Skip to main content

Hello all,
I've heard people mention gain staging as being a very important step in setting up a recording chain. Can someone give me a rundown of what & why?
I rarely record vocals, so a typical chain for me would go like this Virus>Mangler>Fireworx>828 MKII>DAW. The guys in the middle all have an input trim and output level knob. It's not that I'm having problems at the present, but if I can do it better I will.

Thanks,

Comments

Zilla Mon, 11/07/2005 - 17:04

Briefly, gain structure is the practice of optimizing signal levels for each individual device in your signal path. What I mean by optimizing, is that you trim the gain so that you maximize signal-to-noise ratio and minimize distortion. Each device in your signal path will accommodate a given nominal signal level plus headroom. Now sometimes it is desired to abuse the gain structure to illicit some form of effect, like clipping. But generally good gain structure allows your devices to work at designed conditions, have low noise and distortion, and produce a higher fidelity recording.

saemskin Mon, 11/07/2005 - 18:29

interesting.
so, for example, my Kurzweil Mangler has green/yellow/red (-48/-12/clip) leds. The manual tells me what I knew, to keep away from clip and only partially in the yellow (-12) region.

So there's really nothing more to it than keeping a strong signal without clipping?

So, ass backwards would be to turn the Input level way down and crank up the output? We're looking to do the opposite.

groovy 8-)

anonymous Mon, 11/07/2005 - 21:40

saemskin wrote:
So there's really nothing more to it than keeping a strong signal without clipping?

Yep, there's more to it than that. Re-read the original answer. Your goal for proper gain staging is to maximize your s/n ratio and minimize distortion. Typically that means you'll want as much input gain as you can get while maintaining appropriate headroom and output levels at nominal.

The reason is that your input stage of any device is going to be your quietest stage (maximizing s/n). The output stage is going to have any inherited noise plus any noise from the unit itself. So using more input and less output gives you the minimum noise from that stage (hence the name).

So why do we use outputs at nominal rather than even lower? Can't we turn up the input MORE and lower the output? Well, yeah, but then you fuck up the other half of proper gain staging--reducing distortion. Your output controls will exhibit the best frequency and phase response at their nominal levels. For example, on a mixer the fader will add the least amount of distortion when it's sitting at 0, with distortion increasing as you pull the fader down towards negative infinity. So it's a give and take.

..