Any Caveats about In-The-Box mixing?

NealF · Post by **NealF** » Wed May 16, 2007 4:07 pm

I used to own a Mackie D8B. A friend who was a much better engineer than I advised me to keep the channel faders low and the master fader high (about +5) for a better sound.

I'm now mixing in the box and was wondering if there were any things to watch out for (other than the common sense gain stages) in DP5.

Any thing that you might not normally think of.

Thanks.

David Polich · Post by **David Polich** » Wed May 16, 2007 4:18 pm

This topic has been discussed in a lot of previous threads. Don't ask which is "better" - in-the-box or outboard summing - unless you want to start anothe rlong and perhaps pointless discussion about that.

As far as mixing in the box goes - it helps to have an understanding of headroom in digital. Bob katz's book "Mastering Audio" has a lot of info about this. Basically, at 24-bit, you have 48db of headroom above what you have at 16-bit. So levels don't have to be that high.

You actually can't really "run out of headroom" in digital, other than going beyond 0dbfs. That's digital full-scale, and beyond it is ugly horrid distortion and ruined audio. That said, use your ears and keep in mind that frequencies DO build up and can lead to clipping and overs with faders up high and lots of EQ boosts and things like that.

The normal mixing do's and don'ts regarding compression, EQ, panning, rolling lows and highs off of aux effects, etc., still apply in digital, just like they do in the analog world. Maybe the one golden rule to keep - don't put anything on the master bus when doing your mixdown. And don't maximize your mix or anything else to make it "super-loud and punchy". Save that for the mastering people to take care of.

I dunno - the more work in audio I do, the more I just go back to "just make it sound good".

NealF · Post by **NealF** » Wed May 16, 2007 4:38 pm

I know what you're saying.

I always use my ears mainly. For my first in-the-box mix I made stems using auxes. One for rhythm, one for strings and one for group vocals.
Used them as groups. Seemed to be a nice method of mixing.

I don't care about opinions about which is better- inthebox or with an external mixer. I have no choice now anyway. Just wasn't sure if there were any things to avoid that aren't in the manuals.

I'll check out the other threads.

Thanks.

gearboy · Post by **gearboy** » Wed May 16, 2007 5:05 pm

Here's what I have learned and applied in the past year and a half, some here at Unicornation, some at TapeOp, and some through reading Bob Katz and others:

Keep the audio between -18dB and -12dB full scale when working in a 24-bit environment. In Digital Performer, this translates to between -12 and -6 on the meters. Try to keep everything here, including peaks. You will allow enough headroom for your plug-ins, and definitely not clip them. I find that plugs like Waves stuff clips easier and faster than others. When your audio is above -6dB on DP's meters (-12dB full scale) you will run into problems with the occasional peak upsetting this balance.

You should definitely aim to record all of your material in this realm. When you end up spreading audio across 24+ tracks this way, your master Fader will be peaking around 0 to +3 in DP when the fader is at Unity.

Remember, Mastering engineers like headroom, too. Some prefer -6dB (Zero in DP), some prefer -3dB. Mix for the Mastering Engineer, or be safe and split the difference.

Aim for these numbers in 24-bit and you'll be happy.

Jeff

buzzsmith · Post by **buzzsmith** » Thu May 17, 2007 5:44 am

Jeff, just to clarify...

Neal's post was about ITB mixing, but I'm just checking that when you say...

You should definitely aim to record all of your material in this realm

Are you saying that when you lay down your tracks- drums, brass, vocals, whatever- that you attempt to keep those signals in the -12 to -6 range, as well?

I come from the old analog world of recording where you generally tried to record as hot as possible (signal to noise, etc.) and am just making sure that I understand a, perhaps, better procedure!

Thanks!

=buzz=

TOD · Post by **TOD** » Thu May 17, 2007 6:15 am

What Dave Polich said, "Use your Ears"

I also record as hot as possible without clipping in 24 bit, cause, Gosh Darn It, I want to use every bit!

Try to make your mix the best it can be without anything on the master buss, then fine tune with EQ-comp-EQ-limit, (or whatever your flavor of the week is)

my .02

TOD

jlaudon · Post by **jlaudon** » Thu May 17, 2007 6:54 am

Lately I've mixed with lower levels ITB, which seems absolutely fine with well-recorded tracks, and leaves headroom for certain plugins to colour the sound often in good ways (like UAD La2a for example). I'm even getting good 'reviews' for my mixes - the big difference for me was changing monitors (used to use Mackie 824s, and now I use Adams A7s).

Just a handy tip you might know about - press 'w' and lower the master fader, and it will bring all the seperate track levels down the same amount (obviously only before having volume automation enabled on various tracks).

cheers

daniel.sneed · Post by **daniel.sneed** » Thu May 17, 2007 7:03 am

another dead horse thread but :

Depending on wich plugins you use, your way to work considering max levels should be completely different.

For exemple DP and DP plugins are floating 32 bits (a wonder !, thanks Motu !), oppositly Waves plugins are fixed 64 bits (shame on them !).

I curently use only floating 32 bits plugins.
So my only concern is input level (AD) and output level (DA). Nothing else between both matters in terms of level.
That is to say I can overlaod any channel or bus with no distorsion. A very easy way to go !

If you are in doubt about that :
- record any signal at proper level in DP
- set a master fader at -40dB
- boost input channel 40dB with DP trim plug
Channel and master fasder have all gone red, but sound is clean !

If you use any third-party plugin (I guess you do), you must know for shure about this plugin "fixed or floating" status.

gearboy · Post by **gearboy** » Thu May 17, 2007 7:42 am

buzzsmith wrote:Jeff, just to clarify...

Neal's post was about ITB mixing, but I'm just checking that when you say...

You should definitely aim to record all of your material in this realm
Are you saying that when you lay down your tracks- drums, brass, vocals, whatever- that you attempt to keep those signals in the -12 to -6 range, as well?

Yes. 24-bit allows you headroom. You should not record as hot as possible. The "filling every bit" theory worked well for 16-bit because of less headroom, but for 24-bit you are looking at clipping fixed 64-bit plug-ins. I know because I went through this same scenario from March until August of last year. I thought that you needed to "fill every bit", and I did. And guess what? I had to insert a Trim plug-in on every track and subtract 12 to 18 dB because I was clipping TRacks Limiter and RenComp, and had all of my channel faders at like -20 to -30 something each. Then I stumbled upon several of these discussions on the net and realized that I was recording way too hot.

There is also distortion that adds up over tracks when you process through plug-ins and your signal is too hot. Not clicking distortion that lets you know that you are peaking, but a subtle thing. Increasing your headroom lets your mixes breathe and keep things sounding open.

buzzsmith wrote: I come from the old analog world of recording where you generally tried to record as hot as possible (signal to noise, etc.) and am just making sure that I understand a, perhaps, better procedure!

Well, with digital, there is no noise. There is preamp gain noise, sure, as you increase gain into the computer, and you still have to watch out for adding up 24+ tracks that have hiss from gain pots. But digital zero is silence. You're not fighting anything.

You can also just aim for zero on the faders and keep Trim plug-ins on every track before your other insert plug-ins or sends. But this is unnecessary CPU usage in my mind.

Jeff

Phil O · Post by **Phil O** » Thu May 17, 2007 8:03 am

There was a time when I used to keep peak levels at about -12dbfs because that's what sounded best. I later discovered that it was at the analog end of my system that was making the difference. My BOARD sounded best when it was at a level that resulted in a -12dbfs digital signal. Since then I've completely changed my set-up. My board is used primarily for zero latency monitoring and most of my signals go from dedicated pres or direct boxes directly to my converters. I now find that what sounds the best is to go for maximum signal without clipping. I usually give myself 3 to 9 db of headroom, depending on the instrument and the musician. I've gotten pretty good at quickly assessing where a musician will go and coming up with a level that will avoid clipping. When you do your level checks, ask the musician/s to play from what he/she expects to be the loudest section of the piece. You can expect most musicians to go 3 or 4 db over that in the excitement of the actual performance. (At least that's been my experience.) Less experienced musicians may need more leeway.

But with all that said I think it would be wise to heed the advice of those who have said, "Use your ears."

Phil

Phil O · Post by **Phil O** » Thu May 17, 2007 8:16 am

daniel.sneed wrote:Depending on wich plugins you use, your way to work considering max levels should be completely different.

I guess I haven't given this much thought, because I live in the isolated world of MY pluggins and the way I WORK. But I think you make a very valid point!

Phil

ryst · Post by **ryst** » Thu May 17, 2007 8:25 am

Before this thread turns into another 12 page epic, I thought it would be a good idea to provide you with links to most, if not all, of the other threads from various websites that discuss this very topic. I keep these bookmarked because I go back and read them to freshen up on anything I might have forgotten. Get a lot of caffine in you, because this might take awhile. But eventually, it should answer all your questions.

http://recforums.prosoundweb.com/index. ... 918/0/0/0/

http://recforums.prosoundweb.com/index. ... #msg_20115

http://www.gearslutz.com/board/masterin ... db-me.html

http://acapella.harmony-central.com/for ... id=1420205

http://www.gearslutz.com/board/masterin ... indle.html

http://recforums.prosoundweb.com/index.php/t/15038/0/

http://www.bigbluelounge.com/forums/vie ... 02ff6630b7

gearboy · Post by **gearboy** » Thu May 17, 2007 8:36 am

ryst wrote:Before this thread turns into another 12 page epic, I thought it would be a good idea to provide you with links to most, if not all, of the other threads from various websites that discuss this very topic. I keep these bookmarked because I go back and read them to freshen up on anything I might have forgotten. Get a lot of caffine in you, because this might take awhile. But eventually, it should answer all your questions.

http://www.gearslutz.com/board/masterin ... db-me.html

From that Gearsluts posting:

If the tracks are tracked where the gear wants to be (somewhere around 0dBVU / -18dBFS) it's almost difficult to get a mix to peak that high without trying.

I don't mix an awful lot anymore - But most of the time, they just "naturally" have peaks that rarely exceed -8 or -9dBFS.

Don't worry about "hot" - You've got more headroom than at any time since recording was just an idea. Use some of it.

Sidenote: 90% of the time, mixes that come in with "hot" levels, overly compressed for the sake of sheer volume at ANY stage (at the track level, at the group level, on the main buss, etc.) are the ones that don't have the potential to be "in-your-face" loud in the end. On the contrary - Mixes that come in peaking at -12dBFS, with barely any compression that didn't actually *serve the mix* -- Mixes that were well cared for, with generous amounts of headroom at every stage in the game - THOSE are the ones that have the most potential for sheer volume in the end. They're the ones that can handle it, they're the ones that have room for it. They're the ones that can "take the abuse" when the client says "make it freakin' loud."

But again, that's only about 90% of the time...

SO - Rule of thumb - If you're even the slightest bit concerned about your levels, they're probably much too hot.
__________________
John Scrip - Massive Mastering - www.massivemastering.com

John also posts over at TOMB a lot, and he knows his stuff. I say, try a bunch of different methods, and use your ears. I worked hard at learning what works for me and now I have a tried and true method to build upon.

Jeff

OldTimey · Post by **OldTimey** » Thu May 17, 2007 9:55 am

TOD wrote:
I also record as hot as possible without clipping in 24 bit, cause, Gosh Darn It, I want to use every bit!

TOD

what gearboy said about 24-bit vs. 16-bit...no need to fill every bit..just fill more than 16 of them going in, and you are in pretty damn good shape.

also, by trying to fill every last bit, you run the chance of possibly overloading the analog stages of your gear. if you run a preamp really hot, then you it is going to add high level distortion to the signal. which may sound "nice and warm" if it is a tube amp, or "punchy and raw" if it's a solid state pre...or if it's a cheap pre, just add nasty noise and signal distortion.

over the course of 24 tracks or so...all the "color" you have added to your tracks by consistently chasing every last bit (and as a consequence always running your analog gear in it's hottest stages) the distortion and noise can add up...

TOD · Post by **TOD** » Thu May 17, 2007 10:05 am

OldTimey wrote:
TOD wrote:
I also record as hot as possible without clipping in 24 bit, cause, Gosh Darn It, I want to use every bit!

TOD
what gearboy said about 24-bit vs. 16-bit...no need to fill every bit..just fill more than 16 of them going in, and you are in pretty damn good shape.

also, by trying to fill every last bit, you run the chance of possibly overloading the analog stages of your gear. if you run a preamp really hot, then you it is going to add high level distortion to the signal. which may sound "nice and warm" if it is a tube amp, or "punchy and raw" if it's a solid state pre...or if it's a cheap pre, just add nasty noise and signal distortion.

over the course of 24 tracks or so...all the "color" you have added to your tracks by consistently chasing every last bit (and as a consequence always running your analog gear in it's hottest stages) the distortion and noise can add up...

Thank you OldTimey and others for enlightening me. I have noticed "too hot" can lead to some mild distortion, (especially when tracking some powerful vocalists recently). I will pull the knob back down from "11"