Help With Mastering(if thats even what i'm doing)

B

bdam123

Member
So after I get my mix complete I bounce the mix as a stereo two track and import it into another sessions where I might add some air and hit it with an L2. Now I am trying to get these recordings to be as loud as possible (I'll explain why in a second).

I've listened to a few professional mainstream recordings and I'm shooting to be right under them. I guess between -14 to
-12db RMS. The problem I'm facing is that in order for me to get there I have to slam the L2 pretty hard; like around 8-9db of gain reduction (the threshold is at like -11). I feel like this is extreme. Although I don't hear any distortion, when I analyze the curve its showing that I'm over(right around 62Hz).

My question is am I really over? I set my ceiling on the L2 at -0.2 and my analyzer's meter are saying thats where I am at but the visual EQ curve is showing me a spike. This is only true when I ask it to detect peaks but isn't a peak thats going over gonna give me distortion? Like I said, I can't hear it through my monitors I just wanna be safe when I let someone else listen to it through a different system.

Also are there other techniques for me to achieve optimum loudness without hurting the recording so bad? My pre-master mix is peaking at about -3db.

I'm a part of a songwriting/production collective and we're getting ready to bring these mixes to meetings with A&Rs. I know that the people judging my music aren't the smartest of the bunch so if loudness = better to them (which is usually the case) I'm gonna play their little game and get these as loud as I can.

Thanks to anyone who lends me some advice. Peace.
 
I can't quite decipher what you mean by looking at the "visual EQ curve"... do you mean a frequency spectrum analyser?

Anyway, you may be encountering what are known as intersample peaks, which are peaks arising from the interpolation between samples during digital-analog conversion... the samples themselves may not (and cannot in fixed-point files) have values above 0dBFS, but there can be peaks with higher values when the wave represented by the samples is reconstructed.

In most situations these peaks will only be to +0.1dBFS or so and, in my experience anyway (though I may get shot by some of the mastering guys), aren't anything huge to worry about. Most, if not all playback gear nowadays (including cheap MP3 players, etc) has ample headroom and ability to deal with these peaks without any problems or clipping.

Note that the FFTs used in the frequency analysers it sounds like you are using may not be accurate enough to rely on for stuff like this, but you seem to have the right idea by trusting your ears more than what you see on the screen.

Some limiter plugins and level meter plugins deal with intersample peaks better than others... its all a matter of oversampling, interpolation, etc, and complex stuff that most of us folks don't need to worry ourselves with :p

As for what you can do to the mix... if you keep peaks and dynamics under control (but not killed completely) and get everything gelling together consistently whilst mixing, you won't have to smash it through the limiter so hard. For example, rather than relying on the limiter to deal with loads of drum transients, trying routing the drums through a bus and apply some compression there (or compression on individual drums) to get them sitting in the mix more, but you can control this better and get the drums compressed as you want them rather than crossing your fingers and hoping for the best when you try and squash them into the mix with the L2. Take this advice with a pinch of salt... you might get some responses from the MEs that frequent these boards, in which case I would trust them more :)
 
IME, most A&R folks aren't going to give a rat's patootie about the mastering quality, volume level, etc., since none of that will apply if they sign you and stick you into a studio to re-record anyway. And since you're not going to be stuck in the middle of a playlist when they listen, "competitive" volume levels are not relevant.

They are listening solely to the act itself - mostly the first three bars of the first song. If you don't have them interested by then, the interview is probably going to be over pretty quickly.

That said, though, I'd give two recommendations on your mastering for loudness (I'll skip the usual advice: "Don't". ;) ) First, throw away the numbers, the meters, the frequency response graph, and all those visual aids. They mean nothing. Use your ears to determine how far your mix wants to go before it starts sounding artificially pushed, and then back off of that level by a dB or so. The numbers will be what the numbers will be, who cares - the result whatever the numbers will be a maximum volume level suitable for your mix.

Second, try not to push the whole mix up against the wall in one fell swoop. Take two or three smaller bites using less extreme compression settings instead of one big slam with a brick limiter. listen to the in-between results, and if you get something starting to stick out due to the compression (like perhaps your 62Hz bump), then tame it with some targeted EQ evfoe moving on to the next compression stage.

G.
 
The pro ME's will chime in soon enough to help you out, but if your limiter is set at -0.2 then you simply won't "clip"...that's the brick wall. That doesn't mean you won't have distortion because you can squash the hell out of a mix, keeping it at -0.2 limited, and distort the death out of it. If you're presenting a collection of songs to some legit A&R guys I highly recommend you have one of the ME's here master your tracks...if for no other reason than boosting your db to commercial level also can drastically alter the sonic balance of your mix and the next thing you know it's loud as hell but your low end is booming and your highs are harsh. There are many qualified guys here to not only educate you but to provide their services if you wish. A good ME can make a decent mix good, a good mix great and a great mix absolutely stellar. If you're not going for the whole redbook CD master for duplication at this point, many offer online services where you can simply upload your files, they do their thing and send them back. The rest would be up to you. I'm a huge proponant of professional ME's to make the most of your tracks (obviously)...do with my 2 cents what you will. Good luck, hope you land a record deal (no sarcasm intended whatsoever), and the intelligent folks will be around soon enough. :D Cheers
 
I agree with southside except for one thing. He's right about these guys and how the meetings go...I've been in a couple (one went well, one ending before the first song was over). That being said, I personally think that it does matter what your production sounds like. There's an old adage that a great song will find its way past the production no matter what. For a true artist with a good ear for "hits"...whatever the hell that means before the public has heard it, this may be true. But to an A&R guy with lesser ears for true artistic talent (and man, they are out there), maybe not. A good song with great production can sound great to some of these guys if you've got a decent hook. A great song that sounds amiturish (not saying your songs do) can leave some of these guys thinking the song isn't up to par. Sad but true. The best path forward in my opinion is to walk in there with the best sounding mix you can muster and let the chips fall where they may. At least then you can sleep at night knowing that you were judged based on the strength of your songs without other factors in the mix. As for commercial levels, southside is certainly right. Your disc will be put on and play and most likely not "competing" with others in this particular setting...they don't have the time to A/B and even then, levels are not what they're looking for. This doesn't mean that some real mastering wouldn't do you some good. Judgment call...
 
Don't get me wrong, I'm not advocating going in with a bad-sounding mix. I just think that the amount of effort put into *mastering* is kind of wasted on A&R, and as to whether the song winds up RMSing at -12 or -16 isn't going to much matter, because that's not even what they're listening to.

When they listen to a demo, what they hear is not what is on that disc, but what they think it will wind up sounding like after their label gets their greasy paws on it - which may be an entirely different production style altogether form the demo.

Make it sound very good, sure, but you're not going to trick anybody into hearing or believing anything beyond what they intend to listen for. Let them hear your band and it's talents, not a bunch of studio sheen and crapola that will just annoy them because it just makes it harder for them to hear what they really need to hear to do their job.

G.
 
We've dealt with a few A&Rs and we're orbiting the industry pretty hard right now and we're fully aware of the fuckery that is going on. I would say 2% of these guys actually know what they are talking about. A great ear could hear a hit even if the demo was sung over a single acoustic guitar but thats just not the case nowadays. If I can get the ball in my court by representing these songs as best as I can/tricking these A&Rs i.e great reference vocals, mixes, and levels thats what I'm gonna do.

I'm pretty happy with what I'm hearing right now. I just know that my ear isn't as fine tuned as some producers, mixers, or ME so I gotta come on here and ask these questions. Thanks for the replies.
 
bdam123 said:
Also are there other techniques for me to achieve optimum loudness without hurting the recording so bad? My pre-master mix is peaking at about -3db...
Click to expand...
You might want to try tracking your source less hot at around -20dbFS and then mix peaking at -6dbFS. That will give you a bit more headroom where you can max out your RMS without squashing your mix too much.
 
NYMorningstar said:
You might want to try tracking your source less hot at around -20dbFS and then mix peaking at -6dbFS. That will give you a bit more headroom where you can max out your RMS without squashing your mix too much.
Click to expand...

Could you explain how this works?

All my instruments are this point are virtual so should I just bounce all the wavs at about-20db? The same with the vocals? Is there such a thing as having the pre-master too low?
 
bdam123 said:
Is there such a thing as having the pre-master too low?
Click to expand...
You'd have to try really, really hard to have a mix that's "too low". A quick look at the math:

Assuming you're recording/mixing at 24-bit (which you should be at minimum), you have 140dB of "canvas" range between clipping on the top and the digital noise floor on the bottom on which to paint your sonic picture. Your average analog signal records with an average maximum dynamic range of maybe 65dB. This means you could record that signal with peaks as low as -75dBFS without having to worry about the noise floor.

You're recording direct and not analog, but even if you added another 25dB of dynamic range that way (which you're probably not quite), you could record with peaks of -50dBFS and still be OK.

I'm not advocating actually going that low, because you'll just wind up having to boost things eventually anyway, but it should give you the idea that there's a lot more room below than there is above, so it's a good idea to give yourself as much headroom as you need for mixing. The numbers NYM gave are pretty much in the right ballpark for most applications.

G.
 
How does having more or less headroom effect my efforts in mastering? NYM mentions that if I come in at -6db as opposed to -3db that I could reach my RMS goals without squashing my mix to hard. Even if I came in at a lower level with my pre-master wouldn't I eventually have to take it up to that brickwall to get my RMS where i want it?

Sorry, I'm a total newbie at all of this and I'm just trying to wrap my head around all this technical stuff to better prepare for getting these records where I want them.
 
A big part of it has to do not just with the final mixdown levels, but with the levels all the way down the line. A lot of folks want to "use all the bits" and record hot mix hot and master hot. But by keeping the levels nominal from the get-go, from the original tracking on down, it can tend to naturally "force" a cleaner sound all the way down the line, whether one can hear it or not. This results in a cleaner mix that can take more folding and mutilating in mastering before falling apart; i.e. the cleaner the signal path from start-to-finish, the more robust the mix and the harder you can push it before distortions reveal themselves.

G.
 
Ok, I get it.

So a quick a question about my tracks I import into my session and my vocal recording.
I use Logic to compose all my instrumentals and I bounce the tracks roughly leveled. Should I make sure the loudest instrument is at about -20db?

Whats a good level to record my vocals to take advantage of what we've just been talking about? Around -20db also?
 
bdam123 said:
I use Logic to compose all my instrumentals and I bounce the tracks roughly leveled. Should I make sure the loudest instrument is at about -20db?

Whats a good level to record my vocals to take advantage of what we've just been talking about? Around -20db also?
Click to expand...
Yes and yes. Your -20db on your digital meters is approximately 0db on the analog meters. Coming in at that level is the sweet spot for your a/d converters.
 
You must log in or register to reply here.
Back
Top