How much is Tooooooooooo Much Compression

  • Thread starter Thread starter radzikk
  • Start date Start date
R

radzikk

New member
I'm just wondering, I'm very new to the idea of compressing vocals or any mic-ed inst for that matter. And it seems like I compress vocals on the way in and than add more later to get them to sit better in the mix.
So I'm wondering if I am going too far. How do you know when you have compressed a vocal to far? Does it start sounding unnatural, because I don't think I pushed anythig that far yet. And I have pushed it pretty far.
Any ideas, Splain to me--I think I have pretty decent ears, but I do find that different effects take time to hear and understand fully.
Thanks,
-k
 
heres a good thing to do.... take something, and compress the hell out of it... max out the ratio and threshold. listen to it, hear how it "pumps"... pumping is a bad thing. So to your question... there are no set rules for compression, just use your ears... i usually compress to where i like it, then back off the threshold or ratio a little. for vox, i might have a 5:1 ratio. good luck.
 
Your sound will start sounding lifeless - as if someone took the colour out of a picture. Right when you take that sort of compression off, the audio springs back to colour. I love the bypass button on my RNC as I will A/B between compressed and not. You will certainly know when its too compressed. The pumping and breathing, to me, is more a screw up (unless intentional) on the attack and release settings more than overcompressing - but in any case - its usually not a good thing.
 
I'm running a RCA Varacoustic into a Joe Meek 3Q (which has some sort of variable compression thing) and I have to crank the pre gain (ribbon mic thing) and I run the comp around 5. But then in the mix it needs to come forward a little more so I add a Waves comp at like 2.5:1 and it sits better, more up front. But I think I know what you mean about lifeless, you lose the dynamics in there. I guess I should try pushing some other things back first.
Thanks,
-k
 
There is no set rules on compression. You can run it as squashed as you like, or as light as you like. What works for the track is what works. There is a big difference between compression and limiting however.

Compression, or pure compression, that being ratios under 10:1 are best used as an effect. As such, use as much of it, or as little as you see fit. The ear knows best, and it will tell you what you want to hear.

As for limiting, that being ratios over 10:1, well they can squeeze off a good deal of the signal, but it has its uses. Careful use of the attact and release are important to gain control of the signal as to when you want the compressor or limiter to start working, and then letting it go.

It takes time to get good at it, so the more you use it, the better you will become at it. Then you will find out that on some tracks, less is more....
 
Great replies so far. I've found the most critical areas of compression are the attack and release settings. Many people (including myself a while ago) mess up stuff with bad settings on the attack and release, even more than ratios that are too high.

If you want a really nice test to understand what compression is doing start with a bass (either an electric or a keyboard/sampled electric bass) patch it through to the compressor. Start bringing down the ratio till you're getting about 6 to 10 dbs of gain reduction on the meter. It doesn't matter what the attack and release settings are now.

Now, start playing with either the attack or release settings (one at a time). I bet you will notice the most with the release settings. At higher release settings you should notice the sound thickening up bit. It's kind of hard to descibe, but it feels like the envelop of the sound is ending longer/slower. Notice at lower release settings, the effect of the compression ends quicker and the envelop faster. use the bypass switch and you'll notice at faster release times the compressor is actually changing the sound to be more percussive than the original. At a certain point, pumping will occur because it will sound like the compressor can't keep up with the sound (because it's leaving the sound earlier than the sound is sustaining).

Try this and also play with the attack. Think of the attack as when (in time) the compressor engages. At slower attack times, not much of the sound is passing through the compressor. Faster times, the compressor starts up a faster with the sound. Compare faster attack times with slower attack times, you'll notice a dulling of the high frequencies at lower (faster) attack times. Focus on the high end of the bass when doing the comparison of the faster and slower attack times. I've found electric bass is a more useful sound for learning compression. Once you learn on this, you'll better understand what compression is doing on the voice.
 
Just wanted to say that I read this board all the time, and almost never post anything (my knowledge is so limited). But it's threads like this one that keep me coming back again and again. Good info and very practical. For a Newbie like myself it's the kind of stuff I need. I'm learning tons. Thanks to everyone who takes the time to share their experience.
 
Just read over my last post. Edit it to this at the second paragraph.

If you want a really nice test to understand what compression is doing start with a bass (either an electric or a keyboard/sampled electric bass) patch it through to the compressor. Start with a ratio of maybe 4:1 or 6:1. Start bringing down the threshhold till you're getting about 6 to 10 dbs of gain reduction on the meter. It doesn't matter what the attack and release settings are now.
 
Rev E said:
Now, start playing with either the attack or release settings (one at a time).
Click to expand...

Rev E has really good advice here. One thing I would add is focus on a single bass note, and zoom it a lot so you have msec resolution. Compare the before & after waves, that can be informative.
 
You might also like multing, which is as follows: you compress the vocal, but let the output of the compressor return on another channel of the board.

You can add the compressed vocal to the uncompressed now, mix to taste.

Peace, Han
 
Multing the compressor also sound really natural, plus you get the added benefit of compression. Multing also works really well for drums.
 
Compressing on the way in!

OK, just my two cents...
Do not comperss to tape, DAW, whatever you are recording to. If you need to controll peaks, use peak limiting. Set the limiter's threshold a few db lower than your "over", and do what the pros do; ride the gain. Using a compressor to ride gain automatically will only render your vocals with less dynamics, or dull. The limiter is just a safety, in case the vocalist supprises you. If you are recording yourself, use the mic like an instrument. Back off when you belt, get close for that proximity effect when you want it. Set the threshold a little lower. Hot, Flat, and Dry gives you the most options in mixing and mastering. Remember, what the compressor is doing, in essence, is allowing you to raise the level of quieter program material while keeping the louder stuff from going up as well. It's like an automatic fader, and the same can be done with a fader and should when possible. The compressor catches what you can't. This is all assuming you want a transparent sound for vocals with lots of dynamics. Compressing bass or guitar or drums is a different kettle of fish. Bass is usually compressed to remove the dynamics and smooth out the sound. For guitar its usually added for sustain.
All this being said, some famous engineer said if he needed to put the knobs on backwards to get the sound he was after, and it worked, then it was right.
The trick is to have clearly in mind the sound you are after, and tweek from there. Also, with the way people are squashing their mixes to be the loudest song on radio, I often wonder "why bother?"
Good luck
 
:eek: I'm way too afraid to limit to tape. Personally, the effect of a bad limit is even more disastrous than overcompression. Most limiters aren't too nice and they never play fair. I like gentle compression on the way in (if you know how to use a compressor). And a bit more gentle compression in the mix. But to each his own.
 
Just talking about peak limiting.

I'm just talking about peak limiting for safety. That means the threshold is set for just below what your peak overload level is. The limiter should not engage unless you are going to distort anyhow, so it may save a great take from being useless. :)
 
I compress a lot. I have often compressed vocals well over 20 dB, but you need a good compressor and a good ear to make sure you do it well.

As far as how much is too much? Just imagine how much compression you think is WAY too much and you will have about half the compression used by most of the mixers doing all the big hit rock records. I love a lot of Jack Joseph Puig's mixes, but it sure is an exercise in insane amounts of compression.
 
I compress my vocals a lot too. To tape and after the fact. You just have to be careful. I've had takes ruined from too much/lousy compression. Especially when I was using an awful compressor like the Nady I had (forgot the model number).
 
I never thought this would get this many responses. When I think too much compression I hear that "THIS SUNDAY, SUNday, sunday" thing in my head. I guess its like any other effect. I look back at stuff I recorded a few years ago and thought I put just a "touch" of reverb on...I listen know and try to figure out why a vocal dosen't sit in the mix and there it is, I was recording from the far side of the Grand Canyon...
-k
 
Charles Dayton said:
I'm just talking about peak limiting for safety. That means the threshold is set for just below what your peak overload level is. The limiter should not engage unless you are going to distort anyhow, so it may save a great take from being useless. :)
Click to expand...

I figured you were talking about peak limiting. But with 24 bit recording, there's really no need to be recording takes as close to zero as in the 16 bit days. The theoretical dynamic range of 24 bit recording is 144 db... probably more like 120ish in the real world due to limitations of modern A/D converters. Even at -8 below zero, you're still above the theoretical dynamic range of 16 bit recordings (96 dbs). So you're still preserving enough to fill up all 16 bits with data when dithering down to the 16 bit world.

The reason I'm making such a big deal about this is because peak limiters create a very drastic effect on the material, including doing weird stuff to the noise floor when the limiter engages. In the past I've done what you do with peak limiters, and I much prefer recording with a bit of mild compression and/or recording safely below zero. But if limiting works for you then great! :)
 
You must log in or register to reply here.
Back
Top