Within FL when starting a mix after arranging (mostly hiphop/rnb), I generally adjust levels and EQ to give each instrument room to sit in the mix. However I was told that gain staging isn't necessary in the digital age and was useful for when analog was prominent. How true is this and if Gain staging is still necessary or helpful, how would one approach it to procure headroom for the mastering phase?

I used to gain stage like this guy

and it seemed to create headroom in the mastering phase but since music is subjective and there are no hard rules, I feel this tutorial is not the most accurate representation of gain staging.

Wouldn't bringing every instrument to -12db create new issues and resetting the master fader create the "illusion" of headroom. I've had mixes where i applied this technique and there was still clicks and pops after a WAV export even though the master channel didn't exceed 0db.

 

Since I'm working digital I never used 0VU gain staging, it just wasn't necessary.
The important thing is not to clip any channel (with 32bit FP impossible in practise) or plugin in- or output.

Why that hero didn't want to hit any channel higher than -12dB is a mystery to me. Maybe because he knew that with this amount of channels he will end up at around -5dB in the main out?

What I am doing is peak gain staging to a) get an impression of what a plugin is doing level-wise and b) to avoid clipping in plugins which still work with 32 (or even 16)bit (no FP).

 

This is very true. Or "only true".

It get's more nessecary however if you stack tons of magic noise floor plugins. But then you can use the dolby noise reduction trick to get rid of the noise again if you like.

No Avenger stated everything correctly.

The reason why this myth exists by the way is because if you have all your tracks at a high level you are mixing very loud. And then your judging what sound good and what not get's bad. But if you drive everything loud and turn the master fader very low you'll get the same result as if you manage all of your tracks loudness (in 24/32 bit).

 

Last edited: Aug 1, 2020

Some plugins are calibrated to -18dB

 

My reason for gain staging these days is a calibrated reference volume, aiming for a mixdown at -23LUFS.

 

Last edited: Aug 2, 2020

This is the correct answer.

Gain staging to a lower level does matter if you are using analog emulated plugins in your mix. They are often calibrated to perform better at lower volumes & this is why 0VU (at -18db) is a figure you often hear suggested as a reference.

Simply avoiding digital clipping in your DAW isn't enough.

 

So far I haven't come across any plugin where this was really necessary. I already gave in another thread an example where I could send a source with 10dB more or less into a plugin without affecting the sound and it was explicitly mentionend that this plugin was calibrated to 0VU = -18dB FS.

 

I use a lot of Analog Modeled plugins so I let my peak volume rest around -10 to -7dbs.

This usually gives me enough headroom to get a very leveled, competitive-mastering sound, without introducing that digital distorted sound if you know what I'm talking about. Drake had a lot of "it" on Views From The 6.

With that being said, I do use Waves NLS Buss & Channel on my Mixes. So Gain-Staging has been a necessary effort w/ that Plugin, but I think its for the better considering it taught me how to view the Mixing Engineers mind in practicality.

 

A trick that has been working for me as a starting point is:
-set all my channels gain to -10
-leave all my buses and sends at 0
-and set the mix bus gain to +6

Then as you produce/mix try to keep the mix bus peaking at 0.
So when you're ready to master, remove the +6 and you should be perfectly balanced to around -6 headroom.

 

1. try to find stems for songs or construction kits
2. mute everything
3. unmute each instrument and look at the meter of each track and the master
4. pay close attention to the drums and their levels
5. take notes and learn how to balance your levels and dont overthink it

 

even those with no output or input? doubt it. some plugins have a sweet pot

but on the topic, i guess it depends on your set up. these days i'm using my speakers clipping led to balance levels. blend the kick and bass so it knocks just before it clips. then do everything else around those levels

strangely enough. i found that the louder my mixes are the lower the actual levels were. before i realized this is would think an 808 hitting at -20db was too quiet, i had to be doing something wrong but the issue was simply experience. different 808 different pocket. like all other sounds

 

Use Hornet VU MK4 or LU MK2 for gainstaging. They are best and they are very cheap.

Now,
why did I mention a particular VU meter or LU Meter? Why not any other?
Good question.
Watch one or two videos about these two beasts and you will get your answer yourself.

Again, people who don't 'BELIEVE IN GAINSTAGING' or don't understand the -12dBFS I am pretty sure, you people don't use or have never used true analog modeled plugins because true analog modeled plugins have a sweet spot around -18dBFS RMS which is 0 dBVU [ -18 dBFS peak is not 0 dBVU technically. -18 dBFS RMS is 0 dBVU. VU meter always told us about the average which is close to the RMS] because actual analog gears are calibrated that way and true Analog Modeled plugins emulate each and every aspect of that particular hardware.

0 dBVU = -18dBRMS which is -18dBFS RMS
0 dBVU ≠ -18dBFS Peak

Avoid clipping at Master channel is not the ONLY reason to why people gainstage.

 

Last edited: Nov 27, 2020

I always wonder, whether it's true or not, is this a good behavior of a plugin or a bad behavior? Because it always seems to me, that the people who claim those things are defending this behavior for some (maybe conservative) reason.

 

Last edited: Aug 4, 2020

The point is, in 32 bit float operations it actually does not matter when making design choices as developer.
to take Acustica Audio as an example, non of the actual devs did ever specify a sweetspot as far as I can tell.

could you back that up with some proof? I'm sure that would be mentioned in some sort of manual?

I've made mistake often putting too much effort in proper gain staging, killing time and making songs sound dull.
I just really can't be bothered to do so anymore.

 

Last edited: Aug 4, 2020

Why is it good or bad?
early hardwares were made that way. Everything was calibrated to 0dBVU. Why would it be good or bad? No one is defending anything.
LA2A or LA3A or Fairchilds or 1176s or API EQ or ShadowHills Compressor or Neve consoles or any outboard gears are actually calibrated that way.

 

Channels inside DAW work at 32 bit or even 64 bit floating point. Not all plugins.
To get a proof do this.
take 1 channel. insert 20 gain plugins. boost +30 or +50 dB on 10 plugins(or whatever the max value is on the gain or utility plugin). Say you boosted +30 dB on 10 plugins each. Now, reduce 30 dB in the next plugins each.
Now as DAWs and many plugins work on floating point bit, they will not distort. you will get the same signal as the Input, clean.

Now insert any UAD or Analog modeled Plugin ( which will not do anything. So it will stay idle, but active) anywhere in this chain. I am sure as hell, that you will distort.
But, if you insert any plugin that also works in a floating point bit, like the gain plugin with no gain or boost or many Digital plugins (no analog modeling) then you will get the clean signal again.

Sweet Spot as in, the gain, after which the plugin starts to saturate (like the Analog gear).
You don't need to read anything.
These proofs are DIY.

DO IT YOURSELF.



32 bit floating point gives you close to 1500dB Dynamic Range.
64 bit floating point Dynamic range is close to 12318 dB.

[6 dB per Bit is true for Fixed Point Bit (like 8 bit, 12 bit, 16 bit, 24 bit)]

 

I've spoken about the plugins. Not the hardware.
So, to specify my question. Is it technically necessary for a plugin, to emulate those kind of things?
Or, does it anything good? And if not, why is it done this way then?

 

It is the principle. If you are emulating something, you should try your best to emulate it 100%, otherwise what's the point of emulation?
When you don't have some grands to buy the hardware but want to use an emulation that will be as identical as possible to the actual hardware, you will always want something that is really really close to Hardware.

Here you want to remove that behaviour ???!!!
Then don't use plugins that are analog emulations. That way you will never need to care about any gain-staging. Go bananas on Gain staging per channel.
And it is technically necessary. Those things make what an Analog hardware is and therefore, should exist in its software emulation.

Why is this thing so hard to understand?

 

Please not again! Use the search function. We had already discussed this topic and its aspects several times.

 

which plugins which u use have it explicit stated in the manual ?
pls state them.

which ones ?