Compression..............

[fingers23]

So...........apart from levelling out the sound level on multi tracks........what does it do?

Can I use it on single tracks and for what benefit?

What is its best use?

Sorry Guys....so many questions and thanks for the replies so far to my posts

[cacibi]

For recording purposes it can be used to control or flatten the transient peaks of particularly dynamic instruments such as vocals, snare drum, etc. A little will go a long way and lately I prefer to use it sparingly and set the input level lower. It seems like most people like to optimize the input so that it peaks at 0db, but I've been reading some advice to the contrary lately which indicates (especially for digital recording) that -12db on the track and -6db on the master will allow for more options in the mastering stage.

At any rate - don't want to stray too far off the topic. Some mild compression on vocals and drums will help get an even performance that is easier to mix and less likely to clip - possibly causing distortion in the track.

[Leon Grizzard]

It may be a good idea to bounce your existing vocal track to another open track, using fader moves to smooth out the big lumps first, then a little mild compression. Don't try to do it all with compression.

[cacibi]

Originally Posted by Leon Grizzard

It may be a good idea to bounce your existing vocal track to another open track, using fader moves to smooth out the big lumps first, then a little mild compression. Don't try to do it all with compression.

A very good point.

[fingers23]

Originally Posted by cacibi

At any rate - don't want to stray too far off the topic. Some mild compression on vocals and drums will help get an even performance that is easier to mix and less likely to clip - possibly causing distortion in the track.

Please...please....do stray off topic as the more info the better.

I am soaking it all up.............

Many thanks

[cacibi]

I have a couple books that I've used with great results:

Creative Recording 1: Effects & Processing by Paul White

The other title escapes me, will look it up and post when I get home. But it covers some useful basic information on acoustics, mic placement, etc.

These books helped give me useful starting points for compressor settings, etc. I would personally start gathering your information in the logical progression of tracking/engineering then move on to mixing and mastering techniques - assuming you want to do those yourself as well. They are all completely different animals IMHO - but it's like building a house, if the foundation ain't good.........it don't matter what you build on it!

[fingers23]

Thanks for the details of the Paul White book.

A quick Yahoo, the missus always loves that, and I found this treasure trove http://www.amgard.net/PWWrkShops.htm

I think I will be spending many a happy hour here...........

[StuH]

Here's a really good piece of text from Gareth, an audio eng. from FXpansion from a post at FXpansion, and goes in detail on Cacibi's blurb on recording levels. Hey you siad off topic was fine so here it is:

"In modern 24 bit recordings with reasonable convertors its not so important to record at high levels all of the time. The reason why most people mention -12dbfs is so that when you mix (sum) multiple channels together your main bus outputs will stay at close to 0dbfs. Again there are many arguments for and against this with many people arguing that in a modern 32 bit float based you dont need to concern your self with clipping.

This is in my opinion nonsense. It pays good sonic dividends to keep a reasonable signal level throughout your system and to not run busses or master outputs at very high levels. What I state in the mixing guide is this. When mixing you want to have the faders physically or virtually as close to 0db as possible as this is the area that affords the most resolution on the fader allowing you to achieve the subtle movements often needed.

It is not good practice imo to run the channels hot and to then trim these using a fader so that your working in the low resolution area of the fader. Remember faders have a log response. Likewise it is also not good practice to use a limiter to keep the levels in check.

It IS good practice to mix withing the headroom of the system , this means keeping your individual channel levels at around -12/15/18 dbfs this is still a nice healthy 22//21 bit resolution.

Likewise when recording acoustic instruments like a drum kit I record the kit with a 'natural' level that is , things like kick drums and snares will be recorded with a higher level than something like a hihat. Again the idea being that when I monitor or play it back my faders kick , snare , hat etc are all around the 0db mark and kit will play back with a nice 'self' mixed level.

There is no need again to record a hihat at 0db or anywhere that loud.

Again mic pres unless your driving a particulary hard to drive low output ribbon or trying to record the sound of a mouse snoring noise levels arent an issue. The only reason why the mic signal is noisy is poor gain structure/inappropriate mic for the task or your using a behringer console *evil grin*

There is no real excuse for noisy recordings , mic pres have been quiet for MANY MANY MANY years withing sensible gain levels and most mics have very good self noise , poor cables and interfence from external sources is a bigger problem.

People worrying about lower levels 'years ago' were battling a different system , original 16 bit convertors often sounded very grainy when recorded at lower levels , and when we all enjoyed tape it was tape noise we were battling unless you had a good NR system with its own associated over modulation issues.

So people pushed the levels as hard as they could to get away from the noise floor and get as hot a signal as possible to get the best conversion of the signal in the digital domain.

In the modern digtal world it pays to keep levels more conservative as many DAW's struggle with hot levels at the busses and outputs. "

Quote from Gareth at FXpansion

[Tim Armstrong]

Getting back to the initial question, compression on individual tracks can help focus a track and make it sit better in the mix. Folks often use it to make things louder without clipping, but it's also good for making a quieter track a little more distinct.

I've been using a compressor plug-in on vocals, copying a vocal track and strongly compressing one of the tracks while leaving the other untouched, and then mixing them together to taste...

Cheers, Tim

[Leon Grizzard]

Another point would be: limiter or compressor? Folks use the terms more or less interchangably, but as I understand the terms, compressors both boost the low signal and limit the stronger signal. Some would say use limiting, which limits the stronger signal, enabling you to boost the overall level. This has the effect of making the quieter parts louder, but avoids the "pumping" sound you get when the compressor tries to boost the signal when it drops below a certain level.

[cacibi]

Don't forget levelers and expandors:

From the Sweetwater tech tips site:

A limiter (see WFTD archive "limiter") is a device designed specifically to prevent audio levels from going higher than a specified point. In practice a limiter is basically a compressor with a very high compression ratio (20:1 to infiniti:1). The idea is that once the level reaches a certain point, the limiter will not let it go beyond that. Obviously there are tolerances such as a limiter's attack time that will have some effect on its ability to precisely carry out this function. Some of the more expensive units offer guarantees that no matter what happens on the input the output will not go beyond what you set, even for a micro-second. These strict tolerances can be very critical for people in the broadcast industry, large PA system installations, touring PA systems, and perhaps most importantly, in ear monitoring systems.

A leveler (see WFTD Leveler) acts a little bit like a compressor in that it uses a gain cell to change the level of audio signals it is processing. Unlike a compressor which typically only lowers the level of signals above its threshold, a leveler acts as a sort of automatic level control and will increase or decrease the level of program material to maintain a consistent volume over time. These devices do not usually need to generate rapid changes in a signal like a compressor or limiter, but will change the material gradually over a specified period of time in order to achieve a near constant output level. As such, peaks of a short duration will get through as though no device was in line so it may also be necessary to use a compressor or limiter in conjunction with a leveler. Some levelers have these functions built in. Applications for leveling include music playback systems like in a store or office, theater PA systems, and tape duplication.

[Leon Grizzard]

Originally Posted by cacibi

A limiter (see WFTD archive "limiter") is a device designed specifically to prevent audio levels from going higher than a specified point. In practice a limiter is basically a compressor with a very high compression ratio (20:1 to infiniti:1). The idea is that once the level reaches a certain point, the limiter will not let it go beyond that.

I would call that a brick-wall limiter. I was using limiter to mean a device which only compresses signals above a certain threshhold, as opposed to a stomp box type compressor, which both boosts signals below a certain point, and limits signals above a certain point. On a vocal track, some devices which boosts signals below a certain point will try to boost the silence between phrases, on during breathing, bringing up just the noise floor.

[Ben Harmless]

Originally Posted by Leon Grizzard

Another point would be: limiter or compressor? Folks use the terms more or less interchangably, but as I understand the terms, compressors both boost the low signal and limit the stronger signal. Some would say use limiting, which limits the stronger signal, enabling you to boost the overall level. This has the effect of making the quieter parts louder, but avoids the "pumping" sound you get when the compressor tries to boost the signal when it drops below a certain level.

Actually, they are more or less interchangable, so long as you've got a high enough ratio setting available.

Compressors, (with few exceptions) as part of their compression operation, do not actually boost the low-level parts of the signal independent of their reduction of the peaks. They simply boost the whole signal, (usually with an input gain control) which brings up the whole signal, and unless you raise the threshold, pushes more of the peaks into the compression zone.

The term "limiter" is simply used to describe a device that is capable of preventing a signal from exceeding a set level. This can often be accomplished by a compressor. Think of the compressor/limiter difference as:

"Limiter" - prevents signal from exceeding a maximum set level, usually to avoid peaking an input.

"Compressor" - used to reduce the dynamic range of a signal, often for color, or to provide a signal that exists more consistantly in a mix.

Compression is a HUGE topic. You could write books on this stuff. The skill of the engineer who operates a compressor can make or break a record.

[Ben Harmless]

The "Ben Harmless Vastly Abridged Guide to Compressor Controls"

Listed in order of what I usually touch.

Threshold and Ratio - Don't really worry about these yet. Just make sure threshold is set low enough and ratio is high enough to engage the compression for now.

Attack - The amount of time in between when the compressor detects that a signal is exceeding the setting determined by the threshold knob, and the time that the compressor actually starts to reduce the level of the signal. This is useful if you want to maintain the attack of say, a drum, while controlling the rest of the tone of the hit.

Threshold - The level of signal beyond which a compressor will begin to engage.

Ratio - The degree to which a compressor will decrease the level beyond the set threshold. A ratio of 4:1 indicates that it will take 4db beyond the threshold before the signal is allowed by the compressor to increase by 1db. A very high setting (infinity is avalible on many compressors) will simply not allow any level of signal to rise beyond the threshold.

Release - The amount of time which the compressor will continue to reduce the signal. This control can drastically affect the sound of a signal. It can be used for precise tayloring of a drum hit, or for adding overall color to a track. It's often a good idea to set by ear, while asking onesself "how long do I want this compression to last?" If you're compressing drums, often "until shortly before the next hit" is a good answer to this question, or you'll kill the attack, and things get all wonky. On the other hand, sometimes people do this on purpose.

(Makeup) Gain - Using a lot of compression can reduce the overall level of a track. This is added gain that can compensate for that. If I'm using more than a couple of db of this control, I feel that I'm doing something wrong. Compressors of the typical home-studio budget aren't known for the best quality gain stages. $5000 compressors are often sought after for their gain stages, which are part of the "color" of the compressor.

Knee - When the signal exceeds the threshold, how angry is the compressor? A low-db setting, or simply "hard" is very angry, and will slap down that signal with no warning. A "soft" setting will provide for a more gradual introduction of compression, so that the activation of the compressor will be less likely to be noticable. This setting can be very key in the application of compressor "color."

Note that not all these settings will be available on every compressor. For instance, I have a Dbx 163x, which is a half-rack compressor that has a single horizontal slider marked "compression." The compressor claims to figure everything else out on its own. Well, it does, but it may or may not be what I would have chosen if I had set the compressor using my ears. Luckily, I kinda like the color it adds to bass tracks. Not bad for $40.

Anyone have anything about this to dispute! Do it! Educate!

[Arlo]

A friend of mine Rich Hilton who plays keys and appears on many artists records once told me his secret to succesful compression was simple.

A little bit of it going in.

A little bit of it coming out.

A lot of guys overdo it on the input side.

[Frankie]

Originally Posted by fingers23

Thanks for the details of the Paul White book.

A quick Yahoo, the missus always loves that, and I found this treasure trove http://www.amgard.net/PWWrkShops.htm

I think I will be spending many a happy hour here...........

+1 !! Thanks for the link... As a newbe to doing my own, I need that... Sorry to bounce off the topic...

Now back to our program

[Big John]

Originally Posted by Ben Harmless

Compression is a HUGE topic. You could write books on this stuff. The skill of the engineer who operates a compressor can make or break a record.

+1, and when you have mastered the 'black art' of using a compressor you can start on the equally baffling 'EQ' and "Reverb' ones.

Seriously though, if there is a 'down-side' to digital recording it is the temptation to 'master by sight' by relying on a pre-set this or that to do the job, there's NO substitute for ears or experience so get to it and start practising and getting it to sound 'right'.

[Vol. Knob]

Big John Said

Seriously though, if there is a 'down-side' to digital recording it is the temptation to 'master by sight' by relying on a pre-set this or that to do the job, there's NO substitute for ears or experience so get to it and start practising and getting it to sound 'right'.

I tend to go ahead and run with the presets on my BR1180CD. When I messed around with the compression settings on my own, I found that I wasted a ton of time, became increasingly more dissatisfied with the sound, and began to fuss more and more about details that only I heard. My focus kept slipping from the #1 task of recording a decent song. Once I decided to go with a preset that was "not too much" and seemed to accomplish the task at hand, I found that I got a lot more done and was ultimatly having a great deal more fun.

I confess, I did tend to occaisionally turn the presets down a bit too.