noob gain-staging question

[X14Halo]

I have read enough about gain-staging on this forum to know what it is and why it is needed. I just want to make sure I have it down right though...
So I record all my tracks (with no clipping), then bring them into my DAW (reaper) and put everything under a group. I then lower the volume of this master group so that the highest everything gets to is about -18db (or whatever you choose). I can then start applying fx like compressors, vintage warmers, and other things to this master track, and my goal is to get everything back to about 0db in the end....is this correct?

 

[jackbraglia]

what you're talking about here is clipping, not gain staging. when we talk about gain staging it's typically in the analog world where we are trying to achieve the highest possible signal to noise ratio without distortion. in the digital realm, gain staging would simply be how much extra gain you're applying from one plug-in to the next when in series, so that you don't boost the noise floor too much.

with that said, everything you did mention in your post is good practice...except for trying to hit 0db on the master. You're not mixing down to tape, so there's no need to push your levels that high, you've got plenty of headroom in 24bit, so there's no need to be peaking any higher than -6db on the master, unless you want an angry mastering engineer on your hands

 

[jackbraglia]

also, i don't start adjusting faders until I've appropriately compressed every track, or else you'll just have to go and re-adjust all your faders again after you've compressed.

 

[ashgallows]

tracking at around -18 is better as then you wont have to insert trim plugins and such. also you can snap all faders to zero at that point and have equal power across all tracks, which can be useful for eqing and compression descisions.

 

[X14Halo]

I see...so I should record my stuff at -18db into reaper, and turn up the knob on my headphones volume so that I can hear everything nice and loud?

 

[RedDog]

^ precisely.

 

[X14Halo]

Cool, thanks everyone!

 

[pikachu69]

I have read enough about gain-staging on this forum to know what it is and why it is needed. I then lower the volume of this master group so that the highest everything gets to is about -18db (or whatever you choose). I can then start applying fx like compressors, vintage warmers, and other things to this master track, and my goal is to get everything back to about 0db in the end....is this correct?

Well judging by this I dont think you understand gain staging at all tbh.
It starts right at the initial recording phase when you get your source into you daw. After you have finished tracking a drum track, for example, you should be able to listen to the playback WITHOUT HAVING TO ADJUST FADERS TO MAKE IT SOUND GOOD.

All faders should be left at 0 both on the desk/interface you track through and the initial playback faders in your daw. (pre mix)

As the musician/s play you should adjust the input trim on your pre to balance each mic with each other. Before you start tracking, the drum sound coming out your monitors with the drummer playing live should sound like a nicely balanced drum kit almost to where you want the final sound to be.
This will mean close mics like kick/snare will be recorded hotter than say the OH or room mics. Dont just crank the gain until every mic is just below clipping because alot of these mics will not be used that much in the mix. You can either turn it down while tracking or drive your pre too hard and just end up turning it down in your daw anyway. If you do the latter you can not undo what driving the pre to hard will do to your audio though...

Once the drums are tracked in this way, track everything else with gain that makes its it sit well with the drums. Once you finish tracking you should have a reasonable balance between all instruments with all faders still at 0. Only once you start adding EQ and compression etc will the faders need to move and even then I will tend to group things that need to be moved so the main faders stay at 0 where possible. If you have plugins on your tracks then use the gain controls on those to keep the mix balanced.If you bypass your inserts on any given channel and there is a massive difference in volume pre/post then your gain staging in your plugins is not right either.
I only use faders for automation 95% of the time.

And no the goal is not to get it back to 0. It is the job of the mastering enginner to provide the loudness. The louder you push it the more likely you are to put things like clipping and limiting on your groups or master buss and really just making it harder for the mastering enginner to do his job. If you like to monitor your mix loud (which I dont recommend) then simply turn your monitors up rather than tracking hotter.

 

[mva801]

^^^ This is closer but I don't really agree 100%. God made faders for a reason. If you wanna move one, then move it. You shouldn't record your tracks too hot, but you don't necessarily have to pre-mix your album with the mic pres on the desk. I usually work on an old ass API console, and while it sounds tits, the knobs on that thing are touchy as a mother fucker. I use the faders to trim all the time.

But I think the jist of what he's saying is, set your levels with plenty of headroom, and go for a semblance of balance before you start getting crazy with the faders, and this will insure you don't run in the red. the peak meters on your master should stay below -6. I'll usually let my drums peak around -10ish, but if it's over or under that slighty, it won't really matter that much. Use your ears and avoid red lights!

 

[pikachu69]

^^^^

And I agree with this too. The only difference I can see and the reason we disagree is more to do with our choice of gear. I understand on a desk like you describe using faders a bit is needed and not a bad thing. I use a Presonus studiolive and the faders do not effect what the daw recieves so any adjustment I make there only effects my live monitoring. By relying on the trim pot for balance instead of faders it means when I listen back to the recording it sounds the same as in tracking.
Balancing the whole mix on the way in speeds up mixing considerably for me so I like to do this.

 

[mva801]

Oh I feel what you're saying. I think I'm going to make an effort to balance even more through initial levels than I do now. For some reason when I see an ambient mic on my guitar chain coming in where I want it, but at like -50db, it just feels weird as fuck to me so I bump it up a bit and then roll the fader down.

I just didn't want the guy to think he was doing something wrong if he ever touched a fader. Much of engineering is about adapting to your circumstances, and there's many a time when going to tape, or when wanting some pre-amp saturation on a second snare mic, that I'll run the pre amp up and throw the fader down til I get it where I want it.

 

[aviel]

how will this be done using vst inst? like maybe superior drummer? i've noticed they play at almost 0 db, so i will have to adjust the plug master volume so it plays lower? or trim it later?

 

[mva801]

Just go inside the plug and turn the faders down, or turn the master volume of the whole plug down a bit.

 

[Beneath_the_Bliss]

 

[Uros]

Record with peak level no higher than -6dB. If you are only mixing, use trim pot/trim plugin-not fader to adjust the volume level of tracks so they sit between -12 and -6dB peak. Only then start applying plugins. Try and maintain that level (roughly - by using your ears) the same throughout the whole plugin chain (when you go in and out of a plugin, in other words - unity gain). Avoid plugin internal clipping (except on saturation plugins). Use faders only to mix (or, to adjust overall track level against other tracks in the project). Don't clip any buss, and certainly don't clip your 2nd buss. Done deal.

While there's nothing 'wrong' with using faders to adjust track level, it's good to make it a habit to use faders only for balancing&automation, and not for correcting volume levels. Make it systematic, don't be all over the place.

 

[r3c0rd_3m_411]

Ressurecting the thread a bit. Superior drummer sound kinda weird when gainstaged to about -18db. The sound starts to suck, even with monitors adjusted to sound accordingly loud. I guess the problem is, that i adjusted all tracks to -18, while tracks like hihat and OH's should be even below -18db right? So the drumkit should be not just gainstaged but really balanced?

 

[He's Dead, Jim]

Ressurecting the thread a bit. Superior drummer sound kinda weird when gainstaged to about -18db. The sound starts to suck, even with monitors adjusted to sound accordingly loud. I guess the problem is, that i adjusted all tracks to -18, while tracks like hihat and OH's should be even below -18db right? So the drumkit should be not just gainstaged but really balanced?

Set your relative volume levels of each kit piece to each other kit piece first.
Then turn the master volume in Superior down so it's hitting -18db. I usually use a VU meter (Slate VCC has one; there are several free ones in Reaper) with the equivalent of -18db being 0 dbVU because the needle goes more slowly than a standard RMS meter, so you have an easier time telling whether it's hitting where you want it. FWIW Superior should not sound the least bit qualitatively different when you have it at -18db, since the master should only be about 4 db quieter. The Superior mixer is usually set by default so many of the drums are already pretty quiet, so there isn't a lot of additional volume work needed.

I'm also lazy, so I have an aux drum bus where everything is routed for saturation and compression, and I use that as a volume fader in case I need to modify the kick level but I don't want to go all the way back into the superior master knob. Only problem is that then you have to re-adjust your drum bus compressor thresholds and whatnot.

 

[mva801]

Ressurecting the thread a bit. Superior drummer sound kinda weird when gainstaged to about -18db. The sound starts to suck, even with monitors adjusted to sound accordingly loud. I guess the problem is, that i adjusted all tracks to -18, while tracks like hihat and OH's should be even below -18db right? So the drumkit should be not just gainstaged but really balanced?

Don't overthink it. And are you sure you're setting your RMS levels around -18 and not the PEAK levels? Almost all DAW meters will show you the peak. If you're going by what you see in those meters, you should be peaking closer to -8 or -6. That will give you an RMS closer to -18. The RMS isn't measuring the full extent of the initial transient. Make sure you understand the difference.

Basically, with drums, just make sure your peaks are around -6 or so and that should be good.

 

[abt]

I always track hot without clipping. I don't care what anyone says, my interface sounds better this way. If you do this or someone send you files recorded this way, it's really important to use a trim plug first in the chain to bring the levels back down to where they should be. I'd love a dollar for ever screen grab I've seen where the all the faders are right down the bottom. They're not meant to be used this way. Use a trim plug with you faders set to zero.

Another problem, and I know others have also had this problem is that a lot of plugins are written with no gain staging in mind and don't work very well down there so you have to hit the input at a level where it sounds good.

 

[BearOnGuitar]

When people talk about gain staging to -18dBFS what they really mean is -18dBFS RMS, meaning average level which is different from peak level. In order to gain stage to an RMS or average level you will need a meter which is able to display average levels, like a VU meter. Usually a VU meter will calibrated so that 0VU equals -18dBFS RMS.

It's very important to understand the difference between peak meters and VU meters. Peak meters will show how loud a signal is while VU meters are designed to have slower movement, not reacting much on transients by design in order to closely resemble how we hear or feel music with our ears.

What gain staging to -18dBFS RMS does is that you will record at about the sweet spot that converters are designed at and still have 18db of headroom before you start clipping the transients or the signal, depending on the dynamic range of the signal (difference between average and peak level), meaning your actual peak levels will be hitting higher than -18dBFS but still leaving enough headroom.

In any case it's best to simultaneously use a peak meter to always check on transients, especially on very transient rich material like drums where you could go into clipping long before you even reach 0VU or -18dBFS, making sure no signal peaks higher than -6 dBFS. What it also does is that all your signals will be very consistent in their average level or felt volume from one to the next which is very useful. Further keep in mind most analog emulation plugins are designed to also sound best at around -18dBFS RMS.

If you want to go any further check out www.independentrecording.net/irn/resources/metergain/index.htm which is very detailed on meters and referencing levels.