INTERNAL MIXING

Defining the objective of the mix:

The object of the mix is a warm, clear, deep and pumping sound which all events appear in clear definition.

DIMENSIONS

• Dimension 1: Panorama/Panning - Widening the stereo sound basis

• Dimension 2: Frequency distribution
1st aspect: Use of EQ
2nd aspect: Measuring the levels
3rd aspect: Compression
• Dimension 3: Depth - Reverb and Delay, EQing of reverb and delay

PROCESSING IN GROUPS
Compression in groups to save resources and beef up the sound:
Particularly popular are plugins that are full of character like 1176 or LA2A. It's better to compress
frequently and gently rather than rarely and hard. Working with compressors in groups has a lot of
similarities to working with large analog consoles.
TIP: If there are a lot of individual drum tracks a good compressor is applied to most important tracks such
as Snare and Bass drum, and the group compression is then used to weld the other individual tracks together
with the snare and bass drum to a single entity.
The same applies to lead vocals, doubled tracks and ad libs

EQing groups to save resources and create space:

It's much easier to create space by using EQs on the spectral level (dimension 2) rather than EQs on lots of
individual channels. TIP: For example, if the lead vocals require more space in the frequency range for
warmth, that is, the lower mid range up to 350 Hz, this can be easily and effectively remedied by using group
EQs. A good graphic analyser in group solo mode allows to quickly unmask the keyboard pads (in the case
from the example in the video) which creates a fuzzy sound in the vocal range through overlaps and
counteract EQ.
Correction occurs particularly frequently in the lower mid range misery range between 120 Hz to 350Hz.
This range is pre-destined to overlaps and a fuzzy sound as almost all instruments are represented in this
area.
TIP: It's also possible to use group EQs to increase the trebles of a vocal group in order to increase speech
recognition and to convey the impression that the group in question occupies more of a front position within
the mix.

Widening the stereo image in groups

It's much better to apply basis widening effects to groups than to the total.
TIP: If the number of instruments in a particular frequency range is too great and if it's difficult to
differentiate them clearly, stereo expansion offers an opportunity to create a new dimension or at the very
least widen the horizontal dimension.
TIP: Stereo expander allows to swap high choirs and move them to the extreme left or right. Using stereo
expander on the choirs does a much better job of embedding the lead vocals into the choirs
For guitar walls of sound, you can widen the wall by transferring some of the layer guitars to a separate
group in order to insert a stereo spread.

MIXING WORKFLOW
Define the mixing strategy with a panorama sketch.
Start building up the rhythmic backbone, starting with the bass drum, followed by the snare, making use of
panning, EQing, compression, gates and hall, until the folder track "Drums" represents a powerful and
rounded sound.
Next step is to build up the instruments which provide harmony and warmth. Distribute the elements
according to their complimentary spectral properties to the left and to the right in the panorama. Either in
group solo mode or in conjunction with the "open" drum group.
Create a good lead vocal sound and add it to the centre.
Balance the group levels of all groups edited so far.
Distribute decorations and additions in a spectrally sensible manner around the existing basis.
If a particular event sounds fuzzy, look for a spot within the 3 dimensions where it can be clearly heard. If
you cannot find the spot either with a good panning strategy or with EQing or layering, reconsider the reason
for having this particular event where it is, and possibly mute it or look for a better spot elsewhere in the
song.
Fine-tuning of volumes at extremely loud or quite play volumes.
With a little bit of routine and experience, after 3 to 4 hours the mix is 90% ready. The static mix provides
the starting point for the hard part of the work which follows. 4 hours to finish a mix, the remainer is for fine
tuning and it's the most time consuming, usually 1 or 2 days
In order to shape the final 10 %, you need a healthy level of perfectionism. But don't do this by yourself,
team up with a colleague in order to generate the dynamics required to reach higher targets.
Volume automation for introducing new events.
Volume automation for song structure dynamics
Panorama and stereo expander automation for clearing up the last remaining fuzzy spots.
Carry out further automation.
Creative fine-tuning to refine details.
Experiment, experiment and experiment again in order to improve individual events that do not yet sound
right.
Set the brickwall limiter in the master section to -0.3dB in order to ensure slight and consistent peak limiting.
Export the master with a bit of lead-in and lead-out and without fade-out in 32-bit floating point resolution.

DIMENSION 1: PANORAMA
A good plan for the panorama is the basis of a good mix.
Start the mix by drafting a mixing strategy in which you set out the placement of individual instruments.
Almost like placing them on a stage.
The most important aspect in planning the panorama outline is 3-dimensional thinking.

After placing Snare, Bass Drum, Bass, Vox, Mid tom in the centre, now we look at the 2nd dimension
(Frequency Distribution) because the centre is already crowded with so many events.
The lowest 1½ octaves in the centre are exclusively occupied by the bass. This is not a place for any other
instrument.
Above that is the bottom sector of the bass drum at about 80Hz to a 100Hz. The head creates localization
with the clapping sound. Depending on the bass drum sound, it has quite a broadband presence in the upper
mid-range to 6kHz. Because this section of the bass drum sound is characterized by transients and occurs in ?
rhythmic sequences, this range doesn't generally interfere with the upper mid range of the lead vocals.
Above the bass drum are the toms which we'll ignore here because of the panorama distribution and their
sporadic occurrence.
Next instrument in the centre is the bottom sector of the snare at a 120Hz to 260Hz. Because of it's rhythmic
transient character, the snare coexist peacefully with the lead vocals without any spectral competition.
The main body of the lead vocals sets in at the lower mid range "misery range" responsible for the warmth of
the sound, and from then develops upwards to the higher regions.
The range around 1kHz is responsible for the nasal sound whereas the speech comprehensibility for the male
voice begins at about 2.5kHz and slightly above 3kHz for female voices.
S sounds and similar consonants are located between 6kHz and 8kHz, above that the shinny uppernotes.
The snares broadly occupy the area from the upper mid range to the treble areas.
The only range so far with the potential cause for concern is the lower mid range because it's also occupied
by the bass drum and the bass. However, because this range is less important to bass drum and bass it's
possible to counteract this slightly with parametric filters resulting in a slight thinning of the range between a
180hz to 250Hz.
The distribution strategy becomes more complicated the more instruments and events there are to be
distributed and the more inconsistent their appearance. In this situation, separate panorama outlines for the
various parts. If necessary, instruments switches sides in subsequent parts if this is what's required by the
instrumentation. In complex pop music there's no detrimental effect associated with instruments switching
sides within the panorama. However, any unnecessary panorama switches should be avoided in more
intimate music with consistent instrumentation.

To underpin the panorama strategy follow these rules:

The acoustic rule that deep sounds spread in circular form and can hardly be detected below a 100Hz,
whereas a high frequencies spread directionally and are easy to detect gives rise to the following panning
rule:
Choirs, orchestrations and general instruments in pop productions are generally placed in such a way that
deeper sounding instruments are always placed more towards the centre. The higher the sound event the
farther out there likely to be placed.
Within the choirs group you'd best place the higher voice further to the sides in the panorama and the deep
voices more towards the centre.
The same applies to string on samples, orchestrations and all many voiced sound events. Of course there are
exceptions to this rule depending on the preferences and artistic freedom.
The art of a skilful panorama distribution is to pay due attention to the 2nd Dimension (Frequency
Distribution) which is not part of the outline and that at least 3X as much.
The spectral distribution across the centre is a slightly variable standard and as far as left and right are
concerned it makes sense also to be aware of the spectral distribution. Distribute events in such a way that
the frequency distribution is also balanced in the outer sectors.
If a mix is very complex, you can approach the spectral distribution in a series of slices such as "outer left"
(100% Left) , "9 o'clock to 10:30" (20% to 60% Left), "centre" (0%), "1:30 to 3 o'clock" (20% to 60%
Right), "outer right" (100% Right) in order to find a place for each event. In this instance using the stereo
expander can create an extra dimension

SUMMARY OF THE PAN RULES

Create an outline of the panning strategy prior to mixing, taking into account the frequency distribution in
left, centre and right sectors.
Anything that is not bass, bass drum, snare or lead vocal has no place at the centre.
Instruments present in the same or overlaying frequency sectors should be placed at opposite ends
(complimenting each other) within the panorama.
Once the basic static panning is established, but still there are parts within the song that are unsatisfactory in
their transparency, it is always worth to try to remedy this with panning automation.
Well planned and carefully automated panning often creates greater clarity in the mix than the use of EQ and
should be preferred to unnecessary EQing.
If an event is drowned in the sound mush, the first step in trying to find a solution should therefore be to go
for the panning knob, and only after that for EQ.
Used in a controlled fashion, widening the sound basis can create extra space in the horizontal dimension and
thus ensure a clearer sound. Check using the correlation metre and the mono switch.
If the mix collapses in mono, there are probably too many deep sounds distributed across the panorama or
there are too many phase interferences

DIMENSION 2: VERTICAL = Frequency Distribution

1ˢᵗ aspect: Use of EQ

A very important step in this has already been accomplished with the distribution of events in the panorama,
which already takes the frequency characteristics that individual instruments within the left, centre, right
range into account.
If the signals are delivered in perfect shape, it suffices to use EQs sparingly specially in the lower cut range
in order to remove DC offsets and low frequency artifacts such as breathers, popping sounds and so on.

Frequency List
Vertical axis = the half steps. Horizontal axis = octaves.
If the signals are precisely tonely defined you can use this list for the exact settings of EQ frequencies.

Frequency Bands:

Sub-bass Bass Low mid (mid Mid Upper mid Trebles Upper
range misery) trebles
0 to 25Hz 25 to 120Hz 120 to 350Hz 350 to 2kHz 2 to 8kHz 8 to 12kHz 12 to 22kHz

Frequencies between 0 and 100Hz are called DC offsets and also have to be eliminated.

BASS (20 ~ 120Hz)

Together with the Low mid range (mid range misery), the bass sector is crucial for the perception of
warmth.
The bass sector is first and foremost the sector for the bass.
The deepest note of a 5 string bass "B" has a frequency at 30.9Hz. If the bass drum has a centre frequency at
the bottom of 90Hz, this means that there are almost 1 ½ octaves available exclusively for the bass.
Provided the bass drum isn't too low and there's hardly any interference as a result of only sparring use of the
low cut filter, this range provides enough scope to manipulate the bass independently even in the finished
mix.
Bass dead spots: these are individual notes played on imperfectly made instruments which sound quieter
than other notes because of instrument construction errors. With the aid of the frequency list you can
equalize these dead spots accurately by accessing them precisely using a parametric EQ with steep correction
in order to make sure that surrounding notes are unaffected.
The bass drum should be somewhere between 75 and 100 Hz, or 30 and 150Hz.
For club music, depth is not the same as pressure. In clubs, pressure develops at around 90Hz because the
system cannot powerfully transmit bass drums that are much lower. The lower a bass drum, the more
difficult it is to edit it. Bass drums that extend to 60Hz or deeper have to be carefully adjusted to the key of
the song. Deep bass drums tend to have their own tonality which runs the risk of interfering with the bass and
causing the sound to appear washed out.
Another rule for dealing with very deep sound events is: the deeper, the shorter.
Cut deep events short so that they do not take up too much space with the mix and lock the mix from the
start.

LOW MID "Low Mid Misery" (125 ~ 350Hz)

This range is the second pillar for the warmth of a song but also a potential source of unpleasant misery
You should pay particular attention to this range during mixing because almost all instruments are present
here. Stick consistently to the mixing strategy outlined at the start.
The central part of this section belongs to the lead vocals and the left and the right to the harmonic
instruments providing the warmth. It's better to thin out all other events in this area rather than risk overlaps.
This is the section where one finds most of the recording errors, because during the acoustic recordings (for
example, lead vocals) too little attention is paid to the comb filter effects of room acoustics. Room modes in
particular affect this recording range and add up to produce tedious misery.
If this is the case, use notch or parametric filters to look for the problem frequencies in the affected tracks
and lower then deeply. However, ensure good balance between warmth and dullness. If you deliver a mix
where these frequencies were imprecisely and therefore unsuccessfully eliminated taking away much warmth
while still being dull, the mastering studio can do little to remedy the situation, because the mastering
engineer cannot improve one feature without worsening another.

MID (350 ~ 2kHz)

It's very difficult to draw a set of rules for this range. Any rectifications in this area have to be done very
much on a case by case basis. If too much emphasis is placed in this range, the resulting sound is nasal,
wooden and piercing.

UPPER MID (2 ~ 8kHz)

This range is responsible for speech comprehensibility. As a basic principle, check whether a broad band lift
on the lead vocal between 2.5 and 4kHz would make sense in order to aid comprehensibility.
The skin heads which are important for the rhythmic localization of bass drums are also present in this range.

TREBLES (8 ~ 12kHz)
This is where cymbals, high percussion instruments, "s" sounds, chimes and high range of many instruments
frolic about.

UPPER TREBLES (12 ~ 22kHz)

A broadband lift in this range can aid the airness of the recording. However, an over emphasis can result in
digital or harsh perception.
For a natural sound in the upper spectrum, the trebles gradually decrease in loudness (roll-off or treble fade
out). If the high bands up to 20kHz were faded linearly, the recording would sound unbalanced and harsh.

Basic pre requisites for the targeted use of EQs in mixing:

A concrete and sensible panorama mixing concept, taking into account a balanced frequency distribution
across the left, centre, right range;
A good knowledge of the frequency ranges in which individual instruments develop their full functionality.
Many instruments have 2 crucial frequency spots in which they're effective whereas other instruments only
have a single frequency band in which they operate;
Use EQs first of all, to thin out and remove interferences and artifacts, (i.e.: low cut for the sub bass sector)
or to thin out and create space in frequency areas that offer little importance for the instrument in question.
And only as a second step, use them to emphasize frequency areas that have been under emphasized at the
recording stage

Remove first, then add:

Only lower those frequencies of an instrument that are of less significance for its characteristic sound and
conversely, emphasize only those that are essential for its characteristic sound.
It's impossible to always live up to the requirement that each instrument at solo setting or each individual
group have to have an outstanding sound. However, with the skilful mixing concept combined with slim
instrumentation and good recording, it's definitely possible for each solo switched instrument and the total
mix to sound very good indeed. This doesn't necessarily apply to pop productions, (i.e.: an acoustic guitar
that blends in beautifully into a pop mix offens sounds pitifully thin when switched to solo).
The same can apply to choirs groups. In those cases, start off by mixing a harmonious overall sound for the
group in question and then apply EQing to the group EQ within the framework with the overall mixing
strategy.
If there are too many harmony instruments competing for space in the lower mid range on the left and/or
right of the panorama, lesser important instruments can cope with quite significant reduction in sound and
still find an adequate place in the overall mix without resulting in the overlapping or fuzzy sound. This could
apply to an organ which might lose its lower mid range but places additional emphasis on the upper mid
range with its tinny distortions.
Remember during recording and production stage that each additional instrument raises the demands made
of the mix.
Depending on the complexity of the mix it's important to split the frequency range 3 ways, into left, centre,
right to achieve transparency.

Get the mental picture before anything else, then start tweaking. First clean then enhance with broad Q
curves (Pulteq). Think before you touch, you can do a lot with very few settings, the signal will stay more
together and you get a better mix.
Cutting is mostly done with fairly narrow peaks. Enhancing is best with shelves, if problems arise, do a little
cut within the shelf.
In the Panorama Dimension the vertical frequency range is split at least 3 ways, Left, Centre , Right.
A sensible panorama strategy is the pre requisite for a good frequency distribution.
A balanced frequency distribution across the Left, Centre, Right sectors is the secret for loud mixes. Any
rectifications of imbalances in the frequency distribution at the mastering stage have to be paid for with a
loss of sound quality and dynamics

The Lower Mid Range (120 to 350Hz) is of no importance for the Bass or the Bass drum, is very simple to
create space in this are for the lead vocals.
The bass gains its warmth from the lower 1½ to 2 octaves and its tonal comprehensibility between 400 and
800 Hz. The range between 800 and 1200Hz makes the bass sound more woody and planky.
Bongo EQ spec: Bass 40hz, Low Mid 400hz, High Mid 2.5kHz, Treble 6.3kHz.
Sting Ray only works at about 800hz.
Jazz Bass can go as far as 15kHz. The clicking sound of the frets and the slapping noises are higher.

The bottom of the Bass Drum is at about 90Hz and it's rhythmically localization in the Upper Mid Range (2
to 8kHz). Correspondingly, the Lower Mid Range (120 to 350Hz) is rather unimportant for the sound and
can be reduced. This results in a balanced sound in the centre frequency reduction.
Apply the same principle to the frequency distribution in the left and the right section of the Panorama. The
only difference is a lesser bass presence in the extreme Left and Right.
For very complex mixes, imagine the frequency distribution as split into 5 sections (Outer Left, 9 to 10:30
o'clock, Centre, 1:30 to 3 o'clock, Outer Right) in order to find space for each event.
For the pressure of the bass drum, it's sometimes helpful to apply a very steep low cut at 50Hz in order to
create more space for the bass.

Mixing a Tight Bass Range:

One of the distinguishing features of a good mix is that it can be listened to at a much higher volume than a
bad mix without getting close to clipping. The bass sector has been cleaned of all unnecessary information
and artifacts, the energy reserves are available for essential information.
Acoustic guitars whose main purpose is to create a silky strumming sound can be cut with a soft low cut
filter (12dB per octave) at 250Hz. Choirs and anything else same too.
If you need very deep notes, or very deep bass drum, remember the rule: The deeper, the shorter.
In order to keep these energy intensive sound events from preventing the speakers during their work, keep
very deep sound events very short and pulsive. It becomes very difficult to create a clear mix if the bass
drum has a long fade out at about 50Hz. Such bass drums also tend to blur the tonality.
If you really want to use deep and long bass drums, pitch the bass drum to the respective roots of the current
chords.

Basic EQ rules:
Apply a frequency distribution concept on at least three levels (L – C – R);
Implement a frequency distribution concept or respectively panning strategy on at least three levels (L – C –
R);
Familiarize with the sweet-spot frequencies of the various instruments;
First lower, than raise;
Lower rather steeply, raise over a broad frequency band;
Use at least as many low-cut filters as the project has tracks;
Almost any change in one band will affect the sound in the other bands.

Two EQ setting technique/ Split:

one cutting before the compressor and one boosting after it

PULTEQ tips:
Pulteq 200Hz peak "magical frequency" with female vocals and EDM basses with bright content. 5kHz band
sounds amazing.

Pulteq 2-knob technique:

For full mix it's not very useful because it becomes unpredictable and you don't know what it does and you
have to keep that in mind. It's better to just sweeten and maybe use a parametric before or after to
compensate to stuff that's too much.
For individual instruments is a really fast intuitive tone shaping tool.

Look and remove a bad frequency/ Peak and cut technique (when confused about frequencies):
Stay pretty tame both in the Q and the gain if you have to do peak and cut. Don't do it with a very narrow
peak because you'll hear a bunch of artifacts. Be gentle and don't go too fast, because you're probably
passing a lot of masks you don't even know are there. Don't boost too much, if you boost too much, rippling
is gonna happen, it gonna create masks.

Cut only technique:

Very hard to hear the frequency that bothers you by cutting but it will force the brain to work much harder
and you gonna get there faster.

Very regular EQ'ing situations:

Very narrow Qs are not very useful. Medium Q is the best. High quality EQs designers pick their favourite
Qs and that's a good place to start. Deal with the Q last.
Most of the time cut with peaks, then find the frequency (guess or intuition) and narrow the Q to the strict
minimum possible to get the desirable effect. So you touch the signal the least.
Find a frequency by intuition or peaking, do the processing and reduce the Q to its least amount without
ruining the signal, without going to super narrow Q to avoid artifacts.

2ᶮᵈ aspect: Measuring the levels

The level of individual events is the second aspect having a substantial impact on the designed frequency
distribution. Like in a jigsaw puzzle, one piece determines the other. The better the preparatory work the
easier the finer adjustments of levels. A great advantage in this is working with groups with the result that the
task at hand is only to balance the group levels.
To create an overall balance between channel, group and master bus levels or output bus, keep the master bus
at 0dB, and use a brickwall limiter to protect the output from occasional peaks.
Internal overdrives between channels, groups and master bus are hardly possible. Provided you're working
consistently at 32 bit resolution. Even levels above 0dB can be processed perfectly well at 32 bit floating
point resolution.

Should the vocals be in front as in radio compatible pop productions?

Is it necessary to understand every word?
Should the final result be a dance floor mix with greater emphasis on the bass sector?

Start by adding the harmonic instruments which provide the foundation to the bass group and then set the
level for the lead vocal group.
After that, add the accessories: percussion, effects, fillers and less important instruments.
A good trick is to listen to the mix at extreme low volume. If you can still recognize the tune, bass drum,
snare and bass, you've achieved an initial coherence.
To determine the relation between snares and bass drum, use a peak meter and graphic analyser to set levels.
Depending on the style, kick and snare are generally given the same energy level.
If there are uncertainties regarding the level settings for certain tracks it helps to pull all faders down and
then to increase them again slowly.
The volume level of the tracks has to be so consistent that it's possible to achieve a good transparent result
without the use of fader automation. The static mix is the starting point for fine tuning with fader automation.
If you start the automation before the static mix is coherent, you run the risk of inefficiency of having to
rework the automation repeatedly and thus confusing yourself.
Inconsistencies are better remedied directly at the events or clips.

3ʳᵈ aspect: Compression

Density = loudness
Attack – editing the transients with compression affects the perception, and therefore the level.
Short attack times, the effect is subtle. Long attack times emphasize the transients. Increasing the rhythmic
definition.Release time:
Release is the most important parameter in order to keep the pumping sound the lowest as possible. The
release time thus gives us 3 ways to use compressors:
Fine tuning depending on the song tempo and program material;
Increasing the volume of quiet sections that immediately follow transients, like snare strainer sounds. This
requires short release time;
Increasing the volume of quiet sections that would otherwise be drowned in the mix due to lack of sustain.
With the decreasing input signal level, the VCA jumps back to the starting position thus amplifying the
weakening signal. In that way, even a mediocre bass with old strings gets sufficient sustain for a good
foundation.

RMS peak:
RMS peak is a rarely used parameter defining the characteristics of the release and compression behaviour.
RMS oriented compression is advantageous in the mastering process in order to achieve the loudest possible
results, whereas a peak oriented compression is a pure top end compression with little effect on the overall
loudness.
RMS = loudness oriented compression is based on release behaviour that is analogous to compressors based
on electro controlled VCAs

Electro-Opto like LA2A

The electro mode is another description of RMS oriented release behaviour and it's modelled on analog
compressors. With this release behaviour, the return speed of the VCA to the original setting increases the
closer it gets to the original setting 0.
Opto works the other way around, the return speed decreases in line with the approach to 0. This opto
behaviour is very nice to drums for example,

The Urei 1176 is particularly used for editing vocals, because with a long attack time it adds unmistakable
bite to the vocals which is particularly evident in percussive sounds and hard consonants.
Many legendary lead vocals sounds owe their success to the use of the 1176.
Caution: the attack time works the wrong way around. Knob to the left equals long attack time and knob to
the right equals short attack time.

Give the vocals more bite:

Before compressor: high pass, boost the speech comprehensibility at 3 to 4kHz (using a pulteq). Long attack
times. Release time to taste. (Short release times pull head phone crosstalk more to the front than longer
setting).

Make funky guitars sound more crisper:

High pass and roll off the trebles a bit (if there's too much air). Add silky trebles with a pulteq (raising 10kHz
broadband). Long attack to create artificial transients therefore to refresh the signal. Fast release times to
increase of the quieter ghost notes or strumming sounds to make them clearly audible.

Summary:
At he mixing stage compression is predominantly used to structure the sound and to control or respectively
limit natural dynamic leaps, and not mainly to create greater sound density (loudness) as it's done during
mastering. In this context we repeatedly encounter 3 different compressor uses:
Supporting sustain (presence through level consistency)
Supporting transients (presence through snappy artificial transients)
Raising the level of quieter sound sections. Indirectly this processing results in restricting the signal
dynamics. The compressed signals can thus be louder and more level consistent.
The use of compressors in groups often requires more subtle inaudible setting with low ratios with the goal is
to join individual signals and achieve a subtle restriction of dynamics.
In individual cases, the loudness maximizer may be used in the compression of groups as long as that creates
the desired pressure.

DIMENSION 3: DEPTH - (Layering with Reverb and Delay)

1ˢᵗ aspect: Pre-delay as a design component

There are 2 designed concepts available with the significant psychoacoustic effect on the ability of our ears
to interpret distance: Pre-delay and frequency curve of the reverb return.
Pre-delay is an important parameter to set distance. It's used to set the time span between the direct sound
and the onset of its first reflection. The longer this time span, the greater the distance of the sound source
from the listener. However, for the positioning of percussive instruments this parameter has to be used with
great caution. For all percussion events including drums and bass, reverb should be set without pre-delay or
with pre-delay of up to only about 10ms. While all the time checking for rhythmic coherence. High pre-delay
settings can destabilize the rhythm, particularly a percussive sound events. High pre-delay values of up to
about 60ms are particularly suited for choirs and strings to assign them back to the back "?".

In sound theory it is correct that a longer pre-delay makes a sonic event appear closer to the listener. From
my experience this applies for very acoustical projects, let´s say a recording with two classical guitar
players or possibly also for a live recorded jazz band. However, popular music is in most cases much denser
and at least in the realms of my personal experience it appears that this rule does not work very well when
dealing with dense pop music.

Regarding highs and proximity perception I´d like to mention this:

A dry event put into a reverb tends to be perceived as ONE event. Having a bright reverb tail communicates
to our subconscious that event must be close because the event haven´t travelled that far through the room
where it would have lost proportionally more highs than mids and bass. Therefore creates a bright reverb
tail a close perception of the overall event. But as long mixing is an art everybody is free to do it differently
as long the desired results are achieved.

2ᶮᵈ aspect: EQing in reverb return as a design component

Deeps sounds have more energy than high sounds.
The greater the distance between the listener and the sound event the lower the proportion of high
frequencies in the reverb signal. For that reason, the treble rolloff of the reverb signal is one of the most
effective psychoacoustic means of representing distance to a sound source. Because our ears interpret this
information automatically, subconsciously.
For events that are to be at the front center of your mix you should always select reverb settings that are
richer in trebles and for events at the back, those that are duller.

For a typical pop music setting with a 5 piece band, a handful of doublings and a bit of ornamentation:

Send effect Kinds of reverb Area of application Pre-delay Treble roll-off

1 Small room/ambience Drums and some bass 0 to little slight
2 Medium sized room Snare 0 no
(with gate)
3 Medium-sized room Percussions 14 medium
1,8s + Stereo
exapnder
4 Large reverb Background events 30ms strong
(choir/guitars)
5 Vocal chamber lead vocals 0 no
6 Modulation delay keyboard and triangle - -
7 Delay Rhodes and triangle - -
8 - - - -

Send effect 1
In line with the panning strategy outlined as developed in the section on panorama regarding the placement
of individual instruments, the drums are placed at the back of the stage. Use a subtle little ambience reverb
on stage reverb. This reverb creates just enough information to bring the dry recorded drums or carefully
programmed and totally dry samples out in the spatial dimension. In context, you'll only be able to identify
the reverb as such if you listen very carefully. Only by switching it off again we'll notice the absence of
spatiality. To make sure the bass isn't coming from another dimension, it's only giving enough of this reverb
to appear as a unit with the drums.
In the drum section, it's preferred to give a little less ambience to the bass drum than to the other tracks, in
order as not to overemphasize the ambient part of the bass drum skin sound. Otherwise, the result, maybe a
rather flattery sound of the bass drum.
Send effect 2
Depending on style and genre, an additional reverb is often used for the snare, which is more obvious and
which creates more space for the snare. Caution is advised though, otherwise you might find yourself on a
trip back to the 1980's. If using loops with integrated snare or live snare with crosstalk, separate the snare
first to make sure that the big reverb is applied just to the snare.
For ballads it's quite possible to apply this same send effect to the toms to make them sound more
impressive. If needed, this reverb can be gated and using the gate to reduce the decay. This gives the option
of using a full long reverb which is artificially shortened. This trick also originates in the 1980's and has to be
applied with caution and sensitivity. The gate setting should be adjusted to the song tempo.

Once you've implemented the panorama distribution consistently, your aim is to create a modern and
comparatively dry sound style. Guitars and keyboards without reverb can be placed on the left and on the
right in the panorama and thus become ?peratively far forward. Sometimes it's more useful to use a delay for
the mush free differentiated creation of a subtle spatiality. This moves the events in question less far into the
background than would have been the case with reverb.
Send effect 4:
To place choirs, percussion and other instruments deliberately at the far back requires a large reverb with a
bit of pre-delay and damped high ends. In this case the reverb part has to be set quite high to make our ears
accept this spatial information, because in this, you have to fight the masking effect.
The best advice is to experiment with the stereo expander in the reverb return.
Left-right split reverb return for the large reverb:
If the result is undifferentiated or insufficiently mono compatible, try using 2 identical reverb presets from 2
separate devices, one in which returns in the right side of the panorama, the other in the left. Both reverb
devices receive reverse sent signals so that the instrument in the left of the panorama is revered to the right
and vice-versa.
Send effect 5:
Specially in mixes that tend towards dryness, vocals require a particularly high quality reverb to prevent the
vocals from being low into a reverb cloud. Sometimes all that is needed is a subtle, small and unobtrusive
reverb with attributes similar to a drum booth. Sometimes combining with an additional delay works well,
which is less inclined towards blurring than a medium sized reverb space.
Combinations of reverb and delay or delay in reverb can also be used to this end.

In the present scenario, 4 to 5 reverb devices are all you need, provided that for the large reverb space, you
just need 1 reverb device with the stereo left-right return.
Send effect 1 and 2 have no or only a very short pre delay. Send effect 1 can have a slight treble rolloff to
define spatial level whereas the snare reverb can be without EQ in order to place the snare at the front.
Bear in mind that instruments played laterally in the panorama with the effects being sent via aux send and
returned with the central panorama setting, tend to be drawn into the center. This phenomena increases with
the effect amount and applies particularly to modulation effects. Counter balance that with the extreme
panning of the dry track and should that be impossible, use a second panorama effect, or in that case, insert
the effect into the effected track to replace it as a send effect.
In the program setting, the last entry under VST is the option to link the panorama control for send routing to
the panorama control of the channel. This is a useful function to avoid the use of effects blurring the dry
panorama settings which you have made.
The time and effort of using left-right split effects is only justified if the reverb part becomes too large to
convey all the spatial information because it competes against the masking effect. The more complex the
mix, the greater potentially the time and effort required to place all events accurately within the 3
dimensions.

Layering strategies: Delay instead of reverb

A delay tail has the advantage that the lead vocals appear fuller and have more volume without putting the
frontal placement into jeopardy. The more obvious the delay appears in the mix, the more attention you need
to pay to the stereo width of the effect in order to ensure that the lead vocal retains the center position. In
conjunction with delays, there are 2 interesting variances of ducking or suppression:
Either the tail only appears when the lead vocal is silenced, in other words, there is no input signal over a
certain period or the tail is suppressed as soon as there's no signal.
The first option is useful if the delay blurs the clarity of pronunciation too much or it's not desirable from a
sound aesthetics point of view but does create a good filler effect during pauses;
The second option is useful to thicken the vocal sound without obviously appearing in the foreground as a
thickening agent during pauses in the vocal. The TC 2290 Dynamic Delay is well known for this very
purpose.
This effect is easy to reproduce by exporting the delay as a single track mono or stereo. then edit the desired
effects manually using cut, fade, mute and level tools.
While delay also creates a spatial impression, it's not suited to create a targeted stage depth for different
events. For that reason, delay should only be used as a stylistic aid. The exceptions are the dub delays as used
in dub and reggae which convey the impression of a certain distance (Roland Space Echo RE-201. It's the
combination of a band echo with spring reverb. Because for technical reasons, this tape based device is quite
stingy in the high frequencies domain and it communicates distance and spatial depth when strongly used)

INSTRUMENTS
BASS DRUM

Bottom Head sound

65 ~ 110Hz 3 ~ 5kHz (~ 8kHz)

Occasionally, there are 2 tracks for the bass drum. One track with the mic placed inside the bass drum closed
to the drum head, representing the head sound and one track generally recorded by condenser mic closer to
the front skin, or the hole in the front skin containing the deeper frequencies.
Regardless of whether there are 1 or 2 tracks, in both cases the bass drum have the same frequency
characteristics which are important for its function, and which it needs to be properly modelled:
The bottom or main body between 65 and 110Hz and the head sound in the upper mid range between 3 and
5kHz. In rare cases up to 8kHz.

EQ:
For a good bass drum sound, you can use the bell filter to boost the bottom sound. For this purpose Q
amounts should be moderate.
Pulteq EQ is a miracle cure for bass drums that are somewhat weak in the low frequency. Select 60 or 100Hz
and turn the boost knob up a good amount. The attenuation knob can be used in addition in order to reduce
unnecessary mid range components in the bass drum.
With a second bell filter you can specifically control the head sound and thus the rhythmic definition. Qs that
are too high when levels are raised result in artificial sound. Depending on the bass drum, center frequencies
from the entire upper mid range (2 ~ 8kHz) are possible.
The next step is reduction. In practice you'd always reduce prior to raise frequencies of the main areas. If the
bass drum is very low, you can apply a very steep low cut filter at about 30Hz in order to keep the bass sector
tighty.
A far greater importance is the lower mid range "misery range" (125 ~ 350Hz). I this range the bass drum has
no particular purpose and if it's too overbearing there you can use a bell filter to reduce it in this range. This
way you create more space for the bass for all other events which desperately need this range to fulfil their
potential.
If there are 2 tracks available, you can use this to achieve a clear modelling and joining of the main
frequencies. To that end, you could apply a higher low cut to the track with the head components in order to
eliminate all bottom components from this tracks.

Compression:
For the bass drum the compression has 2 functions:
Restricting dynamics in order to catch scattered individual high level peaks, also referred as top end limiting
or top-end compression;
For creating more punch through artificial transients using a long attack time. With a long attack time the
small overshoots support the rhythmic definition and the percussiveness of the bass drum.

Gate:
If the gated bass drum sounds too artificial, you can double the bass drum track, gate one tracks and mix in
the other unprocessed track as ambience.
A short gated bass drum can be more extremely processed with EQ without significant increases in the bass
range flooding the entire sound. The reason for this is that short impulses tend to be less tonal. You can
always add the open unprocessed bass drum track discretely to the mix in order to preserve the impression of
a natural ungated drum set.
The lower the bottom centre frequency of a bass drum, the shorter it should be gated in order to keep it free
from becoming tonal and interfering with the bass. By contrast, a bass drum with a long sustain and little
damping can be processed with just some slight bass increases to avoid an oversaturation of the bass band.
In this situation, a significant rise with a bell filter can sound slight undifferentiated.

Reverb:
Too much reverb emphasizes the head components in the upper frequency spectrum and makes transients
sound fluttery and imprecise.
Use reverb without pre-delay in order to prevent fluttery due to rhythmic offset.
Add a small dosage of a small drum ambience reverb in order to create a spacial link between bass drum and
the rest of the drum set. A totally dry bass drum can easily sound out of context and disjointed but it's used in
certain genres like hip hop.
Bass drum reverb has to be used to subtly that its absence is immediately obvious when switched off but
when on doesn't sound overpowering or latent.

The top of the bass drum can get in the away of the vocals or the bass, attenuate 3 or 4 at 500Hz with Pulteq
(acoustic or electric Bass drums). That way it may make it fatter.

SNARE

Bottom Strainer sound Top end

120 ~ 260Hz 2 ~ 8kHz 8kHz ~

Like the bass drum, the snare has 2 important frequency spots. The only difference is that the bottom
frequency is somewhat higher than the bass drum that is, in the lower mid range. Depending on the size of
the drum or the sample, the centre frequency of snare drums fall between a 120 and 260Hz.
The top mic is mainly focused on recording the punch in the upper mid range (2 ~ 8kHz) and the strainer one
on the snare sound in the upper mid (2 ~ 8kHz) to top range (8kHz >). In the upper spectrum the snares can
often create a broad hiss up to 20kHz.

EQ:
If you're not happy with the snare sound, there are 2 areas where you could intervene. The bottom and the
overtone spectrum. In addition it's recommended a low cut at 80Hz and if necessary a damping in order to
create more space in the mid range. The low cut is important to avoid crosstalk with the bass drum.
The semi-pitch sounds of metal snares add an additional interesting aspect to this equation. These
frequencies can be specifically emphasized or dampened with a mid range Q of a bell filter.
A snare sound that is in tonal context with the key of the song is easier to integrate stylistically into the song.
If the sound is off key by half step it'll most likely sound awful. Don't hesitate at all to use the pitch shifter
function to pitch the entire snare track to a suitable note (i.e.: the root or fifth).
If you find the sound obtrusive, you can use the technique of track-doubling to move the sound further into
the background. To bring out the snares overtones, it's recommended to use high-resolution high shelving or
parametric filters with low Q factors.

Compression:
Snare compression can fulfil 3 tasks:
Top end limiting/top end compression to restrict dynamics. In this case care needs to be taken to ensure that
the natural transients remain unaffected by the compression, because otherwise the resulting sound might
resemble a cardboard washing powdered drum. It sounds as if someone has let the air out of the snare and it
occurs as a result of too much distortion of the natural transient due to excessive dynamic processing;
Boosting snappiness and percussiveness to the use of long attack times;
Adjustment of the quieter signal components, specially the strainer sound through fast release times. The
faster the release time, the more strainer sound.

Doubling the bass drum or snare track:

The trick of doubling the track is very popular for tracks with microphone crosstalk (between bass drum and
snare for example), particularly if only one microphone track is available for the 2. Editing both tracks
differently allows you more intense manipulation of the signal without resulting in artificial sound.
Once you have short-gated the copy track you can apply extreme EQ and compression edits. Heavy EQing
on short impulse sounds is less obtrusive and signals that are often free from crosstalk can be compressed
much harder. Otherwise, a strong compressor with a short release time would result in unpleasant emphasis
of the quieter signal components of the crosstalk on the ungated snare track. The track that's processed is
then mixed to the natural ungated track to create punchness.
You can also employ this technique in order to split the instruments into their core components: bottom, head
and then join them together again. In this case one track would represent the high frequency components, the
other the low ones. This process is a particular important if the snare is mixed with the large reverb. The
copied and gated snare track with the short truncated snare sounds is used to control the large snare reverb
which will therefore be succinct and clean because it's not blurred by any crosstalk signals. By contrast, only
a small ambient reverb is added to the opened ungated track.
In some styles, it's common to add some meat to the snare sound by equipping the gated track with a very
dense gated reverb. Such reverb can be simulated by adding a noise gate downstream from the reverb device.
Mixing both tracks together ensures that the snare still retains enough natural sound.

OTHER INSTRUMENTS
Toms:
If the toms sound somewhat weak, you can create more punch by using an extreme bell filter with a medium
to high Q. To this end, identify the exact center frequency of each tom.
Depending on the genre, you can also mix a small dose of the large snare reverb to the toms in order to bring
them out.
Hi hats:
The high hat microphone contributes to the overall drum sound as a kind of super close ambience track.
Overheads:
The overhead track is already a preliminary stage to the spacial sound of closed ambience. Here a low cut
filter is used. In this case it's recommended a soft roll off of 12 to 24dB per octave. Use your ears to
determine the frequency. Sometimes it adds to the overall sound aesthetics to reproduce the entire spectrum.
In other cases, a low cut at about 400Hz is the appropriate solution.
Remember that the overhead microphones at 2 meters distance from the snares already have a delay of 6ms
which can result in slight flamming of the bass drum if you set the low cut too low.
If the drummer has forgotten his cymbal cleaner for the recording session, you can successfully use a soft
sounding high shelf filter to add some of 12kHz onwards

TECHNIQUES:

Snare needs to be louder (To bring out quieter sections to make them more audible in context): Snare
separation from any crosstalk from other instruments:
Make a copy of the snare track, high pass it so that mainly the higher sections of the snare remain then
compress it and blend with the original track.

Emphasize the snare's attack, make it crisper: Long attack times

Gate reverb on the snare to add beef.

BASS
EQ
The bass can only handle a low cut if it uses one a very steep bass roll off. 36dB per octave is often not
enough for a very deep bass and it would excessively dampen the lower desirable range from about 32Hz
downwards.
Some bass lacking in bottom can benefit from a level increase between 40 and 80Hz with medium Q. If the
bass threatens to take space away from the other instruments, you can dampen the low mid range misery
(120 ~ 350). The sector between 800 to 1.2kHz controls the nasal, wooden sound components. It also
supports tone definition. Above these frequencies there are only a few quiet overtones that are mostly fret
and muffled sounds.

Compression
Limiting dynamics (from top end compression to limiting) to limit disruptive peaks and effects of a regular
playing;
Create sustain for long, sustained notes, particularly popular for slow song tempi;
Emphasise the quieter lighter tones through a short release time;
Transient support. Boosting the rhythmic definition and percussiveness with long attack times.

In normal popular music, the bass is the most subtly and almost inaudibly reverbed with the ambience of the
percussion. This underlines the unity of drums and bass. For a fretless bass, the use of a subtle spacial reverb
is quite popular, other than this, a reverbed bass belong to a particularly style.
1176 or LA2A is compulsory. A gently adjusted 1176 also contributes to a good sound.

VOCAL
Panorama:
In pop music, single lead vocal tracks are generally placed at the center. In parts of the song that are able to
convey an impression of intimacy it's always better to use single lead vocal tracks rather than doubled ones.
If one track isn't enough to create the required presence, you can always duplicate the track and decorrelate
the doubled track, in order words, make then different by processing them separately using different
compressors and EQs and distribute them in the left/right panorama. However, make sure that you retain the
phantom center, in other words, the impression that the lead vocal is coming from the centre. This strategy
results in giving more punch and presence to the voice.
For some pop stars with rather finer voices, we tend to exceed the critical mass by duplicating the song 40 to
60 times and applying auto tune to each track. After sufficient editing, even this, results in a good sound and
the impression of a single full voice.
Pop music refrains are frequently doubled. 4 duplications per singer and part are the usual average. If there's
one lead singer and 3 choir voices, you can easily end up with 16 vocal tracks. For separation of the sound,
route choirs tracks to a separate group.
The spread of the double lead vocal tracks within the panorama should only be wide enough to ensure that
the phantom centre is retained and that the choirs surrounds the lead vocals in the panorama.
The choirs is always placed in the panorama in such a way that the higher voices are further towards the
outside and the lower voices more centered. Pre mix to attain a closed sound within the group and then
embed it within the mix. You can insert a stereo expander in order to increase the basic width of the choirs in
those sections with the separation of choirs and lead vocals seems to be inefficient.

EQ:
For lead vocals, 80Hz high pass is sufficient and for choirs tracks, up to 400Hz with 12 or 24dB per octave,
depending on individual style or situation.
Vocal recordings often suffer from resonances in the bass and the lower mid range as a result of room
acoustics, associated room modes and comb filter effects, because recording situations generally only
recognizes and eliminate the foreground interferences in the mid to higher range. This interferences are
characterized by a dull mumbling, or a closed sound. In cases like this, identify the resonant frequencies by
scanning the frequency band between 120 and 350Hz using extreme and steep amplification and then
reducing them with notch or bell filters with high Q factors.
The lower mid range must not get too thin, this is where warmth is conveyed, particularly of male voices.
The mid range up to 2kHz controls nasality. For male voices from 2kHz and female voices from 3kHz you
can boost speech comprehensibility with a broad increase (with a low Q). Some slight support in this area is
standard practice because placing vocals at the phantom centre, same loudness left and right creates a comb
filter effect as a result of the muffling effect of the head, which weakens just theses frequencies. Comparing
this to a signal from the centre speaker from the front, the signal reaches the left and the right ear uniformly
and without distortion. Reproducing the signal simultaneously on the left and the right channel results in time
differences, because the signal from the left box also reaches the right ear though slight delayed and damped,
thus adding up to a comb filter effect with the right signal and vice-versa.
Frequencies between 6 and 8kHz are very sensitive, this is where the "s" sounds are located. With due
attention to hissing and "s" sounds, all increases consequently must be subtle and over a broad frequency
range, possibly in conjunction with a de-esser.
You can create openness with a gentle increase of the trebles from about 10kHz upwards. It's quite common
to use shelving filters for this purpose.
To edit the upper band, EQs that work with over sampling and have a soft sound are highly recommended.
Compression:
LA2A and 1176 are very popular. The 1176 is the vocal compressor of choice. Only it will give you the
unmistakable Bob Clearmountain sound. A popular term in this context is the bite in a vocal sound which is
adjusted through the use of long attack times. These allow the initial impulses of the vocals to pass unedited
until the compressor starts working. You can select rations of up to 12:1 for the lead vocals. Even if the
compression sounds extreme in solo mode, the sound of the song blends in well and no longer appears as
being glued to the front.

Reverb and Delay:

One of the endless range of options is using large amounts of small room reverb. These reverbs are often so
subtle and unobtrusive that they're not consciously registered. It's quite common to combine these small
reverbs with large ones and with delay. Skilfully used delay can make to vocals sound fuller without pushing
them to the background.
Example 1:
Lead vocal track is doubled. On track 1 a small room reverb is used. Track 2 has a big room that goes to a
delay and then a gate.
Example 2:
There's a small ambience room on track 1 and a delay on track 2 that then is routed through a reverb.

Aligning different sounding takes:

Sometimes the vocal takes have been recorded on different days. Apart from the singers mental condition
and the physiological condition of the voice, even the smallest alterations in mic setup and the position of the
singer relative to the microphone can have a significant influence on the vocal sound with the result that later
corrections often sound very inconsistent. It's easiest to harmonize different takes on a spectrum level. If a
parametric EQ and a good sense of hearing don't produce the desired result, an intelligent EQ might do the
trick.
The only things that cannot be rectified in this way are subtle sound variations such as a slight different voice
timbre on different days as well as room dependent differences caused by slight different recording positions.

REVERB TECHNIQUES

Delay rather than Reverb for subtle spaciality on the L and R

If once implemented the panorama distribution consistently, the aim to create a modern comparatively dry
sound style.
Guitars and keyboards without reverb can be placed on the left and or right in the panorama and thus become
far forward.
Sometimes it is more useful to use delay for a mush-free differentiated creation of a subtle spaciality.
This moves the events in question less far into the background than that would be the case using reverb.

Apply gate to the reverb send according to the tempo of the song. (from the 80's)

To place choruses, percussion and other instruments deliberately at the far back requires a large reverb with a
bit of pre delay and dumped high ends.
“In this case, the reverb part has the be set quite high to make our ears accept the spatial information,
because in this you have to fight the masking effect.”
The best advice is to experiment with the stereo expander in the reverb return.

If the result is undifferentiated or insufficiently mono compatible, try using 2 identical reverb presets from 2
separate devices, one of which returns the right side of the panorama, the other in the left.
Both reverbs devices receive reverse sent signals (so the cabasa in the left of the panorama is reverbed to the
right and vice-versa).

“in the example, I've used separate reverb devices for percussion and chorus, because I have processed the
reverb return of the percussion reverb with the stereo expander, in order to give more space to the percussion.
In the mix, the reverb is unobtrusive but projects a certain depth”

Silky strumming acoustic guitars can be cut with a soft low cut filter around 12dB per 8ve at 250hz

BASS

EQ

The bass gains its warmth form the lower 1½ to to 2 octaves and its tonal comprehensibility between 400 and
800hz. The range between 800hz and 1.2Khz makes the bass sound more woody and planky.

Bass lacking bottom benefit from level increase between 40hz and 80hz with medium Q.

The section between 800hz and 1.2Khz control the nasal, wooden sound components. It also supports tone
definition.

Above 1.2Khz there are only a few quiet overtones, and mostly fret and muffle sounds.

If the bass threatens to take space away from the other instruments, dampen the lower mid "misery" range
between a 120hz and 350hz. Low mid range is not important for bass. So it's simple to create space in this
area for the lead vocals.

EDM bass can benefit from Pultec EQ boost on the 100hZ. If mask occurs, use attenuation.

Some basses' working ranges:

COMPRESSION

Suggestions: Urei 1176 - (Stillwell Rocket) and/or Teletronix LA-2A - (Sonic Anomaly S.LA.X)

Processing options:
 • Limiting dynamics (from top-end compression to limiting)
• Creating sustain for long, sustained notes (long release time)
for slow song tempo.
• Highlight the quieter secondary sounds (short release time)
• Transient support (long attack time)

Multiband compression technique:

In situations where the traditional compressor cannot sufficiently increase the dead notes to make them
audible within the context, insert it into a copy of the bass, where it emphasizes the relevant spectral sector
through expansion.
Leave it in the solo setting of the selected band so that only the frequency range in question passes through
and it's mixed into the full range channel.

REVERB

In normal pop music the bass is the most subtly and inaudibly reverbed with the ambience of the percussion.
For a fretless bass, the use of a subtle spacial reverb is quite popular.

TIP: CHORUS AND OTHER EFFECTS - Use plugins that allows to process frequencies above 250Hz to
avoid phase interference. If that's not available, copy the track, apply a steep low cut at 250Hz to the copied
track and only use chorus on this copy.

GUITAR

Pultec:
3 kH boost with sharp attenuation (works like a low pass filter after the boost)

VOCALS

In pop music, in parts of the song that conveys intimacy it's always better to use single lead vocal tracks
rather than double ones.
If one track is not enough to create presence, duplicate the track and decorrelate tracks, in other words, make
them different by processing them separately using different compressors and EQ's and then distribute them
letf and right panorama. However, make sure to retain the phantom center.

Pop music refrains:

Four duplications per singer and part are the usual average.

The spread of the double lead vocal tracks within the panorama should only be wide enough to ensure that
the phantom center is retained, and the chorus (back vocals) surround the lead vocals in the panorama.

The chorus (back vocals) is always placed in the panorama in such a way, that the higher voices are further
towards the outside and the lower voices more centered.
Pre mix to attain a closed sound within the group and then embed it within the mix.
Insert a stereo expander in order to increase the basic width of the chorus in those sections where the
separation of chorus and lead vocals seems to be insufficient

EQ

Low cut at 80 hz for lead vocals and 12db to 24db p/ octave filters at 400hz for chorus depending on style
and situation.
Vocal recordings often suffer from resonances in the bass and the lower mid range as a result of room
acoustics, associated room modes and comb filtering effects because recording situations generally only
recognize and eliminate the foreground interferences in the mid to high range.

This interferences are characterized by a dull mumbling, (a closed sound). In cases like this, identify the
resonant frequencies by scanning the frequency band between a 120hz and 340hz using extreme and steep
amplification and then reducing them with notch or bell filters with high Q.

However, the lower mid range must not get too thin, this is where the warmth conveys, particularly of male
voices.
The mid range after 2Khz controls nasality. For male voices from 2Khz and female voices from 3Khz, boost
speech comprehensibility with a broad increase. In another words, with a low Q.

Some slight support in this area is standard practice because placing vocals at the phantom center, same
loudness left and right creates a comb filter effect as a result of the muffling effect of the head which we can't
just with these frequencies.

Comparing this to a signal from the center speaker from the front, the signal reaches the left and right ear
uniformly and without distortion.
Reproducing the signal simultaneously from the left and right channel results in time differences, because the
signals from left speaker also reaches the right ear though slighlty delayed and dumped thus adding to a
comb filtering effect with the right signal and vice-versa.

Frequencies between 6Khz and 8Khz are very sensitive. This is where the "s" sounds are located.
With due attention to hissing and "s" sounds all increases must be subtle and over a broad frequency range
possibly in conjunction with a de-esser.

Create openness with a gentle increase of the trebles from about 10Khz upward. It's common to use shelve
filters for this. To edit the upper band, EQ's that work with oversampling and have a soft sound are highly
recommended.

Between 20 Hz and 100Hz (12dB per octave) - High pass to get rid of unwanted noise

100Hz - Notch out for resonances from small rooms

On male vocalists in small rooms, you may find resonances. Or a very deeply miked female vocal with lots
of room sound.
If something in the vocal is fighting with the bottom of the guitar or top of the bass, look in this area and
gently notch it out with EQ

Between 180Hz to 240Hz (around 300Hz for female vocalists) - Notch out if vocal sounds muffled
It tends to be hell for close miked singers. It sounds like cardboard, woofy.
Go here first, especially for male vocalists.
If the vocals feels a little muffled, wooly (dark) or cottony, instead of reaching for the high end first, go down
to this area and notch it out and see how it feels

500Hz area - Notch out with broad curves if vocal sounds nosy
Be gentle, 1 or 2dBs fairly wide Q.

600Hz to 1.6kHz tends to be fine in most recordings.

1.6kHz ~ 1.8kHz to 2.2kHz ~ 2.3 kHz - Boost to add presence

If it's dull, lack presence, if the vocal doesn't jump at you, try boosting a couple dBs bit around 2kHz range.
To help pull a vocal in front of a very busy mix.
Look around for the key of the song and the resonance of the singer to find the exactly place.
2.2kHz to 3.5kHz - Watch for aggressive digital sound
All the digital sounds, screechy sharp, fast notches stuff. A lot of converters have problems here.
Use a dynamic EQ in this range

5kHz - Boost to open up the top of a vocal

If the singer is recorded too closed, the nice resonance in the top of the head disappear because the mic is
overwhelmed with all the bottom from the proximity effect

6kHz to 8kHz - Watch for sibilance. De-essers are recommended

Most of the time, unless you have a blatant problem, it's better leave it alone.

9kHz to 11kHz - Boost for brightness (treble) without sibilance problems

For a dull vocal, boost here without ruining the s's, the 5kHz shine and the presence at 2kHz

12kHz and above - Boost for air, space above the head

Listen before you look,

Have a vision for what the vocal should sound like before tweaking
It's always better to work in the track, rather than in solo especially when enhancing things.

COMPRESSION

Suggestions: Urei 1176 - (Stillwell Rocket) and/or Teletronix LA-2A - (Sonic Anomaly S.LA.X)

Urei 1176

The vocal compressor of choice. Bobby Clearmoutain sound.

Bite = long attack times. This allows the initial impulses of the vocal unaffected.
Ratios up to 12:1 for the lead, even if the compression sounds somewhat extreme in solo mode, the sound of
the song blends in well and no longer appears as to be glued to the front.

Bad room acoustics make it almost impossible to use extreme compression because the unwanted reverb
naturally moves forward.

REVERB AND DELAY

One of endless options is using large amounts of small room reverb. These reverbs are often so subtle and
unobtrusive that they're not consciously registered. It's quite common to combine these small reverbs with
large ones added with delay. Skilfully use delay can make the vocal sound fuller without pushing it to the
background.

Technique 1

Small room sent to Track 1

Big room going through a Delay and then a Gate sent to Track 2

Technique 2

Small room ambience to Track 1

Delay on track 2 (Spache Echo RE-201 ) routed to a Reverb
 COMPARING REVERBS

PLATES/ tail
Nice bloom(early reflections) right after the vocal and then a bit of tail that thins out and is very smooth

SPRING
It has the biggest sonic imprint.
I gives a thinner bloom (compared to plates) but a denser bloom. Bloom is a little shorter.

HALL/ airy
Instead of being just tail like, there's a little bit of a height quality to it.
Comparing Hall with Plate, the ER of the Hall gives a certain height feeling. They're not surrounding the
vocal, they're more kind of airy above the vocal.

Hall gives a kind of air/height and a Plate gives a kind of tail and a surrounding place around the vocal

ROOM/CHAMBER
Same family with Hall, just a lot closer and brighter. More early reflections.
Comparing with Spring, the early reflections are similar, the tail is different, and the Spring has a ringy
quality.

AMBIENCE Algorithm - from the Lexicon in 1983

Very short delay.
It's basically almost all early reflections blending into each other to create space. Kind of like a very short
Room but with a certain intimacy on the vocal

It removes that rough edge of having nothing between the notes, but it doesnt give any garbage or any tail.
Ambience algorithm is really practical to be able to fill the holes in between two notes on an instrument ,
vocal etc but not any tail or garbage or heavy stuff that you have to pull on and try an EQ or do things.
It's just very discrete

In the mix, the differences between different types of reverb will be fading out rapidly but the ER type of it is
what really matters.

Andrew Scheps:
1.4 kHz - standard frequency for rhythm guitars. It brings out the tone of the notes and a little bit of the
attack but it's below where all the noise will be.

If you ever try to get kind of a vintage sound out of an EQ, add high pass (5kHz)and low pass (20kHz) to it
at very high and low frequencies, and you will find that all of sudden it has a little more character to it.
Low pass filter around 20 kHz can affect the bottom of kick drum.

Sending or not sending kick and snare to the drums stereo compressors:

"I dont always want the kick and snare to go into the stereo drum compressors, because what that can do is
make the cymbals start sucking backwards because they'll get pulled down in the drum compressor every
time there's a kick or a snare. And I need the cymbals to stay up and be constant, because otherwise it can
give the impression of the mix being crushed too much even when the mix is wide open, just because the
cymbals get that extra pump"

Aphex

"This is usually used on vocals but can be used on percussion. It can be used on anything that needs a little
bit of synthesised top end"

Rear bus

"I'm not hitting it very hard but I have a lot of make-up gain. The attack time is as fast as it can be but the
release time is pretty quick, but it's not as fast as it can possibly be. The reason for that is if you've got a very
percussive track, you would actually really hear this pumping and you'd hear it grabbing all the transients
and then letting go right away. This is the only control I might ever look at on the Rear Bus.
If I feel that the Real Bus is doing too much to a mix, then I'd come and quicken up the release time a little
bit. Sometimes that would fix it. All of sudden the mix opens up. I get a little more clarity because it's not
holding the compressor down quite as hard. In a really dense mix, that can be a problem. I don't think I
would ever make it slower. I'm not worried about being in time with the track. There's so much stuff going
into this, it doesnt have time to release in time with the track. We're talking micro to miliseconds here. We're
not talking quarter, eighths, sixteenths notes"