Do you print raised track levels before mixdown?

[vvv]

Inspired by here, I agree with and for the last year have been recording at that magic -12dB level and loving the results.

For ease of editing, however, I find myself boosting already recorded tracks to peak around -3dB, just so I can see 'em better; this also helps in mixdown for quick-setup "default levels".

One result is, my master fader is always at 0, and my track faders are always at some negative (ex., distorted electric rhythm guitars tend to be -12 to -14dB, when the vocal is running around -10 dB at the track faders.)

Anybody see anything not "best" practice about this?

[dave watkins]

if you are doing that just to see the wave forms better you might want to see if whatever daw you are using allows you to just zoom in on the waveform (without effecting the level) , then you wouldn't have to amplify things just to turn them down so you can keep your mix from clipping.

though i'd imagine that what you are currently doing wouldn't really have any serious problems sonically, as long as your outputs aren't clipping.

NOW BEHOLD! THIS THREAD SHALL GO ON FOR WAY TOO LONG!

[vvv]

No clipping issues.

The reason I do this is that while the DAW does allow magnification, it's a PITA to keep re-setting. As well, with tracks peaking at about -3dB I can look at the whole track and get a better sense of the dynamics through the song.

And then there's that "default" mix-level thing ...

[joninc]

i don't know what DAW you are on but it's a completely fast and painless thing to magnify the levels in cubase - i do it ALL THE TIME.

you want to keep the levels low also to not hit the plugins too hard - if you are boosting all your levels to -3, you are still feeding a lot of gain to plugins.

[MoreSpaceEcho]

how are you not clipping the master fader with all the tracks peaking at -3? seems unpossible.

[E-money]

If you are getting good sounding mixes, then go for it, but it does seem like kind of an odd workaround for what amounts to a visual issue.

[vvv]

how are you not clipping the master fader with all the tracks peaking at -3? seems unpossible.

As stated above,
"One result is, my master fader is always at 0, and my track faders are always at some negative (ex., distorted electric rhythm guitars tend to be -12 to -14dB, when the vocal is running around -10 dB at the track faders.)"

And yep, it's a visual issue. I'm sure I could learn new "defaults", altho, of course, the gain increases are done to a specific level which is more precise when applying rote fader-settings. I suppose I could even use -6 or -9db if there's a benefit ...

I use Cool Edit Pro 2.1 because I have for years now, and my work flow is quite fast (to me); I'm using a 2 x 2K P4 with XP Pro, what I am also used to.

Now, this notion of not "hitting the plugins too hard", I read that for the first time (in the other thread I linked?) today; is there any kind of objective info on that?

Bottom line is I am willing to change my habit of raising file levels if it makes a difference (I haven't heard a difference, but haven't done a lot of comparison - though I'd ask all y'all! ), but doing that would interfere with my work-flow. That said, I did lower my recording levels (what also did interfere with my work-flow) ...

No one else does this?

[MoreSpaceEcho]

oh ok, now i understand.

no, i never do that. seems silly to raise the gain on everything only to lower it again with the faders. it can't be that hard to resize the waveforms visually in cool edit.

the plug in thing is just gain staging. if you're peaking at -3, you put an eq plug on and boost +4db somewhere, you're clipping. if you just keep the levels low you never have to worry about that stuff.

[jgimbel]

i don't know what DAW you are on but it's a completely fast and painless thing to magnify the levels in cubase - i do it ALL THE TIME.

+1. I haven't used CoolEdit so I can't comment there, but I use Cubase too and I'm constantly zooming in, way in, out, or magnifying the waveform (vertically, Cubase has a slider for that). It's a constant thing, I don't even think about it, and I can't imagine working with things set static!

[farview]

Every DAW I have used has 3 zoom controls:

1. timeline zoom, to zoom in so you can see a single wave or sample

2. Track zoom, to make the channel width bigger or smaller

3. waveform zoom, to make the waveform look bigger so you can edit.

Doing a gain change so you can see something better is just goofy.

[vvv]

I love and respect you guys, also.

Signed,
goofy

[nordberg]

i use cool edit... well, audition now. zoom is super easy. can't you just go to the scroll bar on the left side of the screen and zoom? i can.

[Nick Sevilla]

Inspired by here, I agree with and for the last year have been recording at that magic -12dB level and loving the results.

For ease of editing, however, I find myself boosting already recorded tracks to peak around -3dB, just so I can see 'em better; this also helps in mixdown for quick-setup "default levels".

One result is, my master fader is always at 0, and my track faders are always at some negative (ex., distorted electric rhythm guitars tend to be -12 to -14dB, when the vocal is running around -10 dB at the track faders.)

Anybody see anything not "best" practice about this?

Hi vvv,

I'll go ahead and post a few observations about digital mixing et al, in order to clarify some misunderstandings about gain, gain structure, and similarly named "stuff".
Firstly, I have to mention "Gain Structure", what it is, and why it is important in any audio situation.
"Gain Structure" is the optimum setting for any device or devices in an audio signal chain, or path, so that the device(s) work at their best performance.
Every single audio device has a Range of optimum operation. This is a window, if you can picture this, within which the audio passes through the device and is processed without causing the audio unwanted artifacts, distortion, defects, etc.
It is important to note that this Window, is NOT the entire dynamic range of the device, but a smaller range of Level in which the device processes the sound best.
A good rule of thumb to find a devices "Window" of optimum operating range, is to run some program material through it both on bypass, and with the unit engaged, especially before using it for the first time. You can find out useful things like at what lowest level does the unit still work correctly without adding too much self noise, and also where the maximum level is before it starts doing the Nasty on the audio.
Every audio processing device has a different "Window". So, for your equipment, it would be great if you found what works with yours, and note it somewhere handy. Now, this is only for ONE device, so far. Typically, more than one device gets used when recording and manipulating audio, in normal every day engineering.
Ok, now for what everyone commonly refers to as "Gain Staging", or setting the optimum gain between more than one device:

Once you know what "Window" or Level Range each of your devices works at best, it is much easier to get a proper signal chain going, which will handle ANY audio going through it properly.

Example: I'll create one signal chain, and describe a few places where putting close attention to the "Window" or optimum gain setting is to be observed.

1.- Loud Microphone on a Loud electric guitar(as an example of source), FIRST Gain Setting (Yes, you DO set a level at the microphone...)
2.- Microphone preamplifier. SECOND Gain Setting (yes, second...I explain in a bit why...)
3.- Equalizer (may or may not be part of the MIc Preamplifier device) THIRD Gain Setting.
4.- Compressor, Fourth Gain Setting.
5.- Recorder (Tape or Digital Converter, same principle) Fifth Gain Setting.

OK, we now have a basic recording chain, and there are already 5 places where you have to pay attention to the Levels going between devices. And they say it's easy as pie to make a record... allrighty then.
You have to set your CLEANEST, BEST, Level of operation BETWEEN the microphone and the Sound Source. This undoubtedly is where many many many people screw things up, and end up using things like your "Normalization" and "Gain" plug Ins in the DAW. WRONG. At least for a recording you are doing yourself, please try to get the Level right here. Too Loud? either turn down the amplifier, place the mic further, or... TA DA: use a Pad, either in the mic, or an "inline" one right after it, but BEFORE the mic preamplifier. This is your FIRST Gain Setting. Get this one right, and the rest is a lot easier, and everyone will marvel at your genius.
Second : The Mic Preamplifier. This is a Crucial point, as depending on the Sensitivity of the mic, "Sensitivity" being the ability of your microphone to turn weak signals into usable electric signals that your mic pre can amplify, you will have to adjust the Impedance, the Input Gain (totally misnamed knob, trust me in this), and then finally the Output Gain to the next chain. Why do I mention Sensitivity Twice? So you pay attention to it. Here is how a Mic Preamplifier REALLY works: It takes a VERY SMALL signal coming from the microphone, and, depending on the preamp type, will AMPLIFY it by a factor of 10 or more (being a Logarithmic thing in some amps, and Linear on others, go figure...) The dBu scale, and other dB scales are mostly Logarithmic, so most mic preamps should be Logarithmic in operation.
And this is why you fretted with the mic before walking over to your mic preamp, so you could get the BEST (not LOUDEST mind you) tiny electrical flow from the mic to the mic preamp. Since you are a good engineer and you did get the BEST signal from the mic and source, all you will do now is bring up that tiny little signal up to a Level that all the other devices further down the Signal Chain can actually "see" and process properly.
Now, on to the next signal processing device, the EQ. Just because it may be in the same box as the mic preamp, does NOT mean you don't have to pay attention to the level going into it. Say you need to Boost 6K by 6 dB, and your incoming Level is already very loud, causing the EQ to distort the signal... that's right, you'll have to turn down the input to it by... 6dB.
Some devices have a Gain knob after the mic pre, and before any EQ or other processing. Most do not, so you have to compromise and turn down the input to the Mic preamp, in order to use the EQ properly. Oops on the part of the mfr.
Ok, now that you have the setting on the mic, mic pre and EQ set to where you are getting the right AMOUNT (Level) of signal, and it is not getting fried or losing resolution (too weak), then you can go on to the Compressor, in our example. Hopefully your EQ device has an Output Gain knob on it. Hopefully. This can also be the Fader on your console (if you are doing a post-fader Direct Out to the compressor.)
You should now be able to feed your signal into the Compressor, WITHOUT ever having to go back to re-adjust any Gain Settings anywhere on the chain up to this point. Your compressor, and I will use one of mine as an example , and LA2A... should have an INPUT Gain (in my LA2A it is simply "GAIN") and you should start with your compressor on BYPASS, at least if it is like an LA2A. So, turn down both the knobs, GAIN and Gain Reduction. This way you will FISRT set the amount of signal going through the compressor, unitl you get the SAME amount going in and going out, in other words NO GAIN CHANGE... at least not yet...
To set the BEST Levle for the next to last device, you then set the compressor to reduce by the amount you want, and only then worry about the Tape input or converter input level. Ideally you would use the compressor in a Console insert, so that the Fader on said console channel is the one that is your Final gain adjustment before recording the signal. If not, you might consider using something to do the final gain setting before hitting the recording device... just a thought. In my case I like flying by the seat of my pants... sometimes... unless I have never recorded the artist at the other end of the signal chain, in which case I will ALWAYS use a console fader as my final gain setting before the recording Input.
Now, finally you have gotten to the Recording devices Input, and are going to set this one so that ANY signal arriving at the recording portion (magnetic head, analog to digital converter chip, whatever) will go through without getting mangled on the way through. On digital devices, mainly you set a common single setting for all your inputs, and leave it at that. On an Analog device, mostly it is the same, although I have seen certain "name" engineers set the inputs to their 2" tape machine differently for different signals (danger Will Robinson, do not attempt at home). They do this because they, like many "Method Actors" of our time, have a certain SET WAY of doing things. SO, they know WTF they are doing, and part of why their records sound the way they do...
Now, after reading this horrendously long and boring post, you ask yourself.... WHY, Oh man, WHY did you post this?

Because, Although Normalization might SEEM like a great tool to use, in the hands of the ignorant Masses, it mostly causes untold and IRREVERSIBLE damage to the poor little mic signal. I am an advocate for the poor little mic signals of the World, may they bring us indefinite and undescribable Joy Forever.

Cheers

[vvv]

Nice post, and I do understand your various points but feel like mebbe mine was missed. Damn, I feel bad - that was a lotta of very good info, and I bet a hassle to type! Thanks!

However, I feel confident in my gain-staging, confident in my incoming signal to the box. Hardware is not the issue (altho' I am still curious about joninc's representation that level is a factor in how a software effect sounds, i.e., processes ...)

My questions have to do with what I do with the already-recorded track, already in the box.

I use 32-bit float 44.1, and my tracks come in, gain-staged, etc. such that I am happy up to this point in the recording process, with a peak level in the -12 to -6 dB area.

It's what comes after that I posted and asked about.

Me, I like to limit spikey signals like rhythm guitars (as mine, for example, often are), even where I don't use compression, and I pretty much pass-filter everything. As I do that processing, then, (or just cleaning up the tracks, "topping-&-tailing", etc.) increasing the signal level of the track is easy, and so I do so, bringing the peak level (limited or not) to the -3dB area, in the box.

Again, this serves to give me a better visual analysis, especially when I'm looking at an individual track for editing, but also when I'm looking at 32 tracks in mixdown. A bonus benefit is for what I referred to as "default settings", where I can easily get a quick rough mix going.

My question, then, was, is there a harm in so doing, in raising the already-recorded tracks (that I am happy with) to a peak level averaging -3dB to obtain the above-listed benefits.

I hope that is clearer.

Signed,
silly & goofy



(BTW, I never use normalization, never have, I just tend to leave my digital files peaking around -3dB after processing ...)

[MoreSpaceEcho]

increasing the signal level of the track is easy

i don't understand why you feel you need to do this. besides being able to see it better. if it's peaking between -12 and -6 already, that's plenty loud enough. i'm forever turning stuff DOWN so i can keep a few db of headroom on the master fader.