4.2
`
`Display Technologies
`
`155
`
`4.2 DISPLAY TECHNOLOGIES
`
`lnterdctive computer graphics demands display devices whose images can be changed
`quickly. Nonpermanent image displays allow an image to be changed, making possible
`dynamic movement of portions of an image. The CRT is by far the most common display
`device and will remain so for many years. However, solid-state technologies are being
`developed that may, in the long tem1, substantially reduce the dominance of the CRT.
`The monocllrommic CRTs used for graphics displays are essentially the san1e as those
`used in black-and-white home television sets. Figure 4. 12 shows a highly stylized
`cross-sectional view of a CRT. The electron gun emits a stream of electrons that is
`accelerated toward the phosphor-coated screen by a high positive voltage applied near the
`face of the tube. On the way to the screen, the electrons are forced into a narrow beam by
`the focusing mechanism and are directed toward a particular point on the screen by the
`magnetic field produced by the deflection coils. When the electrons hit the screen, the
`phosphor emits visible light. Because the phosphor's light output decays exponentially with
`time, the entire picture must be refreshed (redrawn) many times per second, so that the
`viewer sees what appears to be a constant, untlickering picture.
`The refresh rate for raster-scan displays is usually at least 60 frames per second, and is
`independent of picture complexity. The refresh rate for vector systems depends directly on
`picture complexity (number of lines, points, and characters): The greater the complexity,
`the longer the time taken by a single refresh cycle and the lower the refresh rate.
`The stream of electrons from the heated cathode is accelerated toward the phosphor by
`a high voltage, typically 15,000 to 20,000 volts, which determines the velocity achieved by
`the electrons before they hit the phosphor. The control-grid voltage detem1ines how many
`electrons are actually in the electron beam. The more negative the control-grid voltage is,
`the fewer the electrons that pass through the grid. This phenomenon allows the spot's
`
`Heating
`filament
`Electron
`gun
`Focusing
`system
`
`Cathode
`
`Control
`grid
`
`Interior metallic coating
`at high positive voltage
`
`Deflection '
`cons
`
`Screen
`
`Fig. 4 .12 Cross-section of a CRT (not to scale).
`
`TEXAS INSTRUMENTS EX. 1009 - 178/1253
`
`

`
`1 56
`
`Graphics Hardware
`
`intensity to be controlled, because the light output of the phosphor decreases as the number
`of electrons in the beam decreases.
`The focusing system concentrates the electron beam so that the beam converges to a
`small point when it hits the phosphor coating. It is not enough for the electrons in the beam
`to move parallel to one another. They would diverge because of electron repulsion, so the
`focusing system must make them converge to counteract the divergence. With the exception
`of this tendency to diverge, focusing an electron beam is analogous to focusing a Light
`beam. An optical lens and an electron Ieos both have a focal distance, which in the case of
`the CRT is set such that the beam is focused on the screen. The cross-sectional electron
`density of the beam is Gaussian (that i.s, normal), and the intensity of the light spot created
`on the phosphor has the same distribution, as was shown in Fig. 4.3. The spot thus has no
`distinct edges, and hence spot size is usually specified as the diameter at which the intensity
`is 50 percent of that at the center of the spot. The typical spot size in a high-resolution
`monochrome CRT is 0.005 inches.
`Figure 4.13 illustrates both the focusing system of and a difficulty with CRTs. The
`beam is shown in two positions. In one case, the beam converges at the point at which it
`strikes the screen. In the second case, however, the convergence point is behind the screen,
`and the resulting image is therefore somewhat blurred. Why has this happened? The
`faceplates of most CRTs are nearly Hat, and hence have a radius of curvature far greater than
`the distance from the lens to the screen. Thus, not all points on the screen are equidistant
`from the lens, and if the beam is in focus when it is directed at the center of the screen , it is
`not in focus anywhere else on the screen. The further the beam is deHected from the center,
`the more defocused it is. In high-precision displays, the system solves this problem by
`focusing the Ieos dynamically as a function of the beam's position; making CRTs with
`sharply curved faceplates is not a good solution.
`When the electron beam strikes the phosphor-coated screen of the CRT, the individual
`electrons are moving with kinetic energy proportional to the acceleration voltage. Some of
`this energy is dissipated as heat, but the rest is transferred to the electrons of the phosphor
`
`Deflected beam
`converges inside
`faceplate
`
`Focus
`
`Undeflected beam
`converges at
`faceplate
`
`.,-.~
`
`~--- Focal length - - --1
`
`Fig. 4 .13 Focusing of the electron beam. The focal length varies as a function of the
`deflection angle.
`
`TEXAS INSTRUMENTS EX. 1009 - 179/1253
`
`

`
`4.2
`
`Display Technologies
`
`1 57
`
`atoms, making them jump to higher quantum-energy levels. ln returning to their previous
`quantum levels, these excited electrons give up their extra energy in the form of light, at
`frequencies (i.e., colors) predicted by quantum theory. Any given phosphor has several
`different quantum levels to which electrons can be excited, each corresponding to a color
`associated with the return to an unexcited state. Further, electrons on some levels are less
`stable and return to the unexcited state more rapidly than others. A phosphor' s fluorescence
`is the light emitted as these very unstable electrons lose their excess energy while the
`phosphor is being struck by electrons. Phosphorescence is the light given off by the return of
`the relatively more stable excited electrons to their unexcited state once the electron beam
`excitation is removed. With typical phosphors, most of the light emitted is phosphores(cid:173)
`cence, since the excitation and hence the fluorescence usually last just a fraction of a
`microsecond. A phosphor's persistence is defined as the time from the removal of excitation
`to the moment when phosphorescence has decayed to I 0 percent of the initial light output.
`The range of persistence of diffe.rent phosphors can reach many seconds, but for most
`phosphors used in graphics equipment it is usually I 0 to 60 microseconds. This light output
`decays exponentially with time. Characteristics of phosphors are detailed in [SHER79].
`The refresh rate of a CRT is the number of times per second the image is redrawn; it is
`typically 60 per second for raster displays. As the refresh rate decreases, flicker develops
`because the eye can no longer integrate the individual light impulses coming from a pixel.
`The refresh rate above which a picture stops flickering and fuses into a steady image is
`called the critical fusion frequency, or CFF. The process of fusion is familiarto all of us; it
`occurs whenever we watch television or motion pictures. A flicker-free picture appears
`constant or steady to the viewer, even though, in fact, any given point is "off" much longer
`than it is "on."
`One detemunant of the CFF is the phosphor's persistence: The longer the persistence,
`the lower the CFF. The relation between fusion frequency and peri>'istence is nonlinear:
`Doubling persistence does not halve the CFF. As persistence increases into the several(cid:173)
`second range, the fusion frequency becomes quite small. At the other extreme, even a
`phosphor with absolutely no persistence at all can be used, since all the eye really requires is
`to see some light for a short period of time, repeated at a frequency above the CFF.
`Persistence is not the only factor affecting CFF. CFF also increases with image
`intensity and with ambient room lighting, and varies with different wavelengths of emitted
`light. Finally, it depends on the observer. Fusion is, after all, a physiological phenomenon,
`and differences among viewers of up to 20Hz in CFF have been reported [ROG083]. Cited
`fusion frequencies are thus usua!Jy averages for a large number of observers. Elinlinating
`flicker for 99 percent of viewe.rs of very high-intensity images (especially prev-~lent with
`black-on-white raster displays) requires refresh rates of 80 to 90 Hz.
`The horizontal scan role is the number of scan lines per second that the circuitry driving
`a CRT is able to display. The rate is approximately the product of the refresh rate and the
`number of scan lines. For a given scan rate, an increase in the refresh rate means a decrease
`in the number of scan lines.
`Tbe resolution of a monochromatic CRT is defined just as is resolution for hardcopy
`devices. Resolution is usually measured with a shrinking raster: A known number of equally
`spaced parallel lines that alternate between black and white are displayed, and the interline
`spacing is uniformly decreased until the lines just begin to merge together into a uniform
`
`TEXAS INSTRUMENTS EX. 1009 - 180/1253
`
`

`
`1 58
`
`Graphics Hardware
`
`field of gray. This merging happens at about the point where the interline spacing is equal to
`the diameter at which the spot intensity is 60 percent of the intensity at the center of the
`spot. The resolution is the distance between the two outermost lines, divided by the number
`of lines in the raster. There is a clear dependence between spot size and achievable
`resolution: The larger the spot size. the lower the achievable resolution.
`In the shrinking-raster process, the interline spacing is decreased not by modifying the
`contents of a raster bitmap, but by changing the ga.in (amount of amplification) of the
`vertical or horizontal deflection amplifiers, depending on whether the vertical or horizontal
`resolution is being measured. These amplifiers control how large an area on the screen is
`CO\'ered by the bitmap image. Thus, CRT resolution is (properly) not a function of the
`bitmap resolution, but may be either higher or lower than that of the bitmap.
`Resolution is not a constant. As the number of electrons in the beam increases,
`resolution tends to decrease, because a bright line is wider than a dim line. Th.is effect is a
`result of bloom, the tendency of a phosphor's excitation to spread somewhat beyond the area
`being bombarded, and also occurs because the spot size of an intense electron beam is
`bigger than that of a weak beam. Vertical resolution on a raster monitor is determined
`primarily by spo1 size; if the vertical resolution is n Lines per inch, the spo1 size needs to be
`about lin inches. Horizontal resolution (in which the line-pairs are vertical) is determined
`by bolh spot size and the speed with which the electron beam can be turned on and off as it
`moves horizontally across the screen. This rate is related to the bandwidth of the display, as
`discussed in the next paragraph. Research on defining the resolution of a display precisely
`and on our ability to perceive image.~ is ongoing. The modulatiotr transfer functiotr, used
`extensively in this research, relates a device's input signal to its output signal [SNY08Sl
`The bandwidth of a monitor has to do with the speed with which the electron gun can be
`turned on or off. To achieve a horizontal resolution of " pixels per scan line, it must be
`possible to tum the elec:ron gun on at least n/2 times and off another n/2 times in one scan
`line, in order to create alternating on and off lines. Consider a raster scan of 1000 lines by
`1000 pixels, displayed at a 60-Hz refresh rate. One pixel is drawn in about II nanoseconds
`[WHJT84), so the period of an on-off cycle is about 22 nanoseconds, which corresponds to
`a frequency of 45 MHz. This frequency is the minimum bandwidth needed to achieve 1000
`lines (500 line-pairs) of resolution, but is not the actual bandwidth bec:wse we have ignored
`the effect of spot size. The nonzero spot size must be compensated fo: with a higher
`bandwidth which causes the beam to turn on and off more quickly, giving the pixels sharper
`edges than they would have otherwise. It is not unusual for the actual bandwidth of a I 000
`by 1000 monitor to be 100 MHz. The actual relationships among resolution, bandwidth,
`and spo1 size are complex, and only rec:lelltly has progress been made in quantifying them
`[JNFA85].
`Color television sets and color raster displays use some form of shildow-mask CRT.
`Here, the inside of the tube's viewing surface is covered with c.losely spaced groups of red,
`green, and blue phosphor dots. The dot groups are so small that light emanating from the
`individual dots is perceived by the viewer as a mixture of the three colors. Thus, a wide
`range of colors can be produced by each group, depending on how strongly each individual
`phosphor dot is excited. A shadow mosk, which is a thin metal plate perforated with many
`small holes and mounted close to the viewing surface, is carefully aligned so that each of
`
`TEXAS INSTRUMENTS EX. 1009 - 181/1253
`
`

`
`4 .2
`
`Display Technologies
`
`159
`
`the three electron beams (one each for red, green, and blue) can hit only one type of
`phosphor dot. Tile dots thus can be excited selectively.
`Figure 4.14 shows one of the most common types of shadow-mask CRT, a delra-delta
`CRT. The phosphor dots are arranged in a triangular triad pattern, as are the three electron
`guns. The guns are deOected together, and are aimed (converged) at the sarne point on the
`viewing surface. The shadow mask bas one smaU bole for each triad. The holes are
`precisely aligned with respect to both the triads and the electron guns, so that each dot in the
`triad is el(posed to electrons from only one gun. High-precision deltlH!elta CRTs are
`particularly difficult to keep in alignment. An ahernative arrangement, the precision in-line
`delta CRT shown in Fig. 4 . 15, is easier to converge and is gaining in popularity for
`high-precision ( I 000-scan-lines) monitors. 1n this case, the three beams simultaneously
`expose three in-line phosphor dots. However, the in-line arrangement does slightly reduce
`image sharpness at the edges of the tube. Still in the research laboratory but likely to
`become commercially viable is the flat-panel color CKI, in which tile electron beams move
`parallel to the viewing surface, and are then turned 90• to strike the surface.
`The need for the shadow mask and triads imposes a limit on the resolution of color
`CRTs not present with monochrome CRTs. Tn very high-resolution tubes, the triads are
`placed on a.bout 0.21-miUimeter centers; those in home television tubes are on about
`0.60-millimeter centers (this distance is also called the pitch of the tube). Because a finely
`focused beam cannot be guaranteed to hit exactly in the center of a shadow-mask hole, the
`beam diameter (the diameter at which the intensity is 50 percent of the maximum) must be
`about t times the pitch. Thus, on a mask with a pitch of 0.25-millimeter (0.01 inches),
`
`Phosphors
`on glass
`faceplate
`
`~
`
`Red
`
`""'-..Metal
`mask
`
`Fig. 4 .14 Delta-delta shadow-mask CRT. The three guns and phosphor dots are
`arranged in a triangular (delta) pattern. The shadow mask allows electrons from each
`gun to hit only the corresponding phosphor dots.
`
`TEXAS INSTRUMENTS EX. 1009 - 182/1253
`
`

`
`160
`
`Graphics Hardware
`
`Elec:tron
`
`Phosphors
`on glass
`faceplate
`
`i
`
`"-Metal
`mask
`
`Fig. 4 .1 5 A precision in-line CRT: the three electron guns are in a line.
`
`the beam is about 0.018 inches across, and the resolution can be no more than about
`o.fu- = 55 lines per inch. On a 0.25-millimeter pitch, 19-inch (diagonal measure) monitor.
`which is about 15.5 inches wide by 11 .6 inches high [CONR85], the resolution achievable
`is thus only 15.5 x 55 = 850 by 11.6 x 55 = 638. This value compares with a typical
`addressability of 1280 by 1024, or 1024 by 800. As illustrated in Fig. 4 .2, a resolution
`somewhat less than the addressability is useful.
`The pitch of the shadow mask is clearly an important limit on the resolut.ion of
`shadow-mas.k CRTs. As pitch decreases, resolution can increase (assuming bandwidth and
`dot size are appropriate). The smaller the pitch is, however, the more difficult the tube is to
`manufacture. A small-pitch shadow mask is more fragile, making it more difficult to
`is also more likely to warp from heating by the electron beam. The
`mount. II
`flat·tension·mask tube has a flat faceplate, with the mask stretched tightly to maintain its
`position; a pitch of0.15 millimeter is achievable with this technology.
`The shadow mask also limits CRT brightness. Typically, only 20 percent of the
`electronS in the beam hit the phosphors-the rest hit the mask. Thus. fewer electrons make
`light than in a monochrome CRT. The number of electrons in the beam (the beam cu"mt)
`can be increased, but a higher current makes focusing more difficult and also generates
`more heat on the shadow mask, further e~tacerbating mask warping. Because the flat tension
`mask is more resistant to heating distortions, it allows a higher beam current and hence a
`brighter image.
`Most high-quality shadow-mask CRTs have diagonals of 15 to 21 inches, with slightly
`curved faceplates thai create op1ical dislortions for the viewer. Several types of fla1-faee
`CRTs are becoming available, including a 29-inciH:Iiagonal tube with a pilch of 0.31
`millimeter. Of course, the price is high, bul it will come down as demand develops.
`
`TEXAS INSTRUMENTS EX. 1009 - 183/1253
`
`

`
`4.2
`
`Display Technologies
`
`161
`
`The direct-view storage tube (DVST) is similar to the standard CRT, except that it does
`not need to be refreshed because the image is stored as a distribution of charges on the
`inside surface of the screen. Because no refresh is needed, the DVST can display complex
`images without the high scan rate and bandwidth required by a conventional CRT. The
`major disadvantage of the DVST is that modifying any par1 of an image requires redrawing
`the entire modified image to establish a new charge distribution in the DVST. This redraw
`can be unacceptably slow (many seconds for a complex image).
`The ubiquitous Tektronix 4010 display terminal, based on the DVST, was the first
`low-cost, widely available interactive graphics terminal. It was the Model T of computer
`graphics, and was so pervasive that its instruction set became a defacto standard. Even
`today, many display systems include a Tektronix-compatibility feature, so that buyers can
`continue to run their (often large) libraries of older software developed for the 4010. Now,
`however, DYSTs have been superseded by raster displays and have essentially disappeared
`from the graphics scene.
`A liquid-crystal display (LCD) is made up of si)( layers, as shown in Fig. 4.16. The
`front layer is a vertical polarizer plate. Next is a layer with thin grid wires electrodeposited
`on the surface adjoining the crystals. Next is a thin (about 0.0005-inch) liquid-crystal layer,
`then a layer with horizontal grid wires on the surface next to the crystals, then a horizontal
`polarizer, and finally a reflector.
`The liquid-crystal material is made up of long crystalline molecules. The individual
`molecules normally are arranged in a spiral fashion such that the direction of polarization of
`polarized light passing through is rotated 90". Light entering through the front layer is
`polarized vertically. As the light passes through the liquid crystal, the polarization is rotated
`90" to horizontal, so the light now passes through the rear horizontal polarizer, is reflected,
`and returns through the two polarizers and crystal.
`When the crystals are in an electric field , they all line up in the the same direction , and
`thus have no polarizing effect. Hence, crystals in the electric field do not change the
`polarization of the transmitted light, so the light remains vertically polarized and does not
`pass through the rear polarizer: The light is absorbed, so the viewer sees a dark spot on the
`display.
`
`Viewing
`direction
`
`Reflective Horizontal
`layer
`polarizer
`
`Horizontal
`grid wires
`
`liquid·
`orystal
`layer
`
`Vertical
`grid
`wires
`
`Vertical
`polarizer
`
`Fig. 4. 1 6 The layers of a liquid-crystal display (LCD), all of which are sandwiched
`together to form a thin panel.
`
`TEXAS INSTRUMENTS EX. 1009 - 184/1253
`
`

`
`162 Graphics Hardware
`
`A dark Spot at point (x1, y1) is created via matrix addrt!ssing. The point is selected by
`applying u negative voltage -V to the horizontal grid wire x1 and a positive voltage + V to the
`vertical grid wire y1: Neither -V nor+ Vis large enough to cause the crystals to line up, but
`their difference is large enough to do so. Now the crystals at (x1, y1) no longer rotate the
`direction of polarization of the transmitted light, so it remains vertically polarized and does
`not pass through the rear polarizer: The light is absorbed, so the viewer sees a dark spot on
`the display.
`To display dots at (x1, y1) and (x,. yJ, we cann01 simply apply the positive voltage to x1
`and x, and the negative voltage to y1 and y1: that would cause dots to appear at (x~o y1),
`(x1• yJ, (x,, y1) , and (x,, yJ, because all these points will be affected by the voltage. Rather,
`the points must be selected in succession, one after the other, and this selection process must
`be repeated, to refresh the activation of each point. Of course. if y1 = y2, then both points
`on the row can be selected at the same time.
`The display is refreshed one row at a time, in raster-scan fashion. Those points in a row
`that are .. on" (i.e., dark in the case of a black-on-white LCD display) are selected only
`about liN of the time, where N is the number of rows. Fonunate.ly, ooce the crystals are
`lined up. they stay that way for several hundred milliseconds, even when the voltage is
`withdrawn (the crystals' equivalent of phosphors' persistence). But even so, the crystal is
`not switched on all the time.
`Acth-e matrix panels are LCD panels that have a transistor at each (x, y) grid point. The
`ll1lnsistors are used to cause the crystals to change their state quickly, and also to control the
`degree to which the state has been changed. These two properties allow LCDs to be used in
`miniature television sets with continuous-tone images. The crystals can also be dyed to
`provide color. Most important, the transistor can serve as a memory for the state of a cell
`and can hold the cell in that State until it is cbanged. That is, the memory provided by the
`ll1lnsistor enables a ceU to remain on all the time and hence to be brighter than it would be if
`it had to be refreshed periodically. Color LCD panels with resolutions up to 800 by 1000 on
`a 14-inch diagonal panel have been built.
`AdvantagesofLCDs are low cost,low weight, small size, and low power consumption.
`In the past, the major disadvantage was that LCDs were passive, reflecting only incidem
`light and creating no light of their own (although this could be corrected with back.lighting):
`Any glare washed out the image. In recent years, use of active panels has removed this
`concern.
`Nonactive LCD technology has been adapted to color displays, and is sold commercial(cid:173)
`ly as the Tektronix liquid-crystal shutter (LCS). The LCS. placed in front of a standard
`black-and-white CRT, consistS of three la)-ers. The back layt.'f, closest to the CRT, is a
`venical polarizer. to polarize light emitted from the CRT. The layer also has a thin coating
`of a transparent conducting material. The next layer is the liquid crystal, and the third
`(front) layer is a color polarizer that transmits vertically polarized light as red and
`horizontally polarized light as green. This front layer also has a thin coating of the
`transparent conductor. If the crystals are in their normal state, they rotate the polarization
`plane by 90", so the light is horizontally polarized as it approaches the color polarizer of the
`third layer, and is seen as green. If the appropriate \'Oitage is applied to the conductive
`coatings on the front and back layers. then the crystals line up and do not affect the venical
`polarization. so the light is seen as red.
`
`TEXAS INSTRUMENTS EX. 1009 - 185/1253
`
`

`
`4 .2
`
`Display Technologies
`
`163
`
`The crystals are switched back and forth between their t\\0 states at a rate of 60 Hz. At
`the same time and in synchrony, images to be seen as red and green are alternated on the
`monochrome display. Mixtures of red and green are created by intensifying the same spot
`during the red and green phases, potentially with different intensities.
`The LCS is an alternative to the shadow-mask CRT, but has limited color resolution. It
`is possible, however, that this technology can be extended to \\Ork with three colors. If it
`can be, the shadow mask will no longer be a limiting factor in achieving higher-resolution
`full-color displays. Spot size and bandwidth will be the major determinants, as with
`monoch.rome CRTs. Eliminating the shadow mask also will increase ruggedness. Because
`LCD displays are small and light, they can be used in head-mounted displays such as that
`discussed in Section 8. 1.6.
`The plasma panel is an array of tiny neon bulbs. Each bulb can be put into an ''on"
`(intensified) state or an "off" state, and remains in the state until explicitly changed to the
`other. This memory property means that plasma panels need not be refreshed. Plasma
`panels typically have 50 to 125 cells per inch, with a I 0- to 15-inch diagonal, but 40- by
`40-inch panels with 50 cells per inch are sold commercially, and even larger and denser
`panels can be custom-made.
`The neon bulbs are not discrete units, but rather are part of a single integrated panel
`made of three layers of glass, as seen in Fig. 4. 17. The inside surface of the front layer has
`thin vertical strips of an electrical conductor; the center layer has a number of boles (the
`bulbs), and the inside surface of the rear layer has thin horizontal strips of an electrical
`conductor. Matrix addressing is used to turn bulbs on and off. To turn on a bulb, the system
`adjusts the voltages on the corresponding lines such that their difference is large enough to
`pull electrons from the neon molecules, thus firing the bulb and making it glow. Once the
`glow starts, a lower voltage is applied to sustain it. To tum off a bulb, the system
`momentarily decreases the voltages on the appropriate lines to less than the sustaining
`
`of •0:
`....
`::;::::
`·===--
`.0!900
`;:oo
`•••
`
`0
`
`Glass
`plate
`with
`cens
`
`6 7
`
`5
`4
`23
`t Vertical
`grid
`wires
`(x address)
`
`1
`2
`3
`4
`5
`6
`7
`Horizontal
`grid
`wires
`(y address)
`
`VIeWing
`direction
`
`Fig. 4 .17 The layers of a plasma display. all of which are sandwiched together to form
`a thin panel.
`
`TEXAS INSTRUMENTS EX. 1009 - 186/1253
`
`

`
`164 Graphics Hardware
`
`voltage. Bulbs can be turned on or off in about 15 microseconds. Ln some panel designs, the
`individuul cells are replaced with an open cavity, because the neon glow is contained in a
`localized area. In this case, the front and back glass layers are separated by spacers. Some
`plasma panels can also display multiple gray levels.
`The plasma panel has the advantages of being Oat, transparent, and rugged, and of not
`needing a bitmap refresh buffer. It can be used with a rear-projection system to milt
`photographic slides as static background for computer-generat.ed dynamic graphics, but has
`found most use in military applications. where small size and ruggedness are important.
`However, its cost, although continually decreasing. is still relatively high for its limited
`resolution. Color has been demonstrated in the laboratory, but is not commercially
`available.
`Electrolllminescent (EL) displays consist of the same gridlike structure as used in LCD
`and plasma displays. Between the front and back panels is a thin (typically 500-nanometcrs)
`layer of an electroluminescent material, such as zinc sulfide doped with manganese, that
`emits light when in a high electric field (about 10' volts per centimeter). A point on the
`panel is illuminated via the matrix-addressing scheme, several hundred \'OilS being placed
`across the horizontal and vertical selection Lines. Color electroluminescent displays are also
`available.
`These displays are bright and can be switched on and off quickly, and transistors at each
`pixel can be used to store the image. Typical panel sizes are 6 by 8 inches up to 12 by 16
`inches. with 70 addressable dots per inch. These displays' m;Uor disadvantage is that their
`power consumption is higher than that of the LCD panel. However, their brightness has led
`to their usc in some portable computers.
`Electrophoretic displays use positively charged colored panicles suspended in a solution
`of a contrasting color, sealed between twO parallel, closely spaced plates that have
`matrix-addressing selection lines. A negative voltage on the front selection line and a
`positive voltage on the rear selection line pulls the colored panicles toward the front plate,
`where they are seen instead of the colored liquid. Reversing the voltages pulls the panicles
`towa.rd the rear plate, so the colored liquid is seen. The display has a memory: The particles
`stay where they have been placed until moved explicitly.
`Most large-screen displays use some fonn of projection CRT. in which the light rrom a
`small (several-inch-diameter) but very bright monchrome CRT is magnified and projected
`from a curved mirror. Color systems use three projectors with red, green, and blue filters.
`A shadow-mask CRT does not create enough light to be projected onto a large
`(2-meter-diagonal) screen.
`The GE light-•'tll•<e proj«tion S}'Stl'm is used for very large screens, where the light
`output from the projection CRT II.'Ould not be sufficient. A light vulve is just what its name
`implies: A mechanism for controlling how much light passes through a valve. The light
`source can have much higher intensity than a CRT. In the most common approach, an
`electron gun trJCes an image on a thin oil film on a piece or glass. The electron charge
`causes the film to vary in thickness: A negatively charged area of the film is "stretched
`out,'' as the electrons repel one another, causing the film to become thinner. Light from the
`high-intensity source is directed at the glass, and is rerraeted in different directions because
`of the variation in the thickness of the oil film. Optics involving Schlieren bars and lenses
`project light that is refracted in certain directions on the screen. while other light is not
`
`TEXAS INSTRUMENTS EX. 1009 - 187/1253
`
`

`
`4.3
`
`Raster-scan Display Systems
`
`165
`
`TABLE 4 .3 COMPARISON OF DISPLAY TECHNOLOGIES
`Electro-
`luminescent
`fair-good
`good
`excellent
`excellent
`good-excellent
`excellent
`good
`good
`fair
`good
`good
`medium-high
`
`power consumption
`screen size
`depth
`weight
`ruggedness
`brightness
`addressability
`contrast
`intensity levels per dot
`viewing angle
`color capabiliry
`relative cost ranse
`
`CRT
`fair
`excellent
`poor
`poor
`fair-good
`excellent
`good-excellent
`g~xcellent
`excellent
`excellent
`excellent
`low
`
`Liquid
`crystal
`excellent
`fair
`excellent
`excellen.t
`excellent
`fair-good
`fair-good
`fair
`fair
`poor
`good
`low
`
`Plasma
`eanel
`fair
`excellent
`good
`excellent
`excellent
`excellent
`good
`good
`fair
`good~cellent
`fair
`high
`
`projected. Color is possible with these systems, through use of either three projectors or a
`more sophisticated set of optics with a single projector. More details are given in
`[SHER79]. Other similar light-valve systems use LCDs to modulate the light beam.
`Table 4.3 summarizes the characteristics of the four major display technologies. The
`pace of technological innovation is such, however, that some of the relationships may
`change over the next few years. Also, note that the liquid-crystal comparisons are for
`passive addressing; with active matrix addressing; gray levels and colors are achievable.
`More detailed information on these display technologies is given in [APT85; .BALD85;
`CONR85; PERR85; SHER79; and TANN85].
`
`4.3 RASTER-SCAN DISPLAY SYSTEMS
`
`The basic concepts of raster graphics systems were presented in Chapter I , and Chapter 2
`provided further insight into the types of operations possible with a raster display. ln this
`section, we discuss tbe various elements of a raster display, stressing two fundamental ways
`in which various raster systems differ one from another.
`First, most raster displays have some specialized hardware to assist in scan converting
`output primitive