Lesson 01: Compression

  • Thread starter Thread starter killthepixelnow
  • Start date Start date
killthepixelnow

killthepixelnow

Do it right or dont do it
Following the advice of one fellow here in the forum, I think that I should try to explain how a freaking compressor works with a simple language. Hope you find it interesting. I'm not a pro but I think this will give more light on this tricky subject. Let's start...



Lesson 01

Let me explain it with my own words. Compressor is a signal processor. You can find it in most recording software as well as rack units. A compressor reduces dynamic range. What does it do?

The compressor tames the highest peaks of your sound wave and allows the signal to be louder. Given that the sound waves are more even, your wave sounds more punchy and tight. overuse of compressor kill dynamics and that's the reason you must be careful and not over compress you tracks.

lesson01_basscomp.png


In these example you can see the track after and before being compressed. What you might notice first is that the weird peaks you fin in the original wave are less pronounced in the compressed track. Also notice that the overall gain (volume) of the new track is almost the same. How does it happen? Well, let's analyze my compressor setting.

lesson01_compressor.png


Threshold:
That's the point where you signal processor start to work. When the signal surpasses -18.0 dB it starts reducing the volume. How much?

Ratio
That's the reduction you're applying to the portion for the signal that past your threshold. I'm using a 3 to 1 reduction. It means that if I have 3db going past the threshold I'm gonna end with 1db. Any compression stronger than 20:1 is considered limiting.

Attack & release
To put in easy words that's thow fast or slow your compressor active and deactivate. In this example I've used a fast attack and faster release. You can experiment with these knobs. Depending on the kind of source your compressing, these setting could vary (metal kicks benefit from high attack and some cymbals sound better with slow releases)

Knee
It's called hard knee compression when the processor kicks in quickly and soft knee compression when it comes gradually as the signal rises. I've used a soft-knee compression to make it sound more natural.

Gain
So you reduced the overall level of your track. Now what? Well, you can apply some Gain to increase the volume until the compressed track matches the uncompressed track.

You can hear both examples in here:
Lesson 01



It would be nice of the experts to comment on this subject :D.
 
I think more emphasis needs to be put on the attack and release times, as these are often overlooked and seriously affect the sound.

Too many times I've seen people whack a compressor on something (not really knowing what' they're doing) with a really fast attack time which ends up cutting off all the transients - you have to think about how the settings 'shape' the waveform.
 
I agree with mattr. This is a great "primer" for people, but compression can SERIOUSLY make or break your mix if you don't know exactly what you're listening for and trying to achieve; example: presets!

NEVER use a preset for compression! There is no "one size fits all" preset. Period.
'Good' compression settings are completely determined by the audio being fed through it, the mix that surrounds that waveform, and how you want it all to sound. (same goes for every other 'effect' out there..)
 
minor point... limiting basicly is any ratio above 10... though i doubt most of us hear a difference between 10 and say 8... and anything requiring a setting of 20 is unfathomable by me...
 
d.bop said:
NEVER use a preset for compression! There is no "one size fits all" preset. Period
Click to expand...

I think presents can give you a general idea of the compressor to use. In example, a vocal compressor preset (like the one that comes with Pro Tools) has a low threshold, soft-knee and low ratio in order to have a natural transition between the original and the compressed part of the sound wave. An snare compressor preset will give you a fast attack and release, hard-knee and medium ratio. It's an start point and you must work and tweak that preset to make it suitable for your project.
 
I didn't hear much difference in yuor examples. Maybe the compressed one had sharper attack on the struck notes.

If you had an exam and asked for the difference between the two would I get any marks?
 
killthepixelnow said:
I think presents can give you a general idea of the compressor to use. In example, a vocal compressor preset (like the one that comes with Pro Tools) has a low threshold, soft-knee and low ratio in order to have a natural transition between the original and the compressed part of the sound wave. An snare compressor preset will give you a fast attack and release, hard-knee and medium ratio. It's an start point and you must work and tweak that preset to make it suitable for your project.
Click to expand...

That's very true, but I have a feeling there are a lot of people out there who just slap on a preset called "lead vocal" and leave it at that, then are wondering why their mixes aren't sounding right.

But yeah you're right, they certainly can point you in the right direction.
 
Monkey Allen said:
I didn't hear much difference in your examples. Maybe the compressed one had sharper attack on the struck notes.

If you had an exam and asked for the difference between the two would I get any marks?
Click to expand...

Well, so that's the point man, good compression allows you to tame the high peaks without altering the natural felling of your playing. As one guy said before, you can RUIN your music by overusing compression. The signal proccessor is there, working for you without being noticed by the rest of the people.

For this particular examples, first look at the compressed waves, they're kinda even. Now look at the uncompressed waves, they seem a bit whacky. Now try to listen it with good flat speakers or a pair of good headphones. Avoid by all means computer speakers... they tend to empathize certain frequencies. Most noticeable thing in the uncompressed bass line is a tendency to play some notes too much louder than the other. When a compressor is used these notes are tamed, giving your performance a more tight and solid sound.

PS: No grades yet :D
 
:cool:Yo KT +pixels:

THANKS for the clear and organized manner. Each time I tune in here, or record in my studio, I learn something.

Manuals are sometimes tough to wade through; however, "coherence" makes clear and "clear" learns something.

I dig all kinds of "101" lessons. A great asset.

Thanks,
Green Hornet:D
 
I'm not sure how long you are able to edit posts but you (and others) should add to make it a little more elaborate.

It is a good primer for the parameters of a compressor(which was your goal so good job!) but I think you it can develop a little more without losing the attention of a newb.

Hopefully I'm not just saying this so that I could get more from it but some things to possibly touch on...

When do you need compression?

How can you hear how it's processing the signal?

How to not use compression?

Just some ideas to make a good thread better! Possibly with the help of new moderation, we would be able to make this worthy of a sticky? Or maybe worthy of a sticky that will direct people to other newbie primer posts.

I'd answer my own questions but my take on compression may be fairly flawed as I stink with it!
 
mattr said:
I think more emphasis needs to be put on the attack and release times, as these are often overlooked and seriously affect the sound.

Too many times I've seen people whack a compressor on something (not really knowing what' they're doing) with a really fast attack time which ends up cutting off all the transients - you have to think about how the settings 'shape' the waveform.
Click to expand...
+1

+tenfiller
 
You asked for it

Compressor attack & release

So, having so good responses, I decided to give further explanation about attack and release settings. As mattr said, you can really whack up a track if you don't know how the compressor is affecting your signal. It's important to say that you don't want to mess up with dynamics, they're important in your music. Without dynamics, you'll end with a dull and boring track.

For this particular example I used three setting of compression for the same snare track. I'm pretty sure you're gonna hear the difference. Well, my source is an snare track; you can see it has a weak wave and a lot of headroom to use. I applied compression to raise up the level of the track and tame some undesirable peaks.

lesson01_compressor_snare.png





Slow attack & slow release

lesson01_compressor_slow_slow.png


Muffled, kinda damped. Snare is maybe the most important piece on your drum kit. You want it louder, and you want to bite. Slow attack causes your processor to enter in action too late. The slow release creates (to me, correct me if I'm wrong) some kind of "tail". If you have 16 notes going one after the other (like in metal) you're gonna mud your snare.



Fast attack & fast release

lesson01_compressor_fast_fast.png


I find this setting more suitable for my purposes. Fast attack and fast release is a good combination to enhance the attack and the snap. Maybe I'm hearing things but I swear I hear more bottom snare too.



Fast attack & slow release

lesson01_compressor_fast_slow.png


By setting the attack at 500.0 us (too fast) I'm killing the dynamics of my snare. Now it sounds tiny, weak and flat.



So, here are the audio files to understand better the images above.
Lesson 01 updated

PS: Any chance to make this post an sticky one?:D
 
You must log in or register to reply here.
Back
Top