Jump to content
knight46

Compression

Recommended Posts

I have read several references in post, most recently by Les (Lcjones) and Steve (Solidwalnut) about compression in the final mix. After searching the site I can't find any information on this subject.

My questions are:

Exactally what is Compression?

How does it work?

What does it do for your recording?

What are some good compressors?

I would appreciate any information on this subject.

Thanks

Eddie (Knight46)

Share this post


Link to post
Share on other sites

In a very simplistic view, compression takes the peaks and valleys of a sound signal and compresses them together. Not length of the signal but the height of the signal.

I know, Steve and Rocker Bob have a good grasp of compression and I'm sure can explain better than I. It's a complicated subject.

I'll post some examples of how compression can alter a vocal or an instrument track later tonight.

**

LC

Share this post


Link to post
Share on other sites

Les,

Thanks for the reply. I have been doing some online research and I read the technical "stuff", just didn't know the practical. I would much appreciate the examples and some reasoning why you use compression.

Share this post


Link to post
Share on other sites

Compression controls the changes in amplitude (volume). It does not remove the dynamics, but it can reduce changes in amplitude by a ratio. Compressors do nothing to the signal until the "threshold" level is reached. Once the level goes over the threshold, changes in level will be reduced by the ratio. If a 2:1 ratio is set and the input level goes 10dB over the threshold, the output will be 5 dB over the threshold. Then there's the attack, which is the speed at which the compressor reacts when the threshold is reached. Slower attacks can be smoother but will not catch the initial transients - sometimes a good thing.

Now go spend forever playing with a compressor. :D

Share this post


Link to post
Share on other sites

My limited experience of compression is in using the Audacity fx. In Audacity it's a sliding lever so the attack etc is automatically included by the software. My ears are not very high tech, but it seems to even out the sound a bit, but the practical result in the recording is to increase the volume of the whole track. Tekker gave an explanation of why this is so - I think it's because by cutting off the extremes of sound, it allows the rest (middle) to expand.....or something like that.

I started using it after rockerbob gave it a rave, but I found that with a lot of compression I got some distortion, so I usually experiment a bit and try to go with what sounds best to me. Sometimes you have to reduce the track volume after compression to avoid clipping.

Share this post


Link to post
Share on other sites

I hope this helps in some way.......

Compression. Ack! There 's so much going on with compression. Attack and release, threshold, knee, ratio. But of course, the coolest thing about having a rack mount compressor is all the groovy and flashing lights! ;) If you're reading up on compression then you're already heads up on the terminology.

There is no exact science to using a compressor. Each song, well, what I should say is, each track on each song will have it's own life in compression. And even at that, each instrument or vocal will dictate whether compression is needed and if so, the dynamics of the vocal/instrument itself will help guide the engineer to the proper amount of compression. The last thing you want to do is "over-compress". As in all things in recording, less is more. Especially in compression and EQ. Once you get into using compression you become familiar with terms like "squashed". So you don't want to "squash" your signal unless you have a real purpose in doing so.

I've included a screen shot of what compression looks like on a wav sound signal in an audio editor. It should be familiar to most home recordists. I'm not going to give or say what settings I used on the compressor. Primarily because they are only useful on these tracks and the instrument I used. If you have a compressor and are just learning to use it, I'll list a starting point a bit later.

'

A little bit of side information. I'm using all outboard equipment. I am not using software plugin's to build the tone(s) I want. It's merely my preference to do so. There is nothing wrong with using plugins. The disadvantage of doing it "my way" is that if I screw up, I have to start all over again. But that's OK. It makes me learn from my mistakes. The bonus to using plugins, is they allow you to try multiple scenarios on a track without having to do it all over again.

Now to the tracks........

I set the volume (trim and gain) to what I wanted. The volume is identical on each track. The only changes are the addition of the compressor and a JangleBox. (which I"ll explain later)

Track 1:

This is a dry track. I plugged in my guitar to a DI box which is then plugged into a channel on my mixer. By DRY, I mean there are no effects being asserted on the signal. It's simply a dry signal from guitar to the audio editor.

Track 2:

This is the same dry signal, now with the addition of a slight amount of compression.

Track 3:

On this track, I plugged my guitar in to an effects pedal called the JangleBox. The JB is then plugged into the DI and on into the mixer.

The JangleBox is, in effect, a compressor unto itself. It was created to emulate the original sounds of Roger McGuinn's (The Byrds) Rickenbacker 12 string. Which is another story altogether.

Track 4:

This is the tone I like the best. I am not only going through the JangleBox, I am also running compression on top of it all. So basically, I am double compressing the signal. (Highly desirable for me)

****

If you listen to the tracks below you may hear a subtle difference in sound with and without using a compressor. As you're listening notice the actual wav (signal) track associated with the sound track. Notice the peaks and valleys of loudness/softness and how, when compression is asserted on the signal the peak and valleys are "compressed" together.

Two things happen when using compression. 1) You level the sound field of the signal. 2) You also change the tone of the signal. The tone changing is arguable by some with a more commercial/professional sense of recording. But I'm here to tell you, using compression can be a great effects tool for tone as well as a tool to manage the dynamics of a signal.

For me, having a hardware compressor is a great thing. I love playing with it. It's a complicated device and for it to be useful in my recording studio requires me to have lots of restraint so that I don't "over-compress." Over-compressing will suck the life out of your recording. If you are interested in getting a hardware compressor for your studio, I would recommend using several different Compressor Plugin's before spending the money. Plugins will give you a real idea of how using hardware compression can affect your own sound signals. There are a number of free ones available on the Net and really, your audio editor probably has one installed already.

As promised earlier. Below are starting points for using a compressor. These settings have been suggested to me by more accomplished recording engineers than I. (and I am NOT a recording engineer) These are useful starting points for using compression. But as I mentioned earlier, each instrument on each track of each song is going to be dynamically different which may or may not require different compressor configurations.

(you may need to open your browser full screen)

copy and paste this into your favorite text editor and save it)

SOURCE..........|ATTACK.........|RELEASE........|RATIO.........|KNEE...........|GAIN REDUCTION

Vocal (normal)..25-100 ms.......100-500ms/Auto..2:1-8:1.........Soft............3-8dB

Vocal (rock)....25-100 ms.......100-300ms.......4:1-10:1........Hard............5-15dB

Acou guitar.....5-10ms/100-500..100-500ms/Auto..4-10:1..........Medium..........5-12dB

Elec guitar.....2-5ms...........500ms/Auto......8:1.............Hard............5-15dB

Elec (raging)...25 ms...........1-2 sec.........4-8:1...........Hard............5-15dB

Drum Kick/snare.1-25ms..........25-200ms/Auto...4-10:1..........Hard............5-15 dB

Drums (Cymbals).25 ms...........1-2 sec.........2-10:1..........Hard............5-15 dB

Bass............2-10ms..........500ms/Auto......4-12:1..........Hard............5-15dB

Bass (Clicky)...25 ms...........25 ms...........4-12:1..........Hard............5-15dB

Bass (Mushy)....100-500 ms......100-500 ms......4:1.............Hard............5-15dB

Brass (Horns)...1-25ms..........25-300ms/Auto...5-15:1..........Hard............8-15dB

Mixes Fast......None............400ms/Auto......2-6:1...........Soft............2-10dB(Stereo Link On)

General Fast....None............500ms/Auto......5:1.............Soft............10dB

Very low frequencies (below 20 Hz) present in a soundfile should be removed before compression; this makes it easier to monitor audible frequencies while compressing.

7842.attach

dry-1.mp3

dry-2-with-comp.mp3

dry-3-janglebox.mp3

dry-4-janglebox-with-comp.mp3

Share this post


Link to post
Share on other sites

Thanks to you all for your responses. I believe that I am beginning to understand both the technical and practical side of this.

Les those examples were very informative and the chart I believe will be most helpful. A couple of other questions if I might. Do you compress each file or only the final mix. In other words if you are recording an acoustic guitar track and then a vocal, would you compress the guitar and vocal individually or would you compress the mix.

Thanks again to Les, RB, Carol and Kenny for your responses. It is always a pleasure.

Share this post


Link to post
Share on other sites

Compression....

I think the best way to attack compression, no pun intended, is to use it on individual tracks. The key element in getting a good mix is getting all tracks to live in their own sound space complimentary to each other. If, for example you are working with an 8 track tune, and you mix them all down to one and then compress, you are compressing every single nuance of every single track of the song. You will be more likely than not to create a lifeless track that has no dynamics in it at all. If, on the other hand, you compress only what needs (or wants) to be compressed on individual tracks, then in the final mix down you have a dynamic sound scape.

As far as vocals and an acoustic guitar goes, if they are recorded on separate tracks, then you will want to compress each track individually. The vocals may require more or different compression than the guitar, therefore you wouldn't want to compress the entire song for fear of displacing the guitar or the vocals. And that's even "IF" they need compression at all. Just because you have a compressor, plugin or otherwise, doesn't mean you have to or even need to use it. ;)

The more I record, the more I realize the most critical piece of the puzzle is the incoming signal. BEFORE it hits any whirly-gig, big bamboozle, latest technology, wham-bam thank ya maam, thing-a-ma- jiggy gets their electrons on it. Let's face the facts, no matter how good the mix is, if you mixed a bad signal, all you're going to get is a bad signal.

:)

**

LC

Share this post


Link to post
Share on other sites

Excellent thread Knight, great replies, LC you put in a lot of work on your reply thanks for that, made things a whole lot clearer, takes a while to get the hang of effects and to know where to draw the line, things though work a lot better on seperate tracks, still a long way to go with recording but the better results in my limited experience have come from splitting everything up:winkthumb:

Share this post


Link to post
Share on other sites
Again thanks for the informative reply Les. As Chris said, all of your replies are thoughtful and very helpful. Thanks...:yes:

By the way, I use the Behringer MDX1600 Compressor/Expander/Gate device. To chain the comp in, I use a TRS Y cable from the comp to an INSERT port on a mixer channel.

**

Share this post


Link to post
Share on other sites
When a TV commercial "seems" louder than the program your watching is it compression they are using?

That's usually a limiter, like a compressor but with a hard line limit at the top beyond which nothing passes.

Share this post


Link to post
Share on other sites
When a TV commercial "seems" louder than the program your watching is it compression they are using?

Which is very irritating and causes me to hate television. This is a problem prevalent in home recordings and a direct result of mastering. Of which I'm the first guilty party in line.

A very good point is brought up here by 6 string. The point is, if you create a 10 song record (CD or MP3 downloads), you have to absolutely make sure each track (recording) is equal in volume level. What you don't want to happen is one song be louder than another. You don't want listeners to have to "turn down" or "turn up" the volume on their player, whether it's an iPod, a car CD player or mega bucks shelf stereo, in order to comfortably listen to your music. This is where mastering your music comes into play.

I can't give you the answers to mastering your music. Suffice it to say, that if track 4 is 20% louder than tracks 3 and 5, you have a mastering problem that needs fixing.

**

LC

Share this post


Link to post
Share on other sites
I guess there's an idustry standard, so DJ's don't have to work too hard having to adjust the volume for each song broadcasted.

DJ's don't have to worry about anything except que-ing up the next "paid" advertisement.

It's a moot point as far as disc jockeys are concerned. DJ's are not engineers. They are noise makers. In today's world of "radio" it's all computerized, unlike the days of Wolf Man Jack when a DJ had to actually work and earn a living.

Stations have limiters, as Rocker mentioned, that nail down the volume to it's most compatible sound level.

**

LC

Share this post


Link to post
Share on other sites

From what I've gathered{hopefully someone will correct me as needed} you compress the loudest signals, so that after compression you can increase the overall volume{gain} without exceeding the dynamic limits of the medium it's to be played back through...hence the TV/Commercial differences, the advertisers compress to the max they're allowed, set by the broadcasters. Most shows are based around soft dialog and are comparatively uncompressed by the TV stations, which causes the vast difference in percieved levels.

So if I'm absorbing this correctly, for us who record, we would use compression to bring up the volume of a soft{quieter} track to better 'sit in the mix' without effecting the over-all dynamics of said track? Am I getting it?

Share this post


Link to post
Share on other sites

One of the ideas of compression, for example, is taking a singer who's vocals are very dynamic, i.e., very loud and very soft, and adjusting the vocal so that when listening, you as a listener, can hear the entire level of that singers vocal, loudness and softness.

Soft tones, especially, of a singer can get lost in the mix. Using compression can alleviate, to a degree, the lost dynamics of a singers voice. As well, any instrument. "IF" used correctly.

**

Share this post


Link to post
Share on other sites
Which is very irritating and causes me to hate television. This is a problem prevalent in home recordings....

...and in most current commercial/professional recordings as well, it seems. I never really noticed until it was pointed out and I started really listening for it. The dynamics are so squashed that everything is at basically the same sound level, from what should be the quietest to the loudest parts of songs. As with anything, there *can* be "too much of a good thing".

Share this post


Link to post
Share on other sites

Les that's some really great info and detail!

As Les is saying, audio editing compression can be used as either an effect or a swiss army knife for a track or tracks. It can used to alter the sound altogether or be there to lightly compress large transients (sound spikes) and so to smooth out the overall level, or to compress the audio.

When you really compress the audio, the basic thing that happens is that you are compressing the transients (while exponentially sucking the dynamic level out of the audio). Which can be both good and bad. This is why there's no one setting for any audio. You can use it for both good and evil :)

The most basic of use of compression in recording are to compress transients to a varying degree, for three main purposes: 1, to keep a track from becoming to loud at certain points and 2, to augment the apparent volume level of a track and 3, to use compression as an effect.

First let's get into how to hear and to adjust a compressor!

ARRT

When done in this order, you'll be able to have control! First let audio into the compressor. This can be done by either turning up the input or by roughly adjusting the Threshold to sensitive. Not only will you have control, you'll be able to set it and forget it. Or at least set it and know what you have.

The first thing to do is to set the Attack anywhere. Next, adjust the Release to be the fastest. Next, adjust the Ratio to infinity or to the highest setting, 20:1 or higher.

Ready??!

  • Adjust the Attack so the compression hits the portion of audio you want it to hit. You may want to let softer passages through and have the compressor only be there for louder passages, for example. But listen to the front edge of the music and decide how and when you want to have the compressor hit it.
  • Adjust the Release so that the effected audio is held in compression for as long as you desire. This is where you're going to hear 'pumping' as the compressor releases. Adjust the 'pumping' so it is somehow within the groove of the song. Maybe it's timed to release with every kick. Maybe every measure. You decide what is the most musical sounding.
  • Adjust the Ratio so you get just the amount of compression you desire. As Rockerbob said, the ratio thing is all about math where the ratio says that if you put in x amount of audio into the compressor, x amount will be come out of it. At 2:1, if you put in 10dB, 5dB will be output. (This is where Make-up Gain can come in. We'll get to that).
  • Now that you have the compressor doing it's thing like you want, adjust the Threshold so that you let in the proper amount of audio to be effected. It's an over-all throttle of the compression unit, you might say. Or as Rockerbob says, any amount under the threshhold is what will be compressed; any amount over bypasses the compressor.

Make-up Gain. On most compressors, you can set Make-up Gain to either automatic or to manual. When you set it to automatic, program will amplify the output of the compressed audio in proportion to the gain lost by compression. You can set it to manual to set the level yourself, either higher or lower than what auto would give you. Salt to taste.

Subtle adjustments are huge! When you can control a comp like you want, it can really make a difference in your recording/mix.

Uses

As has been said in other posts and here, there are three main uses for this when we record and mix: Use it to level peaks in the audio, to use a compressed track as support for the main mix and to use compressed audio as an effect.

Using it to level peaks in the audio is apparent, and is by far the main way compression is used when we record and mix. But the next two are tricks that are used.

Compression as Support

One of the ways is how to beef up a kick drum and/or a bass guitar. The trick is to split the playback signal and send it to another track (or just buss the audio to another track). You do the regular EQ'ing on the main track, but on the second track, you really squish (compress) the track to taste. Now you send this signal to the main mix sneaking it behind the main kick track in volume level.

This same trick can be done to any track, but often you save this for the most important ones (kick, bass, lead vocal usually).

Compression as an Effect

Same trick, only you compress the second track so it sounds as funky as your heart desires. Add EQ and whatever else. Let's say you want to give the kick a good spiky sound. So you squish this second track but you also add a high EQ to the sound so it really clicks. Sneak this in so you really don't hear the click.

The possiblities are really endless! Here's a great site with tutorials on recording and engineering techniques: Tweakheadz Guide to the Home and Project Studio. Or, here's a link there specifically about compressors.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now


  • Recently Browsing   0 members

    No registered users viewing this page.

×