Understanding a compressor

A

anthonyg123

New member
This is my first time ever using a forum so I apologize for any mistakes and/or stupid questions I have, feel free to poke fun to the point I understand what I am being stupid about because I am open to all criticism or advice. When using a compressor, how much gain reduction is common or appropriate for an instrument with a lot of dynamic range such as bass? I'm not looking for a number because I'm aware every knob should always be set individually, but I'm looking for your personal opinions on what you've discovered so I can understand what I am doing or doing wrong. Do I want to see the gain reduction level hitting the same db level? Is it okay if it fluctuates between different dbs or am I trying to stop this by using a different ratio? I know this is a lot to answer but any help at all is much appreciated. It is frustrating after monthes of still not understanding what is helping or hurting my signal.

Edit: I attached two photos of the same note with a ratio of 2:1 and 3:1. They both have a gain reduction 3 that does not change. I may be wrong but isn't the ratio supposed to change the amount of gain reduction?
 

Attachments

  • Screen Shot 2015-03-16 at 2.10.37 PM.png
    Screen Shot 2015-03-16 at 2.10.37 PM.png
    714.9 KB · Views: 352
  • yup.png
    yup.png
    738.2 KB · Views: 346
Last edited:
Haha I agree this is super important! But I'm really looking for someone experienced to better explain what to look for so that I can experiment with a better understanding instead of continuously changing my settings because I don't think its working correctly.
 
The best thing to do is to play with it, whether it be a hardware compressor or plug in. Fiddle with every knob, setting etc. Take it to extremes, play with it subtly, but play with it on one source that you are familiar with. This applies to ANY piece of gear, be it hardware or software you have.Thoroughly learn it and it's characteristics.

It seems to me with plugin presets, often times people never really learn the gear.


Some day there may be a plug in with one button......"Sounds Great".........but not now. And I hope that day never comes, because that is the day no one will know what they are doing.:laughings:
 
anthonyg123 said:
When using a compressor, how much gain reduction is common or appropriate for an instrument with a lot of dynamic range such as bass?
Click to expand...

That is entirely down to what you're trying to do and what you want to hear.
You might want to compress the life out of a bass for a particular effect. On the other hand, you might want to catch the odd stray slap or peak, or something. (although volume automation would come first).
If you have a lot of held bass notes you might use a compressor quite heavily in order to effectively increase the sustain.

I'm just making the point that there isn't one common goal with compression. People use it to different degrees for different effects.

If you're struggling with the fundamentals of how a compressor works I'd recommend watching a youtube tutorial on avid's dyn3 compressor.
I know that's not what you're using but I find it to be a particularly good compressor for showing you exactly what's happening.

Threshold, knee, ratio, and gain reduction are shown plainly and simply live in the gui.

+1 to the 'working at the extremes' idea. That's great way to really hear what any tool is doing.

Hope that's useful. :)
 
Very true, thanks for the tip. I do see the value in learning and then being able to use something correctly. Also gonna check out that video, anything that does a better job explaining than most of the reading I've done is exactly what I'm looking for. Thank you!
 
when you use compressors, the important thing is to watch and see how much it is compressing - very often the incoming audio is simply below the threshold, so the compressor isn't compressing. It's a very common thing and people strain their ears for signs the compressor is doing something. Ramp up the input, or adjust the threshold so the line with the knee drops down, will start to compress and then you can hear it - then the ratio and other gizmos can be heard changing.
 
anthonyg123 said:
Very true, thanks for the tip. I do see the value in learning and then being able to use something correctly. Also gonna check out that video, anything that does a better job explaining than most of the reading I've done is exactly what I'm looking for. Thank you!
Click to expand...

No problem.
Plenty of people will happily try to explain I'm sure, myself included, but like I say, that comp is particularly intuitive.

From your OP, yes you're right. 2:1 and 3:1 should show a different gain reduction with everything else being equal. I'm not sure what's going on there. :eek:
 
Try zeroing that 'knee out. There's enough soft slope there to perhaps over shadow the rather small ratio dif.
Also, you might see a lot more reduction -and indicated in the meters if you increase the release a bunch. Now you have a wider (and longer) window to compare.

Also.. here's one the might be part of it as well! You have 'limiter like speeds set in attack/release, but the detector is set for RMS. Not a lot of average program going on in those short amounts of time.
 
Last edited:
@mixsit After messing with the knee a bit, it seems to seriously affect how much everything changes. This may be obvious, but when you read about a lot of compressors people tend to say "don't touch the knee" or make it seem unimportant. Do you know anything else about what this particular function does? Switching to peak does help as well where as the rms seems to have more of an affect, but because my settings are "limiter like" should I stay on peak? Thanks mate! :listeningmusic:
 
anthonyg123 said:
@mixsit After messing with the knee a bit, it seems to seriously affect how much everything changes. This may be obvious, but when you read about a lot of compressors people tend to say "don't touch the knee" or make it seem unimportant. Do you know anything else about what this particular function does?
Click to expand...

The knee effect is really obvious on that dyn3 compressor that I was talking about.
Hard knee means there's an instant transition from no compressor to compression right at the threshold.
Soft knee effectively implements a variable ratio around the threshold point, so compression kicks in more gently for audio that only just passes (or approaches) the threshold.
Anything that exceeds the threshold by a bit amount is hit just as hard as it would be with a hard knee, as far as I know.


This article says, and shows, it better than I am doing.
 
Never heard of 'not messing with the knee. It comes down to a taste and a bit (or more :) of sound and reaction difference is all. For what it's worth (nothing) I tend to spend more time in hard knee, and work more with the time and dector options.
 
Steenamaroo said:
The knee effect is really obvious on that dyn3 compressor that I was talking about.
Hard knee means there's an instant transition from no compressor to compression right at the threshold.
Soft knee effectively implements a variable ratio around the threshold point, so compression kicks in more gently for audio that only just passes (or approaches) the threshold.
Anything that exceeds the threshold by a bit amount is hit just as hard as it would be with a hard knee, as far as I know..
Click to expand...

Yep. And at low ratios, you could see were it might make for a quite subtle dif.
 
anthonyg123 said:
Edit: I attached two photos of the same note with a ratio of 2:1 and 3:1. They both have a gain reduction 3 that does not change. I may be wrong but isn't the ratio supposed to change the amount of gain reduction?
Click to expand...

The gain reduction depends on the Threshold combined with the ratio.
It's not just the ratio that's doing the job.

Having said that you can imagine the ratio as a knob controlling how "deep" the compressor digs to the sound.

A 2:1 ratio (low ratio) gives you opportunity to dig deeper cause in order to get aroun -6db of GR (-6db is just an example)
you need to go crazy with the threshold.

If you try to increase the ratio to 4:1 to this example then you're going deeper to the sound (-12 db)
which sounds unnatural, right? So what the next step?

You back off the threshold to make the sound audible again, so you're having the same GR now (-6db to this example)
but since the ratio is 4:1 and not 2:1, what happened since you've got the same GR?

This is what happened: The compressor is starting to act like a limiter, being more aggressive chopping the sound
more and more with each increase of the ratio, so you back off the threshold and as result you've got the same GR but you are not so deep to the sound.

In short: 2:1 to 10:1 ratios work fine to give some consistency , more than 10:1 and the compressor starts to act like a limiter that chops off the peaks (which is not bad I love limiters in modern productions, IF THEY ARE USED CORRECTLY cause you can lose all the dynamics and punch in an instant if you overdo it).

So Play with ratios from 2:1 to 10:1 for gentle leveling (ratios can change the sound a bit too, which can be nice sometimes), and go over 10:1 to chop off the peaks.
The less the ratio the more you dig into the sound.
 
If you try to increase the ratio to 4:1 to this example then you're going deeper to the sound (-12 db)
Click to expand...
I'm starting to understand but what did you mean with this, cause at the end you also said "the less the ratio" digs more into the sound. It does seem to "limit" the sound slightly more when turning up the ratio, if I'm saying that correctly.
 
anthonyg123 said:
I'm starting to understand but what did you mean with this, cause at the end you also said "the less the ratio" digs more into the sound. It does seem to "limit" the sound slightly more when turning up the ratio, if I'm saying that correctly.
Click to expand...

I meant that -12db is way too much for a GR.

That's not the sound you're listening to, that's the sound of the compressor going so deep
that I see adele rolling in it :D

Ok, to be serious, that's the sound of your compressor choking itself.

If you really feel the need you need to use such a low threshold
then it's better to use compressors in series.

What's compressors in series you may ask.

It's when you're using 2 compressors instead of one, to ease off the compression.

So if you're really sure that you need -12db of compression (again the number 12 is random) then use 2 compressors
with a gain reduction of 6.

You get what you're aiming for, plus the sound is natural, cause the compressors are not choking themselves.

Hope this is clears up some things :)
 
To Quote this: I may be wrong but isn't the ratio supposed to change the amount of gain reduction?

Well, not quite right, to put it simple, the ratio is the amount of gain reduction after the threshold has been reached.

A signal hits threshold at 2:1 the gain reduction is 2 db for every 1 db above threshold, if the ratio is 3:1 then the amount of gain reduction is 3 db for every 1 db above threshold.

So the quote is right in that the ratio changes the amount of gain reduction, but the gain still rises above threshold but the amount it raises is reduced.

Clear as mud?

Alan.
 
You must log in or register to reply here.
Back
Top