Rui Wang, Assistant professor Dept. of Information and Communication Tongji University.

Transcription

1 Data Representation ti and Arithmetic for Computers Rui Wang, Assistant professor Dept. of Information and Communication Tongji University it

2 Questions What do you know about other number system? (e.g. Decimal base 10) What kind of number system is used by the computer? Why? 2

3 Positional Number Systems Different Representations of Natural Numbers 27 decimal number base-2 or binary number Examples: (11011) 2 = = 27 (2103) 4 = = 147 3

4 The binary system is also called the base-2 system. ( ) 011) Our decimal system is the base-10 system. It uses powers of 10 for each position in a number. (975.3) Any integer quantity can be represented exactly using any base. (3077 octal or 2BAD hex) ) 4

5 Weighted Position Code The base of a number system defines the range of possible values that a digit may have: 0 9 for decimal; 0,1 for binary. The general form for determining the decimal value of a number is given by: Example: = = (500) 10 + (40) 10 + (1) 10 + (2/10) 10 + (5/100) 10 = (541.25) 10 5

6 Binary representation Why is binary representation used by computer? 1 1=11 Easy to implement 0 0=0 Simple computation rules 1 0=0 1=

7 Binary Numbers Each binary digit (called bit) is either 1 or 0 Bits have no inherent meaning, can represent Unsigned and signed integers Characters Floating-point numbers Images, sound, etc

8 A bit is the most basic unit of information in a computer. It is a state of on or off in a digital circuit. Sometimes they represent high or low voltage A byte is a group of eight bits. It is the smallest possible addressable unit of computer storage. 8

9 A word is a contiguous group of bytes. Words can be any number of bits or bytes. Word sizes of 16, 32, or 64 bits are most common. 9

10 Octal & Hexadecimal numbers Decimal Binary Octal Hexadecimal A B C D E F 10

11 Converting base-r to decimal 10101(B)= =16+4+1= (B)= = (O)= =64+1=65 71(O)= =57 101A(H)= =

12 Converting decimal to binary Example: Convert to base 2. Start by converting the integer portion: 12

13 Now, convert the fraction: Putting them all together, =

14 Repeatedly divide the decimal integer by 2 Each remainder is a binary digit 37 = (100101) 2 stop when quotient is zero 14

15 Nonterminating Base 2 Fraction We can t always convert a terminating base 10 fraction into an equivalent terminating base 2 fraction: 15

16 Decimal to Octal & Hexadecimal 100(D)=144(O)=64(H) 8 Octal number Hexadecimal number

17 Base-k to decimal Decimal to base-k Integer portion: divided by k and take the remainders Fraction: multiplied by k and take the integer parts 17

18 Converting binary to octal / hexadecimal numbers Each octal (hexadecimal) digit corresponds to 3 (4) binary bits (B)= (O) (B)=36F.D4(H) 3 6 F D 4 18

19 Some Excercises Write down the computation steps ( )B= ( )D (422)D=( )H (52)O=( )B ( )=( )H 19

20 Answers to Exercises ( )B= =(55.25)D (422)D=(1A6)H 52 8 = (101 2 )(010 2 ) = = ( )( ) = 6D 16 20

21 Classification of numbers Numbers Fixed Point Floating Point Signed Unsigned Signed Unsigned 21

22 Fixed Point Numbers Each number has exactly the same number of digits, and the point is always in the same place For example: 0.23, 5.12, 9.11(decimal); each number has 3 digits, and the decimal point is located two places from the right 11.10, 01.10, 00.11(binary) 4 binary digits and the binary point is in the middle 22

23 Unsigned Integer Storage Sizes Half Word Byte 8 16 Storage Sizes Word 32 Double Word 64 Storage Type Unsigned Range Powers of 2 Byte 0 to to (2 8 1) Half Word 0 to 65,535 0 to (2 16 1) Word 0 to 4,294,967, to(2 32 1) Double Word 0 to 18,446,744,073,709,551,615 0 to (2 64 1) What is the largest 20-bit unsigned integer? Answer: = 1,048,575 23

24 Binary Addition Start with the rightmost bit Add each pair of bits Include the carry in the addition, if present carry (54) (29) (83) bit position:

25 Hexadecimal Addition Start with the rightmost hexadecimal digits Let Sum = summation of two hex digits If Sum is greater than or equal to 16 Sum = Sum 16 and Carry = 1 Example: carry: Subtraction? 1C37286A E 8 4 B Since A + B = = 21 AFC D 1 0 B 5 Sum = = 5 Carry = 1 25

26 Signed Integers 8-bit Binary Unsigned Signed How to represent a signed value value value number? Signed magnitude Sign bit: 0-positive 2's complement (computation) Divide the range of values into 2 equal parts First part corresponds to the positive numbers ( 0) Second part correspond to the negative numbers (< 0) Sign bit: 1-negative

27 Sign Bit Highest bit indicates the sign 1 = negative 0 = positive Sign bit Negative Positive For Hexadecimal Numbers, check leftmost digit If leftmost digit is > 7, then value is negative Examples: 8A and C5 are negative bytes B1C42A00 is a negative word (32-bit signed integer) 27

28 Ranges of Signed Integers For n-bit signed integers: Range is -2 n 1 1 to (2 n 1 1 1) Positive range: 0 to 2 n 1 1 Negative range: -2 n 1 to -1 Storage Type Unsigned Range Powers of 2 Byte 128 to to (2 7 1) Half Word 32,768 to +32, to (2 15 1) Word 2,147,483,648 to +2,147,483, to (2 31 1) Double Word 9,223,372,036,854,775,808 to +9,223,372,036,854,775, to (2 63 1) Practice: What is the range of signed values that may be stored in 20 bits? 28

29 Signed Magnitude Also know as sign and magnitude, the leftmost t bit is the sign (0 = positive, 1 = negative) and the remaining bits are the magnitude. Example: = = Two representations for zero: +0 = , -0 = Example: (-5) = (-9)D? Solution: 2 s complement representation 29

30 Two s Complement One representation for zero: +0 = , -0 = Positive number: 2 s complement = original number Negative number starting value (negative numbers) = -36 step1: keep the sign bit and reverse other bits step 2: add 1 to the value from step sum = 2's complement representation =

31 Binary Subtraction When subtracting A B, convert B to its 2's complement Add A to ( B), if the result is negative (leftmost bit=1), compute 2 s complement again borrow: carry: (2's complement) (same result) Final carry is ignored :( )B ( = )2 s complement = ( = ) -1 31

32 ( = ) ) = Some exercise

33 33

34 Overflow When two numbers are added that have large magnitudes and the same sign, an overflow will occur if the result is too large to fit inthe number of bits used in the representation The result should be (+130)10,which h is bigger than +127 (maximum for 8-bit) 34

35 Overflow Overflow occurs when Adding two positive numbers and the sum is negative Adding two negative numbers and the sum is positive If the numbers being added are of opposite signs, then an overflow will never occur 35

36 Sign Extension When convert low-bit version number to high-bit version, sign extension is needed Example: convert 16-bit 2 to 32-bit binary number 16-bit bit

37 Sign Extension Step 1: Move the number into the lower-significant bits Step 2: Fill all the remaining higher bits with the sign bit This will ensure that both magnitude and sign are correct Examples Sign-Extend to 16 bits = = -77 Sign-Extend to 16 bits = = +98 Infinite it 0s can be added d to the left of a positive number Infinite 1s can be added to the left of a negative number 37

38 Fixed Point Multiplication and Division Multiplication of unsigned binary integers is handled similar to the way it is carried out by hand for decimal numbers When two unsigned n-bit numbers are multiplied, the result can be as large as 2n bits When two signed n-bit numbers are multiplied, the result can be as large as only 2(n-1)+1 = (2n-1) bits, because this is equivalent to multiplying two (n-1)-bit unsigned numbers and then introducing the sign bit 38

39 Multiplication & division for unsigned integers 39

40 Signed Fixed Point Multiplication If we apply the multiplication and division methods described in the previous sections to signed integers, then we will run into some trouble 2 s complement 40

41 Solution: each partial product is extended to the width of the result, and only the rightmost eight bits of the result are retained. If both operands are negative, then the signs are extended for both operands, again retaining only the rightmost eight bits of the result. The first operand is (2 s complement)-->(-1)10 41

42 Multiply unit of MIPS contains two 32-bit registers called hi and lo. These are not general purpose registers When two 32-bit operands are multiplied, hi and lo hold the 64 bits of the result. Bits 32 through 63 are in hi and bits 0 through 31 are in lo. 42

43 Move the result of a multiplication into a general purpose p register The hi and lo registers cannot be used with any of the other arithmetic or logic instructions mfhi $d # $d < hi. Move From Hi mflo $d # $d < lo. Move From Lo 43

44 What went wrong is that the sign bit did not get extended to the left of the result This is not a problem for a positive result because the high order bits default to 0, producing the correct sign bit 0 44

45 Signed Fixed Point division Signed division is more difficult. We will not explore the methods here, but as a general technique, we can convert the operands into their positive forms, perform the division, and then convert the result into its true signed form as a final step 45

46 Division For N-digit integer division there are two results, an N- digit quotient and an N-digit remainder With 32-bit operands there will be (in general) two 32- bit results. MIPS uses the hi and lo registers for the results 46

47 Character Storage Character sets Standard ASCII: 7-bit character codes (0 127) Extended ASCII: 8-bit character codes (0 255) Unicode: 16-bit character codes (0 65,535) Unicode standard represents a universal character set Defines codes for characters used in all major languages Used in Windows-XP: each character is encoded as 16 bits UTF-8: variable-length length encoding used in HTML Encodes all Unicode characters Uses 1 byte for ASCII, but multiple bytes for other characters Null-terminated String Array of characters followed by a NULL character 47

48 Printable ASCII Codes A B C D E F 2 space! " # $ % & ' ( ) * +, -. / : ; < = >? A B C D E F G H I J K L M N O 5 P Q R S T U V W X Y Z [ \ ] ^ _ 6 ` a b c d e f g h i j k l m n o 7 p q r s t u v w x y z { } ~ DEL Examples: ASCII code for space character = 20 (hex) = 32 (decimal) ASCII code for 'L' =4C (hex) = 76 (decimal) ASCII code for 'a' = 61 (hex) = 97 (decimal) 48

49 Control Characters The first 32 characters of ASCII table are used for control Control character codes = 00 to 1F (hexadecimal) Examples of Control Characters Character 0 is the NULL character used to terminate a string Character 9 is the Horizontal Tab (HT) character Character 0A (hex) = 10 (decimal) is the Line Feed (LF) Character 0D (hex) = 13 (decimal) is the Carriage Return (CR) The LF and CR characters are used together They advance the cursor to the beginning of next line One control character appears at end of ASCII table Character 7F (hex) is the Delete (DEL) character 49

50 Question: how to represent large numbers (e.g )? Solution: Floating point numbers 50

51 Floating-Point Representation Computers use a form of scientific notation for floating-point representation Numbers written in scientific notation have three components: 51

52 Computer representation of a floating-point number consists of three fixed-size fields: This is the standard arrangement of these fields. 52

53 The one-bit sign field is the sign of the stored value. The size of the exponent field, determines the range of values that can be represented. The size of the significand determines the precision of the representation. 53

54 The IEEE-754 single precision floating point standard uses an 8-bit exponent and a 23-bit significand. ifi The IEEE-754 double precision standard uses an 11-bit exponent and a 52-bit significand. ifi 54

55 The significand of a floating-point number is always preceded by an implied binary point. Thus, the significand always contains a fractional binary value. The exponent indicates the power of 2 to which the significand is raised. 55

56 Base 10 Floating Point Numbers Floating gpoint numbers allow very large and very small numbers to be represented using only a few digits, at the expense of precision. The precision is primarily determined by the number of digits in the fraction (or significand, which has integer and fractional parts), and the range is primarily determined by the number of digits in the exponent. Example ( ): 56

57 Normalization The base 10 number 254 can be represented in floating point form as , or equivalently as: , or , or , or , or infinitely many other ways, which creates problems when making comparisons, with so many representations of the same number. Floating point numbers are usually normalized, in which the radix point is located in only one possible position for a given number. Usually, but not always, the normalized representation places the radix point immediately to the left of the leftmost, nonzero digit in the fraction, as in:

58 IEEE-754 Floating Point Formats 58

59 IEEE 754 uses a bias of 127 for single precision (1023 for double precision) Biased exponent means that the value represented by a floating-point number is (-1) S (1+d) 2 (p+b) S: value of sign bit d: value of fraction field p: value of exponent field B: bias value (127 / 1023) 59

60 Number=±1.xx xxx 2 xxx 2 ±p d 1 bit 8/11 bits 23/52 bits Sign E F p+127 P+1023 p+bias d Fraction E: data range F: data precision i

61 Examples:single precision 26.0D= B= * =131= B D=-10.1B=-1.01* =128= B

62 Thank you very much! 62