The Fairlight Explained (Part 4)
Jim Grant gives an outline of how the Fairlight CMI copes with performance effects.
How the CMI provides for special effects and looping — in a language just about everyone can understand.
Last month we dealt with one of the Fairlight's main sound creation methods: sampling. The simple act of pointing a microphone at a sound source and typing 'S' on the alphanumeric keyboard transforms the Fairlight from an expensive computer into a powerful musical instrument. And the keyword here is 'musical'. The ability to create new and interesting sounds (or indeed sample them) is not in itself enough. What is required is control over that sound and, to coin a popular phrase, the control must be real-time. Musicians, of course, call this control 'expression', and it's a particularly difficult feature to build into a computer-based musical instrument.
Consider a typical case in which a Fairlight user might be playing the music keyboard while listening to a sequence pre-recorded on Page 9. Everything is running smoothly: the CMI is reading sequence information from the disk, sorting it out, and sending the data to the voice channels to be played. At the same time, the music keyboard is being scanned for pressed notes and more data sent to the channel cards: notes are stolen if necessary. Next the user may decide to swell a particular voice by moving the appropriate footpedal. Here the CMI is forced to deal with an asynchronous event in the normal proceedings, so the pedal value has to be updated constantly and the values obtained used to scale the amplitude of the voice throughout its duration. And if this were not enough, the Fairlight has 17 parameters capable of being controlled in real-time.
It's the unusual multi-processor architecture of the CMI that enables it to handle so many asynchronous tasks simultaneously, but let's move on to the presentation of the controls and their use.
All the controls are handled by Page 7, and a typical display is shown in Figure 1. The page features all the usual controls associated with processing sound: each of the eight voices loaded can have its own unique control setting and can be saved to disk with a chosen filename and the suffix CO. When a voice is loaded into the CMI, it will pull in a specified control file if it was previously Linked to the voice, using the command LNK.
At the bottom of the display is a box which indicates currently-loaded voices. The control file may have a different name from its intended voice so the two are differentiated visually. The active control file is the name highlighted while the active voice is shown in the top right-hand corner. Other control files can be inspected by pointing the lightpen at the names in the display box or by typing 'V,n', where 'n' is the voice number.
There are six real-time faders and five switches patchable to most parameters. Three of the faders and two of the switches are on the left-hand side of the music keyboard, while the other faders (or footpedals) are accessible via Cannon-type connectors on the rear of the keyboard. In addition, the music keyboard passes key velocity information to the CMI, and this can be patched to Level and Attack as KEYVEL.
Below the voice list in Figure 1 is a complete list of the controls and switches that are available. This is used in conjunction with the lightpen and provides a quick way of patching the controls to various parameters: the lightpen is pointed first at the parameter and then at the control list. A patch can also be established by tabbing a cursor around the display using the QWERTY keyboard and typing in the appropriate name or numeric value. Figure 2 shows one of the 'Help' sheets for Page 7 which provide a quick reference for the range and possible patches available.
Some of the control parameters are self-explanatory, such as 'Level', 'Vib Speed' and 'Vib Depth'. Again, we come across the enigmatic 'Mode' switch, which is best left until Pages 4 and 5 are discussed (the suspense is killing me - Ed). 'EXP' is the other half of the companding process that was an option on Page 8 discussed last month. As you may remember, it's a hardware option and is very rarely fitted to the CMI due to the non-linear sampling data that results from its use. The Filter is a low-pass tracking filter resident on each Channel card, used to attenuate any unwanted high-frequency content present in the voice: the cutoff frequency is raised by simply increasing the value. It's all really a case of swings and roundabouts - a high filter setting gives a bright realistic sound but often with digital birdies warbling in the background, while low filter values suppress any funnies but reduce the sound to a dull noise.
When Portamento is on, each Channel allocated to the voice produces a continuous glide between each new pitch it is to play and the last pitch played, the rate of note glide being set by the Speed control. 'Glissando' differs from Portamento in that the glide is not continuous but chromatic, and all the notes on the keyboard between the start and end notes are played. If both Portamento and Glissando are selected, Portamento takes precedence. 'Constant Time' is a switch which selects between two types of glide: when it's turned on, the same time is taken to travel any musical interval and the rate of change alters according to that interval, hence the name 'Constant Time'. This results in polyphonic portamento or glissando, in which the notes arrive at their destinations at the same time producing a coherent chord. With the switch off, the rate of change remains fixed (determined by Speed) and the time taken to glide varies with the size of the interval.
The Attack parameter has a range of zero to 16,384 milliseconds, and may be patched to 'KEYVEL' for touch-sensitive control of the attack time. It's active only for Mode 4 sounds, and is extremely useful for imposing a degree of artificial enveloping upon sampled sounds. 'Damping' has a range of zero to 65,536 milliseconds, reduced to 16,384 milliseconds in Mode 4. The value determines the final decay time of the voice, ie. from key release to silence. If a loop is active and one or more segments are repeated continuously, the voice plays the loop until the damping time expires (when the key is released), otherwise the voice continues through the remaining segments. Should the end segment be reached before the damping time expires, the voice stops abruptly.
The 'Slur' switch is useful for glissando and portamento effects, as it causes Channel cards allocated to a voice to sustain indefinitely in a loop that may be active until a new note is played. New notes are started at the beginning of the loop without playing any of the preceding segments. 'Sustain' determines the behaviour of the voice once the key is released. Normally, a voice fades out either playing its loop or until it hits the end of the segments, but when Sustain is on, Damping is ignored and the voice loops for the duration of key depression: upon key release, the voice continues to play its remaining segments with no decay of amplitude.
Choosing the correct looping point of a voice waveform can make or break a good sound on the Fairlight. Nasty glitches can occur if an inappropriate sample rate is chosen or if the section to be looped spans a natural change of amplitude.
Imagine trying to loop a percussive sound such as a drum. The three loop controls on Page 7 provide a quick way of finding the best loop, and a typical setting might be as shown in Figure 3. Here Control 1 is used to define start point of the loop while Control 2 sets the length. Switch 1 freezes the effect of Control 1 and Control 2 when off, preventing accidental movement of the looping points once these have been decided. A useful feature is that loop parameters are saved with the voice information as well as any Linked control file, so that the sound is playable even though no performance controls are required. The actual loop points are displayed graphically on Page 4, as shown in Figure 4, where the horizontal axis represents the segment number and therefore time. The loop is indicated by the row of highlighted boxes under the Harmonic Profiles graph, and since the voice shown was sampled, there are none of these present. This Page offers a convenient method of selecting looping points using the light-pen.
'Start Seg' is a powerful expression control. It allows the starting segment of the voice to be chosen according to a control value as a new key is played. To explain: suppose we had sampled the classic synthesiser filter sweep and playing the keyboard resulted in a 'fruity' decay. Using a control fader to set the start segment would then enable us to play the synth sound from different parts of the filter sweep. As the control was moved from segment 1 to 128, the sound would begin with plenty of filter sizzle at low control values but become shorter and more mellow as we started the sound further down the sweep (by increasing the Start Segment number). This technique can also be used to control the amount of 'breath' on sampled wind sounds or the amount of bowing on stringed instruments.
Next month, page 5 and some revelations concerning the mysterious 'Mode'.
Feature by Jim Grant
mu:zines is the result of thousands of hours of effort, and will require many thousands more going forward to reach our goals of getting all this content online.
If you value this resource, you can support this project - it really helps!