TIf,E E'ATRTTGIIg

EXPEATSUED
When the Fairlight CMI was released seven years dgo, both the machine
and its software represented a significant step forward in the application
of computer technology to music. Today, the Fairlight is used in the
rnaking of popular music the world over, as well as performing an important
role in the field of musical and technological education. Despite this, very
few people are fully aware of what the CI}II does and how it does it. Jim
Grant, who's been working with one for a number of years at the London
College of Furniture, has decided to rectify matters by writing a series
dedicated to explaining the Fairlight's workings. Part one appears below.

E & MM AUGUST19 84
 efore the Fairlight'sappearance,
most computer music systems
were the perogativeof mainframes
and th e i r c ont r ibut iont o th e w o rl d o f 0r , 1il8N n
everyday music was slight. Kim Ryrie
(the Fairlight'sfather)and his Australian P IID E I INDEX
colleaguessoon changedall that, how- rnbL < DISK DONTROL
ever and their inventionis now used in rNbL J K E Y B D H R DC O N T R c l L
almost area of music production,to the PRGE4 I I I I R I '1 OI C
N ENUELDPES
extent that many people appreciateits PIi|jE 5 I , I E U E F O RE I l E N E R B ITO N
sound without realising that they're
f n l - 'L b R f i I ,I IN C
I , I H U E F I ] R I I'1
lis te n i n gto ' m us ic by num b e rs ' .
However,despiteits widespreaduse, I|1 IRLI [ HT PBriE 7
P R I J ED
c c | N T R | | LP F R R I l E T E R S
S | J U N DS R I l P LI t I | ]
thereare relativelyfew Fairlightsin general
.100 PFDE S SEUUENDER
circulation- less than in the UK -
and to see one in actionat close quarters PH6E FI f f N f f L O GI N T E R F H I E
is a realtreat.Hereinliesthe rationalefor PBl]E [, C O I 'I P O S E R
this series of articles.What does the PFEE D l , I H U E F O RD HI S P L F I Y
Fairlightdo? How does it do it? And PBriE L DI S K L I B R F R Y
what can the averagemusiciando with I l,l.I PrirlE R R E R L - TI I l E C D 1 4 P | ] S E R
it ? u3 , t 5 , Rl , l l TN b L 5 S C R E E NP RI N T
I.JS E R N FI14E, J I l l DRRNT
Hardware
To take deliveryof a Fairlightleaves l@t
your bank balanceempty and your living
room full. The hardwareconsists of a
Central ProcessorUnit (CPU),one - or
optionallytwo - six-octavekeyboards,a 011f lf f t i D '
typewriter-styleQWERTYkeyboard,and
a VDU with added lightpen.ln addition, DI SK, E X E I l P L E FREE SPHCE. 536
there are some long connectingleads,a USER. J I T '1G R f l N T LIBRffRY NO. 8
Systemsfloppy disk drive and a box of 1 I NU E N T . I N I6 STIDK C|]
2llit r I lE r m t7 ItiuENT .S0
disks containinglibrarysounds. 3 PRE L U D E, I N I8 P R E L U D E. S O
4 UI U F L D I , I I I I9 UIURLDII.SO
5 Bf iS S G T . U C ? B LOCUST .RS
6 BDD 2 UC 2I PBGEI .PX
Software 7 CBE L L O ( ] [ U
} C E2 PFGEz .PX
8 G RN G S R U C 2 3 PFGES ,PX
A foolproof system of connectors- S STIC K .UC 24 SYHTH .PC
and a quick glanceat the manualon the 1B STR I I'ID UC 25 SYNTH . PT
part of the user- ensuresthat the Fairlight 11 I f r N N UC 28 BFSSD .SS
T2 ERS S G T C O 2 7 CNTRL .SS
can be powered up in no more than five 13 BBD z . . .C0 28 SN3 .SS
minutes.The VDU displaysthe expectant 14 CBE L L O O O . C O2 9 TUHEl ,SS
I 5 G RN G S R . C O 3 6 TUNEz ,SS
m e ssa g e' CM l RE A DY ' ,w h i l e th e C PU
hums quietly:thereare threefans pulling I l U L TI CIi N D EL
air throughthe innards,keeping500 watts TRBNS F E R DELETE EUERY clt r EII tlcL
of power dissipationdown to an accept- !.qgH$I .,-l|:| .tvee:-ln$IRU|]1ENi , 24. .
.,, ,ye| l, ..:i,,r,e
abl e te mp er at ur e. . . . v o i c e s 'B D D 2 STICK BFSSGT STR|4
ID
lnsertingthe Systemsdisk in the left- T
hand drive(Drive0) resultsin a faintclick
as the stepper motors engage. The l@
operating software is loaded as a series
of 'fetches'- each section of program
loaded pulls in the next section.When
this process has been completed,the
user is faced with the Index page. See
F i g u re1 .
Here lies one of the Fairlight'smost
powerfulfeatures.The whole system is
menu-drivenand the differentoptions
correspond to different VDU displays
and sets of commandswhich are entered Com'irter
from the alphanumerickeyboard.Each rcthsb@rd
option is referredto as a Page. A Page
has one or more files resident9n the
Systemsdisk which are loadedwhenthe
Page is selected. Page 1 is the Index
Menu itself (Figure 1), while Page 2 Address
managesthe files stored on the disk in bus

the right-handdrive(Drive1).Thesefiles
are user-created, and there are seven
differenttypes, as indicatedby the suffix
afterthe file name.Theseare as follows: Cmlfol
crrcurlry
NAME.VCis a voicefileoccupyingabout
wovelorm RAM
2OkBytes.lt holds waveform data (1610
and extra informationregardinglooping dio signol
and so on. oulpui
NAME.CO holds control information
such as portamento,vibrato frequency
and depth.
28 AUGUST1984 E&MM
 channel cards perform this function
autonomously. The computersectionof
the Fairlightpassesparameterssuch as
pitch,vibrato,portamentorate and loop-
ing points along its data bus, and once
sheet, 2 IfIIiIIEIIIIE and types these have been received,the channel
3 F fim SPRCE o n d isk card outputs the sound until the para-
3. lfnfilfli$ number meters are updated.
3 EfiffilflIn tile s { t4 UL T I & C B N C E L) Overallpitchingof the CMI is deter-
flllffiE d isk, u se n o r f i ]e nenret mined by a systemclock residenton a
4 EIIfII co m m a n d soecialcard known as the Mastercard.
c EN:III| IE in stn u m e n t o n vo ir e fi l es We'll be referringto this on numerous
m l!fiflIn in str u m e n t f ile s
f
occasions over the next few months,
6 I[ll| ifflE vo ice f ile s
since it holds the circuitry for a good
b ]m :filfln sEq u e n ce f ile s
many of the CMI's functions.A 34MHz
7 |lJifllfll somnrend
6 If#IIEiIfE cor'mahd
oscillatoris onboard,and this is divided
I nFffiff:dmftfmlfi pnotect i on
and fed to the individualchannelcards.
1 8, Im m L Ef0 o r r m tsste r k e yb o sr d It's from this clock that the RAM clocking
it
m fiil:Im sh o r tcu ts rates- and thus keyboard pitches - are
generated.The wholeinstrumentcan be
toucn sny l0 X r ^ r ith L IGHT PEN o r type, n(set) tuned by scalingthe masterclock.
u h erE : n = sneel
t ouc h lI f f i or t vP e ' HI ( n e t u r n )

EE!tr4 Pa g e2 Co m m a n ds
Lookingagain at Figure 2, there are
severalcommands at the bottom of the
display.TRANSFERallows files to be
copied from one disk to another,using
TRfiNSFER FILES T O BNOT H
Drive0 as the destinationdrive.This is
YPE. L I GHTPEN , essentialfor creating backup copies of
T,file (retu nn ) (select files) importantmusicand/orsounds.DELETE
or <T R f i N S F E R ) erases unwantedfiles to make room on
T, f ile, f ile, l ile( , . . et c ) ( r et ur n)
r,rhene , file = FiLENEI I E SF or f* or * * - *l or I the disk. Invokingthis command prompts
(s s m e as DELETE; s ee abov e) a confirmationmessage to prevent ac-
cidentalerasureof importantfiles.When
E X RI'4PL ES, T. CH0RUS,I N( r et unn) a file is deleted,FREESPACEincreases
T, 4-1 8, FBLE. UC,2S( net ur n)
F i Ies u i I I b e copied FR0l' 1dis k in RH dniv e ( D I S K f i ) T 0 a n o t h e n by the deletedfile size.
disk in LH d nive ( DI SK B) . G iv e TRRNSF E Rc o t n m a n du i t h s v s t e m At the very bottom of the displdy is an
disk in LH d nive and DI SK R in RH dr iv e. l,lhen the messaEe' exampleof the QUERYcommand.This
PL RCE FI LE DI SK I N LH DRI U E
E pp €tsFS, p lace D I SK B in LH dr iv e. l, lhen t h e t n a n s f e r is tells us that LOCUST.INfile will auto-
cornple ted a fin al m es s age r r i I I neques t t h e r e p l a c e m e n t o f t h e maticallyload eight voices, whose names
system d isk. If a f ile alneEdy ex is t s on D I S K E ( h a s s e m E are shown.
name a nd suf f ix as f ile on DI SK R) , it ui l l N0T be ovenunitten
wit ho ut you n con s ent . See als o PROTECTION.

TYPE, T, t ( net ur n)
Help Pages
l,lhe n the me ssege eppeEr s plac e a EEI I t r f i 1 e d i s k i n L H d r i v e . By this time,you'reprobablywondering
R epla ce syste m dis k r , r hen c om Plet ed. Ne r *rd i s k u i I I u s u a l l y how anybody using a Fairlight ever
shor.r an incne ase in FREE SPFCE av ai lable managesto rememberall the commands,
especiallysince we've only considered
@ Page 2 and there are another 13 still to
go.
The answeris simple:Help Pages.
NAME.INconfiguresthe CMI to a parti-
cular instrumentstate. Voices are auto-
GhannelCards Figures 4 and 5 show examples of
maticallyloaded and spread across the Page 2 Help Pages. In fact, the entire
Insidethe Fairlight,thereare usuallyat user's manual is held on the Systems
keyboard.
least 16 circuit cards, the exact number disk.and sectionsrelevantto the current
NAME.SQ holds polyphonic keyboard
sequencerinformation. depending on various options such as display Page can be inspectedat any
NAME.RSis a realtime sequencer(Page an analogue interface and sync card, time by typing HELP(what else?).
R) file. and eight of these are known as voice or Initially,an index sheet is displayed
NAME.PXcorrespondsto a screendump channel cards. (Figure4).Touchingany of the highlighted
(Page S) to disk. This can be spooled The Fairlight produces sound by a options with the'lightpenresults in the
later to a dot-matrix printer for hard copy. processcalledWaveformSynthesis.Each Help sheet specificto the selected option
NAME.PC,PT,SS are MusicComposition command that deals directlywith sound being loaded and displayed. The user
Language (MCL) files. These are gener- generationmust involveat least a section can flick backwards (BWD), forwards
ated on Page C and hold text files that of a waveform. The waveform itself is (FWD)';or recall a previous place (PRE):
describenotes with duration,dynamics held in 16K of RAM on each channel commandscan be enteredfrom the Help
and so on. card as a direct digitalrepresentation,so sheetswhile viewingtheir correct format.
A voice file can be loaded in a number that increasing amplitudes give larger The CMI then automaticallyreloads the
of ways. Probablythe easiestis to point binary numbers. Therefore, when an display page that called the Help sheet
the lightpenat the volce name and then eight-note polyphonicsound is present in the first place, and executes the
at the command LOAD at the bottom of on the keyboard,eachchannelholds the command.And yes, there are even Help
the display (Figure2). Drive 1 springs same voice data. sheetsthat explainthe use of the Help
into action immediately,and after a Put simply, the channelcards can be sheets.. . .
secondor two the selectedvoiceappears regarded as digital oscillators whose That about wraps up the first part of
on the keyboard. waveformis determinedby the contents what will doubtless become a saga of
So far so good. But where does the of 16K of RAM (Figure3). Differentpitches some duration. Next month, we'll take
voice informationgo, and how does it - as playedon the keyboard- conespond a look at Page 3 - the keyboard map -
result in a sound when the kevboardis to the MM"information being read and and the waveformdisplay page, Page D.
played? converted by a DAC at differentrates;the Jim Grant E&MM
E&MM A UG US T 198 4 29
 TIIE T'AIRLIGHT
E)KPEATTUtrD
The secondpart of our insightinto one of the world's most popular
computer instrumentslool<sat displaypagesand what they tell CMI
owners.Jr* Grant
ast month we describedthe gen-
eral software conceot of the Fair-
light and how much of its powerlies
in its abilityto presentits functionsto the
user as a group of related files called
Pages.We saw that the CMI poweredup
with the Index Page (Page 1) and that
=--*.u+'\\S..S*\
soundand music fileswere managedby
Page 2.
One of the oroblemsassociatedwith
most sound-generatingequipment is
the way in which the sense sight is
excluded from the orocess of sound
\\R
'\\- N \.\._>

formation.Most instrumentsoperateon
inLrL;,{ .IUII-L
.:RUIlF ET
rlll|1Hfitr
fl lIrE 4
lltsuEFrJPF Ir Ir:,FLtii'

l,,.lltil;f, 1
T t:t_ll,tFT,t
l'1t-tIt E ,
tJi i t,l E Fr :r Ftt1
Ir I:iF LtlT
F r:rt'l ''1 H T ,HE tliil litlj--ffEl E;4 :ii !l-El- L E- + ir I

B6 S E P T E M B E1
R9 84 E&M M
the basis of the user twiddlingthe con- nel allocationand keyboardmaps. unique sound, and therefore the maxi-
trols and stabbing the keyboard. Fair Figure5 shows a typical displaywith rhum number of notes that can be play-
enough: of all our senses,the ears are eight separate voices loaded into the ed on the keyboard with any single
by far the most acute. Yet if we consider CMl. RegistersA to H are groupsol one sound is one. This is indicatedby the
all the electronic music equipmentcur- or more of the eight Ch.annelsand corresponding'NPHONY'. lf a single
rently available,the most user-friendly 'NPHONY'is the numberof notesthat eight-notepolyphonicvorce is required,
instrumentshave graphic displays,per- can be played with the sound held in a only RegisterA will be activeand Chan-
haps in the form of panel legendsand Register.A quick look at Figure5 reveals nels 1 to 8 will be allocatedto A. The
LEDs or liquid crystal display. Despite that there are eight active Registers, active Registersare mirrored on Page 2
the fact that sight is a very poor quali- each holdingthe voiceindicated.In this so that they can be loaded with sounds
tative sense, it can be of enormous case, all eight Channelcards hold a from disk. The Fairlightwill flag an error
psychologicalhelp in our field.Basically
it boils'down to: 'lf I can see it. I can T EU H PET
understandit.' f4 u l r E, 4
I,IFU TFD R I'1TII i ;P LR Y
:J.r STEP.l E 4
D^ ^^ r-\
r d.Bu, u
There's no doubt that the Fairlight's
visual presentation is founded on this
premise.Each display Page is graphic
withoutbeingostentatious, and PageD,
the voice waveform display, is a prime
exampleof this. Typing PDfollowedby a
RETURNon the alphanumeric keyboard
will resultin a displayof the type shown
in Figure 1. To appreciatethe signifi-
cance of the display,we must delve a
little deeper into the workingsof the
C MI.
Rememberthat voice informationis
held in 16K of RAM on each channel
card. To simplifymatters,the CMI div-
ides the memory (and thus the wave- INIlEX
form) into 128 sectionscalledsegments. 0 l' ll,lffN Ir,
Each segment consistsof 128 bytes,so
the waveform comprises128 segments
multipliedby 128 bytes to give 16384 fi I I lIr!a
5 2 i Sr I Cll
bytes,ie. 16K. This savesthe musician t' 3 ) BH5:,r.iT
handling unwieldy computer numbers D { i STRI,I I II 1
when dealing with the waveformRAM. E 5 I 'S T F ; 1 4 I D 2
The display shown in Figure 1 is a F 6 r c8ELL00q
psuedo-3D representationof the voice G 7 ) TffNN
H P I rlHNr:SF
called TRUMPET.Each line from left to
right is a segment, and the foremost
segmentrepresentsthe beginningof the
sound. When a keyboard note is :l ,l',1i{::rTEF: FITr::H, l?U
pressed, the CMI reads out the RAM E .!JLffUE si:FLE, t(lT
a 'j
information segment by segment from
the front to the rear of the display.
There are two display formats,A and
B, and a numberof optionswithineach
type. Figure 1 is in formatA, and seg-
ments1 to 128 are shownin stepsof 4.
Figure2 is againformatA, with the end
segment number 32 and steps of 8:
OHHfiND '
thereforeonly five segmentsare shown.
Format B gives an oscilloscope{ype
display,but with each segmentslightly SHXY Lg25
above the precedingone. Again,there SNfiRE 886
are a numberof displayoptions.Figure BDRU14 ?,48
3 shows TRUMPETsegments1 to 128 ORGtiN BAVJ
in steps ot 2, and Figure4 segments1 to
64 in steps of 8.
AlthoughPage D is purelyfor display
purposes and does not support any
sound creationcommands,it's still an
invaluableaid. At its most basic level.it
3 .I,lFSTER PI ICH' I28
answers the ouestions'where has the
6 .SLFUE SCFLE, l ? r -:;
sound gone?' and 'is the waveform V <.U U
zero?'.
l\^
rage 5
Page3 is anotherutility-type
display
Page.lt dealswithvoicetunings,
Chan-
S E P T EMBER
19 84 E&M M
 MUSICIAN
OMPUTER
message if you try to open another to each octave.As the C[/l is a music- qui tean el egantsol uti onsi n ce'm usicby
R e g is t eror inc r eas eth e N PH O N Yb e y - ian's instrument,it supports two six- numbers'requi resl ots of num bercr un-
o n d e ight .A s im pler u l ea p p l i e sth
: es u m octavekeyboardscalledthe Masterand ching, but the other choice, and one
of the active Registers times their the Slave.Usingthe maps,it's possible w hi chi s becomi ngi ncreasingly popular ,
NPHONYmust be less than or equal to to create eight differentkeyboardcon- rs computrngconcurrency.In a basic
eight. figurationsby choosingwhich sounds CMI system, there are four micropro-
Figure 6 shows another exampleof will play on each octave within a key- cessorsof the 6800family.Two of these
the Registerallocations.Here, another board. The Master and Slave can be are i n peri pheral s,i e. one each in t he
set of voices has been loaded.so the linked(as shown in Figures5 and 6) to music keyboardand the alphanumeric
NPHONYand Registersare configured any map by changing the Selection keyboard,and this meansthere can be
differently.Although the Registerand numDers. several independentprocesses being
NPHONY settings may seem a little Theinformation presentedin Page3 is executedconcurrently.
confusing and limited, the exact con- knownas an Instrument file.The filecan The microprocessorin the alpha-
figurationis determinedentirelyby the be savedon the user disk and is given numericscansthe keys and passesthe
musician,and changescan be effected the suffix NAME.IN.When this file is data to the music keyboard when
very quicklyfor evaluation. loaded it will pull the specifiedvoices requested.At the same time, the music
into the CMl, allocate the Registers keyboards are scanned for pressed
automatically,adjust the tuning and notes,key velocitiesare calculated,and
andTuning
Keyboards spread the sounds across the key- the controlslidersand switchesread.
Any voice can be tuned in increments boards. Instrumentfiles are a usefully At the right-handend of the Master
of plus or minus one hundredthof a quick way of bringingthe CMI up to a keyboard is a calculator-stylekeypad
semitoneup to t6 octaveswith crystal playablestate with 'preset'voicesand and alphanumeric displayused for rapid
accuracy. Scale allows the Western tu n i n g . loading of voices in a live situation.
temperedtuning of 12th root of 2.00 to rt I Music keyboard information has the
be changed to any other macro/micro 11arcware highestpriorityof all data in the CMl,
tuning,eg. for quartertones,you simply At the timethe Fairlightwas designed, which respondsinstantlyto the packets
changeScaleto 24th root of 2.00.Pitch the microprocessorwas consideredto of data sent flying down the cables at
is a master tuning control which can be a medium-to slow-speeddevice.To 9600 Baud.
vary tuningof all the loadedvoicesby a increasethe power of any computing Well that about finishes off our des-
quarter of a tone in 256 discrete steps, system, designers have two basic criptionol the utility-typedisplay Pages-
to bring the CMI in tune with other choices. In case you'rewonderingwhat the Mode
instrumentsif necessary. Oneof these(theSynclavierapproach) settingon Page 3 is for, don't worry: all
The KeyboardMaps each consistof a is to base the instrumentarounda dis- will be explained.
keyboardnumber (1 to 6) followedby six crete logic minicomputer, thus utilising Next month, the controls on Page 7
lettersindicatingthe Registerassigned the raw speed of logic chips. This is and samplingon Page8. r
 MUSICIA
OMPUTER

TIIE T'ATn.LTGIIT
EIKPEATTUED
Partthree,anda discussion
of how the CMI samples
a soundandwhy its
own particularsampling
techniquesare employed.Ji* Grant
A t last we've coveredenoughof the CMI resultsin a short sound when the samole is than can be looped is known as a Segrpeit
/.\ basics to concentrate on the more played on the keyboard,and this becomes Here lies the crux of choosing a suitable
, \ interestingsoundcreationPages.Now, shorteras we ascend the octaves.To over- SampleRate.lf we attemptto loop a Segment
the single feature that characterisesthe come this,the Fairlightallows sectionsof the or group of segmentsthat doesn't contain a
Fairlightin manypeople'smindsis its abilityto waveform RAM to be read out repeatedly (or whole numberof cycles,the 'ends' of the loop
sample natural sounds: this aspect is dealt looped)as the key is held down, thus sustain- won't join up withoutcausinga suddenlumpin
with by Page B, and is surprisinglysimpleto ing the sound.The smallestsection of RAM amplitude.Choosingan inappropriatesample
USE.
At the rearof the CMI liesa selectionof line U D Ir - 'E, I
and mic inputsto suit most applications.That
0 t' ll' ll i N Ir, I'rHFiI l'1:1
about takes care of the hardware.because :l r-rl _l H DtH l 'l FL I H rJ
everything else is dealt with by software.
Typing 'S' or touching 'Sample' with the S R I'IP LE FTIi TE Zv I EB Hi E H I,IP LE LE U E L c::.
lightpeninitiatesthe samplingprocess.Aftera FILTE R LI:II,I I TR I GI]E F; LE U E L :
second or so, the Display box shows the FILTE R H IGH E TFIII-;i j E FI TIE LB \. E tl i
sound envelopefor quick monitoringof input Ii t'l FFrE i gr:rFi I 'F F
levels (see Figure 1). lf all is well with the
Keyboard Maps on Page 3, the sampled S fi I'IP LE
soundwill be playableon the musickeyboard.
So far so good. But what are the other IIIS P LFIY
functions for? Well, some of them are self-
explanalory. Sample Level is a software-
based volume control and can be used to
attenuatesignalsthat exceedthe input range
of the Fairlight.Unwantedfrequenciescan be
reiected by using digitally-controlledhigh-
pass and low-pass filters:their cutoff points
are set by FilterHighand FilterLow.Theactual Figure1.
circuitryliveson the ubiquitousMasterCard,
and takes the form of switched resistor
networksusing the much-lovedCMOS 4051 I ilt t : ,
Ul' 1l( l l i {It, R TU N E
chio. l 4 0 D E, 4
Whenevera signalis convertedto a stream I , ] F U E F | ] R I!'II 9 P L H Y
of digitalnumbers,it's necessaryto bandlimit
it to one half, or less, of the Sample Rate.
Statedsimply,this meansthat we musthaveat
least two samplevaluesof the input signal's
amplitudefor the highestfrequencypresent.lf
this conditionis not met. the informationthat
the sarnpling process has captured is not
sufficient lo reconstruct the original signal
without frequency distortion. This type ol
distortion is known as 'aliasing'and is both
extremelynoticeableand rather unpleasant.
The CMI guardsagainst'aliasing'by incor-
porating tracking filters controlled by the
SampleRate.

SampleRate Figure2.
Althoughsamplingitselfis verysimple,and
impressive results can be obtained very | I ll IrE:i - U i J IU E . 1
quickly,it's well worth the troublespenrjing r-rtlt1HflIl L Tt- tt{E
lt
some time adjusting the Sample Rate to a l l D Ir E, .{
I l ,l H rl l E r-1Frl
F 'l MLx'i .
tr
valuethat suitsthe pitch of the input signal. FIFJ'1Fi
T+ 11Ir ::iE t:; ::r?
l-t IiFtfii 5TEF, I

aspectof theCMrthatobtiges
imporrant usro ffili,iii?Fr
payattentron
to thecorrecr^:T?1"^r.u^t:^-
The number of original samples taken ,^
is [l:liili.#,
llii',:;ttlj,, :-. r' --rli./ .,'
fixed and equalsthe length of the waveform l--ts
[,1]lll'*ni:L1:----.=
RAM,ie 16334.The fasterthesesamplesare
taken,the shorterthe durationof the sound L''Figure3.
becomes (althoughthe fidelity increases).This

86 OCTOBER1984 E&MI
I
1

l
OMPUTER
MUSICIAN
rate resultsin a dreadfulglitchwhichincreases conect clocking rates requiredfor digital{o- shortly before the signal to be sampled and
at a rateproportionaltothe pitch playedon the analogueconversionwhen a keyboardnote is used insteadofthesignalitselfto trigger the
keyboard, lf the sample rate is almost right, a pressed. However, for the duration of the samplingprocess.Trigger Delaycan then be
one-segment loop produces a sudden slight samplingperiod the CPU grabs Channel 1, used lo define the precise point at which
pitch shift, and waveform crests and troughs and forces it to produce a streamof pulsesat a samplingwill actuallybegin.This is extremely
'drift' laterallythrougha PageD display.This is frequencyof 128timesthe samplerateshown usefulfor soundswith a gentleattack such as
shown in Figures2 and 3, where a drift to the on Page8. This is suppliedto the ADC, which slow strings.
right (sharp)is caused by the sample being set resides on the Master card, and once the Lastly,the Compressoris a softwareswitch
too high and a drift to the teft (ftat)by it being samplingprocess is finishedthe CPU restores which controls a hardwareoption. Basically,
too low. Channel1 to its originaltask. this turns the conversionprocess into a non-
Figure4 shows a sound which is in tunewith TriggerLevel is the amplitudethresholdat linearsystem,thus enhancingthe dynamic
the system and therefore loops perfectly. At which the sampling process is triggered to range.The electronicsuse the same type of
the other end of the scale, if the sample rate is begin.When the Sample command is given, circuitryas that in manyanaloguecompanding
totally wrong the display becomes a hopeless the system waits until this level is reached systems. However, very few Fairlights are
jumble (Figure5). The relationshipbetweena before proceeding.Once the threshold has fittedwith this option as it can have a strange
whole numberof cycles and each segmentof been exceeded, it's possible to delay the effect on the commandson Page 6.
waveform RAM is also the relationship conversionby using the TriggerDelay,which Well, that about wraps it up for Page B.
required for a visually coherent display. Thus has a rangeof 0-65533 milliseconds. Thiscan Thereisn't room this monthfor a discussionon
samplesthat look good will inevitablysound be especiallyusefulwhen samplingfrom tape, Page 7, so we'll have to leave that for next
good, too. for example,as a tone burst can be recorded month. r
Now, if all this sounds rather complicated
and you're beginningto wonder how anyone IiLEi; I
gets anywhere near choosing the corect tr lJ8 ilIl
IT U I I L .l
sample rate, then take heart. lt's all in the help l , l S u E F r t P i Il r I : : F L f ri ,
pages for Page 8 - see Figure 6. A useful
sample rate table is included,and with a little
practice it becomes quite easy to arriveat the
conect setting within the space of a few trial
samples.

ADC
The actual analogue-to.digitral conversion is
accomplished by a 10-bit converter, even
though the CMI is an eight-bit machine.Only
the top eight bits of the sample values are
stored, while the two LSBs (Least Significant
Bits) are ignored.This improves the linearityol
the conversion, which means that the signal
step size required to cause a conversionvalue
to change by one LSB is fairly constant over
the range of the ADG. Figure4.
/
The relationship between the amplitude of
the input signal and the sample values I't I
generated is linear. When the signal level 0 l' ll' lF NIr' JTUNE
changes by a given amount inespective of the t '1 0 0 E' i
I,IITUEFcJR
DI 'IS F L R Y
absolute value the conversion code always F|:l R I'18T.F 4 32 S T E F '1 2
changes by the same amount. This is where
the Fairlight differs from most other sampling
machinessuch as the Emulator.That uses a
non-linear conversion method (called
'companding')which allowsmore codesto be
generatedfor small signal valuesthan for large
ones. Whenever a signal is represented by a
finite range of numbers - in this case f255
(eight bits) - two things suffen noise and
dynamic range.The noise is only heardwhen a
sampled sound is actually being played
through a DAC, ie. the ADC and DAC are not in
themselves inherently noisy. Making sure that
the peak ol the input signal causes the
maximum ADC code to be generatedensures
that most of the noise is masked by the volume
of the signal on playback.
The Displaybox in Figure1 is an invaluable
aid in this respect.
Dynamic range, on the other hand, is a
measure of the range of different amplitude
values that the ADC can handle. In a linear
system, this is directly related to the numberof
bits used in the process and, roughtyspeaking, S ffI,l P LE R Il TE 9
the dynamic range of the sampled signal is noi e +gvts -E vr -tE : ,"'i
6dB times the number of conversion bits.
Since the Fairlight uses eight bits, this gives 8 14817 29S35 74SS 3?t3
E 15884 TiiEr! : j B 5I
x 6dB = 48dB dynamic range. Companding c 1 8 7 4 4 ', Er:r7e 4 I FE
techniques result in a larger dynamic range 17748r 8r:rlB 4 4 : {5
(about 70dB)forthe same numberof bilsused, r sTss 3337 4EtS
but at the expense of greater noise at low tsst2 iig5E 4 S 7 'i l
signal amplitudes. The reasons for Fairlight's E 2I BSB 1Erg49 5 t 7 '{
choice of a linear converter will become more F 22351 I I I I i, i i ,,:,,1

apparent when we look at the functions on ?3EBB lltr4El SrllF

I U5 4 4 Etr?
Page 6. t3t38 EE45
The actual sample rate is very cleverly
generatedon Channel card 1. Normally,the
onboard circuitry is used to generate the

E&MM OCTOBER1984 6l
 u..:=;

THE EATn'LTGHT
fi
t
E

tr)KPEATTUtrD
just
How the CMI providesfor specialeffectsandlooping- in a language
about everyonecan understand.
Ji* Gra,nl
I ast month we dealtwith one of lhe active voice is shown in the top right- resultsfrom its use. The Filteris a lor'.,-
I Fairlight's main sound creation hand corner.Other controlfiles can be pass tracking filter resident on each
L methods:sampling.The simpleact inspectedby pointingthe lightpenat the C hannelcard, used to at t enuat eanv
of pointing a microphoneat a sound names in the displaybox or by typing unw anted hi gh-frequency cont en:
sourceandtyping'S'on thealphanumeric ' V .n ' .where' n' i s the voi cenumber. present in the voice: the cutoff f re-
keyboardtransformsthe Fairlight froman Thereare six real-timefadersand five quency is raised by simply increasing
expensive computer into a powerful switchespatchableto most parameters. the val ue.l t' s al l real l ya case of svr ings
musical instrument.And the keyword Threieof the faders and two of the and roundabouts - a hi gh f ilt er set t ino
hereis 'musical'.Theabilitylo createnew switchesare on the left-handside of the gives a bright realisticsound but often
and interesting sounds(orindeedsample music keyboard,while the other faders w i th di gi tal bi rdi es w ar oling in t he
them) is not in itself enough.What is (orfootpedals)areaccessiblevraCannon- background, while low filter values
requiredis controloverthatsoundand,to type connectorson the rear of the key- suppress any funnies but reduce the
coin a popularphrase,the controlmust board. In addition,the music keyboard soundto a dull noise.
be real-time.Musicians,of course,call When Portamentois on, each Chan-
this control 'expression',and it's a par- nel allocatedto the voice produces a
ticularlydifficultfeatureto build into a continuousglide between each nei.,'
computer-basedmusicalinstrument. pitch it is to play and the last pitch
Consider a typical case in which a played,the rate of note glide being set
Fairlightusermight be playingthe music by the Speed control. 'Glissando'drf-
keyboardwhile listeningto a sequence fers from Portamento in that the glide
pre-recordedon Page 9. Everythingis is not continuousbut chromatic, and
running smoothly: the CMI is reading all the notes on the keyboard between
sequenceinformationfromthedisk,sort- the stad and end notes are played. lf
ing it out, and sendingthe data to the both Portamento and Glissando are
voicechannelsto be played.At thesame selected, Portamento takes preced-
time,the musickeyboardis beingscanned ence. 'Constant Time' is a switch
for pressednotesand moredatasent to which selects between two types of
the channel cards: notes are stolen if glide: when it's turned on, the same
necessary.Next the usermay decideto time is taken to travel any musical
swell a parlicularvoice by movingthe intervgland the rate of change alters
appropriatefootpedal.Here the CMI is ac_cordingto that interval, hence the
forced to deal with an asynchronous name 'ConstantTime'. This results in
event in the normalproceedings, so the polyphonic portamento or glissando.
pedalvaluehas to be updatedconstantly in which the notes arrive at their
passes key velocityinformationto the
and the valuesobtainedusedto scalethe destinationsat the same time prod-
CMl, and this can be patchedto Level
amplitude of the voice throughoutits ucing a coherent chord. With the
and Attack as KEWEL.
duration.And if thiswerenot enough,the switch off, the rate of change remains
Below the voice list in Figure1 is a
Fairlighthas 17 parameterscapableof fixed (determinedby Speed) and the
comoletelistof thecontrolsandswitches
being controlledin real-time. time taken to glide varieswith the size
that are available. Thisis usedin conjun-
It's t he unus ualm u l ti -o ro c e s saorc
r hi- of the interval.
ction with the lightpenand providesa
tecture of the CMI that enablesit to
quick way of patchingthe controlsto
handle so many asynchronoustasks
si mu lt aneous ly , l e t' smo v eo n to l h e
but
vai'iousparamelers:thelightpenis pointed Attacl<and Damping
presentation first at the parameterand then at the The Attack parameterhas a range of
of thecontrolsandtheiruse.
controllist. A patch can also be estab- zerol o 16,384mi l l i seconds, and m ay be
D^ - ^ -l lished by tabbing a cursor around the patchedto 'KEYVEL'for touch-sensitive
rdtc / displayusingtheQWERTYkeyboardand controlof the attackti me.lt 's act iveonly
Al l t he c ont r olsar eh a n d l e db y P a g e 7 , typlngin the appropriate nameor numeric for Mode 4 sounds, and is extremely
and a typicaldisplayis shownin Figure1. value.Figure2 showsone of the 'Help' usefulfor imposinga degreeof artificial
The page featuresall the usualcontrols sheetsfor Page7 whichprovidea quick ' enveloping uponsampledsounds.'Damp-
associatedwith processingsound:each referencefor the range and possible i ng' has a range of ze r o t o 65, 536
of the eight voices loadedcan have its patchesavailable. mi l l i seconds,reduced to 16, 384 m illi-
o wn uniquec ont r ols e tti n ga n d c a n b e So me of the controlparametersare secondsi n Mode4. Thevaluedet er m ines
sa vedt o dis kwit ha c h o s e nfi l e n a me and such as ' Level ' ,' V i b
s e l f-e xpl anatory, the fi naldecayti meof the voice,ie. f r om
the suffixCO.Whena voiceis loadedinto Sp e e d'and ' V i bD epth'A
. gai n,w e come key rel easeto si l ence.l f a loop is act ive
th e CM l, it will pull rn a s p e c i fi e cdo n tro l a c ro s stheeni gmati c' Mode'
sw i tch,w hi ch and one or moresegment sar e r epeat ed
fileif it was previouslyLinkedto thevoice, is best left until Pages 4 and 5 arc conti nuousl y, the voi ce p lays t he loop
u srngt he c om m andL N K. drscussed(fhe suspenseis killing rne- unti lthe dampi ngti meexp ir es( whent he
At t he bot t om of t h e d i s p l a yi s a b o x Ed). 'EXP' is the otherhalf of the com- key i s rel eased),ol herwiset he voice
which indicatescurrenlly-loaded voices. pandingprocessthal was an optionon conti nuesthrough l he rem ainingseg-
Th egont r olfilem ayh a v ea d i fl e re nnl a me Page 8 discussedlast month. As you ments. S houl d the end segm enl be
from its intendedvoice so the two are m a y re member,i t' s a hardw areopti on reachedbeforethe dampinglime expir:s
drfferentiated visually.Theactivecontrol and is very rarelyfitted to the CMI due the voi cestopsabruptl y.
fi l e is t he nam e hi g h ti g h te d w h i l e th e to th e non-l i nearsampl i ngdata that The' S l ur'sw i tchi s usefulf or olissando
 V, /17'//l

MUSICIAN
OMPUTER
Ij'a n d p o r t am ent oef f ec ts ,a s i t c a u s e s
[' i'tt i,i-::: t.ltt I t-'E , I
lChannel cards allocatedto a voice to r r r l r 'l F r i I r X!IJ2
li
sustainindefinitelyin a loop that may be Figure1. r:rl l 'TFti ,L P FP R tl ET E F:3
ive until a new note is played.New I
r:r:rl l TFtr:rL
FILE ' l E Ir? t: tl
notes are startedat the beginningof the
loop withoutplayingany of the preceding l l rrl rE =4 r:i Ll !:j ti r{IrB =rl FF Lrl rl F r:fi TF:L Lrr r
segments.'Sustain'determinesthe be- E 'l F = r:rFF FL:LFrTFrj E ftT0:0FF
=,_l
LI]I]F' :TH FIT I
LE r.rE L : tE \,rrE L l _i F,E E Ir Lr:i l :rPLl i rl TH I
haviour of the voice once the key is FILTE F: = l ! r_:U fl ::i TTII'l E = ttl t 'I I n E I i 'L r r B :T
released.Normally,a voice fades out
I[i H F Il l E = :':1r:i t,lI Ir ITE P TH= B :l Ll J F l = l : l FF
either playingits loop or until it hits the
tl TTB tl l ,, = E t lllB iFFFn = !:! SU ET BIH = D FF
end of the segments,but when Sustainis
on, Damping is ignored and the voice
loops for the durationof key depression: nnnil ETtCt( lH:l!;rlT :lTFtllI IrI
upon key release,the voice continuesto U i II:E T rlRflGsB
r T F : l '1 1 0 e l i l E L L D c | D T 8 l t f l
play its remaining segments with no
decay of amplitude. rl fl TFrL, I : l ti :1l ,l TtH , l. 4 5
frll irFF iFpfl h-E Y U E L
I oonin
r"'b
q
E@m
p:t,:h* :-
@
Choosingthe correctloopingpointof a ::h,:,r ,1!: r 3n.- r ::: ;ni i F t,t!i L'l * i':,r
!th t,3n F 3r 3fi r * tPr
voice wavelorm can make or break a Figure 2.
good soundon the Fairlight.Nastyglitches [l ,rntr,:I P .:r.i nreter R anq*, F:1,:h
can occur if an inappropriatesamplerate l l Ll trE 1, I
L I,F r - l l l . L- IF F !, . : ht r_F l-:
is chosen or if the section to be looped LE (IE L rrl -:'99 | l .: f r-ti TF,Lt-r,
spans a naturalchangeof amplitude. I I L I LFJ I' l E \ r,rE L
lmagine trying to loop a percussive NHNPI TID Lt- tt:,J 5 t:

sound such as a drum. The three loop B TTTi D K 8-16:l :13 LI

controlson Page7 providea quickway of E L I S 9 N IIO ON , OFF

P gR TIi I'l E N TIJ OT.I,OFF :
i
findingthe bestloop,and a typicalsetting ,]TL L ! 8-l !7 c
might be as shown in Figure 3. Here I:IJH S TR N TT I TIE OH , B FF .f

Control 1 is used to define start point of U IB R IITI IIE FTH 8-I?,7 c

the loop while Control 2 sets the length. U IB P IITD 9FE E D A -I?7 I
LI]D P I]ON TR |]L ON , D FF D
Switch 1 freezes the effect of Control 1 LD D P S Tf,R T c
and Control 2 when off, preventingacci- L00P tElrETH S-128 D
dental movement of the looping points S T B R TS E D M E N TI - I 2 8 c
once these have been decided.A useful SLUR ON , OFF i
9 U 5 T FI N
I 0n,0FF
featureis that loop parametersare sved
with the voice informationas well as any
UUILL:
Linked control file, so that the sound is gTE I' IID I
playable even though no performance
controls are required. The actual loop Figure C ON TR OLP R R F}l E TE R S
points are displayedgraphicallyon Page TON TR OL FILE ' LOOP CD
4, as shown in Figure 4, where the
H oI'E = 4 r:Ll ggFti D o 5l ,l TcH 3 Lri 0P C i l rR L = ::H Tr-HI
horizontal axis represents the segment E X F,= I]FF P N R TFIIIE N TO OFF LO|JP S TIi R T = IT{ TF;L1
number and therefore time. The loop is LE U E L = K E Y U E L S P E E IJ 3{ LIJI]P LN .GTH = D N TR LZ
indicatedby the row of highlightedboxes FILIE R =8 rl D ti S TTl H E OTI :JTffR T sE G = I
under the HarmonicProfilesgraph,and D FI'l P Il l G = S E U IB IIE P TH = C N TR L3 S LIIR = S l ,l Tr-H a
since the voice shown was sampled, R TTf,C K = IF U I8 S P E E D = S S U 9TFIN = tl FF
there are none of lhese present. This
Page offers a convenient method of
selectingloopingpoints using the light-
pen.
E IIII? S TIC K B B 9S GT E I:$T|IIT
'Start Seg' is a powerful expression UOIEEs, 9TFTN IN S C B E LLOB D TR H N GR N GS R
control.lt allowsthe startingsegmentof
the voice to be chosen accordingto a INTRL. S I,ITC H .
control value as a new key is played.To
explain: suppose we had sampledthe
classicsynthesiser filtersweepandplay- Iti l E ti
r:rtl l'1F tt Il S TR tl l Irl
ing the keyboard resulted in a 'fruity' q
H FIR tl 0N IC H 0I|E ,
decay. Using a control fader to set the
start segmentwould then enableus to
play the synth sound from ditferentparts l
of the filter sweep. As the controlwas I
\
i
movedfrom segment1 to 128,thesound
would begin with plentyof filtersizzleat J
low control values but become shorter I
and moremellowas we startedthesound
II I
further down the sweep (by increasing I
the Start Segment number).This tech- I
\
nique can also be used to control-the I

amount of 'breath' on sampled wind L00p l [ l |n PL D T I

sounds or the amount of bowing on I
stringedinstruments.
tt
Next month, page 5 and some revel- II
ations concerning the mysterious
'Mo d e' . r \
\
I

.1
 THE T'AIn'LTGIIT
ETKPEATTYED
Samplingmay be the CMI'smost talked-aboutfeature,but asthis article
shows,definingsoundsusingharmonicinformationcanbe just asdramatic.
f ' a\
Izm Grant J

o far we have discussed only one soundsneed lots of numbersto represent antwaveformis visuallyverysimilarto the
method of actuallycreatingsounds them. realthing,and perhapsmore importantly,
with the Fairlight sampling.This So what's the use of Mode 1? Well, soundsi ndi sti ngui shable.
aspectof soundformationis probablythe calculatinga time waVeformfrom har- However, the powerof the CMI liesin its
singlemost importantfeatureof theCMl, monicdala can be quitetime-consuming, abilitynot only to compute a complex
and was cerlainly the focus of public especiallyif the supplieddata is detailed waveformfrom a set of Fourier com-
attention when the machine was an- and enablessubtlenuancesof soundto ponentsso that it can be played on the
nounced. Horvever,the abilityto specify be generated.However,more oftenthat keyboard,but alsoto computea different
sound by means of harmonicinformation not, only a simple waveformis required, wavelormfor each segment. Every seg-
can not only resultin someveryinteresting and to calculatetheRAMwaveformfor all men\ has a unique Page 5 display,so
sounds, it's also rather useful in an 128 segments when a short loop is all while a Mode 1 sound has 32 sets of
educationalenvironment. that's needed is rather wasteful,to say faders,a Mode 4 one will have 1281The
Two display Pages,4 and 5, allow the the least. There's no hard and fast rule currentsegmentnumber is indicatedon
construction of waveformsby harmonic aboutwhich Mode a soundshouldbe in: thedisplay,and thisallowsa synthesised
data. They deal with exactly the same the choice is entirely the musician's. sound to change drasticallythroughout
informationbut presentit to the musician However,usinga voiceas the destinatlon
in different ways. for samplingdalaalwaysresultsin all 128
Firstof all,though,let'sclearup a little segmentsbeing ovenvritten,even if the FBue2
mystery that's been evadingus for some voice selectedis Mode 1.
months - the Mode switch.Actually,this
is very simple and is thereforesomething h?
of an anti-climax.When a voiceoperates rage )
in Mode 1, only the first 32 segmentsof
waveform RAM (4k bytes) are used to Figure1 shows a typical Page5 display.
representthe sound.An unloopedMode This page displaysthe harmonicovertone
1 sound will stop at the 32nd segment, series as a set of 32 'faders' similarto
even though another 96 segments of those on a graphicequaliser.Eachfader
RAM exist. ln order to compensatefor is logarithmicin natureand hasa rangeof
shorter note event time as playedon the zero to 255, allowing a good degreeof
keyboard, each of the 32 segments is control over harmonic amplitudesand
looped severaltimes beforemovingon to thus enablingthe applicationof a Fourier
the next segment:this maintainsa fairly type harmonicseries. itsduration,simplyby the userfillingeach
constant net event lengthfor any pitch. As an example, Figure 2 shows a segmentwith a differentwaveformcalcu-
latedfrom its own Page 5 fader settings.
Mode 4 uses the entirewaveformRAM squarewave generatedby the CMl,com-
(all128segmentsof it)and is alwaysused puted from the valuesof Fouriercom- ln fact,the techniqueof usingdifferent
ponentsshownon the faders.Theresult- waveformsegmentsas the sound pro-
for samplingsince long, high bandwidth
gressesis very much the domainof PPG
Frgue 1. synthesis. Generallyspeaking,thesepro-
gressions are known as wavetables,
N X U OIC T' I and in the PPG,a sound consistsof a set
0 fl rlFH D , S11UBRE
HRRNONI C F fiDERS ODE' 1
11 of 64 waveformsthat reside initiallyin
EPROM(theyare transferredto RAM on
cUK KL N T SEGI'lEN T I power-up)which are read out sequentially
when a key is pressed.The idea behind
this systemwas to circumventthe need
forfiltersby constructingwavetablesthat
held a set of reoresentativewaveforms
of, say,the classicfiltersweep. Unfodu-
nately,this resultsin a very hard,metallic
1 r 1 t l l
sound, as the sound changes abruptly
I I I I 22 ? ? ? ? 2 ? ? 2 3 3 3
I c J 4 56 78 I 0 r 234s e7 890I 23 4 5 6 7 8 9 0 l 2 from one waveformto another, slightly
differentone.lt's stilla good sound,but in
zL t{ 0 c0tlPl J E the interests of flexibility, PPG have
chosen to incorporatethe usual VCFs
andADSRsas wellas exiensivewavetable
modulatlon.
The CMI is also capableof this forrni,l
synthesisto a limited degree,using the
loop controlson Page 7. For example,
supposewe had filled all 128 segments
with waveformsthat changeveryslightly
as we progress through the waveform 0 t' tH8 NIt,
(see Figure 3). Now, if the loop controls
I,IFIU E F|j P IlN I S P LR Y
were set up as shown in Figure4 (thisis a
Page 7 display),moving CNTRL1on the
music keyboardwould resultin a different
timbre when the note was played.Using
this techniqueallowsfor some expressive
playing,sincethe principleis ratherakin
to varyingthe filterfrequencycontrolon a
synthesiser, the only difference being
that the actual timbres can be radically
different from one segment to the next.
To increase the timbral movement
within a sound, CNTRL2can be patched
to 'LOOP LENGTH'on Page 7, resulting
in sections of differgntwaveformsbeing
read out repeatedly.Since the waveform
data is computedby the CMl, it's always
constructed so that the waveform fits
exactly into one segment, thereby over-
coming looping problems.

FourierSeries Ftguo4

Well,with all this talk of 'Fouriercom-

ponents' and the like, some of you may S I ,IEEP
reasonablybe thinking'what's it got to do C O N T R O LP S R R I 'l E T E R S
with music?' The answer,of course,is
not much. Only scientistsand engineers C O N T R O LF I L E . I,IR
delight in quantifyingthe world which our H O D E= IILISSFNI0
4 ut I L|]|]P C N TR L ON
senses seem to handle perfectly ade- EXP = 0FF P0RTFHENT0 |]FF LOOP 9TH R T C}ITR L l
quately. However, in order to exprbss LE U E L = KEYUEL SPEED a LLl OP LN D TH I

ourselves explicitly and unambiguously FI L T E R = 1l tl0NST TIt4E OFF J tl K I CLb CN TR L T

about a wide variety of concepts (someof DFH P I N G= l A g U I B D E P T H =& S L U R = 0 FF
which may be abstract) we need to use f iTT f f C K = 0 U I B S P E E I ] =g S U g T F I N = 0 FF
the language of mathematics. Fourier
analysisand synthesisare mathematical
statementsabout somethingwhich is not
intuitivelyobvious: the fact that any truly
periodic waveform can be decomposed UO I I E S , FilNTiiI
into an infinitesum of sinewaves,usually
called harmonics.Similarly,any periodic
waveform can be constructed from the 1:-t45
sum of an infinite number of sinewaves. FI
The sinewaves have frequencies that are
relatedto the fundamentalof thewaveform
in such a way that the second harmonic
Flgnr,o5. Fgtre8.
lies at twice the fundamentalfrequency,
the third harmonic lies at three times the
fundamental,and so on. The fundamental
itself is often referred to as the first
i'i"i"i
harmonic.
Of course,obtaininga reasonablerep-
resentationof a desired waveform does
not require an infinite number of sine-
ilItl
IT T I
,l
waves: more than 16 is enough to give a
good approximation.The Fairlightusesa
maximum of 32 harmonics,which enables
most waveformsto be synthesisedwith a
fair degree of accuracy. Figures5 to 8
show the developmentof a squarewave
by successivelyadding furtherharmonics
and using Page5 to compute the resultant
waveform.The squarewave doesn'tcon-
tain any even harmonics (2, 4, 6 and so
on) and we can see that the lower
harmonic numbers set the basic square
shape while the higher ones fill in the
bumps and sharpenthe edges.Evenwith
all the odd harmonics the Fairlightcan
compute, the square wave is still not
visuallyperfect,but it soundsOK. I

1984
E&MM DEGEMBER
 MUSICIAN
O MPUTER

TIIE T'ATRLTGIIT
E)KP&ATNED
Waveforms,lightpensand interpolationall come under examinationin
this instalment
of our FairlightCMI GrandTour.Ji* Grant
ust when you thought it was safe to
open up a copy of E&MM without Along the bottom of the displayare the
reading anything about the world's harmonic numbers 1 to 32. A small
most influential comoutermusicalinstru- triangleunderthe numberindicatesthat
ment, your intrepidreporterreturnsfrom the time profileof that harmonicis being
a New Yearhangoverwith anotheraction- displayedon the graph, while a cross
packedepisode.This month we look at shows that the proflle has a non-zero
the informationpresentedby Page5 in a vatue.
slightlydifferentlight. You'll remember So what does all this mean? Have a
that Page 5 held the values lor 32 look at Figure2. Two profilesare shown,
harmonicfadersand computedthe result- one of which is the Firstharmonic(leftto
ant waveform for the current segment. rightdownwards)and the otherthe Third
You shouldrecallalsothatthe onlywayto (left to right upwards). On receipt of a
createa completesoundof 32 segments Computecommand,the CMI will fill the
was to define the fader levels for each waveformsegmentswith sound which
segment and compute over the whole initiallyat least,hasa strongfundamental
waveform; or define a few segments NIIEi;
and Fill the harmonicdata to the rest of rl tl |1ri ll tl H FR 145
the segmentsbeforecomputing.lt's not H fi Fi I'ION
I ROFILES l ' l rl l rE , I
hard to see that this method of creating
soundsmay be very precisebut can also
be extremelytedious.In a lot of cases,all
we needis a wayof tailoringthe harmonics
as the sound progresses: harmonic
envelopes,in other words. EnterPage4.
Figure1 is a typicalPage4 display,and
shows that it's one of the two Fairlight
display Pagesto be almost exclusively
lightpen-driven. The largedark areais in
fact a reversevideoimage,and pointing
the lightpenin this regionresultsin an mrlfl FLr:rT
arrowcursorappearingon the screenat
lllttl r lllr
the currentlightpenposition.For those l!i45r I [!:i
! 2:rrJ :j l
t1 ! :tE t1c "
unfamiliarwith the term, the lightpenis
now a fairlycommoncomputeradd-on, tr n tr
mainlybecauseof its simplicityof opera- I.L EHF, [IILE TE i tF t- t FF' I-I11Eff
tion.Containedwithineverylightpenis a
fast photoelectricdiode or transistor
whichproducesa voltagepulseas theW I llIr Ei: | r :r l r :E, I
linepassesbeneathit. Usually,the pulse r: r:r11f1H fl Il HBFII,1 ::;
is squaredup and passedto the video H Fi F]11[iI HI: FF]i IF I LE i f l n l E' I
controllerchip, which storesthe W line
numberand positionalong the line in a
coupleof registers. This informationcan
then be used by the programmer to
initiatepredefinedevenlssuch as plotting
a point or executinga command.
The Fairlight system is no different,
except that the video controlleris con-
structed from discretelogic chips and
resides on a single eight-inch board
withinthe CPU.In additionto latchingthe
TV co-ordinateswhen the lightpen is L r:rr:rF Irirfll FLrrT
used, it generatesan interruptto the
processorsto executethe selectedtask. .!i:i:ji:i
i tsj rl it!
At firstglance,the graphareain Figure
1 l o o ksa bit c onf us ing
b u t i t' sre a l l yq u i te tr
straightforward: the verticalaxis repre-
l: f1 li'fti
sentsamplitude whilethehorizontal shows
time and hence the seqment number.
M A R C H1 9 8 5 E&M M
't
I

l.

:)ut degenerates into dominant third

rarmonlc. l[r f1l' 1Hf1Il triF :
r l : It
l l i rrE FI:'F l 1 IrI :FL H ,
t 1 1! _ . _ r {r , _ l : j _ : , t _ ! E B _ . _ _ : T E F [| . c :

,.lankprofilegraph.

i"lore About Mode

l l l _ r I l-:
One of the previously mentioned
l l F Frl .' :
featuresof a Mode 1 voice is the way in H i rF;l l i rl lI r-: P F:rtFI LE g l'1[r Ir I
which the first 32 segmentsare looped
several times to maintain the net event
timeof the soundacrossthe keyboard.In
fact, we havesome controlover how long
a segmentlasts beforeeverythingmoves
on to the next one, and this is accom-
plished vrathe profile. Figure4 shows a
harmonicprofilegraph with the duration
profile indicated by a double line: the
defaultvalueis approximately50mSper
segment and increasesas the profile is
drawn higherup the graph.This is parti-
EE LU I:IP mfnl rL,: j
cularly useful for creating sounds with a
short click at the beginningof each note, ' --t " ;
suchas that of a Hammondorgan.Avery
shortdurationvaluecan be drawnfor the
el45 * , . l *u I tl t
:'a :'
r li
;l
F i Ii
lrii
rr:l E r I
.-,-.t.-,.-.
i ili F
C I o
first one or two segments, and then a
longer profile for the remainderof the I::L EF F , IIE LE TE F.E '1E T l 'IFr:i r:i JtlFU TE ':r-tl LE It1 TE F:F r-' . E f,ti
sound. lf the duration profile is made
zero,the sound degeneratesto a Mode 4 Resetcausesa confirmationmessageto contain rapid changes throughout the
condition,except that it only lastsfor 32 be printed and also results in Page 4 durationof the sound.
segments. being restored to a complete default Incidentally, becoming proficient at
Another interestingaspect of Mode 1 condition.Zeroisn'tquiteso drastic,and usingthe lightpenfor drawingcan take a
sounds is their ENG profile.This is an resultsonlyin the currentprofilebeingset lot of practice,so the CMI helps out by
artificialenvelopethat's superimposed to zero. Everytime you give a Compute providinga Join Plotselector:when Join
on the waveform in much the same waY command, a new energy profile is is active, any two points struck on the
as the more usual ADSR principle.But generated.Scale is the opposite: it re- graph are immediately connected by
this one's a lot more flexible.When the drawstheharmonicprofilesfrom a modi- a straight line, while fine detail can be
Compute command is given, the CMI fied energyprofile.This is not without its drawn by selecting Plot. I imagine that
calculates the waVeform segment by dangers, however, as it can result in most of you will be familiar with the
segment and scales the amplitudeso some harmonic profiles being scaled Fairlight'sLoop function by now, so I
that it fits exactlyinto the dynamicrange beyondtheirmaximumamplitude,which won't go throughit all again.Sufficeto
of eight bits. The ENG profile is also leadsto clipping.The Fairlightw7linform saythenthat Page4 offersa quick way ol
generated(and its shape implied)by the you of the situation when it occurs (by drawingthe loop startand length.Mode 1
harmonicdata, but can be alteredby the displaying'Overflow')but is powerlessto soundsare alwayscalculatedso that the
lightpen to control the amplitude of the preventit happeningif theScalecommand waveform fits perfectly into a segment:
sound on playbacltr is issued. The only way to recover the the first harmonic does one cycle, the
soundthen is to reloadthe voice. second harmonictwo cycles, and so on-
Timenow to introduceanotherconceot Gone are the Bad Old Days of trying to
Auxrlary Functrons
A''I'F
with which some of you may not be overly sample a sound to make it fit segments
familiar.lnterpolationis the skillof guess- evenly.All you haveto do now is use any
Along the very bottom of the display ing an unknownvaluethat lies between old loop to span the sectionsof a wave'
Page are a number of usefulcommands. two knownones,and is commonlyused formthat are of interest,and Bob Moog's
Clear deletes all the displayed profiles to predictvaluesof pointson graphsthat your uncle.
from the graph, but they remain active aren't the actual ones originallyplotted. Next month (yes, there's still more to
and can be broughtbackto lifesimplyby When the Interp switch is On, each come),we'll take a look at Page6 which,
the programmerselectingthe harmonic waveformsegment is computed from a arnongmany other weird and wonderful
numberswith the lightpen.Delete,on the mix betweenthe harmonicnrofilesof that things,allowsyou to splicea sounddown
other hand,merelyremovesthe currently- segmentand those of the next one. The to no more than 16,384thof its length.
selectedprofile.Thesamesoft of structure differencebetweenthe two is subtle,and Ahd you thought a razor blade was
appfies to Reset and Zero. Invoking is only reallynoticeablewhen the profiles powerful. .. r
E&MM MARCH1985
 TIIET'AIRLTGIIT
E)KPEATTUED
When a CMI pageisashelpfula sourceof musicgraphics asPage5, you can
bet your lifeit needsa lot of explaining-sothismonth'sepisodeis
dedicated entirelyto it.J i,nt()ru,rtt
I f you're tryingto createa new sound I il[ r t t , , : lrir T r- C

I on the Fairlight,whether it's from tl

r:rl'1ll H ll Il
l t H | | E F r:rF11 IrF:B !i I ttrj
I existingsampleddatausingPage8 or
harmonicprofilesgeneratedon Pages4
and 5, you can be surethat Page6 willbe
referredto over and over again.
At its simplestlevel,Page6 is a siatic
'oscilloscope'containingone segmentof
the waveform RAM. Like Page 4, the
largeblack areais a reversevideoimage
and representsthe region that can be hit
by the now-infamouslightpen.But unlike
an oscilloscope,which acts only like a
window on the information,Page 6, in
conjunctionwith its commandsandlight- : : ' TEF F:B T E ,
pen, behaves more like a door through
iTBF i T
which we can directlyaccessthe wave-
form data. r jlr _r F

The ManualApproach Figure1,

Regularreadersof this serieswill pro-
bably recallthat each segmentconsists
of 128 bytes,and that eachbytecanhold 5lr E * f llntllEfliJilIl I::FiBFH
tillf,G[m r::E[i t{ [ f{ T !:::: i .-rr, r, Er, t
a rangeof valuesfrom -128to 127,giving IiIEf,I[E';,'11i1n -3;1';{
a grand total of 256 possiblevalues.Now, FIEf,ffiEIID :-t:F'p i r' -r il i !F 1 :?
we can highlightany particularbyte by [IftMEm r n j r ] F f '- 1 r n i 1, :i,l i t i ': : t i ': r r
positioninga narrowverticalwindowover I?fjltrro{E[ r.I.3L.'e f 'rr rr nri r:ii I i i 3t i,:'r-l
fiirilUfiflfifi :-rri tf h
the display area. The byte in question, Effil] ,: ,:,n,n,3r;il
shown in Figure1, has its positionindi- m[film L-:i ft,Iir.-rr'il
ca te d b y P 0I NT an d i ts v a l u e b y nt[ilImfiffil'., ilrt t,:,r Ll .J Pl 'r * r 3tl ,:,r l
L EV E L: thesecan be changedby typing lililI] ,: ,:,ri rlr3i'ril
in differentvaluesfrom the alphanumeric tifffn ': ,:,ri I' 3n il
tlf{iliti ,: i, r,rL,3r, '|:l
keyboard.So, theoreticallyat least,you IIIIH i ii tritr 3r-ril
could type in 128 levels for each byte mlffi[i ':,:,rirti':ti,:l
shown in Page6, and repeatthe process nEfitrFEm ,: i,L,r,i3tril
for all 128 Page 6 displaysto coverthe R!roI{#Tl i,-l r,ri, i r-,il
Iff,illEnnB ,:i, r,,n-Jritl
entiresound,giving16,384levelsin all.In lifiE : ':,r,ir,, i,r',:]
practicethough,such a feat would take t[illfl rilitti,-1 r,
an unbelievable amountof manuallabour l![AlffilfffE :-tl'r'rin+r'i : rr,:l
I r''|
(exactly what much of the Fairlight's I r f : : : : , , r, rt l ', L I r . H T i I i i , r r - T 'i F E
r ,r h':t'E
software was designedto eliminate)to [[IG 'rr T i F E
achieve,and this has led me to wonder
whether this feature will be dropped
when the Series lll Fairlight,complete ul' lNr if lIl l 'l ri R Itl tr l
with waveform RAM of megabytedimen- r:ii r U fl Ir i ti l 'IFLIN C
sions, makes its appearancelater this
year. : : ; H I 'l P L ER I i T E I 4E j i j B 5 F I 'I F . L EL E U E L t5E
A more practical approach is, of F I LTEFI LUI,I I T F I I T : G E EL E I , I E L t5
FI L T E E H I L i H .1 T RI I I ] E R I I E L H T u 1,1!
course,to make use of the lightpen.This t r - r H F 'F r E g S u R r:rFF
is as simpleas drawingon the back of a
bus ticket, but just in case even that is
beyondthe user'sartisticcapabilities, the iHt'1 F L t
J 0 I N and P L0T f unc t i o n sa l s op re s e n t
on Page4 are availablehere,too,to help TI I: l F L H \ .
out with the drawing of geometric
snapes.
In fact, trying out a few samplesket-
ches soon revealsthat waveformswhich
look drasticallydifferentmay sound sur-
prisinglysimilar.Scientificallyspeaking,
this is probablydue to the fact that as a
A P R I L1 9 8 5 E&M M
 OMP U T EM
RU S IC IA N
species,we have little prior experience Ifiii i rt : I t : f I
gainedfrom our sensesthat relatesvisual - ' ' r lllr ilr i i ' 1 8 F, I 1 1 I B
ti me dom ain wav ef o rm sto w h a t w e
perceivein terms of harmonicsor par-
ti a l s. F]
: i i I'FL I FH IL 1.1i :
'J
: . H I ] F 'L ELE I,IE L :59
But Page 6 waveform-drawingis of FIt.IIi Lr:rt] T F : l r . r : I F , LE IIE L r5
enormoushelp,firstbecauseit givesthe I-ILItF'H II.H
' r-ri JFF[:':11F
T F 'I I : i L , E FII' E LB I I l'1':,
user extensive and direct control ol r_rF i-
waveform RAM with one sweep of the
hand, and second becausethe abilityto _ fl tr I t
bringabout such drasticsonrcchangeso
simply and visually(almost artistically, It]:FIti ,'
some might say) is a practicalitylight
years away from the sophistication and
complexity of the underlyingmachine
and its software.Or to put it anotherway,
it makesthe CMI seemlessof a beastand
more of a pet.
F ig u r e4 .

The Practical
Approach I lJ I r t f l
Alterationof the display imageis not u lr l' lF fltl
reflectedin the waveformRAM untilthe r:ir:rl_llj Ir ::iHl\lF,L l il [
F I L L commandis used.Thisallowsthe
reverse video field to be used as a :i H IIP LE E FITE I qi l :l U H i i B I {P L E L E I . I E L i 55
scratchpad area in which waveshapes F I LTE F Lr_ri l I T F t l r : r : ; E FL E r . r E L t5
FILTE R H IIJH iJ T EI i r : ; E F : t r E L h i ' E I' l l
can be developed before they are finally r : Lrl 'l FFi E i trJR r-l FF
committedto RAM.
Lookingagainat Figure1, noticethere
are two 'slider'controlssimilarto those i lil'lP L E
on Page 5. These control the segment
numberto be displayedand the stepping II IS FLItY
ratethroughthe entirememoryinstigated
when the command Step is issued.Also
shown are three classic waveforms-
triangle,sawtooth,and square- which
deposit a perfectly fitting single cycle
wave into the segment.The pulsewidth
Figure 5.
of the square can be varied from 17oto
99% by changingthe valueheld next to tu rn e de nd on end by usi ng RE V ER SE , l i magi ne N 0IS E w i l l b e f air ly self -
the squarewave symbol. and this resultsin the backwardssounds explanatory to eventhe leastclued-upof
that have become familiarto pop music this column'sreaders:it fills sectionsof
followersthe world over. the RAM with the output of a random
AuxiliaryCommands R E F L E C T i s a l ess commonl y-used numbergeneratorhiddeninsidethe sott-
In additionto the graphiccapabilities of but if anythingmore interestingvariation w are. Funni l y enough, t he GA I N
Page6, thereare a numberof commands on this theme. lt allows us to olace an command is also used to tailor the
that can only be entered from the Fair- imaginarymirror in the sound and reflect amplitudeof the noiseto that of the rest
light's alphanumerickeyboard.Figure2 everypart of the waveformin front of this of the sound...
shows the range of functionsavailable, mirrorto waveformRAMbehindit. Takea N ow , B LE N D i s a str angelyout of
as presentedby the H E LP sheetmenu. look at Figures3 and 4 for before and place command. Personally,I think it
First off are a f6w utility-typecom- afterviews:the 'mirror'is at segment64. shouldbe on Page4, since its role is to
mands.GA I N scalesthe displayeddata Now on to anotherof the seemingly help smooth out glitches caused by
by a specifiedpercentage:if the rescaled insurmountable problemsbroughton by imperfectloopingpoints.lt may be thatto
waveformis about to exceedthe ampli- the onset of new musical technology. finda good loopit makesextensiveuseof
tude rangeof the system,the CMI willask Very often, when a sound has been extrapolationtechniques,and that see-
you whetheror not you want to proceed sampled,the beginningof the captured ing as this is a centralfeatureof the MI X
and therebyinduceclipping.lf you reply data does not occur at exactlythe startof and ME R GEcommands,Fair lightdeci-
in the affirmative,the command will be the RAM, perhapsdue to some extra- ded to bungit on Page6 alongsidethem.
d u l yex ec ut ed.M ean w h i l eth
, e I N VE R T neousnoisepretriggering theADC.How- However,ratherthan discuss the re-
commandinvertsthe phaseof the wave- ever, if the effect isn't too severe, a mainingPage6 commandsin any detail
form:this is usefulas a preludeto someof convenientmethod of correctionis to rightnow, I thinkit's probablybest for all
th e o t her f unc t ion s s u c h a s M IX, R0 T A T E the soundwithinthe waveform concernedif I demonstratetheirpowerin
MER G Eand A DD. RAMto bringthe startin linewith byte 1. the context of a typical edit session in
A parlicularlyneat little corirmandis This has the (oftendesirable)side-effect whicha soundis createdfromscratch,ie.
ZERO ,whic h allowsu s to c re a tea n u l l of shiftingthe first part of RAM to the end. withoutany sampleddata. So, now you
voice in preparationfor the A DD com- SeeFigures5 and 3 for anotherrevealinq know what l'll be talkinq about next
mand. The entire waveform RAM can be beforeand after month.
' AH IGE T T ING A B IT OVERDRAW N'
H^{ M/I/1I!A,
..JHINK I'LL ST AR T 5Ai IN U T ES LAT ER
A T T H E B A N K , T IME To Co/YlPoSE WITH A + ,' + RHY T H M ,..
AN oTTHHEERR ME CA
A{o GA -S M ASH Ht7 ...AN OfH ER A AST ER P I E CE
SIN6 L E . I T H INK .' F IN ISH ED J T H IN K I'LL J U S T D O
.F ouR R EA^IXESO F T n t S O N E J , 1
S /r{EA HEAD-
: JUST TR/ING

TERL''TE CHI

E & MM APRIL19 85
 THEE'ATRLIGIIT
E)KPEATTUED
In which we takea lookat the differencebetweenlinearandlogarithmic
andincidentally
conversion, endupcreating asoundusingthe Fairlight's
synthesis
faciIiti es.J im Grant
ostpeopleinvolved inthemodern
R D D C O Il
musicindustryarealready aware IIRUTFORN D I SPLSY
noD E, 4
poten-
incredible
of theFairlight's
tial as a music prodrction tool. These
days, you switch on the television,radio
or record player in the almost certain
knowledge that a CMI will make its
presencefelt somewherealongthe line,
and when you considerjust how usefulits
specificationis to studio Ongineersand
producers,that's hardlysurprising. Even
in 1985, there aren't many machines
capable of spreading six octaves of
sampledsound acyossthe keyboard,and
manipulating that sound within user
sequencesto the nth degreeof preciSion.
But if you're fortunateenoughto sit in
front of a CMI for any length of time
without any production deadlines to l l oD E' 4
meet, you'll soon discoverthat its creat- nautFoR nD IspLny
ive power lies as much with sound
synthesisas it does with musicproduc-
tion per se. Pages 4 and 5 are good
examples of this in that they offer the
fairly standard synthesis tools of har-
monic sliders and profiles,but Page 6,
which we introduced last month, is
something of a softwareoddity,since it
allowscontrol overthe wholewaveform-
from a single byte to macro type com-
ma n d s s uc h as G A IN , Ii 1IX a n d
IIER G E .
Now, for any command or processto
be really useful in the field of sound
synthesis, it must be responsiblefor
some radicalchangein the soundstruc- 0i t18tl D , BD D C 0n
l l oD E' 4
ture that's both intuitive and easily f.BU tF oR n D Is ptBy

understandable. Forexample,theVCFof
an analoguesynth changesthe sounda
great deal, and can be simplyexplained
and understood in terms of the attenu-
ation of harmonics.FM synthesis,on the
other hand, also resultsin vast timbral
differences, but comprehensionof the
processes involved (and their possible
results)is a lot more difficult.That'swhy
actuallyarrivingat a pre-specifiedsound
on sonrethinglikea YamahaDX7requires
so much in the way of practiceand
patience - and why so many musicians
prefer programming analoguesynths,
evenif the ultimatesonicpotentialisn't as them directlyinto the samesegmentsof sinewaveboth residentin differentchan-
great. the currentlyselectedvoice, scalingthe nel s of the C MI: the resul t of ADDing
Fortunately, the Fairlight'sinternal amplitudeto avoid clippingif necessary. them togetheris shown in Figure3. The
configurationside-stepsmost of these lf you were to A DD all the segmentsinto addition's proportion can be varied by
operational problems, and a good ex- another, playing the keyboard would using the GA I N command prior to the
ample of how this is done is the A DD result in both soundd being heard acti on,or by repeatedl yADDing one
command. This takes a choiceof seg- togetherbut only usingone voice.Figure voiceto anotherto increaseits amplitude
ments from one loadedvoiceand adds 1 shows a square wave and Figure2 a relativeto the comDositesound.
 OMPUTER
MUSICIAN
onr Ai l D , fl I xcon

mtn tmum

MSB=1

1111 VoiceWhveform
(Triongle)

FilD, SYNT0f4

Sl ]U N D SR N PLIN G

S8I1PIE R R T E I 4O8O H Z SAi l PLE LEU EL 255

F ILT ER to'l I T R IG C ER LEU EL 15
F ILT ER H IG H 8 T R IC G ER IEL8Y O NS
C O I1PR ESSO R OF F
Figure 4. Linear conversion.
SSIl PLE

D ISPLR Y

0111

Figure 8.

0t1l 1R N D , S Y i l To r' l

SO U N DSBIl PLI N G

SR IIPLE R ST E I4O 8A H Z SBI1PLELEU EL 255

F I LT ER LO II 1 T RI G G ER LEU EL IS
F ILT ER H IG H 8 T R IG G ER D ELR Y 9 NS
C OI1PR ESSO R OF F

SR Il PLE

D I SPLfi Y

Figure 5. Non-linear (logarithmic) convercion. Figure 9.

VUIUL:
11txc0t! FE R G C O I ' 1
noDE, 4

l"

waveform is stored in 16K of RAM, in The term 'linear'refersto the relationship

Linearity which each byte has a binary value that between the actual amplitude and the
Now seems as good a time as any to c\oresponds directly to the amplitude of value of the binary number used to
explain sornething we've mentioned the waveform at that point. A byte representit. lf all this sounds a bit on the
many times in the past but haven'treally consistsof eight bits, and consideringall technicalside (and it ought to), have a
discussed in any detail, namely the the possible combinations of these res- quick glanceat Figure4. This shows that
differencebetween linearand non-linear ults in the amplitudeof the sound at any zero amplitudecorrespondsto binary 0,
voice data. As you'may remember, a point being limitedto one of 256 levels. while maximum negatiVe excursion' is

E&MM MAY 1985 85

 IIRN! PFtl l l i fl N D , TE S T
il O D E, 4 noDE' 4
I.IBUEFORI,I
DISPLf,Y l r f,uEF oR r l D Is PLBY

X IIII;T UUILL. I
l 1 l 4 8 N D, T EST tl AN D . TE S TI
I'l ol E ' 4 l ' l o D E' 4
r f,uEFoRi DISPLBY r ,l nuEF oR r ,iD Is PLBY

|-\ re p re s ent edby 1111 1 1 1 1 ,o r 2 5 5 ,a n d sion process under its commonly used sampled using Page 8 and shown in
the maximum positive value is held as name of companding,and it's a system Figure8. lt's pretty clear that the sample
Well,
0 1 1 1 1111,or 127.S t i l lc o n fu s e d ? used by many hardwaremanufacturers ends in a dither of noise. Now, suppose
the valueof the Most SignificantBit holds includingLinn and E-mu. we needed nothing more than a short
the key. lf that valueis 0, the waveformis So if it's so good, why doesn'tthe CMI percussive strike, and that only the
positive,while a 1 gives negativeexcur- use it? Simple. Rememberyour school beginning of the sound was of any
sions. Anything else in between is in days when you added log numbers to interestto us. A quick solution would be
simpleproportion.Thisform of represen- multiply?Well,thisis whatwouldhappen to turn to P age 6 and ZERO , say,
tation results in a ratio between the i f th e e AD D command w as used w i th Segments64 to 128, halve the sound,
smallest and largest signal that can be sounds held in the form generated by and then MI X from Segment 45 to 64.
handled (or in other words, dynamic Figure 5: instead of adding the sounds Lookingat Figure9 shows the result- a
range) of about 48d8. You might con- togetherto producea mix, we'd get the sound that dies away evenly to a noise-
sider that to be not a particularlyimpres- product, and end up with VCA-type free end, much to the relief of all
sive figure, since it means that at low effects at low frequenciesand strange concerned.
signalamplitudes,the sound is moreor sidebandsat higherones.Whichisn't,all l .l E R GEi s fai rl ysi mi l arto M I X - wit h
less surroundedby hiss. things considered,a particularlydesir- one fundamentaldifference.Again, a
The fact is, storage (in one form or able state of affairs. form of crossfadeis generated between
another) of low-amplitude sounds is a startand end segments,but this time, the
perennialengineeringproblem,to which l.a'
previous contents of intermediateseg-
the most common solution is some sort l -l l x l n g ments don't figure in the result. Quite
of noisereductionsystemsuch as Dolby. Anyway, enough of the lecture and simply,the segmentsin betweencontain
In the digitalworld,and as a directresult backto the Fairlight.The MI X command a decreasing proportion of the start
of research into digital telephony,a can also drasticallyalter the waveform segmentand an increasingproporlionof
ditferentsolutionis to use more binary RAM. Essentially, it generatesa cross- the end segment.Figures6 and 10 (oh
bits of the byte to representlow-signal fade between two specified segments yes, very logical- Edl revealall. The MI X
levelsthan you use for the highones.This which must not be adjacent, ie. there and ME R GE commands ar e t r em en-
is shown in Figure5, in which the lower must be at leastone segmentin between. dously powerful for splicing together
valuesof the trianglewaveuseup moreof The waveformmemoryof each segment soundsof differingoriginsand producing
the binary bits than a corresponding between the stad and end points con- an even fade from, say, a violin bow
increaseat largetriangleamplitudes. The tains a proportionof the existingwave- attack to a sung 'ahh'.
binarydata is now no longerlinear- it's form in that segmentand that of the end
logarithmic.For the waveform to be
recovered, the data must be passed
segment: this is best illustrated by
examiningFigure 6 and Figure 7 for
Creating
a Sound
through a DAC that has a curve bent the before and after views. Rememberthat So, now that we've discussedmost of
opposite way to 'straightenout' the the new contents of each segment is a the commands available, let's try to
sound.I know all thissoundsmorethana mix between where you are in the create a sound using everythingexcept
littleinvolved,but it doesbringthe magic waveformand the destinationsegment. the Fairlight'sPage 8 samplingfacility.
dynamic range ratio up to about 72d8, Thus from Segment 2 onwards, the Thequestionis: am I allowedto use Page
which is at leastrespectable. Onlycatch waveform simply fades up to a square 2 and pulla soundoff-diskto work with?
is, the processonly achievesthis with a wave. MI X is most commonlyused to Well.I've decidedl'll have to cheat a bit
correspondingincreaseof quantisation add a clean fadeout to a sound that because I already have a thoroughly
noiseat largersignallevels,thoughthisis decays to noise or doesn't decay pro- marvelloussoundcalled PAN.VC.which
maskedby the volumeof the signalitself. perly in 128 segments. attacks with the characteristic breath
You might be familiarwith thisconver- Have a look at the percussionsound chiff of pan-pipes.
B6 MAY 198s I E&MM
 O M PU T ER
MU SIC IAN
First off,
First off , we configure Page 3 to work on the body of the sound itself. look closelyat the differencesbetween
generate two voices, one with an Beforeany soundcan reallymakethe Fi gures13 and 14, you shou ldn'thave
NP H0NY of 7 in RegisterA to be played gradeas far as aesthetics areconcerned, muchdifficultyidentifying the variationin
on the keyboard,the other monophonic it must have plenty of timbral and ampl i tude,especi al liyn the so und'sf ir st
in Register B as a scratchpadvoice. amplitudemovementwithin it. A good quaner.
Using Page 2, we load RegisterB with way of producing harmonically-rich All that remainsis to inserlloop points
PAN.VC,which can be sbenin Figure11. waveformsis to use Page5 and createa on Page 7 or Page 4, and adjust the
The breath chiff is clearlyvisibleat the few segmentsspreadacrossthose un- attackanddampingon Page7. I
beginning of the sound, but unfortu- used by the chiff. Figure 12 gives the
nately,the samplingstadeda fractiontoo general idea - note that the created
soon, and the waveform has a few initial waveforms are all ditferent. So what
segmentsof low-levelrubbish- nothing about the segmentsin between?Well,
to do with Electronic Soundmaker, you th i s i s w h ere ME R GEcomes i n handy,
understand. The cure is to rotate the fi l l i n g i n the ZE R Oed segments,and
voice left to bring the start of the sound usingthe segmentscreatedon Page5 as
proper coincident with the start of the the start and end points.
RAM. OK, so far it sounds quite interesting
The next step is to flick to Voice 1 in timbrally(and looks it too, as Figure13
R e g i sterA and Z E RO it. U s i n g a n e w shows) but it's still in need of some
c o mma n d, T RA NS F E R , th e fi rs t fe w amplitude variation. An easy way to
segmentsfrom PAN.VCcan be copiedto achieve this is to invert a couple of
the blank voice currentlyselected.Stab- segments(numbers32 and 96, say)and
bing the keyboardat this luncturereveals l.lI X from segments1 to 32, 64 to 32, 64
that all is well,so the nextthingto do is to to 96 and 128 to 96, using Page6. lf you

t've tusr aou6l{T THts NErrrr \ IT CONTAINS ALLOF THE TECHNIQUES ilow I CANCOI PoSE LIK! A{oZABT
'meon-Ruro-col PoSER'-SoFTWARE ANO STYLES OF IHEWONLDs GREAT. AltD BEEtHoylr frL Dl lll6^-f^/{out
ONLY COST T5OOI EST COMPOSERS!.?ANTUi}| THE IN NO TI'',IE! lETs SEE WKAT THE
'I\E /l osr BoRDtGAxD,r{tDtocRr solt6. PRO6RAM DOESWITH TITATPIECE
,flo.l$""r!L.os1 t cnstp
Er r ADL oF nAtBlENT rraUSlCt DtD IAST rltoHT.t

i+/.€oD