.
That is at least on the tests I have done.
Old 23rd October 2013
  #27
Lives for gear
 
polybonk's Avatar
Actually I just checked again and in fact it phase cancels to - 138.5dB. I thought that it was a null because nothing I used measures that far down.
But still that is so far down I don't know why you would bother dithering.
I cant see anyone saying: "Wow. Listen to that reverb tail crunch out as it slip's from -138 into -144!!! Ruins the whole track."
Old 23rd October 2013
  #28
Deleted 691ca21
Guest
Apparently the distortion can accumulate later down the line with further processing though. But if it's a final bounce, yeah, probably not worth worrying about.
Old 23rd October 2013
  #29
Lives for gear
Quote:
Originally Posted by XiviX View Post
I'm trying to figure out how to deliver a track with headroom for mastering. I'm using cubase 5. ME asked for -6 db's of headroom. It's a house tune.

I've learned to start my projects at lower levels to leave headroom, specifically the kick, I try to start around -8 db...however, this project has gone on for months now and has dozens of tracks/group channels ext....and somehow I'm finding that my Kick(loudest track) is now close to 0 db.

I want to create headroom so my first idea was to turn each track down evenly starting with the kick -6 db's, but as I proceeded to do this, I realized that I have a ton of level automation and I can't select all, and turn down.

My next idea is to create a new group channel and route everything to that channel and turn it down there before going to bus.

- My question: Is this a bad idea? Also, is there something I'm missing here? The kick track is the one giving me problems(the only one close to peaking), and when I turn it down, it kills the vibe of the track. The only other solution I can imagine is to turn each track down one by one, including the ones with automation which is not easy to do precisely.
The way I see it these days is that the compressors/limiters are the volume faders and the volume faders are the power/weight faders. I usually start somewhere in the -7 to -10 dBFS region, depending how much power I find I want initially. During post-mixing I push it to a target sound, with the aim of having the power/weight/volume structure pretty much fixed at that point. During pre-mastering I ensure I have a good enough signal for further limiting during mastering.
Old 23rd October 2013
  #30
Lives for gear
 
X-Pand Sound Mastering's Avatar
 

Verified Member
I'm doing Mastering for many different styles of music, including Electronic music, wich requires to be loud, as to be competitive, well thats usually what the client ask. But if you want to obtain the best "loudness" possible, without squashing the tune like hell at the end of the Mastering process, you need to work with files that have "real" headroom.
What I mean by "real" is that, you shouldn't at any time lower the Master fader so as to have "headroom" at the output. Because you're actually only downsizing the overall bit scale digitally speaking.
To have "real" headroom, you would need to have a signal under the clipping point without lowering the Master fader.

Its pretty easy to verify, for exemple in cubase :

Just open an loud song (0 DBFS) on an audio track, then raise the gain of it until it start clipping on the master fader by let say 3 db. Now lower the Master fader by 3 DB. Ok, now the meter (on the master output) shows you a green signal at 0 DBFS. But thats because the meter is showing you things "post" fader. So now, right click on that meter, and set it to show you things "pre" fader. You're actually clipping by + 3DBs.
If I'm asked to, I'm able to produce "loud" masters, and with only 0.5 to 1 DB of gain reduction on the final limiter. But each time somebody sent me a track with the master fader lowered (for exemple by 6DB) it wasn't possible to have something "that" loud in the end, that still sounded good.
Topic:
Post Reply

Welcome to the Gearslutz Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Forum Jump
Forum Jump