Lesson 53

Compression

The 2 most powerful tools you have as a mix engineer are EQ and compression. Get them right and your mixes will sound great. Get them wrong, and you can guess how the mix will sound. But if music is an art, how can there be a right and a wrong? 
Because music isn’t just an art. It’s also a science. It’s a combination of waves along a time grid. When combining these waves, there’s certain things you can do that will make them combine smoother, not clash with each other, and just sound more natural.

A compressor reduces the dynamic range of an audio signal. It makes the quiet parts louder, and the loud parts quieter. 
Have you ever watched a movie... and had to turn up the volume to hear the dialog, but then turn the volume down when there’s a loud part? If you have done that... then you were being a compressor. Thankfully, us audio geeks have this automatic tool that does this for us... we just have to dial it in.

When recording a song, the natural volume fluctuations of our voice, or instrument playing, are much to extreme to sound good in a mix, or sound good when amplified. Compression is almost always needed, on every track. But, too much comprehension will take away the feel of the song... so there’s a delicate balance between keeping the life in the sound, and getting the best volume level.
Let’s take this drum beat for example as I use drum sticks on this table. The feel comes from the dynamic changes. If I compress it so they are all the same volume, it sounds like this. There’s no life left.
It sounds great without any compression because it is the only thing you hear. Let’s add some simple piano chords to it. Now, without compression, the quiet hits are lost, they’re not heard, and the harder hits seem too loud for a nice balance. So, let’s put just a bit of compression on this drum part. There we go... now you can hear the quiet parts, and the loud parts aren’t harsh, and it still has its life. If I overcompress it.. it’s louder, you can hear it more clearly, but the life is gone.
You need to be careful of ear fatigue and poor room acoustics when using compression.
Ear fatigue will cause you to not notice the life being lost little by little.. as you compress more and more because it keeps getting louder, and therefore, fooling us to think it sounds better. If you do an A/B comparison of 2 identical sounds... if one is louder, it will always sound better. I do a lot of microphone testing, and its imperative that I match the volumes perfectly, because an inferior microphone can beat a better microphone if it is just a little bit louder when listening back and forth. So, be aware of that as you keep dialing the compressor up more and more. Trust your ears... only a little bit. Also trust your meters. If the vocals sound good, but you see the gain reduction meter slamming like -20db of gain.... you might be over-compressing and not hearing it. Most of the time, over compressing will sound like a pumping, or an unnatural feeling. Better quality compressors are capable of providing more compression before it is noticeable, and the attack release settings also can help get a more natural sound. The attack release settings can also be used artistically to shape the sound.
First let’s look at the logic compressor. This is a pretty standard layout. You have your threshold, ratio, attack, release, and make-up gain. All of these controls are important to sculpt to sound perfectly, but the most important 2 controls are the threshold and the ratio. 

The threshold is at what volume level the compressor kicks in. The ratio is how much will it compress.
So if you set your threshold to -20, then any sound that goes over -20 will be compressed. 
If the ratio is set to 2:1 then for every 2 dB of volume over the threshold, it only outputs 1 dB.
I’ll take this same example a bit further, with the compressor set to a threshold of -20, and a ratio of 2:1. If the input volume is -30 dB, the output is -30db.
If the input is -20, the output will be -20. If the input is -10, the output will be -15. Keep in mind, these numbers are negative, so -20 is louder than -30. Any sound that is -20 or quieter is not affected, but any sound louder than -20 is reduced in volume by half of the amount that’s over -20. 
If the ratio is set to 4:1, then the volume is reduced to a quarter of the input, above -20. So in this case, if the input signal is -10, the output signal will be -17.5. If the input is 0 dB, the output will be -15.
If the ratio is set to 10:1, then an input signal of -10 will have an output of -19, and an input signal of 0 dB will have an output of -18db.
I know this sounds confusing, but re watch this explanation.. it’s hard to explain, but it does make sense, and you need to understand this to use a compressor properly.
The attack release settings control how quickly the compressor responds to the changes in volume. On percussive parts you’ll probably want faster settings, and on instior vocals, slower settings tend to sound more natural. You don’t want to hear the compressor pumping in and out.

Now, let’s look at this waves compressor 

Next let’s look at this drawmer compressor. 


And last, the uad 4710d. I love this compressor because of it’s simplicity, and the fact that it’s integrated right into the preamp. The compressor itself is at a fixed threshold, and a fixed ratio. You control the amount of compression by raising or lowering the input gain. Raising the gain sends a louder signal to the compressor, and therefore will result in more compression. And the output gain controls the volume of the output. Pretty straight forward. It has a fixed release, and 2 options for attack, fast and slow. Fast works better for drums and I like it better on acoustic instruments, slow works better on vocals. It’s simple, and sounds great.