'Gain Staging' concerns

[questionss]

I'm fairly new to the world of Pro Tools and would like to start by learning appropriate gain staging that can be "universal" to most mastering engineers and other PT technicians that work after me.

First things first, what is proper/great gain staging in Pro Tools in YOUR eyes? I'd like some opinions here. where would YOU, the engineer generally peak your drums or even the entire mix?

secondly, what is the difference between LE's mono TRIM/GAIN Plugins and when would I want to use either over the other?

thanks in advance.

[daeron80]

Wow. The term "gain staging" is seldom used around here. Makes me feel warm and fuzzy all over.

If there's a consensus among reputable recordists about how hot to track - and this applies to any A/D system, not just DAWs and certainly not just PT - it would be to shoot for somewhere in the -18 neighborhood for typical peaks, -12 max, assuming 24 bit dept. If you have an occasional single peak in the -6 range, no biggie. Pushing cheap converters harder than that is unwise. If you operate your analog gear at nominal levels, you should be fine. Digi interfaces should see 0 dBVU as somewhere in the negative upper teens.

In LE, mix levels almost don't matter in one sense, because it's floating-point math. With HD/TDM, you have fixed-point math, and anything over 0 dBfs at any stage will hard clip. But with LE's floating-point math, there's really no ceiling. You can have every track over by 72 dB, and just pull the master down until it's not clipping there, and you're fine. Theoretically.

Now, you do have to know your plug-ins, and consider how they handle hot inputs. Some do better than others. Some plugs distort internally even using float. My favorite de-esser has a hard ceiling that I have to be careful not to hit (Spitfish VST). If you hit a dynamics plug at +dBfs levels, it may not have been programmed with sufficient threshold range or whatever to be able to give you what you want from it. Hitting a saturation modeling plug really hard might get ugly. So, it's still good practice to gain stage at reasonable levels throughout, even though it's not strictly necessary. Headroom is good.

Unless they changed it with 8.x, Gain is Audio Suite only, and Trim is RTAS only. Other than that, the only difference I can think of is that Trim has a polarity switch. I never use Trim for gain staging. I use it to level out a cock-eyed stereo image or to do nulling tests and experiments and such. That's it. To change the gain of a channel, I usually just change the fader level or trim the volume automation.

[weighbor]

in my case i regard -6dBFS as the highest peak of all of my tracks.. i situate something like kick drum at that level and the rest proportionally smaller as they have more upper frequency components.. is this ok? when i record, i tried to set fader in a way that my source at peak will be at -6dBFS .. is this ok quality wise or should i go down even lower to be safe (although i guess i have to trust my ears and intuitions :) ?

[Sounds Expensive]

Originally Posted by weighbor

in my case i regard -6dBFS as the highest peak of all of my tracks.. is this ok quality wise or should i go down even lower to be safe (although i guess i have to trust my ears and intuitions :) ?

I would go lower just to be on the safe side. Leaves lots of headroom at the 2 buss and for plug-ins, plus you can concentrate on the take, not the levels possibly clipping - cause you never know.

[daeron80]

As far as mix levels, it depends on where it's going from there. There are different kinds of mastering. If you're sending it off to a Bob (Ludwig, Katz, Olhsson, ... ), ask them what they prefer. Probably mix peak no higher than -6. But if the mastering facility is going to do nothing more than sequence it and maybe tweak a level here and there, you'll want to peak at about -1 or maybe a hair higher. If you peak higher than -0.7, especially brickwall limited, your transients will likely distort on playback (ever wondered why most recent music tires your ears so much more than the old stuff?). They'll also likely clip when lossy encoded (mp3/ACC/ogg, etc) because perceptual encoders, like D/As, render intersample peaks.

Remember than 24-bit has 256 times the resolution of 16-bit - that's 48 dB of headroom. As long as your mix peaks above -48 you'll have greater resolution than the final 16-bit CD master. You know, if you're still hawking your wares on that obsolete medium. So don't feel like you have to squeeze every last drop out of the final level.

[Bob Olhsson]

There's lots of ridiculous mythology about "mastering headroom."

The two main reasons I use moderate levels when mixing are:

1. to not waste mental energy on which plug-ins need headroom and which don't

2. to make absolutely sure the monitor D to A analog stage isn't being stressed which could easily affect my signal processing choices just like a fuzzy viewfinder could result in an unintentionally out of focus picture.

[samataaudio]

Originally Posted by daeron80

As far as mix levels, it depends on where it's going from there. There are different kinds of mastering. If you're sending it off to a Bob (Ludwig, Katz, Olhsson, ... ), ask them what they prefer. Probably mix peak no higher than -6. But if the mastering facility is going to do nothing more than sequence it and maybe tweak a level here and there, you'll want to peak at about -1 or maybe a hair higher. If you peak higher than -0.7, especially brickwall limited, your transients will likely distort on playback (ever wondered why most recent music tires your ears so much more than the old stuff?). They'll also likely clip when lossy encoded (mp3/ACC/ogg, etc) because perceptual encoders, like D/As, render intersample peaks.

Remember than 24-bit has 256 times the resolution of 16-bit - that's 48 dB of headroom. As long as your mix peaks above -48 you'll have greater resolution than the final 16-bit CD master. You know, if you're still hawking your wares on that obsolete medium. So don't feel like you have to squeeze every last drop out of the final level.

Hi! I am new to the Protools LE and Iwant to know that what is the maximum lebel should be in case of mixing or mastering of any audio project.....because when I am mixdown a file, its not go up to the mark.

[daeron80]

Don't worry about the mark. Just don't clip. If it sounds good, it's fine.

[SKI]

Just because the Master fader doesn't clip doesn't mean you don't have bad clipping or distortion.
You could easily be overloading a channel or plug in input.

[daeron80]

Originally Posted by SKI

Just because the Master fader doesn't clip doesn't mean you don't have bad clipping or distortion.
You could easily be overloading a channel or plug in input.

Not too easily, actually. Possible in certain rare cases. I have one plug-in that is poorly coded and can clip, so I have to watch it with that one. But the LE engine uses floating point math, which yields what amounts to unclippable internal processes.