|
.
|
12.12.06 - I was working on a Separations
project and things were going well, just a few tweaks were in order.
The client mentioned that toward the end of their song, the "volume dropped."
This was on a loud rock CD that was totally pedal to the metal!
Their dilemma: They had added in more
stuff to their tracks at that particular point which then saturated the
overall signal! More sound on their tracks
displaced something
else... the famous "Teeter Totter Principle" at work! Creating insanely
loud CD volume comes with a price - you have to limit the amount of stuff
that you pile on top of other stuff - because the waveform changes as it
gets louder. Here's a huge message
that all the LOUD CD fans can consider....
JV's email said
to the client: ...you have drums, guitars, bass, synth and vocals
going on... and then at the place where your master gets "softer," look
at your tracks.... the cymbals increase,
the guitars increase, the synths
increase, the lead vocal is
louder than anywhere else in the
song, and the backup vocal comes in louder
than anywhere else in the mix...
I have to tell you that in this world of loud digital music. we are at a
point of NO HEADROOM WHATSOEVER. No matter what, there are NO BONUS dbs
after 0dbfs. What that means
is that if you add something to the ALREADY-MAXED sound, something has to
give. The SUM TOTAL
of the waveform will be what the COMPONENTS are that you put together.
(Tip: Don't think of the little red "over" lights the way you would think
of the red area on a VU meter... it's not even close to being the same...
VU meters are missing
11db above the +3 mark....)
Back to your song.... If I mute the vocals and backups, the volume stays
fine. If I mute the synths, the volume is fine. If I mute the
guitars, it's fine. With EVERYTHING added in at once, the energy
of the increased instruments simply pushes down (it's the teeter-totter
principle) the bass and the full bodied stuff - and it just sounds like
the mass decreases in volume!
Note! Anything
music that sounds great on an exceptionally loud CD has been
musically
arranged to accommodate
a saturated waveform while still leaving some trace of musical muscle in
tact. If something is loud, it is because
space has been created
behind it to allow it to come forward and be heard. To keep this all
working on your song, I must decrease something in order to accommodate
your more-of-everything musical arrangement. The saying "Less is more."
means less stuff piled
on top of everything will allow for
more sound to come through.
I will try cutting off the low end of the backup vocals at that point of
your song. Low frequencies have more energy, so if I reduce some lows,
it will be a thinner additive that
comes in on top of the already-maxed sound. If this project was not
using Separations, you would have to set the level of the entire song based
on this last portion of the music... make that right... and then live with
the rest of the song at a lower volume... or you would have to figure this
phenomena out and remix this portion of the song.
The major lesson:
at loud levels, only so much sound will come through and retain it's punch.
Add more and more instruments and fill up more sound space, and the punch
gives way to longer more sustained
and even (square?) waveforms. Punch comes from something being louder
than something else that
is softer. If NOTHING is softer, then the result is smooth sustain,
the oposite of wide excursions of varying sound wave levels (and therefore
corresponding varying speaker movements).
This becomes frustrating when you make your mix - and it sounds fine. But
then it goes to mastering and I'm adding all this gain in the "no bonus
room" department. Try adding 15db of level to your mixes and
see how they react! Mastering didn't used to be a place where the
mixes got squashed into whipped butter! Big-label mixing engineers
often understand just how much stuff must be removed at mix time in order
to allow for that big sound to stay big. Quincy Jones, I think, once
said that a mix was finished when there was no more stuff that could be
taken out of the mix without losing the flavor. Sparse arrangements
= loud punchy music. Take rap for example. Often, it's sparse
instrumentation mixed back so that the drums are huge by comparison.
It's all about what's RELATIVE to something else.
<>
Given all that.... some loud CDs can sound very good, but over a long period
of time, squashed waveforms lose their natural breath - which is part of
what keeps music interesting.
| Why the big level
contest anyway?
Record companies know that consumers listen to music in
the OPPOSITE WAY that musicians play it. Musicians play music at
a normal volume level, and when they want it to be more exciting,
they punch it louder.
The home listener, on the other hand, turns their stereo to where
it's at a comfortable (or desired) listening level. Then if they
hear something softer, they perceive it as being weaker.
Since record companies know about those carousel CD changers with
5 CDs on it, and they know about the shuffle button (that
randomly plays different songs from all the various CDs loaded in
the carousel) - and they sure as heck don't want their artist's
CD to sound weaker than anybody else's CD ...'cause it wouldn't
sound as exciting as the other guy's record....
|
|
|
Here's the scoop. Some artists are flattening
out their sound.... rendering it punchless and distorted... with no low
end muscle, soggy mid-highs and flat high end. They're using tons of plug-ins
and compressors (more info on compression here)
to approach what mastering does to get those CDs hot.
The rub....
as the CDs of 2007 get even hotter than 2006 and 2005, you almost
have to slam and compress your mix...
because the mix friggin' changes
as you limit those pesky dynamics!!!!! How else will you know if you're
putting enough kick in the mix...? or enough clarity on the vocals (cause
the instruments just keep a comin' up as the mix gets squashed....)?
How do you know what to do to PREDICT how to mix for a hotter final CD???
Rule: In order to cut a slammin' hot
CD, you must smoothly limit the peaks in order to not create clipping which
can sound harsh and fatiguing in time. Yet, the whole effect of a
musical peak (like say on a kick drum) is that the waveform surges out
over the music. The peak
sound has a cool impact that stands up from the other sounds. This is punch
that comes from a wider speaker excursion caused
by that bigger peak, which means the speaker
is actually moving more air... farther - therefore... more punch.
When you flatten out all these peaks by over-compressing or hard limiting
and bring up the softer material that is all around those peaks, the speaker
excursion (distance it moves) is relatively smaller (in relation
to the softer parts) - thereby reducing the punch of the music. Granted,
the overall sound is louder coming off the CDR, but the overall sound is
flatter and less open sounding. Everything is more in-your-face all
the time.
High-tech hint: There are different
kinds of loudness. One that is sometimes overlooked is called "apparent
loudness" which is a frequency-dependent kind of level. This is a phenomena
whereby the meter level doesn't increase but the apparent volume does. Use
your ears, and pay close attention to the arrangements of top-drawer commercial
CDs. Less is more!
Q) John, can you reduce the guitars on the sides of my mixes and bring up
the drums in the center for more punch? -Jeff
This question was submitted before we started doing
Separation Mastering which solves issues like
this. However if someone is using a traditional stereo mix, we can
reduce the width of the side guitars, but as the drums in the center come
up, the vocal and other elements in the center will come up as well.
Taking the guitars down will "teeter-totter" everything up in the center
of a 2-track mix. A subtle change may do the job - it's all about
preferences and references.
Before Separations, we would say "we can't make an apple into an orange.
we can, however, shine up the apple." Using Separations, we can in
fact actually create an "appleorange." In all cases, creating the
final balance is always about the teeter-totter effect. If we tilt
one thing back, something else comes forward. The same is true with
the arrangement of the instruments - add more and more instruments and it's
harder to hear the drums. The teeter-totter effect is the same for
actual frequencies - you have to carve a "frequency hole" for other elements
to come through and be heard. It's important to know how mastering
affects mixes, so experiment with pre-squashing your mixes to see what needs
to be done to keep the music punchy. </>
Problem: How do you keep your mixes hot and punchy without owning a high-end
mastering rig?
Solution: Create
Separations. No more alternate mixes,
no more late night debates - get your mix to sound great and create separated
musical elements -- and in mastering, it all comes together. If you
have 14 songs and #8 needs the vocal brought up a bit, simply open the non-destructive
file format and bring up the vocal. If song #10 needs more drums...
go for it. If you prefer the stereo mix, it's all there in the format
- nothing is sacrificed.
Most important: Save time by doing homework,
and get the mix correct. The best way to do that is to A/B your mixes with
commercial CD productions on a level-matched system. Leave
headroom in your mixes (2dBFS is good) and build the excitement into the
individual tracks for a musically vibrant, punchy sound. Then let
the mastering engineer send it through the roof for ya. With $100,000 worth
of eq's, compressors & such in the processing chain, the chances are that
you'll end up with just as much level with more punch and musical impact...
and a CD that is more listenable over a longer period of time... because
it will breathe more naturally than a total squashed-to-the-max stereo mixdown.
And yes, the mastering process might cost a bit more than that cool $150
SuperGodSmasher Plugin for your PC... add in the flexibility of the
Separations format, and so long as you've made good sonic/recording/arranging/production
decisions during your project, you'll have a competitive final sound.
Put more energy into your mixdown "performance".
It's more exciting and more magical, and it's worth it. For this reason
I don't recommend mixing through a Finalizer unless you are very conservative
with the settings. It makes it too easy for everything to sound good. I've
mixed this way, and I immediately knew what the problem was once I got my
own mix into the mastering room.
Hey, this is a no-rules business, and certainly everything you've done up
to this point may be exactly what the doctor ordered for your project to
hit the top ten - exactly as it is.... so use your best creative judgment,
have fun, and DON'T get addicted to the technology!!! Keep your music first
and use top-notch talent every step of the way - at a price that's appropriate
for you. Get people involved, promote relentlessly, and have a good time
along the way.
Remember the rules:
1. There are no rules.
2. If you can't make Separations, make alternate mixes so you aren't boxed
into a creative corner.
3. Label your masters. Clearly. Id numbers, song names, take notes, etc.
4. Don't send 2nd generation audio CDRs in for mastering - use original
24 bit files!
5. Leave a second of "air" before your song starts on data files. Put at
least 30 seconds of dead air on a DAT before rolling tape. Same thing on
analog tape (and be sure you print alignment tones on analog tape).
|
"I have been to "world famous" mastering
facilities and John's abilities and ears are right in that league.
As long as I make CD's, I will have John Vestman master my work."
-Andy Roth, God Help Me! Music
|
|
