Gain structure in a Digital world

C

Cave Dweller

New member
I'm just wondering how people are running their gain structure, especially regarding digital. It stems from a conversation some co-workers and i were having.

I guess i should start off by saying I am a live sound tech, and I approach recording pretty much the way I would set up a live mix: I set all my faders at "0", and adjust my gains until i have a pretty good mix, then use the faders to tweak the mix.

This usually means that sources that are generally up front in the mix (ie kick, snare, bass, vocals) are closer to "0" db than, say hi hat, or guitars. My co-worker argued that in digital everything should be as close to "0" as possible to get the most out of the chosen bit rate. To me this seems like turning up the gain, only to turn it down later, which is a big no-no in analog gear.

So lets strike up a discussion! I want to hear some opinions and arguments for or against either side. What do you think?
 
First off... let's clarify one thing -- it's BIT DEPTH or WORD SIZE, not "bit rate"!!

IMO, at a wordsize of 24-bits, it is FAR less important to worry about bringing levels close to 0dBFS to "maximize resolution." A case could have been made for it at 16-bit or lower, but not with converters of good quality at 24-bit.

The other thing you have to clarify is the frame of references between 0VU (analog), and 0dBFS (digital).... they are completely different scales... given a common gain structure, a signal that reads 0Vu on analog meters can read anywhere from -18 to -12 dBFS, depending on where the manufacturer calibrated their input levels to be.
 
What

Cave Dweller said:
To me this seems like turning up the gain, only to turn it down later, which is a big no-no in analog gear.
Click to expand...
Why is that a no-no in analog? The reason for the gain control is to bring the input signal to line level in order to keep the signal to noise ratio in check. The fader is where you are supposed to turn it down.
jason
www.farviewrecording.com
 
There are a number of reasons why this is a big no-no. You actually get MORE accumulated noise if you are turning up one gain, only to turn it down later in the signal path. (or vice versa, turning it down, only to turn it back up). The noise builds up if this is done on every track (or channel, in the case of live audio)
Another reason is most fader pots have an optimum area ("0")where their frequency response is most accurate. The fader also responds better at "0" db,where a small bump doesnt mean a big change in level, as it does if you were running in the -30 area.

Generally speaking, a sign of a good live engineer is if he/she has a decent mix with all the faders near "0". Everything is pretty smooth sailing at this point, and extreme gain and attenuation of outboard FX, monitor sends etc. arent needed.

I have found that a mix based on bringing up the input to "0" db on the meters, then adjusting the fader to the right level usually results in a "mushy" and muddy sounding mix. aux send levels are unpredictable, and things have the potential to turn into a nightmare. it's just way way easier to leave the faders at "0", and adjust the gain accordingly.
 
Maybe it's just a live thing. Maybe you don't have a lot of compressors and outboard eqs and other stuff on the inserts.
 
Are you talking about tracking or mixing? If you are tracking then all the tracks should be relatively hot. If you are talking about mixing on a DAW then I don't understand where you are applying gain.
 
You must log in or register to reply here.
Back
Top