earlier this year there has been a little chat about dynamics processing, so it seemed like a good topic to explore in a blog post.
First we’ll tackle what it actually is, then go into what dynamics control Syntә has and why.
what it actually is
Ok, so we’re all familiar with how to change the level, volume, gain, loudness (I don’t know why it has so many names) of audio. You reach for a knob or fader, rotate or move it and hey presto!
Now, sometimes the sound is too quiet or too loud - so what do we do? Adjust the volume 😀 Simples.
Congratulations - you’ve just done some dynamics control :)
but
But what about all those boxes and plugins and compressors and limiters, expanders, etc?
Oh ok, so let’s talk about the human version. We said sometimes the sound is too quiet or too loud - how do we know that - use our ears of course hehe.
It goes something like this:
sound -> ears -> change volume
Now, if we wanted a machine to do this for us it would need to have ears. Hmm tricky. Or maybe it can have another way of sensing how loud something is?
And - let’s come up with a daft name that makes no sense, lets call it a sidechain. No not sidechick, no not sidedick, no not bikechain - sidechain.
I guess if we need to confuse people further we could also call it an envelope follower. No, thats not a pet that chases after letters in the post ;)
Then, I suppose we need a way to change the volume. Lets call this an amplifier even though it’s usually turning the sound down. And it would be fun to call it voltage controlled because everything is done digitally these days.
So far, so bad.
components
Lets take a look at how these things work.
Imagine a waveform. This is a squiggle that is like a sound drawn on a piece of paper, or usually shown on a screen.
How loud is it? Let’s guess that when it’s bigger its louder. Ignoring all the wierd ways we perceive loudness for now.
So how do we measure how big it is? Chop it in half of course!
Wait, what? How does that make any sense?
Let’s slice it along the middle like a banana split, or a hotdog bun, or the reflection of some mountains in a lake (split at the horizon).
This hasn’t really done much but it has made it easier for the machine. Cos it can look at how high the ‘mountains’ are and that’s how loud it is.
Otherwise it would get confused squiggling up and down with the waveform.
now is a good time to look at things from a different angle
We’re familar with distortion in sound? makes things very crunchy and often very loud too!
But what if you wanted to make something loud without all the crunchiness? Hmm, is that even possible?
Maybe we should try to understand where the crunchiness comes from.
When we turn the sound up, it will hit some kind of limit. Usually this is called headroom - at last a term that makes some kind of sense?
When it smashes against this, it slices the top off the mountain range, which really messes with the shape of the sound.
What about if… instead of smashing against that limit, we try to gracefully aim towards it, more gently and smoothly?
Well, we could use our machine for changing the loudness of something in response to how loud it is?
Because it’s a machine it can do things very fast, so it could probably respond in enough time while still boosting things up.
If we respond very slowly we’ve just turned everything down and that won’t make things LOUD.
do we need things to be loud?
I dunno, its up to you, there are no rules.
but we’re getting a feel for what the thing called dynamics control is. Its called that because we’re dynamically controlling how loud something is.
but
Surely there must be a way of crowbarring in more weird terminology?
Lets call how fast the machine responds the ‘attack’. ‘cos that makes sooo much sense.
and how quickly it goes back to how it was before we can call the decay. I mean, if you squint, that one kinda makes a bit of sense?
Lets also throw in some other nonsense like RMS and peak detection blah blah blah.
now
now, there is a saying that if something walks like a duck and quacks like a duck then it is a duck. I dunno anything about who says that or why, but lets prove them wrong by calling dynamics control ‘ducking’.
Wait, what??
Yes, get over it.
Ok, so when the compression (the volume being turned down) happens, lets say it makes it duck. Quack.
BUT, morpheus voice what if I was to tell you that we could make something else turn down instead… Eh?
That would mean that one sound has ‘ducked’ another sound.
So the ’ears’ are listening to one sound, then travelling via the ‘sidechain’ to ‘duck’ another sound. Weird.
But it does make quite pumping dance music. 🦆
Let’s talk about Syntә.
one of the things I wanted was to do any old craziness like unlimited feedback and get away with it.
Which has happened, though more by luck than chance. oops.
So I knew I wanted some kind of automatic dynamics control in case things got too loud.
However, that is going to be the subject of another (possibly much longer) post :)