Jump to content

Craig started a ruckus


Extreme Mixing

Recommended Posts

  • Members

Craig,

 

You have really started a ruckus over at Whateverworks with your article recommending normalizing every line of vocal performances to make them sit right in a mix. Just wanted to give you a chance to respond in person. I don't do that my self, and I know some very high profile mixers, and none of them do that in the normal course of mixing a song.

 

What's up?

 

Steve

Link to comment
Share on other sites

  • Replies 56
  • Created
  • Last Reply
  • Members

Interesting. I've just never thought about doing it, partially because people always told I'd be "destroying the waveform" if I did...but regardless, the thing is if it sounds good and it's helpful, then cool.

 

I just do a lot of gainriding with the fader, along with some compression and helping out the vocalist a little bit by getting the performance consistent, and that seems to work for me.

Link to comment
Share on other sites

  • Members

 

I'm a big fan of reversing every line of my vocal performances, that's the true master technique right there...

 

 

I actually sing things backwards, and then reverse the waveform.

 

And then I normalize.

 

And then I keep the faders almost all the way down. That {censored}'s loud!

 

Then I add a ton of reverb.

 

Then I compress it.

 

Then I select the track and delete it.

 

It's a Zen thing. You wouldn't understand.

Link to comment
Share on other sites

There's two ways of "normalizing" - peak and RMS.

 

RMS can clip off peaks, or require compression or limiting to handle what would otherwise cause clipping or overs. Peak normalization normally increases the amplitude only to the point where the highest peak hits just short of clipping. Additionally, sometimes you can peak normalize to a specific value below 0 dBFS, such as -6 dBFS or whatever. While I normally don't feel the need to do any normalizing due to taking care with my levels when I track, if I got something from someone that was recorded very low, I might do so just to get the faders into a more useful part of their range. I've also been known to trim a track down due to it being recorded so hot that I had to pull the faders down to get them into the right range.

 

Maybe Craig is discussing corrective measures instead of standard practices? I don't know - I'll let him clarify what he means. :)

Link to comment
Share on other sites

  • Moderators

I've seen him mention this before and I always thought it was a clever idea. But one that could be used blindly to bad effect.

 

By normalizing phrases you retain the internal dynamic of the phrase while still getting the compressor to hit consistently throughout the tune. The softer parts still get a nice rounding off and the loud parts won't be squashed to death. Like I said, clever, but scary if you don't pay attention. You could inadvertently alter the performance dynamic negatively.

 

I would think peak normalization would work best for this as RMS I believe actually incorporates compression too? I think?

 

I just thought Craig has respect for his audience and assumes a certain level in smarts on their part.

Link to comment
Share on other sites

  • Members

I actually sing things backwards, and
then
reverse the waveform.


And then I normalize.


And then I keep the faders almost all the way down. That {censored}'s loud!


Then I add a ton of reverb.


Then I compress it.


Then I select the track and delete it.


It's a Zen thing. You wouldn't understand.

 

:lol::lol::lol:

 

Reminds me of a stupid joke I always say when working on a project. When the artist finishes a take and says, "Can I hear that one back?" I say, "No, we only record things here, we don't play them back."

 

It does seem to get a laugh and get them to relax a bit if they're nervous. :lol:

Link to comment
Share on other sites

  • Members

 

I suspect the red flag here is the word "normalize."


I've certainly adjusted the level of clips when comping to adjust for varying levels and mic distance (I've also been known to adjust the EQ to match prox effect from clip to clip where there was a diff).

 

 

I sometimes adjust the level and EQ for proximity effect as well, so yeah, who knows, maybe normalizing isn't that weird. But I think a lot of the reaction on the other website is from "normalizing each phrase to even everything out" which some people may interpret as doing this regardless of what the vocal sounds like...in addition to reacting to the word "normalize", as you say.

Link to comment
Share on other sites

  • Members

Well I can't log in to Pro Sound Web to reply, and when I submitted a "forgot password" request, I never received a reply. So, I'll post my response here, and hopefully, someone who can log on to PSW and access the thread mentioned can post it for me. I would also encourage these people who have problems with my articles to address me about it directly (what a concept!): It's not like I'm hard to find.

----------------------------------

I hope that those who assume Mr. Eaton has accurately summarized the article would read it. The articles starts as follows:

 

"As far as I'm concerned, the vocal is the most important part of a song: It's the conversation that forms a bond between performer and listener, the teller of the song's story, and the focus to which other instruments give support.

 

"And that's why you must handle vocals with kid gloves. Too much pitch correction removes the humanity from a vocal, and getting overly aggressive with composite recording (the art of piecing together a cohesive part from multiple takes, and the subject of a future Vocal Cords) can destroy the continuity that tells a good story. Even too much reverb or EQ can mean more than bad sonic decisions, as these can affect the vocal's emotional dynamics.

 

"But you also want to apply enough processing to make sure you have the finest, cleanest vocal foundation possible - without degrading what makes a vocal really work."

 

Let me address a few specific comments:

 

"Craig does not seem to encourage LISTENING to the signal."

 

See above for the comments about "kid gloves" and making sure you don't do anything that "[degrades] what makes a vocal really work." You can't make these kinds of judgement calls without listening (at least I can't). Also, I assume that the readers of EQ are not idiots and, by golly, they use techniques WHERE APPROPRIATE. It seems like the peak of condescension to assume someone reading one of my articles would apply every element, regardless of whether it makes musical sense or not. I present tools: How, when, why, and where people choose to use them is up to them.

 

Mr. Eaton sez: He disagrees that the "phrase by phrase" thing "will retain more overall dynamics."

 

He fails to remember this is all in the context of handling the vocal with kid gloves. Is there anyone here who HASN'T heard an overcompressed vocal? Show of hands? Okay, I didn't think so. What I presented, and which was made clear in the article, was an alternative to excessive compression by raising peak levels for greater dynamic consistency BEFORE applying any compression, and that by using this technique, the engineer would not feel the need to add as much compression. This was specifically to avoid the problem where an engineer uses compression as a blanket treatment to fix problematic volume variations, thereby squashing the louder parts excessively.

 

Mr. Eaton sez: "And that's simply not possible, because he's destroying the dynamics of the take from the get go."

 

Only if you choose to. I specifically mentioned this in terms working with the vocals of someone who doesn't "have the mic technique of a k. d. lang," who in case you haven't seen her, uses a mic like a slide trombone to maintain consistent levels.

 

Of course, if a phrase is softer because it's SUPPOSED to be softer, you leave it alone. As I said, I don't assume our readers are idiots. But if it's softer because it's out of the singer's range, he/she backed away from the mic more than normal (I talk earlier in the article about minimizing restrictions on the vocalists if it hurts the take, and instead, fixing in the mix where possible), etc., then the phrase-by-phrase normalization technique does wonders without having to invoke compression, unless you want to do so later on. We only have so much space in a magazine, and I don't want to spend it saying "Now remember, don't use these techniques if they make the music sound bad!" Some things you just have to assume are self-evident (well, at least to most people). I mentioned trying to achieve a "consistent" vocal, not a "uniform" one. I chose one word over the other deliberately.

 

I might also add that I use this technique *all the time* on narration, which is another aspect of recording vocals. It makes a huge difference in terms of producing vocals that are intelligible over a wide variety of systems, without resorting to slamming with compression. If it didn't work, I assume that people would stop hiring me to do narration, and stop telling me how great the narration sounds :)

 

As to errors within EQ, there was a period before I became Executive Editor when the were indeed loads of technical errors. Since becoming EE, we have a policy of publishing corrections regarding any errors that are pointed out to us. We have many ways to contact the editors with any problems: Snail mail, email, and a web site with forums. It seems odd that someone would choose a venue to complain about the magazine where the only way I would find out about it is if someone stumbled on it and told me about it, which is in fact what happened.

 

It's always possible to take words selectively out of context to make a particular point. More sophisticated readers will take individual elements of an article in the context of the entire article. I cordially invite Mr. Easton, and anyone else who requires clarification of what I've written, to come here and address me directly, as it's a far more efficient method of communication. And if I'm wrong about something in print (which doesn't happen very often because I'm not the only set of eyes that reads an article before it goes to press), I have a record of admitting that and correcting it publicly.

Link to comment
Share on other sites

  • Members

Thanks! It's been so long since I've posted on PSW that I can't remember my log-in, etc., so I can't post your reply there. But I'm sure someone else will. Anyway, as I said previously, I'll have to see the whole article. And now that you've reprinted more of it, it makes much more sense.

Link to comment
Share on other sites

  • Members

whateverworks are a bunch of blowhard douchebags [with the exception of michael wagener if he still helps moderate there].... {censored} those dudes.

 

they really arent even worth a thread here about.

 

as for making vox sit right, its a huge balance of a lot of things. i like compression combined with riding faders. rarely do i EQ unless it absolutely needs it [or going for an effect].

 

i dont normalize ANYTHING ever. peak or rms. and -6 is a bit high for peaks and you dont lose resolution by peaking at -18dbfs... do your bit math first before making statements like that. -144dbfs -18db=-126dbfs. what is analog gears [real world] noise floor anyone? and what is your recording space noise floor? there is your "window" to maintain level over the LSB.

Link to comment
Share on other sites

  • CMS Author

There's really no difference between peak normalizing and just turning up the level. RMS normalizing is sort of like compression but not quite. Whenever I've tried RMS normalizing, it always sounds like the kind of compression that we have learned to hate in mastering, but I guess some folks like it.

 

In the world of 16 marketing bits, turning up the gain digitally wasn't a very good idea because it broght up the level of the poorly resolved quiet parts, making the lack of resolution more obvious. In the 24 marketing bits world, things are considerably better. Now when you increase the level digitally mostly what becomes more audible is ambient noise and mic preamp electronic hiss.

 

It's always better to get all the gain you need at the front end, but if you find that it's easier to start out with a track that has its peaks at full scale it's usually not a big deal. But I don't think it should be a "best practice" kind of thing.

Link to comment
Share on other sites

  • Members

noise floor anyone? and what is your recording space noise floor? there is your "window" to maintain level over the LSB.

 

Good points. I just do what sounds good, perhaps my perceptions aren't sound? ;) Thanks to Mike for clarifying what I tried to say. :eek:

 

All I know is peak normalizing is your best friend when trying to bring up the level on all those old cassette demos. ;)

Link to comment
Share on other sites

  • Members

People get antsy when you say "normalization". I suggested it once as a method of salvaging a spoken word recording and got a huge rash of {censored} from people who said I should ride the fader. *Right* Ride the fader for 45 minutes of spoken word that no one paid attention to in the first place. :rolleyes:

Link to comment
Share on other sites

  • Moderators

People get antsy when you say "normalization". I suggested it once as a method of salvaging a spoken word recording and got a huge rash of {censored} from people who said I should ride the fader. *Right* Ride the fader for 45 minutes of spoken word that no one paid attention to in the first place.
:rolleyes:

 

 

I work with telephony files and believe me... normalization can be your friend. Thousands of files. 50,000 plus files. When I read Craig's article my thought was, "Hey, that's pretty dang clever. I sure hope no one toasts their vocal track though."

 

But like I said earlier, Craig assumes a certain level of intelligence. There is an undo to every action you take on an editing program, yes? You are listening to each move you make, correct? So, really, given that he spent a considerable amount of time qualifying the article at the beginning, where's the harm? He stated very clearly. I'll paraphrase... "DON'T F$%^ UP YOUR VOCAL. IT'S THE MOST IMPORTANT THING."

 

It's a clever technique that I imagine could really do some magic on a pop vocal and I intend to try it.

 

Some folks over at the "other" forum, while obviously credible engineers, tend to get a little (or a lot) structured in their thinking. Some might say stale even.

 

Only use analog!

The D112 sucks

You can't use that?!?

Only mix L, C, R

Digidesign is really owned by Bill Gates

A bass only has 4 strings... etc.

 

Of course I do agree with a lot of that :) but they can get a little knee jerk at times. Times like now me thinks...

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.


×
×
  • Create New...