some thoughts/questions about mastering

P

producerkid

New member
okay i want to throw out two different mastering scenarios and i want you to tell me if im fairly accurate on my methods.

preface: i know for a fact that mastering involves several different approaches ranging from just strapping a limiter across the buss to corrective mastering techniques like multi-band compression.

the less is more approach (suited for spot-on mixes and excellent performance):

input>multiband parametric eq>limiter (possibly dynamic range compression inserted between input and the eq)

the super-slick approach (make it loud, make it hot, make it slam):

input>hi-pass>multiband parametric eq (shaping)>multiband comp (tweaking)>dynamic range compression>"sizzle" eq or analog simulation for sweetening>limiter or clipping (and i mean LIGHT clipping if your routing through a good converter)

if im way off in either of these methods, why? is my chain correct in assembly?

also, i think im going to try running a 0db white noise track to a/b my resulting sound with (as a reference point for bigness/overall level). sound like a plan? or is there another reference signal for absolute volume that i should use instead of white noise. comparing with other records is ok for a general test, but i am talking absolute 0db, not a variable.

eager to get some responses on this one folks....:D
 
producerkid said:
or is there another reference signal for absolute volume that i should use
Click to expand...
Your ears. If you're going to "master" (here's yet another thread that has no idea what "mastering" actually is supposed to mean or to do), you had better learn to hear the two seperate properties of loudness and quality seperately but at the same time, because the best reference to use is the mix you're working on itself.

Every mix - and this is especially true with more amateur mixes - has a loudness point where the louder you push the mix the more the sonic quality decreases. This point is unique to every mix.

You need to - before anything else - actually hear and recognize this yourself, and not just let loudness trick you into thinking that things sound better just because they're louder. Then, once you actually have the ears to do what you want to do, you can decide just how much, if any, quality you're willing to pay as a trade off for extra volume. You think that Coldplay or Fastball or RHCP album that has been pushed to the volume wall sounds okay? You should have heard how much better they sounded before the ME smashed the life out of them.

As for the rest, there are no recipies. Do what your experience tells you you should do based upon what your ears tell you the mix needs.

G.
 
okay, maybe i shoyuld have prefaced this with "i know making something loud makes it sound like shit". i know the difference between volume and quality because i am constantly forced to cross the lines in the type of work that i do.

i make commercial music. lol

trust me, i grew up on 70's rock so i know the whole "feel like ya cant turn it up loud enough" thing that today's recordings just dont have. but if you cant beat em, join em. everybody wants their recordings to be loud as fuck, and i intend on giving them just that. fidelity and volume are two complete different bastards that co-depend on eachother.

to be completely honest, i think the "super-smash" sound kicks ass when its right. its just that modern sound. however, yeah, when you try and make a bad mix loud of course it sounds like shit because it was a bad mix to begin with.

so basicly, getting past all the "shouldn'ts and "don't do's", would you say the theoretics of my first post are in the ballpark?

p.s. you must be talking about "Californication" with mention of RHCP. that album has always sounded like shit to me. lol im not even sure mastering worsened it, theres just some things about the drums and guitar sounds on that album that really bother me (mainly the drums though). i feel pretty safe to say it was rushed at mix. has that vibe to me because i end up there all the time :rolleyes:
 
It sounds like you're approaching this way too pseudo-scientifically. As Glen said there are no recipes. Every song has a different approach.

Learn the techniques of how to apply equalization, compression, M/S processing, parallel comp, MBC, etc. and what they are used to correct. Develop your ear to the point where you can recognize when and how these techniques should be used.

Then it's all just a matter of listening, correcting, and listening again to the results until you've reached your intended goal.

White noise at 0db is likely a bad choice for a sonic goal.
 
producerkid said:
so basicly, getting past all the "shouldn'ts and "don't do's", would you say the theoretics of my first post are in the ballpark?
Click to expand...
You completely misread or misunderstood my post.

I didn't tell you (as much as I'd like to) that you shouldn't sacrifice quality for volume, I said the choice is yours as to how much of one you wish to sacrifice for the other. But in order to make a good choice you have to be able to hear the difference. If you can hear the difference, and if you can make that trade off decision honestly, then the rest of your question is rendered academic.
producerkid said:
i make commercial music.
Click to expand...
Please explain what that means. What is "commercial music"?
producerkid said:
would you say the theoretics of my first post are in the ballpark?
Click to expand...
IMHO, and with respect, they're not even close. The entire concept of there being an ideal intrinsic RMS level to reference against is misguided, as is the concept that there is a standard recipe for signal chain which one should follow. You're looking for recipes/presets/shortcuts that are entirely independent of any evidence you may have in front of your ears, and it just don't work that way. Not, at least, if you want to do a quality job and do it consistently.

Look, every mix has it's own character. The assumption that every mix will need limited bandwidth compression instead of parametric shaping followed by broadband compression (or vice versa), or that everything will need to be slammed into a hard-knee limiter instead of multiple stages of moderate compression followed by a clipped re-route through an A/D converter (or vice versa), or that the mix will not ask for mid/side processing (or vice versa) is ludicrous. There is no standard signal chain. You need to give the mix what it needs, and every mix will be different.

It's the same thing with loudness. You said yourself that bad mixes will sound worse when pushed than good ones will. Then what is it that makes you think that there is a standard reference level to shoot for? You think Californication sounds awful? Just try pushing some amateur tracking that loud; it can be unlistenable.

And this doesn't even begin to cover the fact that the musical content plays into this as well. The denser the sonic material, the higher it's natural RMS level. A wall of guitar sustain will have a higher natural RMS than a clean picked solo. Even more importantly, it will have different apparent volumes at the same RMS level; put them both at the same RMS, and the clean guitar will actually sound louder to the human ear. For this reason, more sparsely arranged ballads will require a different RMS than dense, in-your-face anthems in order to sound the same volume, and a blues, reggae or folk recording will commonly require different treatment than a hard rock, metal, or big band for the same reasons.

The real process, in words, is simple. To engineer audio well requires only three things; the ear to hear and analyze the sounds you have in front of you, the knowledge/experience of how your gear works and exactly how that gear actually affects what you will hear, and finally a clear mental image - not just an idea or concept, but an actual sonic "image" that you can hear in your head - of how you want the final result to sound.

When you have those three, then all your questions about signal chain will answer themselves, and the resulting apparent volume will fall into place on it's own.

G.
 
producerkid said:
also, i think im going to try running a 0db white noise track to a/b my resulting sound with (as a reference point for bigness/overall level). sound like a plan? or is there another reference signal for absolute volume that i should use instead of white noise.
Click to expand...


That's awesome. I never thought the day would come that a newb would be "seriously" asking that very question. Am I understanding this correctly? Are you honestly asking if 0 db of white noise is a valid reference point?

That's awesome! :D Rock on.

Hey, if your stuff can't hang with 0 db of pure, unadulterated white noise ... then you might as well be mixing Karen Carpenter or James Taylor!

I can't wait for all the "How to get my mixes to sound like white noise" threads. :D Looks like they're already here! Thanks for making my day.


.
 
Just throwing it out there- Californication would've sounded pretty bad before getting mastered; of course I'm not discrediting the part that mastering played in getting that album to sound sonically terrible, but it's certainly not just the ME's fault. The album sounds like a bad demo, and I've heard better mixing from my limited time on this site. In fact all the RHCP albums from Californication on have sounded pretty bad, given the budget that they must get for recording.
 
for fucks sake

okay, i would quote a few things, but im not gonna cuz im lazy after a long night of guitars (not that im bitching - mesa dual rec/jcm2000/bogner and mesa cabs....mmmm)

first and foremost, im not a n00b.

there i said it lol.

i've had 11 years of experience with this stuff (started on a 4-track and a set of bookshelf speakers when i was 9). my life has been about nothing but audio since i figured out that quality top end was a thing of beauty. however, in saying this i am not claiming to "know it all". i am just saying that i have been around the proverbial block as thoroughly as anyone could several times and i know sonic integrity and the the importance of it. im trying to learn as much as i can whenever i can and im just trying some new aproaches now. ive made a lot of shit records, and ive made a lot of really immaculate sounding records that you would never think were the result of the starting material (that being said, alot of my pseudo-scientific way of thinking comes from the fact that ive probably had more experience in polishing turds than mixing proper recordings).

point is, i know i aint right about alot of stuff, but i dont think im completely ignorant in my beliefs.....or....wait....:eek:

the white noise scenario is not intended to be an overall goal for volume, but a reference point to make sure im NOT squashing the everlasting shit out of what im working on (at least as much as i can help it) while still keeping the volume levels up.

btw, when i say i make commercial music, i just mean that my production/mixing style is relatively top 40 (not limited to, but consisting mainly of) so most of the stuff i do is aimed at being "pop" and is generally radio-ready.

i forgot the point i was trying to make.....oh well :rolleyes:

hell, while ive got your attention, check this stuff out and give me some feedback (or at least what you can tell from the myspace streaming) if you feel so inclined.

http://www.myspace.com/erisonmusic

http://www.myspace.com/dividedbyfriday (just finished this and the mix/master was a major rush job - just a disclaimer)
 
No offense intended, it was just a very odd question from someone with your experience.

In comparing loudness levels frequency content and density of the material has to be considered just as much as what the meter is showing (see The Phon). Unless you are mixing/mastering white noise it doesn't make much sense IMHO to use it as a reference for anything.

As a result one needs to use a reference that is comparable not only in level, overall character of the frequency spectrum, density of material, genre and style. Even using a proper reference is only a "first step" or "sanity check". Ultimately you need to do what is best for the individual song. You can only determine this from experience in using the tools and listening, as well as the goals as specified by the client.

That said, pink noise is used to setup and measure monitoring systems because it is a consistently measurable source. That's science, not music.
 
producerkid said:
the white noise scenario is not intended to be an overall goal for volume, but a reference point to make sure im NOT squashing the everlasting shit out of what im working on (at least as much as i can help it) while still keeping the volume levels up.
Click to expand...
Producerkid, also no offense, but there are *many* things erroneous just in that single quote. I'd like to explain:

Even if - IF - a noise source had any relation whatsoever to the loudness of a mix - which it doesn't, as Tom pointed out - white would not be the color of noise to use. You should know by now that white noise is the even and random distribution of energy across the spectrum. As such it is totally unrealistic in appliction to audio for three reasons:

a) the amount of energy required to drive a certain frequency to a certain intrinsic volume level varies over the frequency range; in general the lower the frequency, the more energy that is required. You see an effect of this every day when you know that the mo' betta' the bass you want the mo' bigga' the amplifier you need. This means that white noise is going to be unrealistically biased to the high frequencies as compared to a realistic audio mix, and cannot serve as a realistic reference.

b) in the past 11 years surely you have come across the Fletcher-Munson curves, which illustrates how the sensitivity of the human ear also varies by frequency. Just like with the relation of frequency to energy, the ear's realtion of frequency to sensitivity makes a flat line reference like white noise irrelevant to reality. This is why engineers have come up with things like pink noise, A-weighting, etc., to at least roughly shape response curves away from the flatness of white noise and closer to reality.

c) this is the big one: noise, whether white or pink or gray or green, is going to have a crest factor approaching 0dB. This means virtually *no* dynamics whatsoever. You don't need that as a reference; just push your mix all the way into a brick wall limiter with infinite gain reduction so that the silent parts of your mix are as loud as the peaks, a < 0dBRMS flat line at 0dBFS, and you have more or less the same result. Tell me, do you really need that as a reference to let you know that you've gone to far?

Use your ears, 'kiddo. If it sounds squashed, pull back. If it doesn't, you're OK. If you don't know because you're ears aren't there, you're hosed.

G.
 
Last edited:
Glen, you have insulted my inteligence. Thank you.

No seriously....

That was the information i was looking for and i think it just became apparent to me that i am getting to the point where I am hearing more than i know im hearing. This is where my thoughts about going to audio school come into play. I would probably have the answers to alot of my questions if i knew more about the technical theory.

I still think a loudness reference would be useful as a general check, but you are right, white noise would be a bit unrealistic. I just figured a flatline 0db signal of some kind would be sufficient enough to let the room characterisitcs be more obviously apparent but still be a good example of ultimate volume. Running a given mix full on into a limiter sounds way more accurate in determining errors with mix processing than any kind of noise.

I think like a fucking robot lol, so the science behind the beauty is where my head is most of the time. However, im missing a lot of crucial information which is limiting me now (no pun intended). I've definately got my ears because thats how i've managed to even get this far. I just need more GEAAAARRRR. :rolleyes:

I need to listen more, you're right. End of story lol. I get so boxed in mixing in headphones all the time because my room is insufficient, and esepecially recently since i just hung my monitors and havent gotten the drivers back in or the back wall absorber built yet.

Thanks for the reality check guys. I'm still gonna try the volume reference thing though and see if it even makes a damn.
 
producerkid said:
i think it just became apparent to me that i am getting to the point where I am hearing more than i know im hearing.
Click to expand...
That is a very keen and profound observation. Therefore, I love it. :D

Actually that is a very good point on a couple of levels. One can read that as a whole 'nother way of realizing the idea of "trust you ears"; trust them because they may know more than one may think they do. One can also read that as pointing out the difference between "hearing" and "listening", which is the doorway to the difference between entertainment listening and critical listening, the later being necessary to discover the engineering behind the success or failure of the former. Very nice.
producerkid said:
This is where my thoughts about going to audio school come into play. I would probably have the answers to a lot of my questions if i knew more about the technical theory.
...
I just figured a flatline 0db signal of some kind would be sufficient enough
...
I think like a fucking robot lol, so the science behind the beauty is where my head is most of the time.
Click to expand...
There's nothing wrong with having your head in the science. I can tell you understand that this is one of those rare (a)vocations that is just as much left brain as it is right brain, much like the creation of music itself. The science is the underlying infrastructure on which the beauty is built; so an understanding of that is indeed important.

And, with that in mind, I might suggest that what might help in your approach and understating of the whole mastering levels thing is to look at some of the science and art on managing levels all along the way on the path who's final step is the mastering level.

As we all agreed early on, the better the tracking and the better the mix, the better and further one may potentially push the final levels before they start sonicly falling apart. IMHO, and in the HO of many other of the more seasoned folks around here, one of the key principles in this is making sure one's levels are optimal throughout the production process. One of the ironies that comes out of an understanding of this process is that more conservative gain levels early on in the process more often than not allow for louder potential levels in push mastering; OTOH, pushing the levels early on in tracking and or mixing can actually cripple one's potential to push the final mastering levels as loud as one would like.

Why this comes to my mind now is your inclination to reference to 0dB. Actually, on the analog side of the process, where the dB scale we're using is dBVU, this is a pretty good basic principle to start from. It's the referencing on the digital side to 0dBFS where things get a bit off the mark. The understanding that "0dB" on the analog side and "0dB" on the digital side are two different animals with two different meanings, and the further understanding of how the two actually relate and matter to the whole process can go a long way to getting the most out of the last step of mastering.

You may already know about this principle by it's common name of "gain structure" or "gain staging". If this is old news to you, my apologies for sounding condescending; I'm still feeling you out as to where you are with all this stuff.

But if this sounds like something new or interesting to you, it's something I'd definitely recommend schooling on a bit more. And by some strange freak of coincidence :-)rolleyes:), I happen to have something on my website that you might find interesting. Head on over to www.independentrecording.net and in the right column for Online Resources, click on the "Metering and Gain Structure" logo for an on-line applet talking about just this subject.

G.
 
If you're really into top end quality why don't you get something better in addition to the myspace sites. Soundclick is better quality though still compromised even for streaming & lightning MP3 will allow you to demonstrate your wares with much higher quality MP3s.
I've been h/recording for longer than you've been alive but I know nothing & come here to learn.
As for mastering - the word MASTER is very indicative of the process & processor. Once you've mastered listening, then mastered your tools & finally mastered the art of creating an aural image to work toward you'll be on the way to being a master of the task.
Me - I don't master. I do work hard on a mix, I do listen to mixes on a variety of reference set ups and sometimes I PLAY with mastering plugs. I do know, however, that my work is done when the mix is as good as I can get it.
One day I'll send a couple of tracks to be done by someone I've listened to & liked the results of. 1st I'll have to do something good enough as well as save up.
 
You must log in or register to reply here.
Back
Top