Acoustic Guitar Compression

M

mrbozo

New member
Hi

I play fingerpicked acoustic guitar with clean recordings of moderate but not loud volume. I am a newbie mixer and have spent the last few hours researching compressor, and I believe I understand the logic of it and it's different controls (threshold, ratio etc...) however I feel that I am no closer to knowing what settings to use. I am willing to experiment to get the right results, though I was wondering if anyone might have any pointers to get me started? Specifically in regards to the threshold and ratio. What values of these two are more likely to give you subtle compression? What values are more likely to give you drastic compression?

Any ideas would be a help, thanks.
 
Try starting with 4:1 ratio. The threshold will be dependent on the level hitting the compressor, so start out with a very hi thresh, and bring it down slowly until you start getting a little compression. Lower thresh means more signal is getting compressed. I usually end up with about 5 or 6:1 to get ac. guitar to sit well in my recordings, but that can vary wildly depending on what works. I have often ended up with 10:1 on big strumming rhythm ac. Finger picked is very percussive attacks, so you may want to fiddle with the attack control to either let the initial attack get through or not. This can change the sound of ac. gtr a lot also. Slower attack settings will let the initial "pluck" through and make it more punchy...just my two cents
 
I suggest tweaking first with your ears. You have to know how it sounds on a track. Here is my method, inspired by Stavrou (get his book: Mixing with your mind). Put the threshold really low and the ration very high (10:1). Put your attack and release at zero. Then begin to tweak your attack. You'll hear how the transients are affected. That's where you decide which portion of the sound you compress most (attack or sustain). You have to know that from 0 to 50 ms. approx., you'll chop off your attack (a dull sound). An attack of more than 40-50 ms, will let the peaks going through, without being compressed. Then, what I normally do is to reset the threshold at the highest value and the ration at the lower one. From there, you decide if you want a more subtle or drastic effect. For a more subtle comp. use a 2 to 4:1 ratio. 6 to higher will give you a more drastic comp.
Then, you adjust your threshold until you get the desired amount of comp.
Keep in mind that lower ratio goes with lower threshold and higher ratio with higher threshold.

Finally, adjust your release time so you don't get pumping or breathing. Be aware of too short or too long release. I noticed that my release are most of the times between 150 and 500 ms. For an accoustic guitar, depending on the tempo of the song (so the sustain of each chord), If not a fast strumming part, my release will be around 300 to 600 ms.

Hope it helps.
 
My favourite quick goto compressor for acoustic guitar is the Acoustic guitar preset on my TL Audio Fat Man, the manual says the settings for this preset are: Soft Knee, Fast Attack, Fast Release ratio 1:5.

I don't always use the fat man as it's sometimes being used for other things, but the funny thing is that the settings in the fat man preset are about what I end up with anyway, the main difference between mixes and music type is how much I drive the input to the compressor.

Alan.
 
Thanks very much for your replies everyone! That's very helpful info and definitely gives me a better idea of what to start.
 
I like the LA2A (I have the plugin) or similar compression on everything. It's simple with 2 knobs - gain, and peak reduction. I try to get 1-2db of gain reduction and if you need more, it's nice to add another and do the same thing. Layer them.

Sometimes I layer 3 or 4 random compressors with presets and then adjust the ratio, threshold or gain to get 1-2db of reduction on each one to get a subtle but compressed sound.
 
As others have said, anybody else's settings probably won't help you much since the track levels and dynamics and the type of compressor used will always make big differences. However, last finger picked acoustic I mixed had the following settings:

Threshold -17
Ratio 3:1
Attack 26.0 ms
Release 220 ms.
 
Damn, I find I'm having to use 6:1 ratio, -10ish threshold with +4 to 6db gain to get it close to the levels to recording artists I'm comparing it to.

Still, it doesn't sound bad by any means, in fact, while the difference in volume is noticeable compared to the original recordings, the difference in quality is hardly noticeable at all - so I think compression is the winner!

The thing I am not noticing make much change is the attack. I can put it mega fast and mega slow and the difference to me is almost nothing. I wonder if I just don't have the ear for it yet? I'm using garageband, if anyone was wondering..

Thanks again for all your advice.
 
mrbozo said:
Damn, I find I'm having to use 6:1 ratio, -10ish threshold with +4 to 6db gain to get it close to the levels to recording artists I'm comparing it to.

Still, it doesn't sound bad by any means, in fact, while the difference in volume is noticeable compared to the original recordings, the difference in quality is hardly noticeable at all - so I think compression is the winner!

The thing I am not noticing make much change is the attack. I can put it mega fast and mega slow and the difference to me is almost nothing. I wonder if I just don't have the ear for it yet? I'm using garageband, if anyone was wondering..

Thanks again for all your advice.
Click to expand...

I don't wanna stomp on your new skills, but NOOOOOOO!!! lol :eek:

The thing about commercial tracks is they have been mastered. One of the main things that mastering does is increase the volume of your tracks. You will never get a good sounding mix at the same volume as a radio ready track with a compressor. In the mixing stage, your goal is to get things sounding as good as possible, not as loud as possible. I know that for some of us, those two seem very related, but there are other ways to get the volume you like while mixing... The knob on your speakers or interface. If you try to maximize level on your individual tracks within your mix, you will clip things all over and end up with a distorted mess. For now, don't worry about volume. Actually, keep your meters around 2/3 of the full range to be safe. After it's mixed, you can make the full stereo mix louder.

For now, don't compare the volume of those other tracks, just compare the quality. Drag your reference track into your daw and turn it down to match the level of your track. Now you can work without being fooled by volume.

As for the attack setting, it does take time to figure it out. If you can, try it on a bunch of different sources and see how it works on those too. Use a suggestion from above and lower the threshold till the the track is being squashed really hard, and then mess with the attack. You should hear the effects of that for sure.
 
One other thought: if you need a threshold of -10 (implying peaks going rather higher than that) you may be tracking just a bit too hot.
 
sixer2007 said:
I don't wanna stomp on your new skills, but NOOOOOOO!!! lol :eek:

The thing about commercial tracks is they have been mastered. One of the main things that mastering does is increase the volume of your tracks. You will never get a good sounding mix at the same volume as a radio ready track with a compressor. In the mixing stage, your goal is to get things sounding as good as possible, not as loud as possible. I know that for some of us, those two seem very related, but there are other ways to get the volume you like while mixing... The knob on your speakers or interface. If you try to maximize level on your individual tracks within your mix, you will clip things all over and end up with a distorted mess. For now, don't worry about volume. Actually, keep your meters around 2/3 of the full range to be safe. After it's mixed, you can make the full stereo mix louder.

For now, don't compare the volume of those other tracks, just compare the quality. Drag your reference track into your daw and turn it down to match the level of your track. Now you can work without being fooled by volume.
Click to expand...

To be honest, my goal was never to try and boost the volume to super high levels, I only thought I'd manage that a little, it was just something I became interested in along the way, surprised that I could boost each individual track quite a way without it sounding worse. Also, I wasn't comparing it to super loud pop tracks that play on big commercial radio stations, it definitely still would have been quieter than those. I was just comparing it to indie folk albums that I like, to which it was comparable..

Also, with mastering, I've never understood why when you're finished mixing you would want to boost the whole thing, if you've got already got it to a point you are happy with? And isn't that boost of volume in the master achieved by compression anyway, but to the whole thing instead of each individual track? If not, how else is it achieved?

Please excuse my inexperience..

sixer2007 said:
As for the attack setting, it does take time to figure it out. If you can, try it on a bunch of different sources and see how it works on those too. Use a suggestion from above and lower the threshold till the the track is being squashed really hard, and then mess with the attack. You should hear the effects of that for sure.
Click to expand...

Yeah I tried that setting, I could definitely hear it then, only it sounded rubbish either way, fast or slow, and I couldn't figure out which rubbish I liked better haha, but when I used a more natural ratio and threshold I noticed that I could hardly tell the difference. It will probably just take me time to hear the subtleties...

Bobbsy said:
One other thought: if you need a threshold of -10 (implying peaks going rather higher than that) you may be tracking just a bit too hot.
Click to expand...

Garageband doesn't allow me to see the db level of my highest peaks (do other programs?), which is something I think I would find very useful, so without that I was just using my ears, but I'll certainly take that on board and keep experimenting with it. Cheers.
 
mrbozo said:
Also, with mastering, I've never understood why when you're finished mixing you would want to boost the whole thing, if you've got already got it to a point you are happy with? And isn't that boost of volume in the master achieved by compression anyway, but to the whole thing instead of each individual track? If not, how else is it achieved?
Click to expand...

Yeah, that can be hard to catch onto I think... I'll do my best to explain, but I'm no expert on the subject!

How I understand it is this... In the digital world, 0 means 0! This is to say that if you cross 0dB on a channel (any channel), you will be in the realm of digital clipping, which is the nastiest form of distortion ever. So to avoid that, we mix at conservative levels.

I usually aim to keep any individual channel at 2/3 of the meter at most (i don't know how loud that actually is, but whatever). The same is true for the master fader. So, if all the channels are pretty low like that, you have no digital distortion and your mix will sound much more clean. I learned this the hard way, and I bet many others around here did too. However, this also means that your mix will be WAY quieter than any commercial release, whether a rock track or an indy track you're working with now.

But, that is A-ok my friend! You are correct that a mastering engineer will use compression to gain that volume, but actually, they usually go with a limiter, which is simply an extreme version of a compressor.
The difference with using a limiter AFTER all your mix is finished, or trying to gain a bunch of volume for each channel, is that limiting after results in no distortion if done correctly.

With a limiter, you can set a "ceiling" so that nothing ever crosses 0. Basically, think of it as taking a soda can and pressing it against your ceiling at home. What happens? The can crushes, but never goes through the roof (the roof is 0dB). The bottom of that can gets higher and higher though, which is the volume of your mix.

Idk if that answers your question or not. I hope so though... :drunk:
Just never try to make everything in your mix loud. Make the mix loud after it all sounds honky-dory.

One last thought... You have a knob on your compressor that says "output" or "make-up" gain. In essence, this works the same as your fader for that channel; it increases or lowers the volume. What it's meant to be for is comparison. See, compression lowers the overall volume of your track. So you need to increase that make-up gain to recover the lost volume. If you boost it too much though, when you bypass the compressor, you'll think "oh wow! Compressors kick ass! My track is way louder with it!" In reality, you're fooling yourself with volume.

/end rant/
 
The threshold and output gain depend on how load the tracking is and how loud you want the guitar, so these things depend on how it sounds, worry more about the attack release and ratio and use the threshold to control the amount of compression applied and the output gain to get the level sitting in the mix. You can never recommend threshold and output gain settings as there are too many factors to consider.

Alan.
 
sixer2007 said:
Yeah, that can be hard to catch onto I think... I'll do my best to explain, but I'm no expert on the subject!

How I understand it is this... In the digital world, 0 means 0! This is to say that if you cross 0dB on a channel (any channel), you will be in the realm of digital clipping, which is the nastiest form of distortion ever. So to avoid that, we mix at conservative levels.

I usually aim to keep any individual channel at 2/3 of the meter at most (i don't know how loud that actually is, but whatever). The same is true for the master fader. So, if all the channels are pretty low like that, you have no digital distortion and your mix will sound much more clean. I learned this the hard way, and I bet many others around here did too. However, this also means that your mix will be WAY quieter than any commercial release, whether a rock track or an indy track you're working with now.

But, that is A-ok my friend! You are correct that a mastering engineer will use compression to gain that volume, but actually, they usually go with a limiter, which is simply an extreme version of a compressor.
The difference with using a limiter AFTER all your mix is finished, or trying to gain a bunch of volume for each channel, is that limiting after results in no distortion if done correctly.

With a limiter, you can set a "ceiling" so that nothing ever crosses 0. Basically, think of it as taking a soda can and pressing it against your ceiling at home. What happens? The can crushes, but never goes through the roof (the roof is 0dB). The bottom of that can gets higher and higher though, which is the volume of your mix.

Idk if that answers your question or not. I hope so though... :drunk:
Just never try to make everything in your mix loud. Make the mix loud after it all sounds honky-dory.
Click to expand...

I understood this, although you have helped me understand it a little better I think. If the peaks on your individual tracks are as close to 0db as they can be without peaking, then there would be no use for the master boost, right? (Or the master boost would be used only to bring the bottom end up, but not the top end?) But as that is too risky, or maybe too fiddlesome, to achieve, you take a more conservative approach, keeping the peaks safely below 0db in the individual tracks to avoid distortion, knowing you can boost it safely later with the limiter. Is that right?



sixer2007 said:
One last thought... You have a knob on your compressor that says "output" or "make-up" gain. In essence, this works the same as your fader for that channel; it increases or lowers the volume. What it's meant to be for is comparison. See, compression lowers the overall volume of your track. So you need to increase that make-up gain to recover the lost volume. If you boost it too much though, when you bypass the compressor, you'll think "oh wow! Compressors kick ass! My track is way louder with it!" In reality, you're fooling yourself with volume.

/end rant/
Click to expand...

When I set the threshold at certain levels however (at the higher end), it very definitely boosts the volume of the track. It doesn't decrease it. Or have I been confused all this time?

For instance, as I mentioned earlier, I had a threshold of -10db. This was definitely louder than when the threshold was at -20db. Does this mean actually that at -10db less compression is being applied, not more, because there is less information at/above that volume?

If that's true, it would make sense to me only if the compressed track was still quieter than the original, but it wasn't, it was definitely louder than the uncompressed track. Or would that have been the gain causing that increase in volume? The gain was about 3db at that point...

Yesterday I finally felt like I understood compression, now I feel like I'm back to the start...

I really appreciate your comments though.

witzendoz said:
The threshold and output gain depend on how load the tracking is and how loud you want the guitar, so these things depend on how it sounds, worry more about the attack release and ratio and use the threshold to control the amount of compression applied and the output gain to get the level sitting in the mix. You can never recommend threshold and output gain settings as there are too many factors to consider.

Alan.
Click to expand...

Thanks for your reply, as I mentioned above, I'm feeling more and more lost the more I read...

I understand compression reduces the dynamic range. If one of the results of this then is a reduction in a volume, which you use gain to boost back up again, meaning the overall level of the track stays more or less the same, then what is the actual point of compression to begin with? I always thought (most of the things I've read suggest this) that it was used as a tool mostly to increase a tracks volume, but that theory seems to have been turned on its head now. So what are you actually trying to achieve when you use compression? And if the answer is to improve, or clean up the sound somehow, wouldn't you just use EQ's instead? If the answer is to provide more balance in a recording, this I can understand, but is there any point applying compression if the recording is already balanced?

These questions are in relation to the mixing stage, not the mastering.

Cheers..
 
mrbozo said:
I understood this, although you have helped me understand it a little better I think. If the peaks on your individual tracks are as close to 0db as they can be without peaking, then there would be no use for the master boost, right? (Or the master boost would be used only to bring the bottom end up, but not the top end?) But as that is too risky, or maybe too fiddlesome, to achieve, you take a more conservative approach, keeping the peaks safely below 0db in the individual tracks to avoid distortion, knowing you can boost it safely later with the limiter. Is that right?





When I set the threshold at certain levels however (at the higher end), it very definitely boosts the volume of the track. It doesn't decrease it. Or have I been confused all this time?

For instance, as I mentioned earlier, I had a threshold of -10db. This was definitely louder than when the threshold was at -20db. Does this mean actually that at -10db less compression is being applied, not more, because there is less information at/above that volume?

If that's true, it would make sense to me only if the compressed track was still quieter than the original, but it wasn't, it was definitely louder than the uncompressed track. Or would that have been the gain causing that increase in volume? The gain was about 3db at that point...

Yesterday I finally felt like I understood compression, now I feel like I'm back to the start...

I really appreciate your comments though.



Thanks for your reply, as I mentioned above, I'm feeling more and more lost the more I read...

I understand compression reduces the dynamic range. If one of the results of this then is a reduction in a volume, which you use gain to boost back up again, meaning the overall level of the track stays more or less the same, then what is the actual point of compression to begin with? I always thought (most of the things I've read suggest this) that it was used as a tool mostly to increase a tracks volume, but that theory seems to have been turned on its head now. So what are you actually trying to achieve when you use compression? And if the answer is to improve, or clean up the sound somehow, wouldn't you just use EQ's instead? If the answer is to provide more balance in a recording, this I can understand, but is there any point applying compression if the recording is already balanced?

These questions are in relation to the mixing stage, not the mastering.

Cheers..
Click to expand...


For the first piece, yes, that's the idea. You could try to get all the tracks near 0, but it's a risky business of not going into the red. Mix at low levels and limit the mix later for the volume you're after (which is more complicated than that, but it's the jist)

For your second part, I think what is happening is you have a compressor with "automatic makeup gain". In many compressors, like the stock ones in cubase, there is a button to turn that feature off, but it's on by default. I recommend looking for it on your compressor.

What that feature does is, when you lower the threshold, and more gain reduction happens (which you should be able to see on a meter in the plug in) the make up gain is auto adjusted to match the initial level you started will before compression. The issues I've found with this feature, are 1) you can't really A/B to test your settings, and 2) the auto make up is often too much. This means while you compress, the make up is over compensated for, leaving the after signal louder than you started with. This is the worst kind of trickery about a compressor, because we always think that louder is better. Even if you're compressing the snot out of a track and it might not need it, it's louder and sounds better, but it's a mirage.

Look at your GR (gain reduction) meter and try to add the same amount in make up as you're taking away. It's often not precise though, so bypass the comp, listen to the level, then flip it back in. Is it louder or quieter? Compensate with the make up till it sounds even.

Lastly, in it's simplest uses, compression is used to level out a performance. Vocals not all that even? compress them some to smooth them out.
 
I normally use compression for the sound of compression more than smoothing out the dynamics or making something louder. For example, a quick attack and release with a high ratio will bury the pick noise and bring out the sustain. A slow attack and release will accentuate the attack of the pick. Depending on how the guitar needs to sit in the mix, you would want different settings.

Mastering is the point where all the songs are brought together and processed so that they all sound like they belong together, transition smoothly and translate well within the context of the genre. Its not really about making things loud, but that is where all the mixes get the final volume. Mastering limiters are a slightly different animal than the compressors and limiters that we tend to use for mixing. They are specially designed for this purpose.

If you wanted to get your guitar louder without affecting the sound as much, a limiter would work well. This is because the transients don't equate to loudness. Its very easy to have a quiet recording of an acoustic guitar that has peaks approaching 0dbfs. A good limiters can smack down those peaks without changing the envelope of the rest of the sound.
 
mrbozo said:
So what are you actually trying to achieve when you use compression? And if the answer is to improve, or clean up the sound somehow, wouldn't you just use EQ's instead? If the answer is to provide more balance in a recording, this I can understand, but is there any point applying compression if the recording is already balanced?
Click to expand...

I use compression to make something more bold, like a lead vocal or bass. It's also helpful for pushing something to the back without it disappearing, like vocal harmonies.
 
sixer2007 said:
..the make up gain is auto adjusted to match the initial level you started with before compression. The issues I've found with this feature, are 1) you can't really A/B to test your settings, and 2) the auto make up is often too much. This means while you compress, the make up is over compensated for, leaving the after signal louder than you started with. This is the worst kind of trickery about a compressor, because we always think that louder is better. Even if you're compressing the snot out of a track and it might not need it, it's louder and sounds better, but it's a mirage. ..
Click to expand...
Wow. + 1000% re auto makeup gain.
IDK.. What would that be designers thinking they're doing us some favor ..Not :rolleyes:
 
mixsit said:
Wow. + 1000% re auto makeup gain.
IDK.. What would that be designers thinking they're doing us some favor ..Not :rolleyes:
Click to expand...

Yeah, It's a real pain in the a$$ isn't it?! Besides a limiter from time to time, if the comp won't allow me to adjust the output I just quit using it. Mostly because it's impossible to A/B which I do constantly with compressors. I'm still figuring them out to be honest.
 
sixer2007 said:
For your second part, I think what is happening is you have a compressor with "automatic makeup gain". In many compressors, like the stock ones in cubase, there is a button to turn that feature off, but it's on by default. I recommend looking for it on your compressor.

What that feature does is, when you lower the threshold, and more gain reduction happens (which you should be able to see on a meter in the plug in) the make up gain is auto adjusted to match the initial level you started will before compression. The issues I've found with this feature, are 1) you can't really A/B to test your settings, and 2) the auto make up is often too much. This means while you compress, the make up is over compensated for, leaving the after signal louder than you started with. This is the worst kind of trickery about a compressor, because we always think that louder is better. Even if you're compressing the snot out of a track and it might not need it, it's louder and sounds better, but it's a mirage.

Look at your GR (gain reduction) meter and try to add the same amount in make up as you're taking away. It's often not precise though, so bypass the comp, listen to the level, then flip it back in. Is it louder or quieter? Compensate with the make up till it sounds even.
Click to expand...

Thanks a ton - I can't find any 'auto make up gain' button, but the gain metre is naturally set to +4db by default, so that must have been the cause. What a nuisance..

sixer2007 said:
Lastly, in it's simplest uses, compression is used to level out a performance. Vocals not all that even? compress them some to smooth them out.
Click to expand...


Farview said:
I normally use compression for the sound of compression more than smoothing out the dynamics or making something louder. For example, a quick attack and release with a high ratio will bury the pick noise and bring out the sustain. A slow attack and release will accentuate the attack of the pick. Depending on how the guitar needs to sit in the mix, you would want different settings.

Mastering is the point where all the songs are brought together and processed so that they all sound like they belong together, transition smoothly and translate well within the context of the genre. Its not really about making things loud, but that is where all the mixes get the final volume. Mastering limiters are a slightly different animal than the compressors and limiters that we tend to use for mixing. They are specially designed for this purpose.

If you wanted to get your guitar louder without affecting the sound as much, a limiter would work well. This is because the transients don't equate to loudness. Its very easy to have a quiet recording of an acoustic guitar that has peaks approaching 0dbfs. A good limiters can smack down those peaks without changing the envelope of the rest of the sound.
Click to expand...


coreytx said:
I use compression to make something more bold, like a lead vocal or bass. It's also helpful for pushing something to the back without it disappearing, like vocal harmonies.
Click to expand...


...And now I feel like I am understanding compression properly. Thanks to all.
 
You must log in or register to reply here.
Back
Top