CS9211 datasheet(29/62 Pages) NSC

29 / 62 page

Revision 2.1

www.national.com

Functional Description (Continued)

3.2.6

Dithering

Dithering creates intermediate color intensities by mixing

available colors. Human vision sees an average of the

intensities of adjacent pixels on a screen. Although dither-

ing provides additional shades, it does so by sacrificing

spatial resolution.

3.2.6.1

Theory Of Dithering

The number of colors that a given panel displays can be

enhanced beyond the intensity combinations generated by

frame rate modulation by way of a technique called dither-

ing. The drawback is that fine spatial details are lost in this

process, and boundaries between regions of differing color

intensities become blurred.

For example, consider just the red color component of a

2x2 square of pixels. If the only two options for the red color

component were to be turned on or off, there would be only

two colors, black and the brightest red. However, if two of

the pixels’ red color components in the 2x2 square were

turned on and two were turned off, the human eye would

blend these adjacent pixels and the 2x2 pixel square would

appear to be half as bright as the brightest red.

This process is illustrated in Figure 3-9. Suppose each

pixelina2x2 square had6bits of dataassociated with it.

The frame rate modulator is using the upper four most sig-

nificant bits, so the lower two bits would be lost or truncated

without the support of the dithering process. Consider the

arbitrary 6-bit pixel value 38h = 11_1000; the upper four

bits of 38h are 1110, which in hex is “E”. Without dithering,

pixel values 39h (11_1001), 3Ah (11_1010), and 3Bh

(11_1011) would all be displayed the same as pixel value

38h (11_1000), since the upper four bits are the same for

each value (“E”). Since pixel value 3Ch (11_1100) has a

different set of upper four bits (1111 instead of 1110), 3Ch

would appear brighter than 38h. So, without dithering, it

would seem that the panel could accurately display only

pixel values 38h and 3Ch. When the two LSBs are

removed, these become values Eh and Fh, respectively.

Dithering provides a means of displaying the “missing” val-

ues 39h, 3Ah, and 3Bh, by displaying combinations of the

values the panel is able to display in a 2x2 square. The

average intensity of the pixels in the 2x2 square becomes

the intensity of the ‘missing’ values, as illustrated in Figure

3-9. In order to leave room at the top of the intensity scale,

value 3Bh is passed through unchanged, and values 38h,

39h, and 3Ah are modified by the dither algorithm.

One of four dither patterns are chosen by the two LSBs,

bolded below the “Case n” text in Figure 3-9. A zero in the

dither pattern (middle column of Figure 3-9) indicates the

input value will be passed through unchanged. A one in the

dither pattern indicates the displayed value should be dec-

remented to the next available intensity value.

In Case 1) of Figure 3-9, all four pixels want to be value

38h, which is no problem for the panel since 38h (or Eh) is

one of the values it can display directly. However, the dither

pattern contains three ones, so three of the pixels in this

square are dithered down to the next available brightness,

“Dh”. In Case 2), all four pixels want to be intensity 39. Two

pixels are dithered down to intensity “Dh”, and two are

passed through unchanged as Eh. In Case 3) , selected by

dither bits 10, only one pixel is dithered down to brightness

Dh, and the other three pass through unchanged. In Case

4), the dither pattern contains all zeros, so the value Eh is

passed through unchanged for all four pixels in the 2x2

square. Moving from Case 1 to Case 4, one less pixel is

dithered down in each case In Case 5), the sequence

begins again with the next-brightest intensity, Eh, being the

one that is dithered-down to.

3.2.6.2

Pre-Programmed Dither Patterns (ROM)

Theexample discussedwithreference toFigure3-9 is 2-bit

dithering. In 2-bit dithering, four patterns are used, as

shown in Cases 1-4.

Figure 3-9. Effect of Two-Bit Dithering

Cases1-4 canberedrawn as asinglepicture.Refer to the

dark outlined 2x2 box contained within the 8x8 pixel pattern

on the “2-bit scheme” in Figure 3-10. The numbers in the

pixels indicate the value of the lower two bits: 00 = blank,

01 = “1”, 10 = “2”, and 11 = “3”. When the value is “00”, only

the pixel shown as blank will retain the input color intensity.

The other three pixels will be decremented to the next

available intensity value. As the lower two bits of any inten-

sity value increase from 00 to 01, the pixel labeled “1” will

retain the input value and the other two will be decre-

mented to the next available intensity. When the lower two

bits are 10 (“2”), then the pixel labeled “2” will also retain

the input value, and the remaining one pixel will be decre-

mented to the next available intensity. When the lower two

bits are 11 (“3”), then all four pixels retain the input value.

39h 39h

39h

3Ah 3Ah

3Ah

3Bh 3Bh

3Bh

3Ch 3Ch

3Ch

38h 38h

38h

Input

display

Case 1)

Case 2)

Case 3)

Case 4)

Case 5)

38h

39h

3Ah

3Bh

3Ch

dither

pattern

Actual

display

“00”

“01”

“10”

“11”

“00”

CS9211 Datasheet(PDF) 29 Page - National Semiconductor (TI)

CS9211 Datasheet(HTML) 29 Page - National Semiconductor (TI)

Html Pages

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62

Datasheet Download

Link URL