Compression Effects

[Christophe Stoelinga]

On Mon, Nov 29, 1999 at 12:57:49PM +0000, Plinio Barraza wrote:
> My ears definitely told me there was a light 
> boxy effect going on.  I guess that one could 
> think that the high frequencies riding ontop of 
> a compressed wave could get attenuated.  But one 
> could also think that the wave is getting more 
> square (steeper rise flatter top -especially in 
> a limiter kind of compression) and therefore it 
> would make the overall sound brighter.
> 
> Perhaps the filtering that takes place is 
> different for the type of compression used 
> (threshhold, attack, decay), cause if the attack 
> is slow I guess there is no reason for the 
> actual sound to change)

I don't know if this is exactly what hou mean, but if the attack time is set
too fast there is some harmonic distortion due to the fact that the VCA
changes gain during one period of the wave that is compressed. This is not
normal operation for a compressor. 
In a recent issue of the journal of the AES there was an article on
compressors where the HD was calculated in a compressor with an attack time
of 100us. 
normally the attacktime is at least a few ms so the compression ratio
changes only very little during a period of an (not too low) frequentie.
A little confusing is the talk I heart in a local music store where someone
said digital compressors can be set to attack time zero. I don't know what
they mean by it in technical terms, but a ordinary analog compressor with
attack and decay of zero would make an (probably bad sounding) distortion
effect (like the guitar stomp boxes).

hope this clears out a bit,
Christophe