Login / Register  0 items | $0.00 New What is KVR? Submit News Advertise

Compressor analysis

1913 posts since 13 Oct, 2002

Postby Breeze; Sat Jan 24, 2015 8:16 am Compressor analysis

How many compressors do you need? It seems the flood never stops! With advanced configurable compressors like Pressor and Compassion, it seems to me many compressor behaviours can be emulated, maybe not perfectly but quite closely. What can't be emulated of course is the non-linearities of the emulated audio circuitry, but are the compression curves themselves also non-linear? I guess that's my first question, and the answer is very likely "yes".

My main question is how does one go about examining compressor behavior and curves? I'm thinking setting up a number of broadband noise samples of varied attack times, Fq distributions, durations and releases, feed them into the compressor under test, recording and analyzing the results. Is there a better way? Thanks for any input.
User avatar
89 posts since 16 Mar, 2012

Postby tomicious; Sat Jan 24, 2015 8:37 am Re: Compressor analysis

I am Virtual Analogy [Paraphrasis]
User avatar
9 posts since 9 Nov, 2005, from Germany

Postby MyCo; Thu Feb 05, 2015 3:31 pm Re: Compressor analysis

Breeze wrote:How many compressors do you need? It seems the flood never stops!

When you watch some of the production videos on youtube, you'll notice instantly: there can't be enough. Some guys put 2-3 different compressors just on the kick mixer channel :dog:

A good way to analyse a compressor is to
1. correlate input to output to get the overall delay
2. subtract the correlated input from output and then remove the DC
3. do some basic spectrum analysis

You'll get a basic idea of the distortion just by looking at the frequencies.

But to be honest I wouldn't bother. Just build it and when it sounds good, release it.
My Blog, although I don't blog a lot :hihi:

Moderator: Moderators (Main)

Return to DSP and Plug-in Development