The Logic Off Topic list

Hi

 > > > Mikael Adle originally asked:
 > > > Could someone explain the difference between peak and rms setting on
 > > > a compressor and when to use what?

in response to which Kool Musick in part wrote:
> > Rule of thumb ... peak for percussion, RMS for pretty much everything else.
> > Break this rule often and you'll be a good mix engineer!!

unfortunately for Kool Musick it wasn't over yet because blue alien THEN 
wrote:>
>Why would you compress a signal and then expand?  Don't "pro" engineers
>often use these types of effects in this order during mastering?

Yes indeed, "pro" engineers do use these various types of effects in that 
order when mastering -- although occasionally also the opposite.

First off, I am not a "pro engineer". Just an enthusiastic amateur without 
a single hit or even a professional mix job that someone has paid me for to 
my name. I just enjoy messing around with stuff in my sitting room and 
garage. So ... please take the little I know with the proverbial pinch of 
salt. Hopefully, someone who knows a LOT more will correct anything I say 
that's wrong.

The trouble with compressing and/or expanding -- in fact with ANY signal 
processing -- is that you never affect just the one thing you set out 
trying to affect (or effect). Sound is much more organic or holistic than 
that, and you affect everything. You might, for example, want to use your Q 
setting on your eq to hit that horrible sound you are hearing at say 6Khz, 
but what you in fact do is hit everything else -- vocals, bass guitar, that 
nice acoustic etc -- that's also sitting in your mix at around that 
frequency. All you can do then is use your ears to dowse your nightmare 
sound in the best way you can, while still leaving enough of everything 
else present to make it sound "good". For example, after you've done your 
cutting you might use a parametric to boost the signals you do want a 
little bit on either side of where you've just cut and so that the sounds 
you DO want bleed back in and so fill up the hole you just created. Problem 
here is, of course, as you all know, that you are very likely going to 
reintroduce at least a little more of that nightmare sound you were trying 
to get rid of in the first place!!

OK. So music's organic and you can't really isolate anything.

The compressing/expanding issue deals -- as far as I know -- with much the 
same kind of terrain. If you're going to compress, then you're going to 
darken the signal a bit. By the very nature of the compression, the aural 
energy of the lower frequency sounds tends to be brought into increased 
dominance by the whole compressing process. Trebles end up having pretty 
much the same ratio as basses. The whole sound can therefore lose quite a 
bit in its 'sparkle' or definition. When using the compressor you have to 
balance this up against the dynamic effect you are looking to create.

Once you've done your compressing, you may then decide that your mix has 
lost a bit of its 'sheen' -- that sheet of high frequency transients that 
made the original sound more alive -- even though the original contained 
that dynamic range you wanted to suppress. The problem then is ... how do 
you regenerate some of those high frequency transients without also 
reintroducing the dynamic peaking associated with those transients ... and 
that probably made you want to compress in the first place?

Some very expensive compressors (and plug-ins) try to address this by 
putting in some very nifty pieces of circuitry which are effectively high 
frequency harmonic content generators so while you are actually 
compressing, you can add back in to the sound you are creating some of 
those 'aurally pleasing' high frequency detail contents ... but still 
achieve the squashing or compressing effect you want.

Sometimes, also, you want to compress, and to create your wet sound ... but 
then add that compressed signal back in to the original dry sound ... and 
yet STILL not get your transient distortions etc. This would be the case if 
you wanted a nice fat sounding bass, for example. You would isolate the 
bass frequency, compress it, and then add it back to the original with not 
really much danger of distortion etc, because the sound you are bringing 
back is probably not going to affect your high transients over much. It's 
possible to compress, alter by EQ, and then add back in all simultaneously.


If you are using a compressor, you are probably also using a gate -- and 
immediately you have the problem of 'swallowing' those high transients. You 
thus lose some of your reverb tails and nice decays. Lots of things can happen.

Effectively, therefore, the process of compressing and then gating is an 
attempt to overcome the problems caused by compression. The signal you are 
left with after compression is dynamically acceptable when compared to the 
overall dynamic reference level you are dealing with, but on the 
equalization side it may be a bit suspect. But ... the sound is probably 
'safe' in that you have taken out some of the high frequency transients. 
Therefore, if you now expand, you are working with a controlled and 
controllable piece of sonic input, and you can try to re-introduce your 
high frequency sheen effects without losing the overall dynamic effect you 
wanted.

I hope this makes sense.

I have tried to be clear and also explain my reasoning so that if it's 
faulty we'll all know where and why in about five minutes time when some 
'pro' mastering engineer steps up to the plate and tells you all what a 
complete drongo I am and how I don't know nothing. But ... that's MY 
understanding of what's going on, and that's the theory I use behind my own 
work.

HTH

Kool Musick
Keep Musick Kool


_________________________________________________________
Do You Yahoo!?
Get your free @... address at http://mail.yahoo.com