Compressors, can anyone explain what they do?
Moderated By: mods
-
johnnyseven
- .
- Posts: 3998
- Joined: Sun Jun 03, 2007 2:42 pm
- Location: London, England
Compressors, can anyone explain what they do?
I've never used a compressor, it seems that some people love them and some people hate them. However i've never really understood what they do and how they affect your sound. Do people on here use them and can anybody explain to me what they do?
Basically they quieten the louder part of your playing, so inherently make the quieter bits louder by comparison. The overall effect is a smoother, less "peaky" kinda sound, also making the notes sound like they sustain for longer.
Beloved of widdle merchants and funk/country players alike.
[youtube][/youtube]
Zip to 0:40 for some before/after action.
Beloved of widdle merchants and funk/country players alike.
[youtube][/youtube]
Zip to 0:40 for some before/after action.
My interpretation is a bit cock-eyed but you can google for a proper scientific explanation. Anyway, I see it two lights - either as an effect or a tool. Sometimes I think it'd be helpful if more people saw it this way too; it'd dispell a lot of arguments.
By effect I mean use it to add a different colour and change your tone into something else. With this I mean the squished sound where it sounds like your amp is about to burst and you also hear extra jangly swirls and other strange overtones you don't usually pick up. I think a dimed dynacomp with a clean Telecaster on youtube would give a good example of that.
By tool I mean use it to pad out and regulate the tone you already have. i.e get a more accurate string spread, fatten the sound up and give a more consistent volume. This is more about subtlety but it's really helpful to have.
Obviously there's middle ground between these two.
People often say that compressors take away your expression and dynamics but it's totally a double edged sword. I would argue that with compression I can bring out the sounds of playing very softly up to a more audible level, and also bring down fierce picking to the same level - something you wouldn't be able to do. Can, in a way make your playing more dynamic, or at least more textured.
The volume regulation is also great if you're not able to or don't want to focus on playing smoothly:
- singing
- dancing about
- generally being drunk
Hope this doesn't sound completely stupid but that's my own thoughts on it. I wouldn't pretend to say it's the truth.
By effect I mean use it to add a different colour and change your tone into something else. With this I mean the squished sound where it sounds like your amp is about to burst and you also hear extra jangly swirls and other strange overtones you don't usually pick up. I think a dimed dynacomp with a clean Telecaster on youtube would give a good example of that.
By tool I mean use it to pad out and regulate the tone you already have. i.e get a more accurate string spread, fatten the sound up and give a more consistent volume. This is more about subtlety but it's really helpful to have.
Obviously there's middle ground between these two.
People often say that compressors take away your expression and dynamics but it's totally a double edged sword. I would argue that with compression I can bring out the sounds of playing very softly up to a more audible level, and also bring down fierce picking to the same level - something you wouldn't be able to do. Can, in a way make your playing more dynamic, or at least more textured.
The volume regulation is also great if you're not able to or don't want to focus on playing smoothly:
- singing
- dancing about
- generally being drunk
Hope this doesn't sound completely stupid but that's my own thoughts on it. I wouldn't pretend to say it's the truth.
Did he really say "nice" after the first bit of compression playing. QualityDoog wrote:Basically they quieten the louder part of your playing, so inherently make the quieter bits louder by comparison. The overall effect is a smoother, less "peaky" kinda sound, also making the notes sound like they sustain for longer.
Beloved of widdle merchants and funk/country players alike.
[youtube][/youtube]
Zip to 0:40 for some before/after action.
plopswagon wrote:I like teles and strats because they're made out of guitar.
robroe wrote:I dont need a capo. I have the other chords in my tonefingers
-
endsjustifymeans
- Grown Up Punk
- Posts: 19442
- Joined: Tue Feb 10, 2009 4:02 pm
- Location: Ball So Hard University
One of the problems in understanding and using compression as a guitarist is that pedals are designed for a combination of simplicity and seemingly to cater for the stupidity of guitarists. For example a lot of pedals will have controls labelled things like 'squash' and 'sustain' which don't really mean very much in terms of what you're actually doing to the signal, they just attempt to describe the result.
With studio compressors things are almost always labelled in terms of what you are doing to signal. At first glance it seems more complex but it's actually a lot easier to understand what's going on, and easier to get more specific results. On the other hand it's not so 'grab and go' so adjusting things quickly isn't as easy which is another reason guitar focused compressors are so simple.
The line that starts in the bottom left and continues after the threshold mark as a feint line is a continuous representation of an increase audio signal. The input and output levels are matched, with no change to the signal. The idea with a compressor is that you set a threshold to define at what signal level the compressor becomes active and then a ratio to set the amount of dynamic reduction (compression). If you had the ratio set as 1:1, for every 1db that comes in it sends 1db out. at 10:1 you're only sending out 1db for every 10db increase above the threshold. For example if you had a threshold of -30db and your signal (for simplicity imagine a single hit like a snare drum) was at -20db (10db louder) then at 1:1 the output would be -20db, at 2:1 it would be -25db and at 10:1 it would be -29db. The numbers don't mean very much on their own, and it's a good idea to play with a software compressor to see what's going on.
Those are the main two functions of a compressor. But you also have attack which controls the amount of time between when the signal crosses the threshold and when the compressor kicks in. Having that set as 0ms can make it sound very unnatural as it will kick in immediately. Having it a little higher allows the initial transient to get through and can give you much of the benefit without making it sound like you have a compression pulling the dynamics down. Similarly you have release which controls how long the compressors stays active after the signal goes back below the threshold. A lot of software will have auto settings for attack and release which will move depending on the input signal and at the stage when you have a basic understanding of it that's probably a sensible setting to leave it on.
There are two more common settings. One is knee which changes the point at the threshold. A hard knee and it's a sharp angle. When signal goes above the threshold it's changed by the ratio in the way the graph shows. Wih a soft knee it's as though that angle is rounded. So as it crosses the threshold it's compressed slightly at first, and then moves towards the straight output line you see. Some compressors will have a simple hard/soft button and some will have a variable control for that. A lot won't have it at all but it's quite a common feature still. A softer knee can help make it sound more natural if your input material is around the level of the threshold. The final control is make-up gain which is a way to allow you to add volume back to the signal. Sometimes it will be called 'output level' but make-up gain is the standard name.
And here is an example of a software compressor which has all of those controls except knee. Pay attention only to the left, as the 2 and 3 parts are other forms of dynamic effect. Although it's worth mentioning that a limiter is nothing more than a simple compressor with a very high ratio. This is a default compressor that's built into Cubase and Nuendo.
On a guitar pedal, you sometimes see controls like 'attack' but I've never seen one say 'threshold' or similar. 'volume' would be 'make-up gain' and when they have an EA control that's independent of the compressor (it will likely be after the compressor because having it before would change the way your signal is compressed and although that can be useful in mixing it's not what guitarists are wanting on a pedal, so it's like having an EQ straight after the compressor that you turn on with it). And when it says 'comp.' or 'sustain' it's doing a mixture of threshold and ratio settings. It's difficult to guess as to exactly what, but if you understand how they work when used in recording then it's easier to understand the stripped down version.
Reading this page on the Diamond pedal it's all marketing nonsense other than that they tell you it's an optical compressor and the EQ comes after the compressor. It doesn't actually tell you anything about how it works.
With studio compressors things are almost always labelled in terms of what you are doing to signal. At first glance it seems more complex but it's actually a lot easier to understand what's going on, and easier to get more specific results. On the other hand it's not so 'grab and go' so adjusting things quickly isn't as easy which is another reason guitar focused compressors are so simple.
The line that starts in the bottom left and continues after the threshold mark as a feint line is a continuous representation of an increase audio signal. The input and output levels are matched, with no change to the signal. The idea with a compressor is that you set a threshold to define at what signal level the compressor becomes active and then a ratio to set the amount of dynamic reduction (compression). If you had the ratio set as 1:1, for every 1db that comes in it sends 1db out. at 10:1 you're only sending out 1db for every 10db increase above the threshold. For example if you had a threshold of -30db and your signal (for simplicity imagine a single hit like a snare drum) was at -20db (10db louder) then at 1:1 the output would be -20db, at 2:1 it would be -25db and at 10:1 it would be -29db. The numbers don't mean very much on their own, and it's a good idea to play with a software compressor to see what's going on.
Those are the main two functions of a compressor. But you also have attack which controls the amount of time between when the signal crosses the threshold and when the compressor kicks in. Having that set as 0ms can make it sound very unnatural as it will kick in immediately. Having it a little higher allows the initial transient to get through and can give you much of the benefit without making it sound like you have a compression pulling the dynamics down. Similarly you have release which controls how long the compressors stays active after the signal goes back below the threshold. A lot of software will have auto settings for attack and release which will move depending on the input signal and at the stage when you have a basic understanding of it that's probably a sensible setting to leave it on.
There are two more common settings. One is knee which changes the point at the threshold. A hard knee and it's a sharp angle. When signal goes above the threshold it's changed by the ratio in the way the graph shows. Wih a soft knee it's as though that angle is rounded. So as it crosses the threshold it's compressed slightly at first, and then moves towards the straight output line you see. Some compressors will have a simple hard/soft button and some will have a variable control for that. A lot won't have it at all but it's quite a common feature still. A softer knee can help make it sound more natural if your input material is around the level of the threshold. The final control is make-up gain which is a way to allow you to add volume back to the signal. Sometimes it will be called 'output level' but make-up gain is the standard name.
And here is an example of a software compressor which has all of those controls except knee. Pay attention only to the left, as the 2 and 3 parts are other forms of dynamic effect. Although it's worth mentioning that a limiter is nothing more than a simple compressor with a very high ratio. This is a default compressor that's built into Cubase and Nuendo.
On a guitar pedal, you sometimes see controls like 'attack' but I've never seen one say 'threshold' or similar. 'volume' would be 'make-up gain' and when they have an EA control that's independent of the compressor (it will likely be after the compressor because having it before would change the way your signal is compressed and although that can be useful in mixing it's not what guitarists are wanting on a pedal, so it's like having an EQ straight after the compressor that you turn on with it). And when it says 'comp.' or 'sustain' it's doing a mixture of threshold and ratio settings. It's difficult to guess as to exactly what, but if you understand how they work when used in recording then it's easier to understand the stripped down version.
Reading this page on the Diamond pedal it's all marketing nonsense other than that they tell you it's an optical compressor and the EQ comes after the compressor. It doesn't actually tell you anything about how it works.
Shabba.
I would add that it's not love or hate. Some people can make good use of it, for some it's not worth it all and a lot of people don't understand it. We're not talking about marmite here.
Also it's worth noting that almost all forms of distortion/overdrive include compression in how they work. They compress the signal in a very different way but the output signal is compressed.
Also it's worth noting that almost all forms of distortion/overdrive include compression in how they work. They compress the signal in a very different way but the output signal is compressed.
Shabba.
-
Fran
- The Curmudgeon
- Posts: 22219
- Joined: Thu Apr 20, 2006 5:53 am
- Location: Nottingham, Englandshire.
What a load of nonsense.James wrote:One of the problems in understanding and using compression as a guitarist is that pedals are designed for a combination of simplicity and seemingly to cater for the stupidity of guitarists. For example a lot of pedals will have controls labelled things like 'squash' and 'sustain' which don't really mean very much in terms of what you're actually doing to the signal, they just attempt to describe the result.
With studio compressors things are almost always labelled in terms of what you are doing to signal. At first glance it seems more complex but it's actually a lot easier to understand what's going on, and easier to get more specific results. On the other hand it's not so 'grab and go' so adjusting things quickly isn't as easy which is another reason guitar focused compressors are so simple.
The line that starts in the bottom left and continues after the threshold mark as a feint line is a continuous representation of an increase audio signal. The input and output levels are matched, with no change to the signal. The idea with a compressor is that you set a threshold to define at what signal level the compressor becomes active and then a ratio to set the amount of dynamic reduction (compression). If you had the ratio set as 1:1, for every 1db that comes in it sends 1db out. at 10:1 you're only sending out 1db for every 10db increase above the threshold. For example if you had a threshold of -30db and your signal (for simplicity imagine a single hit like a snare drum) was at -20db (10db louder) then at 1:1 the output would be -20db, at 2:1 it would be -25db and at 10:1 it would be -29db. The numbers don't mean very much on their own, and it's a good idea to play with a software compressor to see what's going on.
Those are the main two functions of a compressor. But you also have attack which controls the amount of time between when the signal crosses the threshold and when the compressor kicks in. Having that set as 0ms can make it sound very unnatural as it will kick in immediately. Having it a little higher allows the initial transient to get through and can give you much of the benefit without making it sound like you have a compression pulling the dynamics down. Similarly you have release which controls how long the compressors stays active after the signal goes back below the threshold. A lot of software will have auto settings for attack and release which will move depending on the input signal and at the stage when you have a basic understanding of it that's probably a sensible setting to leave it on.
There are two more common settings. One is knee which changes the point at the threshold. A hard knee and it's a sharp angle. When signal goes above the threshold it's changed by the ratio in the way the graph shows. Wih a soft knee it's as though that angle is rounded. So as it crosses the threshold it's compressed slightly at first, and then moves towards the straight output line you see. Some compressors will have a simple hard/soft button and some will have a variable control for that. A lot won't have it at all but it's quite a common feature still. A softer knee can help make it sound more natural if your input material is around the level of the threshold. The final control is make-up gain which is a way to allow you to add volume back to the signal. Sometimes it will be called 'output level' but make-up gain is the standard name.
And here is an example of a software compressor which has all of those controls except knee. Pay attention only to the left, as the 2 and 3 parts are other forms of dynamic effect. Although it's worth mentioning that a limiter is nothing more than a simple compressor with a very high ratio. This is a default compressor that's built into Cubase and Nuendo.
On a guitar pedal, you sometimes see controls like 'attack' but I've never seen one say 'threshold' or similar. 'volume' would be 'make-up gain' and when they have an EA control that's independent of the compressor (it will likely be after the compressor because having it before would change the way your signal is compressed and although that can be useful in mixing it's not what guitarists are wanting on a pedal, so it's like having an EQ straight after the compressor that you turn on with it). And when it says 'comp.' or 'sustain' it's doing a mixture of threshold and ratio settings. It's difficult to guess as to exactly what, but if you understand how they work when used in recording then it's easier to understand the stripped down version.
Reading this page on the Diamond pedal it's all marketing nonsense other than that they tell you it's an optical compressor and the EQ comes after the compressor. It doesn't actually tell you anything about how it works.
Im sticking with the Gilmour theory
-
Mike
- I like EL34s
- Posts: 39170
- Joined: Thu Apr 20, 2006 8:30 am
- Location: Edinburgh, Scotland
- Contact:
I adore the compression inherent in a good overdrive pedal or amplifier channel. That and the dynamic harmonic distortion content are what I find crucial in defining my taste.
I love the way Dyna Comps sound on guitar but I would only personally use it for clean work that needed some help to stick out in a live band mix
I love the way Dyna Comps sound on guitar but I would only personally use it for clean work that needed some help to stick out in a live band mix
-
johnnyseven
- .
- Posts: 3998
- Joined: Sun Jun 03, 2007 2:42 pm
- Location: London, England
If having a compressor will make me play like this dude I don't want one.Doog wrote:Basically they quieten the louder part of your playing, so inherently make the quieter bits louder by comparison. The overall effect is a smoother, less "peaky" kinda sound, also making the notes sound like they sustain for longer.
Beloved of widdle merchants and funk/country players alike.
[youtube][/youtube]
Zip to 0:40 for some before/after action.
-
dots
- BADmin (he/him)
- Posts: 1022402
- Joined: Mon Apr 17, 2006 4:16 pm
- Location: Esco-A-Go-Go
- Contact:
compression's just another one of those effects that got a bad rep from overuse. i've heard folks make claims like they'll never let compression touch their tracks ever, and it seems silly to me since any song on the radio gets compressed. i look at it like an essential weapon in your arsenal when called for; just don't take the "turn it on and walk away" approach, and you'll be fine. as for live situations, if i was willing to have a huge pedal board, i could see including compression/sustain in the chain. just depends on what kind of sound your going for. something really processed like a trent reznor tone, i don't see how you could get by without it, for example.
-
hotrodperlmutter
- crescent fresh
- Posts: 16665
- Joined: Sat Apr 04, 2009 10:29 pm
- Location: Overland Park, KS, USA
-
samuelcotterall
- .
- Posts: 99
- Joined: Sun May 09, 2010 10:03 pm
- Location: Manchester
Yeah, agreed. This is awesome and informative. More threads like this.hotrodperlmutter wrote:awesome/informative thread is awesome/informative.
Personally though, I don’t really use compression on my guitar unless I’m playing two-hand-tap math stuff just to even out the notes, but then I don’t do that very often because I’m not very good at it and my compressor pedal doesn’t have an LED on it.
However, I’ve found it sounds good between microphones and mixers, which I’m sure is technically wrong in a large number of ways.
It's only wrong in the sense that it's not adjustable after recording. A lot of engineers compress a little on the way to 'tape', be that digital or analogue. If you over-do it or just make a mistake with it you need to re-record the take. If you're happy that you've got it right, there's no harm in compressing as you record.samuelcotterall wrote:However, I’ve found it sounds good between microphones and mixers, which I’m sure is technically wrong in a large number of ways.
In a sense it can be better because it forces commitment. You can continuously tweak if you do it after and sometimes it's best to get it pretty close and be done with it than never get it finished because you can't get it spot on.
Shabba.
-
samuelcotterall
- .
- Posts: 99
- Joined: Sun May 09, 2010 10:03 pm
- Location: Manchester
Ah, I’m with you.James wrote:It's only wrong in the sense that it's not adjustable after recording. A lot of engineers compress a little on the way to 'tape', be that digital or analogue. If you over-do it or just make a mistake with it you need to re-record the take. If you're happy that you've got it right, there's no harm in compressing as you record.samuelcotterall wrote:However, I’ve found it sounds good between microphones and mixers, which I’m sure is technically wrong in a large number of ways.
In a sense it can be better because it forces commitment. You can continuously tweak if you do it after and sometimes it's best to get it pretty close and be done with it than never get it finished because you can't get it spot on.
I tend to treat reverb that way because it can be a pain when you’re mixing, but I think of my compressor as something to get a bit more out of the microphone. For me, I think a lot of recording stuff is mental, rather than technical.
Actually reverb is better after because it's easier to put everything in the same acoustic space that way. If you're going for a surf tone that's different as it's part of the inherent sound. If you're using reverb to give things some space and to make things meld together, you should be looking to do it after to be able to make it consistent for all instruments then.
The thing with this stuff is of course that it's always down to what you want. You might not want consistent reverb. You might also have a nice enough reverb sound in the room that you want to get that in which case it is of course right to record it straight off rather than add it after.
In general though, if you're adding reverb to close mic'd recordings it's best to do it when you mix.
Using a compressor between a mic and a mixer doesn't get anything extra out of the mic. It's pretty much the same as using one after. What it can do is protect against accidental clipping but that's a whole different kettle of fish.
The thing with this stuff is of course that it's always down to what you want. You might not want consistent reverb. You might also have a nice enough reverb sound in the room that you want to get that in which case it is of course right to record it straight off rather than add it after.
In general though, if you're adding reverb to close mic'd recordings it's best to do it when you mix.
Using a compressor between a mic and a mixer doesn't get anything extra out of the mic. It's pretty much the same as using one after. What it can do is protect against accidental clipping but that's a whole different kettle of fish.
Shabba.
-
samuelcotterall
- .
- Posts: 99
- Joined: Sun May 09, 2010 10:03 pm
- Location: Manchester
