BLM5207

Computer Organization
Prof. Dr. Nizamettin AYDIN
[email protected]
http://www3.yildiz.edu.tr/~naydin

Data Formats

1
 Data Formats
• Computers
– Process and store all forms of data in binary format
• Human communication
– Includes language, images and sounds
• Data formats:
– Specifications for converting data into computer-
usable form
– Define the different ways human data may be
represented, stored and processed by a computer

2
 Sources of Data
• Binary input
– Begins as discrete input
• Example: keyboard input such as A 1+2=3 math
– Keyboard generates a binary number code for each key

• Analog
– Continuous data such as sound or images
• Requires hardware to convert data into binary numbers

Computer
Human form: data Input
abcde1453 device Computer representation:
1101000101010101…

3
 Common Data Representations
Type of Data Standard(s)
Alphanumeric Unicode, ASCII, EDCDIC
Image (bitmapped) GIF (graphical image format)
TIF (tagged image file format)
PNG (portable network graphics)
Image (object) PostScript, JPEG, SWF (Macromedia
Flash), SVG
Outline graphics and fonts PostScript, TrueType
Sound WAV, AVI, MP3, MIDI, WMA
Page description PDF (Adobe Portable Document
Format), HTML, XML
Video Quicktime, MPEG-2, RealVideo,
WMV
4
 Internal Data Representation
• Reflects the
– Complexity of input source
– Type of processing required
• Trade-offs
– Accuracy and resolution
• Simple photo vs. painting in an art book
– Compactness (storage and transmission)
• More data required for improved accuracy and resolution
• Compression represents data in a more compact form
• Metadata: data that describes or interprets the meaning of data
• Ease of manipulation:
– Processing simple audio vs. high-fidelity sound
• Standardization
– Proprietary formats for storing and processing data (WordPerfect vs. Word)
– De facto standards: proprietary standards based on general user
acceptance (PostScript)
5
 Data Types
• Numeric:
– Used for mathematical manipulation
– Add, subtract, multiply, divide
– Types
• Integer (whole number)
• Real (contains a decimal point)
• Alphanumeric:
– Characters: b T
– Number digits: 7 9
– Punctuation marks: ! ;
– Special-purpose characters: $ &
• Numeric characters vs. numbers
– Both entered as ordinary characters
– Computer converts into numbers for calculation
• Examples: Variables declared as numbers by the programmer (Salary$ in BASIC)
– Treated as characters if processed as text
• Examples: Phone numbers, ZIP codes
6
 Alphanumeric Codes
• Arbitrary choice of bits to represent characters
– Consistency:
• input and output device must recognize same code
• Value of binary number representing character
corresponds to placement in the alphabet
– Facilitates sorting and searching
• Representing Characters
– ASCII:
• most widely used coding scheme
– EBCDIC:
• IBM mainframe (legacy)
– Unicode:
• developed for worldwide use
7
 ASCII
• Developed by ANSI (American National
Standards Institute)
• Represents
– Latin alphabet, Arabic numerals, standard punctuation
characters
– Plus small set of accents and other European special
characters
• ASCII
– 7-bit code: 128 characters

8
 ASCII Reference Table
MSD
LSD 0 1 2 3 4 5 6 7
0 NUL DLE SP 0 @ P p
1 SOH DC1 ! 1 A Q a W
2 STX DC2 “ 2 B R b r
3 ETX DC3 # 3 C S c s
4 EOT DC4 $ 4 D T d t
5 ENQ NAK % 5 E U e u
6 ACJ SYN & 6 F V f v 7416
7 BEL ETB ‘ 7 G W g w
111 0100
8 BS CAN ( 8 H X h x
9 HT EM ) 9 I Y i y
A LF SUB * : J Z j z
B VT ESC + ; K [ k {
C FF FS , < L \ l |
D CR GS - = M ] m }
E SO RS . > N ^ n ~
F SI US / ? O _ o DEL

9
 EBCDIC
• Extended Binary Coded Decimal Interchange
Code developed by IBM
– Restricted mainly to IBM or IBM compatible
mainframes
– Conversion software to/from ASCII available
– Common in archival
ASCII EBCDIC
Space 2016 4016 data
A 4116 C116 – Character codes
b 6216 8216 differ from ASCII

10
 Unicode
• Most common 16-bit form represents 65,536
characters
• ASCII Latin-I subset of Unicode
– Values 0 to 255 in Unicode table
• Multilingual: defines codes for
– Nearly every character-based alphabet
– Large set of ideographs for Chinese, Japanese and
Korean
– Composite characters for vowels and syllabic clusters
required by some languages
• Allows software modifications for local-languages
11
Unicode Assignment Table

12
 2 Classes of Codes
• Printing characters
– Produced on the screen or printer
• Control characters
– Control position of output on screen or printer
• VT: vertical tab LF: Line feed
– Cause action to occur
• BEL: bell rings DEL: delete current character
– Communicate status between computer and I/O
device
• ESC: provides extensions by changing the meaning of a
specified number of contiguous following characters

13
Control Code Definitions

14
 Keyboard Input
• Scan code
– Two different scan codes on keyboard
• One generated when key is struck and another when key is
released
– Converted to Unicode, ASCII or EBCDIC by
software in terminal or PC
• Advantage
– Easily adapted to different languages or keyboard
layout
– Separate scan codes for key press/release for multiple
key combinations
• Examples: shift and control keys
15
 Other Alphanumeric Input
• OCR (optical character reader)
– Scans text and inputs it as character data
– Used to read specially encoded characters
• Example: magnetically printed check numbers
– General use limited by high error rate
• Bar Code Readers
– Used in applications that require fast, accurate and repetitive input with
minimal employee training
• Examples: supermarket checkout counters and inventory control
– Alphanumeric data in bar code read optically using wand
• Magnetic stripe reader:
– alphanumeric data from credit cards
• Voice
– Digitized audio recording common but conversion to alphanumeric data
difficult
• Requires knowledge of sound patterns in a language (phonemes) plus rules for
pronunciation, grammar, and syntax 16
 Image Data
• Photographs, figures, icons, drawings, charts and
graphs
• Two approaches:
– Bitmap or raster images of photos and paintings with
continuous variation
– Object or vector images composed of graphical objects
like lines and curves defined geometrically
• Differences include:
– Quality of the image
– Storage space required
– Time to transmit
– Ease of modification
17
 Bitmap Images
• Used for realistic images with continuous variations in
shading, color, shape and texture
– Examples:
• Scanned photos
• Clip art generated by a paint program
• Preferred when image contains large amount of detail and
processing requirements are fairly simple
• Input devices:
– Scanners
– Digital cameras and video capture devices
– Graphical input devices like mice and pens
• Managed by photo editing software or paint software
– Editing tools to make tedious bit by bit process easier
18
 Bitmap Images
• Each individual pixel (pi(x)cture element) in a
graphic stored as a binary number
– Pixel:
• A small area with associated coordinate location
– Example:
• each point below represented by a 4-bit code corresponding
to 1 of 16 shades of gray

19
 Bitmap Display
• Monochrome:
– black or white
• 1 bit per pixel
• Gray scale:
– black, white or 254 shades of gray
• 1 byte per pixel
• Color graphics:
– 16 colors, 256 colors, or 24-bit true color (16.7
million colors)
• 4, 8, and 24 bits respectively

20
 Storing Bitmap Images
• Frequently large files
– Example: 600 rows of 800 pixels with 1 byte for each
of 3 colors  ~1.5MB file
• File size affected by
– Resolution (the number of pixels per inch)
• Amount of detail affecting clarity and sharpness of an
image
– Levels: number of bits for displaying shades of gray
or multiple colors
• Palette: color translation table that uses a code for each
pixel rather than actual color value
• Data compression
21
 GIF (Graphics Interchange Format)
• First developed by CompuServe in 1987
• GIF89a enabled animated images
– allows images to be displayed sequentially at fixed
time sequences
• Color limitation: 256
• Image compressed by LZW (Lempel-Zif-Welch)
algorithm
• Preferred for line drawings, clip art and pictures
with large blocks of solid color
• Lossless compression
22
GIF (Graphics Interchange Format)

23
JPEG (Joint Photographers Expert Group)
• Allows more than 16 million colors
• Suitable for highly detailed photographs and
paintings
• Employs lossy compression algorithm that
– Discards data to decreases file size and transmission
speed
– May reduce image resolution, tends to distort sharp
lines

24
 Other Bitmap Formats
• TIFF (Tagged Image File Format): .tif (pronounced tif)
– Used in high-quality image processing, particularly in publishing
• BMP (BitMaPped): .bmp (pronounced dot bmp)
– Device-independent format for Microsoft Windows environment:
• pixel colors stored independent of output device
• PCX: .pcx (pronounced dot p c x)
– Windows Paintbrush software
• PNG: (Portable Network Graphics): .png (pronounced ping)
– Designed to replace GIF and JPEG for Internet applications
– Patent-free
– Improved lossless compression
– No animation support

25
 Object Images
• Created by drawing packages or output from
spreadsheet data graphs
• Composed of lines and shapes in various colors
• Computer translates geometric formulas to create
the graphic
• Storage space depends on image complexity
– number of instructions to create lines, shapes, fill
patterns

• Movies Shrek and Toy Story use object images

26
 Object Images
• Based on mathematical formulas
– Easy to move, scale and rotate without losing shape
and identity as bitmap images may
• Require less storage space than bitmap images
• Cannot represent photos or paintings
• Cannot be displayed or printed directly
– Must be converted to bitmap since output devices
except plotters are bitmap

27
 Popular Object Graphics Software
• Most object image formats are proprietary
– Files extensions include .wmf, .dxf, .mgx, and .cgm
• Macromedia Flash: low-bandwidth animation
• Micrographx Designer: technical drawings to illustrate
products
• CorelDraw: vector illustration, layout, bitmap creation,
image-editing, painting and animation software
• Autodesk AutoCAD: for architects, engineers, drafters,
and design-related professionals
• W3C SVG (Scalable Vector Graphics) based on XML
Web description language
– Not proprietary
28
 PostScript
• Page description language:
– list of procedures and statements that describe each of the
objects to be printed on a page
• Stored in ASCII or Unicode text file
• Interpreter program in
computer or output
device reads PostScript
to generate image
– Scalable font support
• Font outline objects
specified like other
objects

29
 Representing Characters
• Characters stored in format like Unicode or
ASCII
– Text processed and stored primarily for content
• Presentation requirements like font stored with
the character
– Text appearance is primary factor
• Example: screen fonts in Windows
• Glyphs:
– Macintosh coding scheme that includes both
identification and presentation requirement for
characters
30
 Bitmap vs. Object Images

Bitmap (Raster) Object (Vector)

Pixel map Geometrically defined shapes

Photographic quality Complex drawings

Paint software Drawing software

Larger storage requirements Higher computational

requirements
Enlarging images produces Objects scale smoothly
jagged edges
Resolution of output limited by Resolution of output limited by
resolution of image output device

31
 Video Images
• Require massive amount of data
– Video camera producing full screen 640 x 480 pixel true color
image at 30 frames/sec  27.65 MB of data/sec
– 1-minute film clip  1.6 GB storage
• Options for reducing file size: decrease size of image,
limit number of colors, reduce frame rate
• Method depends on how video delivered to users
– Streaming video: video displayed as it is downloaded from the
Web server
• Example: video conferencing
– Local data (file on DVD or downloaded onto system) for
higher quality
• MPEG-2: movie quality images with high compression require
substantial processing capability 32
 Audio Data
• Transmission and processing requirements less
demanding than those for video
• Waveform audio: digital representation of sound
– Audio CD sampling rate = 44.1KHz
• Height of each sample
saved as:
– 8-bit number for radio-
quality recordings
– 16-bit number for high-
fidelity recordings
• 2 x 16-bits for stereo

33
 MIDI
• MIDI (Musical Instrument Digital Interface):
– instructions to recreate or synthesize sounds
• Analog sound converted to digital values by A-to-D
converter
– Music notation system that allows computers to
communicate with music synthesizers
– Instructions that MIDI instruments and MIDI sound
cards use to recreate or synthesize sounds.
• Do not store or recreate speaking or singing voices
• More compact than waveform
– 3 minutes = 10 KB

34
 Audio Formats
• MP3
– Derivative of MPEG-2 (ISO Moving Picture Experts
Group)
– Uses psychoacoustic compression techniques to
reduce storage requirements
– Discards sounds outside human hearing range: lossy
compression
• WAV
– Developed by Microsoft as part of its multimedia
specification
– General-purpose format for storing and reproducing
small snippets of sound 35
.WAV Sound Format

36
 Data Compression
• Compression: recoding data so that it requires fewer
bytes of storage space.
• Compression ratio: the amount file is shrunk
• Lossless: inverse algorithm restores data to exact
original form
– Examples: GIF, PCX, TIFF
• Lossy: trades off data degradation for file size and
download speed
– Much higher compression ratios, often 10 to 1
• Example: JPEG
– Common in multimedia
• MPEG-2: uses both forms for ratios of 100:1
37
 Compression Algorithms
• Repetition
–0587000034000 01587043403
– Example: large blocks of the same color
• Pattern Substitution
– Scans data for patterns  Pe  pi  ed

– Substitutes new pattern,  er  ck  pe

 Pi
makes dictionary entry
• Example: 45 to 30 bytes plus dictionary
– Peter Piper picked a peck of pickled peppers.
–  t   p    a   of  l   pp  s.

38
 Internal Computer Data Format
• All data stored as binary numbers
– Interpreted based on
• Operations computer can perform
• Data types supported by programming language used to create application
• Simple Data Types
– Boolean:
• 2-valued variables or constants with values of true or false
– Char:
• Variable or constant that holds alphanumeric character
– Enumerated:
• User-defined data types with possible values listed in definition
– Type DayOfWeek = Mon, Tues, Wed, Thurs, Fri, Sat, Sun
– Integer:
• positive or negative whole numbers
– Real :
• Numbers with a decimal point, whose magnitude (large or small) exceeds
computer’s capability to store as an integer
39
Representing Numbers - 32-bit Data Word

40
 Unsigned Numbers: Integers
• Unsigned whole number or integer
• Direct binary equivalent of decimal integer
4 bits: 0 to 9 16 bits: 0 to 9,999
8 bits: 0 to 99 32 bits: 0 to 99,999,999
Decimal Binary BCD
68 = 0100 0100 = 0110 1000
= 26 + 22 = 64 + 4 = 68 = 22 + 2 1 = 6 23 = 8
99 = 0110 0011 = 1001 1001
(largest 8-bit = 26 + 25 + 2 1 + 2 0 = = 23 + 2 0 23 + 20
BCD)
= 64 + 32 + 2 + 1 = 99 = 9
9
255 = 1111 1111 = 0010 0101 0101
(largest 8-bit
binary) = 28 – 1 = 255 = 21 22 + 20 22 + 20
= 2 5 5
41
 Value Range: Binary vs. BCD
• BCD range of values < conventional binary representation
– Binary: 4 bits can hold 16 different values (0 to 15)
– BCD: 4 bits can hold only 10 different values (0 to 9)
No. of Bits BCD Range Binary Range
4 0-9 1 digit 0-15 1+ digit
8 0-99 2 digits 0-255 2+ digits
12 0-999 3 digits 0-4,095 3+ digits
16 0-9,999 4 digits 0-65,535 4+ digits
20 0-99,999 5 digits 0-1 million 6 digits
24 0-999,999 6 digits 0-16 million 7+ digits
32 0-99,999,999 8 digits 0-4 billion 9+ digits
64 0-(1016-1) 16 digits 0-16 quintillion 19+ digits

• Binary representation generally preferred

– Greater range of value for given number of bits
– Calculations easier
• BCD often used in business applications to maintain decimal rounding and
decimal precision
42
 Signed-Integer Representation
• 2’s Complement n 2
 2 n 1
an  1   2i ai
+3 = 00000011 +2 = 00000010 i 0
+1 = 00000001 +0 = 00000000
-1 = 11111111 -2 = 11111110 -3 = 11111101

43
Floating Point Representation (IEEE-754 fp)
s biased exp. fraction
32 bits: 1 8 bits 23 bits

N = (-1)s x 1.fraction x 2(biased exp. – 127)

• Sign: 1 bit
• Mantissa: 23 bits
– We “normalize” the mantissa by dropping the leading 1 and
recording only its fractional part
• Exponent: 8 bits
– In order to handle both +ve and -ve exponents, we add 127 to
the actual exponent to create a “biased exponent”:
• 2-127 => biased exponent = 0000 0000 (= 0)
• 20 => biased exponent = 0111 1111 (= 127)
• 2+127 => biased exponent = 1111 1110 (= 254)
44