Newbie question regarding waveform size and volume levels etc

forget what your waveforms look like.

just get a good clean recording.

taking peaks up to -3 is cutting it a bit close.

but the main thing to consider, is average RMS value.

you don't want that down around the -22 mark, -14 would be nicer.....

you have to learn how to control your input peaks, to get your overall rms up higher...


takes external compression and limiting.


you can always slam a limiter on a track at mixdown, and then it really doesn't matter how skinny your guitar waves were, as long as you're not bringing up noise floor that is apparent at the higher levels....

the trick is using compression on the way in, and having it be colorful but NON DESTRUCTIVE

all you're trying to do, is control the occasional wild peak that flys out, and drives your overall level down.


it's easier to do with distorted guitars, i find i rarely have to run external compression while tracking high gain guitars.
but loud chugs can swing it....
and loud clean guitar with a lot of depth of tone, can swing it.

control thy signal.

print it out, put it on your wall.

i was talking about tracking, not mastering.


and yes, there's a huge difference.

going in, you hit your convertors to translate analog into digital.

it is at THAT MOMENT...... that you commit your signal, headroom and all.
or lack of.

or with clipping.


now, if you have a wildly swinging signal that is peaking, that will drive your overall RMS value down.


think of a soundwave of the guitar track your recording, as the grand canyon.
versus a 2' deep ditch.


you want the BOTTOM of that wavelength, to be as close to the top as you can get it, but youi are limited by the peak of the wave.

using a compressor, TAMES that peak, allowing you to drive the signal into the convertor hotter.


now, if you skip that step, go ahead and set your levels on your convertor so that you don't clip your signal, you'll end up with a signal that is 'weak' in terms of overall RMS value.


i use compression, immediately after my mic preamp, and BEFORE the convertors, for two reasons:
1. tame the peaks.
2. color the sound the way i want it to be.


i've done this for 25 years or more.
i've gotten good at it.

the difference is, i can bring a track into sonar with peaks around -6db, giving myself headroom for later mixing tweaks....
and KNOW that my RMS is sitting pretty at about -12db, which is perfect level for me to mix.


slapping a compressor on AFTER your signal has already been committed to digital, forces you to deal with the problem AFTER it's already in digital form.

i trust my analog gear to do that a lot more than some plugin.

but that's ok, i use plugins too......
so, if it were me, i'd put a C4 compressor across that weak track, and find the sweet spot, where i haven 't amplified too much noise (my track are typically mid-low noise floor), and i haven't ruined the dynamics with TOO much compression, get my attack and release right, and give some WEIGHT to that track.



now, on mixdown, i like to leave at least 6db under 0, for mastering...
and typically my RMS of my mixes are anywhere from -18 to -12db.



on mastering, i drive the RMS to about -11db, peaking at -.1db.

but that's an entirely different conversation.

I still dont really get it, whats wrong with just setting your levels so they obviously are in a range that doesnt clip and then just use compression after to tame and colour it, i dont see why it is really an advanatge to compress before only to maybe reduce the risk of overloading a "good" recording take?!

sorry to op for steering this topic off a little.

Cian

I can see where batsbrew is coming from but also whack you are correct. I think a problem with tracking levels is that people have no reference when they are tracking and that is where they can get into trouble. I use K system level setting and metering when tracking especially. If you choose a K ref level of eg K - 14db then it is just a simple matter to get your rms levels sitting around - 14 db. You have 14 db of headroom above that and it is unlikely you will clip. You need decent VU's meters though and that helps in setting track levels. Using a DAW's peak level meters is not that useful. They don't tell you anything about rms levels really, only peak.

People get their peak meters consistent while tracking but they tell you nothing about the rms levels underneath. So rms levels often end up all over the place instead. It is better to keep rms levels consistent on tracks and don't worry about what the peaks are up to. They will take care of themselves. But you need VU's to do this properly.

The better the musician I have found the more consistent the input levels are as well. The advantage of this approach is that you can get a clean transparent sound going in and it stays that way inside your DAW. Then you can add the colour and sound you may want later. Running everything through a compressor on the way in will also impose that sound onto every track you record and sometimes you may not want that.

Digital plugins are fantastic now and if you are prepared to spend a little more on them you can easily match the sound and quality of the best analog units.

A compressor set for limiting can be useful on the way in set with a high threshold so it is not doing much most of the time. But it will catch a wild peak. You need your attack setting fast in this mode too otherwise that wild peak will slip through.

So you can get correct tracking levels everytime and you don't need any outboard gear to achieve it just common sense. I have been doing it for years.

Jeff,

Are you saying that you are using a digital compressor in the FX bin during tracking?  Is n't that non destructive then?  Is that limiting the wave form while tracking and compressing those stray peaks.  I must be missing something in your explaination.

I understand what Bat is saying, how an analogue compressor limits the peaks at the source; I guess I am not understanding how the digital compressor is limiting at the source?

You are very correct, it is easy to rush through this process of getting your levels optimally adjusted; I am certainly guilty of that.  Especially easy when you are recording yourself.  I do all of my compressing and limiting after/ ITB as well.  I used to be guilty of pushing my levels too hot; now I find myself over compensating and often having them set a little lower than I probably should.  I think I need to read up more about K-ref and maybe get some more accurate metering; spend more time getting consistant levels in the tracking phase.  It's funny, I spend a lot of time moving the mic around these days and listening for the best mic position, but then I will often forget to check and see where I am metering afterwards.  A little move of the mic here or there can make a big difference.

No. Anything in the Fx bin is not printed when you're tracking. The only way to print with any sort of effect is what you insert in your input chain, usually hardware, prior to hitting the AD.

 
Yep, this is what I always understood.  I was reading Jeff's message above and thought that maybe I have always been missing a way to do this while tracking with software.  I read most all of Jeff's writings and learn quite a lot from them, I thought maybe I had been missing something all of these years.
 
If I had a really good hardware compressor, I certainly would do as Bat is suggesting, which is control some of those peaks during the tracking.  I think it would just make things easier during mixing.  As long as you are not adding color or compressing too much.  As Jeff and Danny have mentioned on numerous occassions, software is getting so close to the real thing, that it is probably wiser to keep putting my money in plug ins; rather than hardware these days.  Saves on rack space anyway.
 
Incidentally, I have been having much success with layering compressors; not trying to make one do to much at one time, but rather squeezing a little at a time.

This last few posts clears this up thanks guys. I will stick with the post compression strategy since you dont have to commit to anything, like makeshift said iI had a good quality hardware analogue compressor then I would make us of it before the converters.

Cian

I'm in the 'compress on the way in' crowd. But slightly, just enough to keep your signal tight. And that's not to say you won't compress once the track is printed, there are lots of compressor types and techniques for using them. Many times its good for the track (and the mix) to do so.

You need to do a few test records, take a look at your signal level, and adjust accordingly, THEN record your track. The better the input signal quality, the less you'll have to adjust later on.

starise,

it is a low end compressor, that i mostly use on vocals.

i use a cheap DBX MC-6, it's based on the same circuit in the dbx 160, and i use the overeasy function.

it is an older classic sound that i've always been fond of, and i use it to 'color' the sound of a lot of my tracks, and it has become an integral part of my vocal sound.

i typically use a 1176-type compressor for level control if i need it, but with my bass tracks, running thru a sansamp and my A Design Audio MP-1 mic pre, there's usually not much need for compression, my guitar tracks are compressed enough going thru the boogie head, drums are typically samples from Superior Drummer, so it ends up being vocals, acoustic guitars, and percussion that i want to tame.

but the little MC-6 is a secret weapon.
i cannot get that sound out of any other compressor i've tried to squeeze it out of (i've used compressors by UA, presonus, avalon, but i have to say, the one time i used a Distressor, i really loved that machine)

that would be 1st on my list of 'wanna-get-one'

Almost any comp will flatten out the sound levels some.  Basically, most of us most of the time (I think) use a comp going in to squeeze down the dynamic range some.  It makes it easier to mix.  A mid-range unit should work fine for some basic leveling as long as you don't push it.  I had a DBX 163 that worked well on bass and some other stuff but was too noisey for many vocals.  But as Bat sez, sometimes that makes your sound.  

@ 

Thanks alot for all the replies, plenty of info to answer my question! :)

The journey continues.....