CS2115 Computer Organization

2025/2026 Sem A

Chapter 2: Number Systems

Department of Computer Science

City University of Hong Kong

1
 Outline
• Decimal, Binary, Octal, Hexadecimal Systems
– Conversion between different systems
– Arithmetic
• Signed Number Representation
– Sign Magnitude Number
– Two’s Complement
• Floating-point Number
• BCD, ASCII, Gray …

2
 Outline
• Decimal, Binary, Octal, Hexadecimal Systems
– Conversion between different systems
– Arithmetic
• Signed Number Representation
– Sign Magnitude Number
– Two’s Complement
• Floating-point Number
• BCD, ASCII, Gray

3
 Decimal System
• 972.5 = 9×102 + 7×101 + 2×100 + 5×10-1

radix
10
𝑛− 1 Radix-10 System
𝑁 = ∑ 𝑎𝑖 ×10 𝑖, 𝑎𝑖 ∈(0 , 1 ,2 , 3 , 4 , 5 , 6 , 7 , 8 , 9)
𝑖=− 𝑚 generalize

R
𝑛− 1 Radix-R System
𝑁 = ∑ 𝑎𝑖 × 𝑹𝑖, 𝑎𝑖 ∈(0 , …, 𝑹−𝟏)
𝑖=− 𝑚
4
 Binary, Octal and Hexadecimal
• Binary system (Radix-2) 𝑛− 1
2

12.125 = (1100.001)2 𝑁 = ∑ 𝑎𝑖 ×𝟐 𝑖 , 𝑎𝑖 ∈(0 ,1)

12.125 = 1×23+1×22+0×21+0×20+0×2-1+0×2-2+1×2-3 𝑖=− 𝑚

• Octal system (Radix-8) 8

𝑛− 1
12.125 = (14.1)8 𝑁 = ∑ 𝑎𝑖 ×𝟖 𝑖 , 𝑎𝑖 ∈(0 ,1 , …, 7)
12.125 = 1×81+4×80+1×8-1 𝑖=− 𝑚

• Hexadecimal system (Radix-16) 16

𝑛− 1
𝟏𝟎,𝟏𝟏,𝟏𝟐,𝟏𝟑,𝟏𝟒,𝟏𝟓
12.125 = (C.2)16
𝑁= ∑ 𝑎𝑖 ×𝟏𝟔𝑖,𝑎𝑖 ∈(0,1,… ,9, A ,B ,C , D , E , F)
12.125 = C×160+2×16-1 𝑖=− 𝑚

5
Binary  Octal  Hexadecimal

6
 Binary  Octal  Hexadecimal

Octal 2 6 3 5 . 5 4 4 Hexadecimal 5 9 D . B 2

Binary 0 10110011101.101100100 Binary0 10110011101.10110010

7
 Conversion to Decimal
– (10101)2 = (1×24+ 0×23+1×22+ 0×21+1×20)10 = (21)10
– (1276)8 = (1×83+2×82+7×81+6×80)10 = (702)10
– (9DF8)16 = (9×163+D×162+F×161+8×160)10 = (40440)10

– (101.11)2 = (1×22+1×20+1×2-1+1×2-2)10 = (4+1+0.5+0.25)10 =(5.75)10

– (8A.E2)16 = (8×161+A×160+E×16-1+8×16-2)10
= (128 + 10 + 14/16 + 8/256)10
= (138.90625)10

8
 Conversion from Decimal
• For integers: Radix divide method

(12)10 = ( ? )2
(12)10 = ( 1100 )2

(1582)10 = ( ? )16 (1582)10 = ( 62E )16

9
 Conversion from Decimal
• For the fractional part: Radix multiply method
(0.72)10 = ( ? )2 (0.65)10 = ( ? )16
0.72×2 = 1.44 0.65×16 = 10.4

0.44×2 = 0.88 0.4×16 = 6.4

0.88×2 = 1.76 0.4×16 = 6.4

0.76×2 = 1.52 …
0.52×2 = 1.04 (0.65)10 = ( 0.A666 … )16
…

(0.72)10 = ( 0.10111 … )2

10
 Arithmetic
• Direct computation:
– Following essentially the same rule as in decimal, with a different radix
1001 1001 1001 1001
(1001)2+ (1101)2
+ 1101
1
+ 1101
1 + 1101 + 1101
0 10 110 10110

8B 8B
× 23 × 23
(8B)16 × (3)16
1 1A1

11
 Arithmetic
• Indirect computation
– First convert everything to decimal, then compute, finally convert back

(1001)2+ (1101)2 (9)10+ (13)10 (22)10 (10110)2

(8B)16 × (3)16 (139)10 × (3)10 (417)10 (1A1)16

12
 Outline
• Decimal, Binary, Octal, Hexadecimal Systems
– Conversion between different systems
– Arithmetic
• Signed Number Representation
– Sign Magnitude Number
– Two’s Complement
• Floating-point Number
• BCD, ASCII, Gray

13
 Sign Magnitude Numbers
• Naïve representation: the leftmost digit represents sign (sign
digit), the remailing represents the absolute value
– sign digit =0  positive; sign digit = 1  negative
sign magnitude

• Drawbacks
– Addition/subtraction need to consider different cases of different signs
– Complicate circuit and more computation effort
– Rarely used in practice

14
 2’s Complement
• Rules:
– Positive numbers are represented in the same way as in sign
magnitude numbers
– Negative numbers are represented as the complement of positive
numbers

• Complement: change 1 to 0, change 0 to 1

15
 2’s Complement
+127 01111111
pos(+)
• Consider a 8-bit binary system +126 01111110
+125 01111101
– Cut the number system in half
– Use 00000001 – 01111111 to indicate +1 00000001
positive numbers
0 00000000
– Use 10000000 – 11111111 to indicate -1 11111111
negative numbers -2 11111110

-127 10000001
neg(-)
-128 10000000

16
 Sign Bit
+127 01111111
pos(+)
• The most significant bit (MSB) of the +126 01111110
binary numbers is the signed bit +125 01111101

– The MSB is (0) for all positive numbers

• Except 00000000 +1 00000001
0 00000000
– The MSB is (1) for all negative numbers
-1 11111111
– Similar to sign magnitude numbers, this
-2 11111110
allows you to instantly determine whether a
number is positive or negative
-127 10000001
neg(-)
-128 10000000

17
 2’s Complement Process
• To calculate the representation of a negative number – using
2’s Complement Process

– First, complement all of the digits in a number.

– Second, add 1.

18
 2’s Complement Process
Example #1 5 = 00000101
Complement Digits

11111010
+1
Add 1
-5 = 11111011

Example #2 -13 = 11110011

Complement Digits

00001100
+1
Add 1
13 = 00001101
19
 2’s Complement Arithmetic
• Use the 2’s complement process to add the following numbers

POS 9 NEG (-9)

+ POS  + 5 + POS  + 5
POS 14 NEG -4

POS 9 NEG (-9)

+ NEG  + (-5) + NEG  + (-5)
POS 4 NEG - 14
20
 POS + POS → POS Answer
• If no 2’s complement is needed, use regular binary addition.

9  00001001
+ 5  + 00000101
14  00001110

21
 POS + NEG → POS Answer
• Take the 2’s complement of the negative number and use regular binary
addition.

9  00001001
+ (-5) + 11111011
4  1]00000100
8th Bit = 0: Answer is Positive
Disregard 9th Bit

00000101
 2’s
11111010 Complement
Process
+1
11111011
22
 POS + NEG → NEG Answer
• Take the 2’s complement of the negative number and use regular binary
addition.

(-9) 11110111
+ 5  + 00000101
-4  11111100
8th Bit = 1: Answer is Negative

11111100 00001001
To Check:   2’s
Perform 2’s
Complement 00000011 11110110 Complement
On Answer +1 Process
+1
00000100 11110111
23
 NEG + NEG → NEG Answer
• Take the 2’s complement of both negative numbers and use regular
binary addition.
2’s Complement
(-9)  11110111 Numbers, See
Conversion

+ (-5)  + 11111011 Process

In Previous Slides

-14  1]11110010
8th Bit = 1: Answer is Negative
Disregard 9th Bit
11110010
To Check: 
Perform 2’s
Complement 00001101
On Answer +1
00001110
24
 Outline
• Decimal, Binary, Octal, Hexadecimal Systems
– Conversion between different systems
– Arithmetic
• Signed Number Representation
– Sign Magnitude Number
– Two’s Complement
• Floating-point Number
• BCD, ASCII, Gray

25
 Binary with Fractional Part
• Recall that in binary number system
– (1001.1010)2 = 23 + 20 +2-1 + 2-3 =9.625

• If we fix the position of the point Point is here

0 0 0 0 0 0 0 1 0 1 1 0 1 0 1 1

• We are not able to represent small numbers with high precision

Point is here
1 0 0 0 0 1 0 1 0 0 0 0 0 0 0 0
• We are not able to represent large numbers

26
 Floating-Point Numbers
• 32-bit Binary Floating-Point Numbers
• value = (+/- Significand × 2E’)
– Sign bit: 0positive; 1negative
– Biased Exponent (E)
• Biased exponent E = real exponent value E’ + bias
• bias=2K-1 - 1, K=8 (8-bits of exponent), bias = 127
– Significand
• Normalized number: represent 1.xxx
• The most significant digit is always 1, so we do not need to store this information
• Thus, 23-bit is used to store 24-bit significand

27
 Floating-Point Numbers
• 32-bit Binary Floating-Point Numbers
1 10000100 10100000000000000000000
• value = (+/- Significand × 2E’)
negative
– Sign bit: 0positive; 1negative Biased exponent E = 132
– Biased Exponent (E) Real exponent value E’ = 132 – 127 = 5
Significand = 1 + 0.101000... (binary)
• Biased exponent E = real exponent value E’ + bias
= 1 + 0. 625 (decimal)
• bias=2K-1 - 1, K=8 (8-bits of exponent), bias = 127
So the value of the number is
– Significand -1.625 × 25
• Normalized number: represent 1.xxx
• The most significant digit is always 1, so we do not need to store this information.
• Thus, 23-bit is used to store 24-bit significand

28
 Outline
• Decimal, Binary, Octal, Hexadecimal Systems
– Conversion between different systems
– Arithmetic
• Signed Number Representation
– Sign Magnitude Number
– Two’s Complement
• Floating-point Number
• BCD, ASCII, Gray

29
 BCD: Binary-Coded Decimal
• 4-bit binary form to encode ten numbers from 0 to 9

30
 BCD: Binary-Coded Decimal
• 4-bit binary form to encode ten numbers from 0 to 9

31
 BCD: Binary-Coded Decimal
• 4-bit binary form to encode ten numbers from 0 to 9

32
 BCD: Binary-Coded Decimal
• 4-bit binary form to encode ten numbers from 0 to 9

137 10001001 0001 0011 0111

33
 ASCII Code
• ASCII: American Standard Code for Information Interchange
• ASCII encodes 128 specified characters into 7-bit integers

34
 Gray Code
• An encoding of numbers so that adjacent numbers have a
single digit differing by 1

35