Compression and saturation, in what order?

[pekadan]

Would you first saturate a signal then compress it, or the other way around? Why?

[tahome]

Depends on your material. Both are non-linear processes which means that the order actually matters a lot.

I typically use slight compression for mastering (typically multiband via Dynasone) and apply some saturation via Satur8 Pro afterwards.

But there is no general rule of thumb, it all depends on your application (kindof like asking "what EQ settings do you use for voice").

HTH,
--th

[pekadan]

I'm thinking mainly for processing individual tracks, e.g. vocals, drums, guitars.

I don't really have a specific example or problem or so. I'm more generally interested in understanding what difference the order will make.

I mean, what is the result of saturating a signal with lots of dynamics compared to one with high loudness?

[tahome]

Saturation only has an audible effect if your signal volume exceeds a certain threshold, so it might be better to compress its dynamic range first before applying saturation, otherwise saturation might not have any effect at all.

Applying compression afterwards might make the distortion effects of saturation more audible (especially when using MBC).

I guess that are the two important things to keep in mind here.

--th

[Krakatau]

As a clear example :

Guitar then compressor then distortion

Intense saturation effect with long and sustained chords or single notes, perfect for heavy metal....but you'll have to deal with high background noises, hisses and larsen effects**

Guitar then distortion then compressor

Even if the signal is highly saturated, the typical pumping effect due to the compressor unit dominates the overall treatment and will be first noticable from your guitar track inside your mix

Keep in mind that the following effects always dominates the preceeding ones and the only exception to the rules concerns effects that works mixed with the original signal ( chorus, phasers, reverbs and a few others )

hope that helps..

______

(** this converges in its way to the explanation tahome gave to you : the preceeding compression will highly increase overal saturation level )

[tahome]

Keep in mind that the following effects always dominates the preceeding ones and the only exception to the rules concerns effects that works mixed with the original signal ( chorus, phasers, reverbs and a few others )

More precisely, this applies to all linear effects. For linear effects (EQ/filters, reverb, chorus, delay) the order of the effect chain doesn't make any difference.

"larsen effects": what's that again? Never heard of it...

--th

[pekadan]

This is really interesting and useful information guys. Thanks!!!

What does it mean exactly that an effect is linear or non-linear?

[Krakatau]

Larsen effect ( in frensh " EFFET LARSEN"...maybe an unapropriate translation ) ...you know what i mean the unwanted feeback between microphone and speakers !

[tahome]

@pekadan: "Linear" is a definition from DSP, you might want to look it up on a dsp web site (look for "LTI"). The bottom line of it is that the combinations of linear processes are interchangeable.

@Krakatau: you're referring to a feedback loop... I didn't know there was a name for it. Do you have any reference for this naming convention?

--th

[Krakatau]

http://fr.wikipedia.org/wiki/Effet_Larsen (...from the Danish physician Soren Larsen )

in frensh i'm afraid...


________


...but you can have a look at this too :

http://www.ok-centrum.at/english/presse/larsen.html

[Krakatau]

Seems to be much more common in frensh though

( 938 matches in google for "effet Larsen" and only 5 for "Larsen effect" )

[pekadan]

Thanks again! Yeah, I definitely have to learn more about linear vs non-linear effects. That seems to be very very useful knowledge!

[bobby yarrow]

For linear effects (EQ/filters, reverb, chorus, delay) the order of the effect chain doesn't make any difference.

What?

Delay>Reverb you get the sound of the dry signal & delay signals going thru the reverb. Could sound like, for instance, a guitar player using a delay in large room.

Reverb>Delay you get the sound of reverb-treated signal, and then you get it again as the delay repeats. Could sound like, for instance, a guitar player in large room being recorded thru a space echo.

Same applies to eq, filters, and chorus. Try it.

[bobby yarrow]

Oh, and, to answer the original question: I'm not exactly sure what you mean by saturation. If you mean an exciter or a distortion simulator, I rarely use one at all, but when I do I treat it as EQ which I (generally) prefer after compression. It will sound different, so try both. In particular, try not using exciters.

[tahome]

No, not true. You might have misunderstood what I was referring to.

If you have a signal chain like this

Sound -> Reverb -> Delay -> EQ -> Output

then the net effect is the same as if you had

Sound -> EQ -> Delay -> Reverb -> Output

or any other combination.

Of course, if you have a dry path that goes around one of the blocks you are not only changing just the order but also the mix of the individual block's effects. In that case you will of course have two different setups. Only the sequence of effects may be changed without altering the effect, if you change the mix you are changing the system.

As for the second question: saturation means hard or soft limiting. Essentially this is (soft)clipping.

--th