Skip to main content

Let's say I had 2 compressors in series with the following parameters:

Threshold = -5dB
Ratio = 2: 1
Attack = 50 ms
Release = 40 ms

Would there be any difference (technical / sonic / theoretical) if I had a single compressor configured as below?

Threshold = -10 dB
Ratio = 4: 1
Attack = 100 ms
Release = 80 ms

Thanks in advance.

Comments

kmetal Mon, 06/18/2018 - 14:44

Sonically using two in series will result in less extreme gain reduction ie less pumping, or artifacts. Doesn’t mean either way is wrong, more that one way may be more appropriate for the artists vision.

Particularly with digital compressors seeming to be faster acting than the analog counterparts (all other things equal as possible), and having no self noise (in theory) running them in series isn’t a bad idea in general.

When running in series your compressors are hitting different points in time and amplitude, one right after the other, where a single compressor hits the signal at a single point in time and amplitude.

kmetal Mon, 06/18/2018 - 16:16

Boswell, post: 457736, member: 29034 wrote: For the two identical compressors in series, the time-related parameters do not change.

Can you clarify that? I want to make sure I understand this properly. If a signal flows from one compressor to the next one in series, aren’t they hitting the signal ever so slightly at different points in time? Otherwise it would be parallel compression right?

I’m not sure if my understanding is incorrect or we are talking about different things?

Johnny Blade Mon, 06/18/2018 - 17:23

Hey, guys, thanks a lot for supporting my question! I'm learning a lot with you all. (y)

I'm not good in English, so I would like to reinforce it.

I think I'm talking about serial compression, because I'm working in the slots of Samplitude's Object Editor. I'm not sure, but I don't think it's possible to do a parallel compression in Object Editor. It seems that T-RackS does a better job in this way (parallel compression). Anyway, this is a minor issue.

The comparison I wanted to make is simple: do 2 compressors (Comp1 & Comp2) identically configured and working in series (one after another = one slot after another), produce the same result as a single compressor (Comp3) configured with the summed values of those (Comp1 + Comp2)?

The Make-up gain of the 2 compressors would have identical values; the single compressor, then, would have the added value of those two.

In short: are 2 equally configured compressors the same thing as a single configured with the sum of those? :)

kmetal - it seems you prefer one single compressor, right?

Boswell - so, only threshold, ratio and make-up gain I should add the values from Comp1+Comp2 to Comp3, right?

kmetal Mon, 06/18/2018 - 18:21

Johnny Blade, post: 457742, member: 50338 wrote: The comparison I wanted to make is simple: do 2 compressors (Comp1 & Comp2) identically configured and working in series (one after another = one slot after another), produce the same result as a single compressor (Comp3) configured with the summed values of those (Comp1 + Comp2)?

This will produce different results. If I’m thinking of it correctly, the (comp 1) will modify the raw waveform, then the (comp 2) will effect the modified (compressed by comp 1) waveform. So your compressing a compressed signal.

With a single compressor your only compressing the raw waveform.

I am curious about the sonic differences, since I’ve never tried this experiment, matching settings ect.

Johnny Blade, post: 457742, member: 50338 wrote: kmetal - it seems you prefer one single compressor, right?

In general no, I tend to use a compressor on the track, and then on the group bus that the track is fed to. The compressor on the track is generally for basic control, or for some effect, then group bus compressor is for ‘glue’ to kind of help blend all the signals together.

Example:

3 snare tracks- each get an 1176

These compressed snares are routed to
‘Snare Bus’,

The ‘Snare Bus’ then gets an API Bus Comp.

I’m not sure if this is technically considered ‘in series’ since there may be Eq or other effects in between.

An example of when I’d use technically ‘in series’

Vocal track- basic compression for level consistency (comp 1), then (comp 2) set just to catch random peaks.

Not sure if that contributes anything worthwhile, this is an interesting thread.

Boswell Tue, 06/19/2018 - 04:31

Johnny Blade, post: 457742, member: 50338 wrote: The comparison I wanted to make is simple: do 2 compressors (Comp1 & Comp2) identically configured and working in series (one after another = one slot after another), produce the same result as a single compressor (Comp3) configured with the summed values of those (Comp1 + Comp2)?

The Make-up gain of the 2 compressors would have identical values; the single compressor, then, would have the added value of those two.

In short: are 2 equally configured compressors the same thing as a single configured with the sum of those? :)

Boswell - so, only threshold, ratio and make-up gain I should add the values from Comp1+Comp2 to Comp3, right?

In engineering, there is the concept of a Linear, Time-Invariant (LTI) system. In principle, these can be stacked end-to-end (in series), and the order of stacking does not affect the final result. An example from the audio field is an analogue filter. The EQ section of an analogue mixer usually has 3 or 4 sections covering overlapping frequency ranges, and you, as a user, have no way of knowing in which order these are implemented inside the mixer, igoring noise and non-linear effects such as amplitude limits.

By contrast, compressors are not LTI systems. They work in a non-linear fashion, and the series order matters (when the settings are not identical on the two compressors).

I think the problem in phrasing the question in the way you did is the use of the word "sum", because summing is essentially a linear operation. If you change the original question to be something like "Is it true that you cannot set a single compressor to give the same result as two separate compressors of the same type used in series?", it's clear that the answer is that you can't, and that there will always be a difference. So Kurt's long-winded reply of "yes" to the initial question is right on the money.

Johnny Blade Tue, 06/19/2018 - 09:05

WOW... thank you very much, Mr. Boswell ! It' all very clear to me, thanks for the lesson!

I did a single test an I would like to share here.

In Samplitude, at level of Object Editor, I mastered the same music in 2 different tracks.

I just use compressor and limiter.

Compressor - T-RackS Opto Compressor.
Limiter - Samplitude's sMax11

Track 1 = 2 compressors + Limiter
Track 2 = 1 compressor + Limiter

On Track 1, these are the parameters of the same both 2 compressors (Opto Comp.):

Threshold = -10 dB
Ratio = 1.5:1
Attack = 50 ms
Release = 40 ms
Output = 2dB

On Track 2, a single compressor in this way:

Threshold = -10 dB
Ratio = 3:1
Attack = 50 ms
Release = 40 ms
Output = 4 dB

The result is that Track 1 is "bigger", "stronger", with "more punch", much more solid, more vivid, more thick, more mass, more full-bodied.

I think they should sound the same, but Track 1 pleased me more. o_O