IntroductionWelcome| 00:04 | Hi! I am Brian Lee White and
welcome to Foundations of Audio:
| | 00:07 | Compression and Dynamic Processing.
| | 00:10 | In this course, we'll look at the most
common dynamics processors, including
| | 00:14 | compressors, limiters, de-essers,
expander gates, and transient shapers.
| | 00:22 | I will start by covering the
fundamentals of sound waves and amplitude. Then I
| | 00:26 | will break down the common controls of
the compressor and show you how to add
| | 00:30 | punch and sustain to tracks in your mix.
| | 00:33 | (music playing)
| | 00:40 | I will demonstrate how to maximize mix
loudness using brickwall limiters, use a
| | 00:45 | de-esser to de-ess a vocal track, and
apply gates and expanders to eliminate
| | 00:50 | unwanted noise and bleed on your tracks.
| | 00:53 | I will then explain how to best utilize
multi-band compressors and use transient
| | 00:58 | shapers to scope the attack
and decay of individual notes.
| | 01:01 | (music playing)
| | 01:08 | I will cover techniques
including parallel compression, ducking
| | 01:12 | compression, side-chain setups,
mixed-bus compression. and using software
| | 01:19 | models of vintage hardware.
| | 01:21 | Lastly, I will discuss best practices
for when and how to use compression and
| | 01:25 | limiting in your mixes.
| | 01:28 | Throughout the course I will also
provide you with guided exercise content in
| | 01:31 | the form of Get in the Mix
demonstration sessions that you can open up in your
| | 01:35 | own digital audio workstation.
| | 01:38 | Watch the video about the Get in the Mix
content to learn more about this unique
| | 01:41 | learning experience.
| | 01:43 | Now, let's get started
with Foundations of Audio:
| | 01:46 | Compression and Dynamic Processing.
| | Collapse this transcript |
| What you should know before watching this course| 00:00 | In this course, we'll be covering
many basic and advanced topics about
| | 00:04 | compression and dynamic processing.
| | 00:07 | While I don't presume that you have
any knowledge of these topics, I do
| | 00:10 | recommend having some basic working
knowledge of a digital audio workstation.
| | 00:14 | If you need a refresher on the basics
of digital audio recording, mixing, and
| | 00:18 | signal flow, you may want to check out
the essential training course appropriate
| | 00:22 | for your digital audio workstation in
the lynda.com Online Training Library;
| | 00:28 | otherwise, if you're ready, I am ready.
| | 00:30 | Let's get started.
| | Collapse this transcript |
| Using the exercise files| 00:00 | If you're a Premium member of the
lynda.com Online Training Library or if you're
| | 00:04 | watching this tutorial on a DVD-ROM,
you also have access to the raw audio
| | 00:09 | material used to create the exercise
content, as well as all other audio
| | 00:12 | examples featured throughout the course.
| | 00:15 | Inside the Exercise Files folder,
you'll find the folder for each chapter,
| | 00:19 | containing the WAV files
used throughout the course.
| | 00:21 | These files can be imported into your
own digital audio workstation and used to
| | 00:26 | follow along with the material.
| | 00:28 | If you're a Monthly member or Annual
member of lynda.com, you don't have
| | 00:32 | access to the raw audio files, but you can
follow along from scratch with your own assets.
| | 00:37 | For members at all subscription
levels, I have provided you with guided
| | 00:40 | exercise content in the form of a Get in
the Mix demonstration session that you
| | 00:44 | can open up in your own DAW.
| | 00:46 | Watch the video about the Get in the Mix
content to learn more about this unique
| | 00:50 | learning experience.
| | 00:52 | Let's get started.
| | Collapse this transcript |
| Using the "Get in the Mix" Pro Tools and Logic Pro session files| 00:00 | This course features Get in the Mix
exercise content, living sessions with
| | 00:05 | built-in demonstrations and practice
material for you to use with your own
| | 00:09 | digital audio workstation, or DAW.
| | 00:12 | Using your DAW's video track
capabilities, I'll guide you through automated
| | 00:16 | audio examples, demonstrating a
number of the concepts and techniques
| | 00:20 | discussed in this course.
| | 00:22 | All you need to do is press play.
| | 00:24 | Since the files are actual native, high-
fidelity project files purpose-built for
| | 00:28 | your specific DAW, you can
manipulate the audio examples yourself.
| | 00:32 | So feel free to pause, rewind, repeat,
and zoom in on sections during the
| | 00:36 | demonstrations to solidify your knowledge.
| | 00:39 | Get In The Mix project files also
feature additional practice tracks, so you can
| | 00:44 | explore the techniques you
just learned on your own.
| | 00:47 | These tracks are labeled Practice, and
their content is located at the end of the
| | 00:51 | demonstration material.
| | 00:53 | Before using Get In The Mix content,
you must first download the package
| | 00:57 | prepared for your specific DAW from
this course's page in the lynda.com
| | 01:01 | Online Training Library.
| | 01:03 | Inside this package, you will
find the Get In The Mix files.
| | 01:07 | Throughout the course, I will direct
you to open these files when appropriate.
| | 01:11 | If you have this course on DVD, the Get
In The Mix files are included on the DVD.
| | 01:17 | Unlike premium exercise content, Get
In The Mix content is available to all
| | 01:21 | lynda.com subscribers.
| | 01:24 | If you are a Premium subscriber, you
also have access to the raw audio material
| | 01:28 | used to create the exercise content, as
well as all other audio examples featured
| | 01:33 | throughout the course.
| | 01:34 | So download that content package and
get in the mix with Foundations of Audio:
| | 01:38 | Compression and Dynamic Processing.
| | Collapse this transcript |
|
|
1. Audio DynamicsWhat is amplitude?| 00:00 | Before getting into the nuts and
bolts of dynamics processors such as
| | 00:04 | compressors, limiters, and gates, it's
important to take a step back and examine
| | 00:08 | some of the basics of sound.
| | 00:11 | Sound is made of waves of energy that
oscillate back and forth through a medium.
| | 00:15 | Usually the medium is air, but
sound can also travel through solids, like
| | 00:19 | drywall or liquids, like water.
| | 00:22 | Soundwaves create an invisible push and
pull of the air particles around us and
| | 00:26 | our ears perceive and translate these waves
into nerve impulses that are sent to our brain.
| | 00:32 | For example, when you play music from
your speakers, the speaker cone moves in
| | 00:36 | and out, creating changes in the
pressure of the surrounding air. The
| | 00:41 | resulting soundwaves are picked up by
our ears and our brain translates them
| | 00:44 | into sound information.
| | 00:47 | Soundwaves are generally
measured across two dimensions:
| | 00:50 | frequency and amplitude.
| | 00:54 | Frequency is the oscillation speed of the wave
of the rate of push and pull of air particles.
| | 00:59 | Higher frequencies produce higher-
pitched sounds, while lower frequencies
| | 01:03 | create lower-pitched sounds.
| | 01:04 | While frequency and amplitude go hand
in hand, in this course, we'll focus on
| | 01:09 | measuring and reacting to
changing a waveform's amplitude.
| | 01:12 | So let's dive a little
deeper into what amplitude is.
| | 01:16 | When measuring the amplitude of a
soundwave, we chart the changes in
| | 01:19 | atmospheric pressure.
| | 01:21 | When particles of air are packed
together tightly, indicating higher pressure, we
| | 01:26 | chart this push, or positive value on the
graph, and it's called compression.
| | 01:30 | Pulls are negative values in the graph
where air particles are more spread out,
| | 01:34 | are called rarefactions.
| | 01:37 | The height of these compressions
and rarefactions in the graph indicate
| | 01:40 | the amplitude, which is directly
proportional to the loudness of the sound perceived;
| | 01:45 | in other words, the greater the
amplitude of a soundwave, the louder we will
| | 01:49 | experience the sound.
| | Collapse this transcript |
| Measuring amplitude| 00:00 | Our ears and brains help us sense that
louder sounds have higher amplitudes and
| | 00:04 | softer sounds have lower amplitudes;
| | 00:07 | however, to put that information to
better use in a musical application, we need
| | 00:12 | to be able to measure a sound
wave's amplitude more exactly.
| | 00:16 | Then we can create rules of what to
do when a specific amplitude level is
| | 00:20 | reached, and control the range of a
soundwave's amplitudes to our advantage.
| | 00:25 | This measurement process forms the
basis of all dynamics processors.
| | 00:29 | If we can measure it, we
can start controlling it.
| | 00:33 | There are many ways to measure amplitude.
| | 00:34 | Inside our digital audio workstation,
we will use dBFS, or Decibels Full Scale.
| | 00:40 | This scale defines our system's
maximum and minimum amplitude values,
| | 00:44 | otherwise known as dynamic range.
| | 00:47 | This helps us measure and control the relative
amplitude of audio signals within that range.
| | 00:53 | Sometimes the dBFS can confuse people
because it counts up from negative
| | 00:57 | numbers and ends at 0.
| | 00:59 | Why does it do this?
| | 01:00 | In a digital audio system, values
above 0 are simply cut off, or clipped,
| | 01:04 | creating digital distortion; thus
there is no higher value than 0.
| | 01:09 | The decibel scale is a
logarithmic measurement scale.
| | 01:13 | Turning up an audio signal's volume
by 1dB is basically imperceptible.
| | 01:17 | For most of us, it takes at least
a 3dB increase for us to notice it. However,
| | 01:22 | because of its logarithmic scaling,
the loudness escalates quickly.
| | 01:26 | A 10dB increase represents a perceived
doubling of loudness, and an increase of
| | 01:31 | 20 dB is about four times as loud.
| | 01:34 | These numbers are helpful
to know when mixing a song.
| | 01:37 | Use them as a guide, but let
your ears be the ultimate judge.
| | 01:41 | The listener doesn't care about the
dBFS level of your snare drum, just how it
| | 01:46 | feels when mixed with the rest
of the instruments in your song.
| | 01:49 | Measuring amplitude using the dBFS is
the first step in helping us control an
| | 01:53 | audio signal's dynamic range.
| | Collapse this transcript |
| What is dynamic range?| 00:00 | Dynamic range is the
difference between two extremes.
| | 00:03 | In audio, dynamic range represents the
ratio between the loudest and softest
| | 00:08 | signal being produced over any given interval.
| | 00:11 | Every day we experience this
concept of dynamic range almost constantly
| | 00:15 | across all of our senses.
| | 00:17 | And as humans, we're
quite adept at recognizing it:
| | 00:21 | light and dark, hot and cold, hard
and soft, and for the purpose of this
| | 00:25 | course, loud and quiet.
| | 00:28 | A sound's amplitude directly
correlates with its loudness.
| | 00:31 | But what is loud? What is quiet?
| | 00:34 | Is a concert loud? Is a
library quiet? That depends.
| | 00:39 | But I can assure you, if you walked
out of a library and into a rock concert,
| | 00:43 | you would have a pretty good idea of the
difference in loudness between the two spaces.
| | 00:47 | Loudness in our minds is not finite.
| | 00:50 | It's all about perception.
| | 00:52 | The key thing to understand is that we
need both ends of the spectrum to fully
| | 00:56 | evaluate what's loud and what's quiet.
| | 01:00 | The distance between these
extremes is what we call dynamic range.
| | 01:03 | For example, if you heard a book fall off
a desk in a library, it would startle you.
| | 01:09 | This is because the difference
between the sound of the book falling in a
| | 01:12 | library versus the library's
quiet environment is fairly extreme.
| | 01:17 | This same effect is used
by movie sound designers.
| | 01:20 | Ever jump out of your
seat from a loud explosion?
| | 01:23 | At its limits, the human ear can
experience roughly 120 dBs of dynamic range,
| | 01:28 | with 0 dB being imperceptible and
120 reaching the threshold of pain,
| | 01:33 | although much of our day-to-day
experience of dynamic range exists in a
| | 01:37 | much narrower space.
| | 01:39 | For example, a modern pop mix may
only have around 8 to 10 dB of dynamic range
| | 01:44 | between the peak and average
amplitude values over the entire song,
| | 01:48 | while a modern film may have around 20
dBs of peak-to-average dynamic range,
| | 01:53 | which is why you would be more
likely to jump out of your seat at a movie
| | 01:56 | theater during a loud explosion than you
would be listening to a song in your car.
| | 02:00 | Let's listen to some examples.
| | 02:03 | This first example has a wide dynamic range.
| | 02:07 | Imagine a quiet park
interrupted by car horns and sirens.
| | 02:10 | (ambient sound)
| | 02:18 | (car honking)
(sirens)
| | 02:33 | Now, here's an example of reduced
dynamic range, a fully mastered pop mix.
| | 02:39 | (music playing)
| | 02:58 | Unlike the quiet park scene that's
interrupted by loud horns and sirens, the
| | 03:01 | pop mix maintains the same
perceived loudness, thus exhibiting a much
| | 03:05 | narrower dynamic range.
| | 03:08 | As a side note, 16-bit recording
systems offer 96 dBs of dynamic range.
| | 03:13 | In comparison, 24-bit recording offers
a much wider range of 144 dBs, exceeding
| | 03:19 | the range of human perception, which is
generally accepted to be around 120 dBs.
| | 03:25 | Because of this, 24-bit recordings can
accommodate a wider range of amplitude
| | 03:30 | values before digital clipping.
| | 03:33 | Now that we understand the basic
concept of dynamic range, we can begin to
| | 03:37 | harness the power of
dynamics processors to control it.
| | 03:40 | Just as we can measure the change in
perceived loudness of quiet dialogue in a
| | 03:43 | film to a loud bomb-explosion sound effect,
| | 03:46 | we can also measure the dynamics over
shorter periods of time, like the volume
| | 03:51 | differences in phrases of a vocal track
from line to line or the amplitude curve
| | 03:55 | of the crack of a single snare hit.
| | 03:58 | And since we can measure these dynamic
range relationships, we can change them
| | 04:01 | if they aren't working for us.
| | 04:03 | We'll do just that using dynamics processors.
| | Collapse this transcript |
| What are dynamics processors?| 00:00 | In its simplest form, a dynamics
processor is like an automatic volume control,
| | 00:05 | turning the volume up when it's
too soft or down when it's too loud.
| | 00:10 | In the audio world, dynamics
processors adjust the dynamic range of an audio
| | 00:14 | signal by measuring a signal's
amplitude over time, and setting up rules to
| | 00:19 | react to any changes to that amplitude.
| | 00:21 | The dynamics processors we will cover
in this course are compressors, limiters,
| | 00:28 | expander gates, de-essers, and transient shapers.
| | 00:37 | Dynamics processors allow us to
manipulate a signal's amplitude and dynamic
| | 00:41 | range in a variety of different ways:
sometimes reducing it, sometimes
| | 00:46 | increasing it, and often
doing both over a period of time.
| | 00:50 | Before adjusting the amplitude, we
must first specify the amplitude level at
| | 00:55 | which the processor begins to react.
| | 00:57 | This is called the threshold.
| | 00:59 | Then we need to create a rule for what
happens when that threshold is breached.
| | 01:04 | For example, in a compressor, when the
amplitude level of a signal crosses the
| | 01:09 | threshold level, the signal
above the threshold is compressed.
| | 01:15 | All dynamics processors will work on
this basic idea of an action, generally a
| | 01:19 | specified signal level, creating a reaction,
some form of dynamic change to the signal.
| | 01:26 | All you are really doing is telling
the processor what to look for and what
| | 01:29 | to do when it happens.
| | 01:31 | Here is a simple example that
almost everyone can relate to.
| | 01:35 | Let's say you are watching your
favorite TV show and you have got the volume on
| | 01:38 | your surround sound system all
dialed in so you can hear your favorite
| | 01:42 | actor's every word.
| | 01:44 | Suddenly the commercials come on and you
are blown out of your seat by the loudness.
| | 01:48 | Of course, you instinctively reach for
your remote control to turn the volume
| | 01:52 | down, as to not wake up the entire neighborhood.
| | 01:55 | As soon as you find that perfect
level on your remote for the commercial
| | 01:59 | break, inevitably your favorite
program comes back on and you can't hear
| | 02:03 | anything the actors are saying.
| | 02:05 | Again, you instinctively reach for your
remote to turn the volume back up to the
| | 02:09 | pre-commercial level.
| | 02:12 | In this common scenario, even if you
didn't realize it, you are acting as a
| | 02:16 | dynamics processor, a compressor to be specific.
| | 02:20 | When that commercial break comes on,
the relative difference between the level
| | 02:23 | of your show and those loud commercials
triggers a threshold in your brain that
| | 02:27 | says "way too loud!" and you reach for
your remote and turn it down, effectively
| | 02:33 | compressing the dynamic range of the TV
show relative to the louder commercials.
| | 02:38 | Once the commercials end and the
program comes back on, a threshold in your
| | 02:42 | brain is triggered again saying "I
can't hear what they just said!" and you
| | 02:47 | return the volume back to its previous level.
| | 02:50 | Now think about the TV example I just
described and try to relate that to a
| | 02:54 | real-world mixing dilemma.
| | 02:56 | How about the words in a vocal track
rising and falling above the level of
| | 03:00 | the background music?
| | 03:01 | We want to hear all those lyrics
clearly, but we also don't want them to jump
| | 03:06 | out of the mix too much.
| | 03:08 | Dynamics processors to the rescue.
| | 03:10 | A dynamics processor on your mixing
console, in your rack of outboard gear, or in
| | 03:15 | your DAW does the same exact thing
that you do while you channel-surf on your
| | 03:18 | couch, except that it's much faster
and has a higher degree of accuracy.
| | 03:23 | Once you get the hang of the basic
concepts behind dynamic processing, these
| | 03:27 | processors will quickly become some of
the most useful tools in your studio,
| | 03:31 | empowering you to create a
tight, punchy, and focused mix.
| | Collapse this transcript |
| Hardware and software dynamics processors| 00:00 | It seems like a year never passes
without at least a dozen new compressors,
| | 00:04 | limiters, and other similar
tools coming onto the market.
| | 00:08 | It's no surprise that producers and
engineers often become quickly overwhelmed
| | 00:12 | with the number of choices.
| | 00:14 | Which one should I buy?
| | 00:15 | Do I need all of them to make my mix sound good?
| | 00:18 | Will this year's model change my life
like the advertisement says it will?
| | 00:22 | Even though all the flashy interfaces,
lights, and knobs may look radically
| | 00:26 | different from each other, the reality
is that all dynamics processors today are
| | 00:31 | pretty much designed on the same
principles as the ones from a few decades ago,
| | 00:36 | and learning the fundamental concepts
behind them and the basic techniques for
| | 00:40 | using them in your mixes translates
surprisingly well across both hardware and
| | 00:44 | software processors.
| | 00:47 | Think about it this way.
| | 00:48 | If you can drive a Toyota, you can
certainly drive a Ford or a Chevy.
| | 00:51 | It might take a few minutes to get
used to the controls and you might not
| | 00:56 | understand the full feature set the car offers,
| | 00:58 | but once you know how to drive, you
can pretty much drive anything and get
| | 01:03 | from point A to point B.
| | 01:05 | Now some cars are flashy and faster around
the corners, while others are designed to
| | 01:09 | be smooth and elegant.
| | 01:10 | Dynamics processors are no different.
Some sound best on certain instruments,
| | 01:15 | like drums or vocals.
| | 01:17 | Some color the signal, adding warmth
and punch, while others don't color the
| | 01:21 | signal and are considered transparent-sounding.
| | 01:25 | Dynamics processors can exist as
plug-in-based software programs that run
| | 01:29 | inside your DAW--otherwise known as in
the box--or as outboard processors that
| | 01:34 | are built into the channel strip of a console,
or exist as separate hardware pieces in a rack.
| | 01:40 | In this course, we will take
advantage of some of the most popular factory-
| | 01:43 | bundled DAW plug-ins included with
Pro Tools and Logic, as well as the
| | 01:48 | industry-standard Waves plug-ins, a
popular third-party choice that works in
| | 01:53 | almost any DAW. as well as many digital consoles.
| | 01:55 | We will also take a moment to look at
using analog hardware-based dynamics
| | 02:00 | processors in your workflow.
| | 02:03 | Although we will look at a number of
fantastic tools throughout this course,
| | 02:07 | everything I will be showing you
will easily translate to the dynamics
| | 02:10 | processors you have access to.
| | 02:12 | Let's get started!
| | Collapse this transcript |
|
|
2. CompressorsIntroducing compressors| 00:00 | Many instruments tend to be very dynamic
and when recorded, can result in a wide
| | 00:05 | range of note amplitudes, with
some very loud and some very soft.
| | 00:10 | It could be difficult to balance these
very dynamic tracks in a mix and still
| | 00:14 | hear every nuance of the performance.
| | 00:16 | Compressors can be used to restrict the
dynamic range of an audio signal, making
| | 00:21 | it easier to find a more
steady level that works in your mix.
| | 00:25 | Let's listen to a simple example.
| | 00:26 | (music playing)
| | 00:36 | The vocal track is very dynamic and
parts of it are getting lost behind the
| | 00:40 | music, while others stick out too much.
| | 00:43 | Now I will apply compression.
| | 00:44 | (music playing)
| | 00:55 | As you can see, and hopefully hear, by
applying compression, I was able to reduce
| | 01:00 | the dynamic range of the vocal phrase
to help it sit better in the mix and be
| | 01:04 | lyrically intelligible at all times.
| | 01:07 | In the original uncompressed vocal line,
the difference between the softest and
| | 01:11 | loudest word was about 15 dB on average.
| | 01:14 | After applying compression, the
dynamic range is only 5 dB on average.
| | 01:20 | What's happening here is the
compressor detects and grabs hold of the louder
| | 01:23 | words in the phrase and turns
them down, or compresses them.
| | 01:27 | After compressing these louder words,
I can then raise up the entire level of
| | 01:31 | the vocal track, so that both the soft
and loud notes sit comfortably within
| | 01:35 | the rest of the mix.
| | 01:36 | While I will cover many compression
techniques throughout this course,
| | 01:41 | this technique is one of the
most common uses of compression.
| | Collapse this transcript |
| Understanding threshold| 00:00 | Compressors, like all dynamics
processors, work by measuring the incoming
| | 00:04 | signal's amplitude against a user-
defined reaction point, where the
| | 00:08 | compressor will begin to work.
| | 00:10 | This reaction point is known
as the compressor's threshold.
| | 00:14 | Think back to our example of
watching TV and turning down the volume at
| | 00:17 | a commercial break.
| | 00:19 | We all have a unique threshold of how
loud is too loud and what will force us
| | 00:22 | to grab the remote and turn down the volume.
| | 00:25 | That volume level is our threshold, and
is actually the most important component
| | 00:29 | of any dynamic processor.
| | 00:31 | In a compressor, a signal level above
the threshold will cause the compressor to
| | 00:35 | react, while a signal level below
the threshold is left unaffected.
| | 00:40 | The threshold's value is generally
measured in the dBFS to match our DAW's metering;
| | 00:46 | therefore, a value of 0 dBFS means
that threshold is sitting at the digital
| | 00:50 | clipping point and can go no higher.
| | 00:51 | The Threshold level goes down as you
dial it deeper into the negative numbers.
| | 00:56 | For example, a threshold setting of -20
dBFS would cause the compressor to react
| | 01:01 | to any signal whose amplitude is
measured at over -20 dBFS up through 0 dBFS.
| | 01:08 | Therefore, a signal of -10 dBFS could
trigger a reaction in the compressor,
| | 01:12 | while a signal of -25 dBFS would not.
| | 01:16 | Now that we know what threshold is
and how to read its values, how do
| | 01:20 | we actually use it?
| | 01:21 | Well, that depends on a number of other
compression parameters, including ratio,
| | 01:26 | which we will discuss in the next movie.
| | Collapse this transcript |
| Utilizing compression ratios| 00:00 | We now know that a compressor's
threshold determines the point at which the
| | 00:04 | compressor activates, but how much
the signal is actually compressed is
| | 00:08 | controlled by a parameter called ratio.
| | 00:11 | A compressor's ratio control determines
how much any signal over the threshold
| | 00:15 | is attenuated and is generally expressed as
a larger number over one, for example, 4:1.
| | 00:22 | Many times compressors will express the
threshold and ratio relationship using
| | 00:26 | an X-Y graph called a transfer curve.
| | 00:29 | A transfer curve graph plots the
signal's input into the compressor on the X
| | 00:33 | axis and the signal's output on the Y
axis, so you can easily trace different
| | 00:38 | scenarios just by following the line.
| | 00:41 | The kink in the line, called the knee,
is the point at which the compressor's
| | 00:45 | response becomes nonlinear.
| | 00:47 | That is to say, the input and the
output do not match because the compressor is
| | 00:51 | now reducing the gain of any signals over the
threshold by the amount defined by the ratio.
| | 00:57 | A ratio of 1:1 means no compression takes
place, and lower ratios may be barely visible.
| | 01:04 | Higher ratios create a more aggressive
kink or even a horizontal output level volume.
| | 01:10 | The easiest way to wrap your head
around ratio is to flip the number and
| | 01:14 | treat it like a fraction.
| | 01:16 | A ratio of 4:1 flipped into a
fraction would be one-fourth.
| | 01:20 | So every signal that passes the
threshold is reduced to a quarter of the
| | 01:23 | original input level. Sound complicated?
| | 01:27 | You get the hang of it fairly
quickly after a few examples.
| | 01:30 | Let's take a look at a simple input-
output table where the threshold is -20 dBFS
| | 01:35 | and the compressor's ratio is set to 2:1.
| | 01:38 | With a threshold set at -20, an input
signal at -25 does not breach the threshold,
| | 01:45 | so no compression is applied.
| | 01:46 | If the signal is increased to -10, that
is, 10 dBs over the threshold, at a ratio
| | 01:52 | to 2:1, the overage of 10 dB is half
to 5 dB, so the Output level is -15 dBFS.
| | 02:01 | If the Input level is increased even
more, to -5, the Output level would be -12.5.
| | 02:07 | Why? Because the input signal is 15 dBs over
the threshold and half of that is 7.5.
| | 02:13 | Add the 7.5 to -20 dBFS threshold
value to get -12.5. A ratio of 2:1 is
| | 02:21 | considered light compression.
| | 02:23 | Increasing the ratio and lowering the
threshold of the compressor will cause the
| | 02:27 | compressor to attenuate the
incoming signal even more.
| | 02:31 | Compressors with ratios of 10:1
and higher provide much stronger
| | 02:35 | compression and even go by a
different name, limiters, which we will be
| | 02:39 | covering later in this course.
| | 02:41 | If the graphs and ratio numbers I
have shown you here have you shaking
| | 02:44 | your head, don't worry.
| | 02:45 | I rarely concern myself with the
actual numbers while I am working with the
| | 02:49 | compressor and generally apply the
compression by ear, using strategies that
| | 02:53 | I'll explain later in the course.
| | 02:55 | Now let's move on to some
other compression parameters.
| | Collapse this transcript |
| Understanding makeup gain and gain reduction| 00:00 | When an audio signal is compressed, its
amplitude above the threshold is reduced
| | 00:04 | according to the ratio.
| | 00:06 | The gain reduction, sometimes labeled GR,
tells us how much threshold interaction
| | 00:11 | and ultimately how much gain reduction
or signal attenuation is taking place in
| | 00:15 | the compressor at any given time.
| | 00:18 | In simple terms, gain reduction tells us
how much the compressor is turning down
| | 00:22 | our signal, measured in decibels.
| | 00:25 | Gain reduction can be
monitored in most compressors.
| | 00:28 | When audio is played through a
compressor, the amount of gain reduction at any
| | 00:32 | moment is shown on the GR meter.
| | 00:35 | Watch the Gain Reduction meter on
this Wave's plug-in as I play some
| | 00:38 | compressed audio through it.
| | 00:39 | (music playing)
| | 00:49 | The lower we set the threshold, the
more audio signal is compressed, resulting
| | 00:54 | in more gain reduction.
| | 00:55 | Also, if the ratio is set high, say
8:1, then we will also see more gain
| | 01:00 | reduction than with lower ratios.
| | 01:02 | Because our working compressor is
actively attenuating or lowering the level of
| | 01:07 | the signal, most compressors feature
makeup gain, sometimes labeled just gain or
| | 01:13 | output, to recover any lost volume
after the compression has been applied.
| | 01:17 | I like to think of it this way.
| | 01:19 | The threshold comes down to interact
with the louder parts of the program
| | 01:23 | triggering gain reduction and
attenuating the louder signals.
| | 01:27 | Makeup gain is then used to raise the
overall level of the signal, bringing up
| | 01:31 | the uncompressed values to fill in
the signal and blend it into the mix.
| | 01:35 | For basic compression tasks, start by
lowering the threshold to a value that
| | 01:39 | achieves between 4 to 6 dBs of gain reduction.
| | 01:43 | That's a good place to start.
| | 01:45 | Next, since parts of the signal have
been compressed, use the makeup gain to
| | 01:49 | match the signal's level to
around the same as before compression.
| | 01:53 | This trick allows you to turn off the
effect and evaluate your processing before
| | 01:56 | and after without a net gain change.
| | 01:59 | So you can really hear if you are
getting somewhere, rather than the "oh, it's
| | 02:02 | louder and must be better" reaction
so many engineers make the mistake of.
| | 02:06 | I like to use my ears to do this, but
you can often get away with setting the
| | 02:10 | makeup gain to whatever the gain
reduction meter is showing on average.
| | 02:15 | So if that hovers between 6 and 8 dBs,
try using a makeup gain of 7 dBs.
| | 02:20 | Some signals will require more gain
reduction, while some will require less.
| | 02:25 | This is where using your ears in the
context of the material becomes critical.
| | 02:29 | No preset on a compressor plug-in can
tell you how much gain reduction is going
| | 02:33 | to work for your specific signal.
| | 02:36 | So close your eyes, open your ears,
| | 02:38 | and tweak accordingly.
| | 02:40 | Note that some compressors do not
have controllable thresholds or any
| | 02:43 | visual threshold at all.
| | 02:45 | These compressors are said to
be fixed threshold in design.
| | 02:49 | The UA 1176 is an example of a
famous fixed-threshold compressor.
| | 02:54 | With these types of compressors, it's
even more important to read the gain
| | 02:58 | reduction meter and use your ears
to adjust the makeup gain properly.
| | 03:03 | So if you can understand and operate
the threshold and makeup gain controls, as
| | 03:06 | well as read a gain reduction meter,
you can successfully use most compressors.
| | Collapse this transcript |
| Understanding attack and release| 00:00 | We have established that a compressor
reacts by turning down the signal whenever
| | 00:04 | it gets louder than its threshold setting.
| | 00:07 | But just as it takes you a few seconds
to turn down the volume of your TV when
| | 00:10 | those loud commercials come on, and a
few more seconds to turn the volume back
| | 00:15 | up when the program returns, a
compressor takes time to react to a signal also.
| | 00:20 | A compressor's attack and release
controls determine the reaction time of any
| | 00:24 | gain reduction once the threshold is breached.
| | 00:27 | When a signal is determined to be too
loud and shoots over the threshold, the
| | 00:31 | attack time is how long it takes for
the compressor to grab hold of the signal
| | 00:35 | and turn down the volume.
| | 00:37 | Likewise, when that same signal falls
back below the threshold and stops being
| | 00:41 | too loud, the release time is how long
it takes for the compressor to let go of
| | 00:45 | the signal and return it
to its uncompressed level.
| | 00:48 | Well, it might take you five to ten seconds to
grab that remote and adjust the volume
| | 00:53 | on your TV, compressors generally
work much faster, sometimes even reacting
| | 00:58 | instantly to the incoming signal.
| | 01:00 | A compressor's attack and release are
usually measured in milliseconds, or even
| | 01:04 | microseconds. Some compressors offer
look-ahead processing, where the signal is
| | 01:09 | buffered and previewed by the
compressor's algorithm to preempt a threshold
| | 01:12 | breach before it can happen.
| | 01:15 | Attack and release settings are based
on the signal you are trying to process,
| | 01:18 | but at a minimum, the attack should be
set fast enough to grab hold of a signal
| | 01:22 | before it completely decays, and the
release should be quick enough for the
| | 01:26 | compressor to recover
before the next note or beat.
| | 01:29 | For example, if the compressor is set
for too long of an attack time, the signal
| | 01:34 | might sneak through completely
before the processor has time to grab it,
| | 01:37 | kind of like trying to catch a ball
after it's already passed through your arms.
| | 01:42 | Likewise, too long of a release time
may have the compressor holding on for too
| | 01:46 | long, continuing any gain reduction
into the next note or phrase, even if that
| | 01:51 | note is below the threshold.
| | 01:53 | Definitely watch out for this with
percussive signals like drums that tend
| | 01:57 | to have very short decay times,
sometimes well under 100 milliseconds from
| | 02:01 | attack to complete decay.
| | 02:04 | Now that we understand what attack and
release are, in the next movie we will
| | 02:07 | cover strategies for applying attack
and release settings within a compressor.
| | Collapse this transcript |
| Applying attack and release| 00:00 | You might think that using the
compressor's fastest attack and release times
| | 00:04 | might be the best practice
for compressing an audio signal.
| | 00:07 | While sometimes this is true, very fast
attack and release times have their own
| | 00:12 | set of potential problems, and you will
often hear the compressor's character
| | 00:15 | more apparently using faster settings.
| | 00:19 | Too fast of an attack time can make a
signal sound dull and shave off its attack.
| | 00:24 | The difference between a one-
millisecond and a five-millisecond attack on a
| | 00:27 | snare drum can be very apparent in the
bite of the transient, especially with
| | 00:32 | more aggressive threshold settings.
| | 00:34 | (music playing)
| | 00:43 | A fast attack can be a positive thing too.
| | 00:46 | A super-fast attack can mellow out a
signal's transients so that they don't
| | 00:50 | stick out of a mix.
| | 00:51 | Consider an aggressive pick rake
of an over-strummed acoustic guitar.
| | 00:54 | (music playing)
| | 01:01 | A fast attack setting on your
compressor can help tame the signal's bite and make it
| | 01:05 | play nice with the other instrumentation.
| | 01:08 | (music playing)
| | 01:14 | Now let's talk about release times.
| | 01:16 | Fast release settings can be great for
bringing out the sustain and drawing out
| | 01:20 | the tail end of notes, words, and breathes.
(music playing)
| | 01:27 | But too fast of a release setting on a
low-frequency signal can create a nasty
| | 01:31 | distortion as the compressor literally
attempts to follow the oscillation of the
| | 01:35 | slower-moving bass frequencies.
| | 01:37 | It is generally a good idea to use
release settings of 20 milliseconds or
| | 01:41 | greater on anything containing significant
low-frequency material to avoid this artifact.
| | 01:46 | (music playing)
| | 01:56 | Remember, fast and slow in terms of
attack and release is relative to the
| | 02:00 | signal you're processing.
| | 02:02 | A 50-millisecond attack on a vocal or
bass may work great but do absolutely
| | 02:06 | nothing on a snare- or bass-drum hit
that lasts only 100 milliseconds in total.
| | 02:11 | When in doubt, use your DAW's time
ruler to get a sense of how long a note or
| | 02:16 | phrase actually lasts in milliseconds, or
how much of a gap between notes you have
| | 02:20 | to work with if you want the
compressor to recover between each note or beat.
| | 02:25 | Compressing eighth notes on the snare at
a faster tempo, like 130 BPM, will likely
| | 02:30 | need a faster release time to recover
between hits than would a ballad at 60
| | 02:34 | BPM with the snare drum beats 2 and 4.
| | 02:37 | Adjust settings in smaller
increments when working on fast-decaying
| | 02:41 | percussive material.
| | 02:43 | Some engineers even take this time-
between-notes concept to the next logical
| | 02:47 | step by timing their attack, and more
specifically release, times to the BPM of the song,
| | 02:52 | using the calculation of eighth note at 120
BPM equals 200 milliseconds as a starting point.
| | 02:59 | While I too like to set my release times to
the beat of the tune in certain scenarios,
| | 03:04 | I generally always set my
attack and release times by ear,
| | 03:07 | as I have found that most compressors
are very signal dependant when it comes to
| | 03:11 | the reaction time and rarely exhibit
attack and release performance down to the
| | 03:15 | exact millisecond setting on the knob.
| | 03:18 | Certain compressors feature automatic
attack and release controls which adapt
| | 03:22 | and change based on the incoming signal.
| | 03:24 | So for signals with faster envelopes,
the compressor will use faster attack and
| | 03:28 | release times, and vice versa.
| | 03:31 | These types of compressors can be
great for transparent dynamic control, as
| | 03:35 | you're less likely to hear the
compressor working so hard, grabbing and
| | 03:38 | releasing the signal based on
static attack and release values.
| | 03:43 | Be sure to check out your
compressor's user's manual.
| | 03:45 | You would be surprised how many
features you may be missing out on because you
| | 03:48 | didn't realize they were there.
| | 03:50 | So to recap, here are some general
ideas when setting attack and release times.
| | 03:56 | Shorter attack times control
transients and plosives, but settings under 1
| | 04:00 | millisecond can take the bite
or brightness out of signals.
| | 04:04 | Longer attack times allow more of
the signals transient to get through.
| | 04:07 | This helps add bite or punch
to drums as the transient passes
| | 04:10 | through uncompressed--
| | 04:12 | 1 to 10 milliseconds on
drums or 10 to 50 milliseconds on
| | 04:15 | non-percussive instruments.
| | 04:17 | Shorter release times help inflate or
add sustain to signals but can cause
| | 04:22 | breathing or pumping under
heavier threshold and ratio settings.
| | 04:26 | They can also lead to
distortion on low-frequency content.
| | 04:30 | Longer release times offer less
compression artifacts, but compression may not
| | 04:34 | recover fast enough to react to
the next unique note or phrase.
| | 04:39 | Because every signal is unique and
every project has a unique tempo, feel, and
| | 04:43 | energy level, learning to set the
attack and release controls by ear using the
| | 04:48 | basic behaviors I have outlined will
be so much more beneficial to your mixes
| | 04:52 | than trying to find prescribed
settings on the web or with presets.
| | 04:56 | Again, think about what you want the
signal to do for you in the context of your mix.
| | 05:01 | Punchy, flat, firm, dynamic,
what does the song need?
| | 05:06 | I like punchy snare drums as much as
the next guy, but they just don't work on
| | 05:09 | every single song I mix.
| | 05:11 | So I let the song guide my decision
process, right down to every attack and
| | 05:15 | release setting, in how they affect
the other elements in the mix and the
| | 05:18 | emotional delivery of the music.
| | Collapse this transcript |
| Demystifying compression controls: soft knee vs. hard knee| 00:00 | The kink in a compressor's transfer
curve graph is known as the compression knee,
| | 00:05 | due to its distinctly bent shape.
The knee sits at or around the threshold point
| | 00:10 | and represents how much and at what
amplitude level the compressor will
| | 00:14 | attenuate the signal.
| | 00:16 | Some compressors allow us to change this
knee from what is known as hard knee, or
| | 00:20 | the full ratio of compression, as soon as
the signal passes the threshold to what
| | 00:25 | is known as soft knee, a more gradual
form of compression where the signal is
| | 00:29 | eased into the compressor's ratio
setting over a larger threshold range.
| | 00:34 | I say range because in a soft-knee
compression curve, the threshold is no
| | 00:39 | longer a single point on the graph,
but begins and ends over a larger range
| | 00:44 | of signal values centered around the
original threshold setting, starting with
| | 00:48 | a lighter ratio and working its way to the
full amount as defined by the ratio control.
| | 00:54 | Again, this concept is best
represented by a transfer curve graph.
| | 00:59 | Imagine a compressor with the ratio
of 10:1 and a threshold at -20 dBFS.
| | 01:05 | In a soft-knee compression curve, the
compressor would begin to attenuate the
| | 01:08 | signal earlier, at around -30, with a
ratio of 2:1. Maybe you're amping up to
| | 01:14 | 4:1 at -25, 8:1 at -20, and finally reaching
10:1, just pass the threshold, at around -15 dBFS.
| | 01:24 | On some compressors, the knee value
will be a simple switch, allowing you to
| | 01:28 | choose between a hard or soft
compression curve, while on other compressors,
| | 01:33 | this control may be defined in decibels,
representing the threshold range over
| | 01:37 | which the compression will
ease into the full ratio.
| | 01:40 | Many compressors will not have this
control at all and will either be fixed as
| | 01:44 | hard-knee or soft-knee
compressors or somewhere in between.
| | 01:47 | For instance, the Dbx 160 doesn't have
a knee control, but it's known for its
| | 01:53 | over-easy soft-knee setting, which
often sounds great on vocals and bass.
| | 01:58 | As a general strategy, I like to use
hard-knee compressors when I am working
| | 02:02 | with the internal dynamics of a
signal's envelope, especially on drums and
| | 02:05 | percussion, like drawing out the
attack of a snare or kick drum.
| | 02:09 | The hard-knee transition at the threshold
gives me tons of control over the signal's envelope.
| | 02:15 | Likewise, soft-knee compression is
great on less percussive material like
| | 02:19 | vocals, bass, and guitar, where you
might not want to hear as much of an edge,
| | 02:23 | but still have a nice firm
signal sitting in the mix.
| | 02:26 | Soft-knee compression can also work
well on full mixes, where you're trying to
| | 02:31 | subtly glue signals together without
a lot of attitude from the compressor.
| | 02:35 | So if you've never experimented with a
compressor's knee, I think you'll find
| | 02:38 | that it adds a lot of
flexibility to dynamic control.
| | Collapse this transcript |
| Get in the Mix: Using compression to even out a vocal performance| 00:00 | The human voice, like most acoustic
instruments, tends to be very dynamic when recorded.
| | 00:06 | While this enables us to produce a
wide variety of sounds, from intimate
| | 00:09 | whispers to barbaric shouts, that variety of
dynamics can be difficult to fit into a song's mix.
| | 00:15 | In fact, there aren't many
instruments, the human voice included, that are
| | 00:19 | designed specifically to sit in the
middle of a huge hundred-track pop mix,
| | 00:23 | so they often need a little help to reduce
their dynamic range using the compressor.
| | 00:28 | My strategy when it comes to
compressing vocals in a mix is pretty simple:
| | 00:32 | make the singer sound like a star.
| | 00:34 | This means confident and larger-than-life
vocals that sit in the mix like they were
| | 00:38 | meant to be there, not strapped on to a
background track like a bad night at the karaoke bar.
| | 00:43 | Now, I am not going to lie;
| | 00:44 | a great sounding vocal track
comes from a great vocalist.
| | 00:47 | There isn't any mix magic or a super-
expensive compressor that's going to
| | 00:51 | take a lifeless half-baked vocal track
and turn into the performance of the lifetime.
| | 00:56 | The singer needs to sell the
performance during the recording stage.
| | 00:59 | Compression will only help take a
great take and make it sound better.
| | 01:04 | Now, it's time to get in the mix.
| | 01:05 | Pause this movie and open the
appropriate file for your DAW.
| | 01:09 | If you don't have access to a DAW
right now, you can continue watching this
| | 01:12 | video to see the Get in the Mix demonstration.
| | 01:19 | Let's take a listen to an example
from the song Say Yes by Iyeoka.
| | 01:22 | Here is a sample of the lead
vocal track without compression.
| | 01:26 | Listen to how some words pop
out while others can't be heard.
| | 01:29 | (music playing)
| | 01:48 | Now listen to the vocal with
compression and how the level is more consistent.
| | 01:52 | (music playing)
| | 02:11 | Take a look at the muted track
labeled Vocal Processed for a visual
| | 02:15 | representation of what the
compressor is doing to the waveform.
| | 02:18 | If you'd like, pause the session
and zoom in to take a closer look.
| | 02:21 | An uncompressed vocal often sounds a
bit disconnected from the mix, like it
| | 02:27 | doesn't really belong with
the rest of the instruments.
| | 02:30 | Regardless of the volume level I set
the vocal at in the mix, they are bound
| | 02:33 | to be certain words that stick out too far
and others that get buried behind the music bed.
| | 02:39 | I like to think of compression on the
vocal as serving two main functions:
| | 02:43 | first, being simple dynamics ctrl, so
I can hear what the lyric is saying and
| | 02:47 | the second being tonal shaping and
firming, so the vocal takes on a larger-than-
| | 02:52 | life quality and really
connects with the listener.
| | 02:55 | Again, listen to the mix and pay attention as
I automate the settings to apply compression.
| | 03:00 | And because compression doesn't
live in a vacuum, after adjusting the
| | 03:03 | compression, I will add a bit of reverb
and delay to sit the vocal into the mix.
| | 03:07 | (music playing)
| | 03:37 | Dynamics control the vocal is a two-
stage process of tucking in the louder words
| | 03:41 | and phrases and turning up the result,
allowing the softer notes to sit at or
| | 03:46 | near the level of the louder ones.
| | 03:48 | This is achieved by pulling down
threshold until the louder words start to
| | 03:52 | trigger compression--
| | 03:53 | about 6 dBs on average of gain
reduction in this case--and then making up the
| | 03:58 | loss in signal level using the output gain.
| | 04:01 | Notice that I am using a ratio of about
6:1, an attack of 10 milliseconds, and a
| | 04:05 | release time that allows the
compressor to recover between words.
| | 04:09 | I am also using a soft-knee setting to
help the compressor ease into the ratio
| | 04:14 | and sound more transparent.
| | 04:15 | Once I have the vocal sitting in the mix
using compression, I will then consider
| | 04:20 | adding volume automation to perfect the
balance over each section and fine-tune
| | 04:25 | any trouble spots that still
stick out or fall below the mix.
| | 04:29 | As I listen to and adjust other tracks,
I may come back to the vocal and adjust
| | 04:33 | some of the settings to increase or
decrease the total amount of gain reduction
| | 04:37 | or try a different compressor,
depending on how the mix is shaping up.
| | 04:42 | This same technique works for
almost any instrument in your mix.
| | 04:45 | Once you determine that a track needs a
bit of compression, start small and work
| | 04:49 | your way into the sweet spot.
| | 04:51 | Throughout this course, I will tend
to process audio examples a bit more
| | 04:54 | aggressively for greater educational
impact, but you generally want to avoid
| | 04:59 | over-compressing the signal, unless
that's they affect you're going for.
| | Collapse this transcript |
| Get in the Mix: Using compression to add punch and sustain to drums| 00:00 | Let's take a look at how to apply
compression to a single note, using it to draw out
| | 00:04 | the attack or sustain of a signal and
letting that note punch through dense
| | 00:08 | musical sections of a mix.
| | 00:10 | This type of compression is often referred
to as envelope shaping or a transient shaping.
| | 00:14 | A shape of a note's
waveform is called its envelope.
| | 00:17 | The envelope describes how
a note evolves over time.
| | 00:21 | Think of an envelope as the trip a
signal's waveform takes from initial
| | 00:25 | development through final decay.
| | 00:27 | Let's take a look at a
snare drum note's envelope.
| | 00:30 | We start with the sharp transient.
| | 00:32 | This is referred to as the attack.
| | 00:35 | This sharp transient is followed by a
brief sustain and final release period,
| | 00:39 | as the sound dies out.
| | 00:41 | Because a compressor reacts to a
signal's amplitude as it changes over time, we
| | 00:45 | can use compression to play with
the shape of a signal's envelope.
| | 00:49 | Now it's time to get in the mix. Pause this
movie and open the appropriate file for your DAW.
| | 00:53 | If you don't have access to a DAW
right now, you can continue watching this
| | 00:55 | video to see the get in the mix demonstration.
| | 00:57 | Listen to the snare drum without any
compression, pay attention to the sound of
| | 01:01 | the attack and release.
| | 01:02 | (music playing)
| | 01:12 | Now listen to the same snare with
compression applied and listen for how
| | 01:15 | the sound has changed.
| | 01:16 | (music playing)
| | 01:27 | Notice how in the compressed version
the attack of the signal is pulled up
| | 01:30 | and has more punch.
| | 01:32 | This is achieved by adjusting the
attack time of the compressor to allow a bit of
| | 01:35 | their original transient
through before being compressed.
| | 01:38 | An attack time of between 1 and 10
milliseconds works great for this, but any
| | 01:44 | longer and I risk missing
the transient all together.
| | 01:47 | After making up the gain from the
compression, what we end up with is a larger
| | 01:51 | transient, or initial attack portion of
the sound, than we started with, because
| | 01:56 | the body or sustain portion of a sound
has been attenuated, thus changing the
| | 02:00 | dynamic relationship between the two.
| | 02:03 | A muted copy of the processed waveform
has been provided as a visual reference
| | 02:08 | underneath the active snare track.
| | 02:10 | Now, listen again to the snare in the
context of the full mix and pay attention
| | 02:14 | as I automate the settings
to apply the compression.
| | 02:16 | I am exaggerating a bit here so you can
really hear the compression, but notice
| | 02:21 | how the punchier snare helps
the drum kit drive the song.
| | 02:24 | (music playing)
| | 02:42 | Dynamics are all about relative
relationships and amplitude, and in this case I
| | 02:47 | have changed the relationship between
the attack and the rest of the snare hit.
| | 02:51 | The more aggressive the threshold and
ratio settings I use, the more distance I
| | 02:55 | create between the transient
and the rest of the snare hit.
| | 02:58 | We can achieve the opposite
effect by using fast release times.
| | 03:03 | Let's take a listen to very fast
release time on the snare drum.
| | 03:06 | Again, I have exaggerated the effect,
so you can really hear the change in the
| | 03:10 | shape of the snare envelope.
| | 03:11 | (music playing)
| | 03:29 | By allowing the compressor to compress
the initial transient body of the snare
| | 03:33 | drum and quickly release the
compression before the soft or decay section, we
| | 03:37 | can use the makeup gain to inflate the tail
of the signal and draw it out longer in time.
| | 03:42 | Notice how I am driving the compressor a
little harder here, achieving more gain reduction.
| | 03:47 | This check works best with more
aggressive threshold and ratio settings.
| | 03:51 | Feel free to zoom in and look at the
processed example tracks waveform to better
| | 03:55 | visualize what's happening.
| | 03:56 | If you want to watch your attack when
you reach these levels of gain reduction,
| | 04:01 | even a small amount of transient
escaping through uncompressed can eat up all
| | 04:05 | your headroom and clip the output.
| | 04:08 | This trick doesn't work on every kind of
material with every compressor, so use your judgment.
| | 04:13 | Sometimes I split the track into two
and process one for a sharper attack and
| | 04:17 | another for more sustain
and then blend the two to taste.
| | 04:22 | Some compressors are better at
each of these tasks than others.
| | 04:25 | In order to achieve effective
envelope shaping with the compressor, we're
| | 04:29 | generally looking for a processor
with a very fast attack and release time.
| | 04:33 | A slower compressor, or one without
attack and release controls, may not be fast
| | 04:37 | enough to really draw out the envelope precisely.
| | Collapse this transcript |
| Intelligently using compression presets| 00:00 | Say you've just got a fancy new compressor
and notice it was loaded with tons of presets.
| | 00:05 | Descriptions like "best vocal sound ever"
and "snare to die for" fill this list, and
| | 00:10 | you're thinking, this is it, I
finally unlocked the secrets of mixing by
| | 00:14 | purchasing this plug-in and
all my problems are solved.
| | 00:17 | While there's no doubt that
presets are excellent starting points for
| | 00:21 | discovering the features of a new tool,
| | 00:23 | when it comes to presets for dynamics
processors, there are a few simple rules
| | 00:27 | to follow to set yourself up for success.
| | 00:30 | First off, dynamics presets can
actually be pretty awesome and often do exactly
| | 00:35 | what you're looking for right out of
the gate--if you know how to use them.
| | 00:39 | The secret to using compression presets
is to understand that the preset has no
| | 00:44 | idea what kind of signal level is coming in.
| | 00:47 | Therefore, the default threshold and
makeup gain settings are almost always
| | 00:51 | going to be completely wrong,
| | 00:53 | either giving you too much compression, too
little compression, or even worse, clipping.
| | 01:00 | Here's how I approach compressor presets.
| | 01:02 | What I've got here is the C1 comp from
waves, and the first thing I am going to
| | 01:06 | do is call up a preset.
| | 01:10 | Pick this one that says Great Vocal here.
| | 01:13 | Now what I'll notice is that the threshold
and the makeup gain already have settings.
| | 01:19 | So the first thing I am going to do after
calling up that preset is reset them both to 0.
| | 01:27 | Next, I am going to play back my session.
Then I am going to adjust my Threshold
| | 01:31 | to get the right amount of gain
reduction that I want and then use my makeup
| | 01:35 | gain to bring that back up.
| | 01:36 | (music playing)
| | 02:07 | I like to start with 4 to 6 dBs of
gain reduction and listen from there.
| | 02:13 | At this point, I'll start tweaking,
adjusting the threshold to achieve more
| | 02:17 | or less gain reduction. Or I will adjust
the other controls to better suit my signal.
| | 02:22 | Again, the trick to using any kind of
preset is to accept that the person who
| | 02:26 | made the preset has no idea what your
signal sounds like or what you're trying
| | 02:30 | to accomplish in your mix. How could they?
| | 02:33 | Unless they have a time machine or
magical powers, they are probably building
| | 02:37 | the preset based off some
recording they have access to.
| | 02:41 | This is why we need to adjust the
parameters that we know aren't going to
| | 02:44 | match up from the get-go.
| | 02:46 | In the case of dynamics processors,
we start by adjusting our threshold and
| | 02:50 | makeup gain to suit our signal's input level.
| | 02:53 | Remember, because some compressors do
not have a dedicated threshold control,
| | 02:57 | you may have to adjust the input and
output controls to achieve the right
| | 03:00 | amount of gain reduction.
| | 03:02 | In these scenarios, let the
gain reduction meter be your guide.
| | Collapse this transcript |
| Recording with compression: Why or why not?| 00:00 | One of the most common questions I get
asked in any mixing course I teach is,
| | 00:05 | should I record with compression?
| | 00:07 | And my usual answer to that is, if
you're asking me that question, the answer
| | 00:11 | is no, you shouldn't. Why not?
| | 00:13 | Well, if you mess up this signal with a
botched compression setting, like if you
| | 00:17 | overcompress the signal or
unwittingly add unwanted distortion while
| | 00:21 | compressing, that bad compression sound
will be recorded and it's really hard,
| | 00:26 | if not impossible, to undo.
| | 00:28 | Instead, record the signal with no
compression and add compression later in
| | 00:33 | the mixing process.
| | 00:35 | Just to be clear, by recording with
compression, I mean using a hardware or
| | 00:39 | software compressor before your
signal is recorded to hard drive or tape.
| | 00:43 | This is not the same as inserting a
real-time plug-in on the track you are recording to.
| | 00:48 | If a compressor plug-in is inserted on
the actual audio track you're according
| | 00:52 | to, the compression is added after the
signal has been recorded, allowing you to
| | 00:57 | affect the track in a nondestructive
way, and you could freely change the
| | 01:01 | compression setting after recording.
| | 01:04 | This technique gives you the
most control over your sounds.
| | 01:07 | That said, many well-regarded engineers
do actually record with compression. Why?
| | 01:13 | Generally, there are three major
reasons to record with compression.
| | 01:17 | First, sometimes you just know how you
want something to sound and you go for it.
| | 01:21 | For example, say you've worked with a
particular vocalist before and you know
| | 01:25 | that she performs very
dynamically when recording.
| | 01:28 | Adding some light compression with a
particular compressor while recording helps
| | 01:32 | to control the dynamic range of her
performance and makes less volume adjustment
| | 01:36 | necessary when mixing.
| | 01:38 | The second reason stems from the practice
of maximizing the input signal strength.
| | 01:43 | In the days of analog recording,
many engineers compressed signals while
| | 01:47 | recording to improve the signal-to-
noise ratio going to tape, because if you
| | 01:51 | compress the signal after it had gone to
tape, you would risk bringing up the noise
| | 01:55 | floor and tape hiss.
| | 01:57 | However, with 24-bit digital
recording systems and modern analog-to-digital
| | 02:01 | converters, their minimal input noise
and wide dynamic range provides such
| | 02:06 | a high signal-to-noise ratio that I personally
find no need for compression in that scenario.
| | 02:12 | The third reason to use compression
while recording is to change the input sound.
| | 02:16 | If I have access to a great compressor
and want to run my signal through it to
| | 02:20 | warm it up a bit, I might take the opportunity.
| | 02:23 | Well, what I'm looking for in that
scenario is not really compression per se.
| | 02:27 | It's tonal shaping, adding
colorization, character, or depth to the signal
| | 02:31 | before it hits the hard drive.
| | 02:33 | While I certainly don't use
compression on every track I record, it
| | 02:37 | certainly can be useful.
| | 02:39 | What I would suggest is if you want
to experiment with compression while
| | 02:42 | recording, start with very small
amounts until you really understand the sound
| | 02:47 | of your gear and get a feel for how
everything plays out in the final mix.
| | Collapse this transcript |
| Recording with compression: How to do it| 00:00 | Now that we know the strategy behind
recording with compression, let's take a
| | 00:04 | look at some hands-on examples,
both outside and inside the box.
| | 00:08 | Let's start by recording a bass
guitar through a hardware compressor.
| | 00:16 | When recording through a hardware
compressor, the audio signal flow is going to
| | 00:20 | travel from the bass into a
preamp with an instrument level input,
| | 00:24 | then from the output of the preamp
into the hardware compressor's input, and
| | 00:29 | from the compressor's output into
the DAW interface's line-level input.
| | 00:34 | Using the studio's patch bay, I'll route
the output of the preamp into the input
| | 00:39 | of the LA-2A compressor.
| | 00:40 | The output of the LA-2A is
then patched into our DAW input.
| | 00:46 | When using multiple pieces of equipment
in the signal chain, it is important to
| | 00:50 | watch the levels of each one of your
gain stages and be sure you aren't clipping
| | 00:53 | at any point in the chain.
| | 00:55 | It's very easy to add too much gain at
one processor, thus clipping the output
| | 00:59 | of that unit while hiding that
clipped signal from your DAW's metering by
| | 01:04 | attenuating the next unit's
input or output in the signal shape.
| | 01:08 | To avoid this, you can bypass the
compressor if it has a bypass or insert
| | 01:12 | button, or make sure it's not
adding any gain to the signal chain.
| | 01:17 | In the LA-2A here, because it doesn't
have a bypass, I am going to pay attention
| | 01:21 | to the compressor's input and output
metering to ensure proper gain staging.
| | 01:26 | Now I will set up the gain on my
preamp to get a nice clean unclipped
| | 01:29 | signal coming into my DAW.
| | 01:31 | Once that's set up, I can then move
over to my compressor and adjust the
| | 01:35 | controls to achieve the
desired amount of gain reduction.
| | 01:38 | Then I'll adjust the output gain of the
compressor to get about the same level
| | 01:42 | signal going into my DAW as
I had using only the preamp.
| | 01:45 | Now that it's all set up, let's record a take.
| | 01:49 | The bass track will be
recorded with compression;
| | 01:52 | therefore, its waveform will exhibit a
reduced dynamic range in comparison to an
| | 01:55 | uncompressed-based waveform.
| | 01:58 | (music playing)
| | 02:40 | In most DAWs you can record through a
compressor by routing the signal through
| | 02:44 | an auxiliary track before
recording it to an audio track.
| | 02:47 | Here, the input of the aux track
receives the signal from the base and the
| | 02:52 | output of the aux track feeds an
internal bus, which is then routed into a
| | 02:56 | separate audio track that records the signal.
| | 03:00 | In this setup, any plug-in that's
placed on the aux track will affect what
| | 03:03 | is recorded to disk.
| | 03:05 | In contrast, any plug-in that's placed
on the audio track is simply monitored in
| | 03:09 | real time and not recorded.
| | 03:11 | So now that you know how to record
through a hardware or software compressor,
| | 03:15 | experiment with this technique the
next time you're recording a very dynamic
| | 03:19 | performer to tame their input signal.
| | Collapse this transcript |
|
|
3. LimitersIntroducing limiters| 00:00 | A limiter is simply a compressor with a
ratio of greater than 10 to 1. BY using
| | 00:05 | such a strong ratio,
| | 00:06 | the compressor effectively limits the
signal's amplitude to the threshold so
| | 00:10 | that very little, or in some cases
nothing, can pass beyond the threshold.
| | 00:14 | When we look at a limiter's transfer
curve graph, we find that the knee levels off
| | 00:19 | completely, effectively
creating a wall at the threshold.
| | 00:23 | Since compressors and limiters are cut
from the same cloth, as far as dynamics
| | 00:27 | processors go, we'll typically use
them for all the same reasons we use
| | 00:31 | compressors, that is, to restrict
and hone the dynamic range of overly
| | 00:35 | dynamic signals and increase the
average level of a signal without
| | 00:38 | increasing peak level or clipping.
| | 00:40 | Think of limiters as just
more aggressive compressors.
| | 00:44 | You'll find that many limiters feature
the same threshold attack, release. and
| | 00:49 | gain controls as regular compressors,
but often without a ratio control.
| | 00:53 | since this is generally assumed
to be very high or near infinite.
| | 00:57 | Limiters are especially useful while
mixing because many signals contain great
| | 01:01 | variations between their peak and average level.
| | 01:04 | For example, a signal may
contain many very short peaks or high-
| | 01:08 | amplitude transient spikes
| | 01:10 | while the average amplitude
level of the signal is much lower.
| | 01:14 | These peak points in the signal tend to
eat up headroom and are hard to fit into
| | 01:19 | a mix, because as we turn up the
average levels of a signal up enough to hear
| | 01:24 | it, the peaks tend to
stick out or clip in the mixer.
| | 01:27 | Using a limiter helps us control this
type of signal, allowing it to both be
| | 01:31 | audible throughout while not
jumping out of the mix or clipping the DA
| | 01:35 | converters on a mix buzz on the peaks.
| | 01:38 | Listen as I raise the level of this
drumbeat. The peaks of the signal quickly
| | 01:41 | clip the output of my DAW and I
begin to hear digital distortion.
| | 01:45 | (music playing)
| | 01:48 | Using a limiter can help me control these peaks,
| | 01:51 | while still allow me to raise up
the level of the track in the mix.
| | 01:55 | In the next movie, we'll discuss
the two major types of limiters.
| | Collapse this transcript |
| Types of limiters| 00:00 | There are two major classes of limiters.
| | 00:02 | The first type is one we're already
familiar with, as any compressor with a
| | 00:06 | ratio control can be used as a limiter,
simply by increasing the ratio above 10 to 1.
| | 00:12 | These types of limiters are useful
for controlling signals that need more
| | 00:15 | aggressive gain reduction than
compressors with lower ratios, such as very
| | 00:19 | dynamic vocal performances.
| | 00:21 | The second type of limiter is a
special limiter whose ratio is
| | 00:24 | essentially infinite.
| | 00:26 | These limiters are often called
brickwall limiters because no matter what the
| | 00:30 | input amplitude is, the single
will never pass over the threshold.
| | 00:34 | Brickwall limiters are typically used
to maximize loudness and average level
| | 00:38 | while protecting the signal from
clipping. Because no signal is allowed to pass
| | 00:42 | the threshold, these types of limiters
generally do not have attack controls;
| | 00:47 | the attack time is instant.
| | 00:49 | Actually the limiter buffers the
incoming signal and instantly reduces any high-
| | 00:53 | amplitude signal that'll
overshoot the threshold.
| | 00:56 | This feature is often referred to as
look-ahead processing and is why
| | 01:00 | most brickwall limiters introduce a
small amount of latency or delay on the
| | 01:04 | signal they're processing in
order to fill the look-ahead buffer and
| | 01:08 | anticipate any overtures.
| | 01:11 | Look-ahead processing is one of the
main reasons these types of limiters are
| | 01:14 | used to maximize the overall loudness
of a signal or entire mix, as the peaks
| | 01:19 | are reduced instantly with no overshoot.
| | 01:22 | Also, the average level or body of
the signal can be raised up without fear
| | 01:26 | of clipping, effectively increasing the
perceived loudness while reducing the dynamic range.
| | 01:32 | Because many brickwall limiters are
designed to work transparently and on full
| | 01:36 | multi-track mixes, the release
controls are often adaptive or automatic.
| | 01:41 | This allows the limiters to react to
peaks very quickly without digging into the
| | 01:45 | material that follows.
| | 01:46 | While limiters can be extremely
powerful tools for honing the signals dynamic
| | 01:51 | range because of their aggressive
ratio an instant attack and release times,
| | 01:56 | this brute-force method of peak control can
become very nasty when over-used or abused.
| | 02:01 | There is a very fine line between a
nice amount of dynamic control and totally
| | 02:06 | squashing the life out of your signal.
| | 02:09 | If the limiter's threshold digs too deeply
into the signal, grabbing and releasing
| | 02:13 | the signal too quickly and
aggressively, this creates unwanted artifacts,
| | 02:17 | including distortion and what some call pumping.
| | 02:21 | Let's talk about why this is the case.
Because brickwall limiters often use
| | 02:25 | extremely fast release times, low-
frequency material may cause the limiter to
| | 02:30 | distort as it attempts to
trace the waveform's oscillation.
| | 02:33 | For example, 100 Hertz tone cycles through
its waveform at a hundred times per second.
| | 02:40 | It takes the wave from 10
milliseconds to fully develop as one complete
| | 02:43 | compression, or push, and rarefaction, or pull.
| | 02:48 | If the limiter's release time is set
faster than the waveforms oscillation
| | 02:52 | speed, there is a risk that the limiter
will follow the literal push and pull of
| | 02:57 | that waveform, grabbing the signal any
time the waveform has any amplitude as
| | 03:01 | either a compression or rarefaction.
| | 03:05 | With a deep threshold setting, the
insanely fast attacking release times of the
| | 03:09 | brickwall limiter can easily distort the signal.
| | 03:13 | Using a slower release setting can
sometimes alleviate the artifacts of
| | 03:16 | distortion by allowing the low-frequency
waveforms to complete their full cycles
| | 03:21 | before the limiter releases
the signal from gain reduction;
| | 03:24 | however, these artifacts are generally
a component of deep threshold settings
| | 03:29 | and changing their release to slower
settings may simply trade distortion for
| | 03:33 | audible pumping or breathing, where
the attack and releases now slow enough
| | 03:37 | to hear going in and out, like heavy breathing,
which in most cases doesn't sound much better.
| | 03:43 | We'll hear examples of
distortion and pumping in the next movie.
| | 03:47 | So to avoid distortion on low-
frequency material like bass and kick drum use
| | 03:52 | the processor's automatic release
control or set the release times to greater
| | 03:55 | than 20 milliseconds.
| | 03:57 | Remember that retaining some
dynamics in your mix is a good thing;
| | 04:00 | they're what make the speakers move
and the kick drum punches in the chest.
| | 04:04 | So when working with any kind of limiter,
remember that a little can go a long
| | 04:08 | way and always make sure you evaluate
the signal with no net gain change so
| | 04:12 | you can tell just how far you've gone.
| | Collapse this transcript |
| Get in the Mix: Maximizing mix loudness with brickwall limiters| 00:00 | A typical audio signal, whether it's
one individual instrument or a multi-track
| | 00:04 | mix, is usually made up of many quick
transients whose amplitude values extend
| | 00:09 | much further than the average
level or body of the signal.
| | 00:12 | These transients prevent us from
raising the overall level the signal too high,
| | 00:17 | because if we did, those transients
would exceed 0 dBFS, clipping the output
| | 00:22 | converters and causing distortion.
| | 00:23 | Now this presents a dilemma for our
human hearing, because our ears are tuned to
| | 00:28 | average out loudness over a longer
period of time than a millisecond-long
| | 00:32 | transient hitting the top of our dynamic range.
| | 00:35 | So how do we get the average level of
our mix up to a comparable level with the
| | 00:39 | rest of the songs in our music collection?
| | 00:42 | By clamping down on those P
transients and raising up the average level of
| | 00:46 | our mix using a brickwall limiter.
| | 00:49 | Think of a mix going into a brickwall limiter
like a spring being pushed into a concrete wall.
| | 00:55 | Since the spring can never go further
than the wall it's pushing against, the
| | 00:58 | coils will simply move closer together,
compressing the total length of that
| | 01:02 | spring into a smaller space.
| | 01:05 | Let's check out how a brickwall
limiter works in a musical context.
| | 01:09 | It's time to get in the mix. Pause this
movie and open the appropriate file for your DAW.
| | 01:16 | Most brickwall limiters are easy to
operate and have similar controls.
| | 01:20 | I'm going to use the limiter built
into my DAW for this example, but all the
| | 01:24 | major DAWs come with factory-
installed brickwall limiters,
| | 01:27 | in addition to dozens of third-
party limiter plug-ins you can purchase.
| | 01:31 | Typically how we use limiters to
maximize the volume of a mix is by placing
| | 01:35 | them as the last insert in the signal
chain, usually on the master fader or
| | 01:40 | output of the mixer. This allows the
limiter to increase the average level
| | 01:44 | while clamping down the peaks internally before
passing the final signal out to the D/A converter.
| | 01:51 | In this example, I've placed the
limiter directly on the stereo example track,
| | 01:55 | since I want to avoid limiting the
guide dialogue track you're listening to.
| | 01:58 | I'll start by first lowering the output
control to just under 0 dBFS, say around 0.1 dBFS,
| | 02:07 | so that the output from the limiter
never exceeds the converter's maximum output.
| | 02:12 | Now I can simply lower the threshold
control, which simultaneously increases the
| | 02:17 | gain into our 0.1 dBFS wall.
| | 02:20 | Let's take a listen.
| | 02:21 | (music playing)
| | 02:41 | We can see that it's working by looking
at the gain reduction meter. Take a look
| | 02:45 | at the waveform of the muted mix
process track for a visual representation of
| | 02:49 | what the limiter is doing.
| | 02:51 | You don't want to push things too far
or you'll literally flatten your mix and
| | 02:55 | take all the life out of it,
potentially resulting in distortion.
| | 03:00 | Take a listen to what happens if I pull
the threshold down too far. Then I will
| | 03:05 | slow down the release control to help
reduce distortion in the low frequencies.
| | 03:09 | (music playing)
| | 03:36 | Sounds nasty, huh? Even if I slowed the release control,
the distortion of a deep threshold is
| | 03:41 | merely replaced with audible
pumping as the limiter breaths in and out,
| | 03:46 | wheezing on the signal.
| | 03:48 | Generally I'm just going to
go a few DB and check my work.
| | 03:52 | Each track and limiter plug-in is
unique and will be able to handle different
| | 03:56 | amounts of limiting before breaking up.
| | 03:59 | At this point, we don't want to add
any plug-ins after the final brickwall
| | 04:02 | limiter, as any additional level changes
to our signal may clip the converter's output.
| | 04:08 | I like to evaluate what my brickwall
limiter is doing by pulling down the output
| | 04:12 | ceiling to the same level as the
threshold, so I can really hear the limiting
| | 04:16 | and not get fooled by the
additional gain I've added.
| | 04:18 | One of the biggest mistakes novices
make with these things is they simply grab
| | 04:23 | the threshold, pull it down a
whole bunch, and say "Awesome. It's so much louder.
| | 04:27 | It must be better." But when we
evaluate the processing without any net gain
| | 04:32 | change to the signal, we can hear exactly what
the limiter is taking away from our dynamics.
| | 04:37 | Take a listen too, as I reduce the
output to the threshold value and switch the
| | 04:41 | processing on and off.
| | 04:44 | I will purposefully use too much
limiting so you can hear how the limiter is
| | 04:47 | affecting the track when
monitored without makeup gain.
| | 04:51 | Notice that if we go too far we will
lose all the snap and bite out of our
| | 04:55 | drums and percussion.
| | 04:56 | (music playing)
| | 05:24 | One thing to note is that not all
brickwall limiters are created equally.
| | 05:27 | Some let you eek out a bit more average
level without stealing your punch, while
| | 05:31 | other start sounding
pretty bad when you push them.
| | 05:35 | Mixes with less brickwall limiting
typically sound better when cranked up,
| | 05:38 | because the transients of the instruments are
really allowed to work the speaker cones.
| | 05:43 | Tracks destined for the club or big
PA systems will do better with less
| | 05:47 | brickwall limiting and
greater peak-to-average ratio.
| | 05:51 | Tracks destined for further mastering
should almost always forgo the final stage
| | 05:55 | of limiting, as this process can
severely limit, no pun intended, what the
| | 05:59 | mastering engineer is able to accomplish.
| | Collapse this transcript |
| Solving common mix problems with limiters| 00:00 | Using limiters to squash your entire
mix into the loudest song ever created
| | 00:04 | isn't the only way we can
put them to work in our mix.
| | 00:08 | We can also use them to better
control a signal's peak to average level on
| | 00:12 | an individual track.
| | 00:13 | Let's listen to this example.
| | 00:15 | (music playing)
| | 00:21 | Notice that the signal contains many
quick peaks that reach very close to 0 dBFS.
| | 00:27 | Adding any compression without using
a near-instant attack would clip the
| | 00:31 | output of the compressor as
those quick peaks sneak through the
| | 00:33 | compressor's attack stage.
| | 00:36 | Watch the compressor's output and
listen to the quick transients escape through the
| | 00:40 | compressor's slower 20-
millisecond attack stage and clip.
| | 00:43 | (music playing)
| | 00:51 | Likewise, adding any boost via
EQ would cause a clip as well.
| | 00:58 | Again, watch the EQ's output meter and
listen to the EQ clip as the last bit of
| | 01:03 | headroom is gobbled up with the EQ boost.
| | 01:05 | (music playing)
| | 01:13 | Instead of a regular compressor, by
applying a brickwall limiter, we can retain
| | 01:17 | most of the dynamic feel of the signal
while keeping those peaks from jumping out
| | 01:21 | too far from the mix.
| | 01:23 | I'm going to apply waves L1 to
tame some of the biggest peaks,
| | 01:28 | linking my threshold and output
control, so no makeup gain is applied.
| | 01:32 | Listen as I drag down the
threshold and output controls.
| | 01:36 | (music playing)
| | 01:48 | Because the limiter works
instantly and recovers nearly as quick, no
| | 01:52 | peaks passed through.
| | 01:54 | This keeps the single from sounding
like it has been compressed and works great
| | 01:58 | as a transparent form of dynamics control.
| | 02:01 | Listen again as I play back this strong
pick scrape with and without the limiter.
| | 02:05 | (music playing)
| | 02:17 | By pulling down the threshold and output
simultaneously, I'm not using any makeup gain,
| | 02:22 | since my goal is to simply tame
those peaks while leaving the rest of the
| | 02:25 | signal unprocessed.
| | 02:27 | This technique is also great for
controlling headroom between plug-ins inside your DAW.
| | 02:32 | While a regular compressor would
generally allow a bit of the signal's attack
| | 02:35 | through, a limiter clamps down on
those transients that may cause digital
| | 02:39 | clipping within a plug-in.
| | 02:41 | This allows us to regain some headroom
below 0 dBFS, so we can apply gain in
| | 02:46 | another way, like if we wanted to
boost some frequencies with an EQ.
| | 02:50 | Using limiters on individual tracks and
between plug-ins really enables you to
| | 02:54 | control the dynamics of all
parts of your signal flow.
| | Collapse this transcript |
| Using layered dynamics processing| 00:00 | Now that we know we can use limiters
inside our mix to tame transients and
| | 00:03 | plosives, let's take a look at an
example where taming those nasty things
| | 00:08 | actually helps other dynamics
processors in our mix work better.
| | 00:12 | In this example, you can clearly see
that the signal's waveform has a few spots
| | 00:16 | that really stand out from the rest.
| | 00:18 | This signal is in need of compression
to even out the performance and sit it
| | 00:22 | firmly in the mix, but all those hot spots
may cause the compressor to work too hard,
| | 00:26 | since I would need to pull the
threshold down so far under where the transients
| | 00:30 | are sitting to grab the meat of the
signal that I really want to compress.
| | 00:34 | What I like to do in these situations
is insert a limiter before the compressor
| | 00:38 | to create a layered-compression signal flow.
| | 00:41 | The limiter will be set with a higher
threshold to catch and tame the hot spots,
| | 00:45 | while the compressor will work more on
the true average level of the signal.
| | 00:50 | Listen as I play and adjust the L1
limiter to tame just the peaks of the vocal
| | 00:54 | track. Watch the gain reduction
labeled Atten as it grabs the hot spots.
| | 00:59 | Again, I'm linking the threshold
and output so I get no makeup gain;
| | 01:03 | I'll add that later with the compressor.
| | 01:05 | (music playing)
| | 01:17 | And I'm just going to find a sweet spot that
just grabs those, right around -6.
| | 01:23 | (music playing)
| | 01:38 | Now I allowed my compressor to work
on the average or body of the signal.
| | 01:42 | I've already set the compressor up
with a good starting point for this vocal.
| | 01:47 | Notice that with the limiter engaged,
the compressor's gain reduction is working
| | 01:50 | more uniformly on each word.
| | 01:52 | (music playing)
| | 02:24 | Again, the limiter brings the
transients down and the compressor works more
| | 02:28 | evenly over the entire piece, not
compressing so hard on the peaks spots that
| | 02:32 | the limiter has been able to grab already.
| | 02:35 | This helps the whole compression signal
flow some more transparent and natural,
| | 02:39 | as no one processor is doing all the work.
| | 02:42 | This trick is especially cool when you
want to bring out the tonal character of
| | 02:46 | one compressor, say as slower tube
compressor, but need more extreme dynamics
| | 02:50 | control. Using a faster compressor or
limiter to grab the stuff that is really
| | 02:54 | moving around allows you to focus
the tube compressor more on the average
| | 02:58 | level of the track.
| | 03:00 | Most compressors have a point where
they begin to work too hard and they
| | 03:03 | start sounding iffy.
| | 03:05 | Layering your compression and
limiting in small amounts can really help
| | 03:08 | overcome these obstacles.
| | 03:09 | I've even seen mastering engineers
discreetly layering four to five compressors,
| | 03:14 | each with tiny amounts of gain
reduction, in very small ratios to create a loud
| | 03:18 | master without pushing any compressor
limiter in the signal flow too much.
| | 03:22 | So the next time you're doing a mix
| | 03:24 | try this layering trick on any
instrument of vocal that has a lot of transient
| | 03:27 | material that's confusing the
compressor, like guitar chicks or pick scrapes,
| | 03:32 | vocal Ps or Ts, or anything that
sticks out too far that might cause the
| | 03:36 | compressor to work too hard.
| | Collapse this transcript |
|
|
4. Special Dynamics ProcessorsUnderstanding and using de-essers| 00:00 | Sibilance is a term that usually
refers to the hissing effect produced when a
| | 00:04 | vocalist speaks or sings a hard S sound.
| | 00:08 | This S sound can be overwhelming and
irritating and could be especially bad when
| | 00:12 | certain singers are recorded
with less-than-optimal microphones.
| | 00:16 | Hi-hat cymbals and some other high-
frequency instruments can also create sibilance.
| | 00:21 | The challenge to eliminate this
unwanted sound is so prominent in the recording
| | 00:25 | world that a specific type of dynamics
processor was developed to tackle it,
| | 00:29 | called a de-esser.
| | 00:31 | In a typical compression or limiting
scenario we know that the processor looks
| | 00:36 | for and reacts to a signal's
amplitude going over the defined threshold.
| | 00:40 | It doesn't matter if that signal is a
low-frequency signal or a high-frequency
| | 00:44 | signal, a bass note or treble note.
| | 00:47 | Any amplitude value measured over
the threshold will cause the compressor
| | 00:50 | to react and reduce the gain as prescribed
by the ratio, attack, and release controls.
| | 00:55 | But what about special scenarios where
we're looking to attenuate the signal
| | 01:00 | only in a certain frequency range?
| | 01:02 | More specifically, what if we want to
engage the compression only when the
| | 01:06 | hard S of a vocal is detected,
but not anywhere else? That is the main
| | 01:11 | purpose of the de-esser.
| | 01:12 | Listen to the following example. Listen
specifically to how the S's sound in each example.
| | 01:18 | First we have a vocal track without a
de-esser on it. Notice the prominent S sound.
| | 01:24 | (Male speaker: There's a girl named
Sally who sells seashells.)
| | 01:29 | And now here's the same vocal
track with the de-esser engaged.
| | 01:32 | (Male speaker: There's a girl named
Sally who sells seashells.)
| | 01:41 | Technically, de-essers are frequency-
specific compressors because they're set up
| | 01:45 | to react only to the amplitudes in a
specific range of frequencies in a signal,
| | 01:50 | rather than reacting to all
the signal's amplitude uniformly.
| | 01:54 | This frequency-specific
compression is achieved by feeding a filtered
| | 01:58 | narrow band of boosted frequencies,
generally centered around 6k to 8k, where the
| | 02:02 | S sound lives, into the
compressor-detection circuit.
| | 02:06 | This makes the compressor much more
sensitive to those frequencies and thus
| | 02:10 | any loud S sounds will breach the compressor's
threshold and trigger compression on the signal.
| | 02:16 | This compression can be set to occur
over the entire signal, compressing all the
| | 02:20 | frequencies at once, or only over a
defined portion of the signal, such as the
| | 02:24 | higher frequencies above 5k.
| | 02:28 | De-essers generally have two major
controls: threshold and target frequency.
| | 02:32 | The threshold works exactly the same
as it does on a compressor or a limiter,
| | 02:37 | defining the point at which the de-essing
will kick in and reduce the sibilant sound.
| | 02:41 | Set it an appropriate level to
reduce the sibilance without drastically
| | 02:45 | affecting the other frequencies on the track.
| | 02:48 | The target frequency determines where the
de-esser will look for sibilance in the signal.
| | 02:53 | Set this frequency where the signal
is most sibilant and harsh sounding.
| | 02:56 | Your de-esser may have the option to
de-ess only the high frequencies on a track.
| | 03:01 | This option allows the de-esser to be
more transparent, as it only reduces the
| | 03:05 | level of the high-frequency
material passing through it, leaving the
| | 03:09 | low-frequency material unchanged.
| | 03:11 | De-essers are extremely useful tools
because they tame sibilance without having
| | 03:16 | to lower the overall high-frequency content.
| | 03:18 | They could be used on both
individual tracks and also entire mixes.
| | 03:23 | While many people think about using
them only on vocals, de-essers are
| | 03:27 | indispensable tool in mixing drum tracks
| | 03:29 | that are a little too bright in the wrong
areas or when we mastering an entire mix.
| | 03:34 | The beauty of de-essers is that they're
frequency-specific compressors that we
| | 03:38 | can tell to work only when
a problem area is detected.
| | 03:42 | Let's dive into an example of
de-essing in the next movie.
| | Collapse this transcript |
| Get in the Mix: De-essing a vocal track| 00:01 | We now know that de-essing is
frequency-specific compression;
| | 00:04 | let's put a de-esser to work on a vocal track.
| | 00:07 | It's time to get in the mix. Pause this
movie and open the appropriate file for your DAW.
| | 00:14 | Take a listen to this vocal passage,
paying particular attention to the S sounds.
| | 00:19 | (music playing)
| | 00:37 | Because these S sounds will likely get a
bit crispy as we add some top-end EQ to
| | 00:41 | the mix, especially on the
words asleep, chills, and spine,
| | 00:45 | we can use the de-esser to tame that sibilance.
| | 00:47 | I'd like to use my de-essers
before applying compression and EQ
| | 00:51 | so I can get the bad stuff out of the
way before it hits my other processors.
| | 00:55 | Now listen again as I activate the de-esser.
| | 00:57 | (music playing)
| | 01:16 | It's subtle but effective.
| | 01:18 | Using a high-frequency-only de-esser,
all I'm doing here is just taking a bit of
| | 01:22 | the edge off those particularly sibilant words.
| | 01:26 | Many de-essers allow you to preview
the side-chain signal and tune in the
| | 01:29 | frequency band to match the track,
| | 01:31 | so you might try selecting a
particularly sibilant passage to play back in a
| | 01:35 | loop and sweep through the frequencies
until you hear it get really nasty and resonant.
| | 01:41 | Setting up the target frequency is
key because you don't want to de-esser to
| | 01:45 | react to non-sibilant passages of the vocal.
| | 01:48 | In this case, I've set my target frequency to
6k, a good starting point for a male vocalist.
| | 01:54 | By dialing down the threshold, I can
choose the amount of gain reduction or
| | 01:58 | overall de-essing that I want to achieve.
| | 02:00 | I want to pay particular attention
during this stage, as too much de-essing will
| | 02:05 | create a lisp in the performance.
| | 02:07 | I've found that on vocals I'd like to use
split-band de-essing so that only the
| | 02:11 | high frequencies are
compressed during the sibilant sections.
| | 02:14 | This helps the de-esser sound more
transparent to my ears, and I can get away
| | 02:18 | with more gain reduction
without introducing a lisp.
| | 02:22 | Take a listen to this passage as I
dial in too much full-range de-essing,
| | 02:26 | giving the singer a lisp.
| | 02:27 | (music playing)
| | 02:46 | Now, unless I'm playing a cruel track on
the vocalist, I hope you could hear that I
| | 02:50 | don't want to abuse the de-esser like this.
| | 02:52 | Ultimately, you may find that no matter
how well you set your target frequency,
| | 02:56 | other non-sibilant material
will trigger the de-esser too.
| | 03:00 | This is normal. So use your ears and
find a sweet spot that does the best job
| | 03:03 | without triggering significant
compression on the non-sibilant passages.
| | 03:08 | One trick I like to use to evaluate
my sibilance is to listen to the mix on
| | 03:12 | the smaller speakers,
| | 03:13 | as they tend to exhibit harsh
sibilant sounds in more pronounced ways.
| | 03:18 | On particularly tough sections, I might
also use volume automation to reduce the
| | 03:22 | specific sections of extreme sibilance.
| | 03:24 | If you are constantly struggling with
overwhelming sibilance in your vocal
| | 03:28 | tracks, you might try re-recording the
track with a different mic, as certain
| | 03:32 | voices do not work well
with certain models of mics.
| | Collapse this transcript |
| Understanding and using gates| 00:00 | The next time you're at the
supermarket pay attention to how the automatic
| | 00:04 | doors open and close.
| | 00:06 | Walk towards the door, the motion detector
senses your presence and opens up to let you in.
| | 00:11 | Once you walk through, it senses you're
no longer there and closes automatically
| | 00:15 | to keep the heating and
cooling costs down. Simple, right?
| | 00:19 | Well, you've just experienced a real-life
version of the dynamics processor we call a gate.
| | 00:24 | A gate and its close sibling
the expander are extremely common
| | 00:28 | dynamic processing tools.
| | 00:30 | They're generally used to allow the
desired clean signal to pass through while
| | 00:34 | removing any unwanted low-level noise,
like hum from a guitar amp or bleed from
| | 00:40 | a track, like the sound of a headphone
mix that bleeds into a vocal mic during
| | 00:44 | a vocal recording session.
| | 00:46 | Take a listen to this drum track and notice
the bleed of other pieces in this kit.
| | 00:50 | (music playing)
| | 01:02 | Now listen with the gate in effect.
| | 01:05 | Notice how the kick sound is much
cleaner and the bleed is eliminated.
| | 01:08 | (music playing)
| | 01:20 | We use gates and expanders to remove
or reduce the bleed between notes or
| | 01:24 | phrases, which allows us to process a
signal in isolation without processing the bleed.
| | 01:29 | Here is another example. Listen to the
snare drum with compression and reverb
| | 01:34 | applied to it, but without a gate.
| | 01:37 | Notice how the compression and reverb
are drawing out the bleed of the other
| | 01:41 | drums in the snare track,
especially the kick in the hi-hat.
| | 01:43 | (music playing)
| | 02:01 | Now let's listen with the gate engaged.
| | 02:03 | The compression and reverb are allowed
to work cleanly on the snare signal in
| | 02:07 | isolation, without all that muddy
kick bleed or hi-hat exciting the reverb.
| | 02:11 | (music playing)
| | 02:29 | Gates were extremely popular in the past
because of the noise floor issues with tape.
| | 02:34 | Using extreme compression or limiting
would tend to bring up the tape hiss in
| | 02:37 | silent passages significantly,
| | 02:39 | so gates and expanders were used to
treat signals before heavy processing,
| | 02:44 | maintaining a reasonable signal-to-noise ratio.
| | 02:46 | A gate works a little bit like a
reverse compressor, in that instead of
| | 02:51 | attenuating signals over the threshold,
| | 02:53 | it actually allows them to pass through,
just like those doors at the supermarket.
| | 02:57 | Gates feature the same threshold, attack,
and release controls as compressors.
| | 03:02 | Think of the attack time as the
amount of time it takes the door to open up
| | 03:06 | when it senses motion, and the release time
is how fast it closes after you walk through.
| | 03:11 | Some gates feature a hold parameter
that allow the gate to remain open a
| | 03:15 | pre-specified amount of time, regardless
of whether or not the signal has fallen
| | 03:19 | back below the threshold.
| | 03:22 | The hold control is handy for keeping
the gate from overreacting or chattering
| | 03:26 | as it attempts to ride the wide
dynamic variations of complex signals.
| | 03:31 | Again, think about that automatic door.
| | 03:33 | It generally stays open for a little
bit just to make sure you get through,
| | 03:36 | before slowly closing shut.
| | 03:39 | If we look at a gate's transfer curve,
we can see that the output of anything
| | 03:43 | under the threshold is
infinitely attenuated, or gated,
| | 03:46 | while the signals above the
threshold are passed through in a linear
| | 03:50 | one-to-one fashion.
| | 03:52 | Now, just because I can, will I
gate every track that has bleed?
| | 03:57 | Not necessarily. Sometimes a little bit
of bleed or air in the signal can help
| | 04:01 | glue or mix together, and sometimes
getting the gate to trigger correctly just
| | 04:04 | causes more problems than it solves.
| | 04:07 | On certain tracks I'll opt
to edit out the bleed by hand,
| | 04:10 | removing the portions of the waveform I
don't want using my DAW's editing tools.
| | 04:15 | This could be a much cleaner way of
dealing with instruments that aren't
| | 04:18 | always playing, like toms.
| | 04:20 | Be extra careful not to
over-gate things like vocals.
| | 04:24 | You never want to cut off the
beginning and ends of your words unless you're
| | 04:27 | using it as an effect.
| | 04:28 | You don't always want to kill all the
breaths, as these give the track a human
| | 04:32 | quality that I like to
retain more times than not.
| | 04:35 | In the next movie, let's look at a
variation of a gate called an expander.
| | Collapse this transcript |
| Understanding and using expanders| 00:00 | Gates with ratio controls are called
expanders. Instead of completely cutting
| | 00:04 | off a signal below the threshold,
expanders simply reduce the volume of the
| | 00:08 | signal below the
threshold by the specified ratio.
| | 00:11 | In this case they're truly a reverse
compressor because they expand the dynamic
| | 00:15 | range of the signal instead of reducing it.
| | 00:18 | In an expander, the signals that
passed over the threshold are unaffected and
| | 00:22 | the signals that fall under
the threshold are attenuated.
| | 00:25 | This creates a greater dynamic range
between the softest and loudest parts of a signal.
| | 00:30 | Expanders are perfect for scenarios
when you don't want to kill the bleed but
| | 00:34 | just want to turn it down a bit or
separate it more from the main signal.
| | 00:38 | This can help give a signal a little
more bounce as the dynamic range between
| | 00:42 | the below and above threshold
parts is extended or expanded.
| | 00:46 | Listen to the snare through an
expander with a ratio of 2:1.
| | 00:50 | Notice how the bleed is
reduced but not eliminated.
| | 00:53 | (music playing)
| | 01:10 | Expanders, as well as gates, may also
feature a range control that allows you to
| | 01:14 | define the lower and upper end of the
threshold so that the transfer curve
| | 01:18 | remains linear both below and
above the range of gating or expansion.
| | 01:23 | This control gives you more
flexibility in separating signals from their
| | 01:26 | underlying bleed while
still maintaining some of it.
| | 01:30 | In the following movie, let's look
at how to apply a gate and expander to a track.
| | Collapse this transcript |
| Get in the Mix: Gating a drum track| 00:00 | Now that we know what a gate does and
understand its controls, let's jump into
| | 00:04 | an example of using a gate to
clean up some live drum material.
| | 00:08 | It's time to get in the mix. Pause this
movie and open the appropriate file for your DAW.
| | 00:15 | Listen to this un-gated snare track.
Pay attention to the bleed from the other
| | 00:19 | pieces of the drum kit.
| | 00:20 | (music playing)
| | 00:30 | In this case, I'll use a gate to reduce
the bleed from the other drums that have
| | 00:34 | a clean isolated snare signal to
further process with EQ and compression.
| | 00:39 | Most of the time I like to place my
gate's first in the signal chain so I
| | 00:43 | can get stuff I don't want out of a signal
before passing it on to any other processing.
| | 00:49 | Now listen to the snare track with
the gate engaged, noticing the dramatic
| | 00:53 | reduction in the bleed from the other drums.
| | 00:55 | (music playing)
| | 01:04 | Notice when the rest of the drum tracks
come in, the gate doesn't sound as dramatic.
| | 01:08 | This is a good example of
always checking things in context.
| | 01:12 | It doesn't matter where it sounds like soloed.
| | 01:15 | No one's going to hear the isolated
parts when you release the final mix.
| | 01:19 | Now listen as I adjust the gate from scratch.
| | 01:21 | I'm going to begin by raising my
threshold to a point where the signal that I
| | 01:25 | want to keep triggers the gate
open while the bleed remains gated.
| | 01:29 | This step will likely take a bit of
tweaking, so take the time and check the
| | 01:33 | threshold of various sections of the material.
| | 01:35 | (music playing)
| | 01:45 | After finding a nice threshold setting, I can
then begin to tune my attack and release times.
| | 01:50 | As you may have heard, the attack is
currently too slow and is shaving off nearly
| | 01:54 | all the transient or punch of the snare.
| | 01:57 | Also, it releases so fast that
it's choking off a lot of the decay.
| | 02:01 | Instead, I want the gate to open
quick enough that the transient of my
| | 02:05 | signal is not shaved off or softened,
| | 02:08 | so I'll use the setting
that is almost instant here.
| | 02:11 | For the release time it really depends on how
aggressive of a gated sound I'm looking for.
| | 02:16 | In this case, I want to maintain from
that decay of the signal as to keep its
| | 02:19 | sound natural, but not so much that I
pick up a ton of bleed after each strike.
| | 02:25 | In scenarios that require aggressive
gating, I can use a bit of short room
| | 02:29 | reverb to add back a bit of
artificial decay to the signal.
| | 02:33 | Listen as I adjust the attack and
release settings to get the sound that I want.
| | 02:36 | (music playing)
| | 02:55 | In situations where we have a lot of
intricate snare work--drags, roughs, and
| | 02:59 | little ghost notes--
| | 03:00 | it may not be possible to use a hard gate.
| | 03:03 | In this scenario, I usually have to
make a decision to either embrace the bleed
| | 03:07 | or try using a lighter expander
that retains some of the bleed.
| | 03:11 | One trick I like to use in this
situation is to duplicate the track I'm working
| | 03:14 | with, so I can hard gate one and do my
extreme processing in isolation but blend
| | 03:20 | back in the un-gated duplicate track to taste.
| | Collapse this transcript |
| Understanding and using multi-band compressors/limiters| 00:00 | Remember how I said a de-esser was a
frequency-specific compressor working only
| | 00:04 | on the sibilant sounds within a track or
mix. A multiband compressor follows the
| | 00:09 | same idea, but works across
the entire frequency range.
| | 00:13 | By splitting up the compression into
multiple parts, or bands, an engineer can
| | 00:17 | focus dynamic control within a
specified frequency range, leaving other
| | 00:21 | frequencies uncompressed.
| | 00:24 | Because of these multiple bands of
gain control, multiband compressors are
| | 00:28 | especially handy in situations where
only certain parts of the signal need
| | 00:31 | dynamics control, like cleaning up
the low-end resonance of a signal while
| | 00:35 | leaving the high-frequency content
uncompressed, or less compressed, or super-
| | 00:40 | compressing the top end of a vocal to
achieve that pop polish without all the
| | 00:44 | harshness that would come from using EQ only.
| | 00:48 | Mastering engineers will sometime
use multiband compression to tighten up
| | 00:52 | elements of a mix that didn't receive
enough compression during the mixing
| | 00:55 | stage, like a bass guitar that was
left uncompressed, creating unpredictable
| | 01:00 | low-frequency response from note to note.
| | 01:01 | While they might look scary at first
glance, a multiband compressor is simply
| | 01:06 | multiple single-band compressors, each
working on their own dedicated frequency range.
| | 01:12 | One compressor might be working on the
bass notes of a signal, while another
| | 01:15 | works on the mids and high-frequency content.
| | 01:18 | Each band will have its own
threshold, ratio, attack, release, and makeup
| | 01:22 | gain, and in most multiband processors, these
controls can be linked together for ease of use.
| | 01:27 | Instead of a single threshold
triggering compression, each threshold of a
| | 01:31 | multiband compressor looks for
amplitude within its specified band.
| | 01:35 | When the threshold of any band is breached,
| | 01:38 | the compressor will start to attenuate
the signal, but only within its specified
| | 01:41 | range of frequencies.
| | 01:43 | Just like a de-esser, the detection in
the compression is frequency-specific,
| | 01:47 | giving us total control over when
and how the compression is applied.
| | 01:52 | In this sense, you can think of a
multiband compressor as a dynamic EQ of sorts.
| | 01:56 | It attenuates a specific frequency band
just like an EQ would, but only when
| | 02:00 | certain dynamic conditions are met.
| | 02:03 | Let's listen to the same example we
worked on with our de-esser, only now, using
| | 02:07 | a multiband compressor instead.
| | 02:09 | In this example, only the offending
frequency band containing the sibilance has
| | 02:13 | been allowed to compress, while
the other bands are made inactive.
| | 02:16 | (Male speaker: There's a girl named
Sally who sells seashells.)
| | 02:22 | As you can see, and hopefully hear,
multiband compressors and limiters are
| | 02:26 | extremely powerful tools, giving the
engineer total control over both the
| | 02:29 | dynamic and frequency response of
a signal within the same processor.
| | 02:34 | But all this power comes
with added responsibility.
| | 02:37 | Multiband processors are among the most
misused and abused tools in the mixing world.
| | 02:41 | I can't stress enough how degrading
multiband compression can be when used
| | 02:45 | incorrectly across an entire mix.
| | 02:48 | While you'll find multiband
limiters built into nearly every all-in-one
| | 02:51 | mastering tool, pro mastering
engineers tend to use multiband compression
| | 02:55 | only as a last resort and only when
going back to the individual tracks of a
| | 02:59 | mix is not possible.
| | 03:02 | Be careful not to flatten your mix out.
| | 03:04 | I personally find multiband tools most
useful when processing individual tracks
| | 03:08 | or small groups, rather than an entire mix.
| | 03:11 | When I do use multiband compression on
my entire mix, I often disable most of
| | 03:15 | my bands to focus my work on a specific range
of frequencies, like the top end or sub bass.
| | 03:22 | Think of it this way: multiband
compression is just another tool in your toolbox
| | 03:26 | of creative ideas, not an
automatic solution for mixing or mastering.
| | Collapse this transcript |
| Controlling frequency content with multi-band compressors| 00:00 | Now that we have a good idea about why
and how multiband compressors work, let's
| | 00:05 | take a look at a specific example of
using multiband compression to control
| | 00:08 | frequency content in a mix.
| | 00:11 | Take a listen to this loop.
| | 00:13 | Notice how on Beat 1 of the loop we
have a super-bassy 808 drum that seems to
| | 00:18 | overtake the low end of everything else.
| | 00:20 | Now, this can be cool, but I think
the low end is a little too much,
| | 00:24 | especially if there's going to be
other low-frequency elements added to the
| | 00:27 | mix, like another kick drum.
| | 00:28 | (music playing)
| | 00:40 | Because this is a loop and I don't
have the 808 hit on an isolated track, I
| | 00:45 | can't simply turn down the 808. Nor can
I use EQ to treat the low end, because
| | 00:50 | that would affect the low
end of the bass synth as well.
| | 00:54 | What I can use is a multiband
compressor on the low frequencies of this loop to
| | 00:58 | see if I can get it to give me a
little gain reduction only on the downbeats,
| | 01:02 | when the 808 is strongest.
| | 01:04 | To accomplish this, I'll first pull up
my multiband compressor on the loop track.
| | 01:09 | I'm using the Waves C4 plug-in here.
| | 01:13 | I'll start by bypassing all of
the frequency bands except the
| | 01:17 | low-frequency band.
| | 01:20 | Now, I can start working on the
threshold setting to find the sweet spot that
| | 01:23 | only trigger serious compression on
those 808 hits, when the bass from the 808
| | 01:28 | and the synth really add up
and overpower my subwoofer.
| | 01:31 | (music playing)
| | 01:52 | Because I'm dealing with sub-bass here,
I want to make sure my release time is
| | 01:55 | not so fast that it causes
the compressor to distort.
| | 01:59 | I'm using a release of 45 milliseconds.
| | 02:02 | This is plenty of time to keep the
compressor from distorting on the
| | 02:05 | oscillations of low-frequency waveforms.
| | 02:08 | I can also use the Range control to
limit the total amount of gain reduction to
| | 02:12 | no more than a specified value,
regardless of the threshold interaction.
| | 02:16 | Certain compressors feature range or max
| | 02:19 | gain reduction controls that allow
us to use deeper threshold settings
| | 02:22 | while restricting the total amount of
gain reduction to the DB amount shown
| | 02:26 | on the range control.
| | 02:28 | This can help make sure the gain
reduction is triggered more evenly across all
| | 02:31 | sections, regardless of how soft or
loud the passage is, and help louder, more
| | 02:36 | dynamic signals preserve their power
and punch, receiving less gain reduction
| | 02:40 | than they would have otherwise.
| | 02:42 | I chose this specific example to
showcase how multiband compression can be used
| | 02:46 | to solve a problem when
separate tracks are not available.
| | 02:50 | If I had not been working with the
loop and had access to the individual
| | 02:54 | elements as separate tracks,
| | 02:55 | I would work with those individual
tracks, using level control automation, single-
| | 02:59 | band compression, and EQ, to get them
to add up to what I was looking for.
| | Collapse this transcript |
| Understanding and using transient shapers | 00:00 | By harnessing the power of a
compressor's attack and release controls, we can
| | 00:04 | achieve some really cool envelope
shaping on drum transients, helping to add
| | 00:08 | punch and clarity inside a dense mix.
| | 00:10 | However, envelope shaping
with a compressor isn't ideal.
| | 00:13 | While a compressor can draw out the
transient's attack, it will also heavily
| | 00:17 | compress the rest of the waveform, altering
the decay of the signal in ways we may not want.
| | 00:22 | Fortunately, we have another option.
| | 00:25 | Enter the envelope shaper.
| | 00:27 | This tool is designed to dry out the
attack or sustain of a signal without
| | 00:30 | affecting the other parts.
| | 00:32 | There are quite a few of these in the
market now, but the secret weapon that
| | 00:35 | most engineers associate with transient
control is the SPL Transient Designer.
| | 00:41 | The plug-in version I'm using here
is modeled after the famous hardware
| | 00:44 | version of the same name.
| | 00:46 | This unit only has two simple controls,
| | 00:48 | attack and sustain, allowing you to
increase or decrease the attack or sustain
| | 00:53 | portions of your signal
without affecting the other half.
| | 00:56 | Let's take a listen to this snare drum.
| | 00:59 | As I increase and decrease the attack,
listen as the bite of the snare comes and
| | 01:03 | goes, and watch the
corresponding output on the oscilloscope.
| | 01:06 | (music playing)
| | 01:23 | Now, I'll increase and decrease the sustain.
| | 01:25 | (music playing)
| | 01:47 | In this example, the Transient Designer
helps the snare have a little more bite
| | 01:50 | in the mix without changing its overall
tone or character, which a traditional
| | 01:55 | compressor would tend to do.
| | 01:56 | Listen to the snare and the context of
the mix, and notice how much bite it has
| | 02:00 | when I add the SPL
Transient Designer in the mix.
| | 02:03 | (music playing)
| | 02:30 | Using envelope detection, the Transient
Designer identifies both the attack and
| | 02:34 | sustain components of the signal
portion, and either raises or lowers their
| | 02:38 | level without adding any
artifacts to the rest of the sound.
| | 02:41 | Think of it like selecting a piece of
the snare drum's waveforms visually in
| | 02:44 | your editor and
automating its volume up or down.
| | 02:49 | One of the coolest things about
this tool is that it can work in the
| | 02:51 | opposite direction.
| | 02:52 | I can actually reduce the bite of a
signal's transient to soften it up in the
| | 02:56 | mix and make room for other things, or
I can reduce the sustain of a ringing
| | 03:00 | tom or kick drum that's clouding up my low end.
| | 03:03 | While these tools are typically used on
drums and other percussive material, get
| | 03:08 | creative and try them on other signals,
like spiking out a guitar or bass note.
| | 03:12 | You might be surprised at the result.
| | 03:14 | These tricks are difficult, if not
impossible, to do with just a pure compressor,
| | 03:18 | so I'm glad these special processors
exist, and now that you know how to use
| | 03:21 | them, I think you'll find them
especially helpful in your mixing workflow.
| | Collapse this transcript |
|
|
5. Advanced TopicsEffectively using side-chain inputs| 00:00 | Up until now, we've worked with the
compressor by feeding the signal we wish to
| | 00:04 | compress directly into the
threshold circuit of the compressor,
| | 00:08 | but what if you could control the compression
of one signal using the dynamics of another?
| | 00:12 | Listen to the swell of the synth
pad to the beat of the kick drum.
| | 00:16 | The kick drum is
triggering the swell on the pad.
| | 00:18 | (music playing)
| | 00:35 | We'll discuss how to set this up
specifically within your DAW in the next movie.
| | 00:40 | The technique of controlling one's
signal dynamics using another signal's
| | 00:43 | interaction with the compressor's
threshold is called side chaining.
| | 00:45 | Instead of the compressor listening
and reacting to the input signal, the
| | 00:50 | processor is told to listen to another
signal, known as the key, and base its
| | 00:54 | compression decisions on that.
| | 00:57 | One of the most common uses
of side chains is ducking.
| | 01:00 | Think about when a radio DJ talks over
the music he's playing and how the music
| | 01:04 | seems to automatically turn itself down,
much quicker than a human can react
| | 01:08 | with their hand on the volume fader.
| | 01:10 | This is a basic example
of side-chain compression.
| | 01:13 | By using the announcer's voice to
trigger compression on the music, the music is
| | 01:17 | compressed, or turned down,
whenever the announcer speaks.
| | 01:21 | Because no makeup gain is used, the
result is an automatic attenuation of the
| | 01:25 | signal whenever the
announcer chooses to drop in.
| | 01:29 | Most dynamics processors feature side
chain or key inputs, and most DAWs feature
| | 01:34 | an easy way to tap into their power.
| | 01:36 | Again, the concept to understand here
is that we're feeding only the detection
| | 01:40 | circuit of the dynamics processor with
another signal, and the side-chain signal,
| | 01:45 | or key, is not getting compressed, but
merely telling the compressor how to work
| | 01:49 | on the target signal.
| | 01:51 | Side chains can be used for utility
purposes, such as ducking and de-essing, as
| | 01:55 | well as for creative and
sound-design-oriented tasks.
| | 01:59 | In the next movie, we'll dive into a
specific example of using a side chain
| | 02:03 | for creative purposes.
| | Collapse this transcript |
| Using side chains creatively| 00:00 | What used to be reserved for the
underground electronic music scene has since
| | 00:04 | become a go-to technique
for pop arrangers everywhere.
| | 00:08 | Compressing a synth pad or a bass with a
side-chain kick drum is a sound that you can't mistake.
| | 00:13 | Let me show you how to set it up.
| | 00:14 | Listen to this
unprocessed kick drum and synth pad.
| | 00:18 | (music playing)
| | 00:28 | Now, listen as I enable the
side-chain compressor on the synth.
| | 00:31 | (music playing)
| | 00:43 | Here's how I set it up.
| | 00:44 | As with any side-chain workflow, we must
first define what will be receiving the
| | 00:48 | compression and what will act
as the key of that compressor.
| | 00:52 | In this case we want to compress or duck
the synth pad every time the kick drum hits.
| | 00:57 | Now, because the kick drum is plain
and nice even four-on-the floor beat, this
| | 01:02 | creates a cool breathing or pumping
effect on the synth pad, as if it was
| | 01:05 | swelling to the beat of the music.
| | 01:08 | I'll start by applying the
compressor to my synth pad.
| | 01:11 | In this case, I'm using the Waves C1.
| | 01:13 | Now, if I stopped here, I would simply
be compressing the synth pad using its
| | 01:18 | own signals to trigger the threshold.
| | 01:20 | But that isn't what I want.
| | 01:22 | I want to get the kick drum to talk
to the compressor on the synth track.
| | 01:26 | Now, this next step varies from DAW to DAW.
| | 01:29 | In Pro Tools what I need to do is use
a bus send to route part of the kick
| | 01:33 | drum's signal into the key or side-chain
input of my compressor on the synth track.
| | 01:37 | So what I'm going to do is remove this
bus that I've already set up and now I
| | 01:42 | can select from my Sends > bus, and
I'll use Bus 1 as my side-chain key.
| | 01:50 | Turn that up to around 0, to feed that Bus 1.
| | 01:54 | All dynamics plug-ins that support key
inputs will have a bus selector at the
| | 01:59 | top left-hand side of the plug-in window.
| | 02:01 | In other DAWs, like Logic, for example,
I might simply see a track selector
| | 02:06 | directly on the plug-in that allows me
to choose what track I want to key from
| | 02:09 | without having to set up any sends or buses.
| | 02:13 | Now that I have the side-chain
signal fed into the compressor, I want to
| | 02:16 | adjust the threshold setting during playback
so I can start to get a feel for the effect.
| | 02:21 | Some compressors might require that
I click a button to activate the key
| | 02:24 | input, while others, like this C1, automatically
detect the key and start using it immediately.
| | 02:30 | (music playing)
| | 02:49 | This is one case where I will not use
the makeup gain on the compressor, because
| | 02:53 | I want to gain reduction to attenuate
the signal to achieve that swelling effect
| | 02:57 | as the compressor releases.
| | 02:59 | Now I can tune the attack and release by
ear to get the right amount of lag time
| | 03:03 | between the kick hits and
the offbeat synth swells.
| | 03:06 | (music playing)
| | 03:37 | Now, let me switch out the
compressor for a gate so we can hear the
| | 03:40 | difference between the two.
| | 03:42 | Notice how the gate achieves the
opposite effect but is just as cool and usable
| | 03:46 | as a creative effect.
| | 03:47 | I've already set up the C1 gate with
the correct side-chain input and set the
| | 03:52 | attack, release, and threshold settings.
| | 03:54 | (music playing)
| | 04:13 | Sometimes I like to feed my side-chain
with a pre-processed version of the key
| | 04:17 | signal, living on a duplicate muted track.
| | 04:20 | For example, the processed key can be
filtered with EQ to cut out all the high end
| | 04:26 | of a busy loop so the
side chain just hears the low-end kick.
| | 04:30 | Remember, you don't have to actually
hear the signal play back in your mix;
| | 04:34 | it just has to be an effective
key for the side-chain compressor.
| | 04:37 | So experiment with this concept.
| | 04:40 | Generally, I like to use a fast attack and
tune the release to feel good with the beat.
| | 04:45 | Every compressor will work a
little bit differently, so experiment.
| | 04:48 | I can have better luck with faster
compressors that I can really dial in the
| | 04:52 | attack and release settings I'm looking
for, but sometimes compressors without
| | 04:56 | attack and release
settings can sound killer too.
| | 04:59 | At any rate, this is definitely a
situation where you want to let your
| | 05:02 | ears guide you.
| | Collapse this transcript |
| Keying gates and compressors (and/or ducking)| 00:00 | In post-production mixing or audio for
video, it's fairly common to use side-chain
| | 00:05 | compressors to duck music
and effects under dialogue.
| | 00:08 | This trick is especially handy when
deadlines won't allow for time-consuming
| | 00:12 | volume-automation passes.
| | 00:14 | Here's how it works.
| | 00:15 | Listen to this sample dialogue against
the unducked music, and notice how the
| | 00:19 | music is way too loud for the
dialogue to be clearly audible.
| | 00:22 | (music playing)
| | 00:25 | (Male speaker: This is an example of audio ducking.
My speech triggers compression on the music track, but when I stop--)
| | 00:34 | Instead of using automation to turn the
music down when the dialogue comes in,
| | 00:37 | I'll try a different approach
using the side-chain compressor.
| | 00:41 | Listen after I engage the side-chain compressor.
| | 00:45 | (music playing)
| | 00:48 | (Male speaker: This is an example of audio ducking.
My speech triggers compression on the music track, but when I stop--)
| | 01:01 | So, I start by applying a
compressor to the music or effects submix.
| | 01:04 | I'm using the C1 here.
| | 01:06 | Now, because I want the compressor
to react to the signal coming from the
| | 01:09 | dialogue, I need to get the dialogue
track to talk to the compressor on the
| | 01:13 | music track, using the side-chain input.
| | 01:15 | In Pro Tools, I do this by
creating a send to an internal bus.
| | 01:19 | That send will feed the compressor's side-chain
input, which is set to listen to the same bus.
| | 01:25 | In other DAWs, like Logic for example,
the key is simply selected from a track
| | 01:29 | list at the top of the plug-in window.
| | 01:31 | So, I'll choose a bus, Bus 1 in this case, and
I'm going to bring that level up to around 0.
| | 01:41 | I'm going to make sure that in my key input
I've set to that same bus Bus 1 in this case.
| | 01:49 | Once my side chain is set up and the
key is active, I can lower the threshold
| | 01:54 | until I start to get a bit of
attenuation on my music mix. Take a listen.
| | 01:59 | (Male speaker: This is an example of audio ducking.
My speech triggers compression on the music track, but when I stop--)
| | 02:07 | (music playing)
| | 02:13 | (Male speaker:--the music is allowed to
return to its normal volume.)
| | 02:17 | (music playing)
| | 02:20 | When using the compressor to duck music
under dialogue, I prefer to use longer
| | 02:23 | release times so the music doesn't
creep up between words, especially if I'm
| | 02:27 | doing a significant amount of ducking.
| | 02:30 | Release times of one to two seconds or more
in this case can really give the ducking
| | 02:34 | effect that smooth automation ride sound that
doesn't get choppy or cut into the dialogue.
| | 02:38 | Now, even though many compressors can
look ahead to anticipate the attack, they
| | 02:43 | don't look ahead far enough to
anticipate and duck the music in a subtle way.
| | 02:48 | But this is the beauty of the side chain.
| | 02:50 | We can tell the compressor to
work from any signal we want.
| | 02:53 | So, I might send the compressor a
signal of a duplicate track of the dialogue
| | 02:56 | that has been pulled earlier in
time by one to two seconds and key with that.
| | 03:01 | This earlier dialogue track can be
muted and not heard in the actual mix, while
| | 03:05 | still sending signals through
the compressor's side chain.
| | 03:08 | In certain DAWs like Pro Tools, you'll
need to use pre-fader sends to achieve this.
| | 03:13 | That way the track's signal still passes audio
at its send, but its main output is muted.
| | 03:17 | I've gone ahead and set this up already
in a hidden track called Dialogue Early
| | 03:23 | and I've pulled this back one
second so that it triggers early.
| | 03:26 | It's been muted with the send set to PRE here.
| | 03:31 | So, let's take a listen.
| | 03:35 | (music playing)
| | 03:38 | (Male speaker: This is an example of audio ducking.
My speech triggers compression on the music track, but when I stop--)
| | 03:45 | (music playing)
| | 03:51 | (Male speaker:--the music is allowed to
return to its normal volume.)
| | 03:55 | (music playing)
| | 03:58 | Again, ducking dialogue is a great
technique when automation rides would take
| | 04:02 | too much time or you need to affect
the work in real time with a live input.
| | 04:07 | Try this technique the next time
you're working on dialogue and music in
| | 04:10 | the same mix.
| | Collapse this transcript |
| Managing gain staging and headroom and fixing over-compressed tracks| 00:00 | As we've been discussing dynamic
processors and dynamic range, you may have
| | 00:04 | heard me throwing around terms
like headroom and gain staging.
| | 00:08 | Headroom is the amount by which the
signal-handling capabilities of an audio
| | 00:12 | system can exceed the average
working level of the system.
| | 00:15 | Headroom can be thought of as a safety
zone, allowing transient audio peaks to
| | 00:20 | exceed this average working range
without exceeding the signal capabilities of
| | 00:24 | an audio system--digital clipping, for example.
| | 00:27 | Another way to think about this is to
imagine the ceilings in your house.
| | 00:31 | Unless we're all NBA players, most of
us will likely walk around the house
| | 00:35 | just fine without hitting our heads on the
ceiling, and depending on the height of your ceilings,
| | 00:40 | there's probably a good two to four
extra feet of headroom in case you get
| | 00:43 | excited and want to jump up and down.
| | 00:46 | Headroom in the analog world allows
the signal breathing room before the
| | 00:50 | circuit starts to distort.
| | 00:51 | For example, in an analog mixing
console, or tape machine, think of headroom as
| | 00:56 | an amplitude range in which the
signal's input and output response remains
| | 01:00 | linear, that is, not compressed or distorted.
| | 01:04 | However, in many cases we like to push
the boundaries of this range to introduce
| | 01:08 | harmonic distortion or interesting
artifacts that can be quite ear-pleasing.
| | 01:12 | It is important to recognize that
unlike analog gear, in digital systems, this
| | 01:17 | gray area range beyond the
nominal headroom does not exist.
| | 01:22 | Digital systems remain linear in the response,
all the way to the clipping point at 0 dBFS.
| | 01:27 | While tape tube and other analog gear
might push back and introduce natural
| | 01:31 | compression to a signal as its headroom
is breached into distortion, a digital
| | 01:36 | system will simply clip the signal
hard, creating a nasty square wave out of
| | 01:40 | your waveform peaks.
| | 01:42 | Think of it as the difference between
having a soft layer of pillow on your
| | 01:46 | ceiling before reaching hard drywall
of concrete. Well in the digital domain
| | 01:50 | there is no pillow, just the hard surface.
| | 01:53 | Gain staging is the term used to refer
to the level at which you pass a signal
| | 01:57 | from one part of signal change to another:
| | 02:00 | from plug-in to plug-in, from
insert to mix bus, and so on.
| | 02:04 | Cranking up the volume in one
processor can leave you without any headroom to
| | 02:08 | move on to your next stage of processing.
| | 02:10 | If I ratchet up a snare's gain in a
compressor, throwing the transients up into
| | 02:14 | the extreme end of the DAW's headroom
range, and then pass that signal into an EQ
| | 02:19 | and boost 60 dBs on the top
end, what do you think happens?
| | 02:23 | Clipping, distortion, and all kinds of nastiness.
| | 02:26 | Some DAWs and plug-ins utilize higher
internal bit-depth processing, like 32 bits or 48 bits
| | 02:31 | that potentially provide additional
headroom and avoid clipping. But that
| | 02:35 | additional headroom still can't save
your signals from being cut off at the
| | 02:38 | converter at 0 dBFS.
| | 02:41 | Ultimately discussing the dynamic
peculiarities of every plug-in, mixer, and piece
| | 02:45 | of outboard is way beyond
the scope of this course.
| | 02:49 | Take away from this the simple fact
that level matters, and at some point in
| | 02:53 | your mix, there will be a finite
amount of dynamic range and headroom that
| | 02:56 | you have to respect.
| | 02:58 | Digital distortion can be very audible,
in which case, you should immediately
| | 03:02 | take action. But digital
distortion can also be very subtle,
| | 03:06 | something that adds up track to track,
something that creates an indescribable
| | 03:10 | haziness over your work that you
might not be able to put your finger on.
| | 03:14 | Manage your levels and I guarantee you,
your mix will sound better for it.
| | Collapse this transcript |
| Compression first or EQ first? | 00:00 | One of the most common questions I
get from my students is whether they
| | 00:03 | should EQ their tracks before or after
compression, and the answer I always give
| | 00:08 | them is, that depends.
| | 00:10 | I like to consider an addition-
subtraction approach when deciding whether or not
| | 00:14 | I want to EQ or compress first:
| | 00:15 | what I'm going to get rid of and
what do I want to play up or add?
| | 00:21 | In the case of subtraction, I think
about what part of the signal's frequency
| | 00:24 | response I did not like or what I'm going to
want to remove before it hits the mix bus.
| | 00:28 | I generally like to get rid of any
frequency material I don't want before I hit
| | 00:33 | my compressors or limiters.
| | 00:35 | That way the compressor's threshold
won't be triggered by material I'm going
| | 00:38 | to get rid of anyway.
| | 00:39 | For example, if I have a loop with a
ton of low end that I don't need, I might
| | 00:44 | using EQ to filter out all the bass
frequencies before hitting the compressor.
| | 00:49 | Since those bass frequencies
will likely make the most of that
| | 00:51 | signal's amplitude, they would
likely influence the compressor's
| | 00:54 | threshold in an undesirable way.
| | 00:56 | (music playing)
| | 01:12 | Sometimes what happens when you use a
lot of compression or limiting is that
| | 01:16 | signal's frequency response, or tonal
characteristics, can get a bit flattened
| | 01:20 | out, especially in the low and high frequencies.
| | 01:23 | In this case if I want to do additive
or boost an EQ, I might consider saving
| | 01:27 | that for after any compression, as a way
of restoring some of the tonal response
| | 01:31 | or shape to the signal--post-dynamics processing.
| | 01:35 | Don't confuse this discussion of EQ
before after compression with the order that
| | 01:40 | you approach them when starting to mix a song.
| | 01:42 | What I'm talking about here is the
physical order that your signal will take
| | 01:45 | from one processor to the next as it
travels in line through your inserts,
| | 01:50 | not whether you decide to add
compression to your track when you first open the
| | 01:53 | session, then add EQ thirty minutes later.
| | 01:55 | While I generally reach for whatever
processor that will take me in the right
| | 01:59 | direction, sometimes I find it
helpful to apply compression before I start
| | 02:03 | adjusting the EQ curve, as it helps
firm the track's dynamics up in the mix,
| | 02:07 | giving me a better sense of what kind
of EQ it's going to need to sit with
| | 02:10 | the rest of the tracks.
| | 02:12 | Otherwise trying to EQ a dynamically
wild track can be a bit like trying to
| | 02:16 | hit a moving target.
| | 02:18 | Remember, this adding or subtracting
ideology is just a framework that you can
| | 02:22 | use to think through your processing chain;
| | 02:24 | it isn't a hard-and-fast rule
that you have to follow religiously.
| | 02:28 | In many cases, the order just won't
matter all that much. Like if I'm doing a
| | 02:32 | few dB of gain reduction on vocal track,
I don't always have to use two separate
| | 02:36 | EQs before and after
compression to achieve my ultimate goal.
| | 02:41 | With DAWs and plug-ins, it's so easy
to play with the signal chain that it's
| | 02:44 | almost a no-brainer to try out different
approach, just to hear what they might sound like.
| | 02:49 | So experiment, listen, and think
about how the compressor might react to
| | 02:53 | different frequency material.
| | Collapse this transcript |
| Understanding mix bus compression| 00:00 | Mix bus or simply bus compression is
the practice of using a compressor on
| | 00:04 | a group of tracks sub-mixed together,
as opposed to only compressing each
| | 00:09 | track individually.
| | 00:11 | For example, I may use a compressor
on my kick, snare, toms, and overhead
| | 00:15 | individually, but I might also use a
compressor on the entire drum sub mix
| | 00:19 | to achieve an effect
| | 00:20 | that's not possible through adjusting
the individual compressors in isolation.
| | 00:26 | Bus compressors are generally associated
with the phrase "gluing the mix together"
| | 00:30 | and the pedigree has a
fairly interesting history,
| | 00:33 | dating back to the first bus
compressors installed in large analog consoles.
| | 00:37 | The A&R executives used to call the
bus compressor insert button the record
| | 00:41 | button because it instantly glued
the entire mix together and made the mix
| | 00:45 | sound like a record.
| | 00:47 | Take a listen to this mix before and
after bus compression through a plug-in
| | 00:50 | version of the famous SSL bus compressor.
| | 00:53 | Here is the mix before
adding the SSL bus compression.
| | 00:57 | (music playing)
| | 01:21 | And here is the mix with the
SSL bus compression active.
| | 01:25 | (music playing)
| | 01:50 | The change is very subtle
but significant nonetheless.
| | 01:54 | The overall goal with bus compression is
very similar to any other type of dynamics
| | 01:58 | control: by taming the peaks or the or
the transients, I can reduce my overall
| | 02:02 | dynamic range and pull up some of the
lower-level material to inflate the body
| | 02:07 | of the track and give it a little
extra push in average loudness.
| | 02:11 | The bus compressor also grabs hold of
any straggling transients that might jump
| | 02:15 | out of the mix too far, gluing the
elements of the mix together better.
| | 02:20 | Many engineers like to mix through
their bus compression, placing it on the
| | 02:24 | master bus before starting their mix.
| | 02:26 | This can help them mix quicker by
influencing their processing decisions as
| | 02:30 | everything passes through the
compressor. Because everything is getting
| | 02:34 | compressed at the master bus, you tend to
use less compression on individual tracks.
| | 02:39 | This technique can work well if
you're an experienced mixer and know what to
| | 02:42 | listen for and how to tweak the
mixed-bus compression to changes in your
| | 02:45 | individual track processing.
| | 02:48 | This can also work against you if
the volume levels of your mix change
| | 02:51 | radically over the course of the mixing process.
| | 02:54 | If you don't update your bus
compressor's threshold, you may end up hitting it
| | 02:58 | too hard, which in turn may cause you
to make less-than-ideal decisions about
| | 03:02 | the processing of individual tracks.
| | 03:05 | Personally, I like to use a slow
attack and a fast release setting, and never
| | 03:09 | higher than a two-to-one or four-to-one ratio.
| | 03:12 | Most times the idea is not to compress
the entire mix very hard, just to control
| | 03:16 | the dynamics gently in a transparent way.
| | 03:19 | So my advice is, if you're going use a
master bus compressor over your entire
| | 03:23 | mix, use a light touch.
| | Collapse this transcript |
| Get in the Mix: Using mix bus compression| 00:00 | Now that we understand why bus
compression is so powerful as a mixing tool,
| | 00:04 | let's get hands on with an
example and get in the mix.
| | 00:08 | Pause this movie and open the
appropriate file for your DAW.
| | 00:13 | Take a listen to this full mix
with the bus compression bypassed.
| | 00:17 | (music playing)
| | 00:30 | It's sounds pretty good, but I want
to get just a little more lift out of
| | 00:34 | the lower-level details and tuck in the
transients a bit to glue everything together.
| | 00:39 | Now listen with the bus compression enabled.
| | 00:42 | (music playing)
| | 00:55 | Here I'm using an attack of 5
milliseconds to allow some of the transients to
| | 00:59 | punch through and a release of 200
milliseconds, or about an eighth note in this
| | 01:03 | song, based on the tempo.
| | 01:05 | Generally, I only get no more
than a DB or two of gain reduction,
| | 01:09 | just enough to tie things together and
give me that lift and tuck on the whole mix.
| | 01:14 | In this example, I exaggerated my
threshold settings just a bit to make
| | 01:18 | that effect more audible, but in a
real mix I might go back and ease off
| | 01:22 | the gain reduction.
| | 01:23 | While mixed-bus compresses are great for
gluing a mix together, they can often do
| | 01:29 | more damage thing good when pushed too hard,
| | 01:32 | flattening your mix and
sucking all the life out of it.
| | 01:35 | Listen to this example of
too much bus compression.
| | 01:38 | (music playing)
| | 01:52 | Hopefully you heard the pumping and
breathing of the compressor digging into
| | 01:55 | the mix too aggressively.
| | 01:56 | While certain styles of music take
advantage of this aesthetic, it is generally
| | 02:01 | a sign of too much bus compression, so
be extra careful with processors that
| | 02:04 | affect your entire mix, especially
when it comes compression, as it's very hard
| | 02:09 | to undo later in mastering.
| | 02:12 | Many times I'll use no master bus
compression in a mix or only bus compress
| | 02:16 | specific groups of tracks.
| | 02:18 | It all depends on the style of music and
what I think will benefit the song most.
| | 02:22 | For instance, in electro music, the pumping
effect over the entire mix as part of genre.
| | 02:29 | The four-on-the-floor kick drum
drives the compressor into gain reduction
| | 02:32 | on every beat, causing the compressor to
swell and breathe to the beat of the music.
| | 02:37 | Tweaking the attack and release settings
of the bus compressor can really change
| | 02:40 | the attitude and transparency of this effect.
| | 02:43 | To create this effect, try using slower
attack and faster release times with a
| | 02:48 | decent amount of gain reduction.
| | Collapse this transcript |
| Get in the Mix: Working with parallel compression| 00:00 | Parallel compression is the process of
combining an uncompressed version of a
| | 00:04 | signal with the compressed
version and blending to taste.
| | 00:07 | This best-of-both-worlds approach
preserves the dynamics, openness, character,
| | 00:12 | and frequency response of the
unprocessed signal, while solving the issue of an
| | 00:16 | overly dynamic track getting lost in the mix.
| | 00:19 | When a compressed waveform is
combined with an uncompressed waveform, the
| | 00:23 | result is purely additive.
| | 00:25 | The peaks of the uncompressed
waveform are preserved, while the body of
| | 00:28 | the signal is raised.
Now let's see this in action and get in the mix.
| | 00:32 | Pause this movie and open the
appropriate file for your DAW.
| | 00:38 | Let's take a listen to these uncompressed drums.
| | 00:41 | (music playing)
| | 00:49 | Now listen to the uncompressed drums
inside a dense section of the mix.
| | 00:53 | Notice how they have a little bit
of trouble punching through the mix.
| | 00:56 | (music playing)
| | 01:08 | The problem here is that the drums
get lost in the denser parts the mix,
| | 01:12 | the chorus in this example,
where more stuff is going on.
| | 01:15 | Simply turning them up and push up
the transients and eat up headroom.
| | 01:20 | Compressing them will bring them out
more but will also eat up the transients,
| | 01:23 | causing them to lose a lot
of their punch and impact.
| | 01:26 | This is a scenario where parallel
compression is extremely useful.
| | 01:30 | By applying compression to a
duplicate track or duplicate submix of the
| | 01:34 | material, I can overcompress the
duplicate and bring up its volume to fill
| | 01:38 | out the original uncompressed track.
| | 01:40 | I will start by adding a duplicate track and
apply aggressive compression to this track.
| | 01:46 | Take a listen to the
compressed drums on their own.
| | 01:48 | (music playing)
| | 01:59 | Now listen as I bring in the
parallel compressed drums underneath the
| | 02:03 | uncompressed drums with the mix.
| | 02:05 | Once it's in, I will mute it out for a
measure so you can really hear how much
| | 02:09 | it's supporting the drum sound in the mix.
| | 02:11 | (music playing)
| | 02:34 | Notice how the sustain of the track
becomes a bit inflated, allowing it to cut
| | 02:38 | through the dense sections of the mix better.
| | 02:41 | This trick works on almost any kind of
material where you want the benefits of
| | 02:45 | compression without a lot of artifacts.
| | 02:47 | Try it on vocals, guitars,
drums, or even entire mixes.
| | 02:52 | Try automating the level of the
compressed track up and down throughout a mix.
| | 02:56 | For example, turn up the volume on
the compressed track to really drive the
| | 03:00 | drums hard into the last chorus of the song.
| | 03:04 | Whenever you're creating parallel
processing chains like this, you want to be
| | 03:08 | extra sure your DAW's automatic delay
compensation is enabled and functioning,
| | 03:13 | because the separate plug-in chains
may create mismatched latencies in the
| | 03:16 | parallel tracks. Even a sample of
difference will result in very nasty comb
| | 03:21 | filtering as the two recombine.
| | 03:23 | Most DAW's delay compensation is always on,
| | 03:26 | while in Pro Tools, you need to
explicitly enable it in your session under
| | 03:30 | the playback engine.
| | 03:32 | Many compressors and limiter plug-ins
and even some hardware processors feature
| | 03:36 | Wet/Dry mix knobs that allow you to
achieve parallel compression inline,
| | 03:40 | without breaking the
signal off into another chain.
| | 03:43 | This avoids any potential
problems due to latency and makes the set-
| | 03:46 | up ridiculously easy.
| | Collapse this transcript |
| Working with "modeled" vintage compressor/limiter plug-ins| 00:00 | The old is new again, at
least as far as compressors go.
| | 00:04 | In the last decade, it seems a year
hasn't passed that some company hasn't
| | 00:07 | introduced a new and improved
plug-in model of a vintage compressor.
| | 00:12 | Aside from the company simply needing
new products to sell, computers have also
| | 00:15 | gotten much faster over the years, so
the models could be more CPU-intensive and
| | 00:19 | thus recreate every detail of
the original with more precision.
| | 00:24 | The good news is that new vintage model
dynamics processors have never sounded
| | 00:28 | more like their analog counterparts.
| | 00:30 | I'm going to show you how to use two the
most common vintage-modeled compression
| | 00:34 | plug-ins: the Universal Audio
1176 and the Teletronix LA-2A.
| | 00:39 | There are at least five companies that
I can think of at the top of my head
| | 00:42 | that have modeled these classics into plug-ins.
| | 00:45 | And some will not feature the exact
same name, due to licensing reasons, but as
| | 00:49 | soon as you see the interface, you will
definitely recognize the similarities.
| | 00:53 | Here, I am using the Waves CLA
compressors as an example of the UA 1176 and LA-2A.
| | 00:58 | The 1176 is a classic example of a FET,
or Field Effect Transistor, compressor.
| | 01:05 | These circuits were originally
designed to emulate tubes, but achieve a
| | 01:09 | completely unique sound all the round.
| | 01:12 | FET compressors are extremely fast,
clean, and reliable, and sound good on almost
| | 01:16 | any kind of signal material.
| | 01:17 | The 1176 itself has a very bright and
present quality to it, almost like using an EQ.
| | 01:24 | It can really bring out the presence on
anything you use it on, especially vocals.
| | 01:29 | The 1176 features a fixed threshold, so
to control it's gain reduction, you must
| | 01:33 | drive the threshold with the input control.
| | 01:35 | Once you've achieved the desired
amount of gain reduction, you can use the
| | 01:39 | output control to return the
output to a reasonable level.
| | 01:42 | Watch and listen, as I adjust the
input to drive the threshold and use the
| | 01:45 | output to make up any gain reduction.
| | 01:48 | (music playing)
| | 02:18 | The attack and release controls on the
1176 can be especially confusing because
| | 02:22 | they are not marked by millisecond time
values, but simple numbers one through
| | 02:26 | seven, with seven being the fastest
setting and one being the slowest.
| | 02:31 | This confuses many people because
intuitively you think that a larger number
| | 02:35 | would be slower when you are used
to measuring in actual milliseconds.
| | 02:38 | The 1176's attack time is extremely fast.
| | 02:42 | A setting of one is still only around
800 microseconds, while a setting of seven
| | 02:47 | is virtually instant, 20 microseconds.
| | 02:50 | The Release time sits over a much
wider range of 50 to 1200 milliseconds, so you can't
| | 02:56 | really visualize the two in the same way.
| | 02:58 | The 1176 features selectable ratio
controls plus an All buttons in control
| | 03:04 | that introduces a radically
different compression curve,
| | 03:06 | perfect for heavy compression, that you
really have to play with and hear to appreciate.
| | 03:11 | Take a listen.
| | 03:13 | (music playing)
| | 03:36 | Introduced the mid-1960s, the
Teletronix LA-2A is a famous electro-optical
| | 03:41 | compressor that uses an electro-
luminescent panel, basically a small light and a
| | 03:45 | photo optic sensor, to apply gain reduction.
| | 03:48 | As the signal's amplitude increases, the
light panel gets brighter and the photo-
| | 03:52 | optic sensor reacts to this by
applying more gain reduction to the signal.
| | 03:57 | Because of the inherent lagginess in
this design, this type of compressor tends
| | 04:01 | to work more on the average level than
the peaks, making it perfect for less
| | 04:05 | attack-driven instruments like
bass, vocals, guitar, and submixes.
| | 04:09 | As an added benefit, the tube make up
gain can be used to excite additional
| | 04:13 | harmonics that help fatten up the
signal, helping push low- and subfrequency
| | 04:17 | content into the more audible low mid
band, making this compressor sound great
| | 04:21 | on bass instruments.
| | 04:23 | Sometimes, I like to use the LA-2A
without any gain reduction just to tap into
| | 04:27 | the tube gain stage.
| | 04:29 | The LA-2A is super simple to use and a
great example of the knob minimalism of
| | 04:33 | classic analog gear.
| | 04:35 | Simply increase the peak reduction
control until the desired amount of gain
| | 04:38 | reduction is achieved.
| | 04:40 | Then use the gain control to make up any lost
level or drive the compressor's tube amplifier.
| | 04:45 | Listen as I adjust the peak reduction and gain.
| | 04:47 | (music playing)
| | 05:19 | Unlike the input gain control on the
1176, the gain control on the LA-2A does
| | 05:24 | not affect the compression,
| | 05:26 | so it can be used stand-alone if you'd
like, to simply drive the tube amplifier
| | 05:30 | and grab some extra harmonic
distortion and warm up the signal.
| | 05:34 | The limit-compress switch
changes the characteristics of the
| | 05:37 | compressor's transfer occur.
| | 05:39 | When in the compressed position, the
curve is more gentle and presents a
| | 05:42 | low-compression ratio.
| | 05:44 | A higher compression ratio results when
the switch is set to the limit position.
| | 05:47 | No matter what models of these classics
you happen to have access to, I'm sure
| | 05:52 | you will find them just as
indispensable as engineers have found their analog
| | 05:55 | counterparts for decades.
| | Collapse this transcript |
| Building healthy compression/limiting habits| 00:00 | I hope you've seen and heard how powerful
dynamics processors can be when used effectively.
| | 00:06 | They have the ability to make tracks
come alive, providing power and strength.
| | 00:10 | But the same tools that can make a
track stand out can also destroy it.
| | 00:14 | Too much compression can take the life
out of otherwise lively, brilliant tracks,
| | 00:18 | making them sound weak and dull.
| | 00:20 | Here are a couple overall concepts to
keep in mind when deciding how much is
| | 00:24 | too much compression.
| | 00:27 | Just as a compressor can exaggerate a
signal's transient response, making it
| | 00:30 | sharper or snappier,
| | 00:32 | it can just as easily take this away.
| | 00:34 | Remember, loud is only relative to
quiet and hard is only relative to soft.
| | 00:39 | If you push too hard, you can actually
flatten out a track's transient response,
| | 00:43 | making it sound flat and lacking punch.
| | 00:46 | Dynamic differences are what
make sound and music move us.
| | 00:50 | They're what make the speaker cones move and
our chest thump to the beat of a loud woofer.
| | 00:54 | You want to control your
dynamics, not obliterate them.
| | 00:58 | Remember, when increasing a
signal's average loudness and reducing its
| | 01:01 | dynamic range, you're making a sacrifice
between total perceived loudness and punchiness.
| | 01:07 | When you use a brickwall limiter
to maximize the level of your mix, be
| | 01:11 | careful not to squash all the impact
out of your drums just to pick up some
| | 01:15 | extra average level.
| | 01:17 | A compressor pushed to the extreme
will in effect flatten out a track's
| | 01:21 | frequency response,
especially in the low and high end.
| | 01:24 | This is why it is common to add
EQ to a signal post-compression.
| | 01:29 | But there's a fine line between post-
compression EQ touchup and totally needing
| | 01:33 | to rebalance your signal's tonal
curve due to overcompression.
| | 01:37 | Every compressor sounds different.
| | 01:40 | Don't be surprised if the exact same
parameters on one compressor don't work on
| | 01:44 | another, and don't be surprised when one
setting that works on your guitar track
| | 01:48 | today doesn't work on the
next guitar track you record.
| | 01:51 | Always approach each situation with a
fresh mindset, willing to experiment and
| | 01:56 | try different things.
| | 01:57 | Presets are great starting
points, but use your ears.
| | 02:00 | If it sounds bad, tweak it.
| | 02:02 | If it still sounds bad, take the compressor off.
| | 02:06 | The most transparent form of
compression is often a track's volume control and
| | 02:10 | your DAW's automation.
| | 02:12 | Don't be afraid to use the
waveform's amplitude display and draw in your
| | 02:15 | own dynamic control.
| | 02:17 | Understand compression's aesthetic use versus
its utility use and use that to your advantage.
| | 02:23 | Extreme compression can be amazing
when used in the right context, but
| | 02:27 | sound amateur when not.
| | 02:29 | It's not uncommon to use too much
compression when you're first starting out,
| | 02:32 | simply because your ears are not
used to hearing small changes in dynamic
| | 02:36 | range. This is normal.
| | 02:38 | Embrace the learning
experience and listen, listen, listen.
| | 02:42 | Check your work against your favorite
mixes and seek constructive criticism
| | 02:46 | from peers and mentors.
| | 02:48 | My best advice is to experiment, a
lot, and find what works best for you in
| | 02:53 | your style of music.
| | 02:54 | By listening and experimenting with
your own tracks, you'll be better able to
| | 02:58 | sonically identify what the
norms are in your favorite genre.
| | Collapse this transcript |
|
|
ConclusionGoodbye| 00:00 | Thanks for watching the course!
| | 00:02 | If you're hungry for more, check out my
other courses in the Online Training Library.
| | 00:06 | For more information about me
and the projects I'm working on,
| | 00:09 | visit brianleewhite.com.
| | 00:11 | Cheers, and see you next time!
| | Collapse this transcript |
|
|
An Interview with the AuthorA session with Brian Lee White| 00:00 | (music playing)
| | 00:07 | My name is Brian Lee White.
| | 00:08 | I am a mixer, producer, and
educator, based in Oakland California,
| | 00:14 | and I like to make records and
show other people how to do the same.
| | 00:22 | I grew up in a musical family. My dad
played lots of instruments, played in bands,
| | 00:26 | so I've been playing
instruments since I can remember.
| | 00:30 | I got into computers at a really young age
too, computer's and technology and the Internet.
| | 00:34 | So to me, mixing kind of really
incorporates this left-brain, right-brain, sort
| | 00:40 | of very artistic right and aesthetics as
well as highly technical and analytical
| | 00:47 | components, and kind of fusing the two
together into this kind of middle-brain
| | 00:52 | spot, to really kind of achieve the
best of both worlds, and deliver something
| | 00:57 | that's really special.
(music playing)
| | 01:03 | I originally got into teaching sort of
as a way to supplement my income in what
| | 01:09 | is really a hard industry.
| | 01:12 | Ultimately, I really grew to love
teaching, you know, being able to share my
| | 01:17 | experiences, share my tips and tricks
with everyone, and I kind of created this
| | 01:25 | kind of nothing-is-secret approach.
| | 01:28 | I came to that point because I would
have students and I would have friends
| | 01:32 | come to me and they would say things
like, "Yeah, you know, I heard this really
| | 01:36 | cool thing that my buddy was doing and
he told me, 'I am not going to show it to
| | 01:41 | you because that's my trick.'"
| | 01:43 | I would hear that and I would
that's just say, "That's ridiculous.
| | 01:47 | That's not the mentality that we need to
have in this community of people making art."
| | 01:53 | And I would immediately go and then
show them exactly what they wanted to
| | 01:59 | learn how to do and then some.
The real value of a mixer, or a producer, a
| | 02:03 | songwriter is what's in here, those
instincts that tell you, you know, not how
| | 02:10 | to do something, not how to turn a
knob, but why you would turn that knob.
| | 02:15 | I think ultimately the mix has to
serve the song, right, it has to serve the
| | 02:21 | emotion of the song, and ultimately when
I'm working for somebody, I'm providing
| | 02:26 | a service, so I ultimately want to
achieve the goal of the songwriter and the
| | 02:32 | producer, sort of where they want
to go with their aesthetic, right.
| | 02:36 | And that could be both
artistically as well as commercially.
| | 02:43 | What I like personally in a great mix
is I tend to gravitate towards mixes that
| | 02:50 | really make a strong statement.
| | 02:51 | Right, so, where the artist and
producer and the mixer have all come together,
| | 02:57 | they're all the same page, and
they kind of say, let's go for this.
| | 03:01 | We are going to go out there. This
could be a polarizing aesthetic, this could
| | 03:06 | be the sounds and the textures and
the way we are presenting them, not
| | 03:11 | everyone might get.
| | 03:12 | You know, we are going to make a really
gutsy move on this, and we are going to
| | 03:15 | make an artistic statement.
| | 03:18 | Those are the kind of
things that really excite me.
| | 03:23 | As a mixer of course,
sonics are a big thing for me,
| | 03:26 | how things sound. Is it
clear, was it recorded well?
| | 03:31 | While that is my job, I think it's
important for me as a mixer and other mixers,
| | 03:37 | all the greatest mixers, what they
recognize is sort of what draws people to a
| | 03:44 | piece of music or a piece of art.
| | 03:48 | It can be totally disconnected
to with how it sounds sonically.
| | 03:52 | You can see it all over the place, right.
| | 03:55 | You see people listening to songs off
their cell phone speaker and just loving
| | 04:00 | it, just digging it, right?
| | 04:02 | You know, listening stuff off YouTube.
It's been recompressed thirty times and it's
| | 04:07 | no different to them.
| | 04:08 | They are just, they are getting it,
| | 04:09 | right. The emotional connection is not lost.
| | 04:13 | People are really stoke on music, and
it makes them happy, and even in very
| | 04:19 | bandwidth-limited presentations where
the sonics are just horrible, it still
| | 04:24 | makes them happy, and they are still
getting 99% of the same enjoyment out of it.
| | 04:28 | And so, I need to take that to heart
and use those concepts to really serve
| | 04:35 | the project I'm working on and try to
present those so that that enjoyment is
| | 04:41 | maximized, no matter what speaker
system or compression algorithm is being used
| | 04:46 | to put that out there.
(music playing)
| | 04:54 | One of my most important job as a
mixer is to really own the aesthetic of the
| | 05:01 | song and the genre and really
present that to the listener.
| | 05:05 | So whether it's a crazy Garage Rock Punk
Tune, that's going to have all kinds of
| | 05:10 | distorted vocals and drums, or some
super-clean R&B that's got a ton of
| | 05:15 | low-frequency and a ton of
high-frequency extension.
| | 05:18 | You know, for me, I
really want to own that genre.
| | 05:22 | I want to own that aesthetic, and
yes, fit it within sort of some genre
| | 05:27 | expectations, and kind of really push
that forward so that listeners of that
| | 05:31 | genre--it's not a super big stretch for
them to kind of take that and run with
| | 05:37 | it--and I think that really makes my job
super fun, because I know, I really get
| | 05:42 | to wear all these different hats and to
paint with all of these different colors
| | 05:47 | from one day to the next.
| | Collapse this transcript |
|
|
