Compression usage

N

nbtech_2001

New member
Hi,

All I'm doing mainly is recording a live track from an electric piano (One man band piano) and vocals together...again live. It's kinda like recording kareoke I guess, so because I'm crazy and over approach things I think that I have over done the compression a bit and over complicated when and how to compress everything. It's mainly the vocals I am worried about but here's my question: Do I tame the dynamics by compressing the voice and then compress again the whole song to master? Or do I leave the vocals alone and compress the two tracks as a whole to tame the dynamics (Which there is no dynamic issues with the piano, it already seems just fine) Also it would be amazing if I could get someone to divulge how they would approach mixing (from EQ application to comp. Multiban comp) this SUPER simple two tracks? One last thing since I got you, How deep a pocket (in Db) do you usually cut out to sit the vocals in the mix?

Thanks!!:D
 
I would question the need to compress an electric piano at all. But if it did need compression, I'd wait until mixing to play with it.

Compression on the vocal is more likely, and - assuming an outboard compressor - I'd probably just go ahead and compress it on the way in. The exception to that would be if I was not yet too sure of the compressor and what I was doing with it. In that case, I'd try to record dry and compress in mixing instead.

But either way, I would not just automatically treat the two tracks the same.

As far as setting the levels, there's no set answer to that. You do what you gotta do for it to sound right. I'd throw away the notion and the mental image of a "pocket" in this case, and just get them to play nicely together. Remember, when you are on stage playing for an audience, there is no "pocket"; there is just you and the piano. There is no reason why your recording can't be the same.

But if you want them to have "room" to do their thing, there's always panning. Throwing your voice to one side and the piano to the other is a common scheme. Also, if you have a stereo piano, throwing it L and R and the voice down the middle is another option.

And remember "side" does not *have to* mean 100% to that side.

G.
 
great, thanks..So yeah, the piano I never really felt like it needed compression (it's a solid level) I use a hardware comp with Limiter settings on the way in for the vocals. Naturally the electric piano is stereo coming in and the vocal is mono, would you mind explaining how I would narrow the vocal to sit in the middle of the piano? Essentially a mono signal is L and R exact copy 100% left and right. So how would it be sitting in the middle and not ON top of the stereo L and R of the piano?
Thanks

oh, I'm using Logic 9 software compression during mixdown
 
There's no need to "narrow" the vocal. You're exactly right that a mono signal that's centered is actually created by sending equal level to the left and right channel, but the result as it sounds to the ear is what counts in this case. As far as your ear is concerned, the vocal *is" centered down the middle, and therefore will not step on top of a true stereo signal that's panned L&R.

If you don't like the way that comes out, you can alwys put the vocal off to the left, say, and take the left side of the piano and throw that down the center with the right side down the right. That will give you the vocal on one side and the piano will still be in "stereo", but just shifted over to the right.

G.
 
Good stuff. So what factors would make me determine using either the pan method or an Eq pocket method to make the vocals sit and everything else fit. I know that it's a very subjective thing to determine, but what would make you do one or the other?
 
I don't do "pockets", personally. Those are probably already predetermined by the arrangement.

Either you and the piano are going to be working in the same key or you're not. If you are, then there's no "pockets" to be had, probably, and most of the spacing between the two is going to be in the arrangement and the physical pan space. If they are not working in the same key, then there's probably not much to worry about in the way of a "pocket".

The thing when working with vocals and piano together is that it can be real easy to make either one sound crappy by trying to carve them up into "pockets" with EQ.

G.
 
One last thing...How do you feel about sidechain compression to clear the way for vocals, sometimes I think I hear this going on in popular songs (Not the Trance sidechain compression) I could be wrong though?
 
nbtech_2001 said:
One last thing...How do you feel about sidechain compression to clear the way for vocals, sometimes I think I hear this going on in popular songs (Not the Trance sidechain compression) I could be wrong though?
Click to expand...
Here's an answer that many on his board will probably hate, but in all the years I've been doing this stuff, I have *never* come across the need to sidechain anything when producing/engineering music recordings. For some broadcast uses, perhaps, but I have never found anything that I wanted to do in a recording production that required sidechaining anything to get it done.

That's not to say that it's a bad thing or that some folks find it legitimately helpful. Just not for me. Yet.

G.
 
It can happen ..'ducking, rather unintentionally. It is to some extent (or can be) a side effect' if you will, with straight in-line comp.
Whenever you have situation where one thing- the vocal' in this case is sticking out bit or hot' re the other parts in the mix, and in a way where it's poking out and the timing and reaction of the comp is such that the comp is triggering off of it, you have essentially that comp to some degree 'ducking the things below it.
The extent.. depends on threshold, ratio, timing of both, how far 'up that thing is, etc.
Take a look at your mix comp, its timing (and presumably we're talking 'not so fast it's triggering off your typical kick/snare hits here..) and consider what might be driving it.
 
Thats kinda the way I always felt about it (Parrallel and sidechain) I have done it before but it never seems to do anything good besides make the vocal stand out WAY too much, I feel as though if you have recorded it going in the right way then you shouldn't really have to sidechain or parrallel compress anything...So is it safe to say that all these methods are merely ways to fix a bad recording or basicly being stuck with working with stems from someone else where you don't have the means to "re-record"?
 
Personally I would have to say I don't believe that sidechaining or parallel compresion is solely a mean to fix a bad recording. There are tons of creative uses beyond simply ducking a bass to a kick or instrumentation to a vocal or fattening up a drum track.

Many people find uses for these techniques and equally many people don't. It's more a question of finding your own path and style of mixing
 
nbtech_2001 said:
So is it safe to say that all these methods are merely ways to fix a bad recording or basicly being stuck with working with stems from someone else where you don't have the means to "re-record"?
Click to expand...
I don't thik so. They both are far more often used (when used properly) to either achieve a type of sound that you can't easily get in a vanilla track, or to quasi-automate a post process that otherwise might be kind of time- or labor-intensive to manually execute. IOW, they are not meant so much to "fix" bad tracks as they are to provide a special effect or sound using a certain technique.

While Ethan and I agree on the sidechaning thing, we diverge when it comes to parallel compression. Parallel compression IMHO can provide a nice custom sound to some tracks in a way that's pretty fast and easy to execute. But like with *any* effect, moderation is not a bad idea, IMHO.

But it's really a personal thing. Some folks prefer some tools over others while other folks prefer just the opposite. What I say and what Ethan says in this regard are just reflections of our personal preferences, not litmus tests of what's right or wrong to do in general.

G.
 
SouthSIDE Glen said:
I don't thik so. They both are far more often used (when used properly) to either achieve a type of sound that you can't easily get in a vanilla track, or to quasi-automate a post process that otherwise might be kind of time- or labor-intensive to manually execute. IOW, they are not meant so much to "fix" bad tracks as they are to provide a special effect or sound using a certain technique..
Click to expand...
Along these lines what I recall as a live Bob Dylan track on the airs several years back (..maybe than 'several?) where his vocal road over a huge dense electric backing.
It was obviously heavily ducking the band, (easily sounding like an 'effect.) But it worked wonderfully IMO at the time.
 
nbtech_2001 said:
Do I tame the dynamics by compressing the voice and then compress again the whole song to master? Or do I leave the vocals alone and compress the two tracks as a whole to tame the dynamics (Which there is no dynamic issues with the piano, it already seems just fine)
Click to expand...
If the dynamics are a problem on the voice and not piano, then apply dynamic control to the voice and leave the piano and mix buss alone. Never do something unless you hear a reason to do it.
Also it would be amazing if I could get someone to divulge how they would approach mixing (from EQ application to comp. Multiban comp) this SUPER simple two tracks?
Click to expand...
I'd pull the tracks up, listen for what they need, and then do that.
One last thing since I got you, How deep a pocket (in Db) do you usually cut out to sit the vocals in the mix?
Thanks!!:D
Click to expand...
Typically... None. I wouldn't record something that buries the vocals in the first place. If I did have a track that buried the vocals, I still wouldn't jump right to EQ. Many times a fader move is all you need.
 
It seems to me you're overthinking this. Electric pianos, and even mic'ed acoustic pianos contain most of their information in the middle of the stereo field. They still feel and sound like they're filling the entire stereo field, and there should be plenty of room for the vocal. Make the electric piano sound good, and put that vocal right up the middle.

If you've overcompressed in the recording process, then you will likely have artifacts and breathing. Now, plug ins don't tend to cause the same kind of artifacts and breathing as hardware units. 1176s get grainy and have obvious EQ artifacts which usually cause a reduction in high end information when they're being hit too hard. Not so with any plugin 1176 I've used. This is an effect that can be used for the benefit of certain tracks, or it could be a sum negative. It seems you've determined it to be a sum negative, and your only choice is to automate the vocal, and possibly split the vocal onto multiple channels with different EQs.

Vocal splits are a last resort, but it's also a necessity, particularly with an inexperienced vocalist and/or recordist. It's not a bad idea to compress or limit a vocal going in, so don't let one failure in this regard prevent it in the future. This is all part of the learning process, and if you don't make mistakes, you never learn, so don't beat yourself up about it.

If there is some dulling of the high end in parts that are overcompressed, split the vocal on those spots, and add the missing high end. You'll also want to ride the shit out of the vocal, as it's probably inverted, meaning the loud parts seem smaller than the soft parts. You can also set a compressor on each split in order to try and match the differences.

I ride the shit out of vocals (unless I'm looking for a particularly spitty, breathy and affected vocal). It takes some time, but if you get the vocal to sit perfectly in the track, you will achieve what you're looking for. This can only be achieved by reducing the true dynamic of the vocal, and that can be done by riding and by more compression (believe it or not).

A vocal with one instrument shouldn't need to be scooped out with EQ. In fact, if you make the vocal big and beautiful (and big relates to having a healthy dose of low end, not necessarily reverb and/or delay (which may or may not be appropriate)), you'll be able to sit it right on top of that electric piano. Once you have the vocal placed, you can investigate riding some of the e. piano, down if it's stepping on the vocal, and up if it's offering something interesting and effective in the vocal holes.

Compressing the 2-bus doesn't hurt either, but I can't stand plugins for this purpose. YMMV.

What's important is keeping the listener's focus on the vocal throughout the song, and using the accompaniment to help push the listener forward through the song at the appropriate moments. You want to avoid ripping the listener's head off. Believe me, when the vocalist starts belting, you can keep it at a similar level to the verse (perhaps slightly louder), and the illusion of dynamic will be achieved.

Enjoy,

Mixerman
 
Last edited:
You must log in or register to reply here.
Back
Top