Stacking compression effects. Valid, or unforgivable crime against recording methods?

cfg

Member
I am still very much in the learning stage with Reaper and capturing some rough tracks to experiment on here and there to hone my recording skills. I just tried recording a simple acoustic guitar with a CAD M179, which I understand is a half decent condenser, and the raw track was as flat as anything, truly lifeless. So I started throwing on plug ins, free ones, but good free ones I've discovered after much reading.

After quite a bit of tweaking and tinkering, I actually have a sound that I think is very respectable. The acoustic guitar track (blues/rock style) is now punchy, with real body and presence, and I wouldn't be ashamed to let someone listen to it. Here's the problem. On the track I've got no less than three (probably four) compressors or effects that claim use subtle compression. Now it's not a problem in the sense that I like the sound I've cultivated with the plugins, it's a problem because according to my internal thinking, there is some recording engineering law against stacking so many compression effects. I'm not sure where I got this notion from or whether I just conjured it, but in any event, I can't seem to capture the sound I want from just one compressor, even if it's capable of a lot of compression. From this keyboard, there seems to be some sonic advantage of using smaller quantities of compression across a number of compression (or compression leaning) plugins.

So, what is the prevailing theory on this topic of using more than two or three compressors on a track to get the sound one likes? Is it acceptable? Should I only be concerned with what the listener might think?
 
It's a "secret" that is now pretty much common knowledge. Several compressors each doing less is almost always better than one compressor doing a whole lot.

A lot of folks will tell you to use a fast "peak" compressor first and then a slow "leveling" compressor after, but I think that's a holdover from the analog days, and usually prefer to do the opposite. Use a slow RMS compressor, preferably with some lookahead to offset the delay in the RMS detection in order to get each "hit" closer to the same average level, and then apply your peak compression so that each hit gets about the same treatment. It helps keep the internal dynamic envelope of each hit and the action of that peak compressor a lot more consistent and just sounds more natural in general.

Now you might put an "envelope control" compressor - like when you're trying to push down the sustain of a tom so the attack comes through more - where that peak compressor was, but it's still better after the general leveling for the same reason. And frankly, this should probably come before any "ceiling protection" type peak compression or limiting anyway.

So now we've got "leveling" RMS compression>"envelope control">peak control>"ceiling" limiting/clipping/saturation.

IISGIIG is not anybody's favorite response, but seriously like nobody knows or cares how many compressors you put on the thing except you and a few nerds on a home recording forum.
 
...it's a problem because according to my internal thinking, there is some recording engineering law against stacking so many compression effects.
There are no rules. Make it sound as good as you can in any way that you can.


Should I only be concerned with what the listener might think?
In a word, yes.
 
...IISGIIG...

I've seen you use this acronym a few times now, and TBH, I have no clue what it means. :D

OK...maybe I'm not 2018 "hip" if this is some very recent new addition to the think-speak dictionary so many people use these days...but I even Googled it and came up with nothing.

So what's it mean exactly...? :)

...but seriously like nobody knows or cares how many compressors you put on the thing except you and a few nerds on a home recording forum.

:laughings: :thumbs up:
 
I do this at times (usually only 2) but lately on my "pseudo mastering" buss I have been using multiple limiters to get the sound I want. The only thing I will say is to remember to check in bypass and level match to avoid the dreaded "louder is better" syndrome.
 
if it sounds good it is good, probably

:thumbs up: ....yeah, that looks right.
Funny that Goggle, which has maybe hundreds if not thousands of these "modern" acronyms listed...they don't seem to know what that means. :D

You know...that's actually a paraphrase of a Joe Meek quote. He apparently use to say: "If it sounds right, it is right" :cool:
 
Last edited:
There's no problem. But I recommend that each should have its own function in the chain. Like I have comp up front, before interface.only function is to keep bad peaks out of the preamp. Some channels may also just for peak taming. Then comp on master, 2:1. Then limiter after that. So there's several, spread out, doing their own thing so they arent working at odds with each other.
 
I've seen you use this acronym a few times now, and TBH, I have no clue what it means. :D

OK...maybe I'm not 2018 "hip" if this is some very recent new addition to the think-speak dictionary so many people use these days...but I even Googled it and came up with nothing.

So what's it mean exactly...? :)



:laughings: :thumbs up:
I hate that! Years ago when I first started mixing, I had a song to the point that it sounded good TO ME. When I asked for critiques on a forum (my have been this one!), the overwhelming majority of people said it was smashed to death and sounded horrible. But it sounded good TO ME. So, if it sounds good it is good doesn't necessarily work, especially for the newer mixer.

I'm not much better now than I was back then, by the way... :D
 
Should I only be concerned with what the listener might think?
There are no rules. Make it sound as good as you can in any way that you can.
In a word, yes.
[MENTION=199965]cfg[/MENTION], I very much agree with Adrian here, but there's a wide range of what can 'sound good'. It comes down to what people expect. And your own expectations matter very much.

So, what is the prevailing theory on this topic of using more than two or three compressors on a track to get the sound one likes? Is it acceptable?
A direct answer - prevailing theory is a clear 'yes'. Prevailing practice is also 'yes.'

I chain up compressors on a regular basis. And I'm not stringy with them either. Here's some of my thoughts on the approach:

- Begin with the ending sound in mind. Know where you want to go with the sound before you start chaining compressors together.
- Teach yourself through experimentation how different combinations of compressors work together. Try this: Waves LA-3A -> LA-2A on a snare. When the 3A needles is at 5db reduction and the 2A needle is at 3db reduction what do I hear?
When the 3A needles is at 2db reduction and the 2A needle is at 5db reduction what do I hear?
If I run the 2A FIRST, mash out 10db reduction then gently follow with the 3A 2-3db reduction what do I hear?

When observing affect of comps in series, best to start with comps that have fixed attack and release times. Because its easier to observe what they do in relation to another compressor. Less variables.

-Also learn the different characteristics of the different circuit families. This will help minimize guess work. Learn how a FET compressor is tonally different than a VCA. Learn how your Vari-Mu family behaves different than your opticals. Learn what the most commonly used compressors are and why they are so popular.

-Demo demo demo. Try everything.
 
Last edited:
Having a week long break from the studio helps me alot.
I used to sit for hoooours and more and more frustrated, and the song i worked on sounded like a trash, and I gave up.
, but then I listened to the same song a week later, with fresh ears, and that gave me the perspective to eventually pinpoint what the hell was wrong with the mix.

If it's relevant to this topic not, I do not know, but distance, and fresh ears definitely helped me.
 
Back
Top