GB2287352A - Signal distribution representation and control - Google Patents

Abstract

A system for representing and manipulating signal distribution for example in a music synthesiser, comprises a two dimensional array of visual indicators representing signal sources by rows and signal destinations by columns. Signal sources are connectableto signal destinations; and a connection is indicated by modification of the visual indicator at the intersection of the respective source row and destination column. The level of the source signal supplied to each connected destination is independently variable. An intersection may be highlighted in order to adjust the level of a source signal passing to a destination. Sets of levels defining signal distribution may be stored in memory. A control processor may periodically interpolate between stored sets of levels, with the interpolation controlled a rotary encoder. Thus complex real time changes in patterns of signal distribution may be achieved. <IMAGE>

Description

REPRESENTING SIGNAL DISTRIBUTION FIELD OF THE INVENTION The present invention relates to an apparatus for representing distribution and a method of representing signal distribution.

BACKGROUND TO THE INVENTION Considerable advances have been made in the development of digital signal processing, which is being increasingly used to replace analog electronic circuits in a wide variety of applications. Digital signal processing has several advantages over analog methods, including reduced dependence on critical components, simplification of manufacture and reductions in cost.

These advantages have led to developments which are only possible with the advent of this new technology, such as the real-time manipulation of digitized video images.

While the complexity of devices incorporating digital signal processing has advanced considerably, representation of the available functions has lagged to the extent that many users view such equipment as a "black box", using only those functions which are familiar to them. Thus in certain respects, developments in products which use digital signal processing are restricted by the design of the interface through which complex functions are controlled.

Musical synthesis has been revolutionised by digital signal processing. However, modern digital synthesizers, while capable of producing a wide variety of useful sounds, lack the controllability of some analog synthesizers due to the lack of a front-panel interface which clearly represents the elements and interconnections of the sound generation process. An improvement in the representation and ability to manipulate the elements and interconnections of digital synthesizers is sought after by many musicians. A solution to this problem may also be useful in other digital signal processing equipment.

It is the aim of the present invention to provide an improved means of representing and manipulating signal distribution in a digital signal processing system.

SUMMARY OF THE INVENTION According to a first aspect of the present invention, there is provided apparatus for representing signal distribution, comprising a two dimensional array of visual indicating means having a plurality of signal sources connectable to a plurality of signal destinations; wherein a connection is indicated by modifying the visual indicating means at an intersection of a respective source and destination; and the level of the source signal supplied to each connected destination is independently adjustable.

Modification of the visual indicator may be such that connection or non-connection at each intersection is represented by activation or nonactivation of the respective visual indicator. In preferred embodiment , the intensity of activation of a visual indicator may give an indication of the level of source signal transferred to a destination at each intersection.

In order to facilitate modification of connections, there may be provided a set of process representations on a control panel. Each process may have one or several inputs and outputs. These may be represented by buttons, for example round buttons representing process inputs and square buttons representing process outputs. Each output button may represent a signal source, represented by a row on the array of visual indicators, and each input button may represent a signal destination, represented by a column on the array of visual indicators. Simultaneous activation of an input button and an output button may result in modification of visual indicators such that the respective intersection is highlighted.

Once highlighted, a connection may be modified using a rotary encoder to adjust the level of source signal passing to the destination at the intersection.

On or off buttons may be used to make or break the connection respectively.

The set of levels defining signal distribution may be stored in memory, such that a distribution pattern may be recalled instantly at any time.

A control processor may periodically interpolate between stored sets of levels, with the interpolation controlled by manual operation of a rotary encoder. Thus complex real time changes in patterns of signal distribution may be achieved. Useful results so produced may be stored in memory.

According to a second aspect of the present invention, there is provided a method of representing signal distribution, wherein a two dimensional array of visual indicating means represents a plurality of signal sources connectable to a plurality of signal destinations; wherein a connection is indicated by modifying the visual indicating means at an intersection of a respective source and destination; and the level of the source signal supplied to each connected destination is independently adjustable.

BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 shows an electronic musical instrument, arranged to generate sounds and including a sound synthesizer; Figure 2 details the sound synthesizer shown in Figure 1, including a control panel and a digital signal processor; Figure 3 details the control panel shown in Figure 2, including oscillator modules, a ring modulator module, a noise generator module, a filter module, an envelope module, an output module and a two-dimensional array of visual indicators; Figure 4 details an oscillator module of the type shown in Figure 3; Figure 5 details the ring modulator module shown in Figure 3; Figure 6 details the noise generator module shown in Figure 3; Figure 7 details the filter module shown in Figure 3; Figure 8 details the envelope module shown in Figure 3; Figure 9 details the output module shown in Figure 3;; Figure 10 details the two-dimensional array of visual indicators shown in Figure 3; Figure 11 shows a flow chart of the algorithm used by the digital signal processor shown in Figure 2 to calculate signal distribution, including a multiplication and accumulation process; and Figure 12 details the multiplication and accumulation process shown in Figure 11.

DETAILED DESCRIPTION OF A PREFERRED EMBODIMENT An electronic musical instrument is shown in Figure 1, consisting of a note signal generating device, such as a keyboard 15, a midi sequencer or a midi controller etc. Note information generated by the note signal generator 15 is supplied, over a midi interface or other similar interface, to a sound signal generator or synthesizer 16, arranged to generate left and right analog audio signals in response to the note information generated by the note generating device 15.

The analog audio signals generated by the synthesizer 16 are supplied to a stereo amplifier 17, or similar device such as an audio recorder. The outputs from the stereo amplifier 17 are then supplied to audio signal transducers, such as loudspeakers 18. Thus manual or automated operation of the note signal generating device 15 results in musical sounds being produced by the loudspeakers 18.

The synthesizer 16 is detailed in Figure 2 and consists of a midi interface 21, control panel 22, control processor 23, digital signal processor 24 and digital to analogue converter 25.

The midi interface 21 converts the midi signal generated by the keyboard (or other device) 15 into a form more suitable for the control processor 23. In particular, the midi interface 21 includes an optoisolator, thereby electrically isolating the synthesizer 16 from the midi signal generator 15. In this way, the midi standard is maintained and the midi interface may be connected to any midi signal generating device.

The midi interface 21 converts the serial midi data stream into an eight bit parallel stream, which is in turn supplied to the control processor 23. The control processor 23 also receives a "byte ready" signal from the midi interface 21, enabling it to latch an eight bit byte from the midi interface 21, when such a byte becomes available.

The midi information defines particular notes by a unique midi number and specifies when such a note is to be switched on and when it is to be switched off.

The control processor 23 is arranged to generate an indication of actual tone frequencies, such that tones may be produced by the digital signal processor 24. In addition the note-on and note-off information supplied by the midi interface 21 may be translated by the control processor 23 into amplitude information which, when supplied to the digital signal processor 24, may cause a tone to start and finish in accordance with the pressing and releasing of a note played on the keyboard 15 (or other midi signal generating device).

The control panel 22 has controls 312 which send information to the control processor 23 about sound attributes such as the frequency, volume and spectral quality of sounds and the way that these may change throughout the duration of a tone.

The control panel 22 has visual displays 331 and 332 which may receive data from the control processor 23 in order to provide a visual indication of sound attributes such as those affecting frequency, volume and spectral quality of a tone, and how these may change throughout the duration of a tone.

In response to control signals generated by the control processor 23, the digital signal processor 24 generates digital representations of tones, which are supplied to the digital to analogue converter 25. The digital signal supplied to the digital to analog converter 25 consists of two eighteen bit values at forty-four thousand one hundred samples per second, one for each of the left and right channels. The digital to analog converter converts these signals into respective left and right analogue signals, using established high quality digital to analog conversion techniques, such as those commonly adopted in compact disc players and digital audio tape players.

The control panel 22 shown in Figure 2 is detailed in Figure 3. The control panel provides representations of several processes 301 to 306 or modules 301 to 306 used by the digital signal processor 24 to generate sound. The modules include three oscillators 301, a ring modulator 302, a noise generator 303, a filter 304, an envelope 305, and an output 306.

Each module 301 to 306 may have input buttons 310 and/or output buttons 311. Input buttons are round and output buttons are square in order to distinguish their function. A module 301 to 306 may also have one or several rotary encoders 312 for adjusting various module characteristics.

Each rotary encoder 312 on the control panel produces fifty digital pulses per revolution, and an indication of the direction of rotation. This information is used by the cdntrol processor 23 to modify settings for module characteristics such as waveform and frequency. Because there is no limit to the number of revolutions, a greater range and accuracy may be obtained from rotary encoders than from rotary or slide potentiometers.

The numeric value produced by a rotary encoder is shown on an alphanumeric display 332. Each rotary encoder has a touch sensitive knob, such that the values produced by said rotary encoders will be shown on the display 332 as soon as they are touched, together with a brief description of the encoder's function, such as "Filter Frequency".

The control panel 22 has a signal distribution controller 320, consisting of an alphanumeric display 332, a rotary encoder 333, buttons labelled "on" 334 and "off" 335, and a two dimensional eleven by ten array of light emitting diodes 331.

The oscillator module 301 shown in figure 3 is detailed in Figure 4. The oscillator module has a phase modulation input button 41, a signal output button 44, a rotary encoder for frequency 42 and a rotary encoder for selecting different waveforms 43. This module may be used to generate waveforms of any pitch in the audio range. Oscillations are produced at a pitch determined by midi note data provided by the keyboard 15 and modified by the frequency control 42. Signals supplied to the phase modulation input 41 may be used to produce dynamic changes in timbre or vibrato-like effects. The oscillator module algorithm produces oscillations using established high quality waveform synthesis methods, such as those commonly adopted in digital synthesizers.

The ring modulator module 302 shown in figure 3 is detailed in Figure 5. It consists of two input buttons 51 and 52 and an output button 53. Ring modulation is achieved by multiplying two numbers together. In this case the two numbers are fractional, in the range -1 to +0.9999..., and having twenty four bit accuracy. The forty eight bit signed product is truncated to its most significant twenty four bits using established high quality digital audio truncation methods. Thus, the two input buttons 51 and 52 represent the multiplier and multiplicand and the output button 53 represents the truncated product.

The noise generator module 303 shown in Figure 3 is detailed in Figure 6. This has an output button 62 and a rotary encoder which affects the colour or spectral quality of noise, often referred to as its "color". Noise is generated using established pseudorandom number generation techniques, and the color variation is achieved by varying the coefficient of a digital filter in order to vary the distribution of noise density across the audio spectrum: for example in order to vary the noise from being a low rumble to a hissing sound.

The filter module 304 shown in Figure 3 is detailed in Figure 7. The filter module 304 consists of two input buttons, one for the signal input 71 and one which provides modulation of the frequency of the filter 72. There are rotary encoders for frequency 73 and resonance 74, and signal outputs which have low-pass 75, high-pass 76, band-pass 77 and notch 78 filter characteristics. The filter is implemented using established high quality digital filtering techniques.

The filter may be used to affect the harmonic content of the input signal in a number of ways.

The frequency rotary encoder 73 affects the resonant frequency of the filter. This may be swept to different parts of the audio spectrum using the rotary encoder 73 and/or by a signal applied to the frequency modulation input 72. The low pass output 75 produces a signal containing harmonics which are mainly below the resonant frequency of the filter. The high pass output 76 produces a signal containing harmonics which are mainly above the resonant frequency of the filter. The band pass output 77 produces a signal containing harmonics in a band of frequencies centred around the resonant frequency. The notch output 78 produces a signal omitting harmonics in a band of frequencies centred around the resonant frequency. The filter resonance, set by the other rotary encoder 74, is used to accentuate low pass, high pass, band pass and notch filter characteristics, and at high values will cause the filter to self-oscillate.

The envelope module 305 shown in Figure 3 is detailed in Figure 8. The envelope provides controls which affect the way the amplitude of the signal passing through it changes with time. There is a signal input button 81 and a signal output button 86. There are four rotary encoders 82 to 85. The first rotary encoder 82 is labelled "attack", and sets the time taken for the amplitude to reach maximum after a new note has been played on the keyboard 15. The second rotary encoder 83 is labelled "decay" and sets the time taken for the amplitude to decay from its maximum value (reached at the end of the attack phase) to the sustain level which is set by the third rotary encoder 84.The amplitude may remain at the sustain level until the note on the keyboard 15 is released, at which point the amplitude will decay to zero at a rate set by the last rotary encoder 85 which is labelled "release".

Thus the amplitude of the signal passing through the envelope from input 81 to output 86 has its amplitude modified according to the settings of the attack 82, decay 83, sustain 84 and release 85 controls and the playing of notes on the keyboard 15, or some other midi source.

The output module 306 shown in Figure 3 is detailed in Figure 9. The output module includes inputs for the signal 91, amplitude modulation of the signal 92 and pan (stereo position) modulation. There are rotary encoders 94 and 95 for volume and pan respectively.

There is no output on this module, as the two outputs generated, for left and right, form the digital audio output stream from the digital signal processor 24 to the digital to analogue converter 25.

The array of light emitting diodes 331 shown in Figure 3 is detailed in Figure 10. The array 331 consists of horizontal rows of light emitting diodes which represent the outputs of the modules already described. These are: First Oscillator 101, Second Oscillator 102, Third Oscillator 103, Ring Modulator 104, Noise Generator 105, Low Pass Filter 106, High Pass Filter 107, Band Pass Filter 108, Notch Filter 109 and Envelope 110.

The vertical columns of the array of light emitting diodes represent inputs to the modules already described. These are: First Oscillator Phase 121, Second Oscillator Phase 122, Third Oscillator Phase 123, Ring Modulator 124 and 125, Filter Signal 126, Filter Frequency 127, Envelope Signal 128, Output Signal 129, Output Amplitude 130 and Output Pan 131.

An example will now be given of the way in which different modules may be connected.

An initial condition is assumed where none of the array's light emitting diodes is lit. This indicates that none of the modules is connected, and no tones will be produced by the digital to analogue converter 25.

Connection of any module's output to an input is done by first pressing an output button, for example the output button 44 of first oscillator module 301. Doing this will cause all the light emitting diodes in the corresponding row 101 on the array 331 to light up.

Next, an input is selected, for example by pressing the input button 81 of the envelope module 305. This will cause all the light emitting diodes in the corresponding column 128 to light up. The light emitting diode where the row and column intersect will return to its initial off condition - this is the connection being highlighted and it is currently switched off. Any input column or output row may be selected in this way, so that any row column intersection on the array 331 may be highlighted.

To the right of the array of light emitting diodes 331 are two buttons labelled on 334 and off 335.

These may be used to make or break the connection highlighted on the array, switching the light emitting diode representing the connection on or off respectively.

The level of signal passing through a connection may be adjusted, while the connection is highlighted, using the rotary encoder 333. The alphanumeric display 332 indicates either "OFF" or a numeric value indicating the level of the output signal that may pass through the connection to the input column. The value representing the level of signal transferred is referred to as the "connection coefficient". Thus, since the array of light emitting diodes has ten rows 101 to 110 and eleven columns 121 to 131 there will be a total of (ten times eleven) one hundred and ten different connection coefficients which, together, define signal distribution.

Complex signal distribution may be represented by the array of light emitting diodes, with each horizontal output contributing a different level of its signal to each of the vertical inputs to which it is connectable.

For example it may be desirable to connect varying amounts of the first 101, second 102 and third 103 oscillator outputs to the input 129 of the output module 306. In order to do this, the input button 91 on the output module 306 is pressed, causing the corresponding column of light emitting diodes 129 to light up. Next the output button 44 on the first oscillator is pressed, causing the corresponding row of light emitting diodes 101 to light up. The highlighted connection may be switched on (if it is not already on) by pressing the on button 334 to the right of the array of light emitting nodes 331. The level of the signal passing to the destination may now be adjusted using the rotary encoder 333 until a satisfactory audible result is achieved.

Next, press the output button 44 on the second oscillator, which will cause the corresponding row of light emitting diodes 102 to light up. The connection now highlighted is the one just below the one that was previously highlighted; the selected column 129 has not been changed. The connection may now be switched on and adjusted in the manner already described. This procedure is duplicated for the third oscillator's output, row 103. The result of these operations will be that all three oscillator outputs will be contributing a proportion of signal to the input represented by column 129, and this is represented graphically by three light emitting diodes lit in a vertical formation in column 129, at the top of the array of light emitting diodes 331.

A plurality of light emitting diodes may be lit at the same time, representing an equivalent number of connections between modules. There need be no limitation on the type of connections which may be made, and it may be legitimate for outputs of modules to be connected to their own inputs. This makes it possible to set up highly complex signal routes, which may feed back on each other and interact in complex, possibly chaotic ways. This has the potential to produce highly original effects.

In this system, a -constant sampling rate is used and all signal connections and module algorithms will be calculated forty-four thousand one hundred times per second, which is the audio sampling rate used by the digital signal processor.

The algorithm required to calculate signal connections is shown in flow chart form in Figure 11.

This algorithm is executed forty-four thousand one hundred times a second.

Process 211 resets the column counter to point to the first column. The column counter indicates which of the columns or module inputs 121 to 131 is being calculated. Process 212 resets the column total to zero.

Process 213 resets the row counter. The row counter indicates which of the rows or module outputs 101 to 110 is currently selected.

In process 214 the row output is multiplied by the connection coefficient for the intersection of the current row and column. The connection coefficient will be zero if no connection is made (the corresponding light emitting diode is off). The product of this multiplication is added to the column total.

Process 215 increments the row counter, and decision 216 loops the process so that the current row versus connection coefficient products for all the rows 101 to 110 are summed to form the column total.

Once all rows have been calculated, decision 216 diverts program flow to process 217, which stores the column total for later use as a module input.

Process 218 increments the column counter, and decision 219 loops the program to process 212 to reset and generate a column total for the next column and so on until decision 219 recognises that all the columns 121 to 131 have been calculated and execution of the connection algorithm for one audio sample cycle has been completed.

Figure 12 details the flow of data that occurs in process 214. Both the row's value and the connection coefficient are twenty-four bit signed fractional binary numbers. These are multiplied in multiplier 221 to form a forty-eight bit signed binary product which is added to a fifty-six bit column total by accumulator 222. The register 223 passes its input to its output on the rising edge of the processor clock. This controls the feedback to the input of the accumulator 222 such that accumulation only occurs once for each new product produced by the multiplier 221.

The final column total, the value stored in process 217, is created by truncator/rounder 224 which creates a 24 bit signed binary fraction. This is a suitable data format for a module input.

The multiplier 221 and accumulator 222 perform a procedure known as multiply-accumulation, which is a highly efficient process on many digital signal processor chips.

The connection algorithm could be implemented exactly as described above, with a zero connection coefficient where no connection is required (the corresponding light emitting diode is off). However this is not an efficient use of the digital signal processor 24. In practice it is very unlikely that more than half the possible connections will be used for any given sound. The efficient solution is for the control processor 23 to compile a short program for the digital signal processor 24 which omits multiply-accumulations where the connection coefficient would be zero. This compilation is done each time a connection is switched on or off.

The control processor 23, in addition to interfacing between the midi interface 21, the control panel 22 and the digital signal processor 24, also provides non-volatile memory storage for connection coefficients and front panel settings of the modules' rotary encoders, such that complex settings of the front panel which produce useful results may be stored and recalled instantly at any time in the future. Each set of stored information is called a "sound". The front panel controls required to store and recall sounds have been omitted for the sake of clarity.

In an alternative embodiment the control processor 23 is used to facilitate sophisticated changes to connections in real time in a manner that will now be described in detail.

For any given sound, signal distribution is defined by the one hundred and ten connection coefficients which individually define the amount of signal that may pass from a source to a destination represented by a row and column on the array of light emitting diodes 331. This set of connection coefficients will be henceforward referred to as a "frame".

The control processor 23 may generate connection coefficient values by periodically interpolating between two different frames, thus facilitating the gradual evolution from one frame to another in real time. The two frames from which interpolated values are calculated may be chosen from sounds which have been previously stored in non-volatile memory. The movement from one frame to another may be controlled by the rotary encoder 333.

The interpolation process referred to here is where a value, for example n, is.derived from two other values x and y: x is subtracted from y and the result multiplied by a control value k. The resulting product is then added to x. In summary: n = k(y - x) + x Thus if k has a value of 0.5, n will be the average of x and y, or, to put it another way, half way between the values of x and y. In interpolating between frames, each connection coefficient is derived by interpolating between two other values which were stored previously in memory. By controlling the value of k using a rotary encoder, the active frame can be made to be more like one or other of the stored frames. Thus it is possible to gradually evolve from one frame to another over time by slowly moving k through a range of fractional values starting with zero and ending up at one.

The evolution from one frame to another may be highly complex, as can be seen from consideration of the evolution of a single connection coefficient. This may start with a value of -1 and go through a range of values until reaching +0.9999... or alternatively start at +0.9134 and go through a range of values until reaching +0.105. Alternatively, if the connection coefficient in both frames is the same, no evolution will occur. Thus, when all one hundred and ten connection coefficients from a first frame, each possibly having a different value, are caused to evolve gradually into a different frame of one hundred and ten connection coefficients, highly complex changes in signal distribution may result.

By controlling evolution using the rotary encoder 333, the evolution may be frozen at a point where a useful sound is produced, and this may be stored as a new sound in memory. The advantage of movement between frames in this manner is that extraordinary sonic effects may be produced easily which would be impossible to achieve by manual adjustment of connections on a one-at-a-time basis.

The control processor 23 may also include the settings of the various rotary encoders on each of the modules to be modified in real time during evolution between frames, thus providing complete evolution between sounds.

In a further alternative embodiment, movement from one frame to another may be linked by the control processor 23 to real time performance data such as that produced by the keyboard or midi source 15 in such a way that midi information representing, for example, the pressure of a finger on a note, is linked to movement between frames.

In a third alternative embodiment, the digital signal processor 24 is improved in order to perform all of the algorithms previously described several times within each sample period, such that a plurality of tone generators, each one represented individually by the totality of the modules and the array of light emitting diodes on the control panel, are available, enabling more than one note to be produced at any time. In this alternative embodiment, in addition to the operations already described, the control processor 23 assigns different notes to different tone generators according to a system of priorities, such that a musician may play several notes at once on a keyboard 15, while remaining largely unaware of the fact that only a limited number of tone generators actually exist.

In such a system, the maximum number of notes that may be heard simultaneously is defined by the number of tone generators available. By providing a large number of tone generators, such as twenty, more tone generators may be made available than can actually be played by one pair of hands on a keyboard. Thus, a keyboard may be perceived as having a single tone generator assigned to each of its notes when, in reality, a much smaller number of tone generators is actually available.

In a preferred implementation of the embodiments previously described, substantially all or a major proportion of the circuitry required for the digital signal processor 24 is contained within a single integrated circuit, chip or chip-set.

The system is not necessarily limited to the midi standard and alternative protocols may be implemented which may, in particular, provide faster and more comprehensive data. Such an arrangement provides a greater amount of information transfer from the note source 15, which may contain information about the velocity with which notes are struck to a degree of accuracy greater than that provided by midi. In addition, information may be transferred representing key movements throughout the duration of a note. This information may be used by the control processor 23 to affect the sounds produced by the digital signal processor 24 throughout the duration of a tone, for example to effect movement between two selected frames.

Claims (20)

1. Apparatus for representing signal distribution, comprising a two dimensional array of visual indicating means having a plurality of signal sources connectable to a plurality of signal destinations; wherein a connection is indicated by modifying the visual indicating means at an intersection of a respective source and destination; and the level of the source signal supplied to each connected destination is independently adjustable.

2. Apparatus according to claim 1, including memory means arranged to store sets of levels defining signal distribution.

3. Apparatus according to claim 2, including means for calculating levels defining signal distribution by interpolating between stored sets of levels.

4. Apparatus according to claim 3, including data entry means controlling interpolation, such that levels defining signal distribution are modified over time.

5. Apparatus according to any of claims 1 to 4, including entry means for adjusting the level of signal passing from a source to a destination.

6. Apparatus according to claim 5, including switching means representing sources and switching means representing destinations.

7. Apparatus according to any of claims 1 to 6 including visual indicating means for highlighting an intersection of a source and destination.

8. Apparatus according to claim 7, including means arranged such that activation of switching means representing a source simultaneously with the activation of switching means representing a destination, causes the respective intersection to be highlighted.

9. Apparatus according to any of claims 1 to 8, including calculating means, such that the signal supplied to a destination is calculated by summing the products of the source signals and the level coefficients associated with the intersection of each source and said destination.

10. Apparatus according to claim 9, including clocked multiplying and accumulating means, such that successive products from said multiplying means are accumulated.

11. A method of representing signal distribution, wherein a two dimensional array of visual indicating means represents a plurality of signal sources connectable to a plurality of signal destinations; comprising: indicating a connection by modifying the visual indicating means at an intersection of a respective source and destination; and independently adjusting the level of the source signal supplied to each connected destination.

12. A method according to claim 11, including storing sets of levels defining signal distribution.

13. A method according to claim 12, including calculating levels defining signal distribution by interpolating between stored sets of levels.

14. A method according to claim 13, including controlling interpolation by data entry means, such that levels defining signal distribution are modified over time.

15. A method according to any of claims 11 to 14, including adjusting data entry means so as to adjust the level of signal passing from a source to a destination.

16. A method according to any of claims 11 to 14, including highlighting the intersection of a source and destination by visual indicating means.

17. A method according to claim 16, including highlighting an intersection by switching means representing the respective source simultaneously with switching means representing the destination.

18. A method according to claim 16 or claim 17, including adjusting the level of a source signal supplied to a destination while the respective intersection is highlighted.

19. A method according to any of claims 11 to 18, including calculating the signal supplied to a destination is calculated by summing the products of the source signals and the level coefficients associated with the intersection of each source and said destination.

20. A method according to claim 19, including accumulating successive products from said multiplying step.