Main Menu

Snowman question

Started by Mickdoo22, November 25, 2008, 09:32:20 PM

Previous topic - Next topic

0 Members and 2 Guests are viewing this topic.


Jem...can you tell me how you achieved the lush airy vocal sound in "Snowman" on the "Will you tell me when it's gone" section?  I know it's stacked vocals, but it is very breathy sounding......killer!  Is that "fun with Pro Tools" or vocal technique?

By the way..I have been feasting on Milliontown while I wait on my copy of EIMA.  I have to say, I almost forgot how truly brilliant that album is.  As much as I love every song, "No Me No You" is fecking great.


I copied this that Jem said over a couple of different threads from the old forum, enjoy! :)

"EQ and compression are the 2 effects I use most of all. They're what makes a mix for me.

As a general rule for me, I have a medium attack (50ms-ish) and fast release (2-10ms) for vocals, you'll get a load of "in your face" that way. And don't be afraid to compress really hard especially if the vocal is very agressive. If you can get your vocals or mixes to "pump" without sounding too obvious, I find you get a real sense of loudness without necessarily having to resort to 2 mix limiting with a Waves L1 or an L3 to get the "squarewave" effect.

Also with vocal EQ, roll off anything under about 350Hz. All there is down there with vocals is basically just ambient rumble, tapping feet, mic stand clonks and other stuff that will get in the way of your mix compression. Your vocals will sound slightly thinner in isolation, but in the mix they'll be absolutely fine. Do your EQ'ing at the same time as your compressing, they really do go hand in hand. EQ first, then compress. Add lots of air to the vocals and then de-ess before compressing. Also, depending on the voice, I find that a wee peak of a few dB somewhere between 4 and 6khz with a tight Q will help the voice (especially male) cut through better without you having to boost the overall level.

And use delay rather than reverb during busy musical sections. Delay takes up much less space in a mix but your brain is still fooled into thinking there's lots of ambience there. If you have something lke Line 6's Echofarm (it's all I ever use for delay), set it to to clock to a dotted crotchet of your source tempo and keep the feedback percentage low. One or two echoes is fine for most applications I find. But don't be afraid to automate mix, feedback and clock rate depending on what you want to emphasise vocally.

Lastly, a great piece of advice I received once was to remember that things like reverb and echo effects are just that - effects. Try not to use them all the time. You'd be surprised how often you don't actually need them. For example on "No Me No You" and "Snowman" there's no reverb at all on the vocals.

Lastly, and this is probably the best advice I can offer, if you're working with a DAW, look away from the screen a lot when you're mixing and use your ears. Sounds obvious, but it's amazing how distracting a screen full of bling can be! ... ight=booth
In terms of Frost, I record them in a corner of The Cube behind an isolation screen. It folds in half to form a 2 walled structure and I've got acoustic tiles behind that so it forms a 3 and a half walled vocal booth. I stick a duvet over the top to form a roof and I have to say that it works a treat.

I should care more about mics I know, but I'm a bit of a luddite about these things. Well, in truth, I believe that when you've got music as dense and as loud as the Frost stuff is, the actual vocal sound becomes a bit of a moot point. If I was recording an album of piano and vocal ballads however, I'd be a bit more thoughtful about mic choice...

But anyway, to prove the point, I started the album with a Blue Baby Bottle. That broke, so I did the rest on a Rhode NT1. When I listened back, I actually preferred the sound of the Rhode.

Since then, I've got a TLM 103 which is a great all round mid priced mic.

Once my vocals were done, I compressed the living sh*te out of them and rolled everything off below about 600Hz during the loud bits. Again, I wouldn't do that with a spoken voice piece, but the music's taking up so much of the frequency spectrum that I try and keep vocals in as narrow a bandwidth as possible. Your brain thinks it's hearing a full frequency vocal, but in reality it's actually eq'd pretty thin.

It's an old old old old old theory I know, but so many people don't do it and that's to EQ according to the sound you're working with. For example, somebody might have a piano sound and I bet that they never think to eq it in a mix. I always roll a ton of bottom end off piano as it'll get in the way of the kick and the bass parts. You don't need all that energy down there competing with everything else so just get rid of it.

Same for acoustic guitar, strings, vocals...all kinds of stuff. Especially vocals actually, otherwise you get rumble, the sound of feet kicking the mic stand, all kinds of stuff. It's amazing how much space you can free up with some selective eq-ing. For rhythm acoustic guitar I filter it all the way up so the only thing you can hear pretty much is the strumming sound during loud bits. It's the rhythm that's the important bit not the chords. The chords are being taken care of with the piano or the electirc guitar. Your brain thinks it's hearing all this stuff, but the reality can be quite different.

Anyway, back to the vox. Once it's recorded, I'll comp it and stick it through a variety of plug ins depending on the circumstances. Automation plays a vital part in it all though. I'm forever riding levels, EQ, compression ratios, attack and release times...that's why it all takes so bleedin' long!  

A typical effects chain might be vox - Focusrite D3, McDSP Filterbank, McDSP MC2000, Waves De-Esser, Digidesign Pitch (for a tiny bit of wideness)

Or : Vox - McDSP Channel G, Waves De-Esser, Digidesign Sci-Fi (add a little ring mod and then automate it to follow the pitch of the vocal, you'll get a fabulous raspy undertone), Sound Toys Pure Pitch (for a bit of formant drop) and maybe the Waves PS22 (automated to f*ck with the phase of the vocal - I did this at the start of the "Only Survivors/Ghost of Yesterday" section of Milliontown, it does my head in on headphones!  )

Other tricks include running a vocoder behind the lead vocal doubling the melody then mixed in out of phase behind the vocal. Or having a subtle reverse reverb effect behind the vox. Or using Clone Ensemble to create a football stadium of singing Godfreys (see the end chorus of The Other Me)

Or putting the vocal through the Digidesign Pitch Shift audiosuite plug-in. Whack it up 24 semitones, then back down again to normal pitch, the artifacts you get are fabulously gritty. I did that with the vocal and the little keyboard riff on "Wedding Day" that used to be in the Toys section.

There's load of fun one can have with a vocal, a ProTools rig and a warped mind!"
Suhr Carved Top - Carvin DC727 - Roland Fantom X8 FOR SALE  - Axe-FX Ultra


Quote from: "Mickdoo22"As much as I love every song, "No Me No You" is fecking great.
Oh heeeelllllllzzzz yeah!
I've seen paupers as kings,
puppets on strings
dance for the children who stare
you must have seen them everywhere


that was a great read Steve, thanks for digging that up.  :)


Thanks much Steve.  Lots of great info there!  I am not sure that addresses my original question, however there is much I can use there.  Outstanding.


The part about doubling the vocal and putting it out of phase is appropriate, and the ring modulator, also the part about pitch shifting vocals up and then back down I imagine :)


"maybe the Waves PS22 (automated to f*ck with the phase of the vocal - I did this at the start of the "Only Survivors/Ghost of Yesterday" section of Milliontown, it does my head in on headphones!)"

That's the bit you want :)
Suhr Carved Top - Carvin DC727 - Roland Fantom X8 FOR SALE  - Axe-FX Ultra


I've always thought the snowman vocals were done with a vocoder of some sort.


I read it time ago, but re-reading it has been great. Thanks for rescueing it, and thanks again to Jem for sharing!


Quote from: "XeRocks81"I've always thought the snowman vocals were done with a vocoder of some sort.

Yeah I think there's a lot of vocoder doubling the melody an octave down at least.
Suhr Carved Top - Carvin DC727 - Roland Fantom X8 FOR SALE  - Axe-FX Ultra


Lovely information. It gave me a tiny hope I might be able to add some vocals to a few of my tracks.... and now I'm at it :)