vocal compression

Posted on

http://www.reverbnation.com/2ndg
Member Since: Nov 27, 2007

Im recording vox atm, my own actually.
Im wanting to compress going in, on my pre amp but having a few issues with settings.
Im currently using a fast attack, the fastest, and about 6 to 700ms on the release. compressing at about 8:1.
I know that sounds like it'll be squashed to hell, and it very well is, but seems to capture everything well.

just looking for some suggestions on what to do, i like the intensity but not squashing all the life out of it.

[ Back to Top ]


Formally Hydrial69
Member
Since: May 06, 2009


Oct 16, 2010 11:32 am

its hard to say really,
there are no hard and fast rules especially for compression.
it depends on the type of compressor and mic and signal etc,
the best thing to do, it to just take a few hours and play around with it, understand what each nob fully does, which you can do through google easily, and simply turn the nobs with the knowledge you have about how a compressor works, and you will eventually be able to set up your compressor in no time when doing sessions,
the only real advice i can give you on the topic sorry,

best of luck!

MASSIVE Mastering, LLC
Member
Since: Aug 05, 2008


Oct 16, 2010 11:44 am

I can't suggest NOT compressing the input signal enough. Especially with very extreme settings --- The risk of overdriving the front end is far more likely, it's impossible to 'undo' later, etc., etc., etc.

"Flavor" units going in? Another story. 1.1 or 1.2:1 with a dB of gain reduction on a nice opto? Go for it. But 8:1 is --- well, it's a higher ratio than I've ever used on a vocal track in the first place (unless I was looking for extreme limiting with an out-of-control vocalist - but even there, I'd probably be at 4:1). Doing that in the context of the mix is one thing - Doing it at the input sounds a little too much like audio suicide.

IMO/E, YMMV, etc., etc., yada, yada.

Typo Szar
Member
Since: Jul 04, 2002


Oct 16, 2010 02:09 pm

since u said "capture everything well" im assuming ur using this very extreme setting to bring about some thickness to ur vocal, but ofcourse ur kind of sacrificing alot just to get a good signal.

Its really more about the pre if ur talking about capturing everything well, getting ur mic in a good space where the vocal can really enter it and then setting up ur gain

this depends on wat pre ur using as well as mic and if their giving the juice u need. Its not always about getting new equipment but if its not working out just try other mics u may have, u never know wat ull find. Sometimes a mic that isnt predictable or typical may capture just the quality ur looking for and then u wont feel the need to open the floodgates with heavy compression to bring in everything.

trust me, u hear a weak signal and u compress and it sounds nice and thick and loud and u think well hey, lets do it, but weeks or months from now when ur mixing and ur going from high volumes to low volumes ull REALLY start to miss those dynamics.

http://www.reverbnation.com/2ndg
Member
Since: Nov 27, 2007


Oct 16, 2010 08:35 pm

Hey guys thanks for the input,
Yeah crux like you said, its brings thickness to the vocal but thickness then spread thin to make more amount of thinness from that thickness.......hmmm yeah you know what i mean.


i might go with MM's suggie, see how i go with 2:1 but ikeep forgeting my other issues is im also in a walk in closet with foam on the walls and its quite squashy in there. gonna have to take it outside into the room itself i think, get some natural air, in more ways than one.




MASSIVE Mastering, LLC
Member
Since: Aug 05, 2008


Oct 17, 2010 01:30 am

2:1 is still an awful lot of compression (depending on your threshold). Far more than I'd want on an input chain.

And for sure, lose the closet.

http://www.reverbnation.com/2ndg
Member
Since: Nov 27, 2007


Oct 17, 2010 03:59 am

well, you'll be casting your hand over it in the end MM so will try to get it as close as i can to how you would like it as far as mastering is concerned.

I have little compression on anything except the bass.
None on guitars and im still fiddling with the drums, im definatley wanting more colour and body to the drums, and i think the compression is killing the **** out of the whole thing atm.

gotta get those right or youre killing everything else to suit.

Typo Szar
Member
Since: Jul 04, 2002


Oct 17, 2010 11:25 am

i think it has to be pointed out that compression isnt a huge no no, its not like if u ursing less compression ur somehow "mixing" better, i think alot of ppl make that mistake (COZ I DID hehe)

the thing ppl r tryin to warn u away from is compressing on the way in, becoz u cant undo that and ur stuck with that lifeless flatline waveform if ur giong hardcore not to mention all the kinds of artifacts outboard compressors will be adding to ur signal so on so forth

wat i think mm is saying about the 2:1 also being alot is jsut the fact that i mentioned before where ur trying to use teh compressor to get a result that would be better gotten through other parts of ur chain such as ur pre or ur room or even just making sure that ur monitoring in a way u can properly hear ur input so ur not compensating for something u dont actually need

long story short everyone is just trying to warn u away from committing to something u dont particularly have to

http://www.reverbnation.com/2ndg
Member
Since: Nov 27, 2007


Oct 17, 2010 06:36 pm

Thanks crux,
i really wanna give the vox a final shot this week.

Ive got the rest of the mix as good as i can get it atm, if im doing anything now its just changing things around rather than makin anything better.

what i might do is post a song here and get some opinions on the drums, im not 100% happy with em but i cant see how else to get em much better, so maybe you lads have a listen and go, oh yeah man, try this to bring out some more life to em etc.

Im using sd2 but i also have BFD2 which maybe i can sample some drums and layer with sd2.


http://www.unitedmusicians.info
Contributor
Since: Nov 11, 2007


Oct 19, 2010 03:03 pm

Oops this topic is pretty much a mirror of what I just commented on. I'll get on my screaming vocal recording soapbox one more time, this time tuned to your more experienced ears vs the ears of the fresh newb.

Screaming is not melodic. You are not carrying the melody with your scream. Since you aren't going for a specific melody, the normal rules for recording vocals, in my opinion, go out the window. It's not about melody, it's about rhythm.

I would record metal vocals with the compression on 5:1, personally. Compressing something this hard is going to destroy the nuances of the performance but do you WANT the nuances of the performance? In my gutteral death rumblings, I certainly don't...I want a big, FAT, rhythmic device. I don't even want the ME to have the option to leave screaming vocals thin. Compression achieves that. Compressing on the way in is definitely counter intuitive to EVERYTHING else I do in the studio, but for this particular sound, I'm pretty sold on the idea.

All that said...MM obviously disagrees, makes AWESOME metal recordings, and is going to master your stuff anyway...maybe you should just listen to him...if it were my metal recording, I'd do a vocal group stem and somehow trick him into squashing that with his nice compressor/s before sticking it back in the mix. Maybe see if you can get one with compression and one without. Also, you should fly out to oversee the process. It's a $30 bus ticket away for me!

Typo Szar
Member
Since: Jul 04, 2002


Oct 19, 2010 10:16 pm

Just to be "that guy" no one is really saying dont compress too hard, its mostly just dont compress on the way in, if the gain for the track is good and its picking up wat it needs, then u can compress to ur hearts content

i do agree with u that screaming is a totally different beast (pun?!?!) and it does require a different approach, but again compressing on the way in will give u more or less the same results as compressing after without the option to fiddle around

http://www.reverbnation.com/2ndg
Member
Since: Nov 27, 2007


Oct 20, 2010 08:13 am

thinking i might compress @ 2:1 going in to get the Joemeek colour, then hit it again a little with one inside the box.

end of the day, im still a novice and the songs will have engineering flaws all through it anyway. just need to try keep em to a minimal if poss.

Member
Since: Jun 05, 2010


Oct 20, 2010 10:47 am

8:1 is almost limiting! I don't understand why you need a so drastic setting at the recording stage. Do you use a 8 bit recorder?

MASSIVE Mastering, LLC
Member
Since: Aug 05, 2008


Oct 20, 2010 11:13 am

Quote:
thinking i might compress @ 2:1 going in to get the Joemeek colour, then hit it again a little with one inside the box.


Just to 'nag' a bit more -- 2:1 can be a serious amount of compression depending on the threshold. If you're looking for color, backing the ratio down to something more like 1.1 or 1.2:1 and hitting a dB or two of gain reduction on the louder parts is going to be much more reasonable to work with in the mix later.

2:1 (let's say with "several" dB of GR) on the way in - and then trying to craft compressor settings that don't argue and moire with the previous compression settings...? AND make that fit the context of the mix...?

I'd imagine you see what I'm getting at here...

http://www.unitedmusicians.info
Contributor
Since: Nov 11, 2007


Oct 20, 2010 03:35 pm

@ MM: What sort of threshold would you put on a compressor set to 1.1:1 for screaming vocals? What sort of arguing would you expect from various compressors? Curious because I don't know, not critical.

@ Laurend: 8:1 is almost limiting, correct. I typically used something like 4-5:1 for my screaming which is still pretty extreme. I do not use an 8 bit recorder. Pre-compression, screaming vocals sound dynamic, which in any other vocal recording wouldn't be a problem. Screaming is not like singing in that dynamic change is often, for most vocalists I've worked with, awkward sounding as it occurs in the real world (in live scenarios, screamers tend to eat and/or cup the mic in their hands which acts as a sort of "piss off the sound guy and cause terrible feedback" compression technique). Dynamics are often wildly all over the place from one word to the next and the most quiet mouth/throat sound can be the most impactful percussive element to the vocal sound. This is definitely a performance created vs a performance captured, but that's the style I prefer for the genre. Like I said earlier, most screamer type vocalists I know are live sound engineers worst nightmare; they're always cupping the mic in their hands and getting their lips all over the grill. If these engineers or the vocalists knew that the sound the vocalist is looking for can be achieved by hard, otherwise obnoxious, compression settings...the live sound reinforcement game would be much less frustrating for all involved.

You can save compression until the final mix down or compress it on the way in, I would likely record everything without compression, create duplicates of performance and use in the box processing as inserts to give the performance a little variation in dynamics. One track with 2:1, another track with 5:1 is typical, and the uber-compressed track is definitely going to be mixed on top for me. I would not leave an uncompressed vocal track in the mix. It's a polished turd by design, in my opinion.

Try checking out music one of the thousands of underground eletronica forums (dogsonacid etc.), if it's not limited extensively at the master bus, the audience hates it. The audience wants pumping, loud dance music and at any cost. They like the sound of obnoxious pumping compression This is not normal and no doubt most people would hate these recordings, but this is the standard in underground electronica (which is a pretty popular niche!) I'm comparing apples and oranges here (vocal chain compression:master bus compression) but the concept of audience perspective is what I'm trying to get across. Just because you think it's wrong doesn't mean the audience does. As a producer, you can produce for yourself or produce for your audience. That's a call the individual has to make.

Another example of a non-screaming, non-electronica band that appeals to the same demographic who's into the wacky compression sound:





I think it sounds digital and clipped to hell. I can tell the audience that, then they will look at me with a blank stare and say "so what?"

http://www.reverbnation.com/2ndg
Member
Since: Nov 27, 2007


Oct 20, 2010 06:16 pm

no wuz then.
a small comp going in then do whatever i want in the program, then turn it off when i get it ready to export.

I tend to have alot more control than the average screamer so hopefully wont be that bad.

MASSIVE Mastering, LLC
Member
Since: Aug 05, 2008


Oct 20, 2010 11:35 pm

Quote:
@ MM: What sort of threshold would you put on a compressor set to 1.1:1 for screaming vocals? What sort of arguing would you expect from various compressors? Curious because I don't know, not critical.

Whatever threshold would give me a dB or three on the really loud parts - I should've mentioned, a ridiculously slow A&R -- Maybe 100ms attack and 3-5 second release.

The "arguing and moire" thing is when you have A&R times that are fighting each other - You can get anything from weird pumping to just plain freaky modulations like a cheap vibrato pedal.

Member
Since: Sep 14, 2010


Oct 21, 2010 07:22 pm

when doctor dre recorded 50 cent he used a sony c800 mic into a neve 1073. AFTER recording he inserted one compressor set 7.1:1 with a thresh of -1. No EQ. SO he pretty much just found the perfect mic, the perfect mic pre, and the perfect mic position, slapped a limiter on it and called it a day.


Maybe you should spend more time developing your technique and breath control. Then try different mic positions and proximities. You should not "fix it in the mix". learn to lean in and scoot back while your performing.

I always say if it takes me more than 10 minutes to set my compressor and EQ, then i need to rerecord.

also try not coming in so hot. I used to use all anolog equipment so coming in hot as possible w/o clipping was the virtue. Now i got a DIGI 003 with a blue bottle mic and i record at 24 bit. SO coming in at -1 alot of times is too hot. I've recorded hip hop vocals peaking at -4 and averaging at -10. Used a comp around 2.5:1/-15.

also try using a low shelf before compression that will eliminate plosives that activate the compressor when it is not needed.


MASSIVE Mastering, LLC
Member
Since: Aug 05, 2008


Oct 21, 2010 09:06 pm

Quote:
SO coming in at -1 alot of times is too hot. I've recorded hip hop vocals peaking at -4 and averaging at -10. Used a comp around 2.5:1/-15.


(Just using this as an example) -----

Even -10RMS with peaks of -4dBFS is around 250% over spec on the Digi's converters (which I believe are calibrated to somewhere between -18 and -15dBFS).

Personally? If I'm hitting -12dBFS peaks at any time on any source, I'm backing it down. And that's the other problem with compressing the input chain -- Say you're actually at 0dBVU with converters calibrated to -15dBFS. You're probably trying to "stay a little hot" on the way in so you push it to -9dBFS. Now, you're running at over 200% of the spec'd voltage of the unit. That's bad enough.

Throw a compressor on the signal and get 3dB of gain reduction (which of course, the typical person is going to add back). Now the *preamp* is running at well over 300% of line level voltage.

This is what leads people to "my mixes sound small." and "I can't get my mixes 'loud' like commercial mixes."

Headroom is king. All day, every day at every possible stage.

24-bit digital has more downward headroom than any piece of gear you own and more dynamic range than the human ear can take advantage of. CELEBRATE that - Don't try to bypass it...

http://www.reverbnation.com/2ndg
Member
Since: Nov 27, 2007


Oct 21, 2010 09:48 pm

hey man this is screaming metal im doing here, i dont think i can develop my tehcnique anymore than i have, in fact, there really isnt alot of explosive voice fluctuations happening to be honest.
Only dynamics from soft to loud, no wham over the top stuff at all.

its not a problem doing this stuff in sections either so im getting the best outta each take.

Did some practice last night with 1:1 on the pre with a long attack and short release, and then a 2:1 inside the box.
i did find it lost a fair bit of presence but definatley do-able.

Im gonna go ahead with what you said MM and then take off the inbox compression, and as we discussed, will send a copy of a song and you can gimme the verdict on what needs to be tidied up etc. so i can setup the rest of the songs them same way

http://www.unitedmusicians.info
Contributor
Since: Nov 11, 2007


Oct 22, 2010 02:51 pm

^^^ Good plan. Maybe you'd share the before and after with me?

Member
Since: Sep 14, 2010


Oct 22, 2010 04:37 pm

@MM. I learned alot with your last post. Can you expand a little?

where can i go to learn the difference between dBVU and dbFS. Thats always thrown me off when reading about things.

Also where can i learn the "spec voltage" or what my 003's converter is "calibrated to".

There is alot of info on the internet: some of it true, some of it well intended, and some of it false. Like i said, i used do this with anolog equipment so hot as possible was the rule. Where on the meter should I shoot for to come in with my current set up?

Do you know where I can read about this sort of thing. Then maybe you don't have to fill a paragraph on here spoonfeeding me.


Member
Since: Sep 14, 2010


Oct 22, 2010 07:03 pm

@MM so basically your saying that my mixes don't have that commercial loudness because I come in too hot when i record?

I beleive ya, just trying to clarify

http://www.reverbnation.com/2ndg
Member
Since: Nov 27, 2007


Oct 22, 2010 09:30 pm

MM will def set you straight with all this, good call too CKGuy,

im interested myself and im sure he's been over this before with me at some point. (blushes)

Memory=Sieve

Quince,
Sure, can do mate, im hoping to crush (Pun?) these vocals today, so then a touch of mixing and then thats it.

My main issue right now is the drums. I want more body and colour and i cant get it.
Im gonna save this song under a diff name and take all the compresion off the kit and see how it sounds.

Seems ive outgrown my mix room. got bass traps and stuff but its not ideal.

MASSIVE Mastering, LLC
Member
Since: Aug 05, 2008


Oct 23, 2010 01:55 am

[quote]@MM. I learned alot with your last post. Can you expand a little?

where can i go to learn the difference between dBVU and dbFS. Thats always thrown me off when reading about things.[/quote]

Don't forget dBu, dBV, dBm, dB(FS)RMS, etc. Long story short though, dBFS is Deci-Bel (one tenth of a Bel) Full-Scale. Digital reading, -0.0 is the maximum signal (period). dBVU (specifically, 0dBVU) is "0" on a VU meter -- Line level. 1.23vRMS. Line level will equate to a certain analog level on a digital meter. Old crappy digital was almost always at -12dBFS -- Which is pretty freaking hot - and clipping is a constant worry. -18 or -20 is a much more usable level.

Quote:
Also where can i learn the "spec voltage" or what my 003's converter is "calibrated to".


It should be in the spec sheet. But you can always run a 0dBVU signal into it and find out what you're getting. I'd bet somewhere between -18 and -15dBFS --- Which is pretty hot, but pretty typical (I keep my tracking converters calibrated to -20 and I pretend they're at -24dBFS).

Quote:
There is alot of info on the internet: some of it true, some of it well intended, and some of it false. Like i said, i used do this with anolog equipment so hot as possible was the rule. Where on the meter should I shoot for to come in with my current set up?


Somewhere at or under line level (0dBVU). That's where all gear is spec'd at. That's where it's all designed to work.

Quote:
Do you know where I can read about this sort of thing. Then maybe you don't have to fill a paragraph on here spoonfeeding me.


I can't think of a specific resource, but if I can come across something, I'll post 'er here.

Ne'er ate 'er
Member
Since: Apr 05, 2006


Oct 23, 2010 04:10 am

1. Enable ears.

2. Record all tracks without going past 0 dB.

3. Add all effects and dynamic compression to individual tracks after they are recorded.

4. Mix down all tracks as per standard in Rules 1 and 2.

5. Refer to Rule 1 as often as necessary.

MASSIVE Mastering, LLC
Member
Since: Aug 05, 2008


Oct 23, 2010 11:21 am

Quote:
2. Record all tracks without going past 0 dB.


That's the issue - "0 dB" what?

-0.0dBFS? For sure - But you shouldn't even be in the same county as full-scale. 0dBVU? Sure - But a lot of people (as referenced by this very thread) either don't have VU meters or don't quite get where that level sits on their digital equipment.

I'd say "record all tracks without going past -15dBFS" or something along those lines. Let the gear work the way it's supposed to work.

Okay, maybe -10dBFS... But I'm a whore for headroom. Lower is almost universally better.

Ne'er ate 'er
Member
Since: Apr 05, 2006


Oct 23, 2010 05:52 pm

Well, OK. Just avoiding clipping is a good place to start.

http://www.reverbnation.com/2ndg
Member
Since: Nov 27, 2007


Oct 23, 2010 10:25 pm

so ok MM,
I have a VU on my pre. The best way is to be recording through my pre @ roughly not over 0 on that VU metre,

but turning down the output of the pre to get a rough -15 in the DAW? am i following or have i gone walkabout?




MASSIVE Mastering, LLC
Member
Since: Aug 05, 2008


Oct 24, 2010 01:00 am

0dBVU on your preamp is going to *be* around -15dBFS in the DAW.

Some things (percussives and such) have attacks that are too fast for the VU meters to react to -- Those, you just rough in peaking around line or maybe just a couple extra dB on the extra hot hits (this is why it's relatively important to know where line level is on your converters).

http://www.reverbnation.com/2ndg
Member
Since: Nov 27, 2007


Oct 24, 2010 10:22 pm

good info there mate thanks.

Related Forum Topics:



If you would like to participate in the forum discussions, feel free to register for your free membership.