Sorry, I have to single this out. This is the single biggest audio bullshit lie ever. What gear do you own that costs less than $1000 that has self noise below the noise floor of a 16bit signal when running i/o stages at normal gain settings (even assuming crazy-as-hell-expensive converters used for capture that have self-noise well below that realm as well)? None? Yeah, that's what I thought.

24bit does f**k all for anyone if your noise-floor prints at anything above -96dbfs when it's digitized. It doesn't matter what resolution you record at the real problem is never bit-depth it's always the actual noise-level.



Basic concepts on why things work? So you mean like the basic concepts of floating point audio streams I guess is what you're saying?



It's funny, because looking at this "point" again just on it's own exposes how little time you've probably ever taken to understand what actually happens when applying processes to an already digitized audio stream with most any modern DAW. Code is a funny thing my friend. Seriously, check into that stuff... it's very enlightening.

i think what the OP in gearslutz was trying to say is that people treat peak meters as VU meters. they're not VU meters. they operate completely differently - VU meter measures sound pressure over a period of time (few hundred ms), while peak meter measures momentary audio levels. 0VU is like -20 dB on a peak meter, so you clearly *can* go into the red on hardware because you've got quite a bit of headroom there. with peak meters, you can't.

but, as rifftrax rightly pointed out, so long as you are inside the DAW and don't use any plugins that depend on input/output level (e.g. don't have "sweet spots", don't clip on 0dB etc.), you can basically mount oozles of dB on your track, then reduce it by the same amount on the master and you're golden. this is how computers work.

now, tracking with high input levels is another topic alltogether and it is not related to how computers work and how DAWs work. and don't confuse having overs *inside* your DAW with having overs *when the sound leaves the DAW*.

Mostly I apply something akin to gain staging for the sake of having a bit reference between projects. I know roughly where I'm going to want to look at for my compressor thresholds, for example. I know roughly where the fader is gonna be for that clap track. I know roughly how much send I'm gonna need toward that reverb. I have to reinvent the wheel a little less often. And it subtly help for troubleshooting.

But in large part, it's because I'm a sucker for technical consistency. So in a way it's a personal dogma, but not a social dogma. I see those as a little different. And it doesn't seem detrimental, nor a major waste or time, in this case. It's kinda like my partly arbitrary preference for round integers on my faders, instead -7.23 or some other unsightly decimal. They're unnecessary idiosyncrasies, but no one will ever, EVER hear a detrimental difference between when they're followed versus not.

yeah but gain staging DOES NOT make any difference in most DAW's.

you're not flattering yourself by exposing your knowledge as a simple set of arbitrary heuristics.

There seem to be two different conversations happening simultaneously in this thread.

Gain-staging DOES make a difference in my DAW because it's a workflow thing, not a sound thing. Keep things low, stack the tracks, no mixer fiddling, consistency across projects and neatness within them. I don't think it makes any difference to the sound or anything (other than the aforementioned flexibility with some effects), since I don't really use analog modelled plugs anyway. I feel a bit uncomfortable calling it 'gain-staging' as it obviously had a very different meaning and function in the pre-DAW world. I mean, I could let all my tracks have random volumes pre-mixer so I get f**k-all visual feedback from my fader positions, and I could go into the red by random amounts all the time on each project, turning the master down each time it happens and thus changing the volume I'm monitoring at, but... why? What's the point? Fair enough if you're recording in, but why would you even consider going near the red anywhere in an entirely ITB project?

yeah two conversations:

"we're not limited to a certain workflow!!"
"we should stick to the tried-and-true, noob."

there's no point to touching the master but then again no point in not doing so. if you like to keep it neat and tidy that's a personal thing. don't pretend otherwise. I could begin blabbering about what colours are best for midi tracks in cubase to minimize confusion but that's not production technique.

kthxbye.

I would say the point is knowing why it doesn't matter much and why it does.

If you're in the zone, banging away, having the time of your life, do know that every single track clipping doesn't matter much and you need not worry about it.
You can fix that later or just live with it.

On the other hand there are good reasons to keep an eye on it but it's not important enough to stop the flow if that's what you're having. First of all the meters don't do much good if they're full scale all the time. Some plugins are input level sensitive and maybe you'll get better results trimming them down a bit.

The important bit is to choose your battles and know what's important and what's not and when it is important and when it's not.

It is, unless you think the human behind the controls isn't a particularly important aspect of your setup.
Last edited by cron on Sun May 13, 2012 6:21 am; edited 1 time in total

Why not think about this way?
Hosts usually work at 32bit internal floating point precision (or even higher, but floating point regardless). Now imagine a host that is working at a fixed point precision or even lower than 24bit. What happens is that you loose out of dynamic range, and therefore a possible headroom/footroom, etc.

Now, this all might not concern you as long as you don't hit the reds on the master bus, but again - what if the host works in lower bitrates? You then would need to utilize proper gain staging.


Also, Gain Staging is important if you want to incorporate hardware with software and the other way around. Eric Beam wrote a great article over at his blog on "RhyhtmInMind". Also if you come from analog days, knowing the "limits" might benefit from mixing within your DAW.


The thing is that analog consoles work different than digital ones. They usually do not measure the digital peak (at 0ms rise/fall) but are lazy in terms of the response. 5ms rise and suddenly you have a shifted response, you don't see as much transients anymore. Change it to 300ms rise (to resemble an RMS meter), and bam - no transients at all.

Most DAW's however don't have such idle settings and therefore show you every transient. But they lack in terms of the loudness readout (RMS meter, VU's). So you have to think different.

A drum hit (snare) at -10dBFS digital has a VU readout (calibrated at -18dB) of maybe -5VU at most. While a bass signal at -10dBFS digital can have a VU readout of +8VU. You would be in the reds and therefore clipping a channel.

Again, with floating point mathematics - you don't care. But it is important if you want to use hardware, incorporate analog-modeled plugins with a reference level, etc, etc, etc.



So is it worth the hassle?
I'd say yes. Larger headroom, better resolution for your faders, less chance of intersample peaks that can cause clipping... Use it or don't - it's up to you.


BONUS:
Read my KVRmarks - I've posted in several threads regarding gain staging.

ok.

now flood the production technique forum with old geezers being superior about which office chair is the best.

Sorry, I didn't mean to be offensive or anything. I just think that people sometimes forget how important good workflow is. Mind you, I might just be overcompensating within Live given the fact that my DAW is filthy 6 year old laptop that usually balances on a chopping board on my lap. I'll try to upgrade that situation before giving advice on office chairs

of course workflow is important but getting rid of a technical limitation enables the development of new types of workflow!

Good point!

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.