The Logic Off Topic list

cosmicdali wrote:

>Great!
>So in other words, a cymbal should do better in rms mode?

Very good question. There's no simple answer to this. It really depends on 
the effect you are trying to achieve, and what else is happening in your 
project.

In general, though, one would probably compress a cymbal in peak mode 
simply because of the range of frequencies where its transients lie, and 
the quick attack it has.

However ... is the cymbal being played e.g. every 1/8th or 1/16th note? Is 
it being stroked or hit? Is it maybe being played regularly on the main 
beat in a chorus? If so, then it might well be more in the category of an 
overall "mood-setter" in the project, and therefore you might well be 
correct treating it through its rms or average values, rather than through 
its peak ones ... as one generally does with percussion sounds.


I tend to agree with monsdrum in that really this is a matter of learning 
to use your ears. There's no definitive answer as far as USING is concerned.



What the RMS value does is give you the ability to work with the overall 
average level of a sound. The advantage here is that sounds of short 
duration are treated exactly like our ears would treat them. They sound 
quieter then they actually are because our ears work with the reference 
level of everything we are listening to, and use that as a basis for 
judging acoustic phenomena. RMS, therefore, gives a more 'natural' sound.

Peak, however, is in fact more accurate in how it treats transients in that 
it works "with how the sound actually is" rather than with how "it 
appears". Please don't take that too literally, which is why I've put it in 
brackets. The advantage that this also gives is that since the peak level 
is giving you what's actually there, it's a lot easier to isolate and 
control it. However, the problem is that one often ends up giving undue 
prominence to that sound simply because it is standing out aurally in a way 
that it normally would not. Also ... you can't just squash or compress just 
the peak sound once it's generally 'in the mix'. As soon as your compressor 
kicks in in peak mode, your whole audio is going to get squashed simply 
because there's a short transient sound around. This is not usually 
regarded as 'a good thing'. Nevertheless, peak is very useful with things 
like percussion sounds because peak gives you a much better idea what's 
actually happening at that time, and a much better possibility of affecting it.

In essence, RMS is more "natural", but against it it will let the 
occasional peak that "should really have been treated" pass through to your 
project. Not always good. In analogue recording this is can be less of a 
problem because this can often lead to a 'nice' or desirable piece of 
distortion. When you go digital, however, this can be more of a headache. 
In the digital domain one really does want better control of peaks in which 
case RMS can be problematic. This is why many compressors these days even 
though they work in RMS are still fitted with peak detectors. Some 
compressors are peak but use RMS sensing ... some are RMS but use peak 
sensing ... and some combine the two in often weird ways, or else give you 
a genuine choice of switching between one and the other. Try to get a 
compressor that lets you do both ... usually possible if you're using a 
plug-in.

Rule of thumb ... peak for percussion, RMS for pretty much everything else. 
Break this rule often and you'll be a good mix engineer!!

I'm afraid ... you're not going to get the short and neat answer you're 
looking for because, really, there isn't one!!

HTH

Kool Musick
Keep Musick Kool


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