Author Topic: Question About Headroom (Read 15632 times)

Lydian · « **on:** May 08, 2016, 06:41:05 am »

Why are people always stressing about headroom? I get that things need to be below -0db and stuff but if anything clips I can just bring the entire mix down in logic pro by selecting all the volume faders and dragging them all down at once. It keeps the balance the same but lowers the levels.

So why do people stress about headroom as if getting it right in the beginning of a mix is so important it's gonna cause irreplaceable damage if it's wrong at first?

manducator · « **Reply #1 on:** May 08, 2016, 06:50:47 am »

I always had the same thought, Lydian.

But I think there are daws that don't have the option to lower all faders at once (like FL 11)? And then people start messin with the master fader.

Headroom is important while recording analog instruments; once it's recorded too hot (distortion) or not loud enough (bad signal to noise ratio) it can't be changed.

But for people like me, only working with in the digital domain, the creation of headroom is just a fader move away.

Miles Dominic · « **Reply #2 on:** May 08, 2016, 09:13:12 am »

Quote from: manducator on May 08, 2016, 06:50:47 am

But for people like me, only working with in the digital domain, the creation of headroom is just a fader move away.

this x100000

Lydian · « **Reply #3 on:** May 08, 2016, 01:50:38 pm »

Makes sense. I can see how a signal recorded past -0.0db would be impossible to bring down if it were recorded like a vocal or a guitar.

Can't fix a clipped signal. I just wanted some clarification on the issues. I wasn't aware that some daws don't offer the option to bring all the faders down at once.

Guess I gotta love logic even more for that. <3

Arktopolis · « **Reply #4 on:** May 08, 2016, 02:03:55 pm »

^ Off-topic, but declipping is actually possible to some extent: https://www.youtube.com/watch?v=Kcq0zsdJRxE.

Lydian · « **Reply #5 on:** May 08, 2016, 02:17:27 pm »

Quote from: Arktopolis on May 08, 2016, 02:03:55 pm

^ Off-topic, but declipping is actually possible to some extent: https://www.youtube.com/watch?v=Kcq0zsdJRxE.

Wow

Thanks for sharing arktopolis. I actually didn't know that. I'm sure that knowledge will be useful in the future.

Babasmas · « **Reply #6 on:** May 09, 2016, 07:43:50 am »

Quote from: manducator on May 08, 2016, 06:50:47 am

I always had the same thought, Lydian.

But I think there are daws that don't have the option to lower all faders at once (like FL 11)? And then people start messin with the master fader.

Headroom is important while recording analog instruments; once it's recorded too hot (distortion) or not loud enough (bad signal to noise ratio) it can't be changed.

But for people like me, only working with in the digital domain, the creation of headroom is just a fader move away.

FL 12 has the option now. Fader can be selected (more than one) and we're able to lower them all at once.

Miles Dominic · « **Reply #7 on:** May 09, 2016, 07:46:45 am »

Quote from: Lydian on May 08, 2016, 01:50:38 pm

Makes sense. I can see how a signal recorded past -0.0db would be impossible to bring down if it were recorded like a vocal or a guitar.

Can't fix a clipped signal. I just wanted some clarification on the issues. I wasn't aware that some daws don't offer the option to bring all the faders down at once.

Guess I gotta love logic even more for that. <3

You can also just put a gain plugin on the output and turn it down. Easier than turning down all the faders

FarleyCZ · « **Reply #8 on:** May 09, 2016, 09:07:38 am »

Same here. Imho there's no point in stressing out about this. I'm not sure if VST interface can accept floating point audio, but as far as the DAW audio engine goes, you can't really clip anything there these days. As far as it's not clipping on master, all is fine.

Lydian · « **Reply #9 on:** May 09, 2016, 02:11:05 pm »

I guess I won't stress about it. I just wasn't sure whether this was leading to poor mixing consequences. I've been lowering the faders proportionally for a very long time but never really understood why so many people would stress the importance of headroom.

I think I understand now a little bit better.

Mussar · « **Reply #10 on:** May 09, 2016, 02:47:07 pm »

Allow me to make an argument for headroom, though it's super subjective, because even inside the digital domain there are a few good reasons to employ proper gain staging.

First, it teaches you good habits like gain staging, which can be translated into the physical world (you want your signal coming into the mic and thru the preamp to be hot enough to not need a lot of boost to the level afterwards while quiet enough to allow a good amount of dynamic range. It also teaches you the very valuable lesson of Red Is Bad, which in my opinion is as valuable as learning the limits to 32-bit floating point audio.

It also prevents unwanted behavior from happening with third party software. Like Farley hinted at, some plugins will respond differently to a redlined signal because of how their algorithms function. Either they're designed to switch and function differently, like certain analog gear would when you pushed a too-hot signal through it, or they're just designed to expect normal levels and break down at that volume.

Finally, it helps you maintain a clear and representative mixing environment. Your DAW might be 32-bit but your interface isn't and your speakers sure as hell aren't. Why would you even want to risk having the signal distort and give you a signal response that's not actually present? Of course, this can be solved just by putting a clean limiter on your master or using a volume trim plugin like Utility or Fruity Balance to pull down everything before going out to your speakers down -5 or -10 dB if you really don't wanna gain stage before that.

Marrow Machines · « **Reply #11 on:** May 09, 2016, 03:17:00 pm »

Quote from: Mussar on May 09, 2016, 02:47:07 pm

Allow me to make an argument for headroom, though it's super subjective, because even inside the digital domain there are a few good reasons to employ proper gain staging.

First, it teaches you good habits like gain staging, which can be translated into the physical world (you want your signal coming into the mic and thru the preamp to be hot enough to not need a lot of boost to the level afterwards while quiet enough to allow a good amount of dynamic range. It also teaches you the very valuable lesson of Red Is Bad, which in my opinion is as valuable as learning the limits to 32-bit floating point audio.

It also prevents unwanted behavior from happening with third party software. Like Farley hinted at, some plugins will respond differently to a redlined signal because of how their algorithms function. Either they're designed to switch and function differently, like certain analog gear would when you pushed a too-hot signal through it, or they're just designed to expect normal levels and break down at that volume.

Finally, it helps you maintain a clear and representative mixing environment. Your DAW might be 32-bit but your interface isn't and your speakers sure as hell aren't. Why would you even want to risk having the signal distort and give you a signal response that's not actually present? Of course, this can be solved just by putting a clean limiter on your master or using a volume trim plugin like Utility or Fruity Balance to pull down everything before going out to your speakers down -5 or -10 dB if you really don't wanna gain stage before that.

I love you mussar.

Lydian · « **Reply #12 on:** May 09, 2016, 03:58:17 pm »

Tbh my main concern was that since the limiter catches the peaks it would change the sound of the mix as I'm mixing into it. I like being able to see which instruments are clipping the master but that's not really an option with the limiter on. The main reason why I use it is because of volume reasons. I have a habit of gain staging instruments really low since digital synthesizers don't really deal with the noise to floor ratio issue. I know you can just say turn up the monitor levels on your speakers which I do follow but the thing is it makes it harder for me to reference mix. When I play the music on iTunes it comes out super loud. Its kind of annoying even though its just a bother that can easily be fixed.

I think from now on ill stick to just using the utility plugin to boost the volume since it maintains the transients.

Mussar · « **Reply #13 on:** May 09, 2016, 04:48:16 pm »

Well remember that loudness should not be achieved by the result of a single plug-in, but as a cumulative effect of dozens of little tweaks and improvements here and there - which is where headroom starts to come into play.

Loudness is not just slapping the Waves L1 on your master and pulling it down until it sounds big. Loudness comes from proper fader and pan pot mixing first and foremost, and then carefully stacking EQ and compression and saturation to sculpt each sound into its own pocket. Then, the sort of mastering effects like stereo imaging, mastering EQ, harmonic excitement, and of course the final compression and limiting glue all of those elements together.

For every point where you don't properly gain stage, there's a chance for an effect to not have as much room to work or not provide as much information as it could if it were within the traditional range. I suppose you could just put a utility plugin before your mastering chain and pull it all down to rest at -6 or -12 dB or whatever you feel is a good amount, but if you're going to get the signal to that level anyways why not just get in the habit of just pulling the faders down before going to mix?

If you want to reference mix, instead of turning your own sounds up why not turn the reference tracks down? Set up a reference track inside your daw with the signal sent directly to your external outputs, and then once your song is done and you're ready to mix just find out where your track is peaking at, pull the reference song down to that volume, and work from there!

Aerithos · « **Reply #14 on:** May 09, 2016, 08:32:33 pm »

Something else to consider: DAW audio is handled as 32bit floating point internally (1bit sign, 8bits exponent, 23bits fractional). Each magnitude (..., 10^-2, 10^-1, 10⁰, 10¹, 10², ...) has an equal precision of 24bits signed.

Examples:

[0.1 to 1.0) has 23bits of precision
[0.01 to 0.1) has 23bits of precision
[-0.1 to -1.0) has 23 bits of precision
Etc.

So each smaller magnitude distributes the precision across a smaller range of values. In the reverse direction, each larger magnitude distributes the precision across a larger range of values.

Examples:

[1.0 to 10.0) has 23bits of precision
[10.0 to 100.0) has 23bits of precision
[-1.0 to -10.0) has 23bits of precision
Etc.

Unfortunately, this means that when your audio exceeds 0db, each magnitude of volume increase spreads the precision across a larger number of values.

Examples:

[0db to 1db) has 24bits of precision
[1db to 10db) has 24bits of precision
[10db to 100db) has 24 bits of precision
Etc.

At the 10db to 100db range, the result should be an audibly decayed bit depth.

So essentially, when your volume exceeds 0db, any samples outside of the -1.0 to 1.0 range have a decreased precision. At least that's how I would expect it to work with my rudimentary understanding of IEEE single-precision floating point.

I am going to give it a test when I get home tonight:
Synth -> 100db Gain -> -100db Gain -> Master

Author Topic: Question About Headroom (Read 15632 times)

Lydian

Question About Headroom

manducator

Re: Question About Headroom

Miles Dominic

Re: Question About Headroom

Lydian

Re: Question About Headroom

Arktopolis

Re: Question About Headroom

Lydian

Re: Question About Headroom

Babasmas

Re: Question About Headroom

Miles Dominic

Re: Question About Headroom

FarleyCZ

Re: Question About Headroom

Lydian

Re: Question About Headroom

Mussar

Re: Question About Headroom

Marrow Machines

Re: Question About Headroom

Lydian

Re: Question About Headroom

Mussar

Re: Question About Headroom

Aerithos

Re: Question About Headroom