ok ive always wanted to know how to do this

[nch3966]

Fast forward to like 58 seconds. Not the chopping but the actuall huge group singing thing I've tried so many timess and it sounds weak, I typically use a ld condenser with us at a fist length away, and also how do you process this? I mean I can do chants ok but can never get singing to glue in the mix, thanks a lot guys, oh and I appologies for the song haha!

 

[pipaguapique]

mmmm....you get a lot of people and you record them??? in a big room maybe?? then compress them...

 

[WarlockStudios]

Only thing I know about is gang shouts, not had the chance to do gang singing yet but I can imagine it's prettymuch the same. Shitloads of takes - obviously gonna be harder with singing! I usually get a minimum of 5 and pan it to taste and it sounds beefy then process similar to regular vocals without de-essing.

I'm also interested in how others do this.

 

[dan weapon]

Lots and lots of takes, with good harmony parts. I recently did a one man choir for a guys project... had about 36 separate takes in there (record 12 tracks, bounced down 3 times)

 

[OneDaySky]

Use a omni directional microphone in a reasonably sized room with a few ppl, and do a few takes and layer them up!

 

[nch3966]

If I use 2 sd condinsers pointed like an omni directional will that work?

 

[pikachu69]

that will make figure of 8, not omni so you will still have some rejection on either side.

 

[nch3966]

So what should I do?

 

[pikachu69]

Something like this with the crowd (singers) standing a few feet back in front of the mics will work. you can use individual stands for each mic for this.

 

[pikachu69]

pan each hard L R.
Multi takes layered.

 

[ArthurD]

Crowd + Big Room + Mid Side Recording + COMP

 

[amarshism]

Omni ldc in the middle with 5/6 dudes about 3 feet back. Couple 58s/57s/whatever lying around in random places pointed at the walls and stuff. 10-20 takes. Pan the takes across the spread in small increments. Compress, heavy handed de-essing and maybe a touch of verb. Check this track we did.

 

[Morgan C]

Something like at the end of this might be a better example?


Sounds badass and I have no meaningful contribution to this thread.

 

[_Protoss_]

You don't have to record a bunch of people together. You must record every person per separated, and then quantize and correct the pitch of every syllable to perfection with melodyne. Then apply some verb and that's it.
All britney's vocals (lead vox and groups) are edited to infinity with melodyne. Even the dynamics are corrected with the amplitude tool.

 

[fade_2_black]

The more the merrier I've found in the past. Stereo separation plays a huge part in getting that 'big' sound. I generally layer multiple takes using a cardiod LDC aimed reasonably centrally at a group, plus a pair of SDCs in XY configuration coinciding with the capsule of the LDC

 

[Jormyn]

You can also try pitching some of the takes up or down a few cents to get a bit more separation. Say, four takes of your group doing the part, four more takes at +5 or +10, and four more at -5 or -10.

 

[dan weapon]

You don't have to record a bunch of people together. You must record every person per separated, and then quantize and correct the pitch of every syllable to perfection with melodyne. Then apply some verb and that's it.
All britney's vocals (lead vox and groups) are edited to infinity with melodyne. Even the dynamics are corrected with the amplitude tool.

I agree with quantising to perfection, but I wouldn't say autotuning is essential (unless there's actual wrong notes or you use terrible singers) if you are doing heaps of takes the subtle slightly off notes create that element of hugeness - especially if it is the same person recording all the parts.

 

[vespiz]

When mixing, watch out for the mud build-up when layering the tracks.
I also tend to cut at about 1.5kHz, blends nicely around the lead vocals (I read this from Gearslutz about Def Leppard backing vocals, IIRC..).

One of the most important factors in BGV's is, of course, the arrangement.
With the right harmonies/lines, you might find that you need fewer tracks than you thought.
(Don't be afraid to use mute!)

 

[Morgan C]

I agree with quantising to perfection, but not I wouldn'st say autotuning is essential at all (unless there's actual wrong notes or you use terrible singers) if you are doing heaps of takes the subtle slightly off notes create that element of hugeness - especially if it is the same person recording all the parts.

I definitely wouldn't autotune 100%, but autotuning to 50% might be work. It all depends on how good the source is.