Snoman R. Dance Music Manual: Tools, Toys, and Techniques

Подождите немного. Документ загружается.

PART 3

Mixing & Promotion

394

Relative Volume

Analytical listening can tire your ears quickly, so it’s always advisable to avoid

monitoring at loud volumes as this will only quicken the process. The ideal stan-

dard monitoring volume is around conversation level (85 dB), but you need to

keep the Fletcher Munson contour control in mind during mixing. After every vol-

ume or EQ adjustment, reference the mix again at various gain levels. It can also

be beneﬁ cial to monitor the mix in mono when setting and adjusting the volume

levels, as this will reveal the overall balance of the instruments more clearly.

E Q

EQ can be used to shape all instruments but you can only apply so much before

the instrument loses its characteristics, so be cautious with any EQ. Always bypass

it every few minutes to make a note of the tonal adjustments you are making but

remember that while an EQ’d instrument may not sound correct in isolation,

what really matters is that it sounds right when run with the rest of the mix.

Cut EQ Rather Than Boost

Our ears are used to hearing a reduction in frequencies rather than boosts since

frequencies are always reduced in the real world by walls, objects and materi-

als. Consequently, while some boosts may be required for creative reasons you

should look towards mostly cutting to prevent the mix from sounding too arti-

ﬁ cial. Keep in mind that you can effectively boost some frequencies of a sound

by cutting others as the volume relationship between them will change. This

will produce a mix that has clarity and detail.

Don’t Use EQ as a Volume Control

If you ﬁ nd yourself having to boost frequencies for volume, you should not

have to boost by more than 5 dB. If you have to go higher than this, then the

chances are that the sound itself was poorly recorded or the wrong choice for

the mix was made.

Remember the Magic

A Q setting of

octaves has a bandwidth that’s generally suitable for EQ’ing

most instruments and often produces the best results. That said, if the instru-

ment is heavily melodic or you’re working with vocals, wider Q settings are

preferred and a typical starting point is about two octaves. Finally, drums and

most percussion instruments will beneﬁ t from a Q of 1/2 an octave.

Shelf EQ

Shelf equalizers are generally used to cut rather than boost because they work at the

extremes of the audio range. For instance, using a shelving ﬁ lter to boost the low

frequencies will only accentuate low-end rumble since there’s very little sound this

low. Similarly, using a shelf to boost the high range will increase all the frequencies

above the cut-off point, and there’s very little high-frequency energy above 16 kHz.

Mixing

CHAPTER 19

395

Fix it in the Mix

‘ Fix it in the mix ’ is the opinion of a poor engineer and is something that

should never cross your mind. If a timbre is wrong, no matter how long it took

to programme, admit that it’s wrong and programme/sample one that is more

suitable. The following chart indicates the frequencies for mixing:

Frequencies Musical Effect General Uses

30–60 Hz

These frequencies

produce some of the

bottom-end power but,

if boosted too heavily,

can cloud the harmonic

content, introduce

noise and make the mix

appear muddy.

Boosts of a decibel or so

may increase the weight of

bass instruments for drum ‘n’

bass.

Cuts of a few decibels may

reduce any booming and will

increase the perception of

harmonic overtones, helping

the bass become more

deﬁ ned.

60–125 Hz

These frequencies also

contribute to the bottom

end of the track but, if

boosted too heavily, can

result in the mix losing

its bottom-end cohesion

resulting in a mushy,

‘ boomy’ sound.

Boosts of a decibel or so may

increase the weight of kick

drums and bass instruments

and add weight to some

snares, guitars, horns and

pianos.

Cuts of a few decibels may

reduce the boom of bass

instruments and guitars.

125 –250 Hz

The fundamental of

bass usually resides

here. These frequencies

contribute to the body

of the mix but boosting

too heavily will remove

energy from the mix.

Small boosts here may

produce tighter bass timbres

and kicks, and add weight to

snares and some vocals.

Cuts of a few decibels can

often tighten up the bottom-

end weight and produce

clarity.

250 –450 Hz

The fundamentals

of most string and

percussion instruments

reside here along with

the lower end of some

male vocalists.

Small boosts may add body

to vocals, and kicks, and

produce snappier snare

timbres. It may also tighten

up guitars and pianos.

Cuts of a few decibels may

decrease any muddiness from

mid-range instruments and

vocals.

(continued)

PART 3

Mixing & Promotion

396

Frequencies Musical Effect General Uses

450 –800 Hz

The fundamentals and

harmonics of most

string and keyboard

instruments reside

here, along with some

frequencies of the

human voice. Cuts are

generally preferred

here, as boosting can

introduce fatigue.

Small boosts may add some

weight to the bass elements

of instruments at low

volumes.

Cuts of a few decibels will

reduce a boxy sound, and

may help to add clarity to the

mix.

800 –1.5 kHz This area commonly

consists of the

harmonic content of

most instruments, so

small boosts can often

add extra warmth. The

‘ pluck’ of most bass

instruments and click of

the drum kicks attack

also reside here.

Boosts of a decibel or so can

add warmth to instruments,

increase the clarity of bass,

kick drums and some vocals,

and help instruments pull out

of the mix.

Small decibel cuts can help

electric and acoustic guitars

sit better in a mix by reducing

the dull tones.

1.5 –4 kHz

This area also contains

the harmonic structure

of most instruments, so

small boosts here may

also add warmth. The

body of most hi-hats

and cymbals also reside

here along with the

vocals, BVs and pianos.

Boosts of a decibel or so can

add warmth to instruments

and increase the attack of

pianos and electric/acoustic

guitars.

Small decibel cuts can

hide any out of tune vocals

(although they should be in

tune!) and increase the breath

aspects of most vocals.

4 –10 kHz

Finger plucks/attacks

from guitars, the attack

of pianos and some kick

drums, snares along

with the harmonics

and fundamentals of

synthesizers and vocals

reside here.

Boosts of a few decibels can

increase the attack on kick

drums, hi-hats, cymbals,

ﬁ nger plucks, synthesizer

timbres and pianos. It can

also make a snare appear

more ‘snappy’ and increase

vocal presence.

Small decibel cuts may

reduce sibilance on vocals,

thin out guitars, synthesizers,

cymbals and hi-hats, and

make some sounds appear

more transparent or distant.

Mixing

CHAPTER 19

397

Frequencies Musical Effect General Uses

10–15 kHz This area consists of the

higher range of vocals,

acoustic guitars, hi-hats

and cymbals, and can

also contribute to the

depth and air in a mix.

Boosts of a few decibels

may increase the brightness

of acoustic guitars, pianos,

synthesizers, hi-hats,

cymbals, string instruments

and vocals.

15–20 kHz These frequencies often

deﬁ ne the overall ‘air’

of the mix but may also

contain the highest

elements of some

synthesizers, hi-hats

and cymbals.

Boosts here will generally

only increase background

noise such as hiss or make

a mix appear harsh and

penetrating. Nevertheless,

some engineers may apply

a shelving boost around this

area (or at 10 kHz) to produce

the bandaxall curve.

This page intentionally left blank

Mastering

CHAPTER 20

399

Mastering can be viewed as the ﬁ nal link in the music-making chain. After any

track has been mixed and processed, it should be mastered. Mastering a record

is important for any musical production since it allows your music to stand up

to the competition in term of balance, EQ and – more importantly – loudness.

If your record isn’t as loud as other records it isn’t going to sound as impressive.

However, it should be noted that mastering isn’t something you should attempt

yourself, especially if you plan on a commercial release. The importance of

employing a professional mastering engineer cannot be stressed enough. Indeed,

all record companies will have their music mastered professionally regardless of

the proﬁ ciency of the original artist, and it is fully recommended that if you value

your music, you should do the same. Simply because you produced and mixed

the music does not mean you are the perfect choice to master it. It is essential to

appreciate that there is a world of difference between mixing and mastering.

Whereas recording and mixing music is a deﬁ nite skill that requires complete

attention to every individual element of the track, mastering involves looking

at the bigger picture subjectively. This subjective viewpoint is something that

is practically impossible for the producer (i.e. you) to accomplish, due to the

producer’s involvement throughout the entire project.

The mastering process depends on the quality of the mix in question, but if

it has been mixed competently then it will usually involve equalization to set

the overall balance of the mix, compression to add more punch and presence,

and loudness maximization to increase its overall volume. The result will be

enhanced impact of the music and spectral balance that will compare well with

every other record on the radio and CD.

This overall balance is vital if you plan to release music for commercial sale,

as the public will expect the record to have a professional polish similar to

Mastering

CHAPTER 20

’ A failure will not appear until it has passed ﬁ nal

inspection… ’

Albert Einstein

PART 3

Mixing & Promotion

400

other commercial releases. When the sound of a record is right, you’ll sell more

records. This is more complicated than it ﬁ rst appears, though, and accom-

plishing it proﬁ ciently requires experienced ears, technical accuracy, knowl-

edgeable creativity and a very accurate monitoring environment.

Of course, some suggest that you could master it yourself and then test the music

on a number of systems to check that it is correct, but this takes time and no mat-

ter how experienced you are, it is doubtful you would achieve the same results as

a professional mastering engineer. That said, many artists will master their own

music if it’s only to be submitted to a website in MP3 format, to a magazine for a

‘review’ or to a professional internet record label. Consequently, some knowledge

of what is involved in the mastering process and how it is employed is useful, so

for this chapter, we’ll look at the basic theory behind it all.

To master any record you’ll need the appropriate tools for the job. These are

available as all-in-one units in both hardware and software form. The most cel-

ebrated hardware mastering device is the TC Finalizer that has all the mastering

tools you could require in one compact single rack unit. Alternatively, digital

audio workstations can be used to master music, and this is usually accom-

plished with an established wave editor and a number of individual master-

ing plug-ins such as Steinberg’s mastering tools, Waves tools or all-in-one units

such as Izotope’s Ozone.

More importantly, though, no matter how good the equipment it is extremely easy

to misuse it. If the equipment is used incorrectly, the sound will deteriorate and the

audio distort. This distortion will not be to the extent that you can hear it; rather it

will make the mix sound more muddled or reduce the stereo perspective.

MASTERING THEORY

As we’ve previously discussed in earlier chapters, the order of processing can

dramatically affect the overall sound. Thus in mastering, each processor should

be connected in a sensible order. Although this can depend on the music being

mastered and the mastering engineer, the most usual order consists of:

■ Noise reduction

■ Paragraphic equalization/harmonic balancing

■ Mastering reverb (although this is uncommon)

■ Dynamics

■ Harmonic excitement

■ Stereo imaging

■ Loudness maximizer

We’ll look at each of these processes in turn.

NOISE REDUCTION

Although any extraneous noise should have been removed during the recording

stages, occasional clicks, pops, hiss or hum can slip through, so it is important

Mastering

CHAPTER 20

401

to listen to the mix thoroughly and carefully on both monitors and headphones to

see if there is any extraneous noise present. This is where mastering engineers will

have an immediate advantage as they will use subwoofers and loudspeakers with

an infrasonic response. Without these, microphone rumble, deep sub-basses or any

jitter noise introduced with poorquality plug-ins can easily be missed.

Any audible noise present will need to be removed before the track is mas-

tered. Stray transients such as clicks or pops will prevent you from increasing

the overall volume, while any hum, hiss or noise present will also increase in

the volume when you increase the gain.

Ideally, the best way to remove any noise problems in a recording is to re-record

the offending part, as this will inevitably produce much better results than

attempting to reduce them at the mastering stage. If this is not possible, then

you will have no choice but to use noise reduction algorithms or close wave-

form editing. This latter method is especially suited for removing pops or clicks

that are only a couple of samples long and involves using a wave editor to

zoom right into the individual samples and then using a pencil tool, or simi-

lar, to reduce the amplitude of the offending click. This approach is only suit-

able for extraneous noises that occur over a very short period of a few samples,

however. If the noise is longer than this it is advisable to use dedicated reduc-

tion algorithms. These have been designed to locate and reduce the amplitude

of short instantaneous transients typical of clicks or pops and are available in

most wave editors or as a third-party plug-in. Depending on the amount of

pops and the mix in which they’re contained, Waves Click Remover or Waves

Crackle Reduction can be suitable for reducing or eliminating the problem.

Hum and/hiss are much more difﬁ cult to deal with and often require special-

ist plug-ins to remove. As the frequency of tape hiss (also the result of a poor

soundcard A/D or low recording levels) consists of frequencies ranging from 8

to 14 KHz, removing these using a parametric EQ can result in the higher fre-

quencies of the mix also being removed. Thus, it is important to use plug-ins

speciﬁ cally designed to remove it, such as Steinberg’s De-Noiser. Again, the reli-

ability of using these types of plug-ins will depend on the source, and if they

are not used with some degree of caution they can ﬂ atten the overall sound of

the mix.

A.C. hum is little easier to deal with than hiss, as much of the ‘noise’ is centred

around its fundamental frequency that lies around 50 Hz (UK) to 60 Hz (USA).

Using a parametric EQ set to a thin bandwidth with a maximum cut can reduce

most hum, but there may also be additional harmonics at 120, 180, 200, 300,

350 and 420 Hz, and removing any of these can compromise the overall tonal

balance of the music, muddying the lower- and mid-range frequencies. Thus,

if a parametric EQ does not remove the problem, it’s worth using a dedicated

plug-in such as Waves X-Hum .

Clicks, pops, hiss and hum can also be removed, sometimes much more reli-

ably, by using plug-ins or wave editing features that proﬁ le the noise. This is

PART 3

Mixing & Promotion

402

often referred to as fast Fourier transform (FFT) ﬁ ltering. A section of the track

that consists of just the noise it can be sampled by the FFT ﬁ lter to create a

noise proﬁ le that can then be applied over the entire mix to remove the extra-

neous noise.

Waves X-Noise and Sonic Foundry’s Noise Reduction plug-ins are typical examples

of this, but some wave editors may also offer this function. Notably, these do

not always produce the desired results, and in some instances they can remove

frequencies that are essential to the music, resulting in a mix that has no bot-

tom-end weight or a ﬂ at top end.

Above all, any form of noise reduction is not something that should be

approached lightly. Reducing any noise within a ﬁ nal stereo mix will often

affect frequencies that you may wish to keep. This is especially the case with

removing hiss or hum. In fact, if there is an obvious amount of noise that can-

not be removed reliably using FFT noise proﬁ ling or plug-ins, then you should

seriously consider re-recording the performance, seeing if a professional mas-

tering engineer can resolve the problem or drop the project and start a new

one. It is much better to write a new record than to attempt to release one that’s

below your best capabilities.

HARMONIC BALANCING AND EQ

Assuming that there is no noise present, or that it has been removed pro-

ﬁ ciently, the next step in the mastering process is harmonic balancing.

Essentially, this consists of EQ’ing the record to achieve an overall pleasing

tone and an overall tonal balance that is sympathetic to our ears.

The quality of the EQ being used obviously plays a large part in attaining this

balance, and many mastering engineers will choose EQ units that have a pleas-

ing resonant quality, helping to give an overall satisfying tone. These are not

necessarily complex parametric or multi-band graphic EQ units, as any precise

EQ adjustments should have been made during the mixing process. Rather,

during the mastering process wide Q settings with gentle transition slopes

(sometimes as low as 3 dB per octave) are used, as these are more pleasing to

our ears.

When mastering, Q ’s of 0.4 –0.9 are the most popular settings.

Generally speaking, mastering EQ can be broadly split into three different sec-

tions: the bass, the mid-range and the highs.

Bass

The only ‘real’ EQ that’s required on the bass end of a mix is a low-shelf ﬁ l-

ter to remove any frequencies below 40 Hz. Frequencies lower than this cannot

Mastering

CHAPTER 20

403

be reproduced by many loudspeakers. When a signal this low is sent to them,

they will attempt to replicate the frequencies, resulting in the bandwidth of the

speaker system being restricted and the overall volume of the mix dropping.

When a high-pass shelving ﬁ lter is set to remove everything below 40 Hz, the

sub-harmonic content of the sound will be removed by opening up the band-

width of the speakers for a higher volume and result in an improvement in the

deﬁ nition of the bass frequencies.

The bass frequencies may also appear quieter than on commercial records,

but it is important not to use EQ boosts in an attempt to add anything that

isn’t already present. The same goes for any maximizers. Although bass-speciﬁ c

maximizers such as Waves MaxBass can be used to introduce additional har-

monics based on those already present, these are best avoided at this stage as

compression can be used to introduce additional low-end punch to the music.

If, however, there is a distinct lack of deﬁ nition in the low-end frequencies,

they will need to be pulled out using EQ.

In most dance mixes the kicks main energy will lie around 100 –200 Hz.

Perceivably boosting this will only result in muddying the bottom end. Rather,

it’s the transient that needs to be more deﬁ ned and this commonly resides

around 1000 –2500 Hz. Thus, you will need to notch out the frequency that lies

just below the transient so that it brings it out of the mix slightly. Possibly the

best way to ﬁ nd these transient frequencies are to set a very narrow Q with a gain

boost of around 6 dB. By then sweeping through the frequency range you should

be able to locate the required transient. Once you have located it, you can reduce

the gain to around ⫺4 dB and sweep to the frequencies just below this.

Mid-range

The mid-range is the most common source of problems during mastering, so

you need to listen carefully to these frequencies and then identify any prob-

lems they may have. The best way to accomplish this is to compare your cur-

rent mix with a professional CD of the same genre using your ears and a good

spectral analyser. Take note of the frequency dips, if there are any, and then try

to employ these to your mix to see if it makes a beneﬁ cial difference.

Similar to mixing, the mid-range problems can be divided into three catego-

ries: it’s too muddy, too harsh or too nasal.

If the sound is too muddy, then, using a fairly wide Q, place a cut around 180 –

225 Hz of a few decibels so that the frequency range between 100 and 300 Hz is

dipped with the largest cut centred around the current frequency setting. Then,

listening carefully, reduce the width of the Q until the muddiness dissipates.

If the overall sound of the mix appears too harsh, using the previously

described principle place a cut at around 1.8 KHz so that there is a transition

slope down from 1 KHz to the centred frequency, and then a transition slope

up to 3 KHz. Then, once again start reducing the width of the Q setting until

the harshness is removed.