Depends what you want. If the vocals are dynamically diverse (ie you want to preserve the volume changes in the take) then you don't want to ride the compressor very hard, just even out the peaks. A ratio of 4-5 is pretty much the standard in my experience for a starting point. For the threshold, attack, and release - you'll just have to learn what they do and use your ears.

Lets discuss what the different settings do ....

Attack is how fast the comp clamps down on the signal. Fast attacks can dull the leading edge transients of the signal ... if that's what you need then do it ... if not, back off the attack a bit.

Release is how soon the comp lets go of the signal ... slow releases can make things pump and breath but can add sustain to guitars and bass. Fast releases usually work best for vocals.

Threshold is at what level the comp clamps down. Set it for 0dB and when 0 dB is exceeded the comp kicks in.

Ratio is how much the comp squeezes the signal. At a 2:1 ratio, for every 2dB the signal exceeds the threshold, a 1dB gain at the the out put will be seen. 4:1 means for every 4dB exceeding the threshold, 1dB is seen at the output. The higher the compression ratio, the more severe the compression is being applied. Ratios of 9:1 or higher are considered limiting.

Make up gain or gain is where you adjust the output of the comp to compinsate for any gain lost in the compression process.

A good place to start
For vocals, 4:1 medium fast attack and a fast release. Set the threshold so that you see 4 to 6 dB of gain reduction on the meter and add 4 to 6 dB of make up gain so the level is the same as when the comp is disengaged.

... Any other thoughts? They are greatly appriciated!!

Parallel compression is easy in DAW...copy your vocal track, so you have two identical ones (on playback, should be absolutely no difference in phase, etc...)

Then take the copy track, and squish the living crap out of it...take away most of the dynamics...in solo mode should sound like hell, but blend in just a bit of this with the original track and your vocals will remain controlled but have a background of power and chaos...

Usually works well for any kind of rock/bubble gum country....if not, then hit delete and try something else...

Like copying the track twice, leave the original alone, moderately squash one track, and then hard knee brickwall limit the other...this will give a different type of "set in the mix" than most anything you've heard before...

Midlandmorgan wrote: Parallel compression is easy in DAW...copy your vocal track, so you have two identical ones (on playback, should be absolutely no difference in phase, etc...)

Then take the copy track, and squish the living crap out of it...take away most of the dynamics...in solo mode should sound like hell, but blend in just a bit of this with the original track and your vocals will remain controlled but have a background of power and chaos...

Usually works well for any kind of rock/bubble gum country....if not, then hit delete and try something else...

Like copying the track twice, leave the original alone, moderately squash one track, and then hard knee brickwall limit the other...this will give a different type of "set in the mix" than most anything you've heard before...

Very cool, very effective. Typically, I try to get a good sound before I get anything into the mic, so using vocal technique is a great way to even out amplitube, then compression on top of that will make an even better sound.

I did not know people even used vocal techniques any more, need a de-esser? Just try to cut off the S when you sing, cut it with your hand (swipe your hand between you and the mic as you sing an S, blocking the S from entering the mic) or dont sing the S's directly into the mic, (sing S's slightly above the mic) quieter S's before the mic means quieter S's in the mix. Just be careful not to move a lot of wind past the diagram of the mic or you will get unwanted noise.

T-Slice wrote: [quote=Midlandmorgan]Parallel compression is easy in DAW...copy your vocal track, so you have two identical ones (on playback, should be absolutely no difference in phase, etc...)

Then take the copy track, and squish the living crap out of it...take away most of the dynamics...in solo mode should sound like hell, but blend in just a bit of this with the original track and your vocals will remain controlled but have a background of power and chaos...

Usually works well for any kind of rock/bubble gum country....if not, then hit delete and try something else...

Like copying the track twice, leave the original alone, moderately squash one track, and then hard knee brickwall limit the other...this will give a different type of "set in the mix" than most anything you've heard before...

Very cool, very effective. Typically, I try to get a good sound before I get anything into the mic, so using vocal technique is a great way to even out amplitube, then compression on top of that will make an even better sound.

I did not know people even used vocal techniques any more, need a de-esser? Just try to cut off the S when you sing, cut it with your hand (swipe your hand between you and the mic as you sing an S, blocking the S from entering the mic) or dont sing the S's directly into the mic, (sing S's slightly above the mic) quieter S's before the mic means quieter S's in the mix. Just be careful not to move a lot of wind past the diagram of the mic or you will get unwanted noise.Oops, that has nothing to do with MidlandMorgans post, I messed up posting in the wrong spot, sorry man.

Jon_The_Lefty wrote: ... Any other thoughts? They are greatly appriciated!!

Compressors are different, therefore the same settings on two comp gives different results. Settings could depend on many things: tempo, expressions of singer, song style, transparency, etc. There is no fixed rule for compressing but in general:

1.Be careful if compress before AD stage with hardware compressor. Overcompression can't be undone after, without signal quality damages. You have to use pretty expensive compressor (Manley ELOP, LA 2, 1176, Distrressor or similar) and mostly no more than 3-4 dB reduction, to preserve transparency, just for color and «smooth gluing».

2.Do not overcompress (breathing, pumping). Dynamic is good, don't kill that too much. If the time (money) is no object, try with gain ride (volume curve automation) on recorded take, before plugin compression. It's more natural, and adding light comp after, lead to better sound.

3. Sound must be better after comp. If not, go to start and try again. Listen very carefully. Compare (A/B) monitors and headphones «pictures» (solo vocal). How parts with different peak/RMS interact under compression? More or less natural?

4. Try with different plugin compressors. Be patient and pedant. After a while you'll find out that the secret is in small and fine adjustments. Practice, practice, practice.

One of my scenarios:

a) Slightly compress (3 dB reduction) on Elop, after preamp and before AD
b) Recording (force the singer on dynamic or record many takes to choose best parts)
c) Use EQ hi-pass for low cut under 80/100 Hz (carefully, don't change tenor into soprano)
d) Compressor-Waves Ren Comp:

Ratio: 1.5:1 (mostly) – 3:1 (2:1– 3:1 without Elop)
Reduction: 6 dB (maximum peak)
Attack: 10 – 20 ms
Release: 150 – 250 ms
Gain: make up close to zero (never over)

If the ratio is 3:1 or more, set reduction to 3 or 4 dB (maximum peak). If comp grab and squash first attack of voice (loud and long a, e, i, o, u) set attack time higher (30 ms). Look at the peak/RMS of take along the timeline. Every word – different peak/rms, so there must be some compromises in compressor settings. Well, if you are «crazy» enough, you can make automation of reduction and gain make up parameters for every word in the song. Play the game and find out interactions between parameters.

e) Deessing. First, if you have only few sibilants, try to cut with sharp volume curve on very short parts (zooooom in). This is good technique for dynamite - p, b, t, k – voices too. Waves Deesser: Find frequency (headphones) and reduce. Sometimes I have to use two deessers one for S one for SCH (different freq.). Sometimes I'm processing only short sibilant parts to preserve rest of high freq on take (be precise with left and right zero crossing points when selecting).

f) 4 band EQ for mixing (...well...another story)
g) L2 limiter (only limiter, turn off the IDR) and limit max. 3-4 dB (depend of peak/RMS of take)
h) Listen and find out the step where you made mistake = unnatural sound.
i) Now you can process (bounce to new track) through whole virtual rack to preserve CPU. Rename and save song and delete old (starting) takes from arrangements (not from disc). You can always move back to fix something.

(j) Sorry for my English :oops:

Be brave! 8-)

If you don't have a pretty expensive hardware compressor, (not everyone can just patch in a manley...) you can still use a cheaper unit like a composer or 3630, just make sure that you insert it between the pre and the AD conversion like the previous dude said. During tracking just make sure that its not peaking or falling to low, and try to do most of this with vocal techniques. If the vocalist sucks at techniqe try a 1.5:1 or 2:1 ratio with the threshhold at around -10db, then play with it til the level is sitting where you want it to. If the singer is a screamer as well, either make them do it on seperate takes, or set up a limiter that squashes right before you peak the ADC. That way it might squash, but still might be usable. Better than clipping.... My two cents.

K

Hi

Another thing you might like to try is manual compression. Look at your waveform, whichever parts seem too low just highlight and adjust the gain. Do this till your whole waveform looks reasonably even. Much the same thing as you are doing with a compresser.

Good luck

Liam

Hey Groff, Man that was impressive; you must be a pro? I have a question for anyone including Groff. I've read comments that speak about compression artifacts. What are they and are they the result of the compressor, technique, mic, or preamps? and Are they fixable or just preventable?

Kswiss wrote: ... (not everyone can just patch in a manley...) ....

Man, don't get me wrong, I'm far away from being rich and spoiled. I dreamed about my first pro level compressor soooo long. I was lucky when find used Elop in good condition for 1000 $. I borrowed some money and...... without regrets. Still it's the only comp I have.
I worked with cheap Alesis, Behringer, dbx, and they altered sound in the way I didn't liked. So, for a time, I was give up for a while and using only plugins.

If I may say, for the starter, maybe is better way to record without outboard comp and apply plugin compression after. This is more controlled situation (magic undo). Later, when become more familiar with «what compressor can do and how it sound» he could rent some mid or high level hardware comp to play with.

best regards

eBrown100 wrote: Hey Groff, Man that was impressive; you must be a pro?

Oh, I wish I am. I just give one of the rough guidelines for vox. But it’s like eq, can’t apply the same settings on every voice in every situation.

In most cases it’s about wrong settings. More compression and you lose dynamic = flat sound. Push some more and you have punching holes, pumps and breathes. You can hear it as unnatural and unpleasing rising and falling in loudness (compressing/uncompressing).

When you compress with outboard comp when recording, just don’t overdo it. You can always ad more compression later in mixing. If you find artifacts, delete the take, revise settings (rise threshold, lover ratio, slower attack (combinations)) and go again. Avoid fixing artifacts, it’s far harder way to go and you lose the quality.

Levels of threshold and ratio are dependant on the signal and character of the sound you want to create for the instrument (i.e. “squashed drums”).

After compressing, very often low level garbage noises comes up. It’s about acoustic and electric environment.

Spend some time with “search” engine, there is many other very usable tricks with comp from guys bigger than me. Surf through the Net too. I’m still digging and learning.

see you

Hey Groff; I experimented with compression on vocals and percussion. Percussion is a little easier to manipulate b/c it's dynamic range is limited and it's purpose in the music still can be fulfilled even if they are not perfect. But vocals ofcourse require more attention b/c if they are not processed right its distracting. Groff, (or anyone else who is knowledgeable) explain how you use multiband compression (or eq) to process male volcals. More specifically, how do you tame the bass frequencies without making them sound thin? In one of my experiments I eliminated some of the bass frequencies with some eq and multiband compression (done with the Izotope's Ozone 2 plug-in). But the vocals sounded thin so I put a touch of reverb on the lower frequencies and they filled out a little bit, but I was still dissatisfied.

eBrown100,

Don't use compressor on everything you have in the song, it could be boring and unexcited. Compress only if you have to do or if you like the sound of compressor (as color).

I don't compress percussion, that helps me to have greater overall dynamics of the song. But if the peaks are almost all what you have (very low RMS) after recording, change the mic or mic technique. Limiter could help, but better way is to find "healthy" signal at the starting point.

Multiband comp for vocals? Probably I'm wrong, but never saw big guys with multi on voices. I'm doing just with one band.

And the best part: There is no way to make top of the world sound only with comp and eq plugins if you have crap mic and pre. Can't make something from nothing.

What's your mic and pre? Any acoustical treatment in rec room?

Bass freq on voice....Well, high pass (low cut) filter (eq) should be enough. You have to find a freq point and proper Q shape where the sub-garbage is gone and the voice remain "ballsy". Usually around 80-100 Hz. Do it before comp.

Maybe you are too close to the mic (proximity effect) or the room is too small and full of standing waves or too big and noisy with large reverberation...

Don't hide behind the reverb or effects. Search for the best fundamental sound without effects.

good luck

a question that should be asked:

what type of music do you mainly record?

I ask this because a lot of this advice might be off the mark depending on your style.

metal=highly compressed, and most everything in the mix is compressed

country=way less compression on everything