What does compression really do?

[muzician21]

I've seen compression/compressors described as an "automatic volume/
gain control" but it seems that's not really accurate. That
characteristic compressed sound indicates there's more going on than
simply reducing the volume as if you were working a fader in real
time.

What do you consider to be an accurate description of what's happening
with compression?

[John Williamson]

muzician21 wrote:
> I've seen compression/compressors described as an "automatic volume/
> gain control" but it seems that's not really accurate. That
> characteristic compressed sound indicates there's more going on than
> simply reducing the volume as if you were working a fader in real
> time.
>
> What do you consider to be an accurate description of what's happening
> with compression?

It depends on the settings.

As used in most pop music, the channel gain is increased to the maximum,
so that the quietest parts of the track give full output, and the
compressor is used to keep the dynamic range as close to zero as
possible. Then they do the same with the final stereo output.

Or one can be used to subtly decrease the dynamic range of a symphony
orchestra so that the quiet bits can be heard over a car engine or room
air conditioner without deafening the neighbours with the loud bits. Or,
preferably in this case, the mastering engineer could either automate
the gain or ride it by hand.

If the channel gain is increased, it will bring the lower levels up
closer to the high level, while keeping the high levels down to the
maximum desired. If channel gain is low, then it can bring peaks down to
a sensible level, and depending on the settings, can work as a limiter.
Then, there is the effect of attack and release times.....


--
Tciao for Now!

John.

[Peter Larsen[_3_]]

muzician21 wrote:

> I've seen compression/compressors described as an "automatic volume/
> gain control" but it seems that's not really accurate. That
> characteristic compressed sound indicates there's more going on than
> simply reducing the volume as if you were working a fader in real
> time.

> What do you consider to be an accurate description of what's happening
> with compression?

"automated fader-operation" is called Automated Gain Control, AGC. To
properly describe it takes also rate of gain change and attack and release.

What describes a compressor is attack time, release time, ratio, strictly
speaking also input gain and output gain. Some designs use the expression
"threshold", but the same effects are possibly with suitable combination of
input gain and fixed threshold, so while actual electronics circuit
implementation may be different it is conceptually different wordings for
the same effect.

Simplified: "ratio" is an expression of how it redraws the entire signal
curve so that the excursion around the center-line gets smaller,
mathematically it is an expression of the steepness of the transfer
characteristic.

I often used multi-curved transfer characteristics when I draw my own
compressor in A3, in virtual hardware they are real easy to implement,
whether they were so in the software source-code is not my problem. The
example "saturator" is a simple version of such a curve, basically it is
just a "soft knee" at the top of the transfer-characteristic.

For well isolated studio-recorded vox (pop, but not only) a soft-knee
characterstic with the knee starting around -20 dB can be quite useful, it
is just not the type of recording I'm currently into doing.


kind regards

Peter Larsen

[yrret]

Its suppose to be a simple volume control. But many analog compressors had
great side effects of introducing subtle harmonics as a side effect to the
volume compression. A 'good' compressor is one that does these side effects
well. FYI an Exciter is a device that adds such effects without
compression.


"muzician21" > wrote in message
...
> I've seen compression/compressors described as an "automatic volume/
> gain control" but it seems that's not really accurate. That
> characteristic compressed sound indicates there's more going on than
> simply reducing the volume as if you were working a fader in real
> time.
>
> What do you consider to be an accurate description of what's happening
> with compression?
>

[Mike Rivers]

On 5/20/2011 4:09 AM, muzician21 wrote:
> I've seen compression/compressors described as an "automatic volume/
> gain control" but it seems that's not really accurate. That
> characteristic compressed sound indicates there's more going on than
> simply reducing the volume as if you were working a fader in real
> time.

A compressor has one primary function - to reduce dynamic range. It has
a lot of secondary mis-uses including adding a tube or transformer (for
those compressors that have them) in the signal path, adding some
distortion, or modifying the waveform envelope to exaggerate or suppress
the attack of a sound. If there's one thing that a compressor isn't,
it's an automatic volume control. There's another device that does that,
sort of, that works on a similar principle as a compressor, but with
very different time response.

There's an article on my web page entitled Compression Exposed that has
a good overview of what a compressor does and how it's used and abused.


--
"Today's production equipment is IT based and cannot be operated without
a passing knowledge of computing, although it seems that it can be
operated without a passing knowledge of audio" - John Watkinson

Drop by http://mikeriversaudio.wordpress.com now and then

[Scott Dorsey]

muzician21 > wrote:
>I've seen compression/compressors described as an "automatic volume/
>gain control" but it seems that's not really accurate. That
>characteristic compressed sound indicates there's more going on than
>simply reducing the volume as if you were working a fader in real
>time.

That's just because you can't work the fader fast enough. The compressor
can be set to smooth out peaks much faster than a human being, and this
causes a somewhat different effect (almost like the removal of tremolo).

>What do you consider to be an accurate description of what's happening
>with compression?

It's like amphetamine-crazed chihuahuas on the faders.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Les Cargill[_4_]]

muzician21 wrote:
> I've seen compression/compressors described as an "automatic volume/
> gain control" but it seems that's not really accurate.

AGC is a specialization of gain reduction.

> That
> characteristic compressed sound indicates there's more going on than
> simply reducing the volume as if you were working a fader in real
> time.
>

SFAIK, you cannot emulate a real human riding a fader in
real time.

> What do you consider to be an accurate description of what's happening
> with compression?

It's dynamic gain reduction.

--
Les Cargill

[Frank Stearns]

(Scott Dorsey) writes:

snips

>>What do you consider to be an accurate description of what's happening
>>with compression?

>It's like amphetamine-crazed chihuahuas on the faders.

Guffaw. Best "educational-humor visual" of comp I've heard.

Frank
Mobile Audio
--

[hank alrich]

yrret > wrote:

> Its suppose to be a simple volume control.

No, it's a tool for reducing dynamic range, something quite different
from a typical volume control.

> But many analog compressors had
> great side effects of introducing subtle harmonics as a side effect to the
> volume compression. A 'good' compressor is one that does these side effects
> well.

Some consider a good compressor to be one that reduces dynamic range
without introducing artifacts.

> FYI an Exciter is a device that adds such effects without
> compression.

Artifacts of compression and effects of excitation are generally
different.

> "muzician21" > wrote in message
> ...
> > I've seen compression/compressors described as an "automatic volume/
> > gain control" but it seems that's not really accurate. That
> > characteristic compressed sound indicates there's more going on than
> > simply reducing the volume as if you were working a fader in real
> > time.
> >
> > What do you consider to be an accurate description of what's happening
> > with compression?
> >


--
shut up and play your guitar * http://hankalrich.com/
http://www.youtube.com/watch?v=NpqXcV9DYAc
http://www.sonicbids.com/HankandShaidri

[david correia]

In article >, "yrret" >
wrote:

> Its suppose to be a simple volume control. But many analog compressors had
> great side effects of introducing subtle harmonics as a side effect to the
> volume compression. A 'good' compressor is one that does these side effects
> well. FYI an Exciter is a device that adds such effects without
> compression.
>
>
> "muzician21" > wrote in message
> ...
> > I've seen compression/compressors described as an "automatic volume/
> > gain control" but it seems that's not really accurate. That
> > characteristic compressed sound indicates there's more going on than
> > simply reducing the volume as if you were working a fader in real
> > time.
> >
> > What do you consider to be an accurate description of what's happening
> > with compression?
> >



I wouldn't compare an exciter with a compressor, but I def concur that
pros use compressors for more than just taming dynamic range. We use
really good compressors to **** with sound.

It's why LA2's & 3's and 1176's are still being manufactured. They are
indispensable for shaping sounds. When you get it right, them boys can
hump your ear.





David Correia
www.Celebrationsound.com