Why track vocals with compressor into a 24 bit system?

[percussion boy]

There is probably some really obvious answer here (e.g, "cuz it sounds good"), but nonetheless:

Why does pretty much everyone in the GO TO VOCAL CHAIN thread -- pages and pages of intelligent people -- use a hardware compressor in the vocal chain, when the audio's destination is a 24 bit digital recorder with enormous dynamic range?

Not that you would never compress, but why compress right away?

Thanks for illumination.

[Waltz Mastering]

It would make more sense if you were going to compress vocals on the way in, to do it to digital. Tape naturally compresses depending on how hard you hit it.

I've never tracked a vocal with eq, but always have tracked vocals with compression... basically to control the dynamic range.

[thethingwiththestuff]

because uncompressed vocals sound like ass! unless someone's going for a "58 straight into a high school PA" kinda vibe, or doing opera, it just sounds cheap.

maybe Albini can get away with minimal compression sometimes, but often the singer's a screamer, or there's a lot of room sound to thicken it up. show me one guy who loves albini's drum sounds and i'll show you five singers who hate his vocal mixes.

don't fear the compressor.

[kslight]

Besides what was said previously...maybe you don't want the added a/d and d/a conversion stage to use the same compressor...and/or you don't want to use a plugin compressor. Dead obvious to control dynamic range...if you have a really dynamic singer and/or song will give you more data to disk...so the lower dynamics won't have a severely shortened word length and to avoid distortion on louder dynamics. If the singer gets really loud and really quiet you will have parts where they are only using 4 bits and parts where they use full range.. Not to mention that many people like to drive their preamp into saturation so you can curb your peaks a bit and hit the gain harder. I do it because I like to get my sounds while recording not while mixing...I also track with EQ and compression on drums and bass, or any other signal where I know I'm going for a certain sound.

[chris harris]

I do it as a tone shaping tool, to get the sounds I'm after in the beginning. But, to be honest, there are plenty of times when I track vocals straight from the pre to the converters and do my compressing at mix time with either outboard, plugins, or (gasp!) BOTH!! I do think that some people just get in a habit of doing things the same way every time. I bet that a lot of people still compress on the way in to "keep from clipping". I also know that a lot of people (including myself when I was starting out) do things by rote as a way to avoid truly understanding what they're doing. But, as with everything else, once you know what you're doing, you know that you have to first listen to the sound that you're recording and then make a decision about how to best capture that sound. So, as per usual, the real answer comes back around to "it depends".

I don't care at all about "extra ad/da conversions" or "using all the bits" and I don't think that in 2010 that anyone else should either. Those are holdover concerns from an era when converters sucked and noise was a huge concern. These days it puzzles me to see people worrying about theoretical signal degradation that they will never, EVER be able to hear, and using that concern to justify avoiding techniques that could potentially really benefit their recordings.

[percussion boy]

I do it as a tone shaping tool, to get the sounds I'm after in the beginning.

Okay, that makes sense.

These days it puzzles me to see people worrying about theoretical signal degradation that they will never, EVER be able to hear . . .

--And that was what raised the question. It's amazing how quiet recording equipment can be nowadays, including some of the less expensive stuff.
24 bits = lots of dynamic range to play with, if you don't hit the system too hard by cranking the input levels.


Thanks for all the inputs . . .

[percussion boy]

because uncompressed vocals sound like ass! unless someone's going for a "58 straight into a high school PA" kinda vibe, or doing opera, it just sounds cheap.

maybe Albini can get away with minimal compression sometimes, but often the singer's a screamer, or there's a lot of room sound to thicken it up. show me one guy who loves albini's drum sounds and i'll show you five singers who hate his vocal mixes.

I was more tripping on being stuck with the compressed track only, vs. having an ugly uncompressed track and fixing it after.

Never cleaning up vocal levels seems kinda rude, like refusing to wear deodorant in public.

[drumsound]

I like the vocal to be on top. It's the melody after all. If I limit the dynamic range of the vocal and I just push the fader up and have it sit. I often track with two compressors, and sometimes add one in the mix.

[farview]

I do it because of the sound of the compression not because I need less dynamic range. That just happens to be the sound I go for most of the time.

[roscoenyc]

maybe Albini can get away with minimal compression sometimes, but often the singer's a screamer, or there's a lot of room sound to thicken it up. show me one guy who loves albini's drum sounds and i'll show you five singers who hate his vocal mixes.

don't fear the compressor.

He makes that shit up for interviews.

[Fletcher]

Several layers of a "little bit" of compression will have an overall lesser negative effect than one layer of "a buncha" compression... so a little on the way in, a little during the mix, maybe another when you're doing a vocal compilation to help smooth things out and it comes out good.

I have [many times] chained 2-3-4 compressors each working a db or two for the cumulative effect of "a ton" of compression... but not so much so that it will end up sounding like Nickleback record.

This is also predicated on the employment of really good sounding units... I definitely would not employ this method with a DBX-160X anywhere near the chain.

Peace.

[thethingwiththestuff]

Never cleaning up vocal levels seems kinda rude, like refusing to wear deodorant in public.

hahaha.... yes, i agree. i should have said along with "dont fear compression" that there's no reason to fear using it right off the bat on your way into the DAW. if you know how to use one, you won't overcompress and come mix time, you'll be halfway there.

and like flethcer said, a couple comps doing a little GR each usually works nicely on vocals, so why not get one level of GR going during the take, and letting the vocalist hear himself a little better in his cans?

[MoreSpaceEcho]

it's a rare singer who doesn't prefer the sound of their voice with some(times a lot of) compression.

[;ivlunsdystf]

A lot of these responses are missing the essence of the question I think. It's not 'why do we compress vocals?'; it's 'why do we compress them on the way in, when we could easily put off that step until mixing?'

I use a hardware compressor lightly on the way into the machine because I'm going to end up compressing later anyway, and it makes monitoring much easier during the process of actually singing. The quiet bits are more audible etc. What I am used to.

[thethingwiththestuff]

right. i think some people take committing sounds to tape for granted and kinda forgot to explain that part. waiting till the mixing stage to process anything is just putting off the inevitable and adds time to the mixing process. if you're comfortable with your gear, you can use it wherever you want.

i think a larger issue of the appropriate amount of dynamic range for a given style is worth thinking about too, though. i think of 24 bits as leaving me plenty of headroom without noise concerns, not necessarily dynamic range. it gives me the option to keep a good gain structure while making things as thick and forward as they should be for pop/rock derived stuff. i really dont want that large of a dynamic swing on almost any instrument in a mix.