Compression Ear Training

[SimonT]

Hi All!

I was wondering if there's a way for me to train my ears as regards 'dynamic range compression'?

I hear the compressor working and can tell when whatever's being compressed sounds bad but just don't know what I'm doing with it or what I should be listening for or achieve with it.

I suppose it depends what I'm compressing, synth, piano, guitar, vocal, right?

I want to know if there's a way to know how to use one to get the most of it, and how to train my ears to know? And yes, I know how the compressor works. Your signal goes in, and you have an Input Gain to decrease or increase that if desired, then a Threshold knob, once it exceeds the threshold, it is then compressed by whatever Ratio you have set. If it's 4:1, every 4db over the threshold it goes it only let's 1 through, so it reduces it by 3. Attack and release I understand too but not sure if again, I hear what they're doing, and then you have Output Gain or Make-Up Gain where you re-boost the signal that's leaving the compressor, to increase the volume back to what it was before you compressed it. Some have a soft knee too which I gather allows heavy compression to be applied a bit softer so it isn't like a like a sudden blast of compression that hit's you, it applies it a bit softer or the compression a bit more gradually.

So, to sum up, what should I be listening for? this I suppose depends on what instrument or sound you're compressing again, so maybe we should narrow it down here to 3 elements...Kick Drum, Snare and Vocal.

If someone could explain (or link me to a vid) that could help me hear what it is I should be listening for (if that's possible) when I apply compression to these 3 elements.

Thanks!

 

[DarkRed]

Hi All!

I was wondering if there's a way for me to train my ears as regards 'dynamic range compression'?

I hear the compressor working and can tell when whatever's being compressed sounds bad but just don't know what I'm doing with it or what I should be listening for or achieve with it.

I suppose it depends what I'm compressing, synth, piano, guitar, vocal, right?

I want to know if there's a way to know how to use one to get the most of it, and how to train my ears to know? And yes, I know how the compressor works. Your signal goes in, and you have an Input Gain to decrease or increase that if desired, then a Threshold knob, once it exceeds the threshold, it is then compressed by whatever Ratio you have set. If it's 4:1, every 4db over the threshold it goes it only let's 1 through, so it reduces it by 3. Attack and release I understand too but not sure if again, I hear what they're doing, and then you have Output Gain or Make-Up Gain where you re-boost the signal that's leaving the compressor, to increase the volume back to what it was before you compressed it. Some have a soft knee too which I gather allows heavy compression to be applied a bit softer so it isn't like a like a sudden blast of compression that hit's you, it applies it a bit softer or the compression a bit more gradually.

So, to sum up, what should I be listening for? this I suppose depends on what instrument or sound you're compressing again, so maybe we should narrow it down here to 3 elements...Kick Drum, Snare and Vocal.

If someone could explain (or link me to a vid) that could help me hear what it is I should be listening for (if that's possible) when I apply compression to these 3 elements.

Thanks!

In regards to ratio you can also think of it like if the ratio is 3:1 the threshold is at -12db and the signal is peaking at -4 dB, like -12 + (8 / 3) = -9.33 dB (peak at -9.33 instead of at -4 dB, the signal reduction is hence the difference of the peaks, -5.33 dB), but it is the same thing, only that it targets the peak perspective primarily and the reduction perspective secondarily.

Anyway, now to your question of what to listen for etc. First of all notice that various compressors have various behaviors depending on how they were designed, there are for instance compressors with a dynamic threshold. Some have rms mode instead of only peak. Some can be set into expand mode as well. Some provide you the option for a knee setting, some are just applying a knee curve dynamically depending on the amount of signal exceeding the threshold so that for instance the more of a "target" the signal is, meaning the more of the signal exceeding the threshold, the softer the curve. Some give you the option to control the hold, some don't. And so on. All of this is up to the designer of the compressor to decide and how much of that character should be possible to change and to what extent. So all in all, a compressor is really a tool to shape the timbre of the signal - attack, decay, to provide the engineer the option to lower the peak to rms ratio as well as to raise the rms level of the signal with the make up gain.

Now, in terms of knowing what is an optimal use of a compressor one has to go back to what mix qualities you want to achieve. Various compressor applications and behaviors have different impacts and hence will move you more closely or farther away from your mix quality goals. The bottom line is hence, be much more specific about what you want to achieve relative to what compressor and setting is optimal. What this means is that it is the engineer's focus, intention and taste that is of importance. Then when you are very skilled at describing what kind of characteristics you want in order to end up with what specific mix qualities, only then is it worth start looking for a matching compressor, with the exception that many times you are no sure and then it is better to try out various compressors just to get the perspective on what various compressors can do to the signal. Having said this, it is maybe more important to have a set of various types of compressors available to choose from, so that you have a wide array of compression options at your finger tips. So what you should do is to get various types of compressors and group them. Software, hardware, tube, opto, electro... Then when it is time to find a good match, you can speed up the process of finding the right compressor, by first selecting in what groups of compressors you can find a good match relative to what you want to achieve. This also makes you able to narrow down the configurations so that they are a bit more compressor type dependent, since various types of compressors will behave and hence sound differently at the same settings.

Use your ears, focus on the routing a lot, use side chain compression, parallel compression and two stage compression to your advantage. Stack compressors when necessary, route various sound sources through various types of compressors to differentiate and hence enrich the dynamics of the mix. Separate when hard compression is adding to as oppose to removing from the mix. Hard compression can for instance work very well on content that already has a low peak to rms ratio, while it can destroy a mix that is very dynamic. Many mixes out there are destroyed by compressors during mastering, because so much mix attack is cutoff so instantly and by so much by the brickwall peak limiters. Then there are compressors like the distressor that you can push very hard on such type of content and you might still like what the hard compression does. Then there are compressors like the 1176 that just tends to do what you want it to do in many cases. So it is compressor, signal and goal dependent, just dive into it and expand your perspectives.

 

[Mad Phonetics]

In regards to ratio you can also think of it like if the ratio is 3:1 the threshold is at -12db and the signal is peaking at -4 dB, like -12 + (8 / 3) = -9.33 dB (peak at -9.33 instead of at -4 dB, the signal reduction is hence the difference of the peaks, -5.33 dB), but it is the same thing, only that it targets the peak perspective primarily and the reduction perspective secondarily.

If it's 4:1, every 4db over the threshold it goes it only let's 1 through, so it reduces it by 3.

This is only true if you take the time parameters out of the equation by setting them virtually to 0ms.

 

[DarkRed]

This is only true if you take the time parameters out of the equation by setting them virtually to 0ms.

Hmm, interesting, need to test this in my DAW, but you have probably done so, I assume you are correct.

Anyway, an easy way to understand the compression curve of a particular compressor is to just run a signal generator at say a peak of -10 dB, then split the waveform and apply a gain boost to say a peak of -4 dB, then apply the compressor on it with the threshold at -10 dB and then print the result to a track for comparison.

Anyway, I sometimes think of the attack as a kick drum beater in slow motion where the time it takes for the beater to touch the skin is the attack time, at the point it has touched the skin it has reached 100% signal reduction. The hold is how long the beater is against the skin and the release is the time it takes for the beater to return back to its rest/original state. I have seen some graphs where it is shown that the release kicks in when the threshold is exceeded again, I am not sure that is correct, logically the release should happen as soon as the attack and hold are done whatever the signal level is at that point, but I could be wrong the best is to simply just find out in the way I mentioned.

Want to add that an important part of succeeding with compression is in your monitoring quality as well as in the freshness/tiredness of your ears. If you have speakers or cans that compress a lot and you sit mixing for hours, then slowly your compression quality is going to not become so great. But great speakers that do not compress but that reveil the raw sound, in conjuction with fresh ears, that is gold when you apply compression. This is one of the reasons why it pays off to mix into compression on not too loud monitoring volume and why it can pay off to break down your entire mixing and mastering processes to smaller shorter sessions with breaks in between so that the ears can de-compress. Now, your creative flow is more important, therefore don't be so strict about this, but when your creative flow is not there giving you the kinds of improvements you want, thats a good moment to de-compress your ears by taking a break.

For speakers I recommend ATC SCM150ASL Pro, for cans I recommend Audeze LCD-2.