Doesn't matter.

 

It's all going to depend on the material you are working with. But what you describe - a pumping or surging action, sounds like you need to increase the time constant (release) a bit so as to make the compression action act a bit longer on the signal and lessening the "in/out" action. The threshold control simply sets the point at where the input signal triggers the comp.

 

With stacked comp setups like this, I almost always end up swapping the FET for a limiter, even with many vocals.

This. Either know what you want to hear, or what you don't want to hear.

I actually try my best to minimize my use of compression, using every technique I know to get things packed up nice and neat before throwing on a compressor, and they're often for character at that point, or for sculpting something so it sits better in a mix, as opposed to using them to tame anything. Ask me how many tape and saturation plugins I own. Go on... ask.

 

I love this thread. Great question!

 

Acoustic guitar is a great example of an instrument where in most scenarios, transparent compression is the desirable outcome. They can easily have all the life sucked out of them and sound very unnatural if over-compressed.

Forgive me if I'm wrong, but it seems you are a beginner at this, so a few basic principles need to be understood in order to achieve the results that you desire.

Compression isn't just about volume, it's about shape and tone. Yes, you read that right, tone. A compressor provides equalisation as well as dynamic control. The transient in an acoustic guitar is where most of the high frequencies are contained whereas the sustain portion is nearly all lower frequencies. Remove all the attack and the guitar will sound dull and lifeless, remove the body and it will sound thin and weedy.

A compressor will affect the groove of a part, either adversely or for the better, depending on the settings. Done wrong it can completely flatten a performance. Dynamics are essential to the rhythmic pulse of music.

The choice of compressor will also affect tone as there's a huge difference between a clean digital comp and an analog (or ITB emulation) which will add harmonic saturation (that's distortion for the uninitiated).

I'll try to refrain from putting specific settings here (you wanted those right?) as that could vary wildly depending on the context and inevitably whatever I write will be completely wrong for your case. Better to follow a methodology than rely on specific "presets".

First, identity what you are aiming to achieve. Do you want a brighter more attacking tone to cut through a mix or do you want a warmer, more sustained one to fill in the holes of a sparse mix?

Are you looking to correct uneven playing? If so, re-record the parts if possible, played correctly.

A good plan is:

Put an EQ first in the chain. Cut any rogue/undesirable resonant frequencies. In a busy mix, shelve or high-pass the lows, they just get in the way. Where exactly that point is and by how much depends on the context but can be anywhere between 80-150 Hz - experiment. Also the low-mids may need a cut somewhere between 200 and 500 Hz (500 Hz is rather high but still possible). Dealing with this before adding a compressor greatly aids the compressor action in that it's not clamping down on frequencies that you don't actually want in the mix and can focus on what you do want.

Place a compressor second, deal with the transients. Do you want less or more? A medium to slower attack of anywhere between 5-50ms and a fast to medium release that is groove dependent will help emphasise the pick and high frequency content. By groove dependent I mean setting it in such a way that it has completely released before the next note is played. This is entirely tempo and performance dependent so could be anywhere between 50-150 ms. Ratios of 2-8:1 are possible here, it's impossible to say without knowing the context.

Next, insert another EQ. This is to shape the tone of the instrument in a broader way. Use wider Qs to push particular areas of the sound that you desire, narrow cuts to remove anything that has popped out due to the compressor.

Follow with a second compressor. This is for a more broadband approach, to even out inconsistencies, emphasise the body of the guitar and glue the performance. Use faster attack times of under 5 ms and again time the release to the playing but with more of an emphasis on slower settings than the first compressor. Gentler ratios are key here, 1.5-4:1 maximum.

Place a third EQ... Only joking, if you've not got it right by this point you shoud review your settings rather than add yet more processing.

Threshold for both compressors should be set in such a way as you only see a maximum of 6 dB gain reduction (unless sound designing which is another case entirely), 1-4 dB is more the ballpark for a natural sound.

It's by no means necessary to do all stages of this chain, sometimes 1 or none of the above works perfectly well.

There are also other approaches that can yield fantastic results. Careful use of multiband compressors can distill all of the above into a single process. Parallel compression is particularly good on acoustic guitars. Tube/tape saturation can remove the need for compressors and also provide a wonderful mojo if desired.

The choice of guitar itself will make a huge difference in how compression affects the dynamics, there's a whole world of difference between a spruce top flamenco guitar and a mahogany jumbo, that's without discussing guitars with pickups.

The style of play is also important. Picked or fingerstyle, long sustained chords, damped staccato chugs and lead lines are all different case scenarios, as is the combination of all of the above.

The context is equally as important, is it a solo performance or a component of a dense mix? Both can require a different approach. With a solo performance, a gentler, less attacking approach, with more low end is often appropriate. Whereas in a busy mix, the transients can be in need of emphasis with a reduction of competing lower frequencies.

Plus, of course, by far the biggest factor affecting compression with any acoustic instrument is the performance. A well controlled performance will mitigate any need for overt compression. Played too hard, the transient will overpower the body, played too soft and it will lack attack. There really is no compressor that is EVER going to be as good as a well controlled performance.

The overall point to take away from this is that compression is just a component in the production of a good guitar sound and that there's a symbiosis at play between all the various elements that are involved.

The fact is, there's that much to talk about and so many nuances that I could sit here writing about this all day and still have only touched on the subject.

I'm glad you didn't ask about microphone choice and placement coz I'd have been here until next month...

 

Well, I'm here for the foreseeable future, so please, feel free to cover my face in sem... knowledge

 

Ah yes, the cumpressor

 

I was in reverse mode and so forget everything that I wrote in this thread. 3...2...1.....forget...forget.

 

arent these compressors too in some sense ?
extrem coloration compressors ?

 

They're compressors with istantaneous attack and release

 

Interesting, results in a different sound (infinite ratio, less harmonics, even shorter times).

Well, you can (slightly) saturate without producing audible compression. And of course ratio and as @franksq98 correctly said, timing is different.

 

You'd be lucky to get enough knowledge out of me these days to sting your eyeballs..

Anyhow, micing is just a case of dropping an SM58 inside the sound hole or taping a PZM to the body isn't it?

Seriously though, I may start a separate thread on microphone technique as it's a subject very much worth discussing.