Tip:
Highlight text to annotate it
X
Hello and welcome to Nebula Explained. This is a free preview section to promote the full Udemy course.
Please go to www.learndigitalaudio.com/nebula for full details on what packages are available
and how you can sign up right away.
This free section of the course is the first of 2 chapters on the very important subject
of gain staging. Today we are going to learn about:
What is gain staging? What are the basic concepts of gain staging?
Why is gain staging so important for Nebula?
We will learn how to set the ideal input level to Nebula, giving plenty of headroom, triggering
the dynamic layers and driving the saturation to taste.
So what is gain staging? It is simply optimizing the volume that audio passes through your
recording chain.
Anywhere in the recording chain where you can turn the volume up or down can be considered
a gain stage.
It is important to remember that in the digital world this includes volume levels inside your
DAW and it's plug-ins.
What are the Basic Concepts of Gain Staging?
In analog and digital recording you should record a loud enough signal to have a good
signal to noise ratio and enough headroom so there is no danger of distortion or clipping.
Headroom is the space between the current level of the audio and the maximum it can
be. Here is a diagram to illustrate the difference between no headroom and plenty of headroom.
We want a signal that is loud enough to not suffer from excess analog noise, or have to
be needlessly turned up in the DAW.
It must not be so loud so that when we sum many tracks together it clips the DAW.
Here is an example of a typical signal chain.
1st we have the microphone, this outputs a very low signal.
We need to boost this with a preamp. This line with an arrow represents the signal flow.
So the microphone signal comes into our preamp and enters its first gain stage. How much
gain should we use? This will directly effect its output level and in turn the input level
of our next stage, the sound card.
The next stage in our chain is our soundcard line in. We want to get a healthy signal into
it to minimize the background noise from our mic and preamp.
What would it sound like if our preamp gain was too low? Let's listen.
Here is my voice recorded at a good gain level.
Here is my voice recorded at a very low level and boosted back up again digitally inside
the DAW. You can hear all the annoying background noise from the analogue stages. Sounds really
bad!
We also have to consider that there are probably going to be many different tracks recorded
into our project that will be summed together to create our final mix.
Lets say we turned the gain right up on the preamp and recorded very loud, almost to the
point of distorting. The output signal comes into the sound card hot, but it doesn't clip
the converters. Everything seems okay.
What if we record everything at this level and sum it together? The result would be the
DAW output clips. The whole system is overloaded. We could go in and turn every track fader
down, but that would be a huge waste of time, and also cause additional problems for a plug-in
like Nebula that emulates analog equipment. Why is Gain Staging so Important For Nebula?
Gain staging is so important for Nebula as its dynamic programs behave exactly like the
analog hardware that they were sampled from. They react differently when fed with different
volumes of audio.
It is not just that we need to avoid overload or insufficient signal levels. There are several
key things we also need to understand when thinking about the volume of the audio we
send to Nebula: If input is too loud we will get a digital
overload. This results in very bad sounding artefacts.
Here is a copy of Nebula running the NAG tape emulation program by STN. Let's see what happens
when we overload it.
Watch the overload indicator here.
The sound breaks up before the overload indicator light comes on.
Eventually at +3dB input Nebula cuts off the sound completely.
It is obvious why we don't want to overload Nebula, but not so obvious why we equally
don't want to 'under load' it. Dynamic programs are different from other
types of program in that they have many sampled layers for different levels of input volume.
If the input is too quiet and it is a dynamic program, we may not trigger the sampled dynamic
layers. We will loose the 3D effect.
For example, a spring reverb may get a 'sploshier' sound the louder the input level.
Here is a diagram showing how Nebula's dynamic layers work.
The red dotted line shows how the envelope follower tracks the volume of the incoming
audio.
The Envelope follower has its own settings as to exactly how it does this and can be
found on the EVFS page.
Each number on the line represents where Nebula takes a reading of the volume. The interval
between each number represents the PROG RATE.
The PROG RATE determines how often the envelope follower takes a volume reading. The faster
the PROG RATE the more accurate the emulation. It can track the changes in the audio more
accurately.
If we start to follow the audio from the first sample we can see how each reading corresponds
to playing a different dynamic layer.
These dynamic layers store the sound of the original sampled hardware at many different
volume input levels. For example, if you play a louder sound into it then you will hear
exactly how the original hardware reacted to a louder input. This is the secret to Nebula's
amazing emulation.
For example sample 1 is relatively high and triggers dynamic layer 2. Sample 8 is very
low and triggers dynamic layer 6.
As the sound is played Nebula interpolates between the different layers. The way it does
this is determined by the SMOOTH algorithm used. This is show by the red line crossing
the dynamic layers.
The important thing to remember about this is to make sure your volume level is loud
enough to trigger the dynamic layers you WANT to hear.
Every program is different and must be learnt as if it were a new plug-in.
Usually the louder the input the layers with more distortion will be triggered, but too
low an input and there will be no interpolation between layers and no 3D sound
If the preset has distortion kernels, the louder the input the more distortion we will
hear. This is simply because we are triggering the layers with more distortion. Ideally we
want to listen carefully and alter this effect to our personal taste.
How much distortion is possible depends entirely on how the program was sampled and from what
hardware. Unfortunately, it is impossible for Nebula to reach extremely high level of
distortion because of technical restrictions. If we want that effect we will have to use
another plug-in.
Having said that, there are some programs that can reach high levels of distortion,
lets take a look at one of them, SCS Treble & Bass Amp EQ by Tim Pethrick.
This program is sampled from a vintage 60's Selmer guitar amp. It has an EQ control for
treble and bass along with input control and saturation level control.
The input control is the same as the GDRV (aka Gdrive) on other programs, it is the
same as turning up the input to Nebula at the same time as turning down the output.
The SAT control is the same as the one usually called DRIVE, it turns up the harmonics to
create more distortion.
Every Nebula program is sampled in a different way. What works well for this program may
not work well on another. Lets listen to how much distortion we can achieve with this one.
As we turn it on we hear a slight 'smile eq'. Here is the EQ plot with the treble and bass
controls set to zero.
So, leaving the additional EQ settings alone, lets listen to the uneffected audio, a mildly
distorted weak flat sounding softsynth.
Ok, it sounds terrible. Lets try and breath some Nebula magic into this lame loop.
Lets now listen to it using the SCS program with all its controls flat. Remember it has
a mild smile EQ built in so you will be hearing a little extra bass and treble.
Now lets add some saturation using the SAT control.
Lets flick between 0 and 30 a few times to easily hear the nice distortion we are starting
to add to the signal.
And now add some DRIVE
We have hit the program's limit and its starting to fall apart. Lets pull that back a bit.
That sound good there.
Now lets check the RMS with Nebula off so we can make a perfect volume match to hear
our results in context.
-20.9 on the left hand channel
So lets reduce the output a bit to compensate.
Check the RMS, almost spot on.
Lets flick between them to make our final judgement.
Sounds excellent!
Summary
Gain staging is simply optimizing the volume that audio passes through your recording chain.
It is vital for Nebula dynamic programs as they behave like analog hardware. They react differently when fed with different volumes of audio.
Too loud and we get artefacts, too low and we loose the 3D sound, in-between we control the amount of saturation.
We can use the GDRV or equivalent control when available to perform quick gain staging.
The DRIVE control increases the distortion, but the results do not reflect the original hardware in many programs.
EVERY Nebula program and sound source being processed is different and needs to have the
gain staging set by ear.
There is usually an input volume sweet spot where we are emulating the hardware being
driven at just the level the 3rd party developer intended us to use it at. Luckily for us many
developers have already given us a visual way to easily find this level using freely
available metering inside your DAW.
In the next section we will set up your DAW so that every track is at the same volume
level before it hits your track faders and insert plug-ins. This will allow all the headroom
Nebula needs to operate at its best and will automatically make your volume levels coming
into Nebula just where they want to be.
If you don't already have a gain staging system in place it will probably change the entire
way you use your DAW for the better.
The magic formula, which may seem meaningless now, is:
In the next section we will find out exactly what this means and how to set up your ultimate
gain staging system for Nebula and your entire DAW based studio!