Gain Staging Structure: Waste of time?

[Compyfox]

have you ever seen a proper argument for "leaving headroom"

Here are some:
- no clipping, may it be soft or hard
- a bigger dynamic range in the final product, resulting in healthy transients and natural feel of the music/recording
- better listening experience - since stuff isn't pressed to it's limits, it's less ear piercing and causing less ear fatigue
- chances are you can reduce filesizes due to less used bits/file resolution - it's like a picture, the more colors, and the more to their limits, the higher the output file
- one of my favourites: you can easily incorporate hardware
- pressing to several mediums cause you no trouble either. Vinyl needs a certain headroom, so do AC3 or DTS streams, cassette tapes also need a certain headroom but also a fairly healthy level


There are several arguments for both sides of the medal. It's not about what is superior or what works and what not. Personally I think the middleway is most important - especially nowadays where there are ton of analog type plugins. Else we will never fight such things as loudness wars, format wars or even metering standards.

It doesn't have to be this way if we'd stick to certain rules. But you need to accept them in the first place. On top of certain hardware limits.

Balance!

[qa2pir]

Here are some:

and here's one against all of them: a master engineer can typically (depending on setup etc. and of course provided he has a fader or other means of gain control) lower the volume if required, thus adding headroom.

talking about "leaving headroom" in the sense of not letting anything peak above say -6 dbfs

leaving headroom in the sense of not squashing the shit out of everything in a poor attempt at mastering is of course desirable.

[MickGael]

As long as the master is not in red while exporting, everything is absolutely fine.

yep.

this reminds me of the old dogm "leave some headroom for the mastering engineer". seriously, does that make any valid sense other than being one more axiom/heuristic that people can substitute talent and experience with when feeling an urge to be superior and condescending?

All due respect: have you ever been to a mastering session or seen a mastering engineer work?

no. have you ever seen a proper argument for "leaving headroom"?

you're proving my point dude.

Had you been to a mastering session, you would understand why more clearly than any rebuttal I can offer.

[aciddose]

yes the purpose of "leaving headroom" is defined under the definition of "headroom".

that is: so that there is room for your head not being squashed. literally.

have you ever noticed analog consoles tend to have -36db, 0db, +18db (or similar) on the meters? why on earth would you ever need +18db?

notice that an equal adjustment on the fader for a channel will match the vu display. so if you see a +10db peak, you can directly adjust the channel -10db by looking at the fader rule and get exactly the result you want, immediately. no trial-and-error.

have you ever noticed professional systems have a mute/monitor/solo mode available to allow you to pick out individual channels or groups?

the whole point is to provide high accuracy measurements and lots of extra dynamic range so you can make fast, accurate adjustments.

in vst this is fine if you have +18db on the meters, but if you can no longer see the peaks it means you're running out of headroom.

headroom is not 0db. headroom is the maximum display range of the tools you're using.

if you throw a limiter on the master at 0db, you've just thrown away all your headroom.

[qa2pir]

Had you been to a mastering session, you would understand why more clearly than any rebuttal I can offer.

proving my point again, mystifying and revering the minor task of a mastering engineer simply because it demands more expensive gear.

[qa2pir]

headroom is not 0db. headroom is the maximum display range of the tools you're using.

yeah.

but the practice of "leaving headroom" as a separate production step refers to making sure that peaks don't exceed some limit like -3 or -6 dbfs. this is an example of another thing that's just redundant in digital audio.

[Funkybot's Evil Twin]

and here's one against all of them: a master engineer can typically (depending on setup etc. and of course provided he has a fader or other means of gain control) lower the volume if required, thus adding headroom.

Not if there's a clipped peak. The minute you start clipping peaks you're not only permanently cutting into your headroom, you're also adding distortion.

[qa2pir]

and here's one against all of them: a master engineer can typically (depending on setup etc. and of course provided he has a fader or other means of gain control) lower the volume if required, thus adding headroom.

Not if there's a clipped peak. The minute you start clipping peaks you're not only permanently cutting into your headroom, you're also adding distortion.

yeah then slap a limiter just for safety measures in case the song should behave differently when you export than all the 100 times you've played through it.

[Funkybot's Evil Twin]

and here's one against all of them: a master engineer can typically (depending on setup etc. and of course provided he has a fader or other means of gain control) lower the volume if required, thus adding headroom.

Not if there's a clipped peak. The minute you start clipping peaks you're not only permanently cutting into your headroom, you're also adding distortion.

yeah then slap a limiter just for safety measures in case the song should behave differently when you export than all the 100 times you've played through it.

Or just leave plenty of headroom so that you're not limiting that stray clipped peak. Instead, there'd be no clipped peak.

Here are the biggest reasons why I like leaving headroom in my mixes:

1. Analog modeled plugins tend to use analog reference levels so they tend not to sound their best when slammed. Not all have input levels, and even adjusting that, is "gain staging."
2. If you're nowhere near clipping the master, then it's never an issue. I'd rather not have to worry about it. Ever.
3. If you record less hot and mix at lower levels (monitor louder), then you've got plenty of space (or headroom) to bring things up in the mix. If everything's hot with faders up and the kick is getting lost, you could end up having to bring down twenty different channels to get the relative level of the kick up. If you left yourself lots of headroom, there's plenty of room to bring that kick fader up. I see this all the time when I mix stuff for some of my less experienced friends.

[qa2pir]

ok so you'd rather waste some bit depth on the entire file than have 100 ms of light limiting at one point in the song?

[cron]

No bit depth is wasted. 24 bit has a ton of headroom to make a 16 bit master before you start losing resolution. That's why there's no point in trying to get close to digital zero ITB pre-master.

[Funkybot's Evil Twin]

No bit depth is wasted. 24 bit has a ton of headroom to make a 16 bit master before you start losing resolution. That's why there's no point in trying to get close to digital zero ITB pre-master.

Cron is correct. If exporting a 24 bit or 32 bit file for mastering, there's not going to be a loss of bit depth (the dynamic range on these resolutions is huge). When it comes to the final master going down to 16 bit, the mastering engineer will have done the work to have raised the levels to maximize the bit resolution.

In short: there's absolutely NO HARM in leaving plenty of headroom, and there are certainly valid arguments against. If one chooses not to leave themselves any headroom, then that's fine with me. But they'd just be making things harder on themselves.

[jupiter8]

have you ever seen a proper argument for "leaving headroom"

Here are some:
- no clipping, may it be soft or hard

0.1 dB is enough. If you're below that you get no clipping,hard or soft.

- a bigger dynamic range in the final product, resulting in healthy transients and natural feel of the music/recording

No you don't if anything you get less. Max peak - noise floor.

- better listening experience - since stuff isn't pressed to it's limits, it's less ear piercing and causing less ear fatigue

Or you could just turn your amplifier down.

- chances are you can reduce filesizes due to less used bits/file resolution - it's like a picture, the more colors, and the more to their limits, the higher the output file

Really ?

- one of my favourites: you can easily incorporate hardware

Ah c'mon you'll need to adjust the volume no matter what,how is that any simpler ?

- pressing to several mediums cause you no trouble either. Vinyl needs a certain headroom, so do AC3 or DTS streams, cassette tapes also need a certain headroom but also a fairly healthy level

Seriously ?

[qa2pir]

if your plug-ins behave linearly, predictably, non-randomly (which most do) you shouldn't have to worry about having to leave headroom. and in terms of quality, some extra bits to work with matter more than satisfying an old myth. that's the relevant headroom.

[MickGael]

Had you been to a mastering session, you would understand why more clearly than any rebuttal I can offer.

proving my point again, mystifying and revering the minor task of a mastering engineer simply because it demands more expensive gear.

Such flamboyant logic.