I tend to shoot for mix levels around -20dBRMS. That should give you occasional peaks around maybe -4 or -3dBFS or so -- generally speaking.

Lower is fine also.

On your individual tracking levels - There's a lot of confusion around that subject... The "-18" thing comes up a lot, as it should. But what you're really shooting for are levels that have a rough RMS equivalent to around -18dBFS (more accurately, -18dBRMS in this case).

Assuming your converters are calibrated to -18 (most modern units seem to be), that's going to equate to an analog signal around 0dBVU, which is where your gear is spec'd to. So you don't need to *peak* at -18 (although, you certainly can if you're comfortable with that) but you want the "bulk" of the signal around there.

Roger that, Bulk of the signal around -18dB, not a hard etched in stone rule, but a good place to be.

Thanks for the help, Im shooting for my song to have the potential to be able to be brought up in volume while still maintaining dynamics as much as possible(who isnt?)

I got Roger Nichols Dynam-Izer, it has a meter that lets you see where a the bulk of the signal is. So that should come in handy.

Another thing, when getting statistics on RMS levels, that is basically seeing where the bulk of the signal is right?

Ahhh this gives me so much more confidence, all this time i was always trying to get levels close to -6db. Which doesnt make much sense due to horrible dynamics in the track. I really dont know anything so if this could be clarified i would appreciate it. Even though the mix will sound quite at -18 db once you get it mastered it will be much louder?? Am i totally off or what?

Ataraxia wrote: Ahhh this gives me so much more confidence, all this time i was always trying to get levels close to -6db. Which doesnt make much sense due to horrible dynamics in the track. I really dont know anything so if this could be clarified i would appreciate it. Even though the mix will sound quite at -18 db once you get it mastered it will be much louder?? Am i totally off or what?

Yes, it could be louder if this is what you are looking for and ask your ME to do so. And no, you are not totally off.

Actually you did a sample for a song in one of my old bands. Vanity Orchard, remember? Tell me if you think you could do much with this http://www.myspace.com/teharbitrarys . I tried some samples with some other people and i must say you definately had the best ear for what we did so maybe in the near future that might be a good idea. At the current moment im more worried about getting more recording done.

Well can you have too much headroom?
For instance
Recording, not hot enough, with majority of the signal at -30dB and some max peaks at -21dB.
Would that be too low?
I mean it would give headroom for days.
But would this not be good practice?

And mainly, if your record at this low level, could you just amplify the signal to a hotter level offline within the multitrack program(adobe audition 2.0) useing the Gain Process
And to the same degree/Vice Versa
Couldnt you, if you recorded too hot, but didnt clip, de-amplify the signal offline, so as to give more headroom?

Could someone clarify.

well you could be to low in level, but if you are printing at 24 bit, you have to be quite low to make a difference. anytime you number crunch and move levels around, there are all sorts of variables. I think it depends on the software and hardware that you are using as to whether or not it's a good idea to change levels. The point is you really shouldn't have to change levels. I really don't care what level I get, as long as eveything is intact. I get them all over the place and it takes me about 5 seconds to push or pull it to where I need.

Dozer: I think Michael did answer pretty well to your questions.

Ataraxia: I remember the name of the band but not the song, sorry. Glad you liked.

I think the MOST important thing when mixing is NOT to over compress.

A Mastering Engineer can easily change levels but, if you compress your mix before hand you will severely limit the task of the Mastering Eng.

I like it when the mix engineer uses a compressor. I think compression is an important part of a mix (given the style) as long as it's used for the sound and vibe. A limiter on the other hand strapped across the 2-bus that's used to get more level is the problem for me.

When I compress the mix I use about a 1.5-2.0:1 RATIO with a threshold relative to the material but, basically to kick in only at certain peaks. I mostly do it to give it some warmth (I use a TUBE unit) and prep the mix, so to speak, to get things in the neighborhood of where I like them to be.

Once in the MASTERING studio, I then let the ME use their compressors and limiters as needed for achieving the final character and volume.

I also do totally un-compressed mixes for certain projects.

Hey this is for massive mastering,
lm reading this while checking my levels i have on a 15 track song and the levels are actually around the -30!! peaking at 24. are these the kind of levels you would expect when turning it in?
this is a good thread topic by the way. it has been on my mind how the levels and eq should be and how much we should rely on a ME to top off the master tape.
also, should the eq be adjusted in order for you to bring it out in the mix or how much control would a ME have regarding what eq the song has? l hope that you understand what it is i am trying to figure out. l think its around the lines of whether every track in the song is audible and clear or not in simple terms but t doesn't hurt to hear this from a ME.

Senior fico

I wouldn't really see any reason to mix that low. It's not "dangerously" low or anything, but you have to be cranking the amps pretty hard to get to a decent mixing level in the room. Absolute peaks at -6, even lower, maybe -10dBFS is great.

But if that's where the mix falls and you're happy with it, that's okay too.

I've been working on a project all evening that's averaging -40dBRMS with a few occasional peaks up around -6 (but mostly around -16 or so). It turns into a gain-staging challenge, but I'd much rather meet that challenge than end up trying to make some bad peaks sound less irritating.

also, should the eq be adjusted in order for you to bring it out in the mix or how much control would a ME have regarding what eq the song has?

That part I'm not clear on for some reason...

I'm with John - it's VERY low, but not horrible.

I'd much prefer this to my normal files that I receive with peaks at 0.0dBFS and average (RMS) levels at -10dBFS. (Seriously - I get so many files that are over-limited to begin with, it's not even funny.)

What's common around here is - bands go to local basement studios and the basement studios limit the hell out of the mix to make it as loud as possible. Then they send that to me and I deal with what I get.

Or worse - they send me the MP3 as burned out of ITunes which apparently limits the track as well (I don't know for sure - I don't use ITunes.)

So...if you are sending your mixes which have plenty of headroom, I'm sure you're making your ME happy. Bear in mind though...if you have THAT much headroom, you might be treading on dangerous ground...when boosted enough to get appropriate levels, you might be bringing up digital black to an audible level (not pretty).

My personal favorites are those mixes which come to me that never peak at 0.0 dBFS, sound pretty good to begin with (usually have some compression on the bus) and have RMS levels around -20dBFS give or take 5dB or so.

, you might be bringing up digital black to an audible level (not pretty).

after reading this i havent responded for two days. i was so shook at the thought of this ..."Digital..Black" Sounds scarier than a ghost

That's some great info ... just one question on (sorta) topic ...

Do you add any processing on the stereo outs? Even EQ? I've heard different things and am not sure what's more common/the proper way to do it.

digital black would be nothing. he probably means bringing up the noise floor.

Adding processing on the 2-bus is done if it's needed during a mix. just remember that once you add it, you have to recall the mix to take it off. If you're not sure if you should add it or not, then don't add it. One shouldn't be putting stuff on the 2-bus "Just because".

Michael Fossenkemper wrote: digital black would be nothing. he probably means bringing up the noise floor.

Yeah...technically digital black is 0000000000000000 (or add 00000000 in 24 bit) and therefore is nothing. I do mean the digital noise floor - which is hiding in digital black (or just above digital black) until printed. Then of course, by raising the amplitude of the mix 20 dB or so (especially with a 16 bit file where digital noise occurs at -96dBFS or higher depending upon dither) definitely brings that noise to a quite audible level.

Obviously if mixing and mastering at 24 bit, a 20 dB increase won't kill bring the digital noise to an audible level (unless you're listening at insane levels) BUT....the file is automatically truncated (or downgraded - no bit truncation actually occuring) to 16 bit noise levels without even adding dither yet.

In rock or pop, this really isn't much of a problem. In classical, jazz or even some country, it's a problem.

...Or what you are probably trying to say, is that you will have to raise the sound of the original least significant bit (the audio turning on and off from digital black to the first digital representation of the audio) up to more audible levels, having recorded more important parts of the audio signal so close the the least significant bit. I doubt this will be a problem, if you are recording/mixing in 24 bit; but with levels this low, you are not even getting to the nominal level of your analog gear anyway.
And I'm not sure what the poster means exactly by this:

the levels are actually around the -30!! peaking at 24

Average levels around -30 with maximum peak amplitude of -24? or maximum average level of -24?
Either way, I would think you are tilting the other way now as far as the benefits of headroom go. There are also disadvantages to leaving too much headroom...

This is a very informative thread. I have one question: Do these rms and peak levels also appy to individual instrument tracks w/in the mix, or just the finished mix?

mobilelab wrote: This is a very informative thread. I have one question: Do these rms and peak levels also appy to individual instrument tracks w/in the mix, or just the finished mix?

No - just the final mix.

An individual instrument may (and likely will) have a VERY low RMS signal.

Mix as usual, just keep the headroom there.