The No.1 Website for Pro Audio
How to compress?
Old 21st February 2016
  #61
Gear Maniac
 

Quote:
Originally Posted by systemheavy View Post
I'd recommend tuning into JJP's explanation of compression (Around 25 mins in, although the whole interview is worth checking out, especially details on the mid range).
It's by far the best explanation I've ever heard:
Pensado's Place - Episode 022 - Jack Joseph Puig - Pensadia
That link doesn't work anymore, this was a working link I found: Watch Videos Online | Pensado's Place - Episode 022 - Jack Joseph Puig | Veoh.com
Old 10th March 2016
  #62
Here for the gear
 

The main problem with compression is not how it works but what to use it for.
Old 18th March 2016
  #63
Quote:
Originally Posted by arseniy View Post
The main problem with compression is not how it works but what to use it for.
agree with this, you only need to compress the weakest links, sometimes no compression or expansion is better.
Old 29th March 2016
  #64
Here for the gear
 

How do you side-chain with compression?

How do you side-chain with compression? Specifically in EDM production.
Old 23rd May 2016
  #65
Gear Head
 

Quote:
Originally Posted by polapade View Post
This is THE best verbal explanation of compression I've come across
. It's by a school called "Sonic Scoop"

-Alex
This is great. Such a succinct summing up of everything we know yet tend to forget.
I'm going to write a bullet point summary of this and stick it to my studio wall.
Old 27th May 2016
  #66
Gear Nut
 
Deadsound's Avatar
 

Here are some great videos that show how to compress in the context of the mix! SO IMPORTANT!

There are many videos out there that explain "The Theory" of compression or "What Is Compression." Those are great but hearing and knowing how to use compression in the mix, we feel, is much more vital to understand and practice

Check these videos out as they all explain compression, the plugins used to compress (so you understand how they are different) and when to use different compression on different sources of audio.

Drums:
https://www.youtube.com/watch?v=kJoQNrk5u8g

Bass:
https://www.youtube.com/watch?v=hgl7prlbYyo

Guitars:
https://www.youtube.com/watch?v=sNr8otdRDc4

Synths/Pianos:
https://www.youtube.com/watch?v=YL8xScAdMys

Kick:
https://www.youtube.com/watch?v=s7exvF7n6PQ

Snare:
https://www.youtube.com/watch?v=0UyyrYa4eeY


If you are into reading vs video this page literally covers all compression types and explains what they do

Compression Archives - Continuum Music Studio
Old 21st August 2016
  #67
Gear Head
 
<Tempest>'s Avatar
 

1. Learn what the controls do.
2. Adjust the compressor on instruments while they're in the mix.
3. Twist the knobs and listen to what happens, make it obvious.
4. Repeat steps 1-3 until you get it. Then try a new compressor.
5. Make a lot of mistakes, including but not limited to: Over compressing everything. Destroying transients. Not paying mind to gain staging. Thinking the equipment will make your tracks sound pro. Buying into marketing hype. Not following steps 1-4.
6. Realize compressors are tools and it really makes no difference what you have or use, only that you know how to to achieve results that benefit the artist(s) vision for the song. You will definitely acquire your own tricks along the way, and steal a lot of others as well.

My perspective, at least.

-<T>
Old 9th September 2016
  #68
Here's a video I did years ago where I explain what the controls do:


So there's step 1 from above.
Old 10th September 2016
  #69
Information in this video is actually incorrect.
Old 10th September 2016
  #70
Lives for gear
 
chrischoir's Avatar
 

Quote:
Originally Posted by vinnie2k View Post
Information in this video is actually incorrect.
it's ok, it's the internet
Old 11th September 2016
  #71
Old 11th September 2016
  #72
Lives for gear
 

I posted this at the home recording site, looks like it'd be fitting here as well.
Like some cool insights on how / what makes various compressors tic'?
The Kush Tweaker.
http://www.mediafire.com/view/3bwd2x...ull_Manual.pdf
Old 29th September 2016
  #73
Lives for gear
My question is how to create a Boss CS3 type compression effect using a plug-in after the fact? (on guitar)

For example, the CS3 includes a 'sustain' pot, not a release setting. It gives a longer compressed ringing sound to guitar chords - and is more of an effect than a levelling function, which you would usually want to be transparent.

So how would I use my DAW / plug-in compressor to achieve this sustain effect?
Old 4th October 2016
  #74
Lives for gear
 

Quote:
Originally Posted by dontsimon View Post
My question is how to create a Boss CS3 type compression effect using a plug-in after the fact? (on guitar)

For example, the CS3 includes a 'sustain' pot, not a release setting. It gives a longer compressed ringing sound to guitar chords - and is more of an effect than a levelling function, which you would usually want to be transparent.

So how would I use my DAW / plug-in compressor to achieve this sustain effect?
Looked at the pdf, Kind'a disappointing when manufacturers gloss over info -like in this case could be instructional. From what I read they could be doing gain, threshold, ratio, release, who knows under a nob called 'sustain.
Old 4th October 2016
  #75
Here for the gear
 

Quote:
Originally Posted by miguellacorte View Post
I have a question about compressing. Why does is sometimes outputs more volume? .
By compressing you make the loud parts softer, if you then boost the total sound you have increased the soft parts relative to the loud parts of your audio signal. The signal is louder this way. Hope this makes sense and helps!
Old 23rd November 2016
  #76
Gear Head
 
83dude's Avatar
I have an easy approach at hand, that gets beginners and students going. It's targeted at starters, but works quite well. Reading up on the workings of a compressor helps!

--

83's compression recipe

1. Level your material.
Do it! Get your operation levels straight (especially, if there's EQs, DeEssers etc. in the daisy chain before the compressor/limiter).
If in doubt, read your operation level up. -18 dB(FS) is a good point to start at, for pre-compressed material aim for -12 dB(FS). Voltage is quite another thing. Read them up! Level handling is what differs proper engineers from ... yeah, none-so.

2. Turn your threshold to maximum (upwards out of the signal processing), and set attack & release shortest.
No worries - you'll change this later on. It'll just make it easier for the next two steps for the time being. If you know what you're doing, you might skip this one.

3. Choose your ratio according to purpose.
This is a tricky one. It mostly depends on what you're doing, since all "purposes" require a different approach. Transient taming requires rather high ratios (e.g. 8:1, 20:1), while you're far better off with lower ratios if you want to gain on RMS level and overall loudness and fatness. This always depends on the material you're using. No two signals are the same, so be ready to experiment. Use your ears!

4. Slowly turn down the threshold and sink it into the signal, until the desired compression is achieved.
This usually requires some experience. Nevertheless, if you keep your purpose in mind, you'll probably know what to listen for. Don't squash your signal too much if not intended, don't go easy if not applicable. A good indicator is your meter for gain reduction: some people tend to go for softer reductions (3-6 dB), some go harder (5-10 dB). Depends on the purpose and the intended result, though. Try what works best for you, especially on gaining loudness you'll soften up attack and release anyway.

5. Adjust attack and release until the sun goes up.
You'll want the shine! This is the moment I prefer analogue equipment over plug-ins, since you'll be hooked on getting the sound right instead of using your eyes for meters etc. If you can, use a controller hear (hehe, pun intended). I'd actually suggest fiddling with attack and release, until you couldn't make the sound better (e.g. "open them up"). If your compression rate is too high, don't use makeup gain! Try turning up your monitoring level instead, but don't forget to turn it down later!
I have had better results with attack values shorter than release values, but this isn't always right. Transient taming mostly requires short attacks, but release depends on what sound you're looking for. If you want loudness and have transients to mostly pass through compression unhandled, you'd be in for longer attack but short release.
In short: There simply is no rule for this. Fiddle around with these parameters until it sounds just about right - try tuning them at the same time (via hardware or controller; a mouse only handles one parameter at a time!), until you feel you can't make it sound better.

6. Level your output.
Try to make up for your level loss by turning up the makeup gain. I'd suggest not to rely on gain reduction meters, their quality mostly depends on slow metering applications if not using digital metering. The best way to achieve this is to check your levels BEFORE compression (bypass, deactivating the insert return; SOLO PFL) and turning up the makeup gain until it reaches the same peak level as before (after bypass, that is). If in doubt, use your ears. Yes, it SHOULD sound louder, but avoid clipping. Roughly setting makeup gain to the level indicated by gain reduction ist rough at most, metering plug-ins CAN help.

--

Some general advice:

a. Use your ears.
There's (almost) nothing to add. If it sounds dull, it IS dull. Try another compressor (plug-in) if you just can't get it right. Usually, one or two compressors (also plug-ins) suffice to achieve the trick. The one you know is better than the one you seek.

b. Experiment!
You'll know that it's right, when it's right.

c. Sidechain filters
They can solve problems more than you'd think. If there's bassy material, try cutting it out. My advice: Cut more than necessary. If you want to keep out that kick drum from your drum bus, do NOT filter around 100 Hz, but go for 250 Hz or something. Fiddling is the key!

d. Multiband vs. Multi-knee
I'd suggest leaving out multiband compression for the time being, if you're not experienced. It's a kind of magic, but like any magic, try to avoid it until you know what you're doing. Finding crossover frequencies is a tedious task at best if you can't get a grip on the material processed. It's better to compress your sound multiple times ("multi-knee") tenderly than to compress once but effectively. Don't save on ressources here!
Quick tip: Try taming transients first, "loudnessing" later in the chain.
Quick tip 2: Sidechain filters are the real MAGIC here without any need to cast it actually. Get your bass straight by trying it out - again.

e. Use your ears.
This can't be emphasized enough. I've seen engineers handling their plug-ins fiddling with makeup gain between 2.45 dB and 2.47 dB to get it right according to their metering application, shouting why it wouldn't allow for 2.46 dB makeup. Don't be that person, don't rely on your eyes, but use your ears. This is also another reason why I prefer (analogue) hardware - you're forced to use your ears (and not your eyes). It's even better with restricting yourself to "steppable" parameters, because you'll always compromise for the best sound. Sometimes, "easy" is the new "superb".

f. If you're stuck, try again.
No two signals are the same. If you've achieved excellent results with one setting, it doesn't mean you'll get it just as nice for another sound source / mix application. Have you checked your equipment chain? Same speaker, same tech? Same room? Try fiddling. Getting the compression basics straight helps!

g. Avoid limiting.
You generally speak of limiting when compressing with high ratios, usually higher than 8:1 (literature tends to differ). Try to avoid it whenever possible. If you NEED a limiter, it usually means you couldn't control your dynamics beforehands (there's exceptions, of course, e.g. in electronic music). But no matter what, several subtle compressions will achieve a far more satisfying result than limiting for all its means.

h. Use your ears.
Please do.

i. Doesn't sound right - chain OK?
If you just can't get that result you wish for, try inserting an EQ beforehand. Or switch it to afterwards. If not enough, use a deesser (before/after), or a noise gate. If that's not enough, record again. There's only so much you can do fixing stuff. You usually work on a DAW, not a STP (sewage treatment plant). Yes, I had to look that up.

--

I have to emphasize that this is just a rough guideline that helped me a lot getting used to compression. It's not perfect, but it gets you there, ESPECIALLY in the beginning. You should always check your levels properly and be aware, that no material is the same as it was hours before. Don't use presets if possible, but use them if you tend to use similar setups a lot (like in marketing productions with similar vocalists, e.g.), but ALWAYS be ready to adjust their settings!

Last of all: A comfortable environment makes comfortable work and an ease of handling you can't substitute. Work where you know you can, whether it's people present or equipment available. Work with the tools you feel comfortable with, and don't listen to people telling you that "this is the only compressor you'll need" - experiment yourself. Know your stuff!

--

P.S.: Feedback appreciated.

Last edited by 83dude; 23rd November 2016 at 10:19 PM..
Old 29th November 2016
  #77
Good compression all relies on the tempo and pacing of the vocals. Listen to the phrasing and how the singer is singing and decide what style you wants. Your attack and
release times will be dependent on his or her rhythms-tmepo-release times of words.....mix it with a little saturation and your golden! No ONE compression setting works
for every song.. every song is COMPLETELY different. Listen to the way the compressor works and ask yourself "how does this setting make the song feel?"
Old 29th November 2016
  #78
Quote:
Originally Posted by ology View Post
Good compression all relies on the tempo and pacing of the vocals. Listen to the phrasing and how the singer is singing and decide what style you wants. Your attack and
release times will be dependent on his or her rhythms-tmepo-release times of words.....mix it with a little saturation and your golden! No ONE compression setting works
for every song.. every song is COMPLETELY different. Listen to the way the compressor works and ask yourself "how does this setting make the song feel?"
you are totally right here, release time is the magic ingredient, getting that to fit in a musical way is so important, that's why I'm not a fan of compressors where you can't control attack and release times.
Old 2nd December 2016
  #79
Lives for gear
 

A good rule of thumb when mixing is when you think you have enough compression where it sounds good, back it off 30~50% so its barely noticeable.
reason being is when you master you'll be using additional compression and limiting over what's compressed on tracks.

Driven guitars rarely need compression because they already have their peaks clipped. Clean guitar where you have a huge dynamic range, Bass, Drums and Vocals can benefit the most.

Dot be afraid to try using a limiter on drums instead of a compressor. A compressor brings the soft notes up between the peaks that are being compressed. A limiter brings down peaks but the floor remains where it is. Rock drums for example can retain a hard small without becoming pumpy or flabby sounding like it can with a compressor. Overall it winds up being much more transparent when used on a drum buss.

You may also want to try using a multiband limiter/compressor in a Buss instead of using individual track compression. A multiband uses several compressors in parallel and you feed each with a section of the frequency spectrum. The low band compressor for example can compress bass and kick together. The mid band can compressor guitars, toms, snare and some of the vocals together, the high bands can tame the cymbals, Vocal top end and snare crack.

Because you're targeting frequencies instead of individual instruments you can get things to gel and work together in those bands. Plus an instrument like bass which may have sub lows and also lower mid frequencies, you can get two compressors taming the same instrument instead of using one for all frequencies like you would on a track. Notes further up the bas neck can be made to punch more instead of getting thin and weak.

You still have to know how to use a regular compressor well but knowing how to use them all is important.
Every compressor has its own style and coloration too. I use maybe a dozen compressors for different purposes. I know them all well enough to when I hear a piece of music I simply grab the tool that will give me the results I want. Don't be afraid to save presets too. The saved preset may not work the same on another track but it can help you zero in on where you need to be.
Old 21st December 2016
  #80
Gear Head
 
Gabriele's Avatar
 

Quote:
Originally Posted by matt thomas View Post
Image you have the numbers:

1,2,3,4,5,6,7,8,9,10

You want to make these louder on average, but you can't go above 10, so you limit these (compress them) to 5, and you have

1,2,3,4,5,5,5,5,5,5

You can then turn them all up (in this case double them) to

2,4,6,8,10,10,10,10,10,10
That's a brilliant way to put it!
Though I believe 6,7,8,9,10,10,10,10,10,10 is more appropriate.
That's how dB work: in this case applying 5 dB of make-up gain.
And IMO mentioning that "average" level being 8 is confusing.
What really matters here is that you have reduced the dynamic range of the signal from a value of 10 (1 to 10) to a value of 4 (6 to 10).
Working the fader you can reposition that instrument/vocal compressed track where it better fits within the mix.

G.

Last edited by Gabriele; 22nd December 2016 at 10:28 AM..
Old 3rd January 2017
  #81
Gear Head
 

Few basic n00b questions...

The amount of gain reduction is determined by ratio: a ratio of 4:1 means that if input level is 4 dB over the threshold, the output signal level is 1 dB over the threshold. The gain (level) has been reduced by 3 dB:

1. Why does the compressor still compress when the threshold is -12, but the ratio is 100:1? There is no signal that is 100db over the threshold, so why is there still reduction?

2. I'm looking at my waveform in Adobe Audition. I'm looking at everything over -12db... some peaks are at -3db. So, I set a threshold of -12db and a ratio of 9:1 because those peaks are 9db over the threshold, so they should be outputting at -11db after compression, right? But my waveform measurement doesn't show the peaks being reduced to that level. What am I doing wrong? I was just experimenting with a waveform to get a better understanding compression, but my stuff isn't getting reduced to level I think they're going to get reduced to. I must be missing a trick here. Feel like a right dummy.
Old 4th January 2017
  #82
Here for the gear
 

PDF file - compression info

Quote:
Originally Posted by matt thomas View Post
I started a thread the other day featuring a great video on EQing techniques.

How to EQ

I was wondering if anyone knows of a similarly good video on compression (dynamic range compression), so we can have a sister thread on that?

Cheers
Matt

ps. I'm going to sticky this for a while to see if we can get any more responses
Hi everyone, here is a pdf file that I've had for sometime now from my previous modules at uni. Its got overall information on mixing including compression and Eq. Let me know if I can be of further help.

I hope this helps for you beginners !!
Attached Files
Old 4th January 2017
  #83
Lives for gear
 

Quote:
Originally Posted by TextHead View Post
Few basic n00b questions...

The amount of gain reduction is determined by ratio: a ratio of 4:1 means that if input level is 4 dB over the threshold, the output signal level is 1 dB over the threshold. The gain (level) has been reduced by 3 dB:

1. Why does the compressor still compress when the threshold is -12, but the ratio is 100:1? There is no signal that is 100db over the threshold, so why is there still reduction?

2. I'm looking at my waveform in Adobe Audition. I'm looking at everything over -12db... some peaks are at -3db. So, I set a threshold of -12db and a ratio of 9:1 because those peaks are 9db over the threshold, so they should be outputting at -11db after compression, right? But my waveform measurement doesn't show the peaks being reduced to that level. What am I doing wrong? I was just experimenting with a waveform to get a better understanding compression, but my stuff isn't getting reduced to level I think they're going to get reduced to. I must be missing a trick here. Feel like a right dummy.
The two assumptions -as to functions or how they're applied is way off (along with trying to apply that quote without the context it needed.
The 'amount of compression is based on-
-The threshold vs the level
-The ratio (of how many dB reduction per dB over threshold
-But only IF the event lasts long enough to honor attack time RATE (a time span) TO FULL REDUCTION after the event crossed the threshold.

I would suggest going back to page one (lots of basic stuff was offered right off.
How to compress?
Old 4th January 2017
  #84
Gear Head
 

Quote:
Originally Posted by Wayne View Post
The two assumptions -as to functions or how they're applied is way off (along with trying to apply that quote without the context it needed.
The 'amount of compression is based on-
-The threshold vs the level
-The ratio (of how many dB reduction per dB over threshold
-But only IF the event lasts long enough to honor attack time RATE (a time span) TO FULL REDUCTION after the event crossed the threshold.

I would suggest going back to page one (lots of basic stuff was offered right off.
How to compress?
I don't understand why that second part is wrong. Is this not correct? -->

If I have peaks at exactly -3...

Threshold = -12

Ratio = 9:1

Fast attack. Long release.

Those peaks should get reduced to -11 in theory, right? I don't see that on my waveform. I know compression is done by ear, so maybe I'm overthinking the theory of it.

I know compression is based on threshold v level of signal, but that still doesn't explain to me why a ratio of 60:1 would still compress something if the threshold is like -6. How does that work? There is no 60db over the threshold, so does the compressor revert back to the highest ratio it CAN use?

Maybe I got the maths wrong or something...
Old 4th January 2017
  #85
Gear Head
 

Well, it appears as if I'm correct. I just generated a tone and put this into practice, and those -3db peaks got reduced down to -11db. My compressor was probably doing auto-make-up gain which had me confused for a minute there.

Still... can you explain to me how a ratio of 60:1 works in the context of a -12db threshold with sound above that. Does it just use the maximum ratio it can in that instance? As there's no 60db over the threshold?
Old 4th January 2017
  #86
The ratio is just that, a ratio. Not a dB amount shaven off your signal.

Nice formula right here: The Ratio of the Audio Compressor. What Does it Really Do? – Audio Issues

O = (I-T)/R+T

So: I = -3, T = -12, R = 9 => O = (-3 - (-12))/9 + (-12) = (-3 + 12)/9 - 12 = 9/9 -12 = 1 - 12 = -11 [dB]
With R = 60: O = 9/60 -12 = -11.85

That's in theory and that's pretty much all wrong because compression doesn't just "shave off peaks" but at least you'll get why 60:1 works with your example.
Old 4th January 2017
  #87
Quote:
Originally Posted by Wayne View Post
TO FULL REDUCTION
There is no such thing
Old 4th January 2017
  #88
Lives for gear
 

Quote:
Originally Posted by vinnie2k View Post
There is no such thing
I'm not sure why you'd say that. Unless its the phrasing used? How about 'final amount of reduction' it will archive given time to honor the attack?
Old 4th January 2017
  #89
Lives for gear
 

Quote:
Originally Posted by TextHead View Post
Well, it appears as if I'm correct. I just generated a tone and put this into practice, and those -3db peaks got reduced down to -11db. My compressor was probably doing auto-make-up gain which had me confused for a minute there.

Still... can you explain to me how a ratio of 60:1 works in the context of a -12db threshold with sound above that. Does it just use the maximum ratio it can in that instance? As there's no 60db over the threshold?
My mistake lumping the two earlier statements together.
The main distinction as vinnie2k point to, is your connecting ratio as if it's related to the levels in the material.
Old 4th January 2017
  #90
Gear Head
The best explanation of compression is by Michael Paul Stavrou in the book "Mixing with your mind".

He talking about compression is 'like cracking a safe'.

Temporary settings:

-ATTACK TO ANYWHERE
-RELEASE TO MINIMUM
-RATIO TO MAXIMUM
-THRESHOLD TO SENSITIVE

Book is great....
Topic:
Post Reply

Welcome to the Gearslutz Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Forum Jump
Forum Jump