adjust the gain of your audio to the appropriate level(s) before processing, using either clip gain or a gain / trim plugin. a good VU meter is important. mvmeter by tb pro audio is a great freebie. i also like the klanghelm vu meter because it has a gain adjustment attached to the meter (one less plugin to open).

once you begin processing, use the input and output controls on each plugin to control the level of the audio from plugin to plugin

once you've finished setting up a signal chain, you can use your DAW fader to control the overall volume of the processed track. if you use the gain plugin from the beginning of your chain it will just mess up the levels across your plugins and alter the sound you've worked hard to dial in. if you prefer to leave your DAW faders at unity gain (or 0), another option is to put a gain plugin on the end of your signal chain and adjust that one as you work. some like to use this trick when programming automation

 

an insert gain plugin first in the chain is not going to change your plugin settings at all. unless you don't insert it from the start of work on that channel. in the Logic inspector you can set +/- gain per region without loading the gain plugin first, like you said Nodoz; to keep faders closer to unity or dealing with really different source volumes.

but generally, you can use them anywhere you want if you just pay attention to your other settings.

 

Vu meters have a fairly slow attack-release time, so it makes sense quick transients won't register. But i was talking about clipping the bus input in FS. If you use -18dBfs as a target for individual channels you have less chances of reaching 0dBfs when adding lots of them while still having the faders in a confortable possition. Now i've seen that work with a target of -12dB, guess it's a matter of taste.

More than a myth i think it's a misunderstanding. -18dBfs=0dBVU when measured with a continuous sinewave, but dBVU and dBFS beheave differently in terms of speed, and i've never seen any manufacturer claim that you need to keep everything near -18dBfs or 0dBVU for the plugin to sound as it should, they just state the way the plugin is calibrated, not where the "sweetspot" is, because in pratice it will differ from source to source. Some manuals (like those of some nebula libraries) might go a bit more specific and give you numbers in dBfs, which is nice. The point is, manuals are helpful if you stick to what they say, and not to what they don't say, otherwise you get these myths.

All that said, I think a target level of -18dBfs for gainstaging works just fine for most applications, you might not need -18dBfs, but is a nice "home" level in case you need to change plugins in a chain, use outbord gear without too much hassle and keep things from clipping later on.

 

I was told the staging was setting the 'stage' correctly for every track at the very beginning, and after that was gain balancing. Seems some people in person and online have slightly differing descriptions. Good to know.

 

It's good to know that one or two words difference in a forum post can lead to it being even wildly inaccurate, that you would probably not even notice in a conversation. just like any forum thread will probably contain some wildly inaccurate information either way

let me ask this then, in reference to getting everything "correct" from the start. Let's say, with an electronic music project with a large number of channels. maybe 20 or more. What would be the correct way of gain staging this "correctly", particularly where samples are being used. (meaning just that any possible sound in the world may be used in any eventual channel). Would it be advisable to start with a certain amount of channels as a group right away? At which point, any common instrument types could then just be organized into those groups? with those groups pre-staged to some approximate for that generic source type

I'm interested in this thread because I have seen many about it, specifically in regard to logic. (but then I'm not looking at cubase forums)

 

Great question and I do not know the exact answer. But with your statements, by default I'd ensure the signals emanating from your VST and/or samples in every channel were not distorting and sitting comfortably under 0db with headroom, wouldn't you? While each DAW is different, you can lower the gain output in each channel by gain itself or wave edit/gain on samples if no built in gain staging is present in the DAW. From being caught with a track that distorted in one track while recording without noticing, that taught me to check every file in case I missed something. .

I do not know how anyone else would do it, but I have recently done a track with 12 recorded drum tracks, 10 percussion tracks, bass, keys, saxophone, brass section tracks (8) stereo keys and stereo mallets. Near enough to 40 tracks and I checked each one individually. I'd use the same principle even if they were samples and VST outputs., but that's my take on it. Sure you Group them, but I was taught checking each was important, not just the group.I may have been taught differently, or in some minds, incorrectly, but this has always worked for me. I missed it once and decided never to again and gain stage right at the very beginning before I add any plugin during and after recording, or using samples/vst's. I also have a tendency to render the VST's to 2x mono's or a single stereo file so I work exclusively with wav files at a certain point.

 

Last edited: Jun 20, 2021

Chirp artifact on Coral Bax if you got any higher than -3dBFS. Also, the input volume on many saturation plugins determines how much harmonics are added. Gain staging to -18dBFS(actually 0dBVU) will usually give a more subtle/clean saturation. It really makes a difference in a mix where all this stuff adds up. Second pic is going into Kelvin at -18dBFS and adding 18dBFS of makeup gain. The third is going in at 0dBFS. The difference is definitely noticeable. It's honestly just a safe bet to go into any analog modeled or saturation plugin at -18dBFS(0dBVU) and using a drive or input gain knob to saturate to taste. I like to use the Hornet VU meter Mk4 for auto gaining to 0dBVU to +3dBVU(depending on how much saturation I want) into Acustica preamps or saturation plugins. Then I balance the volume of my tracks post FX. It adds more depth to the mix.

 

Last edited: Jun 20, 2021

100%? I'm not even 10% on gain-staging. Does anyone have any decent video recommendations?

 

This should help - its not a video but its a quick read (it explained a lot to me), and its by Izotope so they should know what they're talking about

https://www.izotope.com/en/learn/gain-staging-what-it-is-and-how-to-do-it.html

 

Yes, very helpful. I think I'd kind of grasped most of that intuitively, just through playing around with the software. Although I own quite a lot of hardware as well, I've never really bothered recording any of it, so I'd never really looked into what I needed to do to do it properly in the past.

I'm glad I screw around with synths and not with guitars. That whole area sounds like a complete minefield to me.

 

Very interesting topic
All I can say is that since I've implemented gain staging my mixes became much more coherent and better sounding. Especially when using loops and samples with volumes all over the place. Using HoRNet VU Meter MK4 for auto gain staging by the way

 

OK:

 

The subs below 100 cycles are off the wall. The entire spectrum from around 100hz to 20Khz is almost like a 45 degree shelf.
I gather this is EDM? It does explain the radical curve (gain levelling curve probably in this piccy) but it needs eq by the looks of it above in places based on the spectrum. Auto A,I? I'd go back to scratch if I had a spectrum and gain structure that looked like that.
I am looking at the 20hz randomly jumping on the far left and picturing a batch of people in front of sub-woofers throwing up or passing out.

 

Last edited: Jun 20, 2021

Keep it simple.

In FL, I generally use the channel volume knob to reduce the sound to generally make it fit better in its context, and while layering. Then I route each sound to a mixer whose faders I only touch after I have processed the sound and again after I apply processing on the buss, for sounds like chords and leads. It can have one more use, i.e while mixing the entire song. But in that case I mostly need to adjust buss output volume, since individual sounds rarely are a problem if mixed properly before.

Here comes Fruity balance, which I place at the top of master channel and bottom of the buss or individual sounds sometimes. I use it only and only for automations. I use only stock plugins and fabfilter mostly so whatever technical stuff related to analogue plugins is not relevant in my case.

If the mixdown is too loud (which is usually), I just reduce output volume of the master eq so that i gey around 6-8db headroom before i slap ozone

 

Last edited: Jun 20, 2021

Subs? Which subs? You mean the fidgeting below -120dBFS? Nobody cares about that. It's a 100Hz sine whose level I raise in 2dB steps and drive it into Softube Tape. Nothing more.

 

Yea Kelvin has an input knob so you can adjust how hard you drive the saturation with the input knob or even the drive knob. Either way you're still going to be gain staging inside the plugin if you want subtle saturation. Some saturation plugins don't have an input knob so you will have to gain stage going in and out them with a volume plugin. Saturation Knob from Softube is a good example of this. I've included 2 pics where I turn the saturation drive knob up just enough to kick in the saturation. You will see that the first harmonic is 30db louder when the input gain is at 0dbfs compared to the one that's going into it at -18dbfs. There's also more aliasing. When you're doing this with a bunch of tracks it will accumulate and may not be desirable. That's why I say it's "SAFE" to always go into an analog emulation/saturation plugin at 0dbvu. You can always adjust the saturation to taste from there using the drive knob, pretty much every saturation plugin has a drive knob or input knob. Most people love Acustica stuff because of the added depth which is achieved by the subtle saturation. Test any Acustica Pre and you will see that the saturation is very low. You can only get that by gain staging properly if you are using many other plugins. I have tested A LOT of them and this seems to always be the case.


Yea the chirp artifact isn't audible in your example because the bass sound has no high frequencies. It's actually kinda hard to drive Bax hard enough to hear the chirp. You'd have to send like a drum loop or something with higher frequencies in at like +5 to +7 dBFS to hear it which I've seen EDM guys do before. I've seen tutorials where people are clipping the shit out of their plugins and lowering the channel fader afterwards lol. Gain staging at different input gains is important depending on what effect you're going for with the saturation. 0dbvu is almost always the best starting point. That's my whole point. Anyway, you asked for one example. If you think the Saturation knob example isn't good enough I don't know what to tell you. We'll have to just agree to disagree lol.

 

Last edited: Jun 20, 2021