Limited Compression: redundant?

[Raveshaper]

A while back someone asked about adjusting makeup gain and threshold together to keep sounds level. It got me thinking.

You could set an MClass comp on your sound, then send the gain reduction curve to the programmer CV input of a combi that adjusted the input gain on a maximizer chained at the end of the comp. This would create a makeup gain and ride the level to prevent clips.

But would this be redundant? Or silly? I feel like trying it out.

[Olivier]

Wouldn't this (theoretically) result in no dynamics whatsoever ? Why would you want to do that ?

[Raveshaper]

Yeah, I don't know. It sounded odd so I thought I'd explore it. But I don't see an immediate benefit.

[avasopht]

Well because what you've described results in a feedback loop (compressor outputs CV that affects an input parameter) the CV will be processed in the next batch. Bearing in mind as well that CV changes once every 1.5ms (at 44.1k), there's no telling what it might do, but it's still worth trying out.

[Gaja]

Isn't that basically part of what the Selig Leveller does?

[Raveshaper]

I believe the leveler does something like this, don't know for sure.

[Benedict]

This is something I raised in that initial thread. Yes in theory you could create a sort-of make-up for the compression amount by using the Gain Reduction CV value to drive the output volume (either on the same Comp or on a following device as you suggest).

Of course before you get into it you need to know why you are compressing. If you are compressing to change the shape or give more punch then the last thing you want is to turn the volume back up by the same amount you just turned it down. If you are looking to lop-tops then maybe there could be some sense but I would think you would want to smooth that CV out or your track volume would stagger all over the place.

I can't really see much good coming from it though as you need to use your eyes and ears and not a random robot for this sort of stuff.

[selig]

Well because what you've described results in a feedback loop (compressor outputs CV that affects an input parameter) the CV will be processed in the next batch. Bearing in mind as well that CV changes once every 1.5ms (at 44.1k), there's no telling what it might do, but it's still worth trying out.

Not as was described, because they suggested a separate MClass to generate the CV, which in turn controls the input gain of a Maximizer. But what is the intention with this device? If you reduce the gain going into a compressor by the CV of another compressor, it's the same thing as taking the output of the first compressor into the second, right? You are basically "compressing" the input to the Maximizer with this setup. 

EDIT: that being said, "serial" compression is a common way to control dynamics without sounding as "controlled", and I've found going from a low ratio compressor into a limiter produces desirable results. 

[tibah]

Another interesting question that I usually end up thinking about is: How many elements of my track actually do have dynamics? If you are like me and your music is primarily based on virtual instruments and samples.

A compressor can be used for color/flavor of course, but most of the time nowadays I apply nothing else but a little bus compression and/or good old Scream4's Tape setting compression to my tracks.

[kloeckno]

That is kind of similar to the AGC (Automatic Gain Control) circuits used on radios and portable recorders. My Zoom H2 has that feature.