Diagnosing Common Problems (Why does mine not sound like a

commercial track?)
Sounds small, not loud enough. Turning the mix up or down does not solve this
problem (doesnt add clarity)
Sounds dull, other recordings are warm and deep, yet bright and open, all at the
same time. Boosting EQ in the mix will sound harsh and noisy.
Instruments/Vocals sound thin and lack sense of fullness that your favorite
recordings have (compressors leave a mix sounding low dynamically)
The bass doesnt have enough punch. You boost it with some low end EQ, but now
it just sounds louder/muddier, not clearer and punchier
You can hear all instruments in a mix, and they all are panned relatively well, but
the overall image still sounds wrong (not enough width and image)
You had reverb on individual tracks, but it sounds like instruments in their own
spaces instead of one drum kit in the same space.
A Mastering Engineer Has Three Personalities
Ms. Fix It someone who can recognize a problem with a recording. Too much bass,
too little treble, too much dynamic rangewhatever the problem, this person will
work to rebalance things so they work better
Mr. Make-It-Better hot shot who knows how to add that little extra pinch of spice,
sparkle or fullness or depth (take a good mix to great)
Ms. Nuts and Bots the science nerd who makes sure everything is done right. No
technical flaws, least amount of processing
Mixing vs Mastering
Dont mix and master in one step (youre tempted to try to mix, master, arrange,
and maybe even re record within the same session
Separation of mastering is very important, and you primarily focus on the overall
sound of the mix instead of changing patches (dont focus too much of your work on
a single instrument)
Mastering Effects
Compressors/Limiters/Expanders used to adjust dynamics of the mix, may need a
multi band dynamic processor to adjust specific frequencies of the overall mix
Equalizers used to shape the tonal balance
Reverb overall sense of depth to the mix, in addition to the reverb that may have
been added in mixing/arrangement

Stereo Imaging can adjust the perceived width and image of the sound field
Harmonic Exciters add edge or sparkle to the mix
Limiters/Maximizers increase the overall level of the sound by limiting the peaks to
prevent clipping
Dither provides the ability to convert higher word length recordings (24 or 32 to 16
bit for a CD) while maintaining dynamic range and minimizing distortion
Preferred Order of Mastering
EQ Dynamics Post EQ [Harmonic Exciter][Stereo
Imaging][Reverb]Loudness Maximizer
Processors that are less frequently used are in brackets
Loudness Maximizer and Dither should always be at the end of the chain
as an iron-clad rule
*READ IZOTOPES DITHERING GUIDE AT WWW.IZOTOPE.COM/OZONE/GUIDES
To change the order in Ozone, click the Graph button. This brings up the display of
modules, which you can drag around
Do not just blindly begin to go through the process. Identify things that you want to
change and then identify the tool or technique that would be best to make that
change.
Remember to take breaks and listen to other CDs/songs to refresh your ears in
terms of what other stuff sounds like (even seasoned pros will take a moment to
listen to familiar recordings to recalibrate themselves during a session)
Listen on other speakers and systems (burn different masters on a CD and see
which one translates the best)
Check how it sounds in mono this cant be stressed enough, a good ratio
between mono and stereo (correlated to uncorrelated) is very important in many
listening contexts and even vinyl cutting and MP3 creation
Monitor at around 85 dB SPL (C-Weighted). How loud is that?
Turn up your monitors until you can still have a conversation with someone who is a
meter away without having to strain your voice.
Remember to go to bed at night and forget about your master, wake up in the
morning, and be critical. Try to trick yourself into believing youve never heard the
track before.
Use insight to add even more detail to the visual feedback about your audio

Whats the Goal of EQ When Mastering?

We are shooting for tonal balance. Instrument specific EQ has been done during the
arranging and mixdown stages, so were just trying to shape the overall sound into
something that sounds natural. (EQ for both subtle correction and sonic
enhancements use very subtly)
Analog EQ sounds better while making slight boosts or cuts, while the
transparency of digital phase filters is useful when applying surgical
corrections/cuts
Use Undo History to quickly audition between different EQ filter types on the
fly
If you hold down the alt key and click oh the spectrum, you reveal an audio
magnifying glass that lets you hear only the frequencies under your mouse cursor,
without effecting your EQ settings
Useful for pinpointing the location of a frequency in the mix without messing
up your actual EQ
Can set the width of this filter in the options dialog adjust the alt solo filter Q
Useful workflow to alt-solo and find frequency problems, then double click exactly
where the mouse is to create a band at that EQ
FFT Size greater FFT size, greater frequency resolution (4096 is usually good
choice, could go higher for better resolution when dealing with lower frequencies
Too muddy? Try cutting between 100 and 300 hz with a parametric EQ , try cutting
the gain in this region a few DB
Too nasal-sounding? Try cutting between 250 and 1000 Hz
Too Harsh Sounding? Can be caused by frequencies in the range of 2000 to 3500
Hz, try cutting a few dB. Use alt click feature in this part especially (youll get the
most natural sound using relative narrow bands when cutting but wider bands when
boosting they are more musical)
Every song will feel like a problem mix, the art of mastering is addressing
the challenges that each individual mix presents
Mastering Quick Tips
Try cutting bands before boosting others
Cutting or boosting more than 2-4 dB means you probably have a problem
you cant fix from the stereo master, go back to the multi track mixing step
Use as few bands as possible
Use gentile slopes for boosting (wide-bandwidth, low Q) and narrow bands for
cutting
Shelf or highpass filters below 30Hz, it can get rid of low frequency rumble

and noise, but it comes with a price (do not compromise bass for rumble and noise
elimination)
Use your ears and your eyes, compare to other mixes using both senses
If you catch yourself using too narrow of a notch filter, you may be trying to
fix something that the EQ on the stereo mix cant fix
Check Your ABCs
A is reference music that you think sounds good in your room
B is the original mix, are you making it better than the original
C is the new version of your track, is there anything you are doing that is
revealing something unattractive
D is any other tacks/songs that will be part of the same collection does your
new version of the track fit with the others?
TIME AND PRACTICE IS THE TRUE ANSWER
Dynamics
Not just about making things louder and more competitive, so what is the primary
aim?
A consistent listening experience is one of the desirable effects of a good master.
Mostly, the listener does not want to have to constantly reach for the volume
control between the verse and a chorus of a song, or even between songs on an
album
Can help reduce/expand the dynamic range as needed, which helps empower the
listener to enjoy instead of be conscious of level changes
Additional sonic enhancements can occur by highlighting certain frequency
elements or instruments within a mix, or the reverse, smearing transients together
for a gluey and tighter sound
Principles of Dynamics
Effect is subtle if done correctly. Not something you clearly hear, but instead
something that changes the character of the mix. Compression removes something
(dynamic range), so you will hear the absence of something
Use the level histograms and compression meters in Ozone to reference when the
compression is occurring
Not all compressors are created equal. Applying a quality compressor correctly can
smooth the peaks and valleys in your mix to make it sound fuller, smoother, or
allow you to increase the average level (if thats the desired goal)
When you want the pop of a drum to go through, you would set a slower attack to
let the pop through before you compress the other spectral components
Starting point for release could be somewhere around 250 ms, there is no rule here

Power that comes from the compressor is that you can use it to gently compress the
middle or average level of the mix (limiter takes care of peaks, you can add glue
and body to the mix with a low ratio bringing the volume together), pull the
threshold for the compressor down at points where you see 1-3 db of gain reduction
as a good start
Use the multiband dynamics tool to:
Set different attack times for different frequency ranges low frequency signal, such
as fundamental of bass instruments, may take tens of milliseconds to complete one
cycle, whereas high frequencies may take 3-4. With multiband processing, you can
tailor the compression to control each range of frequencies to taste (wont hear the
compressor eat up too much bass energy, maybe decide not to touch the hi hats,
etc)
If you need more low end, but dont want it to get too loud, you can compress and
then boost (EQ is much more suited to tonal control, but multiband can be just what
you need in some situations)
General Dynamics Processing Tips
Set your ratio depending on what youre trying to compress:
Full mix 1.1 to 2.0 values (club tracks may benefit from higher ratios)
Bass, Kick 3.0 to 5.0 values depending on what sound youre shooting for
can go up to 10
Vocals 2.0 to 3.0. Rough suggestion, can change to vocalist.
Bring your threshold down until its just above the average level of the mix (level
histogram) aim for 2-3db of compression at max
Turn up output gain as you see fit to boost the compressed signal
Experiment with attack and release time, no good single tip. Shorter attacks level
off transients, but cause distortion (lowest attack possible without hearing any
artifacts)
Note: If youre just trying to increase the level of the mix a little without
changing the internal dynamic structure, you could use a limiter like the
Loudness Maximizer instead (later)
Loudness Maximizer
Principles of Limiting
Principle behind a limiter is that it limits the peaks at the threshold and then
brings up the rest of the mix afterwards (bulk of the mix can be brought up since
outlier peaks are cut down nothing will overload!)
Tiny bit of limiting is almost unnoticeable, beyond our perception to notice in
subtle applications

Threshold level at which limiting begins. As you do this the sound will tend to
get louder, due to the make up gain that is applied and directly related to the
threshold setting
Compare to original in level matched way
For subtle limiting, bring threshold down past the crest of incoming signal
histogram. Reasonably strong levels in a mix can be remedied by 1-3 db of limiting
A Margin setting of -0.3 to -0.6 will be appropriate as a final output level for your
mix (depending on how much processing will happen after Ozone.
Five Selectable Algorithms for Limiter Mode
Soft uses margin (final output level) as a guide but not as a fixed limit for the
output level. Provides a soft, natural limiting effect at the expense of allowing the
level to exceed the margin
Hard (formerly Brickwall) uses margin point as absolute guide, final output level
will not exceed this point (both hard and soft algorithms provide a natural analog
limiter effect choice is related moreso to behavior of the limiter and output level
than sound quality)
Intelligent (IRC I, II, III) Algorithm provides intelligent digital loudness maximization
of the signal, unlike the previous two analog styled limiters, designed for neutral
(transparent) limiting. Analyzes source material and applies limiting in a
psychoacoustically pleasing manner (reacting quickly to transients to prevent
pumping and slower to steady bass tones to prevent distortion). I uses the least
CPU, and III uses the most.
When using the IRC limiters, the release slider is replaced by character slider since
release it controlled by intelligent algorithms
Intersample Detection when using IRC, select this to allow Ozone to intelligently
predict the behavior of the analog signal reproduced for the listener, and to prevent
clipping from occurring in the analog domain
General Maximizer Tips
Intelligent (IRC) works better for transparent limiting, hard mode for analog limits
with a fixed threshold and the soft limiter for gentile boosts.
Do not set the Margin above -0.3 dB. Technically, you can set the margin to 0 DB
but it will maximize output to the point of clipping.
More aggressive loudness maximizing (lower threshold) will require long release
times generally speaking
ALWAYS listen to the music before and after the limiter in a level matched way, just
to be sure you are not causing unwanted artifacts checking the Automatically

Match Effective Gain box in Ozone preferences will automatically volume match any
A/B listening comparisons
Digital look ahead limiter is a very powerful tool, because you can do extreme
loudness maximation and some of the distortion or artifacts will be initially hidden
from the listener. The only question is if this is something you want or not. More
loudness = less dynamics (emotional highs and lows always a trade off)
Reverb
This comes into play when reverbs dont create the most cohesive sense of space, it
sounds like they are in their own reverb spaces and not together. However, a little
mastering reverb can add overall depth to the sound and push these instruments
into the same reverb space.
Fixes troublesome decays or room modes. Coat of reverb to the final mix can
smooth out imperfections in the original acoustic space.
Short reverb can add fullness to the mix and improve bass intelligibility. (not
impression of reverb, but creating short reverb at low level to fill sound stage)
Sometimes mixes are cut too short, and the mastering engineer doesnt have a long
enough audio tail to do the required fade out slight application of reverb at this
point can help extend natural decay of last chord of music, for more comfortable
fade out opportunities
How to classify mastering reverb Studio Effects and Acoustic Simulations
Acoustic reverbs simulate a realistic acoustic space. Clearly hear early reflections
from the original signal echoing off the walls, decaying into a space with later
reflections (clear sense of positioning)
Studio reverbs are not as natural, have effect of their own, can give overall sheen or
lush ambience to a song
Easy to overdo it and wash your mix down the drain, but just a touch can wash
away imperfections in original mix and give it a nice sheen
Mode there are room based reverbs (acoustic based) and then there is plate
mode, which provides lush, smooth sound
Pre Delay sets amount of delay in milliseconds between the original signal and
the beginning of the reverb (first bounce off walls)
Decay Time Acoustic sense, this controls overall size of room (decay time of
reverb)
If trying to wash over a mix, probably try values in the range 0.8 to 1.2 for
this fader

If the mix has reverb on individual tracks (which it probably does) set the
decay time or length of reverb to slightly longer than the reverb of the original
tracks (wet between 5 to 15 Dry at 100)
Another interesting effect to play around with is using small room size (0.25
0.65), and turn the Wet to 20-30 Dry 100. Can create fuller sound by adding
short reverb (doubling) the mix. Can also make some mixes sound terrible!
Keep reverb width at 100, spreading out an extremely short reverb is not natural
sounding
Greater High Decay values will result in brighter sounding reverb (lower values are
duller)
Many people use Ozone with Low Decay and High Decay settings around 1, but
experiment with these sliders if youre looking for unique reverb effects.
Mastering Quick Tips
In most cases, youll want the width from 100 to 150
As you turn up the width, you perceive more reverb. At higher room widths,
try turning down the Decay Time may seem counter intuitive but give it a listen
(turn it to extreme of 200).
High and Low Cutoffs
Mastering reverb has a spectrum, roll off reverb signal at different frequencies
Can drag nodes to left or right to change the bandwidth of the reverberated signal
that is returned and mixed back into your mix (area under bright red line is reverb
signal you hear)
Dont need to roll off high frequencies for yourself, because Ozone does this already.
However, rolling off highs can remove the tinny ness of reflections
Tend to start with low cutoff of 80 Hz, and high cutoff around 3-5 kHz.
If sibilance is too much, we can move the high cutoff down to around or
below 2 kHz
Reverb Tips
Reverb is easy to overdo. For mastering, use wet amount of 6-8% or lower, and
something more in the region of 2-3% is more appropriate
Bypass the mastering reverb to get a reality check on what the dry world sounds
like (sensed more than heard, if possible)
If you want more reverb, you have multiple options. Increase wet amount, or
increase decay time, or increase room width. Use history window to see which was
most effective out of these four

If youre applying a wide reverb (room width between 100 and 200), keep an eye on
the phase and use mono switch to check to make sure it doesnt completely fall
apart in mono
Stereo Imaging
Spatialization of your mix is like a sonic picture, comprised of the overall panning
and spatial placement of individual components within the stereo mix. In
mastering, it is the manipulation of the perceived image of a mix to enhance the
listening experience. Sometimes this means widening a mix, but can equally mean
narrowing a mix to solve certain problems.
Stereo imaging is a tough task only used specifically and often very gently.
Difficult to get a cohesive mix that still has a sense of space and imaging. (Over
application makes it more difficult to image. Small amount might taste good, but
any more and thats all you taste. Can help recenter bass that otherwise cause
phase and mono compatability problems (particularly when audio is played back via
mono sound sources.
Most pop/rock/hip hop have the most important elements (vocals and drums) in
center pan. Use a stereo widener to emphasize other elements of the mix, to widen
them around the middle panned instruments.
Vectorscope this allows you to see if the left and right channels are uniform or if
there is phase difference (widening). This is useful for seeing different bands and
their phase by soloing different bands in the earlier stages of mastering (dynamics
processor)
It has multiple operation modes:
Lissajous plots per-sample dots on a traditional oscilloscope display stereo
recordings produce a random pattern (taller than wide)
Polar Sample plots per sample dots but uses a polar coordinate display that is
more useful in highlighting the stereo image of incoming signal (Look for signal
outside the 45 degree safe lines which is out of phase audio)
Polar Level See Polar Sample (use 45 degree rule), length of rays on polar
coordinate represents amplitudes
General Stereo Imaging Tips
Generally can do more widening of higher bands (without lower bands) without
destroying mix clarity
May even want to try narrowing lower bands to pull bass to center if it is too wide
Monitoring on headphones is going to give you a false impression of the imaging of
your mix (imaging needs to be done in speakers (headphones will always feel wider)

Check mono capability

Ozone supports automation, which allows you to change values of controls over a
course of a mix one trick is to automate the widening of a mix (widening the mix
on a chorus then tightening it back up for the verse)
Stereoize can be used to give life to old, vintage mono recordings

Harmonics Exciter
Whats the Goal of Harmonics Exciting?
Add power, punch, and other subjective terms like warmth and brightness.
Distortion in small amounts, such as a harmonics exciter uses, is actually quite
appealing.
Saturating tubes exhibit a type of harmonic distortion that is described as warm or
musical. If used with care, distortion creates additional harmonics that add
presence or sparkle to the mix while preserving natural characteristics. Boosting
high frequencies with an EQ is not going to achieve the same effect (turns up
existing harmonics instead of creating new ones as an exciter does)
Tape saturation provides similar effect, more odd than even though (tube has more
harmonics an octave apart musical, while tape is slightly more aggressive
excitation harmonics are a fifth apart)
Principles of Harmonics Exciter
Easy to overdo an exciter, so remember these tips:
Compare it to commercial mixes. In some cases these are overdone as well, but
depends on genre.
Live with an excited mix for awhile. At first listen an exciter is exciting but over
time it can be fatiguing and harsh.
Higher bands can usually bear higher amounts of excitation
If you click on oversampling, it engages more processing power. It greatly
increases accuracy and sound quality of analog modeling in the exciter. Engage this
tool before bouncing and rendering.
Right click on the multi band spectrum, and click learn to automatically analyze
your audio and determine the most appropriate and transparent places to place
crossover points

General Harmonic Exciter Tips

Excitation of the upper bands will usually give the desired effect, but tube emulation
can be achieved on low band mids as well. (Try small amount of equal excitation
across all bands keep amount low and work in single band mode)
Dirty bass effect by applying excitation to the low band (if youre looking for more
bass level, use an EQ or dynamics, but if youre looking to add grunge on those bass
harmonics, sometimes you can use the harmonic exciter. Check out iZotopes Trash
for multiband distortion fun
Tape excitation mode provides bright saturated tone, tube model provides a thicker
midrange rich tone; retro model provides a heavy saturation character
Warm is similar to tube, unique in that it generates only even harmonics that decay
quickly can use warm setting to drive tube style saturation quite hard, without
creating mud in the low mid-range
Electronic mixes that are comprised of more soft synths and samples than acoustic
instruments can sometimes sound lifeless use harmonic exciter to change this
Compare RMS to peak level, for electronica you should have -8 to -12 db compared
to 0dBFS, -10 to -14 for RnB and Pop, -12 to -16 for rock, and -14 to -20 for jazz,
classical, folk, etc
Have the spectrogram running while working in the EQ module to help
focus. Also useful to have one running at the beginning and end of your
mastering chain to keep tabs on what is happening and how your original
mix has been altered
Vectorscope/Correlation/Stereo Image Metering
Most underappreciated aspect of metering, not usually discussed at the beginning
of learning to master
When you are mastering, you want to be sure the main instruments in a mix dont
disappear when you listen in mono a correlation meter gives you visual feedback
so you can be sure the recording has strong orientation toward mono does not
spend too much time with strong orientation toward out of phase information
Vinyl, MP3 encoding, and radio have heavy reliances on mono signal being
preserved and out of phase information will create unpleasant artifacts.
When cutting to vinyl, maintain transient peaks (use little or no limiting in
mastering)
MP3/AAC invariably add to signal level due to intense processing required by the
creation of the audio file. Peak level of source WAV/AIFF file might need to be
lowered by 1 to 1dB before encoding.

Broadcast broadcast chains usually include significant dynamic range processing

(avoid hyper compressing or limiting since theres another stage coming)
Flipping the phase from left channel to right and right to left, and then summing to
mono, lets us know how much energy is in the mid signal vs the side. If you are
processing via a compressor, you may not have enough gain or threshold range to
properly compress both signals leave the side alone as it prevents too much
pumping
M/S EQs can bring the track up to complement a too loud vocal, slightly brighten up
the side signal to enhance stereo width perception, raise the mid if the kick and
vocal need support, take mud out of a live mix by rolling off some low mids in the
side be careful, if you take too much, it starts to sound less live!

Four Bands of Multi Band Processing

Lows 20 to 120 Hz (meat of bass instruments and kick drum)
Low Mid 120 Hz to 2 kHz (fundamentals of vocals and midrange instruments
(warmth of mix)
High Mid 2 kHz to 10 kHz (cymbals, upper harmonics of instruments, sibilance of
words, treble of mix)
Highs Absolute upper frequency range, extending from 10 kHz to 20 kHz
(perceived as air)
Mastering Quick Tips Regarding Automation
Try using stereo automation on stereo widening. Widen a chorus and tighten a
verse
Try changing compression ratios between choruses (hard/high ratio) and verses
(soft/low ratio)
Try increasing the exciter amount on light/quiet parts, back off on the amount
during louder/fuller parts