Nothing Special   »   [go: up one dir, main page]

F3.5 CRT and AMLCD Comparison

Download as pdf or txt
Download as pdf or txt
You are on page 1of 9

Comparison of CRT with AMLCD

1 Introduction
AMLCD technology is the only full-colour display technology sufficiently well
developed to seriously challenge the dominance of the Shadowmask CRT and, indeed,
it has now replaced the CRT as the display medium of choice for many applications. It
is appropriate to give more detailed consideration to the advantages and disadvantages
of each type. In the following discussion, comparisons are made between a six-inch
Taut-Shadowmask periodic-focus CRT and a state-of-the-art RGB vertical stripe six-
inch AMLCD with a colour group pitch of 0.254mm (100 colour groups per inch).
Considerations of cost, size/weight and image viewability are considered, with the latter
concentrating on the key issues of brightness, contrast, colour range and resolution. The
comparison concentrates only upon the performance of the two display types when
displaying raster images. If the information content is suitable and the additional
complexity can be justified, the high-brightness stroke mode of the Shadowmask
display is still superior to the image given by an AMLCD in character formation,
brightness and contrast.

2 Cost
The AMLCD is a highly complex device; however, huge production investment for the
commercial and domestic markets has brought the cost of devices down to well below
that of the specialist-market Taut Shadowmask CRT. The emergence of good-quality
high-volume AMLCD units makes further development of CRT technology quite
unlikely. In addition the AMLCD is more power-efficient than the Shadowmask CRT
(though it requires heating at low temperatures; both types of unit need cooling at high
temperatures). The lack of high-voltage drive circuits result in reduced cost of
peripheral circuitry and greater reliability of what is now a mature product. To
summarise, AMLCD technology is now less expensive than Shadowmask, both in cost
of purchase and in cost of ownership.

3 Size/weight
The AMLCD should have a clear advantage over the Shadowmask in this area. With its
complex, high-powered periodic-focus gun, the Shadowmask CRT is quite long and
will not fit into some cockpits. To provide sufficient ruggedness for airborne use,
together with the ability to withstand the immense tension of the mask, the Taut
Shadowmask CRT is heavy and bulky in comparison to an AMLCD. In addition the
need for EHT circuits leads to a further increase in volume and weight for the complete
display.

4 Brightness
For good viewability in the cockpit under daylight conditions, high brightness is of
critical importance. For the Taut-Shadowmask CRT, the problems associated with
Shadowmask inefficiency make it necessary for the three CRT electron guns each to

QinetiQ/SES/ETPS/GS/F3.5/V1 Page 1 of 9
©QinetiQ Ltd 2006
Comparison of CRT with AMLCD

generate a high-current beam in order to achieve adequate brightness and this requires
careful design of the guns and the focussing arrangements. For the Shadowmask CRT,
each phosphor colour occupies only about 18% of the screen area; this means that over
80% of the energy used in electron beam generation is wasted. A White luminance of
about 200 ft Lamberts (690 cd/m2) in raster mode, with contrast enhancement filter in
place, is achievable. For typical colour group arrangements, each primary colour
occupies about 30% of the screen area of an AMLCD and this might at first sight appear
to give an advantage to the AMLCD. However, the presence of opaque transistors at
each photosite, coupled with the light-absorbing effect of diffusers, polarizers, dye
layers, electrodes and multiple reflecting internal surfaces make the AMLCD an
extremely inefficient transmitter of light. The backlight performance is therefore
critical. Cold-cathode backlights with luminances of 20,000 cd/m2 are available; with
judicious use of collimating reflectors this can provide a luminance normal to the
display surface in the region of 300 ft.L (1000 cd/m2) for a state-of-the-art display,
providing a brighter, more easily viewable image than can be generated by the Taut
Shadowmask CRT in raster mode. Note that at the slower writing speeds used in stroke
mode, a Taut Shadowmask CRT can produce symbology of luminance in excess of 700
ft.L (2450 cd/m2.

5 Contrast
5.1 For good viewability, it is essential that the observer is able to clearly distinguish
between regions of the display which are black (unenergized in the case of the CRT)
and those which are bright (excited by the CRT guns). As detailed earlier in these notes,
a measure of this capability is the Contrast Ratio, which is defined as:

Luminance of symbology
Luminance of background

5.2 CRT contrast: For a CRT the luminance of the energized phosphor is the sum of the
energization light plus the reflected light from ambient illumination, the definition is
more usually expressed as:

(Energization luminance + Background luminance)


Background luminance

It can be seen from this expression that for high contrast, it is important for the
energization luminance ('CRT Brightness') to be high, but equally important for the
background luminance to be low. As the CRT phosphor is white and reflects a high
proportion of the ambient light falling upon it, it is necessary for a contrast
enhancement filter to be placed between CRT face and observer, to reduce the visibility
of the unenergized phosphor.

5.3 Shadowmask CRT contrast: For the Shadowmask CRT it is possible to maximise the
attenuation of ambient light whilst minimising the attenuation of the CRT light by using
a 'Triple Notch' wavelength-selective filter. This type of filter is designed to have a low
attenuation at the (three) wavebands of the CRT red, green and blue phosphors whilst
having a high attenuation to other wavelengths of light. Thus the broadband ambient

Page 2 of 9 QinetiQ/SES/ETPS/GS/F3.5/V1
Comparison of CRT with AMLCD

light is severely attenuated as it passes from the outside world through the filter to the
CRT screen, bounces off the phosphor and again passes through the filter to reach the
observer's eye, whilst the narrowband light from the energized phosphor is minimally
attenuated and only has to make one pass through the filter material. Additionally, in
order to minimise the number of reflective surfaces, the filter is bonded to the glass of
the CRT screen. Using such techniques a contrast ratio of about 3.2:1 is achievable.

5.4 AMLCD contrast: With sunlight shining directly on to the display surface the AMLCD
symbology also undergoes a loss of contrast through illumination of the dark areas of
the display. However, in addition to having brighter symbology to start with, the
AMLCD does not suffer to the same extent as the CRT. This is because some of the
external illumination finds its way through the LC cell to the diffuser, from where it is
reflected back through the display, effectively reinforcing the light from the backlight
and adding to the luminance of the symbology. In practice, an AMLCD contrast ratio of
6:1 is achievable under sunlight illumination conditions – almost twice as good as for
the best CRT technology.

6 Colour range
Under zero ambient illumination, the range of colours that can be exhibited by the
Shadowmask CRT is determined by the emission wavelengths of the three phosphor
types deposited on the CRT screen. The phosphor materials are selected to give a good
balance between phosphor efficiency and colour saturation. Although the circular CRT
spot, with a diameter of 0.5mm (taken at the 50% brightness contour), is significantly
larger than the Shadowmask pitch of 0.24mm, the presence of the Shadowmask ensures
that each electron gun can stimulate only its intended phosphor, thus maintaining the
purity of the colours. This is shown diagrammatically in Figure 6-1. A further difficulty
arises for the Shadowmask CRT, in the presence of strong ambient illumination. The
(white) ambient light reflected off the (white) phosphor surface of the CRT screen
causes significant desaturation of the image. The colour gamut (or range of colours)
available from the CRT is illustrated on chromaticity charts in Figure 6-2 and Figure
6-3, for conditions of zero ambient illumination and full sunlight. The colours that can
be reproduced lie within the triangular shapes shown. For comparison, the colour gamut
of a typical AMLCD is also illustrated.

QinetiQ/SES/ETPS/GS/F3.5/V1 Page 3 of 9
Comparison of CRT with AMLCD

Figure 6-1; Shadowmask CRT geometry

Figure 6-2; Colour gamuts taut shadow mask and typical AMLCD
zero ambient illumination

Page 4 of 9 QinetiQ/SES/ETPS/GS/F3.5/V1
Comparison of CRT with AMLCD

Figure 6-3; Colour gamuts taut shadow mask and typical AMLCD
105 lux ambient illumination

7 Resolution
The resolution, or ability of a CRT to delineate fine detail in an image, is critically
important to the viewability and acceptability of the image. Two CRT parameters affect
the resolution; these are the spacing of the phosphor dots and the size of the spot
generated by the CRT beam. A third parameter, that of beam spot profile, has a bearing
on the 'softness' of the image; this affects the acceptability or otherwise of unwanted
alias components introduced into the image by interference between image and
phosphor spatial frequencies.

8 Aliasing
8.1 Because the screen of a Shadowmask CRT or an AMLCD is not homogenous but is a
mass of small dots or pixels, an image can be displayed only where there is a phosphor
dot or pixel to display it. The screen structure therefore acts as a sampling device, and a
continuous image being presented for display is effectively sliced up into lots of tiny
samples. This is of particular concern when a symbol is being drawn in a primary
colour, as only one of every three phosphor dots or pixels is able to contribute to the
image. The effect of this sampling action becomes important when the image to be
displayed contains features of a size or spacing that is small enough to be compared to
the spacing between the phosphor elements. When such an image is sampled by the
screen, the image spatial frequencies are modified by the spatial frequency of the screen
structure and new spurious or "alias" frequencies are generated which may corrupt the
original image. It can be shown mathematically that when the spatial frequencies
contained in the image approach or exceed half the spatial frequency of the sampling

QinetiQ/SES/ETPS/GS/F3.5/V1 Page 5 of 9
Comparison of CRT with AMLCD

process, alias components will be generated. For any image that is to be sampled there
is therefore a critical upper frequency limit, known as the 'Nyquist frequency' and all
image components must be below the Nyquist frequency to avoid corruption. An
extreme example that illustrates the problem very clearly is the display of a sinusoidal
symbol in primary green on a high performance AMLCD with an RGB vertical stripe
arrangement and a pixel pitch of 0.085mm (100 colour groups per horizontal inch), as
illustrated in Figure 8-1. If the sinusoid is well stretched out (low spatial frequency) all
is well, but if the peaks are too close together for the pixel structure of the AMLCD
severe corruption of the symbol results. In the extreme example of Figure 8-1, the
sinusoid peaks are just over 0.28mm apart, much closer than the minimum size of
0.51mm allowed by the Nyquist limit and the resulting image bears little resemblance to
the original symbol.

Figure 8-1; Aliasing by screen structure

8.2 The above example relates to the sampling performance of the LC panel itself and does
not take account of any anti-aliasing circuitry or signal-processing software that might
be included within the display drive circuitry. Coincidentally, the example is also
representative of a high-performance Beam-Index CRT. Unfortunately, this sampling
effect takes place at the output of the device (ie at the screen surface) and the sampling
frequency cannot therefore be filtered out. The best that can be done is to ensure that the
signal input to the sampling structure only contains signal frequencies that are well
below the Nyquist limit.

8.3 In the airborne environment, serious examples of alias effects can occur when an image
containing small details (for example small characters on a Map display) interferes with
the phosphor structure of a CRT screen or the pixel structure of an RGB vertical stripe
AMLCD. The effect of this interference is particularly evident when displaying features
in a primary colour (red, green, blue) as only one phosphor dot or AMLCD pixel in
each group of three is being used. An example is shown for a display of a 3.5mm high
character ‘R’ in a primary colour on a Taut Shadowmask CRT and on an AMLCD, in
Figure 8-2. In the worst case, if the character is rotated so that the leg of the ‘R’ is
vertical the character might be mistaken for a ‘P’ on the AMLCD because of the stripe

Page 6 of 9 QinetiQ/SES/ETPS/GS/F3.5/V1
Comparison of CRT with AMLCD

structure of the screen. It is therefore necessary to limit the bandwidth of the video by
electrical or software filtering in the AMLCD’s drive circuitry, thus ensuring that image
frequencies greater than half the stripe frequency do not occur. For a 625-line input
video standard, a video cut-off frequency of less than 5.77 MHz is required (the
standard 625-line bandwidth is 15 MHz). One possible method of achieving this
filtering in the drive circuitry of an AMLCD is to simulate a CRT-like Gaussian
intensity profile across a small matrix of colour groups.

8.4 For the Taut Shadowmask CRT, the nominal spot size of 0.5mm is significantly greater
than the pitch between phosphor dots of the same colour. The spot size is therefore the
dominant mechanism for determining the resolution of the CRT. If a series of closely-
spaced parallel vertical lines is drawn on the CRT screen and the spacing between the
lines in the horizontal direction is gradually diminished, the lines appear gradually to
merge with one another as the shoulder of one line adds with the shoulder of its
neighbour to fill in the dark space between the lines. This effect is illustrated in Figure
8-3, and serves to reduce the amplitude of the high-frequency image modulation, thus
lessening the severity of the aliasing. For 625-line video inputs, the video frequency at
which aliasing would become apparent is about 6.9 MHz; however the Gaussian spot
profile is equivalent to a filter operating at a frequency of roughly 6.3 MHz.

8.5 The horizontal resolution of the Shadowmask CRT is therefore possibly a little better
than that of the AMLCD. However, the performance of the Shadowmask CRT is
sometimes not uniform across the entire screen, being worse at the corners and better in
the centre because of variations in the spot size and the possible presence of
misconvergence. Additionally, unless precautions are taken to control the spot size as
the CRT luminance is varied, the CRT spot may shrink as the CRT is dimmed, causing
problems of aliasing in dim ambient conditions. In the vertical direction, the
Shadowmask Gaussian spot profile can be used to mitigate the effects of the raster line
structure (for a 625-line picture, the raster lines are about 0.26mm apart on 6-inch CRT
screens). The RGB vertical stripe AMLCD used as the comparison example has a major
advantage over the CRT in the vertical direction as the pitch between colour groups is
0.085mm, the same as the pixel pitch.

QinetiQ/SES/ETPS/GS/F3.5/V1 Page 7 of 9
Comparison of CRT with AMLCD

Figure 8-2; Effect of screen structure on 3.5mm character

Page 8 of 9 QinetiQ/SES/ETPS/GS/F3.5/V1
Comparison of CRT with AMLCD

Figure 8-3; Merging of gaussian lines

9 Summary of CRT-AMLCD comparison


In summary, each of the two technologies has strengths and weaknesses. In the areas of
weight, size, power consumption and cost the AMLCD scores over its Shadowmask
rival, particularly as the technology continues to mature. In the area of raster video
image quality also, the AMLCD has now overtaken the CRT in the performance race,
with a brighter image and with better contrast control. However, the Taut Shadowmask
CRT is still unbeatable in its ability to display high-brightness stroke symbology of
superb quality. Many assessors would agree that the quality of symbology from either
the Taut Shadowmask or the AMLCD is adequate for use within transport aircraft and
even within fast-jet bubble-canopy cockpits. Neither technology can yet be described as
‘good’ or ‘without problems’ under all viewing conditions. However, if the stroke mode
of the CRT is discounted, AMLCD technology now offers a superior overall viewing
performance and is a clear winner in size, weight and cost-of-ownership.

QinetiQ/SES/ETPS/GS/F3.5/V1 Page 9 of 9

You might also like