How do I use a compressor the right way?
-
- KVRist
- Topic Starter
- 57 posts since 14 Feb, 2015
Hey guys, first of all: I know in general how a compressor works, I just feel like I haven't used compressors to their full extent yet. One example: I have a piano chord progression and I feel like it needs some "thickness" or "force". My experience is that compressors can help with this. The problem is though that the compressor ruins the transients of the piano. I've tried dialing in a longer attack and all but I havent't really come up with anything I like. I have tried using Maximus (I'm using FL), which is a multibandcompressor and it's default setting sounds really good in some cases (nice heavy bass and bright highs), but I'd like to really understand exactly how to use a compressor/MB compressor to get the sound that I'm after. Thanks for your help and advice!
-
- KVRAF
- 3080 posts since 17 Apr, 2005 from S.E. TN
What compression ratios have you tried?
If you just want to mildly level single note melodies interspersed between chords, that sort of thing, you could try a soft knee or "over easy" curve with a mild compression ratio maybe somewhere in the ballpark of 1.5 : 1 or 2 : 1 up to 3 : 1 or even 4 : 1. More likely transparent compression in the lower end of that range.
That might give the piano a little more sustain with more level amplitude without messing up the note attacks real brutally.
With a gentle compression ratio and threshold set low enough to give 6 dB or more gain reduction on the loudest sections, you might be able to use fairly fast attack times without spoiling the piano sound. Maybe even 1 to 10 ms, but longer if your ear likes it better thataway. If you use fairly long attack time, then the release needs to be pretty long as well.
However, for piano compression the release time should be fairly long to avoid intermodulation distortion. Especially two fisted piano playing where short release allows bass notes to intermodulate treble notes. But even noticeable on midrange triads if release is too short.
Different comps mean different things re attack and release times, but I'd maybe use minimum of 100 ms release on piano, perhaps much longer release time.
If you just want to mildly level single note melodies interspersed between chords, that sort of thing, you could try a soft knee or "over easy" curve with a mild compression ratio maybe somewhere in the ballpark of 1.5 : 1 or 2 : 1 up to 3 : 1 or even 4 : 1. More likely transparent compression in the lower end of that range.
That might give the piano a little more sustain with more level amplitude without messing up the note attacks real brutally.
With a gentle compression ratio and threshold set low enough to give 6 dB or more gain reduction on the loudest sections, you might be able to use fairly fast attack times without spoiling the piano sound. Maybe even 1 to 10 ms, but longer if your ear likes it better thataway. If you use fairly long attack time, then the release needs to be pretty long as well.
However, for piano compression the release time should be fairly long to avoid intermodulation distortion. Especially two fisted piano playing where short release allows bass notes to intermodulate treble notes. But even noticeable on midrange triads if release is too short.
Different comps mean different things re attack and release times, but I'd maybe use minimum of 100 ms release on piano, perhaps much longer release time.
-
- KVRist
- 155 posts since 24 Jul, 2012
Hi Moniatre,
the way a compressor react is dependent to the sound level you feed in and is character that's why if you are in luck a preset cand make the job...
To be objective there is an important rule when you compare the original signal with the compressed one, they have to be compared at the same level (meterplugs perception or melda production mcompare plugins can help on this process), in fact you can be fooled by the resulting overall increase in loudness.
http://en.m.wikipedia.org/wiki/Fletcher–Munson_curves
Using a multiband compressor can be very tricky,most of time the single band ones are a better choice.Use mutiband only when is necessary.
The kind of compressor (peak,rms,opto,vca,vary-mu) sound very different to the same source, some work best for drums other for vocals, or affect more or less the transients...the right compressor choice make huge difference in the mix.
if you want to learn have a look to this tutorial,is not a free one but it covers all aspects of compression:
https://www.macprovideo.com/tutorial/ar ... g-dynamics
Hope it helps.
the way a compressor react is dependent to the sound level you feed in and is character that's why if you are in luck a preset cand make the job...
To be objective there is an important rule when you compare the original signal with the compressed one, they have to be compared at the same level (meterplugs perception or melda production mcompare plugins can help on this process), in fact you can be fooled by the resulting overall increase in loudness.
http://en.m.wikipedia.org/wiki/Fletcher–Munson_curves
Using a multiband compressor can be very tricky,most of time the single band ones are a better choice.Use mutiband only when is necessary.
The kind of compressor (peak,rms,opto,vca,vary-mu) sound very different to the same source, some work best for drums other for vocals, or affect more or less the transients...the right compressor choice make huge difference in the mix.
if you want to learn have a look to this tutorial,is not a free one but it covers all aspects of compression:
https://www.macprovideo.com/tutorial/ar ... g-dynamics
Hope it helps.
Last edited by Chandran on Thu May 07, 2015 11:21 am, edited 2 times in total.
-
- KVRAF
- 2549 posts since 13 Mar, 2004
There's this 'method' I keep in mind when setting up compressors.
In short:
https://www.gearslutz.com/board/newbie- ... essor.html
from this book where it's described in detail
http://www.mixingwithyourmind.com/contents.php
edit:
And, as the poster above says, always comparing loudness matched...
In short:
https://www.gearslutz.com/board/newbie- ... essor.html
from this book where it's described in detail
http://www.mixingwithyourmind.com/contents.php
edit:
And, as the poster above says, always comparing loudness matched...
-
- KVRAF
- 3080 posts since 17 Apr, 2005 from S.E. TN
Yep, better to use single-band compression except for tough problems. Compressing acoustic piano isn't usually a real tough problem. Or phrased another way, most pop recordings with piano, have compressed piano, so lots of people have managed to do it "good enough". 
-
- KVRAF
- 4590 posts since 7 Jun, 2012 from Warsaw
Use it along with Dynamic Range Meter to make sure nothing is overcompressed.
If you have compressor with analog gauge, rule of the tumb is to make it swing as far as possible. This makes that you both are compressing hard and have some dynamics in the signal.
If you have compressor with analog gauge, rule of the tumb is to make it swing as far as possible. This makes that you both are compressing hard and have some dynamics in the signal.
Blog ------------- YouTube channel
Tricky-Loops wrote: (...)someone like Armin van Buuren who claims to make a track in half an hour and all his songs sound somewhat boring(...)
Tricky-Loops wrote: (...)someone like Armin van Buuren who claims to make a track in half an hour and all his songs sound somewhat boring(...)
-
- KVRist
- 137 posts since 2 May, 2007
Have you tried NY style/parallel compression? It is simple enough and can help preserve transients while letting you play around. If you use a send channel, you can EQ the channels separately too.
