CN101882423B - Methods for driving electro-optic displays - Google Patents

Abstract

An electro-optic display having a plurality of pixels divided into a plurality of groups is driven by selecting each of the plurality of groups of pixels in succession and applying to each of the pixels in the selected group either a drive voltage or a non-drive voltage, the scanning of all the groups of pixels being completed in a first frame period; repeating the scanning of the groups of pixels during a second frame period, and interrupting the scanning of the groups of pixels during a pause period between the first and second frame periods, this pause period being not longer than the first or second frame period.

Description

The method that drives electro-optic displays to use

Dividing an application of the dividing an application of the Chinese patent application No.200480018385.9 that the application's denomination of invention that to be applicant E Ink Corp. submit on Dec 28th, 2005 is " method that drives electro-optic displays to use " (submit on March 31st, 2008, application number be 200810088490X).

The present invention relates to the method that drives electro-optic displays to use.Method of the present invention especially, although be not, for driving bistable electrophoretic display device (EPD) to design.

The application's book is closely related with international patent application book PCT/US02/37241 (notification number WO03/044765) and PCT/US2004/10091, and the content of these documents is familiar with supposition reader in following description.

When term " electric light " is used in material or display, here with it, the traditional meaning at imaging technique is used, refer to a kind of material with the first and second show states different at least one optical characteristics, by applying electric field to this material, this material is changed into its second show state from its first show state.Although this optical characteristics is generally the appreciable color of human eye, but it can be another optical characteristics, such as light transmission, reflectance, luminous or in the situation that the demonstration designing for machine reading, the pseudo-color in the meaning changing in the electromagnetic wavelength reflectance that exceeds visible range.

Term " gray states " is here used with its traditional meaning at imaging technique, refers to the intermediateness between two extreme optical state of pixel, not necessarily means that the Hei-Bai between these two extremities changes.For example, several patents and published application for patent relate to the electrophoretic display device (EPD) the following describes, and wherein this extremity is white and dark blue, so middle " gray states " is actually light blue.Really, just as already noted, the transformation between these two extremities may not be color change.

Term " bistable " and " bistable state " are used in the traditional sense of imaging technique here, refer to and comprise that its display element has the display of the first and second show states different at least one optical characteristics, make any given element by after the addressing pulse driving with a limited duration, just get its first or its second show state, after addressing pulse finishes, this state will continue change at least several times of minimum length in time of the desired addressing pulse of state of this display element, for example at least four times.At international patent application book WO 02/079869, show, it is stable that some electrophoretic display device (EPD) based on particle can have not only gray scale under its extremely black and white state, and be also stable in Intermediate grey states, the electro-optic displays situation of some other type is also like this.Such display can suitably be called " multistable ", rather than bistable, although for simplicity, term " bistable " here can make to show for both containing bistable, contain again multistable demonstration.

Term " pulse " traditional meaning use at imaging technique with it here, refers to the integration of voltage to the time.But some bistable electro-optical medium plays electric charge transmitter, thereby adopting such medium, can use one of pulse to replace definition, is exactly that electric current is to the integration of time (equaling applied total electrical charge).The suitable definition of pulse should depend on that this medium has been voltage time pulse transducer, or the effect of charge pulse transducer.

As described at above-mentioned WO 03/044765 and PCT/US2004/10091, the electro-optic displays of known several types, for example, rotate Double-color film type, for example, and as U.S. Patent No. 5,808,783; 5,777,782; 5,760,761; 6,054,071; 6,055,091; 6,097,531; 6,128,124; 6,137,467; With 6,147,791 is described, and electrochromism type; For example see the people such as O ' Regan, B., Nature 1991,353, and 737; And Wood, D., InformationDisplay, the article of 18 (3), 24 (March 2002).Separately see Bach, the Adv.Mater. that the people such as U. show, 2002,14 (11), 845.Again for example, U.S. Patent No. 6,3019038, international patent application is announced No.WO 01/27690 and in U.S. Patent application book 2003/0214695, is also described this Nanochromic film.

Having become the electro-optic displays of the another kind of type of nervous research and development object for many years, is the electrophoretic display device (EPD) based on particle.Manyly yield Massachusetts Institute of Technology (MIT) (MIT) and E.InkCorporation (company) or described such display with its patent and the application form of name application; For example, see U.S. Patent No. 5,930,026; 5,961,804; 6,017,584; 6,067,185; 6,118,426; 6,120,588; 6,120,839; 6,124,851; 6,130,773; 6,130,774; 6,172,798; 6,177,921; 6,232,950; 6,249,721; 6,252,564; 6,262,706; 6,262,833; 6,300,932; 6,312,304; 6,312,971; 6,323,989; 6,327,072; 6,376,828; 6,377,387; 6,392,785; 6,392,786; 6,413,790; 6,422,687; 6,445,374; 6,445,489; 6,459,418; 6,473,072; 6,480,182; 6,498,114; 6,504,524; 6,506,438; 6,512,354; 6,515,649; 6,518,949; 6,521,489; 6,531,997; 6,535,197; 6,538,801; 6,545,291; 6,580,545; 6,639,578; 6,652,075; 6,657,772; 6,664,944; 6,680,725; 6,683,333; 6,704,133; 6,710,540; 6,721,083; 6,724,519; With 6,727,881; With U.S. Patent application bulletin No.2002/0019081; 2002/0021270; 2002/0053900; 2002/0060321; 2002/0063661; 2002/0063677; 2002/0090980; 2002/0106847; 2002/0113770; 2002/0130832; 2002/0131147; 2002/0145792; 2002/0171910; 2002/0180687; 2002/0180688; 2002/0185378; 2003/0011560; 2003/0011868; 2003/0020844; 2003/0025855; 2003/0034949; 2003/0038755; 2003/0053189; 2003/0102858; 2003/0132908; 2003/0137521; 2003/0137717; 2003/0151702; 2003/0189749; 2003/0214695; 2003/0214697; 2003/0222315; 2004/0008398; 2004/0012839; 2004/0014265; 2004/0027327; 2004/0075634; With 2004/0094422; With international patent application bulletin No.WO 99/67678; WO 00/05704; WO 00/38000; WO 00/38001; WO00/36560; WO 00/67110; WO 00/67327; WO 01/07961; WO 01/08241; WO 03/092077; WO 03/107315; WO 2004/017035; And WO2004/023202.

Many above-mentioned patents and application form are all recognized, wall around discrete micro-capsule in entrapped electrophoretic medium can replace by an external phase, thereby produce a what is called " polymer-dispersed electrophoretic display ", wherein this electrophoretic medium comprises the external phase of a plurality of discrete electrophoresis droplets and a polymeric material, and discrete electrophoresis droplet can be regarded as capsule or micro-capsule in such polymer-dispersed electrophoretic display, although the capsule membrane not being associated with each each droplet; For example, see above-mentioned 2002/0131147.Correspondingly, for the object of the application's book, such polymer-dispersed electrophoretic media is counted as the subclass of sealing electrophoretic medium.

The electrophoretic display device (EPD) of relevant type also claims " microcell electrophoretic display ".In a microcell electrophoretic display, charged particle and suspension liquid are not to be encapsulated in capsule, but remain on mounting medium, are generally in a plurality of holes that form in thin polymer film.For example, see international patent application bulletin No.WO 02/01281 and U.S. Patent application bulletin No.2002/0075556, both all yield SipixImaging, Inc. (company).

Although electrophoretic medium often opaque (because, for example, in many electrophoretic mediums, particle is block visible light transmission display device substantially), and operate under a reflective-mode, can make many electrophoretic display device (EPD)s operate under a what is called " shutter mode ", one of them show state is opaque substantially, and another is printing opacity.For example, see above-mentioned U.S. Patent No. 6,130,774 and 6,172,798, and U.S. Patent No. 5,872,552; 6,144,361; 6,271,823; 6,225,971; With 6,184,856.Be similar to electrophoretic display device (EPD), dielectrophoresis (Dielectrophoretic) display that still depends on electric field intensity variation may operate under a similar pattern; See U.S. Patent No. 4,418,346.The electro-optic displays of other type also may be operated under shutter mode.

The bistable of the electrophoretic display device (EPD) based on particle or multistable variation characteristic, and other electro-optic displays that present similar variation characteristic (for simplicity, such display can be called " pulsed drive display " hereinafter), with traditional liquid crystal (" LC) display forms distinct contrast.Twisted nematic liquid crystals does not play bistable or multistable effect, but plays voltage transmitter, so apply given electric field to the pixel of such display, just in this pixel, produces a specific gray scale, and has nothing to do with the gray scale that this pixel presented in the past.In addition, LC display only drives in one direction and (from opaque or " secretly " to transparent or " bright "), from brighter state-transition to darker state by reducing or eliminating this electric field, carries out.Finally, the gray scale of LC display picture element is insensitive to the polarity of electric field, and only responsive to its amplitude, really, due to technical reason, business sell LC display generally on interval time frequently upset drive the polarity of electric field.Otherwise bistable electro-optic displays, says like recently, plays pulse transmitter, makes the final state of a pixel not only depend on applied electric field and apply the time of this electric field, and depends on this electric field state of this pixel before that applies.

May first can expect, tackle a kind of like this Perfected process of pulsed drive electro-optic displays, to be so-called " general grayscale image stream ", its middle controller is an image of configuration write each time like this, makes each pixel from its initial gray scale, directly be converted to its last gray scale.But, on a pulsed drive display, write image and inevitably have some error.Some the such error running in practice comprises:

(a) original state dependence; For some electro-optical medium at least, make pixel be switched to a desired pulse of new optical states and not only depend on the optical states of electric current and requirement, and depend on the optical states before this pixel.

(b) residence time dependence; For some electro-optical medium at least, pixel is switched to a desired pulse of new optical states and depends on the time that this pixel has been spent at its different optical states.This dependent precise nature is still very not clear, but in general, this pixel is longer in its current optical states residence time, just requires more pulses.

(c) temperature dependency; Pixel is switched to a desired pulse of new optical states and seriously depends on temperature.

(d) humidity dependence; For the electro-optical medium of some type at least, pixel is switched to a desired pulse of new optical states and depends on humidity around.

(e) mechanical uniform; Pixel is switched to the impact that a desired pulse of new optical states can be subject to the mechanical outflow of population in this display, for example the influence of change of the thickness of electro-optical medium or relevant lamination cementing agent.The mechanical unevenness of other type may be owing to inevitably changing, manufacturing franchise between the different manufacture of medium batch and material change causes.

(f) voltage error; The actual pulse that is applied to pixel will inevitably be different from the pulse applying in theory slightly, because inevitable some error slightly on the voltage providing at driver.

As described at above-mentioned WO 03/044765 and PCT/US2004/10091, the problem that general grayscale image stream has " accumulation of error " phenomenon to cause, on the image of some type, it may produce the gray-scale deviation that general viewers can be discovered.This accumulation of error phenomenon is applicable to all types of error listed above.As above-mentioned 2003/0137521 described, it is possible compensating such error, but just in a limited precision.Thereby, general grayscale image stream will draw good result, must have to applied pulse very accurate control, and have been found that according to experience, at electro-optic displays, under current state of the art, it is infeasible that general grayscale image stream is sold on display business.

Nearly all electro-optical medium all has built-in reset (error limitation) mechanism, and in other words, their extreme (general black and white) optical states, plays " optics fence ".After a specific pulse is applied to a pixel of electro-optic displays, this pixel cannot become whiter (or more black).For example, at one, seal in electrophoretic display device (EPD), after applying a specific pulse, all electrophoresis particle is all forced to withstand each other or head on capsule wall, cannot move again, thereby, produce a restrictive optical states or optics fence.Because in such medium, there is electrophoretic particles size and CHARGE DISTRIBUTION, some particle was met this fence before other particles, set up " soft fence " phenomenon, therefore when the last optical states of a transformation approaches black and white extremity, reduced desired pulse degree of accuracy, thereby when the last optical states changing approaches this extreme black and white state, desired optics degree of accuracy reduces, and when the terminal changing approaches optical range middle of this pixel, desired optics degree of accuracy is increased sharply.

The dissimilar drive scheme that utilizes optics fence that known electric optical display unit is used.For example, Fig. 9 of above-mentioned WO 03/044765 and 10 and relevant description, has described a kind of " showing slides " drive scheme, is wherein writing before any new image, and whole display is driven to two optics fences.Such a drive scheme of showing slides produces gray scale accurately, but while being driven into optics fence, the flicker of display makes beholder vexed.Someone advises (seeing U.S. Patent No. 6,531,997), adopts a similar drive scheme, wherein in this new image, only has the pixel that its optical states need to change to be just driven to this optics fence.But, " limited showing slides " drive scheme of the type, if any, even more make beholder vexed, because the integral body (solid) of the common drive scheme of showing slides flicker is depended on that the flicker of image replaces, wherein write new image with forth screen on the feature of old image and new image with contrary chromatic flicker.

Obviously, as single, grayscale image stream drive scheme can not look to utilizing this optics fence to avoid gamma error, because in such a drive scheme, any given pixel all may experience the infinitely-great change of number in gray scale, and does not touch any optics fence.

In one aspect, the present invention makes every effort to provide and on electro-optic displays, reaches the method that gray-scale Control is used, and it reaches and is similar to the stable of gray scale that the drive scheme of showing slides reaches, but the flicker problem making one nervous of the drive scheme of not showing slides.The preferred method of the present invention can give one of beholder be similar to single as the visual experience that provides of grayscale image stream drive scheme.

In other respects, the present invention makes every effort to provide the method that reaches the meticulous control of gray scale in the display being driven by width modulation.

When driving one there is bistable electro-optical medium to write the Active Matrix Display of grayscale image thereon, apply an accurate pulsed quantity preferably can to each pixel, to reach the accurate control of shown gray scale.Driving method used can rely on be applied to the voltage in each pixel modulation and/or execute the modulation of alive " width " (duration).Because the power supply cost that voltage modulated driver is relevant with them is relatively high, and width modulation is commercially attractive.But in utilizing scanning process of Active Matrix Display of such width modulation, traditional driving circuit only allows to apply a single voltage to any given pixel in any one scanning process of this matrix.Thereby the width modulation of Active Matrix Display drives to be undertaken by this matrix of Multiple-Scan, at zero degree, the driving voltage that several times or all applies in scanning process, depend on the change that specific pixel requires in gray scale.Scanning each time can be regarded a frame drive waveforms as, and complete addressing pulse is the superframe being formed by a plurality of frames in succession.Should be noted that, in each scanning process for the row address time, although driving voltage is only applied on any specific pixel electrode, but in the time course between selection in succession of same a line, this driving voltage maintains on this pixel electrode, only can slow-decay, make in succession driven between selection in same a line of this pixel.

Just as already noted, in each frame process each provisional capital of this matrix need indivedual selected, make for high resolution display (for example, 800 * 600 pixel display) in practice frame rate cannot surpass approximately 50 to 100Hz; Thereby each frame generally continues 10 to 20ms.The meticulous control example that the frame of this length causes being difficult to the electro-optical medium of many quick switchings is carried out to gray scale as, some is sealed electrophoretic medium and substantially between their extreme optical state (once changing about 30L* unit), in about 100ms, completes once and to switch, and the gray scale corresponding to Yue6L* unit moves for the frame of a 20ms of such medium.Such a movement is too large for the accurate control of gray scale; Human eye Dui Yue 1L* unit gray difference is responsive, and only controls this pulse being equivalent on the scale of 6L* unit, probably causes visible distortion, " ghost image " causing such as the original state dependence of this electro-optical medium.More particularly, may experience ghost image, because as discussed, in some above-mentioned patents and application form, the change of the pulse gray scale applying is not linear, and any specific gray scale change needed overall pulse may be along with applying time of pulse and middle gray scale and changing.For example, in simple 4 gray scales (2) display with gray scale 0 (black), 1 (dark gray), 2 (bright gray scale) and 3 (in vain), by a simple width modulation drive scheme, driven, nominally nominally these non-linear actual gray scales that reach after 0-2 transformation that may cause are different from the gray scale reaching after 1-2 transformation, produce extremely undesirable vision distortion simultaneously.The invention provides the method that reaches the meticulous control use of gray scale in the display being driven by width modulation, thereby, the problems referred to above avoided.

Correspondingly, in one aspect, the invention provides a kind of driving and there is at least one pixel, can reach any one at least four different gray scales, comprise the electro-optic displays of two extreme optical state.The method comprises:

On this display, show the first image; With

Rewrite this display, to show the second image thereon,

Wherein, in rewriteeing the process of this display, number of transitions surpasses any pixel of a predetermined numerical value, this predetermined value is at least one, and do not touch extreme optical state, at this pixel driver before its last optical states in this second image, be driven at least one extreme optical state.

For simplicity, the method can be called " limited transformation method " of the present invention.

In a form of this limited transformation method, the rewriting of display is carried out like this, once pixel is driven to contrary extreme optical state from an extreme optical state by the pulse of a polarity, before reaching contrary extreme optical state, this pixel does not just receive the pulse of opposite polarity.

In addition, in this limited transformation method, predetermined value (predetermined number of transitions) is not more than N/2, and wherein N is the gray scale sum that a pixel can show.This limited transformation method can utilize three grades of drivers to carry out, that is the rewriting of display can be by applying voltage-V to this pixel or each pixel, 0 and+in V any one or carry out above.In addition, this limited transformation method can be DC balance, that is the rewriting of this display can be carried out like this, any transformation series that makes to stand for a pixel, and impressed voltage is bounded to the integration of time.

Limited transformation method of the present invention, the rewriting of this display can be carried out like this, makes to be applied in a transition process the initial and last gray scale that this transformation is only depended in a pulse in pixel.As another program, the method can be designed to consider other state of this display, as described in greater detail below.In this limited transformation method, in a preferred form, the transformation of at least one standing at least one pixel from gray scale R2 to gray scale R1, is applied to the pulse of the following form of this pixel one sequence:

TM(R1，R2)IP(R1)-IP(R2)TH(R1，R2)

Wherein " IP (Rx) " representative is from a correlation values all for each gray scale with the pulse electromotive force matrix of a numerical value, and TM (R1, R2) representative is from a correlation values for each R1/R2 combination with the transition matrix of a numerical value.(for simplicity, the pulse train of the type can be abbreviated as " x/ Δ IP/x " sequence hereinafter.) such-x/ Δ IP/x sequence can be applied to the wherein initial whole transformations different with last gray scale.In addition, in such-x/IP/x sequence, it is over half that last " x " section can take maximum update time.This TM (R1, R2) or x value can be selected like this, make the symbol of each numerical value only depend on R1; Specifically, for one or more bright gray scales, these values can be to be just chosen as, and for one or more dark gray, are chosen as negatively, make gray scale beyond two extreme optical state from approaching the direction convergence of extreme optical state.

Above-mentioned-x/ Δ IP/x sequence can comprise additional pulse.Specifically, such sequence can comprise the pulse of additional a pair of [+y] [y] form, and wherein y is the pulse value of negative or positive, should [+y] and [y] pulse insertion-x/ Δ IP/x sequence.This sequence can also comprise the pulse of second a pair of [+z] [z] form of adding, and wherein z is a pulse value that is different from y, and can be negative or positive, should [+z] and [z] pulse insertion-x/ Δ IP/x sequence.Be somebody's turn to do-x/ Δ IP/x sequence can also comprise that a no-voltage puts on the cycle of this pixel.Should " no-voltage " cycle can appear at this-two elements of x/ Δ IP/x sequence between, or in its single element.Should-x/ Δ IP/x sequence can comprise two or more " the no-voltage cycle.

Utilize above-mentioned-during x/ Δ IP/x sequence, this display can comprise a plurality of pixels that are divided into a plurality of groups, and this transformation can be undertaken by following steps: (a) select each in a plurality of phase pixel groups, and to each pixel in this selected group, be not that to apply driving voltage be just non-driving voltage, the scanning of all pixel groups all completes in the first frame period; (b) in the second frame period process, repeat the scanning of this pixel groups; (c) in a pause period process between this first and second frame period, interrupt the scanning of this pixel groups, this pause period is long unlike the one the second the second frame periods.

In this limited transformation method, the rewriting of this display can be carried out like this, makes always by one of same polarity last pulse, to be undertaken to the transformation of a given gray scale.Specifically, the gray scale beyond two extreme optical state can be from the direction convergence of more approaching extreme optical state.

The present invention also provides a method that drives electro-optic displays, and this electro-optic displays has a plurality of pixels that are divided into many groups.The method comprises:

(a) select each in a plurality of pixel groups in succession, and to the every pixel in selected group, be not to apply driving voltage, apply exactly non-driving voltage, the scanning of all pixel groups all completes in the first frame period;

(b) in the second frame period process, repeat the scanning of this pixel groups; With

(c) in a pause period process between the first and second frame periods, interrupt the scanning of this pixel groups, this pause period is no longer than the first or second frame period.

For simplicity, the method can be called " interrupt scanning " of the present invention method hereinafter.

In such interrupt scanning method, general the first and second frame periods are equal in length.The length of this pause period can be the part (sub-multiple) of a length in the first and second frame periods.This interrupt scanning method can comprise a plurality of pause period; Thereby the method can be included in scanning element group at least first, second, and third frame period process, and between the frame period in succession, at least the first and second pause period processes, interrupt the scanning of pixel groups.This first, second, and third frame period can be substantially equal in length, and the total length of each pause period equals a frame period or a frame period deducts a pause period.Generally, in this interrupt scanning method, this pixel is arranged in a matrix with a plurality of row and a plurality of row, and wherein each pixel is defined by the intersection point of a given row and given row, and each group pixel comprises a row or column of this matrix.This interrupt scanning method is DC balance preferably, that is the scanning of this display is preferably carried out like this, makes the transformation of any series of standing for a pixel, and impressed voltage is bounded to the integration of time.

In other respects, the invention provides a method, in order to the electro-optic displays that drives to have a plurality of pixels, this pixel is with applying to each pixel the pulse-width modulation waveform driving of a plurality of different pulses.The method comprises:

(a) storage indication applies a given pulse to a pixel and whether produces one higher or lower than data that require the gray scale of gray scale;

(b) detect when two neighbors all require same gray scale; With

(c) regulate the pulse that is applied to these two pixels, make a pixel lower than the gray scale requiring, and while one other pixel is higher than requiring gray scale.

For simplicity, the method can be called " balance gray level method " of the present invention hereinafter.

In the method, pixel can be divided into two groups, makes each pixel have the neighbours of at least one contrary group, and these two groups are used to different drive schemes.

As above-described, each method of the present invention can be applied to the electro-optical medium of any one the above-mentioned type.Thereby, method of the present invention can with comprise electrochromism or rotate together with the electro-optic displays of double-colored film electro-optical medium, entrapped electrophoretic medium or microcell electrophoretic medium and use.Also can use the electro-optical medium of other type.

Figure 1A and 1B illustrate two parts of this limited transformation drive scheme of the present invention;

Fig. 2 illustrate can be used for the method for the present invention preferred-x/ Δ IP/x sequence;

Fig. 3 schematically illustrates the waveform shown in Fig. 2 and how can revise, to comprise a pair of additional driving pulse;

Fig. 4 illustrates with the waveform that the illustrational mode of Fig. 3 changes Fig. 2 and produces a waveform;

Fig. 5 illustrates and with Fig. 3, illustrates the waveform that mode changes Fig. 2 and produce the second waveform;

Fig. 6 is schematically routine illustrates that how can further revise the waveform shown in Fig. 5 comprises a pair of additional driving pulse;

Fig. 7 illustrates and with Fig. 6, illustrates the waveform that mode changes Fig. 5 and produce a waveform;

Fig. 8-10 illustrate three modifications of the waveform shown in Fig. 2, to comprise a no-voltage cycle;

From the above, obviously the present invention provides several different improvement methods on electro-optic displays driving method.In the following description, various improvement provided by the invention generally will be described separately, although imaging technique technical professional will be appreciated that, single display can utilize above these main aspects in practice; For example, use the display of limited transformation method of the present invention can also utilize interrupt scanning method.In addition, because these improvement provided by the present invention can be applied the method for above-mentioned WO 03/044765 and the described various driving electro-optic displays of PCT/US2004/10091, below describe supposition (reader) is familiar with to the basic driver method shown in WO03/044765 Fig. 1-10 and relevant description.Specifically, Fig. 9 of the application and 10 describes the so-called uncompensated n-prepulse with three root segments (n-PP SS) waveform of showing slides.First, this pixel is erased to a uniform optical states, be not generally white be exactly black.Then, between two optical states, generally white and black again, come and go and drive this pixel.Finally, make this pixel in a new optical states, it can be one of several gray states.This last (or writing) pulse is called addressing pulse, and other pulses (this first (or wiping) pulse and middle (or blank) pulse) are collectively called prepulse.

A major defect of the type waveform is that it has the optical flicker of large amplitude between several images.This can be by making renewal sequence move the superframe time of half use of pixel and improve with high resolving power staggered pixels, as what discuss with reference to its figure 9 and 10 at WO 0303044765.Possible pattern comprises in every line, every row or checkerboard pattern.Note, this does not also mean that and utilizes contrary polarity, that is " from black " and " from white ", because this can cause non-matching gray scale in adjacent pixel.But can by make NEW BEGINNING more postpone one " superframe " (one group of several frame that are equivalent to the maximum length that Hei-Bai upgrades) complete pixel half (that is, first group of pixel completes erasing pulse, when this first group of pixel starts the first space pulse, this second group of pixel starts this erasing pulse).This will require to add the superframe of a total update time, to allow them synchronous.

Limited transformation method of the present invention

Above-mentioned flicker problem for fear of the drive scheme shown in Fig. 9 and 10 of WO 03/044765, and the problem of the general grayscale image stream of discussing before simultaneously also avoiding, preferably according to this limited transformation method configuration driven scheme of the present invention, make any given pixel, by before an extreme optical state (black or white), can only experience the gray scale transformation of a predetermined maximum number (at least one).A transformation of leaving this extreme optical state, since an accurate known optical states, is in fact cancelled any error of accumulation in the past.In WO03/044765, carried out having discussed the different technology of the optical effect that makes like this pixel by extreme optical state (such as the flicker of display) being reduced to minimum use.

In the reset process process of above-mentioned such drive scheme, appear at the black and white flicker on display, to user, that yes is visible, and is tedious for many users.In order to reduce the visual effect of such reset process, way is easily, and the pixel of this display is divided into two (or more) groups, and this different group is applied to dissimilar reset pulse.More particularly, replace black and drive in vain the reset pulse of any given pixel if must use, way is easily, pixel is divided into at least Liang Ge group, and configuration driven scheme, a pixel groups is driven into white, and one other pixel group is driven into black simultaneously.Suppose the careful space distribution of selecting Gai Liangge group, and these pixels are fully little, user will experience this reset process as experience gray scale interval time (perhaps occurring some slight snowflake) on this display, and such a gray scale interval time is generally black and white flicker is generally more less horrible than a series of.

For example, with a kind of form of such one " two groups of resets " step, the pixel that odd number lists can be included into " odd number " group, and the pixel that even number lists is included into second " even number " group.Then, odd pixel can be utilized the drive scheme shown in Fig. 9, in erase step process, these pixel drivers to black state, and simultaneously in erase step process, the modification that can utilize this drive scheme these pixel drivers to white state.Then, two pixel groups all can stand even number reset pulse in reset process process, make the reset pulse that Liang Ge group uses substantially differ 180 degree phase places, and this display all occurs gray scale in whole reset process.Finally, in the process of writing at the second image of this step, odd pixel is from the black final state that is driven into them, and while even pixel is from being driven in vain their final state.For guarantee each pixel reset in the same way for a long time (thereby, this reset mode is not introduced any distortion on this display), this controller is preferably between image in succession and switches drive scheme, make to write display at a series of new images, each pixel alternately writes its final state from black and white state.

Obviously, can use similar scheme, wherein the pixel of the row of odd-numbered forms first group, and the pixel of the row of even-numbered forms second group.In another similar drive scheme, first group of pixel that comprises odd column and odd-numbered line and even column and even number line, and second group comprise odd column and even number line and even column and odd-numbered line, makes these two groups to be arranged to chessboard mode.

Replace (or except) pixel be divided into Liang Ge group, and at one group to differ 180 degree phase configuration reset pulses with another group, these pixels can be divided into the group that uses different reset process, different in the number of pulse and frequency.For example, one group can be by six pulsed reset sequences, and the second group of similar sequence that can use the pulse with 12 these frequencies of twice simultaneously.In a meticulousr scheme, pixel can be divided into four groups, utilizes six pulse schemes for first and second groups, and phase place differs 180 degree each other, and simultaneously third and fourth group use 12 pulse schemes, but phase place differs 180 degree each other.

According to limited transformation method of the present invention, another method of dwindling flicker problem can utilize such drive scheme to carry out, and allows any one given be assumed to non-zero but touching before an optics fence gray states Limited Number in succession.In such a drive scheme, when rewriting display so that while showing a new image thereon, any one has stood the pixel that number of transitions surpasses predetermined value and do not touch an extreme optical state, in this pixel, be driven to before its last optical states, be driven at least one extreme optical state.In a preferred form of such a drive scheme, be driven to the pixel of an extreme optical state after this transformation, be driven to the more approaching extreme optical state of its gray scale, certainly suppose that this requires optical states is not one of extreme optical state.In addition, such a, utilize in a foregoing preferred form of drive scheme of looking into table, the transformation maximum times that pixel is allowed to stand in the situation that not touching optics fence (extreme optical state) arranges to such an extent that equal the number of the previous optical states considered at this transition matrix; Such method does not require extra steering logic or storer.

Touch the rewriting time used that driving method that restriction before an optics fence changes maximum times needn't increase this display significantly.For example, investigate four gray shade scales (2 s') display, one of them is from vain to black transformation or will, with 200 milliseconds, make general grayscale image flow drive scheme and will rewrite display completely with this time on the contrary.On such a display, changing the unique situation that need to revise is when pixel repeats to switch between two central gray scales.If the number of transitions that such a pixel is switched between two central gray scales surpasses this predetermined number, this limited transformation method of the present invention requires this next one switching to be undertaken by optics fence (extreme optical state).Have been found that under such a case, forwarding optics fence to needs 70 milliseconds, and forwarding this gray scale to and will, with approximately 13 milliseconds, make total fringe time approximately only have 200 milliseconds afterwards simultaneously.Thereby this limited transformation method is compared with general grayscale image stream, needn't extend fringe time completely.

Now with reference to Figure 1A and 1B, a limited transformation driving method that reduces the tedious impact of reset process is described.In this scheme, these pixels are divided into Liang Ge group again, and first (even number) group is according to the drive scheme shown in Figure 1A, and this second (odd number) group is according to the drive scheme shown in Figure 1B.In addition, in this scheme, black and white middle whole gray scales are divided into the adjacent dark gray of first group of contiguous black grade, and the adjacent bright gray shade scale of second group of contiguous white grade is divided the same with two pixel groups like this.Best, but be not important, the same number of gray shade scale in Zhe Liangge group, had; If there is odd number gray shade scale, central grade can give arbitrarily any one group.For the ease of illustrating, Figure 1A and 1B display application are in the drive scheme of eight grade gray-scale monitors, and these grades are denoted as 0 (black) to 7 (in vain); Gray shade scale 1,2 and 3 is dark gray levels, and gray shade scale 4,5 and 6 is bright gray shade scales.

In Figure 1A and 1B graphics driver scheme, gray scale to the transformation of gray scale according to following rule treatments:

(a) in the first even pixel group, in turning to the process of dark gray levels, the final pulse applying always tend to white pulse (that is, its polarity trends towards driving this pixel from its its white state of black state shift), yet in the transformation that turns to bright gray scale, the final pulse applying always tends to black pulse;

(b) in the second odd pixel group, in turning to dark gray levels process, the final pulse applying always tends to black pulse, and in the transformation that turns to bright gray shade scale, the final pulse applying always tends to white pulse;

(c) in all cases, after having reached white state, the black pulse of a trend is only with after the white pulse of trend, and after having reached black state, and the white pulse of trend is only with after the black pulse of trend; With

(d) even pixel cannot be with the black pulse of single trend from a dark gray levels to black driving, and odd pixel also cannot be with the white pulse of single trend from a bright gray shade scale to white driving.

(obviously, in all cases, white state is only to utilize the white pulse of last trend to reach, and black state is only to utilize the pulse that last trend is black to reach.)

Applying these rules just makes each gray scale be utilized maximum three pulses in succession to carry out to the transformation of gray scale.For example, Figure 1A represents that even pixel stands the transformation from black (grade 0) to gray shade scale 1.This is to utilize to be denoted as 1102 the white pulse (certainly representing with the positive gradient in Figure 1A) of single trend and to reach.Then, this pixel driver is arrived to gray shade scale 3.Because gray shade scale 3 is dark gray levels, according to rule (a), must reach by the white pulse of a trend, thereby grade 1/ grade 3 changes pulse 1104 that can be white by single trend to be processed, it has the pulse that is different from pulse 1102.

This pixel is driven to gray shade scale 6 now.Because this is a bright gray shade scale, by rule (a), it must reach by the black pulse of a trend.Correspondingly, application rule (a) and (c) calling hierarchy 3/ class 6 change and are undertaken by two pulse trains, in other words, the white pulse 1106 of the first trend, its drives pixel bleach (grade 7), one second black pulse 1108 of trend of heel, it drives this pixel from grade 7 to calling hierarchy 6.

Then, this pixel driver is arrived to gray shade scale 4.Because this is a bright gray shade scale, by a transformation parameter used that is just similar to grade 1/ grade 3 of early stage discussion, to be undertaken, the transformation of this grade 6/ class 4 is undertaken by a black pulse 1110 of single trend.Next transformation is to grade 3.Because this is a dark gray levels, by a transformation parameter used that is just similar to grade 3/ class 6, undertaken, the transformation of this grade 4/ grade 3 is processed by two pulse trains, in other words, the black pulse 1112 of the first trend, it is this pixel driver to black (grade 0), and one second of heel tends to white pulse 1114, and it is driven into this pixel the grade 3 of requirement from grade 0.

Last transformation shown in Figure 1A is from grade 3 to grade 1.Because grade 1 is dark gray levels, according to rule (a), it must carry out convergence by tending to white pulse.Correspondingly, application rule (a) and (c), the transformation of this grade 3/ grade 1 must be processed by three pulse trains, comprise drive this pixel to first trend of white (grade 0) white 1116, drive this pixel from the 3rd of black grade 1 state to requiring, to tend to white pulse 1120 to second trend of black (grade 7) black 1118 and this pixel of driving.

Figure 1B represents that odd pixel carries out same 0-1-3-6-4-3-1 gray states sequence, as the even pixel of Figure 1A.But, will find out, the pulse train of using is very different.1, one dark gray levels of rule (b) calling hierarchy, by tending to black pulse convergence.Thereby it is that one of heel drives the trend black pulse 1124 of this pixel from grade 7 to the grade 1 requiring by driving this pixel to carry out to the white pulse 1122 of the first trend of white (grade 7) that this 0-1 changes.This 1-3 changes and requires three pulse trains, and one drives this pixel to drive this pixel to drive the three trend black pulse 1130 of this pixel from grade 7 to calling hierarchy 3 to the white pulse 1128 of second trend of white (grade 7) and one to the black pulse of first trend of black (grade 0) 1126, one.It is to class 6 that this next one changes, this is a bright gray shade scale, according to rule (b), by a pulse convergence that trend is white, this grade 3/ class 6 changes and is undertaken by two pulse trains, comprises that one drives this pixel to the black pulse of the trend of black (grade 0) 1132, this pixel of driving to the white pulse 1134 of the trend of the class 6 requiring.This grade 6/ class 4 changes and is undertaken by three pulse trains, in other words, one drives this pixel to tend to white pulse 1136, one to white (grade 7) to drive this pixel to the black pulse 1138 of the trend of black (grade 0) and this pixel of driving to the white pulse 1140 of the trend of the class 4 requiring.The transformation of this grade 4/ grade 3 is undertaken by two pulse trains, comprises that one drives this pixel to the white pulse 1142 of trend of white (grade 7), and one of heel drives this pixel to the black pulse 1144 of the trend of the grade 3 requiring.Finally, the transformation of grade 3/ grade 1 is to be undertaken by a black pulse 1146 of single trend.

From Figure 1A and 1B, will find out, this drive scheme guarantees that each pixel follows a saw tooth pattern, wherein this pixel from black forward to white, and do not change direction (obviously, although this pixel can rest on one section short of any middle gray grade or long during), and after this black from forwarding in vain, and do not change direction.Thereby, rule above (c) and (d) can be to replace by following single rule (e):

(e) once pixel by the pulse of a polarity from an extreme optical state (that is, white or black) to contrary extreme optical state, drive, this pixel just cannot receive the pulse of opposite polarity, until it has reached above-mentioned contrary extreme optical state.

Thereby this drive scheme is one " fence is stablized gray scale " i.e. " RSGS " drive scheme.Such a RSGS drive scheme is special circumstances of limited transformation drive scheme, it guarantees that a pixel is at most only to stand to equal the transformation of the number of transitions (or more precisely (N-1)/2) of N/2, wherein N is the gray shade scale sum that can show, and does not require that transformation occurs by an optics fence.Such a drive scheme is avoided the slight errors (for example, the inevitable less fluctuation of voltage being applied by driver causes) in each transformation, and infinite accumulation is until the serious distortion of grayscale image appears at observer in front.In addition, this drive scheme designs like this, makes even number and odd pixel always from given middle gray of contrary direction convergence, that is in one case, the last pulse of this sequence is the white pulse of trend, and in another case, be the pulse that trend is black.If most of district inclusion of this display substantially even number and the odd pixel of equal number writes a single gray shade scale, " reverse direction " feature reduces to minimum the flicker in this region.

Owing to being similar to, relate to above those reasons that other drive scheme is discussed, when realizing Figure 1A and 1B sawtooth drive scheme, pixel is divided into two discrete groups, should carefully note the layout of pixel in even number and odd number group.This configuration will guarantee that any continuous display area substantially all will comprise the odd and even number pixel of equal number substantially best, and the full-size of same group of continuous pixels piece is fully little, and general viewers is difficult for discovering.Just as already discussed, the configuration of the checkerboard pattern of two pixel groups meets these requirements.In addition, random screen technology can be used for configuring the pixel of Gai Liangge group.

But in this sawtooth drive scheme, the use of checkerboard pattern often increases the energy consumption of display.Any given row at such a pattern, neighbor will belong to contrary group, and whole pixels all stand the same gray scale transformation sizable continuum of (one is not abnormal situation) size therein, adjacent pixel will often require opposite polarity pulse in any given time.Any listing to the pixel in succession occurring, apply opposite polarity pulse, while often writing a newline, require row electrode (source electrode) electric discharge of this display and charging again.In the driving of Active Matrix Display, to row electrode discharge and again charging be the principal element of a display energy consumption, this is well-known to technician.Thereby checkerboard is arranged the energy consumption that often increases display.

Energy consumption and avoid between the hope in the large region of continuous pixels on the same group one rationally compromise be to give rectangle to each group pixel, pixel wherein is all positioned at same row, but stretches several pixels along these row.Adopt such layout, when rewriting has the region of same gray shade scale, only have when moving on to next rectangle from a rectangle, just must be to row electrode discharge and charging again.Rectangle is 1 * 4 pixel preferably, and configures to such an extent that the rectangle in adjacent column does not finish in same a line, that is the rectangle in adjacent column has difference " phase place ".Rectangle in each row is given each phase place and can be carried out randomly or in a looping fashion.

An advantage of the sawtooth drive scheme shown in Figure 1A and 1B is that the region of any monochrome of this image is all upgraded with single pulse simply, is not black in white, is exactly white to black, as the overall part of upgrading of this display.Rewriteeing such monochromatic areas maximum duration used is that gray scale changes to gray scale half that desired region rewrites required maximum duration, and this feature can advantageously be used for promptly upgrading image feature, such as user's character input, pull-down menu etc.Controller can check whether the requirement of an image requires gray scale to the transformation of gray scale; If do not require, need the imagery zone rewriteeing can utilize this more new model rewriting of monochrome rapidly.Thereby user can upgrade character, pull-down menu and other user interaction features of the input of this display fast, seamlessly overlaps in the slower renewal of general grayscale image.

The number of transitions that each pixel that limited transformation drive scheme not necessarily requires usage counter to measure display stands, and do not hinder the use of drive scheme (such as the circulation RSGS drive scheme of having described with reference to Figure 1A and 1B), even if the latter not yet reaches predetermined number of transitions, just require some transformation to occur by optics fence, suppose that the algorithm that definite mode of carrying out changing is used does not allow any pixel to stand more than the transformation of pre-determined number not touch optics fence.In addition, will appreciate that, the number of transitions standing in the next given pixel of the situation of not touching optics fence needn't all be carried out by every rewriting primary display image, especially at display with interval time frequently more under news.For example, this inspection can only just be carried out when alternately upgrading, and supposes that whole pixels are not to surpass predetermined number of transitions, and next exactly renewal may surpass this number after being driven to optics fence.

Another preferred limited transformation method of the present invention now will be described, although just in illustrational mode.This preferred method is used for moving four gray shade scales (2) Active Matrix Display, and its uses the transition matrix of the state formerly of only considering an initial and last gray shade scale (indicating respectively " R2 " and " R1 ") for the transformation that will carry out and not considering to add.This display controller is three grade width modulation (PWM) controllers, can be with respect to the public front termination electrode that remains on 0 (level) to apply-V of each pixel electrode, 0 or+V.

This display controller comprises two RAM image buffer zones.Image current on this display is stored in a buffer zone (" A ").Generally, this controller is in park mode, save data make this display driver keep inactive in RAM.The bistable state of this electro-optical medium is preserved image same on this display.When receiving that image is more during newer command, this controller packs this second buffer zone (" B ") into new image.Then, according to current, the initial state R2 (from buffer zone " A ") of the last state R1 of this pixel request (from buffer zone " B ") and each pixel, for each pixel of this display, this controller is all checked (in flash memory) multiframe drive waveforms.

Organization of Data in this flash file becomes a three-dimensional magnitude of voltage V (R1, R2, frame) array, wherein just as has been indicated R1 and R2 each be the integer of from 1 to 4 (corresponding with four available gray shade scales), and " frame " is the number of frame, that is, the number of associated frame in the superframe for each transformation.Generally, superframe can be 1 second long, simultaneously each frame takies 20ms, makes the number of frame can be in 1 to 50 scope.Thereby this array has 4 * 4 * 50=800 item.Because each in array all must can represent this magnitude of voltage-V, 0 and+any one of V, so generally store each magnitude of voltage (array of values) with two.

Very obvious, because each 800 array item can have these three possible magnitudes of voltage any one, so have a possible array (waveform) that number is huge, number is too large to such an extent as to cannot exhaustive search.In theory, have 3 ⁸⁰⁰or approximately 5 * 10 ³⁸¹individual possible array; Because approximately there are 1078 atoms in universe, people's mean lifetime has 10 ⁹second, virtual rating has 200 orders of magnitude at least, cannot exhaustive search.Fortunately, the existing knowledge about electro-optic displays variation characteristic, the especially needs of DC balance on it, to possible waveform to force to add additional constraint and allow search best or that approach best waveform is limited in feasible scope.

As what discuss at above-mentioned U.S. Patent No. 6,504,524 and 6,531,997 and above-mentioned WO03/044765, known major part, if not all, electro-optical medium requires DC current (DC) balanced waveform, otherwise just may occur injurious effects.When using unbalance DC waveform, such impact can comprise the long term drift of the damage of electrode and the gray states in the scope of several L* units (cycle was over one hour).Therefore, it seems whenever making every effort to use the scheme of DC balance drive waveform all should think thrice before acting.

From the above, may first can expect, such DC balancing method may be to be beyond one's reach because any specific gray scale to the needed pulse by this pixel of the transformation of gray scale so that electric current is constant substantially.But, this says like just recently and is only really, and have been found that by experience, at least in the situation that the electrophoretic medium based on particle (other electro-optical medium too), the effect of pulse that applies 50 milliseconds, five intervals to a pixel is not identical with 250 milliseconds of pulses that apply same voltage.Correspondingly, have some dirigibility flowing through a pixel and reach on the electric current of given transformation, and this dirigibility can be used for assisting to reach DC balance.For example, this looks into a plurality of pulses that can store given transformation use with table, numerical value together with each total current providing in the middle of these pulses, and this controller can maintain a register to each pixel, configuration for example deposits in, from being applied to the algebraic sum of the pulse this pixel some the previous time (, maintaining black state from this pixel).When specific pixel will be driven into black state from white or gray states, this controller can check the register being associated with this pixel, determine from before the desired electric current of overall sequence DC balance of transformation of the upcoming black state of black state shift, and select one of pulse from white/ash to black transformation use of a plurality of storages, it is not exactly relevant register to be reduced to zero, be at least remainder reduce to as far as possible little (in this case, this relevant register will maintain the value of this remainder, and be added to institute in transition process subsequently and applied on electric current).Obviously, repeat to apply the long-term DC balance accurately that this processing can reach each pixel.

Must investigate the explication waveform of DC balance in waveform.In order to determine whether DC balance of a waveform, generally use the resistive model of an electro-optical medium.Such a model is not entirely accurate, but can be that supposition is enough accurately to current object.Utilize such a model, the characteristic of definition DC balanced waveform is that applied voltage is bounded to the integration of time (pulse applying).Note that it is " bounded " and for " zero " that this definition will be quadratured.In order to illustrate this point, investigate a monochromatic addressing waveforms, it uses a 300ms *-15V rect.p. to drive from white to black transformation, and drives the transformation from black to white with 300ms * 15V rect.p..This waveform is obviously DC balance, but the integration of each voltage constantly applying is non-vanishing; This integration 0 and ± change between 4.5V-sec.But with regard to this integration bounded, this waveform is DC balance; For example, this integration never reaches 9 or 18V-sec.

In order to further consider DC balanced waveform, it is wise that term is done to some definition.Term " pulse " was defined as in specific interval time, and the voltage applying in a general addressing pulse or pulse elementary process is to definite integration of time (M-sec of unit).Term " pulse electromotive force " by be used to refer to from any starting point (starting points of general considered a series of transformations), be applied to all pulses this display and.In this starting point, pulse electromotive force is at random set to zero, and this pulse electromotive force rises and declines when applying pulse.

Utilize these terms, the definition of DC balance is, when and and if only if this pulse electromotive force while being bounded, this waveform is only DC balance.The pulse electromotive force that has had bounded, just means that people must be able to say each that Pulse Electric will definitely be in the possible situation of finite population.

For the time independently controller (that is, the waveform of its pulse is affected by the initial and last state of considered transformation only, and do not stopped the controller that other factor of number of times, temperature affects, all R1/R2 controllers described above), in order to show that a waveform is DC balance, must be able to prove that Pulse Electric will definitely be bounded after each of the optical states sequence of any endless changes.An adequate condition of such proof is the function that this pulse electromotive force can be expressed as the original state of fixed number, and provide the Working concepts of a DC balance for electro-optic displays controller, that is this pulse electromotive force can be expressed as the function of the previous and current optical states of finite population.The starting point that the end that the pulse electromotive force that note that any pixel of this display upgrades from an image is upgraded to another image is constant, because do not apply during this period voltage.

For each combination of (limited) number original state, this controller applies fixing pulse (the definite pulse of data in above-mentioned flash memory), and these fixing pulses can be enumerated.In order to enumerate them, must be by the number (that is for R1/R2 controller, the number of enumerating original state for this need to define for the combination of two whole original states in rear end) of the original state that at least uses in this controller.

Pulse electromotive force while finishing in order to define this renewal, has known the fixing pulse applying in this pulse process, and the whole states in must enumerating for this are determined the pulse electromotive force at the starting point place of this renewal.This means, the clean pulse that waveform applies must be the function of an original state fewer than the required number of pulse electromotive force that defines uniquely end.In order being changed into, to determine the optimum waveform problem identificatioin that will be applied by a controller, to this means, the Pulse Electric of a waveform certainly will must be the function of the original state that its number is fewer than the number of determining the state that this waveform is used.For example, if a controller has by three state R1, the definite pulse data of R2 and R3 (wherein R3 be immediately the gray shade scale before the initial gray shade scale of considered transformation), each of R1 and R2 is in conjunction with allowing this electro-optical medium be held in the irrelevant pulse electromotive force of same and R3.

In other words, this controller must " know " that this electro-optical medium works as the pulse electromotive force of considered transformation while starting, so it can apply the suitable value that correct pulse produces pulse electromotive force after this transformation.If this pulse electromotive force is allowed to according to whole R1 in above-mentioned example, R2 and R3 and change,, in the next one changes, will have no idea to allow this controller " know " the pulse electromotive force that this starts, because the R3 information of using in the past may be thrown aside.

Just as has been indicated, limited transformation method of the present invention preferably utilize a R1/R2 controller (that is, the pulse wherein applying in any transition process only depend on this transformation initial with last gray shade scale) carry out, and from above discussion, in such a controller, this Pulse Electric certainly will must be defined as the function that only has R1 uniquely.

When determining optimum waveform, more complicated situation is to produce from such one phenomenon that can be called " pulse hysteresis ".Except the rare situation of extremely overdriving under optics fence, the electro-optical medium being driven by the voltage of a polarity always becomes more black, and always becomes whiter by the electro-optical medium that the voltage of opposite polarity drives.But, for some electro-optical medium, specifically for some, seal electro-optical medium, optical states is along with the variation of pulse demonstrates hysteresis; Driven during further to leucismus at this medium, the optical change of the unit pulse that each applies reduces, if but the polarity of the voltage applying overturns suddenly, make this display be driven to reverse direction and change, the optical change of every unit pulse increases suddenly.In other words, the optical change of every unit pulse not only depends on current optical states consumingly, and depends on the change direction of optical states.

This pulse lags behind and produces intrinsic " restoring force ", trend towards making this electro-optical medium to middle gray change of rank, defeat with unipolar pulse (the same with general grayscale image flow process) and drive the effort of this medium from a state to another state, and still maintain DC balance simultaneously.When applying pulse, this medium swims in three-dimensional R1/R2/ pulse hysteresis surface, until it reaches a balance.For each pulse length, this balance is fixed, and is generally located at the center in this optical range.For example, by experience, have been found that driving one seals four gray shade scale electro-optical mediums from black to 100ms of dark gray requirement *-15V unipolar pulse, drives it from dark gray, to get back to the black 300ms * 15V unipolar pulse that requires.Due to obvious reason, this waveform is not DC balance.

A scheme that solves pulse hysteresis problem is to use bipolar drive, in other words, in (potential) non-direct-path from a gray shade scale, drive this electro-optical medium to next gray shade scale, if desired, first apply a pulse and drive this pixel optics fence to enter optics fence, to maintain DC balance, then apply the optical states that the second pulse reaches this requirement.For example, in these cases, people can still by first applying additional negative voltage, then apply positive voltage by applying 100ms *-15V pulse from the black dark gray that becomes, and R1/R2 pulse curve is swum in downwards on black state, become again white from dark gray.Just as already discussed, by fence, stablize gray scale, error accumulation problem is also avoided in such indirect transformation.

The pulse hysteresis phenomenon of electro-optical medium and original state dependence, as discussed above, in above-mentioned patent and application form, require the waveform of each transformation to change along with considered pixel original state resume.As described at above-mentioned WO 03/044765, the optimum waveform of each transformation, can determine (that is the indicator corresponding with above-mentioned data array can be " tuning ") by utilizing initial " conjecture " transition matrix to set up waveform (it is used for being generally pseudorandom or original state completes this electro-optical medium of optical states series access by fixing).A program deducts from the target gray states of each original state the actual optical states reaching same combination, to calculate an error matrix, its size is identical with this transition matrix.Each element in this error matrix is corresponding to an element in transition matrix.If an element in this transition matrix is too high, corresponding element in this error matrix is also pushed away highlyer.Then can use PID (proportional-integral-differential) to control drives error matrix to zero.There is cross term (each element in this transition matrix affects the more than one element in this error matrix), but these effects are smaller, often along with tuning, by repeatedly iterating, undertaken, along with the amplitude of the value in this error matrix reduces and reduces.(note that I or the D constant of PID controller can be set to 0 sometimes, result is PI, and PD or P control.)

When this is tuning while finishing dealing with, the previous optical states of finding some number should be in this transition matrix, to reach the accurate performance of some gray shade scale.For example, utilize this processing specifically to seal electro-optical medium, result obtains a waveform, and wherein this controller records the more previous optical states of a ratio in this transition matrix, and the pulse that utilizes algorithm to calculate in this waveform first paragraph guarantees DC balance.In this waveform, the combination that this pulse electromotive force is allowed to each original state to being covered by this transition matrix is different.

Obtain relevant (" TM size ") between the number of this transition matrix size and the greatest optical error of this waveform, as table 1 below, list:

Table 1

TM dimension	Greatest optical error (L*)
		1	10.6
2	3.8
		3	2.1
4	1.7

Because the limit of general viewers visual experience is about 1L* unit, the data in this table show, it is very useful in this transition matrix, having more than one dimension, and two-dimensional matrix is better than one dimension, and three-dimensional matrice is better than two dimension etc.

Noticing above-mentioned all each points, is two gray-scale Control device design preferred wave shape form of above-mentioned R1/R2.This waveform maintains fixing pulse electromotive force for each last optical states R1, but uses two-dimentional transition matrix.Stablized fence, to reduce the accumulation of error, and design has low divergence in the process of upset, because its consideration pulse hysteresis curve.

Below in the symbol of use, digitized representation pulse.By applying negative pulse at one section of apply-V of given time (that is ,-15V), by apply positive pulse (that is waveform is width modulation) at one section of apply+V of given time, make the amplitude of volt-time product equal the amplitude of this pulse.Voltage modulated can be used as another program and uses.

In this preferred waveform, in renewal process, applying following pulse train each time, from left to right timely reading:

-TM(R1，R2)IP(R1)-IP(R2)TM(R1，R2)

Wherein " IP (Rx) " representative has the correlation values (being vector in this case) of the pulse electromotive force matrix of a numerical value from each gray shade scale, and TM (R1, R2) representative has the correlation values of the transition matrix of a numerical value from each R1/R2 combination.Certainly, for the value of some R1 and R2, TM (R1, R2) can bear.(just as already noted, for simplicity, the pulse train of the type can be abbreviated as " x/ Δ IP/x " sequence hereinafter.)

Value in this transition matrix can regulate as required, and needn't worry DC balance, because the clean pulse of first and the 3rd section of this waveform is always 0.Pulse potential difference values between initial and last state is applied to the interlude of this waveform.

By experience, have been found that this last driving pulse almost always has larger effect than initial pulse to last gray shade scale, so this transition matrix that this waveform is used can be tuning by above-mentioned same PID method.For fixing last gray shade scale, for the value of this pulse electromotive force setting affects the renewal speed of this waveform.For example, all pulse electromotive forces can be set to zero, but it causes a long update time, because this last driving pulse (the 3rd section) is always contrary with initial pulse (first paragraph) equal in length.Thereby in this case, this last driving pulse can not be longer than half of total update time.By careful strobe pulse electromotive force, likely much bigger total update time percent for this last pulse; For example, can reach and account for last driving pulse over half, and up to 80% of maximum total update time.

The length of different pulses is preferably selected by computing machine, utilizes that gradient is followed optimization method, similarly PID controls, limited difference is in conjunction with evaluation etc.

Pointed as WO 03/044765, the transformation in electro-optical medium is generally temperature sensitive, has been found that when all to the transformation of specific gray shade scale during all always from same optics fence, gray shade scale increases the uncompensated stability of temperature.Its reason is very simple, along with temperature, changes, and the switch speed of electro-optical medium becomes comparatively fast or be slower.If in 2 gray-scale monitors, dark gray is ejected from black fence to the transformation of bright gray scale, still the white transformation to bright gray scale is ejected from white fence.If it is slower that the switch speed of this medium becomes, from the bright grey states of black addressing, will become darker, but will become brighter from the bright grey states of black addressing.Thereby for the waveform of temperature stabilization, a given gray shade scale is always from same one " side " convergence, that is the last pulse of this waveform always has same polarity, this is important.In the preferred drive scheme of above-mentioned utilization-TM (R1, R2) IP (R1)-IP (R2) TM (R1, R2) sequence, this requires to select TM (R1, R2) value, makes the symbol of each numerical value only depend on R1, at least for some gray shade scale.A method for optimizing is, allowing TM value is any one symbol in black and white state, but only just uses for bright gray scale, and only use for dark gray, bear, thereby, from more approaching optics fence convergence middle gray.

This preferred waveform and technology are compatible fully, such as this waveform of insertion of short pause period to increase pulse resolution, as what the following describes.

Just as has been indicated, above-mentioned-x/ Δ IP/x pulse train can be revised, to comprise additional pulse.Such modification allows to comprise the pulse of an additional classes, is called hereinafter " y " pulse.The feature of " y " pulse is, it is the form of [+y] [y], and wherein y is a pulse value, can be negative or positive (in other words, this form [y] [+y] is effective equally).This y pulse is completely different with " x " pulse of describing in the past, be that half [x] and [+x] that this " x " pulse is right arranges before this Δ IP pulse and afterwards, yet " y " pulse can be arranged on other position of this pulse train.

Second such modification be on the arbitrfary point in this pulse train, add a 0V " pulse " (that is, one-period when related pixel is not applied to voltage), improve the performance of this sequence, for example, by this gray shade scale that this transformation causes is marginally moved up or down, or reduce or change the impact of previous state information on this pixel final state.Such 0V section is not can insert between different pulse elements, inserts exactly in the middle of single pulse element.

A structure fence is stablized the preferred method that waveform is used, and utilizes at the described indicator of WO 03/044765, and the method is as follows:

(a) numerical value (generally calculating by experience) of the pulse electromotive force that each gray shade scale uses is set, and for each changes the suitable Δ IP pulse of insertion indicator;

(b) for each transformation, pick the numerical value that an x uses, and inserted one before Δ IP-x pulse, and insert one+x pulse (just as already noted after Δ IP pulse, the value of x can be born, so-x and+x pulse can have polarity arbitrarily);

(c) for each, change, pick a y value, and insert one-y and+y pulse arrives this pulse train.Should-y/+y pulse combined can insert this pulse train in any edge of pulse circle, for example before-x pulse, before Δ IP pulse, before+x pulse or+x pulse after;

(d) for each, change any point in this sequence or what time insert n frame 0V, wherein n=0 or larger; With

(e) repeat as required above-mentioned steps repeatedly, until this waveform performance meets the requirements of level.

Now illustrate with reference to the accompanying drawings this processing.Fig. 2 represent a transformation use waveform basic-x/ Δ IP/+x structure, for the purpose of illustrating, the value of supposing two of x and Δ IP is all positive.Except undesired Δ IP and+be provided between x pulse 0V interval time, otherwise needn't be reduced between these two pulses, be applied to the voltage that this is tied, make this Δ IP and+in fact x pulse form a long positive pulse.

Fig. 3 with symbol illustrate [y] [+y] pulse to insert shown in Fig. 2 basic-x/ Δ IP/+x waveform.Should-y and+y pulse not necessarily will in succession occur, but can insert original waveform in different positions.There are two particularly advantageous special circumstances.

The first special circumstances, are somebody's turn to do " y ,+y " pulse to before the starting point of be placed on-x/ Δ IP/+x waveform ,-x pulse, to produce the waveform shown in Fig. 4.Have been found that when y is contrary with x symbol, as illustrational at Fig. 4, this last optical states can be by duration y, even appropriate coarse adjusting comes tuning subtly.Thereby the value of x can regulate for coarse control and the value of y can regulate for the last control of the last optical states of this electro-optical medium.Believe that this will occur, because y pulse increase-x pulse, thereby, change the degree that this electro-optical medium is pushed into its one of optics fence.The degree that pushes one of optics fence is known, to the meticulous adjusting (providing by x pulse in this case) away from this last optical states after the pulse of this optics fence is provided.

At the second in particular cases, at Fig. 5, illustrate, should-y pulse is placed on this again-starting point of x/ Δ IP/+x waveform, before-x pulse, but should+y pulse is placed on the end of this waveform, should+x pulse after.In the waveform of the type, this last pulse provides coarse tuning, because this last optical states is very responsive to the amplitude of y.It is thinner tuning that this x pulse provides, because last optical states generally depends on the amplitude that is driven into this optics fence less intensely.

Just as has been indicated, more than one pair of " y " pulse can be inserted this basic-x/ Δ IP/+x waveform, to allow the gray shade scale of this electro-optical medium to carry out " fine tuning ", and so multipair " y " pulse can differ from one another.Symbol example explanation for Fig. 6, in a kind of like this mode that is similar to Fig. 3, inserts second pair of y type pulse (indicating " z ", "+z ") waveform of Fig. 5.Be readily understood that because should-z and+z pulse can introduce on any pulse border of the waveform shown in Fig. 5, from introducing this-z and+z pulse can cause a large amount of different waveforms.A preferred gained waveform as shown in Figure 7; The waveform of the type is useful to the fine tuning of last optical states, and its reason is as follows.Do not have-z of investigation and+situation of z pulse (that is, Fig. 5 waveform discussed above).This x pulse element is for fine tuning, and this last optical states can reduce by increasing x, and increases by dwindling x.But, be undesirable beyond x is reduced to certain point, because then this electro-optical medium does not bring and stablizes this waveform requiredly enough close to an optics fence.For fear of this problem, by as shown in Figure 7, add-z ,+z pulse pair, makes z have the symbol contrary with x, is not to dwindle x, can (in fact) increase-x pulse and do not change+x pulse.Should+z pulse increase-x pulse, and while-z pulse keeps this transformation under the clean pulse of this requirement, thereby, maintain a DC balance change table generally.

In limited transition waveforms scheme of the present invention, for " diagonal line key element ", (this indicator key element and zero changes corresponding, wherein this initial and last gray shade scale are same, and address is because such key element is on leading diagonal line in one of an indicator common matrix formulation like this; Such diagonal line key element has Δ IP=0) to comprise x and two pulses of y.Any given indicator element can comprise zero group or organize x and/or y pulse more.

Limited transformation method of the present invention can also be utilized the frame period.For rewriteeing completely in the needed frame group of this display (being called easily superframe) process, repeat the scanning of pixel groups, and on a typical electrical optical display unit, more than this scanning will repeat once.Generally, fixing sweep speed is for upgrading, 50Hz for example, and this is to the frame of 20 milliseconds, and this allows.But this frame length may provide optimizing the resolution of waveform performance deficiency.In many cases, the frame of length t/2 is best, for example, in the waveform of general 20 milliseconds of frame lengths., in conjunction with the frame of difference number of times time delay, to produce a pulse resolution, be likely n/2.In order to consider a specific situation, the single frame that can be 1.5*t length inserts the starting point of this waveform, and a similar frame inserts this waveform end, and a similar Zheng Gai waveform end is (immediately before the end of 0V frame, this should occur with common frame rate, and it is generally used for the end of this waveform, to avoid by the caused undesirable impact of the residual voltage in pixel).These two longer frames can be by adding that the 0.5*t between the scanning of two consecutive frames realizes time delay simply.Then this waveform can have following structure:

T ms frame: t/2ms time delay: t ms frame [...] t ms frame: t/2ms time delay: t ms frame (all outputing to 0V)

For a length, be the normal frames of 20 milliseconds, initial and last frame add that their respective delay amounts to and are all 30 milliseconds.

Utilize this waveform, structure, initial and last pulse allow to change 10 milliseconds of length by following algorithm:

(a) if the length of this inceptive impulse can be divided exactly by t, the first frame is driven and is formed by 0V, and the frame of the t ms of a corresponding number is activated to meet the requirements of pulse length; Or

(b) if the length of this inceptive impulse leaves the remainder of t/2 during divided by t, the first frame 1.5*t is activated, and after this initial frame, the t millisecond frame of corresponding number is activated to meet the requirements of pulse length.

Same algorithm is followed in last pulse.Note that and will make this algorithm suitably work, this initial and last pulse must be distinguished starting point alignment and terminal aligns.In addition, in order to maintain DC balance, this initial and last pulse can be that one-x/+x is to corresponding part.

No matter whether adopt pause period, have been found that, realize to change the effect of waveform of use by this waveform in any pulse process or before it, exist a no-voltage cycle (being actually a time delay) to revise, limited transformation method of the present invention can comprise the no-voltage cycle in this waveform in subsequent pulses or between them, that is this waveform can be " discrete ", as this term in the above and use in above-mentioned PCT/US2004/010091.Fig. 8 to 10 illustrate comprising of Fig. 2 in such no-voltage cycle basic-variation of x/ Δ IP/+x waveform.In the waveform of Fig. 8, this-insert a time delay between x pulse and this Δ IP pulse.In the waveform of Fig. 9, at this Δ IP pulse interpolation, enter a time delay, or still the same, this Δ IP pulse is split as two independent pulses of this time delay of being separated by.The waveform of Figure 10 is similar to the waveform of Fig. 9, and just this time delay is inserted in be somebody's turn to do+x pulse.Time delay can be contained in a waveform, reaches and there is no such delay optical states that is just beyond one's reach.Time delay can also be used to the last optical states of fine tuning.This fine tuning ability is important, because in a driven with active matrix, the temporal resolution of each pulse is defined by the sweep speed of this display.The temporal resolution being provided by this sweep speed may be very coarse, so that do not have the additional means of some fine tuning just can not reach accurate last optical states.

Interrupt scanning method of the present invention

Just as already noted, the present invention, for driving the electro-optic displays with a plurality of pixels that are divided into a plurality of groups, provides a kind of " interrupt scanning " method.The method comprises each that select in a plurality of pixel groups in succession, and is not to apply driving voltage to each pixel in this selected group, applies exactly non-driving voltage, completes the scanning of all pixel groups in the first frame period.In the second frame period, (should be understood that, any specific pixel can apply driving voltage in the first frame period, and in the second frame period process, apply non-driving voltage, and vice versa) apply the scanning that repeats this pixel groups in driving voltage process in process.In interrupt scanning method of the present invention, between the first and second frame periods, in pause period, interrupt the scanning of this pixel groups, this pause period is no longer than the first or second frame period.In the method, this first and second frame period is generally equal in length, and the length of this pause period is generally the part (best 1/2,1/4 etc.) of the length in one of each frame period.

Between difference is to adjacent frame periods, this interrupt scanning method can comprise a plurality of pause period.The best length of a plurality of pause period like this equates substantially, and the total length of a plurality of pause period is not preferably to equal a complete frame period, equals exactly a frame period to deduct a pause period.For example, as what will discuss more in detail below, an embodiment of the first method can be used a plurality of 20ms frame periods, and not to be three be exactly four 5ms pause period.

In this interrupt scanning method, pixel groups is generally each row of a traditional row/column active matrix pixel array certainly.This interrupt scanning method comprise each that select in a plurality of pixel groups in succession (that is, generally, scan each row of this matrix), and be not to apply driving voltage in this selected group, apply exactly non-driving voltage, the scanning of all each pixel groups all completes in the first frame period.Repeat the scanning of this pixel groups, and on a typical electro-optic displays, in rewriteeing the desired superframe process of this display, repeat this more than run-down.In the pause period process of the scanning of this pixel groups between the first and second frame periods, interrupt, this pause period is no longer than the first or second frame period.

Although driving voltage is only applied to any specific pixel electrode with a line addressing time in the process of scanning each time, in the time course of in succession selecting between same a line, this driving voltage continues on this pixel electrode, just decay lentamente, when other row of this matrix is chosen, this pixel is continued in this time course driven, and interrupt scanning method rely on this continuation of this pixel driven.In its " non-selected " time course, ignore this moment, in its non-selected time course, the slow-decay of this voltage on this pixel electrode, in the frame period process before this pause period immediately, the pixel that is set to this driving voltage will continue this driving voltage of experience in this pause period process, make for such a pixel, in fact the frame period above extends the length of one section of pause period.On the other hand, be set to the pixel of this non-driving (be generally zero) voltage, in the frame period process before the pause period of this before x immediately, will in this pause period process, continue to experience no-voltage.May wish to regulate the length of this pause period, to allow the slow-decay of the voltage on this pixel electrode, to guarantee that the overall pulse of carrying has the numerical value of requirement in this pause period process.

Presented for purposes of illustration in order to lift simple example of this interrupt scanning method, investigate and there is the simple width modulation drive scheme of superframe for example, being formed by a plurality of (, 10) 20ms frame.Generally, the last frame of this superframe will arrange whole pixels to this non-driving voltage, because when bistable electro-optic displays generally just thinks that shown image changes maybe when thinking just driven in relatively long interval time while preferably refreshing shown image, make each superframe by a long cycle of general heel, wherein this display is not driven, and change rapidly for fear of occurring in some pixel in this long non-drive cycle process, when this superframe finishes, to be set to non-driving voltage be extraordinary to whole pixels.In order to change such a drive scheme according to interrupt scanning method of the present invention, can between two 20ms frames in succession, insert a 10ms pause period, and this simple modification makes applied pulse and complete possible maximum difference between the pulse that a given transformation needs ideally to reduce by half, with this, in the maximum deviation of the gray shade scale being reached, roughly reduce by half in practice.This 10ms pause period is inserted in each superframe after frame second from the bottom easily, but inserts on other point that if desired can be in this superframe.

In practice, in this example, preferably not only insert 10ms pause period, but also an additional 20ms frame is inserted to each superframe.Unmodified drive scheme makes people apply such pulse to any one given pixel:

0,20,40,60...160 ，180 unit

One of them unit pulse be defined as apply that driving voltage 1ms causes pulse.Thereby, the maximum difference Shi10 unit between available pulse and the desirable pulse of a given transformation use.(because the last frame of this superframe is all set to non-driving voltage whole pixels, only have 1/9 frame of this superframe to can be used to apply this driving voltage.Just as already explained, the any pixel that is set to this driving voltage on frame before this pause period all continues to experience one of this driving voltage and equals the cycle that this frame period adds pause period, thereby, for this frame, the pulse of experience Yi Ge 30 units, rather than 20 units.Correspondingly, amended drive scheme allows to apply such pulse to any given pixel:

0,20,30,40，50，, 60 units etc.

Preferably this additional frame is inserted to superframe, to allow this amended drive scheme can transmit a just pulse for 180 units.Because the pulse of the several times of any just Shi20 unit, all require this related pixel in the frame process before this pause period to be set to non-driving voltage, reach one just the pulse of 180 units require the superframe of 11 frames, make any pixel that will receive this 180 pulse in 9 frame processes, to be set to this driving voltage, in frame before this pause period, be set to this non-driving voltage, and (as always) is set to non-driving voltage in the last frame of this superframe.Thereby when utilizing this amended drive scheme, the maximum difference between available pulse and the ideal pulse of a given transformation use is reduced to 5 units.Although (this amended drive scheme can not apply the pulse of Yi Ge 10 units, and its consequence is very little in practice.In order to produce quite consistent gray shade scale, the number of available pulse must be greater than the gray shade scale number of this display significantly, makes any gray scale change all pulses of the unlikely Yi Ge little Dao of requirement 10 units.)

Certainly this pause period can have and reaches the desired any number of the control of the requirement of applied pulse and length.For example, not to change this above-mentioned drive scheme to comprise a 10ms pause period, but this drive scheme can be revised to such an extent that after different 20ms drives frame, comprise three 5ms pause period, preferably returns the driving frame that this drive scheme adds other three 20ms, not heel pause period.This amended drive scheme allows to apply some such pulses to any given pixel:

0,20,25,30,35....170 ，175，, 180 units

With the maximum difference between this available pulse and the ideal pulse of a given transformation use, reduce to 2.5 units, compare and dwindled four times with original unmodified drive scheme.

The polarity problems of the pulse applying has been ignored in discussion before this interrupt scanning method.As above and discuss at above-mentioned WO 03/044765, bistable electro-optical medium requires the pulse of two kinds of polarity all to apply.In some drive scheme, such as the drive scheme of showing slides, before new image writes display, first whole pixels of this display are all driven into an extreme optical state, be not black be exactly white, and after this by the pulse of single polarity, this pixel driver is arrived to their last gray states.Such drive scheme can be revised according to this interrupt scanning method by the mode of having described.Other drive scheme requires the pulse of two kinds of polarity all to apply the last gray states to them this pixel driver.The pulse of two polarity can be to be applied in independent frame, for example, or the pulse of two polarity can be to be applied on same frame, utilizes three grade drive schemes, wherein public front termination electrode remains on a V/2 voltage, and each pixel electrode remains on 0 simultaneously, on V/2 or V.When the pulse of two polarity is applied on single frame, expect by providing at least 2 single pause period to affect interrupt scanning method, a pause period is followed the frame that is wherein applied in a polar impulse, and another pause period is followed the frame that is wherein applied in opposite polarity pulse.But, when utilizing that one wherein the pulse of two kinds of polarity is all applied to the drive scheme on same frame, this interrupt scanning method can only be utilized a single pause period, because as becoming apparent from the above, the effect that comprises a pause period after a frame is, increase the amplitude be applied to the pulse in any pixel that has applied driving voltage in this frame, and irrelevant with the polarity of this driving voltage.

Also as above-mentioned WO 03/044765 and discussed above, the most handy such drive scheme of many bistable electro-optical media drives, it reaches long-term DC current (DC) balance, and such DC balance utilizes such drive scheme to realize easily, the gray shade scale of one of them its pixel substantially immovable DC balancing segment is applied in before the main driving section that will change gray shade scale, selects to such an extent that to make applied pulse algebraic sum be 0 or at least very little for these two sections.If this main driving section is revised according to this interrupt scanning method, strong this DC balancing segment of proposed amendments, to avoid by the caused additional pulse of pause period of inserting accumulation and cause DC serious unbalance.But this DC balancing segment needn't be usingd and revised as this main mode for the accurate mirror image that section revises that drives, because this DC balancing segment can have gap (no-voltage frame), and most of electro-optical medium is not subject to the unbalance injury of short time DC.Thereby, in the drive scheme that utilizes the middle single 10ms pause period of 10 20ms frames of an insertion of discussing in the above, by making the duration of the first frame of this drive scheme, be 30ms, can reach DC balance.In this frame process, to a pixel, apply or do not apply driving voltage, can make overall pulse equal the several times of 20 units, make this pulse balance easily subsequently.At the drive scheme that utilizes three 5ms pause period, first two frames of this drive scheme can be 25 and 30ms (with any order) similarly on the duration, make again overall pulse equal the several times of 20 units.

From the above, will find out, interrupt scanning method of the present invention require extend the addressing time and improve paired pulses and the control of the gray scale that therefore the method produced between weigh the advantages and disadvantages, extending the addressing time is because the pause period for each insertion in each superframe comprises that the needs of an additional frame cause.But, as long as this interrupt scanning method makes addressing time lengthening not many, can provide the improvement of the control highly significant of paired pulses; For example, an above-mentioned superframe comprises that 10 20ms frames revise to comprise the drive scheme of three 5ms pause period, and 40% the addressing time lengthening of can take is cost, produces the improvement of four times in pulse accuracy.

The gray scale method of balance of the present invention

Just as already noted, the present invention also provides a kind of balance gray scale method, in order to the electro-optic displays that drives to configure a plurality of pixels in an array.This pixel drives can apply the pulse-width modulation waveform of a plurality of different pulses.Drive circuit stores indicates whether to apply the data of a given pulse, will produce one higher or lower than the gray shade scale that requires gray shade scale.When two of two neighbors all require in same gray scale, the pulse regulation being applied in these two pixels must make a pixel lower than the gray scale requiring, and simultaneously one other pixel higher than the gray scale requiring.

In form preferred one of the method, pixel is divided into Liang Ge group, indicates hereinafter " even number " and " odd number ".These two pixel groups can be configured to checkerboard pattern (make between the pixel Gai Liangge group of each row and column alternately), or do other configuration, as described at above-mentioned WO 03/044765, suppose that each pixel has the neighbours of at least one contrary group, and Liang Ge group is used different drive schemes.If it is the transformation of requirement gray shade scale by producing substantially that the data of storage are pointed out one of available pulse, the pulse applying for this transformation, both dual numbers pixel, applied this pulse to odd pixel again.But, if storage data point out a specific gray shade scale change desired pulse substantially between these two available pulses half, one of these pulses change for this in even pixel, and other transformations for odd pixel of these pulses.Thereby, if two adjacent pixels are wanted in same gray states (under the most important state of accurate control of gray scale), one of these pixels will have the gray shade scale that is slightly higher than calling hierarchy, and simultaneously another by the gray shade scale having slightly lower than this calling hierarchy.Vision with the average result of optics will be see these two gray shade scales average, thereby, produce one than the grade reaching with available pulse the approximate gray scale closer to calling hierarchy.In fact, this balance gray shade scale method is used small-signal spatial jitter (applying to proofread and correct the error of applied pulse) to overlap in the true gray scale of large-signal, increases the factor of two available impulse ratings.Because each pixel is still in ballpark gray shade scale, the effective resolution of actual display is not compromise.

The necessary calculating of a complete realization provides as follows by MATHLAB false code.This floor function is rounded up to nearest integer downwards, and mod function calculates its first independent variable is divided by the remainder of its second independent variable:

quotient＝floor(desired_impuslse)

remainder＝mod(desired_impulse，1)

if remainder＜＝0.25

even_parity_impulse＝quotient

odd_parity_impulse＝quotient

else if remainder＜＝0.75

even_parity_impulse＝quotient+1

odd_parity_impulse＝quotient

else

even_parity_impulse＝quotient+1

odd_parity_impulse＝quotient+1

end.

In some drive scheme of former description, the circulation RSGS drive scheme of for example describing above with reference to Figure 1A and 1B, the pixel of this display has been divided into two groups, and different drive schemes is applied to these two groups so that the amplitude that meets the requirements of the needed pulse of gray shade scale by be two groups different.Can revise such " two groups " drive scheme according to balance gray shade scale method, but the details that realizes of the method is slightly different from simple situation discussed above.Compared the transformation of available pulse and requirement is needed simply, but calculated individually the error of two groups of gray scales, this error is got to arithmetic mean, and determined whether mobile Yi Ge group will reduce this arithmetic mean to different available pulses.Note that in this case, reduce arithmetic mean and can differently depend on which group is moved to a different pulse, and obviously, mobile which can produce less which on average just to carry out.

Again, the method it is contemplated that the small-signal spatial jitter realizing into top in large-signal interior intensity, with this small-signal, shakes to proofread and correct the pulse error that the limitation of width modulation used causes.Because in this scheme, each pixel still roughly in correct gray scale, and is proofreaied and correct and is just proofreaied and correct approximation error, and the true resolution of display is not compromised.Say it in another way, the method realizes small-signal spatial jitter at the top of the real gray scale of large-signal.

Diverse ways of the present invention can utilize at above-mentioned application form, different additional variation and technology that especially above-mentioned WO03/044765 and PCTUS2004/010091 describe.Will appreciate that, in order to drive on total waveform of an electro-optic displays, at least in some cases, some transformation can be carried out according to diverse ways of the present invention, and other transformation simultaneously can not utilize the method for the present invention, still can utilize the transformation of other type the following describes.For example, diverse ways of the present invention can utilize any one or more:

Discontinuous addressing (see above-mentioned PCT/US2004/010091, paragraph [0142] is to [0234] and Fig. 1-12);

The addressing of DC balance, as upper part (also see above-mentioned PCT/US2004/010091, paragraph [0235] is to [0260] and Figure 13-21) of discussing;

Defined area update (see above-mentioned PCT/US2004/010091, paragraph [0261] is to [0280]);

(see above-mentioned PCT/US2004/010091, paragraph [0284] to [0308] and Figure 22) in bucking voltage addressing;

DTD integration dwindles addressing, and (see above-mentioned PCT/US2004/010091, paragraph [0309] to [0326] and Figure 23); With

Residual voltage addressing (seeing 03/044,765,59 to 62 pages of above-mentioned WO).

Claims (23)

1. for driving a method for the electro-optic displays with at least one pixel, described at least one pixel can reach any one gray shade scale at least four different gray shade scales that comprise two extreme optical state, and described method comprises:

On described display, show the first image; With

Rewrite described display, to show the second image thereon,

Described method characteristic is: in the rewriting period of described display, the any pixel driver that does not touch extreme optical state surpassing the number of transitions of predetermined value by having stood is to it before last optical states in described the second image, by described pixel driver, at least one extreme optical state, described predetermined value is at least one.

2. according to the method for claim 1, wherein, realize the rewriting of described display, once make by a kind of pulse of polarity, pixel to be driven towards contrary extreme optical state from an extreme optical state, described pixel does not receive the pulse of opposite polarity, until it has reached described contrary extreme optical state.

3. according to the process of claim 1 wherein, described predetermined value is not more than N/2, and wherein N is the sum of the pixel gray shade scale that can show.

4. according to the process of claim 1 wherein, the rewriting of described display by described pixel or each pixel, apply voltage-V, 0 and+any one or more realizations in V.

5. according to the process of claim 1 wherein, realize the rewriting of described display, any transformation series that makes to stand for pixel, the voltage that applies to the integration of time, be all bounded.

6. according to the process of claim 1 wherein, realize the rewriting of described display, initial gray shade scale and the last gray shade scale of this transformation only depended in the pulse that makes to apply to pixel between tour.

7. according to the process of claim 1 wherein, at least one from gray shade scale R2 to gray shade scale R1 standing for described at least one pixel changes, and what to described pixel, apply is the pulse train of following form:

-TM(R1，R2)IP(R1)-IP(R2)TM(R1，R2)

Wherein " IP (Rx) " represents from the correlation values each gray shade scale to the pulse electromotive force matrix of a numerical value, and TM (R1, R2) representative is from each R1/R2 combination being had to the correlation values of the transition matrix of a numerical value.

8. according to the method for claim 7, wherein, for wherein initial gray shade scale and the finally different whole transformations of gray shade scale, described pulse train has following form:

-TM(R1，R2)IP(R1)-IP(R2)TM(R1，R2)。

9. according to the method for claim 7, wherein, in described-TM (R1, R2) IP (R1)-IP (R2) TM (R1, R2) sequence, last TM (R1, R2) section takies the over half of maximum update time.

10. according to the process of claim 1 wherein, realize the rewriting of described display, make always by the final pulse of identical polar, to realize to the transformation of given gray shade scale.

11. according to the method for claim 10, and wherein, the gray shade scale beyond described two extreme optical state is from the direction convergence of nearer extreme optical state.

12. according to the method for claim 7, wherein, selects the numerical value of described TM (R1, R2), makes the symbol of each numerical value only depend on R1.

13. according to the method for claim 12, wherein, described TM (R1, R2) numerical value be chosen to one or more bright gray shade scales for just to one or more dark gray levels for negative, make gray shade scale beyond described two extreme optical state from the direction convergence of nearer extreme optical state.

14. according to the method for claim 7, wherein, described at least one transformation also comprises the extra-pulse pair of [+y] [y] form, wherein y be can just maybe can be negative pulse value, described [+y] and [y] pulse are inserted into described-TM (R1, R2) IP (R1)-IP (R2) TM (R1, R2) sequence.

15. according to the method for claim 14, wherein, described at least one transformation also comprises the second extra-pulse pair of [+z] [z] form, wherein z is different from the pulse value of y and can bears or can be just, described [+z] and be inserted into-TM of [z] pulse (R1, R2) IP (R1)-IP (R2) TM (R1, R2) sequence.

16. according to the method for claim 7, and wherein, described at least one transformation does not also comprise executes the alive period to described pixel.

17. according to the method for claim 16, wherein, to described pixel, does not execute between two elements that the alive described period appears at described-TM (R1, R2) IP (R1)-IP (R2) TM (R1, R2) sequence.

18. according to the method for claim 16, wherein, to described pixel, does not execute between the intermediate point in the individual element that the alive described period appears at described-TM (R1, R2) IP (R1)-IP (R2) TM (R1, R2) sequence.

19. according to the method for claim 16, and wherein, described at least one transformation comprises that at least two are not executed the alive period to described pixel.

20. according to the method for claim 7, and wherein, described display comprises a plurality of pixels that are divided into a plurality of groups, and described transformation realizes by following steps:

(a) each pixel groups in a plurality of pixel groups described in Continuous Selection, and each pixel in described selected group applies driving voltage or non-driving voltage, all the scanning of pixel groups completes within the first frame period;

(b) during the second frame period, repeat the scanning of described pixel groups; With

(c) between the pause period between described the first and second frame periods, interrupt the scanning of described pixel groups, described suspending period is no longer than the described first or second frame period.

21. according to the process of claim 1 wherein, described electro-optic displays comprises electrochromism or rotates double-colored film electro-optical medium.

22. according to the process of claim 1 wherein, described electro-optic displays comprises the electrophoretic medium through encapsulation.

23. according to the process of claim 1 wherein, described electro-optic displays comprises microcell electrophoretic medium.

Images