Compression is the primary in a family of audio signal processing called dynamics (the others are limiters, and gates). What a compressor does is reduce the transient spikes of an audio signal to keep them below a user defined limit. Basically it smooths out the spike in the attack of a sound, think of it like a block of clay on a potters wheel. rather than changing the whole shape, it just rounds off the sharp corners.

Typically there are four controls for a full compressor: threshold, ratio, make up (or gain) and knee.

The threshold is the limit that you set where you want the compressor to start working, say -10db.

The ratio is the amount of compression, expressed as a ratio like 2:1, or 3.5:1. This is probably the most important of the controls. The way it works is, once you set the threshold, the compressor will reduce the audio signal that goes over the threshold limit by the amount that you set as the ratio. If you set a threshold of -6db and a ratio of 3:1, then any audio signal that goes over that level will will be reduced by 3db for every 1db that it goes over -6. 4:1 is heavier compression, 2:1 is lighter compression.

The knee control adjust how drastic the compressor takes effect.

Gain will boost the signal to make up the loss in overall volume caused by compression.

Two other important controls are the attack time and release time, measured in milliseconds. The shorter the attack time, the faster the compressor will take effect. The longer the release time the longer it will hold the level down.

One other more sophisticated type of compressor is the multiband compressor, which will act on a certain frequency, or group of designated frequencies, but leave the other unaffected.

For a saxophone I will usually use a compressor if it's anything close to being close miced. Because the saxophone can often have a sharp, hard attack, it sometimes help in reducing that "bite you head off" affect of some styles of playing. Though I almost never go above 3 or 3.5:1 compression.

If you are performing with a sound engineer mixing the band from a console, he or she will usually have a rack of compressors next to the sound console. They will insert it on your channel if they think it needs it.

Compression isn't something that changes or colors the sound. But it can make just about any source easier to listen to by reducing some of the harsh spikes in sound that the microphone/speaker combination can introduce. If you don't know how to use it, it's probably better to not bring it to a first gig. If you don't set it right, it will prevent any sound from coming through your channel. But if you plan on using other effects already, then it maybe something to look into. But find out if the FOH mix position alrady has compressors available. Most do.

Dynamics Processing

Dynamic Range Compression

 

On another post, I read that a sax player gets a particular sound from " compression. " Is this an electronic device? Is it separate from the PA, sound board? I use a clip-on mike with an FM receiver. Is it something I can turn on/off as I please? I also plan on using
a Q Tron filter for a Wah Wah, preset.

Yes, it is an electronic device that goes somewhere in your signal path between the microphone and the PA. Whether it should go before or after the Wah is best left to experimentation (though I'm suspecting it will be after). It is something you can turn on and off, or change the intensity of.

If you want all the dirty details, read this.

 

First of all, compression is older that you are. All the way back to triode tube circuits. I am an audio engineer, have been the better part of 30 odd years. I'm going to make it simple. Compression makes loud things softer and soft things louder, within an envelope set by the user. Big dynamic range is a pain in the asss. Whenever you get po'd when the commercials come on while watching TV and blast you out of your chair, or you can't hear the softer dialogue during a movie you will understand why you need compression. Be aware that it (compression) will also raise the volume of any extraneous noise in the room, during the "making soft things louder stage" Also compression does in fact have the tendency to change the "color" of a signal often in a much desirable manner! There are devices such as the UA LA2A, and Urie 1176 that are much desired and emulated just for that reason! There it is in a nutshell, compression 101. Within this simple explanation there are nuances, but basically it is all you need to know.

 

OK, Thanks...but maybe this is not what I need to know. What I am looking for is....what is it that makes a sax sound like...say Sanborn? He sounds totally different when he is not playing through a sound system, but just " live ". Not that I want to sound like him, but I would like the slightly edgy, raspy sound without having to " growl " my sax all night...get what I mean? Is this done electronically? I guess I had the wrong idea of what the compressor does.

 

That full bodied sax sound that you are hearing on records! Sanborn, Gato, etc?

It is in fact a combination of compression (for compressions sake), compression for *COLOR*, proper EQ to eliminate problems or give a gentle boost to a desired frequency, and the judicious application of effects for spatial depth.

That subtle "fatness"? Mostly it is (in fact) is the compression (as *color* yes!) that you are hearing!

For compression very, very much effects a perceived EQ curve.

See? Your first instincts are in fact, right on the mark!

Compression, properly used, has both the effect of warming the signal and providing "bite" all at the same time. We pay some ridiculous amounts of money for some of these vintage devices! A Fairchild 670 comes to mind! I threw one out in the 80's before I fully understood what a comp really does! Now.......Doh!

Most people if they are not in this biz, or if they are purely FOH people completely misunderstand the power of a compressor. In addition to limiting dynamic range. A properly designed compressor is a a device for adding edge, depth and best of all *color!* to a signal.

The ear itself does a fair amount of what could be construed as "compression" all on it's own. Little wonder that compressors are so very, very useful and misunderstood.

A fair amount of understanding this phenomena can be gleaned from a decent comprehension of the Fletcher Munson curves and their effect upon perception of frequency bands in the human ear. Just Google it and have fun

A good way to think of a normal signal chain is:
1. EQ to correct problems,
2. Compression to decrease dynamic range and add dynamic color.
3 Effects to eliminate a "too dry sound", and then perhaps further applied to add depth or even FFFFXXXX!

Be very aware that much of this is a nuance that the record buyer or in a lessor respect people in the front of house (FOH) will hear. It is very hard to achieve this amount of a "real time" perception of nuance through stage monitors live, as you are performing. The fact is a that a sax is very loud on it's own!

Unless you are playing a *huge* venue the people FOH are very likely hearing a combo of both the acoustic sound of the sax, plus the sound of the horn processed (as earlier outlined) through the PA. You will very likely never hear the sound you hear on a CD or even the sound that people hear out front, on stage from a monitor in a live performance.

 

I think as a sax performer you can mostly ignore compression. Let the engineer use it if he wants.

Compression will nullify some of your volume dynamics and only the sound engineer will know if that is desirable in any specific venue. I've even tried the guitar FX style compressor peddles (singing sustain...) and it just seems worthless to me on sax. (for recording a compressor can be nice but that's different)

If you are honing up on some cool electronic effects, forget compression. Get a delay that you can leave on a single fast echo (slap back echo), which can come in real handy for certain solos, and maybe even another delay with a 3 note longer repeat, especially if you can tap the delay timing with your foot so you can easily set it to quarter note triplets. You already have a Qtron for the buck-buck chicken sound. Cool, but keep it minimal. Try some kind of harmonic doubler, or harmonizer. The most useful that I like is to double yourself an octave lower to give you a synthetic horn section sound, which works best for funk fill lines. Chorus / flanger can be interesting too (new wave?), but all these should be used minimally at most. Ring modulator can give you a robot sound.

That's my story and I'm stuck to it!

 

To SLAB HARDCHEEZ Yes, the added effects you mentioned are just what I was thinking, too. Of course to a minimum, just something to break it up a little.

To CPETE OK, that's what I thought.

Thanks to everyone! Great to have a forum like this, very helpful!

 

JC- Very concise and helpful explanation- thanks.

 

Reverb can help make a sax sound better, either live or on a recording.
The other effect I use to help give a more robust sax sound on a rock-and-roll song is "delay". I don't have a good explanation of how that works, just that it helps fatten up the sound a bit.

If you want a great rock-and-roll sax sound like Clarence Clemons on tenor or Sanborn on the alto, I don't think compression is going to help you much there.

The type of sound in our covers band for which we use compression the most is the vocallists, both lead and backup. I am not sure of this, but I think that most of the vocals you hear in recorded pop music have some degree of compression applied to them.

 

I have an offer from some young lions to join them, doing a feature set of Jazz/Funk/Blues. I would like to modernize my sound. I normally play tenor, but will be playing a lot of alto in this. On another post, I read that a sax player gets a particular sound from "compression." Is this an electronic device? Is it separate from the PA, sound board? I use a clip-on mike with an FM receiver. Is it something I can turn on/off as I please? I also plan on using a Q Tron filter for a Wah Wah, preset.

Others have summarized compression very well. I think it's not useful for you to think about it at this point. You use a clip-on mic with a receiver. What do you send the signal to from the receiver? To a PA system? Into an amp or powered speaker? The system you're sending the signal into can have a lot more effect on the sound than any efx box. Crappy sound equipment will give you crappy sound; to get good sound you need good sound gear.

I would recommend first that you have a good reverb/delay. That can either be in a separate EFX box or in the amp/PA that your sound is going through. That's why it's relevant what kind of system you're going through and who is operating it. In a live setting, the delay will do more to get a modern sound than compression.

You also have a Q Tron Wah. That sounds like enough effects for the time being. Don't get too far ahead of yourself or you'll get run over by your own gear.

There are extensive threads on this site about getting the desired tone, such as the "slightly edgy, raspy sound" you want. Much of the discussions have to do with mouthpieces and reed set-ups (and horns too), much more than electronics. Do a quick search on "bright mouthpiece" and see what comes up. You'll see what I mean.

 

Compression "compresses" the volume fluctuation in decibels of your sound being pprocessed. A 4:1 ratio compression will mean that for every 4 decibls louder you get, it squashes that increas to only 1 decible. So a loud note played 20 decibels louder will be percieved as only 5 decibles louder. This can be important in recording, but can especially be important with less abled sound guys who cant seem to know what t do with sax levels in the first place, or you are simply way down thge chain of importance of things for him to worry about.
Many sound guys mix from the top down, and place your sound level at a level where your loudest note can't be too loud (in relation to the band mix - in his opinion). However, at this safe level, much of the time this means your quiet or more legato notes are lost. So by "compressing" the amount of decible increase (I like 4:1 live) this allows your loudest note to actually not be that much louder than quieter notes, and since your are mixed by your loudest not potential, your quiet notes are percieved as significantly louder with more body than without the compression.

 

Compression is like having a really fast finger on a fader, pulling it down rapidly when a certain 'threshold' is sensed, and then moving that fader back up--hopefully so that it will sound invisible.

Pete is correct about using compression for coloration--but it takes a trained, experienced ear to get it right--and, like many other things in audio, very easy to get wrong.

 

It's best to use no compression unless you have a very expensive one and know how to use it and don't use too much of it.