What Would Two Identical Compressors Next To Each Other Do?

[penguinpajamas]

Hey guys! So I've been working on a mix for a song and I'm running into a problem that I just can't figure out. I'm revisiting a song and I think it was a mistake that I liked and kept, but I have a bass synth line that's being side-chain compressed via a kick for the classic pumping effect but what's unique here is that this compressor is duplicated to an identical one, feeding the output of one compressor into another identical one with the same side-chaining. The two compressors in series definitely adds some dynamics to the bass that I just can't seem to replicate with a single compressor.

So I suppose my question is, why is that I can only achieve this effect by keeping these two identical compressors in series? My intuition would be that if I squared the ratio of the dual compressors, I would be able to duplicate the effect with a single compressor.

Any thoughts are appreciated.

- Charlie

[Richie Witch]

While I think your idea about squaring the compression ratio is a valid mathematical model, I suspect the issue is more about attack/release timing.

Let's say the attack on the first compressor is 5 ms, so it's going to reshape the transient and send the signal along. Then the second duplicate compressor is going to do that again, but with another 5 ms slope on the attack. So you would have two identical, but not quite cumulative attacks, due to the shift in time. There would be a similar time-based effect on the release.

Maybe... just maybe... if you square the attack and release times as well, you might get something similar, depending on how the waveform is being reshaped. Have you looked at the output of each compressor on a scope?

[ChironControl]

Great question! There are a lot of non-linear processing steps. It is not an easy calculation to adapt the settings of two chained compressors to one single compressor. Probably it is not possible to get exactly the same output.

I agree that squaring the ratio is the best to have the gains doubled. But with the same time constants, one compressor would need longer to adapt to these gains. Thus, the times have to be set shorter. This could be the best you can do.

I am not sure if there is a time shift - I think the latency compensation works in a way, that there is no time shift. As both of the two compressors get the same input and produce exactly the same gains, this shouldn't have any effect.

[Angstrom]

many people misunderstand what the envelope controls on a compressor do.
A lot of people think they are like synth ADSR controls, assuuming that when the signal is over the threshold the 'attack' phase is triggered. This is not correct. This is not a 'triggered' AR stage, it is continuous. The A and R here are more like a slew control, or a portamento control with separate up and down components.

Have a play around with AutoFilter which has a similiar method on the 'Envelope', notice how it responds to input continuously.

In the compressor While the signal is above the threshold the Attack value will control how quickly the compression ratio is applied to the signal at each moment. The release value controls how quickly the ratio application falls off.

So, once you start to think of the Enveloping as an ACTIVE process on the sound, like a portamento for volume. Now think what might happen if you stacked two of these "volume portamento" envelopes. This is likely why you hear a difference.

[ChironControl]

...
So, once you start to think of the Enveloping as an ACTIVE process on the sound, like a portamento for volume. Now think what might happen if you stacked two of these "volume portamento" envelopes. This is likely why you hear a difference.

The interesting thing here is that we have side chaining with the same input, i.e. we have two compressors calculating exactly the same gains. Assuming that there isn't any delay or time shift, for each sample the same gain is applied, i.e. a summation of the two gains happens for each sample.

The time behaviour which you describe can also be seen as a smoothing of the gains. It is thinkable that you can achieve similar results by setting shorter times. But that depends on the exact implementation of the "smoothing", and it is not possible to make general statements about it.

[Tarekith]

This should be required reading for all electronic musicians:

https://www.attackmagazine.com/features ... mpression/

[penguinpajamas]

Thanks for the advice guys. I like to think I understand how compressors work, as I've done a lot of reading on the topic, yet I still can't seem to get what I'm looking for.

I suppose my question at this point would be, is there any real reason not to just keep the two compressors rather than reducing them down to one?

[Angstrom]

Keep your two compressors if it sounds good! That's the only rule.

[yur2die4]

Use the two. Not uncommon at all.

[chapelier fou]

Weirdest question ever !