WO1981001068A1

WO1981001068A1 - Method and apparatus for addressing an opto-electric matrix display

Info

Publication number: WO1981001068A1
Application number: PCT/GB1980/000160
Authority: WO
Inventors: R Benjamin
Original assignee: United Kingdom; R Benjamin
Priority date: 1979-10-09
Filing date: 1980-10-08
Publication date: 1981-04-16
Also published as: GB2062929B; GB2062929A; JPS56501264A

Abstract

Matrix addressing of opto-electric displays such as liquid crystal displays and electroluminescent displays of the kind comprising a first set of N conductors (Y1 to YN) and a second set of M conductors (X1 to XM) forming NxM intersections or elements whereby a particular information display is obtained by altering the optical property of the display at selected intersections by application of a predetermined cyclically repetitive set of N drive voltage waveforms, independent of the information to be displayed, to the first set of conductors, and a set of M drive voltage waveforms to the second set of conductors selected in dependence upon the information to be displayed, such that their interaction with the predetermined set of waveforms applied to the first set of conductors, integrated over the repetition period of the predetermined set of waveforms applied to the first set of conductors, integrated over the repetition period of the predetermined set, produces a visual display of the information. In accordance with the invention, the predetermined set of N drive-voltage waveforms, which is advantageously applied to the smaller conductor set having an odd number of conductors, comprises, at any instant, a substantially equal split between two voltage levels, which may conveniently be of equal magnitude but opposite polarity. The waveforms applied to the second set of conductors may simply comprise two-level waveforms, or, where selective weighting of the display brightness of display elements associated with respective conductors of the second set is required, e.g. to obtain uniformity of contrast over the whole display, three or more level waveforms may alternatively be used. The specification describes in detail the application of the invention to small alphanumeric displays as well as waveform or oscilloscope displays operating in either RMS or direct response modes.

Description

METHOD AND APPARATUS FORADRESSING ANOPTO-ELECTRICMATRIXDISPLAY -------------------------------------------------------------------------------------------------------------

This invention relates to matrix-addressing of opto electric displays such as liquid crystal displays and electroluminescent displays, of the kind comprising a first set of N conductors, and a second set of M conductors forming Nx M intersections or elements whereby a particular information display is obtained by altering the optical, property of the display at selected intersections (hereinafter simply called display elements), the change in optical property being achieved by application of appropriate voltage waveforms to the two sets of conductors. Displays of this kind will hereinafter be referred to simply as matrix displays of the kind specified.

Liquid crystal displays are commonly formed by enclosing a thin layer of liquid crystal material between glass plates provided on their inner faces with transparent electrodes. Application of a potential difference between the electrodes can be used to produce a change in the optical properties of the material, removal of the potential difference allows the liquid crystal material to relax back into its original state. The rate at which this switching of the display can be achieved depends on a number of factors including the thickness of the liquid crystal layer, the particular material used and the mechanism by which the change in the optical properties occurs. Typical values for the switching times for different types of liquid crystal display vary from less than one millisecond up to several hundred milliseconds. Various different display configurations are known which selectively switch different parts of the display to produce the desired variation in reflectivity or transmissivity across the display area. A common method of addressing complex liquid crystal displays is to provide crossed sets of spaced parallel electrodes on opposite sides of the layer of liquid crystal material to form a matrix of display elements, each one uniquely defined by the crossover between one of electrodes of one set (row electrodes) and one of the electrodes of the other set (column electrodes). Any element of the display may thus be selectively addressed by the application of suitable drive voltages to a particular column electrode and a particular row electrode. A similar technique may also be used in the addressing of so-called seven segment numeric displays or other pre-determined patterns or matrix displays in which the electrodes are arranged in radial and curved form to provide polar coordinate displays.

Other forms of opto-electric display in which matrix addressing is commonly employed are the electroluminescent and gas-discharge (planar) displays in which the application of voltage difference across a phosphor layer sandwiched between two electrodes causes light emission.

The present invention is concerned with the matrix addressing of such opto-electronic displays in which one set of conductors is driven by a corresponding number of separate, and in general, mutually orthogonal, waveforms which produce a succession of distinct drive; patterns across that set of conductors. For each instantaneous drive pattern appearing on the first set of conductors, each conductor of the other set must be driven at an appropriate voltage determined jointly by this instantaneous drive pattern and by the desired ON/OFF pattern along that one of the second set of conductors.

Consider a matrix display comprising N row conductors and M column conductors, in which each row conductor is driven by a respective one of a set of N distinct, and in general mutually orthogonal,ie uncorrelated, periodic waveforms. Typically these are distinct quasi-random sequences within a common repetition time(period). These waveforms will then produce a succession of distinct drive patterns across the set of N row-drive conductors at the rate of one per autocorrelation interval of the drive waveforms. For each instantaneous row-drive pattern, each column conductor will have to be driven at an appropriate voltage determined jointly by this row-drive pattern and by the desired ON/OFF pattern along the given column conductor. In order to avoid an irritating flicker, due to the fluctuation of the control voltages across the various intersections, the luminence associated with each intersection must be smoothed by some sort of integration process. There are two modes for achieving this. In the first instantaneous, or "direct-response" mode, the response of the liquid crystal (or other medium) is fast enough to follow the voltage fluctuations. Hence smoothing is left to the persistence of vision of the human eye, and the complete cycle (or equivalent sequence) of control voltages must be kept rather shorter than this persistence of vision.

In the second or "root mean square" (RMS) mode, the response time of the medium is rather longer than one complete cycle (or equivalent sequence) of the control- voltage fluctuations, so that the smoothing is performed in the medium, and is independent of the characteristics of the observer. In principle, to obtain any given display "picture" of ON and OFF elements, it would be usual to expect the optimum sequence of N row-drive patterns required to produce that "picture", to be a function of the N desired ON and OFF patterns of elements along the row-drive conductors (and of the M instantaneous ON and OFF patterns of elements along the column drive conductors), concurrently with the converse relationship in the other coordinate. However, such a scheme would be immensely complex when dealing with arbitrary and frequently changing display "pictures", and to reduce the complexity of such a scheme, it has been proposed to apply a predetermined sequence of drive-voltage patterns to one set of conductors, say the row- drive conductors . The required pattern of ON and OFF elements along each column-drive conductor may then be regarded as a vertical feature, and the sequence of instantaneous column-drive voltages produces, or at least optimises, the display of τhat vertical feature.

One method of implementing this form of matrix addressing for liquid crystal displays is described in IEEE

Transactions on Electron Devices, Vol ED-21 , No 2, February 1974, "Scanning Limitations in Liquid Crystal Displays" , P M Alt and P Pleshko, in which a strobe pulse is applied to each of the row-drive conductors in turn, while an appropriate sequence of column drive voltages is applied simultaneously to each of the column drive conductors . Thus the voltages applied to the column-drive conductors during each strobe pulse are determined in accordance with the desired pattern of ON and OFF elements along the particular row-drive conductor to which the strobe pulse is being applied. While in such an arrangement only those elements are switched on which are actually required to be switched. on to generate the desired display feature only limited contrast ratios are achievable, particularly in matrix displays having a large number of rows, because only those elements along one row can be switched on at any given time. The predetermined sequence of drive voltage patterns applied to the N row-drive conductors may alternatively comprise the entire range of possible voltage pattern combinations(2^N combinations for a two-level or binary drive system). However, unless N is very small, this sequence would last longer than the persistence of vision, and possibly also longer than the time for which an unchanged display feature is required to be generated.

According to the present invention, a method of addressing an opto-electric matrix display of the kind specified, includes the steps of applying a predetermined cyclically repetition set of N drive-voltage waveforms, one to each of the first set of conductors, the values of which are independent of the information to be displayed, and simultaneously applying a set of M drive voltage waveforms one to each of the second set of conductors, the instantaneous values of the M drive voltage waveforms applied to the second set of conductors being selected in dependence upon the information to be displayed such that their interaction with the predetermined set of voltage waveforms applied to the first set of conductors, integrated over the repetition period of said predetermined set, produces a visual display of said information, wherein at any instant, siibstantially half the conductors of the first set are driven at a first voltage level, and the remainder are driven at a second voltage level different from the first. Preferably the first set comprises an odd number of conductors, and the voltage distribution of the predetermined set of drive waveforms is such that at any instant, (N+1 )/2 of these conductors are driven at a first voltage level, and (N-1 )/2 are driven at a different voltage level. This slightly uneven voltage distribution in each of the predetermined instantaneous drive voltage patterns applied to an odd number of conductors is preferred in symbolic, alpha-numeric and other matrix displays in which the number of conductors in the first set is relatively small, typically less than 20. However, for a so-called oscilloscope display, involving the display of one or more individual traces, and thus normally requiring large numbers of drive conductors, an even number of conductors in the first set may be used with little reduction in benefit . In such an application, each instantaneous voltage pattern in the predetermined drive sequence will preferably contain an equal split between two drive voltage levels, ie half the conductors (N/2 ) are driven at one voltage level and the remaining half driven at a different voltage level.

Preferably each of the waveforms applied to the first set of conductors is a two-level binary coded waveform consisting of a sequence of logic ones and logic zeros, each representing a respective one of said two drive-voltage levels, which conveniently may be of equal magnitude but opposite polarity.

The sequence of two-level drive voltage patterns applied across the first set of conductors may comprise, in sequence, all the possible pattern combinations involving a substantially equal split between the conductors driven at the two different voltage levels . For example, in the preferred case of an

(N+1 )/ 2, (N-1 )/2 split in an odd number of conductors, this sequence will contain 2[N! ]/{[(N+1 )/2 ] ! [(N-1 )/2]! } patterns. However, the complete set of patterns may advantageously be halved by omitting one of each complementary pair of patterns, and further reductions may be made in the length of this reduced sequence by using only an appropriate subset of these patterns . In order to minimise the length of this sequence for a given NxM matrix display, the predetermined sequence of drive- voltage waveforms is preferably applied to the smaller set of conductors. The information-dependent drive-voltage waveforms applied to the second set of conductors may be two or more level waveforms.

For any desired 2-dimensional display pattern, in selecting the information dependent drive voltage waveforms applied to the second set of conductors, it is convenient to consider the pattern of wanted ON and OFF elements along each of the second set of conductors as a separate line feature of that 2-dimensional pattern, consisting of a wanted line pattern of ON and OFF elements. Thus, for a given two or more level waveform format, ie having a fixed voltage associated with each level, the instantaneous value of the waveform applied to each conductor of the second set is preferably selected, in relation to the values of each instantaneous pattern of waveforms applied across the first set of conductors, so as to maximise the instantaneous mean-squared contrast of the wanted line feature, averaged over that conductor. This will also substantially maximise the mean-squared contrast ratio of the whole information display, averaged both over all the conductors of the second set, and over the duration of the predetermined pattern sequence applied to the first set of conductors.

According to one embodiment of the present invention, the waveforms applied to both sets of conductors are two- level waveforms , comprising respectively two voltage levels V1,V2 of opposite polarity for the first set, and V3,V4 of opposite polarity for the second set of conductors. The magnitudes of V1,V2,V3 and V4 are such that the difference between each pair of voltages of equal polarity is below the switching threshold of the display elements, but such that the difference between each pair of voltages of the two waveforms having the opposite polarity, is above the switching threshold.

Thus, the above-mentioned maximisation ofthe contrast of the wanted line features along respective conductors of the second set may then be achieved by applying to each individual conductor of the second set, that voltage level, V3 or V4, which is of opposite polarity to the instantaneous voltage level, V1 or V2, applied to the conductors of the first set associated with the majority of wanted ON elements and/or the minority of the wanted OFF elements in the line feature along that conductor of the second set, whenever this condition arises.

By ensuring that there is an odd number N of conductors in the first set, one or other of the voltage levels V1 or V2 will always be associated with a majority of the wanted ON elements and/or a minority of the wanted OFF elements in the wanted line feature along any of the conductors of the εenond set.

In a simple embodiment of the invention, the voltages V1,V2,V3 and V4 are all of equal magnitude. Further improvements in both the overall contrast, and in the uniformity of the contrast ratio of each line feature across the display, may be achieved by applying three or more level waveforms to the conductors of the second sot whereby to selectively weight the brightness of the display in accordance with the pattern, and relative numbers, of wanted ON and OFF elements in each line feature. It will have become apparent that, because of the nature of the predetermined sequence of voltage patterns applied to the first set of conductors, involving a substantially equal split in the numbers of conductors which experience two different instantaneous voltage levels, it is not possible to ensure that all the wanted ON elements in any line feature are always switched ON and all the wanted OFF elements are always switched OFF. How closely the achievable pattern of ON and OFF elements matches a wanted line pattern of ON and OFF elements along a conductor of the second set, will vary from pattern to pattern in the predetermined waveform sequence applied to the first set of conductors, and will also depend on the nature of the wanted line pattern. Hence the contrast ratio achievable for a given information display, when integrated over the predetermined sequence of drive patterns, will vary from line feature to line feature across the display. However, this variation in the uniformity of the contrast of the display may be reduced by selectively weighting the drive-voltage levels applied to the conductors of the second set so as to ensure that the integrated brightness of the wanted ON elements in each line feature is substantially equal. Thus, the instantaneous potential appearing across each element of the display may vary between different values corresponding to different levels of brightness between fully OFF and fully ON, the number of different levels and their brightness being dependent respectively upon the number of different levels, and their voltages in the waveforms applied to the second set of conductors. The period of the drive waveforms applied to the two sets of conductors may be such as to maintain the instantaneous potential applied across each element of the display for a period either exceeding the response time of the elements, so that each element is fully switched into a state determined by that instantaneous potential ie direct response operation; or shorter than the response time leading to root-mean- squared (rms) operation, in which each element is switched to a state determined by the rms value of the potential appearing across it. To obtain acceptable performance in the direct response case, the duration of the predetermined sequence of drive voltage patterns applied across the first set of conductors should be shorter than the response time, or persistence of vision of the human eye. The achievement of this condition is facilitated by the present invention in that it enables shorter predetermined drive sequences to be used, with resultant economies in the cost, complexity and power consumption of drive generation distribution.

In general the luminence obtained is a highly non-linear function of the control voltage across an intersection. Thus, with the "direct-response" (DR) mode, the instantaneous voltage will generally switch each element to "black" (OFF) or "white" (ON), and the perceived luminence depends on the relative proportions of the total time for which the instantaneous luminence is "white" and "black" respectively. The eye averages this over the "persistence of vision" interval. Hence, in principle, the repetition period of the excitation waveforms should be shorter than the persistence of vision, and by ensuring that the proportion of "black" to "white" clock cycles does not vary unduly over the the total excitation period, a somewhat longer excitation period may well be perceptually acceptable. Thus a period of 100 m sec will normally be satisfactory. It will be shown later that the waveform period for a 7 x5 element character display can be restricted to only 10 clock cycles. This allows then up to 10 m secs per clock cycle. Hence any material response time constant less than, say, 5 m secs will normally permit the "direct-response" mode of operation. In the RMS mode, the RMS effect of the control voltage is averaged over the repetition period of the excitation waveform (assumed to be less than the response time-constant of the medium). Hence the resultant luminance is steady and is "white" if this RMS voltage is above an upper threshold, and "black" if it is below a lower limit. This mode is less restrictive on timings. If the displayed pattern is to be able to change in not more than say 0.5 seconds, the time constant of the material should not exceed 200 m secs. This peimits a total wavefoim repetition period of up to 80 m secs, and hence a clock cycle time of up to 8 m secs. At the other limit, power consumption may discourage clock frequencies exceeding 1 MHz, corresponding to a waveform period of 10 μsecs and a minimum material time constant of 20 μsecs (if there should be an explicit desire to achieve RMS rather than direct-response operation). Hence it is perhaps unlikely that the RMS mode would be found attractive at speeds higher than this, even if suitable materials were available.

Thus the timing constraints are broadly summarised in the following table:-

According to a second aspect of the present invention, a display apparatus comprises an opto-electric matrix display of the kind specified, means for generating a predetermined set of N cyclically repetitive drive-voltage waveforms, the values of which are independent of the information to be displayed, means for applying respective ones of said drive voltage waveforms to each of the first set of conductors of the display, to produce a predetermined sequence of instantaneous drive voltage patterns across the first set of conductors, means for generating a second set of M drive voltage waveforms, and means for applying respective ones of said M voltage waveforms to each of the conductors of the second set simultaneously with the application of the predetermined set of waveforms to the first set of conductors, the instantaneous values of the waveforms applied to the second set of conductors being dependent upon the information to be displayed, such that their interaction with the predetermined set of waveforms applied to the first set of conductors, integrated over the repetition period of said predetermirod set, produces a visual display of such information, wherein, in each instantaneous drive voltage pattern, in the predetermined set of waveforms, substantially half of the waveforms are at a first voltage level, and the remainder are at a second voltage level different from the first. Preferably, there is an odd number of conductors in the first set, and in each instantaneous drive voltage pattern, (N+1)/2 of the waveforms are at a first voltage level, and (N-1)/2 are at a second voltage level different from the first.

The means for generating and applying the first and second sets of drive voltage waveforms to the first and second sets of conductors may be adapted to operate the display in accordance with a method of the first aspect of the present invention as set forth above. The invention will now be described in greater detail, by way of example only, with reference to the accompanying drawings, of which

Figure 1 shows a plan view of a 7x5 element liquid crystal matrix display in accordance with the invention;

Figure 2 shows a section through the matrix display of Figure 1 along the line II-II;

Figure 3 shows a typical set of information display patterns that can be generated with a 7x5 dot matrix display of the kind shown in Figures 1 and 2;

Figure 4 shows, in block schematic form, a matrix display apparatus in accordance with the present invention and;

Figures 5(a) to 5(d) show various drive configurations for arrays of individual matrix displays in accordance with the present invention.

Referring to the drawings, Figures 1 and 2 show a NxM element matrix display 1 comprising two glass plates 2,3, the plate 3 carrying a set of spaced strip conductors X1 to XM (M=7) on its inner face, and the plate 2 carrying a set of spaced strip conductors Y1 to YN (N=5) on its inner face. The two sets of conductors are arranged at right angles to one another to provide a seven row, five column matrix of display elements, each one uniquely defined at the crossover between respective ones of the row and column conductors. These conductors may be formed in any known manner, typically by sputtering a layer of tin oxide on the appropriate surfaces of the plates 2,3 and then etching through a photolithographic mask.

A layer of liquid crystal material 6, the composition and thickness of which is selected in dependence upon the type of liquid crystal effect to be used and the desired operating characteristics in known manner, is contained between the plates 2,3 by an insulating spacer 7 which also serves to maintain a fixed separation between the plates. The display may be observed in any known manner, eg by light transmission using natural or artificial light illuminating the display from behind, or by reflection from a reflector placed behind the display. This latter arrangement is shown in Figure 2 in the application of the invention to a twisted nematic liquid crystal display, in which the liquid crystal material 6 sandwiched between the plates 2,3 is adapted in known manner to act as a polarisation-converting light guide in its relaxed state. The display 1 is thus sandwiched between a pair of orthogonal polarisation filters 10,11 arranged so that light incident on the top surface is polarised in one plane by the upper polarisation filter 10, and is rotated through 90º by the layer of liquid crystal material 6, so that it is accepted by the lower filter 11. The accepted light is then reflected back along the same path by a mirror 8 to produce (ideally) a 50% reflectance, the other 50% of the "wrong" polarisation having been absorbed by the first polarisation-selective filter 10. On application of an electric field of sufficient strength (of either polarity), in the present example, to individual areas by application of suitable switching voltages to the appropriate X and Y conductors, the effects which give rise to the polarisation in the liquid crystal will be overriden, destroying the polarisation-rotating character of the liquid crystal material resulting in an (ideally) zero reflection, or 'OFF' condition. Alternatively, the polarisation filters 10,11 may be parallel to each other, giving "full" 50% reflectance (ON) in the presence of a sufficiently strong electric field, and "zero" reflectance (OFF) in its absence. For convenience the latter scheme, in which a maximum potential across a display element yields a "full" reflectance white or 'ON' condition, and a zero potential yields a "zero" reflectance black or OFF condition, will be assumed hereafter. Further, the (50%) limiting reflectance achievable will be referred to as "full ON", with varying degrees of reflectance below this limit referred to as an appropriate proportion of "full ON", eg 70% of "full ON" reflectance(ie 35% absolute reflectance) will be referred to as 0.7 of full ON. Each set of conductors X1 to XM and Y1 to YN is driven by respective row and column drive circuits 13,14 arranged to provide two-level drive waveforms therefor. The drive waveforms applied to one set of conductors, advantageously the smaller set of column conductors Y1 to YN in this example, comprise a predetermined cyclically repeated sequence of instantaneous orthogonal voltage patterns across the associated set of conductors, the sequence being indpendent of the information to be displayed. The drive waveforms simultaneously applied to the other set of conductors, in this example, the row conductors X1 to XM, are then selected in dependence upon the information to be displayed, their instantaneous values being selected so as to optimise the display of this information according to the following criteria. Considering first the so-called direct response mode of operation in which the period of the mutually orthogonal drive waveforms applied to both sets of conductors is such that the instantaneous voltage applied across each element of the display is maintained for a sufficient period to enable that element to respond fully to that voltage, either to switch from an OFF condition to an ON condition, or vice versa, or to maintain its original condition as required.

This is achieved in the present examples using two- level binary coded waveforms, ie a series of logic ones and logic zeros represented by equal voltage levels of opposite polarity. Thus, logic one is represented by voltage V1 and logic zero by a voltage - V1. Application of a logic 1 or a logic 0 to both the row and column conductors associated with a particular display element, will produce a zero potential across that element, causing that element to remain or be switched OFF; while application of a logic 1 to one of these electrodes and a logic zero to the other will cause that element to remain or be switched ON.

In accordance with the present invention the predetermined sequence of drive voltage patterns applied across the odd number of column conductors Y1 to YN, is such that at any instant, (N+1 )/2 of the conductors are driven at one of the voltage levels, and the remaining (N-1 )/2 are driven at the other voltage level. Thus at most, this sequence will contain 2[N!]/{[(N+1)/2]! [(N-1)/2]!} different pattern combinations which in practice may be reduced to half this number by including only one of each pair of complementary patterns (ie ones and zeros interchanged.)

Once the predetermined column drive waveforms have been selected irrespective of the information to be displayed, the two-level binary row drive waveforms can then be selected in accordance with this information. For this purpose, it is convenient to consider the row of display elements associated with each one of the row electrodes X1 to X7 as a separate feature of the overall display. For any given information display pattern, each row must appear to contain a particular pattern of ON and OFF elements forming part of that display. Ideally, if thit were possible,each of these elements would be continuously maintained in its required state for as long as that particular information display pattern is required. However, for obvious reasons this is not possible in matrix displays of this kind, and it is therefore necessary to ensure that each element is maintained in its desired state for the major part of the time, and to rely on the persistence of vision of the observer^,s eye to give it the appearance of being in that condition.

Thus, for each column drive pattern in the predetermined drive sequence applied to the column conductors, it is only possible to generate one of two patterns of ON/OFF elements on each row conductor with a two-level drive waveform, by selecting an instantaneous drive of either the 1 or the 0 voltage level. Thus, to obtain the maximum instantaneous mean (or mean-squared) ON/OFF contrast along any given rowfor a particular row feature, it is necessary to select that row drive which will cause a majority of the desired ON elements and/ora minority of the desired OFF elements to be switched ON. For any desired pattern of ON and OFF elements along a row conductor, the instantaneous drive pattern applied to the column conductors will always apply either a 0 or a 1 column drive-voltage level either to a majority of the desired ON elements in that pattern or to a minority of the desired OFF elements, or both, simply because there is an odd number of column drive conductors. Thus, to obtain maximum contrast along that row, it is necessary then to select the opposite row conductor drive level to the column drive level which is associated with the majority of desired ON elements and/or the minority of desired OFF elements. In this way, the majority of desired ON elements along that row will experience the full ON voltage difference of 2V1 volts, while the majority of the desired OFF elements will experience zero drive voltage corresponding to fully OFF.

Thus, it this row drive selection procedure is applied to each row conductor for each column drive pattern in a suitable predetermined column drive sequence (hereinafter referred to as the frame sequence), then all the ON elements in the desired information pattern will experience the same ratio of full ON to full OFF excitations with full ON predominating, and all the OFF elements in this pattern will experience a common ratio of full ON to full OFF excitations, with full OFF predominating. When integrated over the frame sequence, assuming this to be shorter than the persistence of vision time, then each row feature will appear to have visually uniform ON and OFF contrast levels. However, the actual levels achieved will depend upon the total number of ON elements in a row and on the total number N of column conductors.

The performance that can be achieved may be analysed as follows:

Let the total number of ON elements in a row feature be W (representing white) and let the total number of OFF elements be B (representing black) . Further let any particular instantaneous column-drive pattern, in association with the associated instantaneous row drive, engender full reflectance for a majority w of the W ON elements and a minority b of the B OFF elements in the given row.

Hence w > W/2 and b ≤ B/2 (W odd, hence B even) or w ≥ W/2 and b < B/2 (W even, hence B odd) Within a complete sequence of column drive patterns

(containing all 2[N!] /{[(N+1 )/2] ! [(N-1 )/2] !} combinations) there will always be just one drive pattern fitting any one specific pattern of w ON elements and b OFF elements (together with its inverse, with the 0 and 1 drives interchanged) for a given row feature. The total number of such column-drive patterns in this complete sequence (which satisfy the condition of switching ON a majority w of ON elements and a minority b of OFF elements) is given by the number of ways of selecting w ON elements and b OFF elements out of W ON elements and B OFF elements, ie

R ( w) = (W!/(W- w)! wϊ ) . (B !/(B-b)! .b! )

It should be noted that two variants of the right-hand term must normally be considered, entailing b ₁ = (N-1 )/2 - w and b ₂ = (N+1 )/2 -w, since each drive pattern splits N into two slightly unequal numbers . R (w) is then the sum of these two contributions .

Further, because all such drive patterns are spread equally over the W wanted ON elements, during the total sequence of drive patterns, they contribute full ON . reflectance to these W elements for w/W of the time. Similarly, they also contribute an average reflectance of b/B to each of the B elements intended to be OFF. However, by ensuring that w/W> ½ >b/B, although either one at a time can become equal to ½, the W ON elements will always reflect more strongly than the B OFF elements in the row.

The total reflectance of the ON and OFF elements of a row is then given by the resultants arising from the summations of the above functions for all relevant; values of w , weighted by their relative frequencies of occurrence. These overall reflectances, L. (light ) for the ON elements, and D (dark) for the OFF elements, are given by:-

Again it should be noted that, each summation is in fact the sum of two constituents, one being a function of (w,b₁) and one of (w,b₂). The limits of w for these summations are given below:

Table 2 below gives the total reflectances, Lfor the W wanted ON elements and D for the B wanted OFF elements for various numbers of wanted ON elements (w) in a given row, for displays having different numbers of columns (or elements in a row).

For displays having large numbers of columns (N →∞), and when N>>W, then the function of B can be neglected in computing R (w) and the overall reflectance D of the wanted

OFF elements reduces to 0.5, but that for the ON elements becomes :

substantially as before.

However, simplifies to 2^W .

Here the parameter x defines the range of summation, and w is a simple function of x as follows :- w = x when x ≥ W/2 w = W - x when x<W/2

These conditions, which assume that the number of pattern elements as opposed to background elements in a given row feature is small in relation to the number N of elements in the row, are represented at the end of Table 2, a) for a white- on-black display, and b) for a black-on-white display.

The former involves W ON elements in row of N elements, and the latter comprises (N-W) ON elements, ie W should be interpreted as B for this particular case.

In general, the results of Table 2 show that the contrast L/D is best if the wanted pattern is black-on-white

(ie B<W) together with a small number of columns N and a small ratio B/N. Further, a considerable reduction in the spread of L and D reflectance values can be achieved by ensuring that B and W are both always greater than 1.

It will be seen that in any given combination of N, and W values, the principal degradation of contrast arises from the combination of w,b values with the largest b/w ratio.

However, marked improvements in the contrast can be achieved by applying three-level information-dependent row drive waveforms to the row conductors X1 to XM to effectively introduce a weighting for these wors+ case situations. In the present example this may be achieved by introducing a zero voltage level in addition to the +V1 and -V1 voltage levels previously used to represent the logic 1 and logic 0 respectively, and to apply this zero voltage level wherever this worst case situation arises along a row conductor, ie whenever the ratio of b/w is a maximum. This would inhibit any reflectance from both wanted ON and wanted OFF elements at that time, thereby improving the overall contrast, but would substantially reduce the mean reflectance of the ON elements as well as increasing the spread of this ON reflectance level between different values of W. However, any row drive voltage level imteimediate between + VI (representing full ON) and zero volts, providing it is sufficient, in combination with the predetermined column drive voltage levels to cause a majority w of the W wanted ON elements to (at least partially) switch ON, will maintain the same contrast for any particular combination of w, b, values, but will reduce its contribution to the integrated light and dark levels L,D. Hence, as a compromise, the row drive voltage levels may be selected, for different wanted row features in combination with the predetermined column drives, which make the ON reflectance levels L the same for all row features. However, this selective weighting of the row drive waveforms requires an increased number of different drive voltage levels as illustrated in the examples given in Table 3 below for displays having 3,5 and 7 columns. In this table, the entries under w,b define the conditions under which drive amplitudes other than + V1 are required, the entries under "wt" give the required weighting, ie the ON reflectance levels required with the drive amplitude used on these occasions as a proportion of full ON reflectance, and L and D give the resultant ON and OFF reflectance levels relative to full ON reflectance.

Thus for a five-column display (N=5), a weighted row drive voltage of 0 volts is applied whenever the number W of wanted ON elements in that row is 0. When the number W of wanted ON elements is 3 in a given row feature, then a row drive-voltage which produces full ON reflectance is applied whenever the

instantaneous predetermined column drive pattern dictates that two of the three wanted ON elements will be switched ON (w = 2 ) together with only one of the wanted OFF elements (b = 1 ) . For all other values of w, b for that particular value of W, the full

ON voltage of + V1 volts is applied. This situation will arise for example, when the wanted row feature is:—

1 0 1 0 1 (0 for OFF, 1 for ON) and the instantaneous column drive pattern is :— 1 0 1 1 0 (1 for +V1 , 0 for- V1 ) w w b

The labels w and b indicate those elements of the row which contribute to the val"es cf w and b in this case, ie the two out of three wanted ON elements, and the one no t of two wanted OFF elements which are switched ON. In order to ensure that a majority of the wanted ON elements and a minority of the wanted CFF elements are switched ON, the row drive voltage is selected to be of opposite polarity to the column drive level associated with the majority of the wanted ON elements, ie it must be negative. This will cause two of the three vranted ON elements to switch ON (w=2), together with one of the wanted OFF elements (b=1 ), thus meeting the condition specified in Table 3, in which the row drive voltage should be weighted to produce full ON reflectance for those elements which are

actually switched ON.

Two further examples of row drive weighting systems for a five column display are given in Table 4 below, requiring only 4 instead of 5 voltage levels. The second of these weighting systems, producing relative reflectances of 1,0.67 0.25 and 0; yield a lower overally uniform ON reflectance L, coupled with a better contrast ratio, L/D.

The application of the invention to a 7x5 dot-matrix character display, using the display panel 1 of Figures 1 and 2, will now bs described with reference to Figures 3 and 4. Figure 3 shows a typical set of alphabetical and numerical characters together with punctuation marks and other "housekeeping?' symbols . It will be seen that this set of symbols requires only 20 of the possible 32 row features or patterns of ON and OFF elements. Hence any of the characters shown in Figure 3 can be generated by appropriate selection of 7 out of 20 possible information—dependent row-drive waveforms for application to the row conductors X1 to X7 of the display panel.

With regard to the 5 predetermined information-indpendent column-drive waveforms, it can be shown that the number of ways of splitting the 5 columnsinto three logic 1 drives (+V1 ) and two logic 0 drives ( V1 ) is only 10. Thus, by inverting an appropriate set of 5 of these 10 into three logic 0 drives and two logic one drives, all requirements can be met by a sequence of only 10 column drive patterns, each containing five logic 0 levels and 5 logic 1 levels, for example as set out in Table 5 below.

Referring now to Figure 4, this sequence of 10, 5-bit drive patterns, which define the predetermined drive waveforms for the five column conductors Y1 to Y5, is stored in a readonly memory 20 (ROM) from which it is read-out cyclically under the control of a synchronising clock 21 into the column drive circuit 14 of the display panel 1 . The column drive circuit 14 is operative in known manner to apply an appropriate pattern of +V1 and - Vl voltage levels, for the duration of each clock period, across the column conductors in accordance with each successive pattern of logic 1 s and Os in the drive pattern sequence read from the ROM 20. As shown, successive patterns of Os and 1s are transmitted from the ROM 20 to the drive circuit 14 in 5-bit parallel form, although in some applications it may be preferable to serially read out each logic bit from the ROM 20, and to transmit the total sequence in time division multiplexed, format . Each of the 20 different 10-element long row drive sequences, which define the 20 possible row features required to generate the 50 different characters set out in

Figure 3, are also stored in a read— only memory (ROM) 24. Each of these 10-element feature sequences is identified within the ROM 24 by a 5-bit binary word, and the appropriate 7, 5- bit binary words which identify which of the 7 stored row drive sequences are required to generate a particular character are selected by a control unit 26 in response to an output signal from a keyboard 27, or other external source defining the desired pattern of alpha-numeric (or other) symbols . The 7 selected row-drive sequences are then cyclically read out from the ROM 2 into an electronic switching network 28, in synchronism with the column drive sequence read out of ROM 20 under the control of the synchronising clock 21 . The switching network 28 is operative to apply these row-drive sequences to appropriate ones of the row conductors X1 to X7 via the row drive circuit 13 whereby to control the instantaneous drive voltages applied to the respective row conductors.

In its simplest form, in which two-level row-drive , waveforms are used, each element of each row drive sequence will comprise one binaiy bit, 0 or 1 , which determines respectively whether a voltage level +V1 or -V1 is to be applied to the associate row conductor. However, where selective weighting of the row-drive waveforms is to be applied, each element of each row drive sequence contains additional bits defining the weighting factor to be applied. In the case of either of the 4-level weighting systems set out in Table 4, this will require only two additional bits in each element to specify the weighting factor. Thus, each element of the row-drive sequences stored in ROM 24 comprises three bits, one to determine the polarity of the required voltage and two to determine which of the four weighting levels is to be used. Accordingly, the row-drive circuit 13 contains means for generating voltages of the appropriate level and polarity in accordance with this information.

Using a liquid crystal display panel having response and relaxation times both under 3 milliseconds, a flicker free display having a frame repetition rate within the persistence of vision of the human eye can be achieved using a clock period of 6 milliseconds, ie the do level applied across each element of the display is maintained for 6 milliseconds, producing a frame repetition period of 10 x 6 = 60 milliseconds. Cycle times of this order allow ample time for any multiplexing in the generation and distribution of the various row and column drive sequences.

By way of an example, to generate a black-on white display of the figure 8 as shown in Figure 3, using the predetermined column drive sequence set out in Table 5 and working through the sequence from the top to the bottom of the table, the following weighted drive sequences using the 1,0.7,0.33,0 weighting system of Table 4, are applied to the row conductors of the display (working from right to left).

Each entry in Table 6 indicates the polarity and the weighting factor of each element of the row sequence to be applied to each row conductor in synchronism with the application of the two level column drive sequence to the column conductors set out in Table 5. The fact that the weights for all 7 row-conductors are equal arises from the fact that the figure 8 entails only two mutually complementary line features (10001 and 01110). Thus it is not representative of all other symbols.

Although in the above embodiment, a possible selection of 20 different row-drive sequences has been used, it is in fact possible to reduce this to 14, because 6 of the 20 row features which they represent are complements of 6 others (ON and OFF elements interchanged), and so each such pair can be derived from a single row-drive sequence.

While the invention has been described above in its application to the direct response mode of operation, in which the instantaneous potentials applied across each element of the display are maintained for a period exceeding the response and relaxation times of the liquid cyrstal material, it may also be extended to the so-called mean-squared (slow- response) mode of operation, in which the resultant reflectance of each element is related to the root-mean- square (rms) voltage appearing across it. values L and D given on Table 2 are means of a sequence of

1s end Os, they are equally also mean squares, and it will be noted that for a 5 column display, the individual rows all provide L/D contrast ratios of 2.33 or better. These contrast ratios could be achieved in an rms responding display using for example nematic liquid crystal materials having a relaxation time of 100 milliseconds, and using a frame cycle time of 40 milliseconds, achievable using a 4 millisecond clocking period in the above embodiments. Again these speeds would present no problem in waveform generation and distribution, even when matching to much faster crystal response speeds.

However, by applying suitable selective weighting to the pattern-dependent row drive waveforms, it is possible to further increase the maximum contrast ratios achievable.

If the pattern dependent drive voltage associated with a specific combination of w and b values is increased so as to increase the full ON drive voltage from 1 (represented by respective row and column drives of +V1 and -V1 or vice versa) to 1 +

, then w ON elements and b OFF elements

along a row are instantaneously driven to a voltage of (1 +

and the remaining (W - w) ON element and (B-b) OFF elements are driven to a voltage of . Hence, it can be shown that the

"instantaneous" mean -squared contrast is given by:-

Since, by definition, w/W>b/B , application of an increased weighting voltage in situations which cause both w/W and b/B to increase equally (and conversely any decrement in the weighting which causes these terms to decrease equally) can only degrade the contrast.

However, the unwejghted mean-squared drive voltage over the whole cycle is given by:-

Hence, it is advantageous to increase, by means of an incremen , the weight applied to the pattern dependent

column drive voltage in situations involving an above-average value of w/b up to the point at which no further benefit arises . Similarly it is advantageous also to decrease, by means of a decrement , the weight applied to column drive

voltage in situations involving a below average value of w/b. It should be noted that the weighting value

and £used here represent changes in the drive voltage, rather than the effect of these weights on the reflectances as expressed in Tables 3 and 4.

If the mean-squared light and dark levels or reflectances are L _o and D_o in the absence of any weighting, ie using two level waveforms of +V1 for both the row and column drives, then for a particular combination of w and b values in which the pattern dependent drive voltage is incremented by an increment

the resultant contrast S is given by:-

These formulae give the magnitude and sign of the optimum velue of

, both for scaling, up the reflectance contributions from combinations of w and b values with better than average contrast, and for scaling down those with worse than average contrast. Direct application of even the simplified formula is normally sufficient for most applications, although exact calculations using the unsimplified formula requires some iteration, since each value of

modifies the values of L and D to be used in computing others.

Table 6 below lists suitable increments

for the five pattern dependent drive voltages to be applied to a 5 column display (the value

epresentating the proportion of the drive voltage V1, together with its polarity, which is applied to the two-level pattern dependent row drive conductors ) for different values of W with different combinations of w and b, to achieve on overall ratio of 2.7 (ie L/D = 0.94/0.35).

The conditions to be satisfied in determining the appropriate weighting increments (or decrements) are:—

0.94 = L_o and 0.35 =

where L_o and D_o correspond to the values of L and D in Table 2 (ie without any increments Q )

It will be seen that there are a total of 12 increments (including zero), requiring an additional 4 bits to identify each one in the ROM 24 and its output control signals (as opposed to two for a 4 level incrementing system described) .

The matrix addressing system of the present invention also enables a simplification in the topological problems associated with addressing closely spaced arrays of liquid crystal displays, eg 7 x 5 dot-matrix character displays . Typlically, such displays may comprise a plurality of individual matrix displays of the kind described above in which the liquid crystal material of each display is provided between a pair of glass plates common to the whole array. In such an arrangement, difficulties may arise where it is required to make external drive connections to all the individual displays from the edge of the plates .

In the case of the predetermined pattern-independent column drives, all, or selected groups of displays in the array may be connected in series to a common set of 5 column drive connections at the edge of the plates, by providing all the column drive conductors of the interconnected displays on the same plate, say the top plate, of the array. Several possible configurations are schematically shown in Figures 5

(a) to (d ), in which the connecting paths 30 each comprise five parallel conductors, one for each column conductor of the displays 1. In the arrangement shown in Fig 5(a), the order of the column drive sequences will be reversed in alternate columns of the array for identical display characters. A second option is to provide an external fly-back between the columns of the array as shown in Fig 5(b); in this case all the array columns behave identically. Alternatively, a spiral drive path as shown in Figure 5(c) may be used to the same effect. In Fig(d), economies in the space required for the interconnection paths is achieved at the expense of using two sets of 5 feed points and terminations at the edge of the display. With regard to the pattern—dependent row drives in such an array, it is clearly not possible to provide a common set of row—drive waveforms for a number of displays, because each one may be required to display a different character. However, the number of edge connections required at the edge of the display panel may be minimised by the use of time division multiplexed row drives for a number of individual displays, and then serial-to-parallel converting these multiplexed row-drives using shift registers, one placed adjacent the edge of each individual display (or column of displays). Several such shift registers may also time-share a common drive circuit. Although the invention has been specifically described in its application to character displays, it may also be applied to single or multi-trace waveform displays . However, because such displays require relatively large numbers of drive conductors, the benefits of using odd numbers of predetermined-drive conductors becomes marginal. In the case of a single-valued black-on-white waveform display, with the predetermined drive sequence applied to N row .

(horizontal ) conductors, the column drive sequence applied to each of the M column conductors merely replicates the predetermined row drive-waveform of the row in which the one desired OFF element is contained, thus producing zero voltage difference across that element . The remaining elements in each column would all be switched fully ON for the 50% of the time when their pre-determined row drive is of opposite sign to that of the selected OFF element, and fully off during the other 50% of the drive sequence.

For a two or more traced waveform display, this drive scheme is not possible, and it is then necessary to revert to the normal considerations discussed earlier for a character display. In the case of a two-trace display, each column drive conductor would be driven by a waveform which replicates the instantaneous value of the two instantaneous row drives applied to its two wanted OFF elements, when these are both equal (w= 2, b=0), ie both +V1 or both-V1 ; and when they are unequal (w=1 ,b+1 ), it may be arranged to apply either +V1 or -VI with equal probability. However, for direct response operation, since these latter drive voltages would contribute equally to the mean reflectivity of both ON and OFT elements, it is preferable to apply zero column drive voltage in circumstances when the instantaneous row drives to the two wanted OFF elements are unequal.

For a triple-trace display, any such zero- voltage weighting of the column drive waveforms may be applied to all situations in which the instantaneous row-drive voltages for the three wanted ON elements in a column comprise a combination of 2, +V1 drives and 1 , — V1 drive (or vice versa) . ie when w = 2 and and b=1 . The light and dark reflectance levels L and D for double (W =2) and triple— trace (W=3) waveform displays are given in Table 8 for a direct response display. In this table, "wtd" means weighted, and represents zero drive voltage for the w=1 , b=1 condition for two traces, and w=2, b=1 for three traces . Further, for a black-on-white display, the value 'W' should be interpreted as 'B ' .

With such waveform displays, the number of predetermined drive waveforms will normally be too large for the predetermined drive sequence (applied to N row conductors) to encompass all N1/(N/21)² combinations of N/2, +V1 drives and N /2, -V1 drives, assuming these to be equal, in any sensible frame time, and a suitable sub-set of these which ensures that the N constituent row line drive waveforms are mutually orthogonal may be chosen, preferably enabling the frame time to be reduced to a minimum duration of N clock periods . Even then , mean-squared operation may still be necessary where crystal relaxation times shorter than the required clock period are unobtainable.

However, as mentioned earlier, mean-squared operation does not alter the drive conditions or performance for a single trace display. For a two- trace display using the zero voltage weighting technique for the cases w=2, b=1 , some loss of contrast does result . In the case of a three-trace display, a slightly scaled-down column drive is desirable in situations when w=2, b=1 . In such an application, the optimum weight for this situation i

approximated in Table 9 below, by 0.8. In Table 9, the values of the light and dark reflectance levels are given for two and three trace rms displays in the same manner as Table 8 for direct response displays . Again, for black-on-white, 'W' should be interpreted as 'B' .

It will be seen that black-on-white displays will give an RMS L/D rati

2, in all cases concerned. For any of the display systems described above, selection of the information-dependent drive waveforms may involve a small DC component appearing across the display elements. Where this is significantly deleterious and cannot conveniently be avoided by suitable adjustment of the weighting, each sequence of the row and column drive waveforms may be followed by its complement.

It will be understood that invention may also be applied to the matrix addressing of electric-optic displays employing the so called "two-frequency" method of matrix addressing,in which a high-frequency component is superimposed upon the basic drive voltage waveform applied to one or both sets of conductors as described in Benjamin R. Matrix-Addressed Liquid Crystal Displays," pp281-282,The Radio and Electronic Engineer, Vol 50 No 6, June I980. It will be understood, therefore, that any references to waveforms herein are intended to cover the basic drive waveforms as hereinbefore described whether or not they include a superimposed high frequency component used in the "two-frequency" mode. Further, although the invention has been described in its application to displays in which the mode of operation is either purely "direct-response", or purely RMS, liquid crystal materials having intermediate response times, producing a compromise betwee the RMS and direct-response modes of operation, may also be used.

Claims

1. A method of addressing an opto-electric matrix display of the kind specified, including the steps of applying a predetermined cyclically repetitive set of N drive-voltage waveforms, one to each of the first set of conductors, the values of which are independent of the information to be displayed, and simultaneously applying a set of M drive voltage waveforms one to each of the second set of conductors, the instantaneous values of the M drive-voltage waveforms applied to the second set of conductors being selected in dependence upon the information to be displayed such that their interaction with the predetermined set of voltage waveforms applied to the first set of conductors, integrated over the repetition period of said predetermined set, produces a visual display of said information, wherein at any instant, substantially half the conductors of the first set are driven at a first voltage level, and the remainder are driven at a second voltage level different from the first.

2 A method as claimed on Claim 1, wherein the first set of conductors comprises en odd number of conductors, and the voltage distribution of the predetermined set of drive waveforms is such that at any instant (N+1 )/2 of these conductors are driven at a first voltage level and (N—1 )/2 are driven at a different voltage level.

3 A method as claimed in Claim 1 or Claim 2 wherein the number of conductors in the first set is less than 20.

4 A method as claimed in Claim 1 , 2 or 3 wherein the display has a larger number of conductors in the second set than in the first set. 5 A method as claimed in any preceding Claim, wherein the waveforms applied to the first set of conductors is a two-level binary coded waveform consisting of a sequence of logic ones and logic zeros each representing a respective one of the two drive- voltage levels.

6 A method as claimed in Claim 5, wherein the two drive- voltage levels are of equal magnitude and opposite polarity.

7 A method as claimed in Claim 5 or Claim 6 wherein the predetermined sequence of drive-voltage patterns applied to the first set of conductors contains no complementary pairs of binary drive voltage patterns.

8 A method as claimed in any preceding Claim, wherein the voltage waveformsapplied to the second set of conductors each comprise a sequence of fixed voltage levels applied in synchronism with the sequence of predetermined drive-voltage patterns applied to the first set of conductors.

9 A method as claimed in Claim 8, wherein the instantaneous voltage level of the waveform applied to each conductor of the second set is selected in relation to the voltage levels of each instantaneous pattern of waveforms applied across the first setof conductors, such as to substantially maximise the instantaneous contrast of the contribution to the desired information display, of the display elements associated with that conductor, averaged over that conductor.

10 A method as claimed in Claim 8 or 9 wherein the instantaneous voltage level of waveforms applied to each of the second set of conductors is selected in relation to the voltage levels of each instantaneous pattern of waveforms applied to the first set of conductors such as to ensure that all, or at least a majority of elements required, in accordance with the desired information display, to produce a first of two contrasting optical display properties and/or none, cr at least a minority of those required to produce the second of said two contrasting optical properties, are caused to produce said first optical property.

11 A method as claimed in any preceding Claim, wherein the waveforms applied to both sets of conductors are two-level waveforms comprising respectively two voltage levels of opposite polarity for the first set, and two voltages of opposite polarity for the second set of conductors, the magnitude of the voltages being such that the difference between each pair of voltages of the same polarity is below the switching threshold required to produce a desired optically contrasting change in the optical property of the display elements, but such that the difference between each pair of voltages of opposite polarity is above the switching threshold required to produce a desired change in the optical property of the display elements.

12 A method as claimed in Claim 11, wherein the voltages applied to the first and second sets of conductors are all of equal magnitude.

13 A method as claimed in any of Claims 8 to 10, wherein the voltage waveforms applied to the second set of conductors each comprise three or more level waveforms, and the instantaneous voltage level applied to each of the conductors of the second set is selected in relation to the required contribution of the display elements associated with that conductor to the desired information display and to the instantaneous pattern of drive-voltage levels applied to the first set of conductors, such that the contrast ratio between display elements required to produce optically contrasting optical properties in accordance with the information display, is substantially uniform over the whole display when integrated over the predetermined sequence of drive voltage patterns applied to the first set of conductors.

14 A method as claimed in any preceding Claim, wherein the period of the drive waveforms applied to the two sets of conductors is such as to maintain the instantaneous potential applied across each element of the display for a period exceeding the optical response time of the elements so that each is fully switched into state determined by that instantaneous potential during each said period.

15 A method as claimed in any one of Claims 1 to 13 wherein the period of the drive waveforms applied to the two sets of conductors is such as to maintain the instantaneous potential applied across each element of the display for a period shorter than the optical response time of the elements such that each element is switched to a state determined by the RMS value of the potential appearing across it. 16 A method as claimed in any preceding claim wherein the display is a liquid crystal display.

17 A method of addressing an opto—electric matrix display substantially as hereinbefore described with reference to the accompanying drawings. 18 Display apparatus comprising, an optc—electric matrix display of the kind specified, means for generating a predetermined cyclically repetitive set of N drive-voltage waveforms, the values of which are independent of the information to be displayed, means for applying respective ones of said drive voltage waveforms to each of the first set of conductors of the display, to produce a predetermined sequence of instantaneous drive voltage patterns across the first set of conductors, means for generating a second set of M drive voltage . waveforms, and means for applying respective ones of said M voltage waveforms to each of the conductors of the second set simultaneously with the application of the predetermined set of waveforms to the first set of conductors, the instantaneous values of the waveforms applied to the second set of conductors being dependent upon the information to be displayed, such that their interaction with the predetermined set of waveforms applied to the first set of conductors, integrated over the repetition period of said predetermined set, produces a visual display of such information, wherein, in each instantaneous drive-voltage pattern in the predetermined set of waveforms, substantially half of the waveforms are at a first voltage level, and the remainder are at a second voltage level different from the first.

19 Display apparatus as claimed in Claim 18 wherein said firstset comprises an odd number of conductors, and the generating means associated with the first set of conductors is arranged to drive (N+1 )/2 of said conductors at a first voltage level, and (N-1 )/2 at a different voltage level.

20 Display apparatus as claimed in Claim 18 or 19, wherein the number of conductors in the first set is less than 20. 21 Apparatus as claimed in Claim 18 , 19 or 20, wherein the display has a larger-number of conductors in the second set than in the first set .

22 Apparatus as claimed in any one of Claims 18 to 21 , wherein the waveforms applied to the first set of conductors by said generating means are two-level binary waveforms consisting of a sequence of logic ones and zeros each representing a respective one of the two drive-voltage levels .

23 Apparatus as claimed in Claim 22, wherein the tvro drive- voltage levels are of equal magnitude and opposite polarity.

24 Apparatus as claimed in Claim 22 or Claim 23, wherein the predetermined sequence of drive-voltage waveforms generated by the generating means associated with the first set of conductors contains no complementary pairs of binary drive-voltage patterns .

25 Apparatus as claimed in Claim 22, 23 or 24, wherein said generating means associated with the first set of conductors includes storage means in which said predetermined patterns of drive-voltage waveforms are each stored in the form of a binary word, these w ords being read from the store in the pred etermined sequence, to generatethe appropriate waveforms .

26 Apparatus as claimed in any one of Claims 18 to 25, wherein the generating means associated with the second set of conductors is arranged to apply to each of the second set of conductors a waveform comprising a sequence of fixed voltage levels in . synchronise with the predetermined sequence of drive-voltage patterns applied to the first set of conductors . 27 Apparatus as claimed in Claim 26, wherein the instantaneous voltage level of the waveform applied to each conductor of the second set is such, in relation to the voltage levels of each of the instantaneous patterns of waveforms applied across the first set of conductors, as to substantially maximise the instantaneous contrast of the contribution to the desired information display of the display element associated with that conductor, averaged over that conductor.

28 Apparatus as claimed in Claim 26 or 27, wherein the instantaneous voltage level of the waveforms applied to each of the second set of conductors is such, in relation to the voltage levels of each instantaneous pattern of waveforms applied to the first set, that all, or at least a majority of elements required, in accordance with the desired information display, to produce a first of two contrasting optical display properties and/or none, or at least a minority of those required to produce the second of said two contrasting optical display properties, are caused to produce said first optical property.

29 Apparatus as claimed in any one of Claims 18 to 28, wherein the waveforms generated by the waveform generating means associated with both sets of conductors are two-level waveforms comprising respectively two voltage levels of opposite polarity for the first set of conductors, and two voltage levels of opposite polarity for the second set of conductors, the magnitudes of the voltages being such that the difference between each pair of voltages of the same polarity is below the switching threshold required to produce a desired optically contrasting change in the optical property of the display element, but such that the difference between each pair of voltages of opposite polarity is above said switching threshold. 30 Apparatus as claimed in Claim 29, wherein the voltages applied to both sets of conductors are all of equal magnitude.

31 Apparatus as claimed in any one of Claims 26 to 28, wherein the generating means associated with the second set of conductors is arranged to apply to each conductor of the second set a three or more level waveform, the instantaneous voltage level applied to each of the conductors of the second set being such, in relation to the required contribution of the display elements associated with that conductor to the desired information display, and to the instantaneous pattern of drive-voltage levels applied to the first set of conductors, that the contrast ratio between display elements required to produce optically contrasting properties in accordance with the information display is substantially uniform over the whole display when integrated over the predetermined sequence of drive-voltage patterns applied to first set of conductors.

32 Apparatus as claimed in any one of Claims 26 to 31, wherein the waveform generating means associated with the first set of conductors includes first storage means in which information regarding the drive-voltage patterns of the predetermined sequence is stored, the generating means being arranged to cyclically read the information from the store in accordance with said predetermined sequence, the generating means associated with the second set of conductors includes second storage means containing information relating to the drive-voltage waveforms required to be applied to any of said second set of conductors in order to produce any desired display pattern in the display elements associated with that conductor when integrated over the repetition period ofthe predetermined sequence of drive-voltage patterns applied to the first set of conductors, and the apparatus further includes means for selectively reading from the second storage mpans waveform information required to generate a required information display and means for selectively generating and applying drive-voltage waveforms in accordance with that information to respective conductors of the second set cyclically and in synchronism with the predetermined set of waveforms applied to the first set of conductors.

33 Apparatus as claimed in any one of Claims 18 to 32, wherein the period of the drive waveforms applied to the two sets of conductors is such as to maintain the instantaneous potential applied across each element of the display for a period exceeding the optical response time of the elements so that each, element is fully switched into a state determined by that instantaneous potential during each said period.

34 Apparatus as claimed in any one of Claims 18 to 32, wherein the period of the drive waveforms applied to each of the two sets, of conductors is such as to maintain the instantaneous potential applied across each element of the display for a period shorter than the optical response time of the elements such that each element is switched to a state determined by the RMS value of the potential appearing across it.

35 Apparatus as claimed in any preceding Claim, wherein the display is a liquid crystal display.

36 Apparatus as claimed in any preceding Claim, including a plurality of opto—electric displays in which the first set of conductors of each display is connected to be driven by a common set of predetermined drive-voltage waveforms, the second set of conductors of each display being driven independently of one another. 37 Apparatus as claimed in Claim 36, wherein said displays are interconnected substantially as shown in and as hereinbefore described with reference to any one of Figs 5(a) to 5(d ).

38 Display apparatus substantially as shown in and as hereinbefore described with reference to the accompanying drawings.