mixer compression question

[phantasm777]

my yamaha mixer has a built in compressor for 2 channels. i know what a compressor can do after the fact of recording a track, but why would i use it during recording? wouldn;t it be best, as in effects, to use compression AFTER you record a track and not during?

[AndyR]

I'd actually like it if my mixer had a compressor on two channels

(EDIT: Sorry, long post again. SHORT answer, it's there if you want to use it to reduce annoying peaks and clipping on the signal on those channels which mean you can't turn them up enough... or in our case, record the performance loud enough)

I probably wouldn't use it on most instruments. But I would use it on two main things: vocals and resonator guitars.

These both have quite aggressive peaks that will need sorting out eventually anyway. But on the input to the recording device the peaks can cause distortion and clipping unless you lower the levels somewhere in the chain.

Lowering the levels (turning the gain or whatever down) doesn't help the signal-to-noise ratio. You get a consistent amount of noise from any stage in the chain, and reducing the actual signal at that stage doesn't reduce the noise as well (it only reduces any noise already picked up from a previous stage in the signal chain).

So, when you boost the signal later (eg in the mix after compression/limiting), that original noise gets boosted to much bigger than it was. Not necessarily a big deal if it's on just one track or two, especially for home recording quality like we're all doing. But that noise does add to the general confusion and lack of clarity in the final mix.

The solution is to maximise gain levels throughout the signal chain. On a spikey performance (vocals or resonator for me), this means applying gentle compression (or limiting really) to the source as soon as you can in the signal chain. If you gently smooth the hottest parts of the performance you can then push each gain stage further without peaking. This means you get the hottest signal you can onto "tape" with as little noise as possible.

When I treat my vocals like this, I use the following (got it from various sources, but a post from Flash Harry a while back finally helped me understand what was going on and therefore exactly what sort of settings reach for - thanks Flash!!)

Threshold: -6db at most
Ratio: 8:1 or above (this is turning the compressor into a limiter)
Attack: Fast as possible
Release: Similar

That sorts out the peaks. There's also the "noise floor". This is about losing as much of the background noise on the source performance. So, rustling shirts, kids screaming in the street outside, general hiss form the room/mic, guitar humming while at rest, wife trying to attract your attention... all that annoying stuff that gets on your recording.

For this you use an Enhancer. It's kind of a subtle noise gate. Now, I don't know as much about this because I found it far too fiddly to get it right... and for my home recording stuff, the benefits seem to be minimal for the effort involved in trying to figure out how to set it!! (So I don't actually bother anymore)

Now you can do all this to the track once it's recorded, especially with digital recording. But you're doing it at the expense of increased noise in the final recording. If you want to maximise the signal-to-noise ratio, get as hot and clean a signal onto "tape" as you can, then the best place to do the initial compression/limiting and expansion is right at the start of the chain (before the mixer channel's EQ and fader, before the mixer's output fader, and before the input level of the recording machine).

I suspect that's why the desk has it there.

[phantasm777]

i dont have to worry about peaking too loud, i don;t have the vol. up that much. what i am wondering is fi using this compressor on vox would it be helpfull for the fact that my voice volume itelf varies, being a new singer, i sometimes unknowingly sing a little quiter than i realized at the time. would the compressor bring that up more and bring down the loud end? if so then it would be perfect for my vox.

also i have these buttons - the 80 switch (high pass filter) it cuts off frequencies below 80 hz. what instruments would i even need to use that for?

[64Guitars]

my yamaha mixer has a built in compressor for 2 channels. i know what a compressor can do after the fact of recording a track, but why would i use it during recording?

You should always record at the highest recording level possible without getting clipping. If the sound source you're recording has lots of high peaks (close-mic'd vocals, for example), then you'll either get clipping on those peaks or you'll have to lower the recording level till the peaks fit. If there's a huge difference between the highest peak and your average level, then the average level will be too low. The compressor on the mixer allows you to reduce the difference between the average signal level and the highest peaks so that your average signal level will be higher, giving you a better recording. Of course, you lose some dynamic range from the compression, so you don't want to overdo it. You just want to compress the peaks enough to get a good average recording level.

[AndyR]

^ exactly what he says

And, Phantasm, yes, it will help on your vox - you'll need to fiddle about and experiment a bit, find how to make it work for you. And it's not just new singers ... the less new singers need it as well

Not sure whether I'd use the 80Hz cut at the input myself (but if it was there I might).

It is definitely used on vocals - takes out the effect of breath on the mic for example, or the singer tapping his/her foot and that coming through the mic stand. I only use it if I have a problem on an otherwise usable recorded vocal. So, personally, I'd probably only ever use a cut like this on a vocal after the recording.

In rock music you cut the bass guitar around there as well (I tend to cut below 50Hz, but 80's OK too) - stops the bass booming without losing anything you can hear, and it makes room for the kick drum in the mix. Again, I'd probably do it after recording the track - but I nearly ALWAYS do it, really cleans up the mix and lets me get louder over all.

A lot of people (not guitarists themselves usually ) also rip the bottom out of guitar parts - anything down there on a guitar part is not audible in the mix and just serves to confuse and overload the bottom-end of the whole thing.

Having that "cut the bottom" on a channel means that you don't have to use the EQ controls on the channel strip to do the job, they're free to be used for the EQ you want. So, on a vocal, for example - put the 80Hz cut in place if you need it, and you can still use the EQ controls to add a little boost (or cut) at 200Hz to warm up (or de-muddy) the vocal. If you didn't have the switch, and you had to do the cut, then you'd have to use the EQ to do it, and you couldn't use the EQ for anything else.

It's starting to sound to me like this desk is also designed for live use - in that case the finished product is what's leaving the desk, rather than feeding into a recording machine for future "fiddling with".

Most of these 80Hz cuts I describe, I'd want to do later during mixing. There is always a possibility that you might want the boom or rumble later, and if you cut before recording, it won't sound as good trying to put it back again later during the mix.

But the compressor on there I would use. With some experimenting, I think you could definitely make that compressor work for you and help you get more consistent vocals (that's what I use them for... you should hear the jumble actually coming out of my mouth ... what ends up on a finished recording always sounds a lot more polished than what I sang in the first place...)

[phantasm777]

on my cover song - the cheater - i use the built in compressor for my vox for the first time.