How much GR do you use on vox

[wesley tanner]

for the average poppy song/rocky song, i'm lookin at like 12-15+ dbs of gain reduction on loud notes... this has helped me immensly with making the vox sit on top of a mix. i'm just curious how everyone else comps vox 'round these parts. and don't give me that 'for every project' bs. everyone has some habits, some things that work well for them. i'm talking about rather complex mixes here, in the pop/rock/hardrock genres.

[stash98]

uhmm..i have never once gone by db..i always go by my ears..the vocal has to sound level and every syllable must be clear. not too loud or too soft....the only thing i do is find the loud waveforms, and when i draw automate them, i draw the line to go under the wav, and come back up with it, so it sounds smooth. this is a pretty common technique that is used, but maybe that helps....itsvery hard to describe, but its like you are drawing the volume to shape the wav, just at a lower db.

[chessrock]

Originally Posted by wesley tanner

for the average poppy song/rocky song, i'm lookin at like 12-15+ dbs of gain reduction on loud notes...

That's sounds about right for a dynamic singer.

[wesley tanner]

Originally Posted by stash98

uhmm..i have never once gone by db..i always go by my ears..the vocal has to sound level and every syllable must be clear. not too loud or too soft....the only thing i do is find the loud waveforms, and when i draw automate them, i draw the line to go under the wav, and come back up with it, so it sounds smooth. this is a pretty common technique that is used, but maybe that helps....itsvery hard to describe, but its like you are drawing the volume to shape the wav, just at a lower db.

yeah, i've done this before, got the tip from a berklee friend. i've found with many vocalists (halfway trained/experienced ones) that finding the right compression is more natural in the end. plus, then i can use the automation for changes in the overall dynamic in the song, and then, if neccessary, do word/syllable automation. some singers just DON'T know how to sing through a line without making some things jump off the page.

edit: i used to never watch db on anything, just listened, but my mixes improved MASSIVELY when i started paying attention to the maths that are involved as well as using my ear for final judgement. the most obvious part of this is with gain stageing from gear to gear and then trim to plugin to plugin to master/track to trim of buss, to buss plugs, to buss volume, etc etc...

[Dogman]

I won't say I watch the db on my vocals, and try to record them at a certain db, but I do try and track them quieter than I used to. I used to try and go as close to -3 as possible, then I got to play with some trackes done by scrubs, and his were very quiet, yet it was easy to put them in the mix, and turn them up a bit, so I tried it, and it was easier to get the mix to gel. I now set stuff up to record a bit quieter, maybe -6, or even down to -12, and it does seem to work better for me. And if I have a good take, and it's real quiet, I still keep it and try it. Seems with digital stuff, the clarity is still there, even at lower volumes. This is probably all different if you use any analog stuff. Just my $1.82 worth.
Ed

[stash98]

even though I track in 24 bit, -6 is always my peak. It is a consistent thing w/ me, and it keeps random peaks from popping up over 0. If you are tracking, and it hits over 0, then your take is screwed.

[Massive Master]

Damn straight. Where's the "ThumbsUp" icon in this place...?

[HogansHiro]

Tanner.....new icon? Racoon on crack?

[masteringhouse]

Originally Posted by wesley tanner

for the average poppy song/rocky song, i'm lookin at like 12-15+ dbs of gain reduction on loud notes... this has helped me immensly with making the vox sit on top of a mix. i'm just curious how everyone else comps vox 'round these parts. and don't give me that 'for every project' bs. everyone has some habits, some things that work well for them. i'm talking about rather complex mixes here, in the pop/rock/hardrock genres.

12-15db seems a bit much to me, I think that I would try a little bit of automation at that point. But numbers don't matter if it sounds good to you.

[wesley tanner]

it has nothing to do with TRACKING. I ALWAYS am paraniod about not peaking when tracking, I'll do whatever it takes so it doesn't peak anywhere in the chain... but that's all about gain staging. i'm just talkin' about mixing. tracks all done, you set the threshold and ratio to get x amout of GR.

[xfinsterx]

It depends on what compressor youre using.
In PT ill use the Waves Rvox comp and drop -6 into the vocal and itll sound great.

Ill turn around and drop the same GR on the same vocal with an outboard dbx and it sounds like way too much.

Its a pretty subjective question.

Invariably though, if it sounds good....go with it.

-Finster

[wesley tanner]

i only have experience(in recording situations) with various plugin comps, and the results have been pretty similar with each of them.

[ds21]

I never pay attention to the dbs either, if you don't want to hear the "depends on the track/singer" answer, then I guess I can't say anything. Sounds like your more into limiting.

It might be intresting to post this in the PMC contest and ask people after they've mixed their versions of the song.

[In Tune Audio]

Originally Posted by wesley tanner

it has nothing to do with TRACKING. I ALWAYS am paraniod about not peaking when tracking, I'll do whatever it takes so it doesn't peak anywhere in the chain... but that's all about gain staging. i'm just talkin' about mixing. tracks all done, you set the threshold and ratio to get x amout of GR.

If you track properly you don't need to worry about it as much when mixing. Teach the singer proper technique, and ride the gain while recording, and/or compress a little on the way in, and you will have a track that will sit better and not need as much doctoring later, plus you'll have a louder and more consistant signal all the way through with no chance of clipping, (if you do it properly)

If you can fix things (like a singer who is all over dynamically) on the way in, you'll have a much better track.

[xstatic]

To me a lot depends on what compressor you are using. If you are using some nice compressors, you can really step on stuff. I would NEVER let an Alesis 3630 go down below about 6 DB of compresison. I wouldn't let an RNC go much below that either. If you have a Distressor, you can tweak the attack and release settings and get yourself into 24+ db of compression without much in the artifact department. I have found that if you chain a couple of different compressors together you can get pretty cool stuff. For example, with a medium attack and release I can let my distressor hit a vocal at a 4:1 ratio and constantly compress the signal so that the extreme peaks in a track might hit 16 db of compresison. This might even the track out a bit as well as giving it a more upfront sound. If I follow that with a 3:1 ratio off of a DBX 165a with a pretty fast attack and a medium release, you can get a really thick vocal sound that still has a lot of high frequency clarity and a really upfront punchy sound.

I think the most important thing is to experiment a bit. Test all of your compressors out and see which ones do which better than others. Use them for their strong points and don't be afraid to combine them. One compressor may work better on fast transients and at lower ratios, where another one might have a better sound when it really crushes something. Or, you may even be able to use two of the same compressors in succesion. Often times two identical comps running at 3:1 and only comping about 6 db each might sound smoother and more natural than just one of them running something like a 6:1 ratio and hitting 12 db of compression. Compression is really fun, but it can take a long time to really figure out just what each compressors strong and weak points might be. It may only do certain sounds well at a setting, but others may not work so well on the same setting. Mess with your comps a bit and find out where it stops performing well, and make a note of what characteristics it exhibits with different program material and different attack and release settings.

As far as shear numbers go, I think it is important to watch them a little. Often things may sound good to us, but later on we hear them and wonder what we were ever thinking. I use the numbers to double check my ears. If something sounds good to me but I have a unit running that is really slamming, then I ask myself if that is really the sound I want, and if there is another way of getting it that might be a little more within normal gain staging and level setting principles. I also try to listen more closely and make sure there is nothing in that sound that I may "regret" later.

[FALKEN]

i wouldn't do that with digital comps.

[xstatic]

And why not?

[wesley tanner]

Originally Posted by ds21

I never pay attention to the dbs either, if you don't want to hear the "depends on the track/singer" answer, then I guess I can't say anything. Sounds like your more into limiting.

It might be intresting to post this in the PMC contest and ask people after they've mixed their versions of the song.

i would looove to check that out...


oh, and as far as teaching people... I don't have the time to teach everyone, but I always give a vocalist a spiel about mic technique and not letting words jump off the page... they just usually don't listen(see my 2nd post)

[In Tune Audio]

Originally Posted by wesley tanner

oh, and as far as teaching people... I don't have the time to teach everyone, but I always give a vocalist a spiel about mic technique and not letting words jump off the page... they just usually don't listen(see my 2nd post)

I'm not trying to be difficult, but, if they want to have a good product that's what they need to learn. In order to play the guitar on a recording, you have to know how to play the guitar. In order to sing on a recording, you have to know how to sing properly. If you want the absoulute best record IT STARTS WITH THE PERFORMANCE!

As far as not having time, it takes more time to fix a poorly recorded track, than to record it properly in the first place. That and teaching people is part of the job of recording them. People don't know how to perform properly on recordings, and it is the job of the engineer to provide the best recording, so show them.

[wesley tanner]

i have to disagree. it can take YEARS to form a well trained voice, and i have to crank out performances in a day or two to keep paying the bills...

i believe it's more the art of coaxing the best possible performance out of them in the time you have. every word you say to a vocalist from when you meet them can often determine what they'll perform in the end. A lot of the time(when you can't say no to clients), getting the best take means working with the untrained voice to make it fit with the song.

i'm leaving for california in about an hour, so no fair sayin anything i can't reply to for three weeks!