BCSE205L

Computer Architecture and Organization

Module 2 – Data Representation and Computer Arithmetic

Dr. C.R.Dhivyaa
Assistant Professor
School of Computer Science and Engineering
Vellore Institute of Technology, Vellore
 Number System & Types
• In digital electronics, the number
system is used for representing the
information.
• Number system - different bases
• The base or radix of the number
system is the total number of the
digit used in the number system.
 Binary Number Systems
• Characteristics
• It holds only two values, i.e., either 0 or 1.
• It is also known as the base 2 number system.
• The position of a digit represents the 0 power of
the base(2). Example: 20
• The position of the last digit represents the x
power of the base(2). Example: 2x, where x
represents the last position.
• Examples: (10100)2, (11011)2, (11001)2, (000101)2,
 Binary Number Systems
• Terminologies - Binary systems
• Each digit – 1 Bit – 0/1 - Single Binary Digit
• Collection of 4 bits - Nibble
• Collection of 8 bits - Byte
• Collection of 16/32/64 bits - Word

Depends on Memory
Organization of a
Computer
 Decimal Number Systems
• Contains ten digits from 0 to 9
• base 10 number system
• The successive place value or position, left
to the decimal point holds units, tens,
hundreds, thousands, and so on.
 Octal Number Systems
• Contains Eight digits from 0 to 7
• base 8 number system
• Octal number – represented in only 3 bits.
• The position of a digit represents the 0
power of the base(8). Example: 80
• The position of the last digit represents the
x power of the base(8). Example: 8x, where
x represents the last position.
• Examples:
(273)8, (5644)8, (0.5365)8, (1123)8, (1223)8.
 Hexadecimal Number Systems
• Contains ten digits from 0 to 9 and 6 letters from A to
F.
• The letters from A to F defines numbers from 10 to 15.
• base 16 number system.
• The position of a digit represents the 0 power of the
base(16). Example: 160
• The position of the last digit represents the x power of
the base(16). Example: 16x, where x represents the
last position.
• Examples: (FAC2)16, (564)16, (0ABD5)16, (1123)16, (11F3)16.
 Number Systems
• Numbers
• Unsigned Numbers (+)
• has only magnitude - no sign.
• Signed Numbers(+/-)
• has both magnitude and sign
• Sign bit – 0 – positive number
• Sign bit – 1 – Negative number
• Representing Binary Numbers
• n bits – can be used to represent--- 2n distinct numbers (0 to 2n-1)
▪ For example: If n=3 bits, 23 = 8 distinct numbers (0 to 7)can be represented
▪ (0-000, 1-001, 2-010, 3-011, 4-100, 5-101, 6-110, 7-111)
 Unsigned Number Representation of 4 Bits
• If no. of bits → 4, Unsigned → 0 to 2n-1
• then for unsigned numbers, these four bits can store the decimal values from
0 to 24-1 → (decimal values 0 to 15).
No. of Unsigned numbers that can be stored in
bits (n) n bit representation
3 Decimal Numbers 0 to 7 0 to 23-1
4 D. Numbers 0 to 15 0 to 24-1
5 D. Numbers 0 to 32 0 to 25-1
6 D. Numbers 0 to 63 0 to 26-1
7 D. Numbers 0 to 127 0 to 27-1
8 D. Numbers 0 to 255 0 to 28-1
……. …… ……

Hence, in unsigned representation - if u need to perform operations on the Decimal number 5,

we need 3 bits to represent it ---- number 14 → 4 bits , number 30 → 5 bits , ……….
 Signed Binary Numbers
• When a signed binary number is positive
• The MSB is ‘0’ which is the sign bit and rest bits represents the
magnitude
• When a signed binary number is negative
• The MSB is ‘1’ which is the sign bit and rest of the bits may be
represented by three different ways
• Signed magnitude representation
• Signed 1’s complement representation
• Signed 2’s complement representation
 Signed Binary Number Representation
• Representation
• Sign magnitude representation
• One’s complement representation
• Two’s complement representation

Unsigned → 0 to 2n-1
Signed (Sign Magnitude Representation)→-2n-1+1 to 2n-1-1
Signed (One’s Complement Representation)→-2n-1+1 to 2n-1-1
Signed (Two’s Complement Representation)→-2n-1 to 2n-1-1
 Sign Magnitude Representation of 4 Bits

• For an n bit representation:

• The most significant bit (MSB) indicates the sign (0-Positive and 1-
Negative)
• Remaining n-1 bits will represent the magnitude.
• Includes values for +0 and -0. (Two representation for 0)
• Hence the possible range of numbers –(2 n-1 -1) to +(2n-1-1)
• For 3 bits, the range is (-3,-2,-1,-0,+0,+1,+2,+3)
 Signed and Unsigned Number Representation of 4 Bits
(Sign Magnitude Representation)

All 4 bits - data

1st bit – sign bit

Remaining 3 bits - data

Two representation for 0

 One’s Complement Representation of Signed Numbers
• One’s Complement – One way of representing a negative number.
• One’s complement of a Positive value representation
– no change in both sign bit and magnitude.

• Negative value representation – Take Complement

of every bit of the positive value of
that number (1 to 0 and 0 to 1).
1’s representation of -ve number -A =
(complement of every bit of +A)

1’s(+ve number +A) = +A

One’s Complement Representation of Signed 4 Bits

Sign-Magnitude Representation

Two representation for 0

Disadvantage of One’s Complement Representation
• Hence, One’s complement representation of +6 = 0000 0110
• Hence, One’s complement representation of -6 = 1111 1001

• It's important to note that in 1's complement representation, there are

two representations for zero (+0 = 0000 & -0 = 1111).
• This lead to some complications in arithmetic operations, comparisons and
conversions…
One’s Complement – Not preferred for any calculation
Reason that other signed number representations like 2's complement
are more commonly used.
 Two’s Complement Representation of Signed Numbers
• One representation for 0 (Both +0 and -0 are treated as same)
• For a positive value – Two’s complement representation remains same
as its original value.
• For a Negative value – Two’s complement representation – calculated by
taking One’s Complement of the positive value of that number + 1.
(Complement every bit of the number and then add one to the resulting number)
 (Two’s Complement) Signed → -2n-1 to 2n-1-1

Two’s Complement Representation of Signed Number

• If no. of bits → 4, 1 bit - stores sign bit, remaining 3 bits – stores value
• then for signed numbers, these four bits can store the decimal values from
-(23) to +23-1 → (decimal values -8 to +7).
No. of bits Unsigned numbers that can be stored in n bit Signed numbers that can be stored in n bit
(n) representation representation + 1 bit for sign
3 Decimal Numbers 0 to 7 0 to 23-1 -4 to +3 -22 to +22-1
4 D. Numbers 0 to 15 0 to 24-1 -8 to +7 -23 to +23-1
5 D. Numbers 0 to 32 0 to 25-1 -16 to +15 -24 to +24-1
6 D. Numbers 0 to 63 0 to 26-1 -32 to +31 -25 to +25-1
7 D. Numbers 0 to 127 0 to 27-1 -64 to +63 -26 to +26-1
8 D. Numbers 0 to 255 0 to 28-1 -128 to +127 -27 to +27-1
……. …… …… …… ……

Hence, in Signed representation - if u need to perform operations on the Decimal number 5 or -

5, we need 4 bits to represent it ---- number 14 or -14 → 5 bits , number 30 → 5 bits , ……….
 Two’s Complement Representation of Signed Numbers

Advantages Expansion of Bit Length: Add additional bit positions to the left
• Unique representation of 0. and fill in with the value of the original sign bit.
• Subtraction can be done using addition. Overflow Rule: If two numbers with the same sign (both positive or
• Leads to substantial saving in circuitry. both negative) are added, then overflow occurs if and only if the
result has the opposite sign.
 Two’s Complement Representation of Signed Numbers
• Example: How to represent -4 using 2’s complement representation??
• No. of bits required to represent signed - 4 is 3 bits + 1 sign bit
• Signed representation of +4=0100
• 1’s complement representation of -4 is –(0100) is 1011.
• 2’s complement representation → 1011
+ 1
=1100
• Here again the most significant bit will indicate the sign 0 for positive,
1 for negative.
To conclude: Pros and cons of integer representation
• Signed magnitude representation:
• 2 representations for 0
• Simple
• 255 different numbers can be represented.
• Need to consider both sign and magnitude in arithmetic
• Different logic for addition and subtraction
• 1’s complement representation:
• 2 representations for 0
• Complexity in performing addition and subtraction
• 255 different numbers can be represented.
• 2’s complement representation:
• Only one representation for 0
• 256 different numbers can be represented.
• Arithmetic works easily
 Real/Fractional Number Notation
• Real Numbers - numbers with fractional component
• Two major approaches to store real numbers
• Fixed Point Notation
• Floating Point Notation.
• Fixed point notation - there are a fixed number of digits after the decimal
point (Integer . Fraction)
• In this representation the position of the radix point is fixed. Hence this
representation is called fixed point representation
• Example: (123.431), (4.5)10 = (100.1)2
• Floating point number - allows for a varying number of digits after the
decimal point. (Integer . Fraction)
• In this representation the position of the radix point is allowed to move.
• Example: (-123.334444455555567),
Real Number Notation
 Fixed Point Representation
• The number of integer and fraction bits must be agreed upon by
those generating and those reading the number
• Example: Fixed-point representation of 6.75 using 4 integer bits
and 4 fraction bits in binary
• 6 =110 -→ 0110
• .75 = 110 → 1100
• Binary Representation:
01101100
(6.75)10 = (0110.1100)2
• Convert back to Decimal (Check)
0*23 + 1*22 + 1*21 + 0*20 + 1*2-1 + 1*2-2 + 0*2-3 + 0*2-4
= 6.75
Fixed Point Representation
 Fixed Point Binary Numbers - Arithmetic
Operations
• Addition
• Subtraction
• Multiplication
• Division
 Fixed Point Binary Numbers - Addition
• Follow same rules as in decimal addition, with the difference that when
sum is 2 indicates a carry (not a 10)
• Carry rules Carry 1 1 1 1 1 0
• 0+0 = sum 0 carry 0 Augend 0 0 1 0 0 1
• 0+1 = 1+0 = sum 1carry 0 Addend 0 1 1 1 1 1
• 1+1 = sum 0 carry1 Result 1 0 1 0 0 0
• 1+1+1 = sum 1 carry1

111 Carry Values

0101
+1011
10000
 Fixed Point Binary Numbers - Subtraction
• Borrow rules
• 0-0 = 1-1 = 0 borrow 0
• 1-0 = 1 borrow 0
• 0-1 = 1 borrow 1
• The rules of the decimal base applies to binary as well.
• To be able to calculate 0-1, we have to “borrow one” from the next left
digit. 12
0202
1010
-0111
0011
 Overflow and Underflow
• Overflow Overflow
• Overflow occurs - When the addition of • Assume 4-bit restriction and 2’s complement
two binary numbers represented in ‘n’ • Maximum possible value: 24-1 – 1 = 7
bits - yields a result that is greater than 610 + 310 = should be 910
the maximum possible value that can
be represented with ‘n’ bits. But, signed 9 can’t be represented with 4 bits(-8 to +7)
Hence it produces a wrong result.
• Possible with both integers and floating
point numbers. 01102 610
• Example: adding 1 to the maximum +00112 +310
value of a uint 8 (255) would cause an 10012 -710  not good!
overflow, resulting in a value of 0.
 Overflow and Underflow
• Underflow Underflow
• Underflow occurs - When an
arithmetic operation of two binary • Assume 4-bit restriction and 2’s complement
numbers represented in ‘n’ bits - • Minimum possible value: -(24-1) = -8
yields a result that is lesser than -510 + -510 = should be -1010
the minimum possible value that
can be represented with ‘n’ bits. But, signed -10 can’t be represented with 4 bits(-8 to +7)
Hence it produces a wrong result.
• Underflow is a bit trickier to
understand because it has to do 10112 -510
with precision in floating points. +10112 +-510
01102 610  not good!
Fixed Point Arithmetic Operations-
Multiplication
 Fixed Point Binary Numbers -
Multiplication
• Multiplication
• Generation of Partial Products.
• Each successive partial product is shifted one position to the left relative to
the preceding partial product
• Final product = sum of the Partial Products.

• Multiplication of two unsigned binary numbers of n bit size results into 2n bit
result.
N * N bits input = 2N bits result
 Fixed Point Binary Numbers -
Multiplication
Unsigned Integer representation
• n bit - How many distinct numbers can be represented?
• 2n – Distinct numbers.
▪ For example:
▪ If n=3 bits,
23 = 8 (0-000, 1-001, 2-010, 3-011, 4-100, 5-101, 6-110, 7-111)
▪ 0 is minimum and 7 (2n-1) is maximum
▪ If n=4 bits,
24 = 16 (0 to 15)
▪ 0 is minimum and 15 is maximum
▪ If n=5 bits,
25 = 32 (0 to 31)
▪ 0 is minimum and 31 is maximum
▪ If n=6 bits,
26 = 64 (0 to 63)
▪ 0 is minimum and 63 is maximum
 Multiplication of unsigned numbers
1. Multiply the given numbers 13 x 11
• number of bits required to represent the number 13 is 4 bits
• number of bits required to represent the number 11 is 4 bits
• Convert decimal to binary
• 13→ 1101
• 11→1011
 Multiplication of unsigned numbers
1. Multiply the given numbers 13 x 11

27 26 25 24 23 22 21 20

=(1*27)+ (0*26)+(0*25)+(0*24)+(1*23)+(1*22)+(1*21)+(1*20) =128+8+4+2+1 = 143

Product of 2 n-bit numbers is at most a 2n-bit number.

Unsigned multiplication can be viewed as addition of shifted versions of the multiplicand.
 Multiplication of unsigned numbers

• We added the partial products at end.

• Alternative would be to add the partial products at each stage.

• Rules to implement multiplication:

• If the ith bit of the multiplier is 1, shift the multiplicand and add the shifted
multiplicand to the current value of the partial product.
• Hand over the partial product to the next stage
• Value of the partial product at the start stage is 0.
 Multiplication of unsigned numbers

Typical Multiplication Cell

 Multiplication of unsigned numbers

Combinatorial array multiplier

Multiplicand is shifted by
displacing it through an
array of adders.
 Multiplication of Signed numbers
Signed number representation
• For an n bit representation: MSB- sign bit(0 or 1), n-1(magnitude)

• The possible range of numbers–(2 n-1 ) (Min) to +(2n-1-1)(Max)

▪ If n=4 bits, - 23 to + 23 -1= -8 to +7
▪ -8 is minimum and 7 is maximum
▪ If n=5 bits, - 24 to + 24 -1= -16 to +15
▪ -16 is minimum and 15 is maximum
▪ If n=6 bits, - 25 to + 25 -1= -32 to +31
▪ -32 is minimum and 31 is maximum
 Multiplication of Signed numbers
• Case 1:Multiply the given numbers (+13) x (+11) (Positive
Multiplier)
• number of bits required to represent the number +13 is 5 bits
• number of bits required to represent the number +11 is 5 bits
• Convert decimal to binary
• 13→ 1101 11→1011
• +13 →01101(5 bits=0(sign), 1101(magnitude))
• +11→01011(5 bits=0(sign), 1011(magnitude))
 Multiplication of Signed numbers
Multiplication of 5 bits and 5 bits – 0 1 1 0 1 ( + 13 )
produces 10 bit result … Hence 0 1 0 1 1 ( + 11 )
extra bit added.. (optional)
0 0 0 0 0 0 1 1 0 1

0 0 0 0 0 1 1 0 1
Sign extension is
shown in red
0 0 0 0 0 0 0 0

0 0 0 1 1 0 1

0 0 0 0 0 0

0 0 1 0 0 0 1 1 1 1 ( + 143 )

- 29 28 27 26 25 24 23 22 21 20
= (0*-29)+ (0*28)+ (1*27)+(0*26)+(0*25)+(0*24)+(1*23)+(1*22)+(1*21)+(1*20)
=128+8+4+2+1 =143→ answer
 Multiplication of Signed numbers
• Case 2: Multiply the given numbers (-13) x (+11) (Positive Multiplier)

• number of bits required to represent the number -13 is 5 bits

• number of bits required to represent the number +11 is 5 bits
• Convert decimal to binary
• 13→ 1101 11→1011
• Take 2’s complement of 13→0010+1=0011
• -13 →10011(5 bits=1(sign), 0011(magnitude))
• +11→01011(5 bits=0(sign), 1011(magnitude))
Multiplication of Signed numbers
1 0 0 1 1 ( - 13 )
0 1 0 1 1 ( + 11 )

1 1 1 1 1 1 0 0 1 1

1 1 1 1 1 0 0 1 1
Sign extension is
0 0 0 0 0 0 0 0 shown in red

1 1 1 0 0 1 1

0 0 0 0 0 0

1 1 0 1 1 1 0 0 0 1 ( - 143 )
- 29 28 27 26 25 24 23 22 21 20
= (1*-29)+ (1*28)+ (0*27)+(1*26)+(1*25)+(1*24)+(0*23)+(0*22)+(0*21)+(1*20)
=-512+256+64+32+16+1
=-143→ answer
 Multiplication of Signed numbers
• Case 3: Multiply the given numbers (+13) x (-11) (Negative Multiplier)
• number of bits required to represent the number +13 is 5 bits
• number of bits required to represent the number -11 is 5 bits
• Convert decimal to binary
• 13→ 1101 11→1011
• Take 2’s complement of 11→0100+1=0101
• +13 →01101(5 bits=0(sign), 1101(magnitude))
• -11→10101(5 bits=1(sign), 0101(magnitude))
• Since the multiplier is negative(-), take 2’s complement of both M(+13) and Q(-11)
and proceed as -13 and 11
• Take 2’s complement of (+13)→10010+1=10011(-13)
• Take 2’s complement of (-11)→01010+1=01011(11)
• -13 →10011(5 bits=1(sign), 0011(magnitude))
• +11→01011(5 bits=0(sign), 1011(magnitude))
 Multiplication of Signed numbers
Since the multiplier is negative(-), take 2’s complement of
both M(+13) and Q(-11) and proceed as -13 and 11
1 0 0 1 1 ( - 13 )
0 1 0 1 1 ( + 11 )

1 1 1 1 1 1 0 0 1 1

1 1 1 1 1 0 0 1 1
Sign extension is
0 0 0 0 0 0 0 0 shown in red

1 1 1 0 0 1 1

0 0 0 0 0 0

1 1 0 1 1 1 0 0 0 1 ( - 143 )
- 29 28 27 26 25 24 23 22 21 20
= (1*-29)+ (1*28)+ (0*27)+(1*26)+(1*25)+(1*24)+(0*23)+(0*22)+(0*21)+(1*20)
=-512+256+64+32+16+1
=-143→ answer
 Multiplication of Signed numbers
• Case 4: Multiply the given numbers (-13) x (-11)
• number of bits required to represent the number -13 is 5 bits
• number of bits required to represent the number -11 is 5 bits
• Convert decimal to binary
• 13→ 1101 11→1011
• Take 2’s complement of 13→0010+1=0011
• Take 2’s complement of 11→0100+1=0101
• -13 →10011(5 bits=1(sign), 0011(magnitude))
• -11→10101(5 bits=1(sign), 0101(magnitude))
• Since the multiplier is negative(-), take 2’s complement of both M(-13) and Q(-11) and
proceed as 13 and 11
• Take 2’s complement of (-13)→01100+1=01101(+13)
• Take 2’s complement of (-11)→01010+1=01011(+11)
• +13 →01101(5 bits=0(sign), 1101(magnitude))
• +11→01011(5 bits=0(sign), 1011(magnitude))
Multiplication of Signed numbers
0 1 1 0 1 ( + 13 )
0 1 0 1 1 ( + 11 )

0 0 0 0 0 0 1 1 0 1

0 0 0 0 0 1 1 0 1
Sign extension is
shown in red
0 0 0 0 0 0 0 0

0 0 0 1 1 0 1

0 0 0 0 0 0

0 0 1 0 0 0 1 1 1 1 ( - 143 )

- 29 28 27 26 25 24 23 22 21 20
= (0*-29)+ (0*28)+ (1*27)+(0*26)+(0*25)+(0*24)+(1*23)+(1*22)+(1*21)+(1*20)
=128+8+4+2+1 =143→ answer
 Multiplication of Signed numbers

• Drawbacks of Normal Signed Multiplication

• Different strategies followed for positive and negative multiplier cases
• Multiplicand (+ve) and Multiplier (+ve)

• Multiplicand (-ve) and Multiplier (+ve)

• Multiplicand (+ve) and Multiplier (-ve)

• Multiplicand (-ve) and Multiplier (-ve)

 Multiplication of Signed numbers
– Booth Algorithm
• Allows multiplication of two signed binary integers in 2's complement respectively.
• Advantages:
• Booth Algorithm will treat both positive and negative multiplier in the same way.
• Speed up the performance of the multiplication process - very efficient too.
• It reduces no of partial product addition
• Booth modifies the multiplier bits to reduce the
number of partial product.
 Multiplication of Signed numbers
– Booth Algorithm
• Multiplication Steps:
Step 1: Find binary Multiplier Booth
representation of Multiplicand (M) Multiplier Bit
Bit i Bit (i-1)
and Multiplier (Q) including sign
(Bit to modify) (Bit to the right of i)
bit
Step 2: Add zero to the LSB of 0 0 0×M
the multiplier 0 1 +1 × M
Step 3: Use booth multiplier
recoding table to modify the 1 0 -1 × M
multiplier bits 1 1 0× M
Step 4: Perform Multiplication
 Multiplication of Signed numbers
– Booth Algorithm
Finding the Booth Multiplier
• Example 1: M = 6 , Q = 2 • Example 3: M = - 6 , Q = 2
• 6 → 0110 and 2 -> 0010 • - 6 → 1010 and 2 -> 0010
• Booth Recoding • Booth Recoding multiplier
multiplier
Multiplier Booth
0 0 1 0 0
0 0 1 0 0 Bit i Bit (i-1) Multipl
0 +1 -1 0
0 +1 -1 0 (Bit to (Bit to the ier Bit
modify) right of i)
• Example 2: M = 6 , Q = -2 • Example 4: M = 6 , Q = -2
• 6 → 0110 and -2 -> 1110 • - 6 → 1010 and -2 -> 1110
0 0 0×M
• Booth Recoding • Booth Recoding multiplier 0 1 +1 × M
multiplier 1 1 1 0 0
1 1 1 0 0 1 0 -1 × M
0 0 -1 0
0 0 -1 0 1 1 0× M
 Case 1: 6 X 2 Multiplication of Signed numbers - Booth Algorithm
Step 1: Find binary representation of Multiplicand (6) and multiplier
(2) Including sign bit. 6 → 0110 and 2 -> 0010
Step 2:Recoding multiplier → 00100 →0+1-10
Step 3: Multiplication operation Multiplier Booth
0110X Bit i Bit (i-1) Multiplier
0+1-10 -1→ take 2’complement (Bit to (Bit to the Bit
00000000 of multiplicand modify) right of i)
1111010 0 0 0×M
000110 0 1 +1 × M
00000 (Sign bit of the result) 1 0 -1 × M
1 1 0× M
100001100 answer is 12
(Ignore carry) -27 26 25 24 23 22 21 20 =(0*-27)+ (0*26)+(0*25)+(0*24)+(1*23)+(1*22)+(0*21)+(0*20)=8+4=12
Case 2: 6 X -2 Multiplication of Signed numbers - Booth Algorithm
Step 1: Find binary representation of Multiplicand (6) and multiplier
(-2) Including sign bit. 6 → 0110 and -2 → 1110
Step 2: Recoding multiplier → 11100 → 0 0 -1 0
Step 3: Multiplication operation Multiplier Booth
0110X Bit i Bit (i-1) Multiplier
0 0 -1 0 -1→ take 2’complement (Bit to (Bit to the Bit
00000000 of multiplicand modify) right of i)
1111010 0 0 0×M
000000 0 1 +1 × M
00000 (Sign bit of the result) 1 0 -1 × M
1 1 0× M
11110100 answer is -12
-27 26 25 24 23 22 21 20 =(1*-27)+ (1*26)+(1*25)+(1*24)+(0*23)+(1*22)+(0*21)+(0*20)=-128+64+32+16+4=-12
Case 3: -6 X 2 Multiplication of Signed numbers - Booth Algorithm
Step 1: Find binary representation of Multiplicand (-6) and multiplier
(2) Including sign bit. -6 → 1010 and 2 →0010
Step 2: Recoding multiplier → 00100 → 0 +1 -1 0
Step 3: Multiplication operation Multiplier Booth
1010X Bit i Bit (i-1) Multiplier
0+1-1 0 -1→ take 2’complement (Bit to (Bit to the Bit
00000000 of multiplicand modify) right of i)
0000110 0 0 0×M
111010 0 1 +1 × M
00000 (Sign bit of the result) 1 0 -1 × M
1 1 0× M
11110100 answer is -12
-27 26 25 24 23 22 21 20 =(1*-27)+ (1*26)+(1*25)+(1*24)+(0*23)+(1*22)+(0*21)+(0*20)=-128+64+32+16+4=-12
Case 4: -6 X -2 Multiplication of Signed numbers - Booth Algorithm
Step 1: Find binary representation of Multiplicand (-6) and multiplier
(-2) Including sign bit. -6 → 1010 and -2 →1110
Step 2: Recoding multiplier → 11100 → 0 0 -1 0
Step 3: Multiplication operation Multiplier Booth
1010X Bit i Bit (i-1) Multiplier
0 0 -1 0 -1→ take 2’complement (Bit to (Bit to the Bit
00000000 of multiplicand modify) right of i)
0000110 0 0 0×M
000000 0 1 +1 × M
00000 (Sign bit of the result) 1 0 -1 × M
1 1 0× M
00001100 answer is 12
-27 26 25 24 23 22 21 20 =(0*-27)+ (0*26)+(0*25)+(0*24)+(1*23)+(1*22)+(0*21)+(0*20)=8+4=12
 Multiplication of Signed numbers
– Booth Algorithm
• Hardware for Booth’s Multiplication
 Multiplication of Signed numbers
– Booth Algorithm
Flow Chart
M - Multiplicand
Q - Multiplier
Q0 - least significant bit of QR
Q-1 - previous least significant bit of QR
A- 0
Count - number of bits in multiplier
Example: Multiplication of Signed numbers - Booth Algorithm
Multiply 7 x 3
Multiplicand =7 → Binary equivalent is 0111→M
Multiplier = 3 →Binary equivalent is 0011→Q
-7 → Binary equivalent is 1001→ -M
A0 0 0 0 A 1 1 10
-M 1 0 0 1 M 0 1 1 1
A1 0 0 1 A0 1 0 1
Step A Q Q-1 Action Count

1 0 0 0 0 0 0 1 1 0 Initial 4

2 1 0 0 1 0 0 1 1 0 AA-M
1 1 0 0 1 0 0 1 1 Shift 3

3 1 1 1 0 0 1 0 0 1 Shift 2

4 0 1 0 1 0 1 0 0 1 AA+M
0 0 1 0 1 0 1 0 0 Shift 1

5 0 0 0 1 0 1 0 1 0 Shift 0 Convert binary to decimal and the

answer is 21
-27 26 25 24 23 22 21 20
Example: Multiplication of Signed numbers - Booth Algorithm
Multiply 7 x -3
Multiplicand =7 → Binary equivalent is 0111→M
Multiplier = -3 →Binary equivalent is 1101→Q
-7 → Binary equivalent is 1001→ -M
Step A Q Q-1 Action Count

1 0 0 0 0 1 1 0 1 0 Initial 4
A0 0 0 0
2 1 0 0 1 1 1 0 1 0 AA-M -M 1 0 0 1
1 1 0 0 1 1 1 0 1 Shift 3 A1 0 0 1

3 0 0 1 1 1 1 1 0 1 AA+M A1 1 0 0
+M 0 1 1 1
0 0 0 1 1 1 1 1 0 Shift 2
A 10 0 1 1 Discard 1
4 1 0 1 0 1 1 1 1 0 AA-M A0 0 0 1
1 1 0 1 0 1 1 1 1 Shift 1 -M 1 0 0 1
A1 0 1 0
5 1 1 1 0 1 0 1 1 1 Shift 0
-27 26 25 24 23 22 21 20 Convert binary to decimal and the
answer is -21
Example: Multiplication of Signed numbers - Booth Algorithm
Multiply -7 x 3
Multiplicand =-7 → Binary equivalent is 1001→M
Multiplier = 3 →Binary equivalent is 0011→Q

Step A Q Q-1 Action Count

1 0 0 0 0 0 0 1 1 0 Initial 4
A0 0 0 0
2 0 1 1 1 0 0 1 1 0 AA-M -M 0 1 1 1
0 0 1 1 1 0 0 1 1 Shift 3 A0 1 1 1

3 0 0 0 1 1 1 0 0 1 Shift 2 A0 0 0 1
+M 1 0 0 1
4 1 0 1 0 1 1 0 0 1 AA+M A1 0 1 0
1 1 0 1 0 1 1 0 0 Shift 1

5 1 1 1 0 1 0 1 1 0 Shift 0

-27 26 25 24 23 22 21 20 Convert binary to decimal and the

answer is -21
Example: Multiplication of Signed numbers - Booth Algorithm
Multiply -7 x -3
Multiplicand =-7 → Binary equivalent is 1001→M
Multiplier = -3 →Binary equivalent is 1101→Q

Step A Q Q-1 Action Count

1 0 0 0 0 1 1 0 1 0 Initial 4
A0 0 0 0
2 0 1 1 1 1 1 0 1 0 AA-M -M 0 1 1 1
0 0 1 1 1 1 1 0 1 Shift 3 A0 1 1 1

3 1 1 0 0 1 1 1 0 1 AA+M A0 0 1 1
+M 1 0 0 1
1 1 1 0 0 1 1 1 0 Shift 2
A1 1 0 0
4 0 1 0 1 0 1 1 1 0 AA-M
A1 1 1 0
0 0 1 0 1 0 1 1 1 Shift 1 -M 0 1 1 1
A0 1 0 1 Discard 1
5 0 0 0 1 0 1 0 1 1 Shift 0
-27 26 25 24 23 22 21 20 Convert binary to decimal and the
answer is 21
 Multiplication of Signed numbers
– Booth Algorithm
Practice Problems
• Show the step by step multiplication process using Booth
algorithm when the following binary numbers are
multiplied. Assume 5-bit registers that hold signed
numbers. The multiplicand in both cases is +15.
• (+15) × (+13)
• (+15) × (-13)
 Multiplication of Signed numbers
– Bit Pair Recoding Algorithm
• Fast Multiplication / Modified Booth Algorithm
• Bit-pair recoding halves the maximum number of summands
(versions of the multiplicand)
• - Reduces the number of Partial Products
 Multiplication of Signed numbers
– Bit Pair Recoding Algorithm
Multiplier Multiplier bit Booth
Note:
on the right Multiplier Bit
Bit i+1 Bit (i)
If multiplier bit is (left bit) (Bit to modify) i-1

• +1➔ Partial Product (PP) = M × 1

0 0 0 0×M
0 0 1 +1 × M
• -1→ PP = 2’s (M)
0 1 0 +1 × M
• +2→ PP = M × 10
0 1 1 +2 × M
• -2→ PP = 2’s (M × 10) 1 0 0 -2 × M
1 0 1 -1 × M
1 1 0 -1 × M
1 1 1 0× M
 Multiplication of Signed numbers – Bit Pair Recoding Algorithm
Case 1: 13 x 6
0 1 1 0 1 ( + 13 ) Multiplier Multiplier Booth
´ 0 0 1 1 0 (+6 ) bit on the Multiplier
Bit i+1 Bit (i)
(left bit) (Bit to right Bit
• Bit- Pair Recoded Multiplier modify) i-1

Implied 0 0 0 0 0×M
0 0 0 1 1 0 0 0 0 1 +1 × M
Sign extension 0 +2 -2→Bit Pair Recoding Multiplier 0 1 0 +1 × M
0
0
1 1
+2
0 1
-2
0 1 1 +2 × M
1
0
1
0
1
0
1
1
1
1
0
0
0
1
1
0
1 0
1 0 0 -2 × M
0 0 0 0 0 0
1 0 0 0 1 0 0 1 1 1 0
1 0 1 -1 × M
1 1 0 -1 × M
64+8+4+2=78
1 1 1 0× M
 Multiplication of Signed numbers – Bit Pair Recoding Algorithm
Case 2: 13 x -6
0 1 1 0 1 ( + 13 )
1 1 0 1 0 ( -6 )
Multiplier Multiplier Booth
Bit i+1 Bit (i) bit on the Multiplier
• Bit- Pair Recoded Multiplier (left bit) (Bit to right Bit
Implied 0 modify) i-1
1 1 1 0 1 0 0
0 0 0 0×M
Sign extension 0 -1 -2 →Bit Pair Recoding Multiplier 0 0 1 +1 × M
0 1 1 0 1 0 1 0 +1 × M
0 -1 -2
1 1 1 1 1 0 0 1 1 0 0 1 1 +2 × M
1 1 1 1 0 0 1 1
0 0 0 0 0 0 1 0 0 -2 × M
1 1 1 0 1 1 0 0 1 0 -78
- 29 28 27 26 25 24 23 22 21 20 1 0 1 -1 × M
= (1*-29)+ (1*28)+ (1*27)+(0*26)+(1*25)+(1*24)+(0*23)+(0*22)+(1*21)+(0*20) 1 1 0 -1 × M
=-512+256+128+32+16+2
=-78→ answer 1 1 1 0× M
 Multiplication of Signed numbers – Bit Pair Recoding Algorithm
Case 3: -13 x 6
1 0 0 1 1 ( - 13 )
0 0 1 1 0 (+6 )
Multiplier Multiplier Booth
Bit i+1 Bit (i) bit on the Multiplier
• Bit- Pair Recoded Multiplier (left bit) (Bit to right Bit
modify) i-1
Implied 0
0 0 0 1 1 0 0 0 0 0 0×M
Sign extension 0 +2 -2→Bit Pair Recoding Multiplier 0 0 1 +1 × M
1 0 0 1 1
0 +2 -2 0 1 0 +1 × M
0 0 0 0 0 1 1 0 1 0 0 1 1 +2 × M
1 1 1 0 0 1 1 0
0 0 0 0 0 0 1 0 0 -2 × M
-78
1 1 1 0 1 1 0 0 1 0
1 0 1 -1 × M
- 29 28 27 26 25 24 23 22 21 20
1 1 0 -1 × M
=-512+256+128+32+16+2 =-78→ answer
1 1 1 0× M
 Multiplication of Signed numbers – Bit Pair Recoding Algorithm
Case 4: -13 x -6
1 0 0 1 1 ( - 13 )
1 1 0 1 0 ( -6 )
Multiplier Multiplier Booth
Bit i+1 Bit (i) bit on the Multiplier
• Bit- Pair Recoded Multiplier (left bit) (Bit to right Bit
Implied 0 modify) i-1
1 1 1 0 1 0 0
0 0 0 0×M
Sign extension 0 -1 -2 →Bit Pair Recoding Multiplier 0 0 1 +1 × M
1 0 0 1 1
0 1 0 +1 × M
0 -1 -2
0 1 1 +2 × M
0 0 0 0 0 1 1 0 1 0

0 0 0 0 1 1 0 1
1 0 0 -2 × M
0 0 0 0 0 0 1 0 1 -1 × M
0 0 0 1 0 0 1 1 1 0 64+8+4+2=78 1 1 0 -1 × M
- 29 28 27 26 25 24 23 22 21 20
1 1 1 0× M
 Multiplication of Signed numbers
– Bit Pair Recoding Algorithm
Practice Problems
• Show the step by step multiplication process using Bit-pair
recoding technique when the following binary numbers are
multiplied. Assume 5-bit registers that hold signed
numbers. The multiplicand in both cases is +15.
• (+15) × (+13)
• (+15) × (-13)
 Binary Division
• Long Division is a method for dividing large numbers, which
breaks the division problem into multiple steps following a
sequence.

• Just like the regular division problems, the dividend is

divided by the divisor which gives a result known as the
quotient, and sometimes it gives a remainder too.
Example:
Algorithms for Division

• Restoring Method

• Non-restoring Method
Restoring Division Method
• Restoring Method
• Partial remainder is restored by adding the divisor to the negative
difference
• The restoring algorithm calculates the remainder by successively
subtracting the shifted denominator from the numerator until the remainder
is in the appropriate range. The operation in each step depends on the result
of the previous step.
• Value of Register is initially kept 0 and this is the register whose value is
restored during iteration due to which it is named Restoring.
• Comparison Method
• A and B are compared prior to the subtraction operation.
• If A ≥ B, B is subtracted from A.
• If A < B, partial remainder is shifted left and the
numbers are compared again.
Circuit Arrangement
Hardware Implementation of binary division

➢An n-bit positive divisor is loaded into register M

➢An n-bit positive dividend is loaded into register Q

➢Register AC (A) is Set to 0

➢Initial carry C is set to 0

➢After the division is complete, the n-bit quotient is in register Q and

the remainder is in register AC

Restoring Division (Unsigned Binary Division)

• Shift A and Q left one binary position

• Subtract M from A, and place the answer back in A
• If the sign of A is 1, set q0 to 0 and add M back to A
(restore A); otherwise, set q0 to 1
• Repeat these steps n times
Example Perform Division Restoring Algorithm Dividend = 8 Divisor = 3
 Non-Restoring Division Algorithm for Unsigned Integer

• The non-restoring division algorithm is more complex as compared to the

restoring division algorithm.
• Avoid the need for restoring A after an unsuccessful subtraction
• The Non-Restoring Division Algorithm is a method used to perform division
operations on unsigned integers without relying on restoring intermediate
remainders.

• It's an iterative approach that approximates the quotient and updates the
remainder in each iteration, leading to an accurate division result.
Non-restoring division algorithm, steps are described as follows:
Step 1: In this step, the corresponding value will be initialized to the registers, i.e., register A will contain value 0,
register M will contain Divisor, register Q will contain Dividend, and N is used to specify the number of bits in
dividend.
Step 2: In this step, we will check the sign bit of A.

Step 3: If this bit of register A is 1, then shift the value of AQ through left, and perform A = A + M. If this bit is 0,
then shift the value of AQ into left and perform A = A - M. That means in case of 0, the 2's complement of M is
added into register A, and the result is stored into A.
Step 4: Now, we will check the sign bit of A again.
Step 5: If this bit of register A is 1, then Q[0] will become 0. If this bit is 0, then Q[0] will become 1. Here Q[0]
indicates the least significant bit of Q.
Step 6: After that, the value of N will be decremented. Here N is used as a counter.
Step 7: If the value of N = 0, then we will go to the next step. Otherwise, we have to again go to step 2.
Step 8: We will perform A = A + M if the sign bit of register A is 1.
Step 9: This is the last step. In this step, register A contains the remainder, and register Q contains the quotient.
Non Restoring - Division
 Non - Restoring Division - Steps
If the sign of A is 1,
Shift A and Q left one binary position (Left Shift),
A+M → A;
Otherwise,
Shift A and Q left one binary position (Left Shift),
A-M → A;
n cycles
If the sign of A is 1,
set q0 to 0
Otherwise,
set q0 to 1,
Repeat these steps n times
After n cycles,
If the sign of A is 1,
A+M → A; (Restore)
Perform Division by using Non-Restoring Algorithm
Dividend = 8 Divisor = 3 M=0011 Q=1000
 Floating Point Representation
• Fixed-point notation (e.g., twos complement) - Very large numbers cannot be
represented, very small fractions.
• Floating - point notation

• This number can be stored in a binary word with three fields:

• Sign: plus or minus
• Significand S (Mantissa)
• Exponent E
• Base
• 976,000,000,000,000 can be represented as 9.76 * 1014, and
• 0.0000000000000976 can be represented as 9.76 * 10-14
IEEE – 754 Floating Point Representation
IEEE - Institute of Electrical and Electronics Engineers
• 3 basic components Excess 127 (Single Precision):
No. of bits for Exponent (k)= 8
• Sign of Mantissa Bias = (2𝑘−1 - 1) = (27 - 1) = 127

0 - positive number
1 - negative number
• Biased exponent
- An exponent field needs to represent both positive and negative exponents.
• A bias is added to the actual exponent in order to get the stored exponent.
• Normalized Mantissa
- Mantissa - part of a number in scientific notation or a floating-point number
- consists of its significant digits.
- Here we have only 2 digits, i.e. O and 1.
- So a normalized mantissa is one with only one ‘1’ to the left of the decimal.
 Floating Point Representation
IEEE-754 Floating Point Representation

• Single Precision Format (32 bits)

• Bias – Excess 127 - (E-127) Excess 127 (Single Precision):

Depends on the size of the registers No. of bits for Exponent (k)= 8
• Double Precision Format (64 bits) Bias = (2𝑘−1 - 1) = (27 - 1) = 127

• Bias – Excess 1023 - (E-1023)

Excess 1023 (Double Precision):

No. of bits for Exponent (k)= 11
Bias = (2𝑘−1 - 1) = (210 - 1) = 1023
Floating Point Representation – SP Format
• The representation used is known as a biased representation.
• The leftmost bit stores the sign of the number (0 = positive, 1 = negative).
• The exponent value is stored in the next 8 bits.
• Typically, the bias equals (2𝑘−1 - 1), where k is the number of bits in the binary exponent.
• The final portion of the word (23 bits in this case) is the Significand (Mantissa)
Floating Point Representation – Examples

negative
20 127 + 20 = 147

negative

normalized -20 127 - 20 = 107

The bias equals to (2K-1 – 1) → 28-1 – 1 = 127

 Floating Point Representation – Steps
Steps:
1. Binary Conversion
2. Normalize the Mantissa
3. Identify Sign bit
4. Find the Exponent with Bias Value
Need for Normalization :
• Multiple representation of mantissa
and exponent are possible
– creates confusion
Hence normalization needed
Floating Point Representation –
Normalization

Formula to make normalized FP values in Human Readable form

 Floating Point Representation –
Normalization

Decimal FP: 0.0234 → Not normalized, 0.234×10-1 --→ Normalized

(If the MSB bit after radix point is Non-zero –> Normalized)
Binary FP: 0.01011101 → Not normalized, 1.011101×2-2 → Normalized
(If the radix point is after the MSB ‘1’ -→Normalized)
 Floating Point Representation - Example
• Represent 85.125 in Floating Point Single Precision Format
• Steps: 3. Identify Sign bit
1. Binary Conversion sign = 0
85 = 1010101, 0.125 = 001
85.125 = 1010101.001 4. Find the Exponent with Bias Value
Biased exponent for Single precision :
2. Normalize the Mantissa Excess 127+6=133
1.010101001 x 26 Exponent = 133 = 10000101
Normalized Mantisa = 010101001
(add 0's to complete the 23 bits) E-127=6
E=133
The IEEE 754 Single precision is: Exponent = 133 = 10000101

0 10000101 01010100100000000000000
 Floating Point Representation - Example
• Represent 85.125 in Floating Point Double Precision Format
• Steps: 3. Identify Sign bit
1. Binary Conversion sign = 0
85 = 1010101, 0.125 = 001
85.125 = 1010101.001 4. Find the Exponent with Bias Value
Biased exponent for Double precision :
2. Normalize the Mantissa Excess 1023+6 =1029
1.010101001 x 26 Exponent = 1029 = 10000000101
Normalized Mantisa = 010101001
(add 0's to complete the 52 bits) E-1023=6
E=1029
The IEEE 754 Double precision is: Exponent = 1029 = 10000000101

0 10000000101 0101010010000000000000000000000000000000000000000000
 Exercise Problems
Express the number - (640.5)10 in IEEE 32 bit and 64 bit floating point format
Solution:
• IEEE 32 BIT FLOATING POINT FORMAT
MSB 8 bits 23 bits
sign Biased Mantissa/Significand
Exponent (Normalized)
Step 1: Express the given number in binary form
(640.5) = 1010000000.1* 20
Step 2: Normalize the number into the form 1.bbbbbbb
1010000000.1* 20 = 1. 0100000001* 29
Once Normalized, every number will have 1 at the leftmost bit. So IEEE notation is saying that there is no need to store
this bit. Therefore significand to be stored is 0100 0000 0100 0000 0000 000 in the allotted 23 bits
 Exercise Problems
Express the number - (640.5)10 in IEEE 32 bit and 64 bit floating point format
Solution:
• IEEE 32 BIT FLOATING POINT FORMAT
MSB 8 bits 23 bits
sign Biased Mantissa/Significand
Exponent (Normalized)
Step 3: For the 8 bit biased exponent field, the bias used is
2k-1-1 = 28-1-1 = 127
Add the bias 127 to the exponent 9 and convert it into binary in order to store for 8-bit biased
exponent. 127 + 9 =136 ( 1000 1000)

Step 4: Since the given number is negative, put MSB as 1

Step 5: Pack the result into proper format(IEEE 32 bit)

1 1000 1000 0100 0000 0010 0000 0000 000

 Exercise Problems
Express the number - (640.5)10 in IEEE 32 bit and 64 bit floating point format
Solution:
• IEEE 64 BIT FLOATING POINT FORMAT
MSB 11 bits 52 bits
sign Biased Mantissa/Significand
Exponent (Normalized)
Step 1: Express the given number in binary form
(640.5) = 1010000000.1* 20
Step 2: Normalize the number into the form 1.bbbbbbb
1010000000.1* 20 = 1. 0100000001* 29
Once Normalized, every number will have 1 at the leftmost bit. So IEEE notation is saying that there is no need to store this bit. Therefore
significand to be stored is 0100 0000 0100 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 in the allotted 52 bits.
 Exercise Problems
Express the number - (640.5)10 in IEEE 32 bit and 64 bit floating point format
Solution:
• IEEE 64 BIT FLOATING POINT FORMAT
MSB 11 bits 52 bits
sign Biased Mantissa/Significand
Exponent (Normalized)
Step 3: For the 11 bit biased exponent field, the bias used is
2k-1-1 = 211-1-1 = 1023
Add the bias 1023 to the exponent 9 and convert it into binary in order to store for 11-bit
biased exponent.
1023 + 9 =1032 ( 1000 0001 000)
Step 4: Since the given number is negative, put MSB as 1
Step 5: Pack the result into proper format(IEEE 64 bit)

1 1000 0001 000 0100 0000 0010 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000
 Exercise Problems
Represent -12.62510 in single precision IEEE-754 format.
Solution:
• Step #1: Convert to target base. -12.62510 = -1100.1012
• Step #2: Normalize. -1100.1012 = -1.1001012 × 23
• Step #3: Fill in bit fields. Sign is negative, so sign bit is 1.
Exponent is in excess 127 (not excess 128!), so exponent is represented as the
unsigned integer E = 3 + 127 = 130.
Leading 1 of significant is hidden, so final bit pattern is:
1 1000 0010 . 1001 0100 0000 0000 0000 000
 Floating Point Arithmetic Operations
• Floating Point Operations: –

• Addition – Directly Add mantissas

• Subtraction – Directly subtract Mantissas

• Multiplication Division – Direct multiplication

• Division – Direct Division

 Floating Point Arithmetic Operations
• Normal Addition
X = 1.01101 x 27 1.01101 x 27
Y = 1.10101 x 26 + 0.110101 x 27
10.001111 x 27

• Normal Subtraction
X = 1.01101 x 27 1.01101 x 27
Y = 1.10101 x 26 – 0.110101 x 27
0.100101 x 27
• Normal Multiplication 1.10012
X = 6.2510 = 110.012 = 1.1001 x 22
x 1.10012
Y = 12.510 = 1100.12 = 1.1001 x 23 10.011100012

• Normal Division
X = 3.7510 = 11.112 = 1.111 x 21 0.110011
Y = 95.62510 = 101 1111.1012 1.111 1.011111101
= 1.011111101 x 26
 Floating Point Arithmetic Operations –
Addition/Subtraction
• Steps 0.0234 → Not normalized
0.234×10-1 --→ Normalized
• Check for 0’s in the given decimal and normalize it. (If the MSB bit is Non-zero –> Normalized)
• Choose the number with smaller Exponent.
• Align Mantissas (Shift its Mantissa right to number of positions – difference
between exponents)
• Set the exponent of result = larger exponent
• Add or Subtract Mantissas and determine the sign
• Normalize the result, if necessary

For addition and subtraction, it is necessary to ensure that both operand

exponents have the same value
This may involves shifting the radix point of one of the operand to achieve
alignment
 Floating Point Arithmetic Operations
Some problems that may arise during arithmetic operations are:

• Exponent overflow: A positive exponent exceeds the maximum possible exponent value
and this may leads to + or - in some systems

• Exponent underflow: A negative exponent is less than the minimum possible exponent
value (eg. 2-200), the number is too small to be represented and maybe reported as 0

• Significand underflow: In the process of aligning significands, the smaller number may
have a significand which is too small to be represented

• Significand overflow: The addition of two significands of the same sign may result in a
carry out from the most significant bit
 Floating Point Arithmetic Operations –
Addition/Subtraction - Algorithm
Algorithm:
 Floating Point Arithmetic Operations –
Addition - Example
Step 1: Normalize Mantissa Step 3: Add Mantissa

Step 2: Make exponents equal

 Floating Point Arithmetic Operations –
Step 1: Normalize Mantissa
Subtraction - Example
Step 3: Subtract Mantissa

Step 2: Make exponents equal

Flowchart- Addition/Subtraction
Z=X+Y and Z=X-Y
 1.01101 x 27 1.01101 x 27
X–Y=Z X = 1.01101 x 27
– 0.110101 x 27
SUBTRACT
+ 0.110101 x 27
Y = 1.10101 x 26
10.001111 x 27 0.100101 x 27

Change sign of Y

X = 1.01101 x 27
X+Y=Z 0.100101 x 27
Y = 0.110101 x 27
no no Expoenents yes Add signed Results yes
ADD X = 0? Y = 0? Round result
Equal? significands normalized?

yes yes no no

Increment smaller yes Significand Shift significand RETURN

ZY ZX Z0
exponent = 0? left

no 1.0001111 x 28
1.00101 x 26
RETURN Shift significand RETURN Significand no Decrement
right overflow? exponent

10.001111 x 27 yes
1.00101 x 26
Significand Shift significand no Exponent
Y = 0.110101 x 2no7 = 0? right underflow?

yes

Put other number

1.0001111 x 28 Increment
Report underflow
in Z RETURN exponent

RETURN yes Exponent no RETURN

Report overflow
overflow?
Floating Point Arithmetic Operations –
Multiplication - Algorithm
Floating Point Arithmetic Operations –
Multiplication - Example
 XxY=Z X = 6.2510 = 110.012 = 1.1001 x 22
Y = 12.510 = 1100.12 = 1.1001 x 23
E1 = 127 + 2 = 129
E2 = 127 + 3 = 130
Floating-point E1 + E2 = 259
Multiplication
ET = 259 – 127 = 132

1.10012
x 1.10012
10.011100012

10.01110001 x 25
=1.001110001 x 26
110
Floating Point Arithmetic Operations –
Division - Algorithm
Floating Point Arithmetic Operations –
Division - Example
 X = 3.7510 = 11.112 = 1.111 x 21
Y¸X=Z
Y = 95.62510 = 101 1111.1012
= 1.011111101 x 26
E1 = 127 + 1 = 128
E2 = 127 + 6 = 133
E2 – E1 = 5

Floating ET = 127 + 5 = 132

Point Division

0.110011
1.111 1.011111101

0.110011 x 25
= 1.10011 x 24

113
 Non-Numeric Data Representation

• ASCII - American Standard Code for Information Interchange.

• EBCDIC character code - It is a character-encoding format used
by IBM mainframes. It is an 8-bit code and is NOT Compatible to
ASCII. It had been designed primarily for ease of use of punched
cards.
• Unicode - Under different platform, different software and different
language, able to exchange every character over the world
Character set should be large enough. Each character has only one
meaning.
 Character Representation ASCII

• ASCII (American Standard Code for Information Interchange)

Code.
• It uses 7 bits to represent one character.
• Use number 0 to 127 represent English character set.
• In order to enhance character set, ASCII character set use 8 bits to
represent one character.
• So it can represent some non-English character, common symbols
and mathematical symbols.
 MSB (3 bits)
Character Representation ASCII
LSB 0 1 2 3 4 5 6 7
(4 bits) 0 NUL DLE SP 0 @ P ‘ P
1 SOH DC1 ! 1 A Q a q • A - 1000001 - 64+1=65
2 STX DC2 “ 2 B R b r
3 ETX DC3 # 3 C S c s
• Here MSB - 100 and
4 EOT DC4 $ 4 D T d t • LSB – 0001
5 ENQ NAK % 5 E U e u
6 ACK SYN & 6 F V f v
7 BEL ETB ‘ 7 G W g w
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 m n ~
F SI US / ? O n o DEL
 Extended Binary Coded Decimal Interchange Code
• It had been designed primarily for ease of use of punched cards.
• It was used on IBM mainframes and midrange systems (AS/400).
• Another strength of EBCDIC was the availability of wider range of control characters for
ASCII.
• The character coding in this set is based on binary coded decimal.
• The contiguous characters in the alphanumeric range are represented in blocks of 10
starting from 0000 binary to 1001 binary.
• Other characters fill in the rest of the range.
• There are four main blocks in the EBCDIC code:

There are several different variants of EBCDIC. Most of these differ in the punctuation coding.
EBCDIC character code in hexadecimal indices
 Unicode
• Unicode is a character encoding standard that provides a unique number (code
point) for every character, regardless of platform, program, or language.

• Unicode supports characters from virtually all writing systems in use today.

• UTF-8, UTF-16, and UTF-32 are popular encoding schemes used to represent
Unicode characters.

• Unicode is a universal character encoding supporting a vast range of characters

from various scripts and languages. When working with Unicode, different
encoding schemes like UTF-8, UTF-16, or UTF-32 are used to represent the
characters in a computer-readable format.
Unicode Representation
 Practice Problems
• 1. Represent the following numbers in a) Sign and Magnitude b) 1s
complement and c) 2’s complement representation
i) +24 ii) - 24 iii) -8 iv) +13 v) -37
• 2. Multiply the following numbers using Booth algorithm.
i) 11* 13
ii) 11* -13
iii) -11* - 13
• 3. Show the step by step multiplication process using AC , QR and Q-1
registers for the below number
i) -15* 5
 Practice Problems
• 4. Multiply the following numbers using modified booth algorithm / Bit pair recoding( Try one
problem using the registers(A, Q , Q-1 ) method)
i) 23 * 10
ii) 14 * -2
iii) -12* 8
• 5. Divide the following numbers using restoring division algorithm
i) 14/7
ii) 28/11
• 6. Divide the following numbers using non- restoring division algorithm
i) 18/9
ii) 27/4
• 7. Represent the following number in IEEE- 754 format
i) 12312.13
ii) -14.5