Worksheets For Computer Science Orientation: Daniel Balls

Worksheets for
Computer Science Orientation

Version 1.1, 22 January 2007
Daniel Balls
This document is in the public domain.
Funding for this project was provided by the Hewlett Foundation

through its Engineering Schools of the West Initiative.
CS 160 Name ________________________
Computer Science Orientation Section Meeting Time____________
Logic Gates Worksheet
In order to add, subtract, and compare numbers, computers manipulate 0s and 1s.
Operations that manipulate individual 0s and 1s are called Boolean operations. A logic
gate is a mechanism that performs a Boolean operation by taking one or more 0s and 1s
as inputs and producing a 0 or 1 as output.
The Boolean operations we will study are either unary or binary. A unary operation has a
single input, and a binary operation has two inputs. Finding the output of a logic gate is
simple, once a few principles are understood.
A
A f f(A ) f f(A , B )
B
A unary operation (one input). A binary operation (two inputs).
The only unary Boolean operation is the NOT function. The NOT gate returns the
complement of the operand. An input of 1 results in an output of 0; conversely, an input
of 0 results in an output of 1. The latter of the two possibilities for the NOT operation is
shown below.
0 N O T 1
The NOT gate returns the complement of the input.
The most common binary Boolean operations include AND, OR, and XOR (short for
‘exclusive OR’). If at least one of the two operands is 1, the OR operation gives an
output of 1. The output of the XOR operation is 1 when exactly one of the operands is 1.
The AND operation returns 1 only if both inputs are 1. There are four possible
combinations of input pairs for each of these three gates. One of these combinations for
each gate is shown below.
0 O R 1
1
The OR gate returns 1 when at least one of its inputs is 1.
0
1 X O R 0
1
The XOR gate returns 0 when both operands are 1 or both operands are 0.
1
A N D 0
0
When at least one of the operands is 0, the output of the AND gate is 0.
Other logical gates can be formed by appending the NOT gate to any of the binary gates.
These gates are denoted by NAND (NOT AND), NOR (NOT OR), and XNOR (NOT
XOR). The output of these gates can be found by first determining the output of the
binary gate and then using this output as the input of the NOT gate, as shown below.
1
A N D 0 N O T 1
0
1
N A N D 1
0
The NAND gate is equivalent to an AND gate followed by a NOT gate.
The symbols that we have used above for the different logic gates are easily
distinguishable and widely used, with one difference: the name of the gate is not written
inside the symbol. Henceforth we will follow that practice. The symbols for the seven
logic gates are as follows:
A N D O R X O R N O T N A N D N O R X N O R
These are the symbols for the seven logic gates presented in this worksheet.
1
One way to represent a Boolean function is through the use of a truth table. The first
column(s) give all possible input combinations. The values in the last column(s)
represent the output of the function given the inputs that precede it. Such tables are
called truth tables. An example of the truth tables for the NOT and OR operations are
shown below (I1 stands for the first input, I2 stands for the second input and O represents
the output):
I1 O I1 I2 O
1 0 0 0 0
0 1 1 0 1
0 1 1
1 1 1
Truth table for the NOT gate. Truth table for the OR gate.
An applet that will help you become more familiar with the basic logic gates can be
accessed at eecs.oregonstate.edu/~quinn/education/JavaApplets/applets/LogicGates.html.
1. Make sure the option 1 Gate is selected.

2. Double-click on the rectangle and choose one of the logic gates available.
3. Next change the values of the inputs by clicking on the values.
4. Repeat steps 2 and 3 for all the logic gates.
5. Notice than when gate is selected, a table appears to the right. This is a truth table
for the selected gate. Once inputs have been selected, the row that corresponds to
these inputs is highlighted.
Critical Thinking: Answer the following questions, using the applet whenever helpful.
Q1: Using the space below, explain in your own words the difference between the XOR
gate and the NOR gate.
Q2: Which pairs of input values will yield an output value of 1 for the NAND gate?
Which pairs of input values will yield an output value of 0 for the AND gate?
2
Q3: In the space below, construct truth tables for the five other Boolean operations listed
above.
3
Combinatorial Circuits
An arrangement of more than one logic gate is called a combinatorial circuit. The gates in
a circuit are linked. Often it is the case that one gate has an output that serves as the input
for one or more gates within the circuit. The use of logical gates and combinatorial
circuits play a fundamental role in how computers operate. An example of a
combinatorial circuit is shown below. Notice that in this circuit the lower input serves as
the input for the NOT gate as well as the second input of the XNOR gate.
An example of a combinatorial circuit.
Use the applet

eecs.oregonstate.edu/~quinn/education/JavaApplets/applets/LogicGates.html to help you
understand more effectively the ideas behind combinatorial circuits.
1. Select the option labeled 2 Gates.

2. Double click on each rectangle and select a logic gate for each one.
3. Change the inputs to investigate the nature of the circuit.
4. Create a different circuit by selecting different gates and investigate its nature.
Notice how the truth table changes as the logic gates are changed.
5. Select the options for 3 Gates (2 inputs), 3 Gates (4 inputs), and 4 Gates, and
repeat steps 2 – 4.
Critical Thinking: Answer the following questions using the applet whenever helpful.
Q4: Construct a truth table for the combinatorial circuit drawn above.
4
Q5: For each of the circuits below, identify the pair(s) of input combinations that
produce an output of 1.
Q6: Find a combination of gates (using only AND, NAND, OR and NOR gates—but not
necessarily all of them) that will make the following circuit equivalent to the XOR
gate.
? ?
5
Challenge Question (Optional): Create a combinatorial circuit with three inputs (I1, I2,
and I3) and two outputs (O1 and O2) that has as its truth table the one shown below. For
what purpose could this circuit be used?
I1 I2 I3 O1 O2
0 0 0 0 0
0 0 0 0 1
0 1 0 0 1
0 1 1 1 0
1 0 0 0 1
1 0 1 1 0
1 1 0 1 0
1 1 1 1 1
6
7
CS 160 Name ________________________
Binary Number System Worksheet
Digital computers store information in the form of 0s and 1s. An important use of digital
computers is performing computations. That means we must be able to store numbers as
patterns of 0s and 1s. The binary number system is a way of representing integers using
only 0s and 1s.
Before we explore the binary system, it will be helpful to review the number system we
are more familiar with—the decimal system. The decimal, or base-ten, number system
uses ten digits—0 through 9. The position of each digit in a number is associated with a
specific power of ten. The rightmost position in a decimal representation has a value of
100 (or 1), followed on the left by 101 (or 10), then 102 (or 100), then 103 (or 1000), and
so on. For the number represented in decimal form by 247, the digit 7 is associated with
100, the 4 with 101, and the 2 with 102. The value of the representation (e.g., 247) can be
found by multiplying each digit by its position’s value: the representation 247 has the
value (2·102 + 7·101 + 4·100).
2 4 7 = 2·100 + 4·10 + 7·1

10 10 100
2 1
The value of the decimal representation 247.
The binary number system has a structure similar to the decimal system. In the binary, or
base-two, system there are only 2 digits—0 and 1. The position of each bit (binary digit)
is associated with a specific power of 2—20, 21, 22, 23, etc. (moving from right to left).
For example, the decimal equivalent of the binary representation 1101 can be found by
multiplying each digit by its position value: 1·23 + 1·22 + 0·21 + 1·20 = 13. Therefore,
the binary number 1011 is equivalent to the decimal number 13.
1101 = 1 1 0 1 = 1·8 + 1·4 + 0·2 + 1·1 = 13

23 22 21 20
binary decimal
representation representation
The binary number 1101 is equivalent to the decimal number 13.
When we express a binary number verbally, it is helpful to say the digits in order from
left to right. For example, as opposed to “one thousand one hundred and one,” the binary
number 1101 would be pronounced “one-one-zero-one.”
As a computer science major, you need to become adept with the binary number system.
The following activity will help you achieve this goal.
8
Exercise: Fill in the following blanks with the binary equivalent of the decimal number
given. To make sure you are on track, some of the binary representations have been
filled in for you.
Binary Decimal Binary Decimal Binary Decimal Binary Decimal

__________ 1 __________ 26 __________ 51 __________ 76
__________ 2 __________ 27 __________ 52 __________ 77
__________ 3 ___11100__ 28 __________ 53 __________ 78
__________ 4 __________ 29 __________ 54 __________ 79
__________ 5 __________ 30 __________ 55 __________ 80
__________ 6 __________ 31 __________ 56 __________ 81
__________ 7 __________ 32 __________ 57 __________ 82
__________ 8 __________ 33 __________ 58 __________ 83
__________ 9 __________ 34 __________ 59 __________ 84
__________ 10 __________ 35 __________ 60 _ 1010101_ 85
__________ 11 __________ 36 __________ 61 __________ 86
__________ 12 __________ 37 __________ 62 __________ 87
___1101___ 13 __________ 38 __________ 63 __________ 88
__________ 14 __________ 39 __________ 64 __________ 89
__________ 15 __________ 40 __________ 65 __________ 90
__________ 16 __________ 41 __________ 66 __________ 91
__________ 17 __________ 42 __________ 67 __________ 92
__________ 18 __________ 43 __________ 68 __________ 93
__________ 19 __________ 44 __________ 69 __________ 94
__________ 20 __________ 45 __________ 70 __________ 95
__________ 21 __________ 46 _ 1000111_ 71 __________ 96
__________ 22 __________ 47 __________ 72 __________ 97
__________ 23 __________ 48 __________ 73 __________ 98
__________ 24 __________ 49 __________ 74 __________ 99
__________ 25 __________ 50 __________ 75 __________ 100
Critical Thinking:
a. What is the binary representation of the decimal value 17?

b. What is the binary representation of the decimal value 34?
c. What is the binary representation of the decimal value 68?
d. What is the binary representation of the decimal value 22?
e. What is the binary representation of the decimal value 44?
f. What is the binary representation of the decimal value 88?
9
Q1: Look for a relationship between the binary representations of 17, 34, and 68. The
same relationship should hold between the binary representations of 22, 44, and 88. What
is the relationship? What is the analogous rule in the decimal system?
Q2: How many bits are there in the binary representation of the decimal value 100?
What is the largest decimal number you can represent with that number of bits?
Converting Between Binary and Decimal Representation
There are two methods for converting numbers from their decimal to binary
representations. One method involves successively dividing by 2. The quotient at each
step will always be a whole number with a remainder of 0 or 1 (think about why this is
true). If the remainder is recorded but not included in the next step of the division, the
process will continue until the quotient is 0. The string of remainders (going from last to
first) is the binary representation (written left to right) of the decimal number. An
example is shown in the next figure.
68 ÷ 2 = 34 R0
34 ÷ 2 = 17 R0
17 ÷ 2 = 8 R1
…
8÷2=4 R0 1 0 0 0 1 0 0 = 68 (base 10)

4÷2=2 R0
2÷2=1 R0
1÷2=0 R1
The division method for converting decimal numbers into their binary representation is shown here. The
remainders are recorded and become the digits of the binary number. This example shows 68 base ten is
1000100 base 2.
An alternative method of finding the binary equivalent of a decimal number makes use of
one’s knowledge of the powers of 2—2, 4, 8, 16, 32, 64, 128, etc. The procedure begins
by determining the greatest power of 2 less than or equal to the decimal number being
converted. For example, when converting 68 into its binary equivalent, we notice that
10
64 = 26 is the largest power of 2 less than 68. We subtract this power of 2 from the
original number, and repeat the process: 68 – 64 = 4, so what is the largest power of 2
less than or equal to 4? The answer is 22 = 4. We stop when the difference between the
number and the power of 2 is zero. So we have determined that 68 consists of one 64 and
one 4. That means the binary representation of 68 is found by creating a string of bits that
has a 1 in the places representing 64 and 4, and a 0 everywhere else: 1000100.
Converting a binary number into its decimal equivalent is related to the method just
described. We simply add up the powers of 2 represented by the 1s in the binary
representation. For example, the number 1011001 can be converted to a decimal
representation as follows: 1·26 + 0·25 + 1·24 + 1·23 + 0·22 + 0·21 + 1·20 = 64 + 16 + 8 +
1 = 89. The following exercises will help you improve your ability to convert numbers
between the decimal and binary system.
Critical Thinking: Answer the following questions.
Q3: Convert the decimal numbers 14 and 49 into their binary equivalents using the first
method described above (successive divisions by 2).
Q4: Convert the decimal numbers 19 and 57 into their binary equivalents using the
second method described above (greatest power of 2 less than or equal to the
number).
11
Q5: Using your favorite method, convert the decimal numbers 161 and 585 into their
binary equivalents.
Q6: Convert each of the following binary numbers into their decimal equivalent.
1011000110 1100101101110 10000001111
12
Challenge Question: Professor I. M. Smart has invented a new kind of computer storage
device in which fundamental unit of information is a trit (trinary digit). A trit takes on
one of the three values 0, 1, or 2.
In your own words explain how to convert a number from its trinary representation to its
decimal representation. Then explain how to convert a decimal number to its trinary
equivalent.
13
CS 160 Name ________________________
Octal and Hexadecimal Integers Worksheet
In the previous worksheet we worked with a number system that is very important to
computer operation—the binary system. Another non-decimal number system that plays
an important role in computer science is the hexadecimal system. This number system
has 16 digits. Since there are only 10 Arabic numerals, we need additional symbols to
represent the values 10, 11, 12, 13, 14, and 15. We’ll let the letter A represent the value
10, B represent the value 11, and so on, through F, which will represent the value 15.
An example of a hexadecimal number is 4C7. To compute the decimal value of this

number, it is important to know that each place value is associated with a power of 16 in
the same way the powers of ten are associated with the decimal system and the powers of
two are associated with the binary system. So 4C7 is equivalent to the following decimal
number:
4·162 + C·161 + 7·160 = 4·256 + 12·16 + 7·1 = 1024 + 192 + 70 = 1223.
Another number system that has application in the field of computer science is the octal
number system. This system has 8 digits—0, 1, 2, 3, 4, 5, 6, and 7. As its positional
values, the octal system uses the powers of 8 (but you may have guessed that by now).
Then 452 in the octal number system is equivalent to the decimal number:
4·82 + 5·81 + 2·80 = 4·64 + 5·8 + 2·1 = 256 + 40 + 2 = 298.
The next activity will make you more familiar with hexadecimal and octal numbers.
14
Exercise: Fill in the blanks with the appropriate hexadecimal and octal integers.
Decimal Octal Hex. Decimal Octal Hex.

1 __________ __________ 51 __________ __________
2 __________ __________ 52 __________ __________
3 __________ __________ 53 __________ __________
4 __________ __________ 54 __________ __________
5 __________ __________ 55 __________ __________
6 __________ __________ 56 __________ __________
7 __________ __________ 57 ____71____ __________
8 __________ __________ 58 __________ __________
9 __________ __________ 59 __________ __________
10 __________ ____A_____ 60 __________ __________
11 __________ __________ 61 __________ ____3D____
12 __________ __________ 62 __________ __________
13 __________ __________ 63 __________ __________
14 __________ __________ 64 __________ __________
15 __________ __________ 65 __________ __________
16 __________ __________ 66 __________ __________
17 __________ __________ 67 __________ __________
18 __________ __________ 68 __________ __________
19 __________ __________ 69 __________ __________
20 __________ __________ 70 __________ __________
21 __________ __________ 71 __________ __________
22 ____26 ___ __ _16_ __ 72 __________ __________
23 __________ __________ 73 __________ __________
24 __________ __________ 74 __________ __________
25 __________ __________ 75 __________ __________
26 __________ __________ 76 __________ __________
27 __________ __________ 77 __________ __________
28 __________ __________ 78 __________ __________
29 __________ __________ 79 __ 117 _ _ __________
30 __________ __ __ ____ 80 __________ __________
31 __________ __________ 81 __________ __________
32 __________ __________ 82 __________ __________
33 __________ ___21____ 83 __________ __________
34 __________ __________ 84 __________ __________
35 __________ __________ 85 __________ __________
36 __________ __________ 86 __________ __________
37 __________ __________ 87 __________ __________
38 __________ __________ 88 __________ __________
39 __________ __________ 89 __________ ____59____
40 __________ __________ 90 __________ __________
41 __________ __________ 91 __________ __________
42 __________ __________ 92 __________ __________
43 __________ __________ 93 __________ __________
44 __________ __________ 94 __________ __________
45 __________ __________ 95 __________ __________
46 __________ __________ 96 __________ __________
47 __________ __________ 97 __________ __________
48 __________ __________ 98 __________ __________
49 __________ __________ 99 __________ __________
50 ___ 62_ ___ __________ 100 __________ __________
15
Hexadecimal and Octal Conversions
To convert an octal number to its hexadecimal equivalent, first convert the number to its
binary form and then put the bits into groups of four, working from right to left. Once the
bits are in groups of four, change each group into the appropriate hexadecimal digit.
A detailed illustration of the process follows. Suppose we want to convert the octal
number 736 to its hexadecimal equivalent. First we’ll convert this number to its binary
equivalent. The simplest way to do this is to write each digit (i.e. 7, 3, and 6) in a binary
form consisting of exactly three bits. If a digit requires less than three bits (the digits 0
through 3), one or more leading zeros are placed at the beginning of the string—000, 001,
010, and 011. All the other octal digits (4 – 7) have a binary representation that requires
exactly three bits, and no leading zeros are required. In our example, the digit 7 is
equivalent to the binary 111, 3 is equivalent to the binary 011, and 6 is equivalent to the
binary 110. The binary representation of the octal 736 is found by concatenating these
binary strings: 111011110.
Once the binary equivalent of our original octal number has been obtained, we’ll use the
fact that 16 is also a power of 2. The binary string representing the octal integer is
arranged, from right to left into groups of four bits (because 16 = 24) with leading zeros
added to the leftmost group if necessary. Thus 111011110 is arranged in this manner:
0001 1101 1110. To find the hexadecimal equivalent of the original octal number, we
write down the hexadecimal digit corresponding to each group of four bits—0001 is
equivalent to the hexadecimal 1, 1101 is equivalent to D, and 1110 is equivalent to E.
Therefore the hexadecimal integer 1DE is equivalent to the octal number 736.
To convert a number from its hexadecimal form to its octal equivalent, we reverse the
process. Convert the hexadecimal number to its binary form (by writing the hexadecimal
digits as binary strings that are four bits long), regroup the bits (this time into groups of
three), and write the octal digits that correspond to these strings of size three. For
example, to find the octal equivalent of the hexadecimal number 3C8E, write the binary
equivalents (made up of four bits each) of the digits 3, C, 8, and E: 3 is written 0011, C
is written 1100, 8 is written 1000, and E is written 1110. Thus the hexadecimal number
3C8E has the binary form 0011110010001110. Notice it’s much easier to write long
strings of bits in hexadecimal form! Now we regroup the bits, starting at the right, into
strings of length three: 011 110 010 001 110. Each of these strings of three bits
represents an octal digit: 011 becomes 3, 110 becomes 6, 010 becomes 2, 001 becomes 1,
and 110 becomes 6. We have just converted the hexadecimal number 3C8E into its octal
equivalent: 36216.
16
Q1: Convert the following hexadecimal integers to their equivalents in the octal number
system: CFE, 12D9, and 1F39A.
Q2: Convert the following octal integers to their equivalents in the hexadecimal number
system: 6655, 7426, and 175264.
Hexadecimal and Decimal Conversions
Although the procedure for converting numbers between the hexadecimal and decimal
number systems is not as elegant as converting between octal and hexadecimal, it should
have a familiar feel: to convert a decimal number to its hexadecimal equivalent, first find
the decimal number’s binary form, then arrange the bits from right to left into groups of
four. For example, to find the hexadecimal equivalent of the decimal 269, we first need
to obtain the binary form of 269. Using either of the methods present in the last
worksheet, it can be shown that the binary equivalent of 269 is 100001101. Regrouping
this binary form into strings of 4 bits (and adding three leading zeros) shows that the
decimal 269 can be written as 0001 0000 1101. These three strings correspond to the
hexadecimal 1, 0, and D, respectively. Therefore the hexadecimal equivalent of the
decimal 269 is 10D.
To find the decimal equivalent of the hexadecimal integer 82C, we’ll first write this
number in a binary form: 1000 0010 1100 = 100000101100. Next, we add appropriate
powers of two to obtain the result: 1·211 + 1·25 + 1·23 + 1·22 = 2048 + 32 + 8 + 4 = 2092.
Therefore the hexadecimal number 82C has an equivalent decimal form of 2092.
17
Q3: Convert the following hexadecimal integers to their decimal equivalents: 20A, F39,
and 1552.
Q4: Convert the following decimal integers to their equivalents in the hexadecimal
number system: 161, 2576, and 4095.
Q5: In your own words describe a procedure that one could use to convert an octal
integer into an equivalent decimal integer.
Q6: In your own words describe a procedure that one could use to convert a decimal
number into an equivalent one in the octal system.
18
Q7: Using the method you described in Q5, convert the following octal numbers into
their decimal equivalents: 335, 552, and 2156.
Q8: Using the method you described in Q6, convert the following decimal numbers into
their octal equivalents: 171, 888, and 1945.
Challenge Question: What is the largest integer that can be represented by…
a. two bits?
b. two octal digits?
c. two decimal digits?
d. two hexadecimal digits?
e. two digits from a base-k number system?
19
CS 160 Name ________________________
Two’s Complement Worksheet
Up to this point we have seen how it is possible to represent whole numbers (e.g. 0, 1, 2,
3, …) using bits. But can one represent the integers (i.e., the counting numbers and their
negative counterparts) using the binary system? The answer to this question is yes. The
most common system for representing the integers is called two’s complement notation.
Although most computers today use a 32 bits to represent the integers, learning the two’s
complement system with that many bits would be too cumbersome. Therefore, we will
study the features of the two’s complement system using only 6 bits. As we shall soon
see, using a 6-bit pattern, we are able to represent all the integers from -32 to 31. It is
important to note that the 6-bit pattern and the 32-bit pattern are identical in structure, so
studying the 6-bit pattern does have merit. The 32-bit pattern is used in actual computers
so that more integers (from -2,147,483,648 to 2,147,483,647) can be represented. The
table below contains some of the representations of the integers in a 6-bit two’s
complement system.
Value
Represented Bit Pattern
31 011111
30 011110
… …
5 000101
4 000100
3 000011
2 000010
1 000001
0 000000
–1 111111
–2 111110
–3 111101
–4 111100
–5 111011
… …
–30 100010
–31 100001
–32 100000
In any two’s complement system, the leftmost bit represents the sign (+ or –) of the
integer. For this reason, it is called the sign bit. The integer is negative when the sign bit
is 1; the integer is either zero or positive when the sign bit is 0. To represent the positive
integers leading zeros are added as needed to the left of the normal binary representation
of the integer. For example the number 23, which has a binary representation 10111, is
represented as 010111 in a 6-bit two’s complement system.
20
The easiest way to find the representation of a negative integer is to begin with its
positive counterpart. For example, to find the two’s complement representation for the
integer –23, we start with the two’s complement representation of 23: 010111. At this
point there are two methods that can be used to determine the 6-bit two’s complement
representation of –23. Both methods are described below.
The first method is referred to as the ‘copy and complement’ method. Beginning at the
right, the bits of the positive integer are copied up to and including the first 1 bit. In our
example, since the rightmost digit is 1, it is the only digit copied: __ __ __ __ __ _1_.
The remaining digits of the negative integer are the complement of the remaining bits of
the positive integer. The complement of a pattern is found by changing all 0s to 1s and
all 1s to 0s. Thus, the complement of the remaining portion of 23 is 1 0 1 0 0 1, and
the integer –23 is written 101001 in two’s complement notation
Another method to find the negation of a number in two’s complement notation could be
named the ‘complement and add 1’ method. We obtain the complement of the original
number and add one to it. Using the same example of 23, we begin by determining the
complement of 010111, which is 101000. Adding 1 to this complement yields the same
result as before: 101001. Notice that regardless of what method is used, a positive
integer—which always begins with 0—will always have an opposite or negation that
begins with a 1—and is therefore negative.
Exercise: Using either of the methods described above, find the negation of each of the
following numbers, represented in 6-bit two’s complement notation:
001110 negation: ____________________________________
010010 negation: ____________________________________
110011 negation: ____________________________________
000000 negation: ____________________________________
Critical Thinking: Answer the following question.
Q1: In your own words and based on evidence, explain what happens when you negate
the negation of a number in two’s complement notation.
21
The ability to negate a number written in two’s complement form is crucial in finding the
value of any number written in a two’s complement system—for example what integer
does the number 110110 represent? The first task is to determine whether the integer is
positive or negative. This factor depends on the sign bit—the first bit of the string. If the
first bit is 0, finding the value of the number amounts to converting the number—which
is basically in binary form—into its decimal equivalent. If however the number’s sign bit
is 1 (and therefore the number is negative), treating the number as if it is part of the
binary system will not work. Instead, we need to find the magnitude of the number and
then place a negative sign in front of this magnitude. And what is the magnitude of a
negative number? It is that number’s negation. In our example, the negation of 110110
is 001010 and this number corresponds to the binary number 1010. This binary number
has a decimal equivalent of 10, which means 110110 is equivalent to the negation of 10,
or –10.
Exercise: Convert these 6-bit, two’s complement integers into decimal form.
101000 ______________________________________________
110011 ______________________________________________
111111 ______________________________________________
Exercise: Convert these decimal integers to 6-bit, two’s complement form.
14 ______________________________________________
–27 ______________________________________________
–11 ______________________________________________
Binary and Two’s Complement Addition
One method of verifying that 101001 represents –23 is to add this string to the two’s
complement representation of 23 and check the result. If the sum is 0, which is the sum
of 23 and –23, then we can be sure that 101001 is indeed the representation of –23.
Adding numbers written in two’s complement notation is equivalent to adding numbers

in the binary system. To add two binary numbers, one must be familiar with some basic
base-2 addition facts, much like we must understand basic base-10 addition facts.
Fortunately, there are only a few facts to know!
0+0=0 0+1=1 1+0=1 1 + 1 = 10 10 + 1 = 11
22
The base-10 addition practice of ‘carrying a leftover’ is used when adding binary
numbers. Thus if the binary number 110 is added to the number 111, the result is 1101 as
expected and is described below.
1
1 110
110
+1 1 1 1
+ 111 1101
01
The 1 and 1 in the middle column add to The 1 and 1 in the leftmost column, combined
10 and the 1 (from 10) is carried to the with the 1 carried from the middle column, add to
leftmost column. 11. The leading 1 (from 11) is carried once again.
This method is analogous to addition in the decimal system. However, when adding
numbers in two’s complement notation, a change in the familiar method is necessary
since the number of bits of the sum must be equal to the number of bits in each of the
addends. In other words, when two numbers that are represented in a 6-bit two’s
complement system are added together, the result must also have exactly six bits. What
happens when there is an extra 1 that, according to our decimal rule, should be put into a
seventh bit position? The answer is that it is simply discarded. (This may be
counterintuitive, but it works!)
Using our knowledge of addition in the two’s complement system, we can now compute
the sum of 101001 and 010111, which as shown below does equal 0 (or in 6-bit two’s
complement notation 000000).
This carried 1
is discarded.
1 1 1 1 1
101001
+ 1 010111
000000
The sum of 101001 and 010111 is 00000, which means they are negations of each other.
One of the powerful features of a two’s complement system is that it contains the
structure to not only perform additions, but also subtractions. At the foundation of this
process is the fact that any subtraction problem can be rewritten as an addition problem
[for example, 12 – 8 = 12 + (–8)]. So if a computer has the ability to perform addition
(i.e., if it has an adder) it will also be able to perform subtraction. The subtraction
problem 12 – 8 could be performed by writing the integers 12 and –8 in two’s
complement notation and then adding them. This addition is shown below. Note that
while we could have used a 6-bit (or 32-bit) system, a simpler 5-bit system is sufficient,
23
since the positive integer is less than or equal to 15 and the negative integer is greater
than or equal to 16.
This 1 is
discarded.
1
01100
+ 1 11000
00100
A model of the subtraction problem 12 – 8 using two’s complement notation. The number 01100
represents 12 and 11000 represents –8. The expected result 4 will is evident after the leftmost ‘carried 1’
(created by the addition of 1 and 1 in the leftmost column) is discarded.
Exercise: Using a 6-bit two’s complement system, perform the following addition and
subtraction problems:
23 + 7 –13 – 18 28 – 19 16 + 20
Did you run into a problem with the last sum (16 + 20)? If you did everything correctly,
your sum should have been 100100. But this doesn’t make sense, since the sum of two
positive numbers is positive. The integer 100100 begins with a one and is therefore
negative. The solution to this conflict has to do with a phenomenon called overflow.
This problem is due to the fact that the integer 36, which is the true sum of 16 and 20,
cannot be represented in a 6-bit system. Whenever the sum of two numbers cannot be
represented in the current two’s complement system, overflow occurs. The problem of
overflow is serious because, unless the computer reports the error, the computation could
return an erroneous result.
24
An applet that will help you become more familiar with the concept of overflow is found
at eecs.oregonstate.edu/~quinn/education/JavaApplets/applets/TwosComplement.html.
1. Make sure the option Demo Mode is selected.

2. Click on the individual bits to change them from 1 to 0 and 0 to 1.
3. See if you can get the applet to achieve overflow. The circle above the word
Overflow will turn red when overflow occurs.
4. By clicking on the arrow to the right of number 4 you can change the number of
bits in the system to 6 or 8.
Critical Thinking: Answer the following questions, using the applet whenever helpful.
Assume we are using 6 bits to represent two’s complement integers.
Q2: Under what conditions is the overflow bit set (i.e. what types of integers being
added together will cause overflow to occur?
Q3: What is the fewest number of bits set to 1 that causes overflow?
Q4: What is the most number of bits set to 1 that doesn’t cause overflow?
Challenge Question: Of the two techniques for negating integers described in this
worksheet (‘copy and complement’ and ‘complement and add one’), which do you think
would be easiest to incorporate into a computer’s circuitry? Explain your answer.
25
CS 160 Name ________________________
Machine Language Worksheet
In this worksheet we will explore how the major components of a computer work
together to perform simple tasks. We begin with the architecture of an extremely
primitive computer, which is a slight modification of the computer described in the ninth
edition of J. Glenn Brookshear’s Computer Science: An Overview, Addison-Wesley
(2007).
The central processing unit (CPU) is the circuitry that controls how information
(in the form of bit patterns) is manipulated and stored. The CPU is made up of two parts
—the logic/arithmetic unit and the control unit. The logic/arithmetic unit has the ability
to perform simple mathematical calculations in addition to other operations. The control
unit contains many individual cells, called registers, of which there are two types:
general-purpose registers and special-purpose registers. The function of general-purpose
registers is to temporarily store bit patterns that will be manipulated by the
logic/arithmetic unit. There are two special-purpose registers: the instruction register and
the program counter. The instruction register contains the instruction the control unit is
currently carrying out. The program counter contains the address of the next instruction
to be performed.
Another critical part of computer architecture is the main memory. The main
memory consists of cells. Each cell has a unique address and contains 8 bits. These bits
may correspond to a program instruction or to a data value. Bit patterns are transferred to
and from the main memory to the CPU through a series of wires called a bus.
A Sample Machine
We will discuss how the CPU and main memory of a simple computer work
together to perform basic functions. Our machine has 16 general-purpose registers,
numbered in hexadecimal notation from 0 to F. Main memory consists of 256 cells—
each having as its address a hexadecimal number between 00 and FF, as shown in the
diagram below.
Computer programmers typically write programs in a high-level language, but in

order for computer programs to be executed on a computer, they must be translated into
low-level instructions called machine language. We will deal with the operations of our
computer at the machine language level. In our simple computer, a machine instruction
is a sequence of 16 bits that directs the CPU to perform a particular task. An example of a
machine instruction may be 0010 1101 0011 1100. We’ll use hexadecimal notation to
simplify the reading and writing of these instructions. In this system the example
instruction above would be written 2D3C. We’ll now explain what purpose each of these
four digits has in the machine language.
26
CPU Main Memory
Registers Memory
0 Program Address Cells Address
Counter Bus 00 01
1
02 03
2
04 05
3
Instruction
Register  
  
  

F FE FF
 

The first hexadecimal digit (2 in our example) in the machine instruction is called
the operation code, or op-code, for short. The digit in the op-code field indicates which
basic function is to be performed. These basic functions can be classified into three
categories: data relocation, arithmetic/logic, and control. We will discuss selected
functions from each of these categories.
An important data relocation function is LOAD. There are two types of LOAD
function. One of the LOAD instructions copies a value from the main memory and puts
it into a specified general-purpose register. The other kind of LOAD function —
distinguished from the previous LOAD function by a different op-code (see below) —
puts a constant value into a specified general-purpose register.
Another data relocation function, MOVE, copies a value from one register and
puts it into another register. The STORE function copies a value from a register and
writes it into a specific location in the main memory.
There are two ADD functions in the computer’s arithmetic/logic repertoire. One
ADD function adds integers represented in two’s complement notation. The other ADD
function adds two floating-point numbers. There are also logical functions that the CPU
can perform on bit patterns. For example, the NOT function returns the complement of
the number that is given to it. Other examples of logical functions are described below.
One of the control functions the computer utilizes is the JUMP function. In the
normal sequence of events, the computer executes instructions in order; the program
counter advances from 0 to 2 to 4, and so on. The JUMP instruction may set the program
counter to an arbitrary address.
Op-
Operand Description of Function Example
code
17D2 loads bit pattern found
LOAD bit pattern found in
1 RXY in memory cell D2 into
memory cell XY into register R.
register 7.
2 RXY LOAD the bit pattern XY into 2D3C loads the bit pattern
27
register with address R. 00111100 into register D.
STORE the bit pattern found in 3A03 stores the bit pattern in
3 RXY the register R in memory cell register A into memory cell
XY. 03.
MOVE (copy) bit pattern from 403F copies bit pattern in
4 0RS
register R to register S. register 3 into register F.
ADD integers in registers S and 5B64 adds integers in registers
5 RST T (two’s complement form) and 6 and 4; puts result in register
place result in register R. B.
ADD floating-point numbers in 6AE1 adds the floating point
6 RST registers S and T and place result numbers in registers E and 1;
in register R. puts result in register A.
Combine bit pattern in register S 7D29 combines bit patterns
and register T using the OR found in register 2 and 9 using
7 RST
operator and place result in the OR operator and places
register R. the result in register D.
Combine bit pattern in register S 81F2 combines bit patterns
and register T using the AND found in register F and 2 using
8 RST
operator and place result in the AND operator and places
register R. the result in register 1.
Combine bit pattern in register S 94C3 combines bit patterns
and register T using the found in register C and 3
9 RST
EXCLUSIVE OR operator and using the XOR operator and
place result in register R. places the result in register 4.
ROTATE bit pattern in register R A907 rotates bit pattern in
A R0X
to the right X times. register 9 to the right 7 times.
BDA7 compares bit pattern in
JUMP to instruction located in register D with that in register
memory cell XY IF the bit 0. If they are equal, program
B RXY pattern in register R is EQUAL to jumps to instruction located in
bit pattern in register 0; otherwise cell A7; if they are not, the
continue to next instruction. program moves to the next
instruction.
C 000 HALT the program’s execution. C000 causes program stop.
D18D skips to instruction
JUMP to instruction XY IF the
located in cell 8D if bit pattern
bit pattern in register R is LESS
D RXY in register 1 begins with 1;
THAN ZERO; otherwise
otherwise, the program moves
continue to next instruction.
to the next instruction.
NOT—negate each bit in pattern E03C finds complement of bit
E 0RS found in register S and put pattern in register C and puts
resulting bit pattern in register R. the result in register 3.
One JUMP occurs when the value in specified register is equal to the value in register 0.
The second JUMP occurs when the value in a specified register is less than zero.
Another control instruction is the HALT command, which causes the execution of the
program to stop.
28
For our purposes, each of these (and several other) commands has been assigned a
specific hexadecimal digit, as shown in the Op-Code portion in the above table. Next to
each op-code in the table is a description of the operand that provides details necessary
for the instruction to be carried out.
In our example instruction (2D3C), the first hexadecimal digit 2 indicates that the
machine is to load a particular bit pattern. The other hexadecimal digits (D3C) in the
instruction are known as the operand. The letter D in this case represents the address of
the register into which the bit pattern will be loaded. The 3C signifies that the bit pattern
to be loaded is 00111100. Taken together, the instruction 2D3C indicates the machine
will load the bit pattern 00111100 into the general-purpose register with address D.
Another example of an instruction is 87F2. The op-code 8 instructs the machine

to combine two bit-patterns into one using the AND function. The hexadecimal digits F
and 2, tell us to combine the bit patterns stored in those two registers and place them in
register 7. For the next few paragraphs we will assume that bit pattern 01101110 is in
register F and the bit pattern 11110011 is in register 2. To combine these two patterns
into one using the AND operation, we first stack one bit pattern on top of the other so
each of the eight digits is aligned:
0 1 1 0 1 1 1 0
1 1 1 1 0 0 1 1
Next we’ll start at the right (though this is arbitrary) and, according to rules of the
AND operation, place a one under this column of numbers if they are both one, and a
zero if they aren’t both one. The same procedure is applied to each of the remaining
seven columns of numbers, and the result is represented below:
0 1 1 0 1 1 1 0
AND 1 1 1 1 0 0 1 1
0 1 1 0 0 0 1 0
The instruction 87F2 would combine 0110110 and 11110011 (which are the bit
patterns in registers F and 2, respectively) using the AND operation and put the resulting
bit pattern, 01100010, into register 7.
The combining of two bit patterns into one can also be performed by other logical
operators—OR, XOR, etc. You may want to verify that the result of combining the bit
patterns in registers F and 2 using the OR operation will yield the bit pattern 11111111,
while the resulting bit pattern using the XOR operation is 10011101. The NOT operator
requires only one bit pattern and simply negates each of the bits in the pattern. For
example, the instruction E07F will place the bit pattern 10010001 in register 7, which you
may also want to verify.
The ROTATE function rotates the bits in a bit pattern a certain number of
positions, in a circular fashion. For example, the instruction AF04 will rotate the bit
pattern in register F four positions to the right. If the original pattern is 01101110, the
resulting bit pattern is 11100110, as shown below.
29
0 1 1 0 1 1 1 0  1 1 1 0 0 1 1 0
original byte rotated byte
In order to help you become more familiar with the intricacies of machine
language, we will walk through a few examples. First, let’s add two values that are
stored in the main memory and place the sum back in the main memory. Suppose 02 and
08 are the values of the bytes stored in the main memory cells with addresses F0 and F1,
respectively. Our objective is to write a machine language program that adds these two
values and places their sum in the main memory cell with address F2. The steps that we
will need the computer to perform in order to realize this objective are outlined below:
1. Load the value of the memory cell with address F0 into register 0.
2. Load the value of the memory cell with address F1 into register 1.
3. Add the values in register 0 and register 1 and place the result in register 2.
4. Store the value of register 2 into the memory cell with address F2.
5. Halt the program.
For each of the steps above, there is an associated two-byte instruction written in machine
language that the CPU can use. These instructions are:
10F0
11F1
5201
32F2
C000
We will now represent the changes that occur in the main memory and CPU as
these instructions are carried out at the machine level. Notice that the machine language
instructions are stored in the main memory in cells 00 to 09. As the CPU progresses
through the five instructions, watch the program counter, the instruction register, the
effected registers and main memory cells change accordingly.
CPU Main Memory

Address Address
Registers 00 10 F0 01
Program
0 Counter Bus 02 11 F1 03
1 00 04 52 01 05
06 32 F2 07
2
Instruction 08 C0 00 09
3 Register
 -- -- -- --

 F0 02 08 F1
F F2 F3

30
CPU Main Memory
Address Address
Program
02 0 Counter Bus 02 11 F1 03
1 02 04 52 01 05
06 32 F2 07
2
3 Register
 10F0

 F0 02 08 F1
F F2 F3

CPU Main Memory

Address Address
Program
02 0 Counter Bus 02 11 F1 03
08 1 04 04 52 01 05
06 32 F2 07
2
3 Register
 11F1

 F0 02 08 F1
F F2 F3

CPU Main Memory

Address Address
Program
02 0 Counter Bus 02 11 F1 03
08 1 06 04 52 01 05
06 32 F2 07
A0 2
3 CPURegister Main Memory
 5201 Address Address
Registers
 00 10 F0 01
 Program F0 02 08 F1
02 0 Counter Bus 02 11 F1 03
F F2 F3
08 1 08 04 52 01 05

06 32 F2 07
A0 2
3 Register
 32F2
 31
 F0 02 08 F1
F F2 A0 F3
CPU Main Memory
Address Address
Program
02 0 Counter Bus 02 11 F1 03
08 1 0A 04 52 01 05
06 32 F2 07
A0 2
3 Register
 C000

 F0 02 08 F1
F F2 A0 F3

There are a couple of items that should be noted. Since each instruction is two
bytes long, each line of instruction takes up two memory cells; consequently, the program
counter only identifies the first address of the cell containing the line of instruction. This
address is always an even number Also, due to space constraints, only those cells that are
necessary for the program are shown in the graphical representations. This convention
will be used in the following exercises. It’s now time for you to try your hand at machine
language.
Question 1: Based on the following representations of the CPU and main memory,
describe in your own words what the following machine instruction program
accomplishes. Include in your description the reasons for any changes in general register
or main memory cells.
CPU Main Memory

Address Address
Registers 00 17 B1 01
Program
- - 0 Counter Bus 02 28 C4 03
 00 04 99 78 05

 06 39 B2 07
-- Instruction
7 CPU 08 C 0 Memory
Main 00 09
Register
 Address Address
- - 8
Registers -- -- --
Program 00 17 B1 01
-- -- 09 Counter Bus B0
02 28 89 0 54
C B1
03
 02
- - A B2
04 9- 9- 7- 8- B3
05

 06 39 B2 07
05 7 Instruction 08 C0 00 09
Register

- - 8 17B1
32
- - 9 B0 89 05 B1
- - A B2 -- -- B3
CPU Main Memory
Address Address
Program
- - 0 Counter Bus 02 28 C4 03
 04 04 99 78 05

 06 39 B2 07
Register

C4 8 28C4
- - 9 B0 89 05 B1
- - A B2 -- -- B3
CPU Main Memory

Address Address
Program
- - 0 Counter Bus 02 28 C4 03
 06 04 99 78 05

 06 39 B2 07
Register

C4 8 9978
C1 9 B0 89 05 B1
- - A B2 -- -- B3
CPU Main Memory

Address Address
Program
- - 0 Counter Bus 02 28 C4 03
 08 04 99 78 05

 06 39 B2 07
Register

C4 8 39B2
C1 9 B0 89 05 B1
- - A B2 C1 -- B3
33
CPU Main Memory
Address Address
Registers 00 01
Program 17 B1
- - 0 Counter Bus 02 28 C4 03
 0A 04 99 78 05

 06 39 B2 07
Register

C4 8 C000
C1 9 B0 89 05 B1
- - A B2 C1 -- B3
Question 2: Below you will find a list of 8 instructions. Assume that memory cell D2
contains the bit pattern 00001011 and memory cell F0 contains the bit pattern 00011010.
a) Translate each of these instructions into machine language.

b) Draw the relevant memory locations and CPU registers. Put the initial binary
bit patterns in memory locations D2 and F0. The program places values in
memory location E3 and registers 0 through 5. Write in each cell the binary bit
pattern written by the program.
c) To what simple arithmetic problem does this list of instructions provide an
answer?
1. Load the contents of memory cell D2 into register 0.

2. Load bit pattern 00000001 into register 1.
3. Load the contents of memory cell F0 into register 2.
4. Compute the logical NOT the bit pattern in register 0 and place the result in
register 3.
5. Add the values in register 1 and register 3 as if they were integers written in two’s
complement form and place the result register 4.
6. Add the values in register 2 and register 4 as if they were integers written in two’s
complement form and place the result in register 5.
7. Store the bit pattern in register 5 in memory cell with address E3.
8. Stop.
34
In our final example we will illustrate the characteristic of one of the two JUMP
functions in our arsenal. We’ll begin by assuming that the value of register 0 is known to
have the bit pattern 06 and the value of register 5 is 0. Also, for the time being, we’ll
assume the value of the memory cell with address C3 is unknown. We’d like to subtract
the value of C2 from 06, the value in register 0, and place the result in the memory cell
C3. If the value of C2 is also 06, the difference is clearly 0, and in this case we don’t
want to have to run the rest of the program. First we will create a list of instructions that
will accomplish the objective. They are as follows:
1. Load the contents of memory cell C2 into register 1.

2. JUMP to the last line of instruction if the value of C2 is equal to the value
in register 0; otherwise, move on to step three.
3. Load the value 00000001 into register 2.
4. Compute the logical NOT of the bit pattern in register 1 and place the
result into register 3.
5. Add the integers in registers 2 and 3 as if they were in two’s complement
form and place the result in register 4.
6. Add the integers in registers 0 and 4 as if they were in two’s complement
form and place the result in register 5.
35
7. Store the value of register 5 in the memory cell with address C3.
8. End the program.
We’ll assume that these 8 machine instructions are located in the memory cells
with addresses 00 through 0F. Here is the resulting machine language program.
1 1 C 2
B 1 D C
2 2 0 1
E 0 3 1
5 4 2 3
5 5 0 4
3 5 C 3
C 0 0 0
Question 3: Write a machine language program that takes a value from a memory cell,
say 2C, computes its absolute value, and then stores its absolute value in memory cell
2D. An efficient way to do this is to use the JUMP IF LESS THAN ZERO operation. Be
sure to halt the program and to include the following in your solution:
1. A list of instructions using written English descriptions.
2. A list of the instructions translated into machine language.
36
CS 160 Name ________________________
Problem-Solving By Enumeration Worksheet
The ability to solve problems is important in many disciplines, including

computer programming. Computer programmers seek to obtain effective and efficient to
perform desired tasks. Let’s elaborate on three important terms in the previous sentence:
algorithm, effective, and efficient. “An algorithm is an ordered set of unambiguous,
executable steps that defines a terminating process” (Brookshear, Computer Science: An
Overview, 2007). An algorithm is effective if it finds the solution (or a close enough
approximation). An algorithm is efficient if the number of steps it performs is reasonable.
In this worksheet and the ones to follow, we will explore several important
problem-solving strategies. It should be noted that there is no single set of rules that, if
followed, always leads to an effective and efficient algorithm. In fact, problem solving
relies so much on creativity and ingenuity that some people regard computer
programming as an art form! Looking for better algorithms can be a lot of fun. With this
in mind, think of the following worksheets as an opportunity to develop your own
creative problem-solving techniques.
The first technique we are going to explore is called enumeration. Some types of
problems have many possible solutions that are partially correct, but only one that
satisfies all the criteria outlined. Enumeration means listing each of the possibilities and
then systematically eliminating those that don’t fit the requirements set out in the
statement of the problem.
To illustrate problem-solving by enumeration , let’s look at an example. Boronoff,
Pavlow, Revistsky, and Sukarek are individuals who work in different fields of the arts
and humanities: one is a painter, one is a writer, one a dancer, and one is a singer. Using
the information below, determine the profession of each of these four people.
1. Boronoff and Revitsky were both in the audience the night the singer made his
debut on the concert stage.
2. The writer’s biography of Sukarek was a best seller. He also wants to write
Boronoff’s biography, but hasn’t met him yet.
3. Both Pavlow and the writer have sat for portraits by the painter.
4. Boronoff has never heard of Revitsky.
37
To solve this problem by the enumeration technique, we will begin by listing all possible
combinations of the individuals and their professions:
Dancer Singer Painter Writer Dancer Singer Painter Writer

Boronoff Pavlow Revitsky Sukarek Revitsky Pavlow Boronoff Sukarek
Boronoff Pavlow Sukarek Revitsky Revitsky Pavlow Sukarek Boronoff
Boronoff Sukarek Revitsky Pavlow Revitsky Sukarek Boronoff Pavlow
Boronoff Sukarek Pavlow Revitsky Revitsky Sukarek Pavlow Boronoff
Boronoff Revitsky Pavlow Sukarek Revitsky Boronoff Pavlow Sukarek
Boronoff Revitsky Sukarek Pavlow Revitsky Boronoff Sukarek Pavlow
Pavlow Boronoff Revitsky Sukarek Sukarek Pavlow Boronoff Revitsky
Pavlow Boronoff Sukarek Revitsky Sukarek Pavlow Revitsky Boronoff
Pavlow Sukarek Revitsky Boronoff Sukarek Revitsky Boronoff Pavlow
Pavlow Sukarek Boronoff Revitsky Sukarek Revitsky Pavlow Boronoff
Pavlow Revitsky Boronoff Sukarek Sukarek Boronoff Pavlow Revitsky
Pavlow Revitsky Sukarek Boronoff Sukarek Boronoff Revitsky Pavlow
Now we’ll use the clues to eliminate those possibilities that do not meet the first
criterion. Since Boronoff and Revitsky were both in the audience when the singer was on
stage, neither of them can be the singer. Therefore we will cross out (i.e. eliminate) each
row in which either Boronoff or Revitsky is the singer, as shown below.

38
The second piece of information lets us know that the writer must be someone other than
Boronoff or Sukarek. Therefore we will eliminate any row in which Boronoff or Sukarek
is the writer, as shown below.

The third sentence informs us that Pavlow is neither the writer nor the painter and we will
eliminate the rows in which Pavlow is the writer or the painter, as shown below.

After these three rows have been eliminated, we are left with only two rows: one
in which Boronoff is the dancer and Revitsky is the writer (option A), and one in which
Boronoff is the painter and Revitsky is the writer (option B). It is clear that Revitsky is
the writer, and, according to the third piece of information, Revitsky sat for a portrait by
the painter (who is Boronoff if option B is true). The last clue, however, states that
Boronoff has never heard of Revitsky, which means option B cannot be true. Thus
Boronoff must be the dancer, Pavlow is the singer, Revitsky is the writer, and Sukarek is
the painter.
39
Exercises: Find a solution to the following problems using the method of enumeration.
P1: In a certain bank three employees—Sonya, Kirk, and Dakota—work as a teller,

branch manager, and loan officer (not necessarily in that order). Use the information
below to determine who fills what position at the bank.
1. The teller, who is an only child, earns the least amount of money.
2. Dakota, who married Sonya’s brother, earns more than the branch manager does.
40
P2: One weekend a reporter from a campus newspaper came by to do a feature on four
college roommates—Keon, Randall, Yoshio, and Fernando, but each was out of town.
The reporter asked a neighbor if he knew anything about the roommates. After thinking
for a while, the neighbor replied, “One of them is a computer science major, another is an
avid golfer and mows the lawn at the country club, a third has had a mohawk since 5th
grade, and the last one is allergic to kiwi.”
The reporter then rudely reminded the neighbor that he was writing a feature and needed
a little more information than that. The annoyed neighbor replied by giving the following
intentionally veiled facts: “Fernando, who is bald, usually waits for the golfer to finish
his putting practice before they walk to campus together. Unlike Keon, who prefers a
traditionally conservative hairstyle, Yoshio is thinking about getting a mohawk after the
term. The roommate with a kiwi allergy always cooks enough food for Keon who is
deathly allergic to grass pollen.” With that the neighbor bid the reporter good day and
went back into his home.
Based on the neighbor’s information, match up each roommate with his unique
characteristic.
41
Challenge Question: Five sisters all have their birthday in a different month—February,
March, June, July, and December. Using the information below, determine each of the
sister's birthday months. (Adapted from Puzzlers Paradise Website,
http://www.puzzlersparadise.com/article1021.html.)
1. Paula and the sister born in June have their birthdays earlier in the year than Brenda.
2. Abigail and Brenda have their birthdays in consecutive months.
3. Within a calendar year, Paula’s birthday is between Mary’s and that of the sister born in
February.
42
CS 160 Name ________________________
Problem-Solving by Greedy Methods Worksheet
The next problem-solving technique we will explore is called problem solving using
greedy methods. The greedy method solves a problem by focusing on particular local
aspects of the problem, making a choice that is best for the local sub-problem, and hoping
that a series of locally optimal choices will lead to the best overall solution.
For example, suppose a department store clerk owes $17.68 in change to a customer who
has just made a purchase. What is the fewest number of bills and coins (assuming the
clerk has at his disposal only $20, $10, $5, and $1 bills and quarters, dimes, nickels and
pennies) the clerk can give to the customer to provide proper change?
If we disregard the ‘fewest number of bills and coins’ phrase, there are a myriad of ways
that this change-making process can be brought about. For example, the clerk could give
the customer 1768 pennies, or 176 dimes and 8 pennies. However, if the greedy method
is applied to this situation, the clerk will try to give the greatest monetary piece (bill or
coin) at each ‘phase’ of the change-making process. First, he will place a $10 bill on the
table, since that is the largest single piece that he can give without giving more than the
amount owed. Next, a $5 bill will be given, followed by two $1 bills. At this point $17
of the customer’s change has been paid and bills can no longer be given. The clerk will
then give two quarters, a dime, a nickel and three pennies. This solution of four bills
($10 + $5 + $1 + $1) and six coins ($0.25 + $0.25 + $0.10 + $0.01 + $0.01 + $0.01) is the
optimal one—there is no way of combining less than 10 of the specified monetary pieces
so that their sum is exactly $17.86. While this rather mundane example is one we are all
familiar with, it is indeed an example of the greedy technique at work—the next piece of
money to be given was always the largest one that could be given at the time, and this
series of ‘local’ optimal decisions led to a ‘global’ optimal solution.
However, there are situations—and some very similar to the one just described—in
which the greedy method fails to provide the optimal global solution. For example, let’s
imagine a society whose monetary system consisted of the following coins: 10-krunk, 8-
krunk, 5-krunk and 1-krunk. If a sales clerk is obliged to give a customer 17 krunks in
change, the greedy method would yield the following solution: one 10-krunk coin, one 5-
krunk coin, and two 1-krunk coins. This global solution of four coins is not optimal.
Rather, it is possible to provide 17 krunks with only three coins—two 8-krunk coins and
one 1-krunk coin. This example illustrates the fact that the greedy method should be used
with caution.
43
Exercises: Try to solve each of the following problems using the greedy method.
Explain your reasoning in each case.
P1: An international thief has gained entry to a high security medical laboratory and
intends to steal valuable powdered chemicals from the lab. There are 8 different types of
powder, and each type of powder is stored in a single large container. She plans to pour
the precious powders from their containers into individual, thin plastic bags and then
replace the containers. After putting the plastic bags containing the stolen powder in a
brief case, the thief will attempt to bypass the lab’s security equipment again. In order to
successfully escape, she has calculated that, after accounting for the weight of the
briefcase and the plastic bags, she can put no more than 25 kilograms of the stolen
powder into the briefcase. Below is a table listing the powders, their weight, and their
value. If the thief wants to escape with the most valuable collection of powders in her
briefcase, how much of each type of powder should she steal? Note: The “Weight of
Powder” column indicates the total available amount of each type of powder. For
example, the thief can steal up to 3.25 kg of IM2, but no more.
Type of Weight of Value of

Powder Powder Powder
IM2 3.25 kg 1975 £
B24 9.50 kg 2350 £
CP 1.25 kg 3000 £
D+ 7.50 kg 1100 £
Rz 8.50 kg 4800 £
Q32 8.00 kg 3600 £
W07 9.50 kg 2600 £
57/J 6.00 kg 4320 £
44
P2: Chris has won a shopping spree at a gourmet donut wholesale bakery. The bakery
owners have agreed to allow Chris to have 1 minute to gather as many boxes of donuts as
possible, with two restrictions: all donuts must remain in their original (bulk) boxes and
no more than ten pounds of donuts are to be chosen. Chris has created a list of the types
of donuts the bakery sells, the weight of each box and the value per box (see table
below). If Chris wants to end up with the most valuable ten pounds of donuts, which
boxes of donuts should he choose?
Donut Type Package Value of

Weight Box
Jelly Filled Donuts 2 lbs $15.00
Maple/Chocolate Bars 5 lb $86.25
Chocolate-Covered Cake Donuts 1 lbs $12.50
Custard Cream Filled Donuts 6 lbs $105.00
Cinnamon Rolls 5 lbs $85.00
Apple Fritters 3 lb $50.00
45
P3: “In regional news, Blueneck County has decided to pave some of the roads
connecting the cities in their county. County officials have decided that there must be at
least one paved route between every pair of cities. Due to the economic shortfalls that
have swept the region, the county also wants to minimize the cost of the project.”
According to the map of Blueneck County shown on the next page, which roads should
county officials have paved?
Appleton Durazno
10 miles
iles
12 m
Berry Hill s
m ile
8m 18 11 miles
ile
s
Cider Canyon 12 miles

10 miles
Elberta
s
ile
m
10
l es
17 mi
Fruit Heights
Challenge Question: The greedy method was successful at determining the optional
solution in two of the three problems above. Which problems were they? What
characteristic distinguishes the two problems that could be optimally solved by the
greedy method from the third problem, which couldn’t be optimally solved by the greedy
method?
46
CS 160 Name ________________________
Divide-and-Conquer Problem-Solving Worksheet
Another problem-solving method that is often utilized in computer science is called

divide-and-conquer. In this paradigm, a problem is divided into smaller sub-problems,
which are then solved. The solutions to these sub-problems are then combined to form a
solution to the original problem. Divide-and-conquer algorithms often require fewer
steps than greedy algorithms when solving the same problem.
For example, suppose we want to arrange the following words in order from longest to
shortest: desire to latent sometimes go misfortune awakens overwhelming on.
One method that would surely work is to apply a greedy method called selection sort. In
this method we seek to find the longest word in the group, and once that is accomplished,
we put it in the first position. Then the second longest word is determined and put in the
second position and so on.
Here is how selection sort would sort the words from longest to shortest.
First, it would search the list for the longest word:
desire to latent sometimes go misfortune awakens overwhelming on
The longest word is “overwhelming,” so it would swap “desire” and “overwhelming.”
overwhelming to latent sometimes go misfortune awakens desire on
In the second iteration, it would search the remainder of the list for the longest word:
overwhelming to latent sometimes go misfortune awakens desire on
The longest word found is “misfortune,” so it would swap “to” and “misfortune.”
overwhelming misfortune latent sometimes go to awakens desire on
The algorithm would continue in this fashion until the list of words is completely sorted:
overwhelming misfortune sometimes awakens latent desire to go on
Certainly the selection sort algorithm is effective—it sorts the words correctly—but is it
efficient? To determine how efficient the selection sort algorithm is, we need to think
about how many times it compared a pair of words. During the first step, the number of
comparisons carried out was eight—one less than the number of words in the list. At the
next step, there were seven comparisons, and six at the next, and so on. In general, if
there are n objects in a list to be compared, there will be (n – 1) comparisons during the
first step of the selection sort process, (n – 2) during the second step, and so on until only
47
1 comparison is made at the (n – 1)st and last step of the process. In general, if the list
contains n elements, the number of comparisons being made using the selection sort
algorithm is: (n – 1) + (n – 2) + · · · + 3 + 2 + 1.
This type of sum occurs so frequently in computer science that we’ll briefly divert our
attention to understanding its features. We would like to have a formula so that
regardless of the value of n, we can always quickly find the result of adding the first n
natural numbers: 1 + 2 + 3 + · · · + (n – 1) + n. When n is small, say 3, the sum is
trivial: 1 + 2 + 3 = 6. But what if n is 30,000? It would be a waste of time to do 29,999
additions, even with a calculator or computer, because there is a simple formula that will
do the job. One way to understand this formula is to look at a geometric argument.
Suppose we have an arbitrary n by n square made up of smaller squares of side length 1.
In total there would be n2 of the smaller squares—n rows and n columns (see figure below
with n = 6). To model the desired sum geometrically we could fill in the first square of
the first row, the first two squares of the second row, and so on until we fill in the first n
squares of the nth row (i.e. we would fill in each square of the last row of squares). The
area of the shaded squares is equal to 1 + 2 + 3 + · · · + (n – 1) + n, the exact sum in
which we are interested. Drawing a diagonal from the top left corner to the bottom right
corner will split the n by n square exactly in half, so that the area to the left of the
diagonal is n2/2. But there is still some shaded area that has not been counted—the n
half-shaded squares. The total area of the half-shaded squares is n/2.Thus the entire area
n 2 n n 2  n n n  1
of the shaded squares is    .
2 2 2 2

Another way to arrive at this formula is to write the numbers in a list and then pair the
first and last numbers, then the second and second to last, and so on, as shown in the
figure below.
1 + 2 + 3 + · · · + (n - 2) + (n - 1) + n
Each of these pairs adds to (n + 1), and if n is one can see that there are n/2 pairs that
n n 2  n n n  1
have been matched up. Thus the sum of the pairs is (n  1)    , which is
2 2 2
the same result obtained using the geometric argument.

48
Question: Answer the following question, showing your work.
Q1: What is the sum of the first 30,000 whole numbers?

(i.e. 1 + 2 + 3 + · · · + 29,998 + 29,999 + 30,000)
Getting back to an analysis of the selection sort algorithm, we can now see that the sum
n1
(n 1) 2  (n 1) n 2  n
(n – 1) + (n – 2) + · · · + 3 + 2 + 1 =  i   . Since the leading
i1 2 2
term in this sum is n2, we say that the selection sort algorithm has order n2, and often
write this fact in the following way: O(n2).
 Sort Algorithm
Merging Lists and the Merge
At this point we will focus on a different and more efficient way of sorting a list of
objects. The algorithm that we will describe uses the divide-and-conquer problem-
solving strategy. To begin, we introduce the idea of merging two lists that have already
been sorted. For example, the two numerical lists that follow have each been sorted from
least to greatest: 5 13 27 32 8 12 17 20. Merging these two lists into one in
which all eight numbers have been sorted will begin, as you might expect, by comparing
the first numbers of each list—in this case 5 and 8. Once these numbers have been
compared (and the lesser number, 5, is placed in the first position of the merged list) the
second number in the first list—13—is compared with the first number in the other list.
The lesser number, 8, takes its place in the second position of the merged list and the next
comparison, between the second numbers in both lists, is made. (See the diagram below
for a complete illustration of this example.)
This process will go on until one of two things happens: the last number in one of the
lists is compared to, and determined to be smaller than, a number from the other list; or
the last two numbers in each list are compared. In the best-case scenario, the last number
in one of the lists—say list A—is smaller than the first number of the other list—call it
list B. In this case we will make four comparisons—comparing each of the numbers in
list A with the first number in list B. Once the last number in list A has been compared
with, and is determined to be smaller than, the first number in list B, no other
comparisons are necessary, for list B is already in order. At this point list B is simply
copied and the two lists have been merged after only ½·n comparisons.
49
In the worst-case scenario, the merging process ends with the last numbers from list A
and list B being compared. In this case there will have been (n – 1) comparisons. In
general, if n is even, it will take at most (n – 1) comparison to merge two lists of size n/2
as we shall demonstrate below.
We will now apply the divide-and-conquer paradigm to the general problem of sorting a
large list of numbers. The process that we will outline below is called the merge sort
algorithm. For sake of example, let’s suppose we have a list of 1024 numbers. We will
view the list as 1024 sorted lists of size and note that our objective is to obtain one sorted
list of size 1024.
At the first step we will merge the first two single ‘sorted lists’ (i.e. the single numbers
viewed as individual lists of size one) into a list of size two. This will require only one
comparison (we’ve only got two numbers to compare). We’ll then do the same with the
next pair of numbers, and so on until there are 512 sorted lists of size two. With one
comparison per merge and (½)·(1024) = 512 merges, there will be 1·512 = 512
comparisons at the first step.
At the second step we will merge two sorted lists of size two into a sorted group of four
numbers. There will be at most 3 comparisons per merge (n – 1 where n = 4 = number of
objects being merged) and (½)·(512) = 256 merges. So during the second step there will
be no more than 3·256 = 768 comparisons made. Recall we may make fewer
comparisons, but we’ll be conservative and assume the worst-case scenario. We will
look at one more step to establish the pattern: during the third step—in which we will
merge sorted groups of four numbers into sorted groups of eight numbers—there will be
at most 7 comparisons per merge and (½)·(256) = 128 merges. Thus at the third step we
are guaranteed to complete all merges using no more than 7·128 = 896 comparisons.
In general, if we begin with an unsorted list of n numbers, at the kth step of the merge-sort
n
process, there will be at most 2k – 1 comparisons per merge and k merges. Thus there
2
n
will be less than 2  k  n comparisons.
k
2


50
An applet that compares the selection sort and merge-sort algorithms is found here:
eecs.oregonstate.edu/~quinn/education/JavaApplets/applets/SortingAlgorithms.html.
1. Change the sorting method on the left from Quick Sort to Selection Sort.
Change the sorting method on the right from Bubble Sort to Merge Sort.
2. With your mouse, drag the blue circle in the Number of Elements box until the
number of elements is 16. Choosing this number of elements will help
illuminate the merge sort process.
3. At a steady rate, continually click the step button, which will move the sorting
process along a step at a time. Once you get the basic idea, you can click the
play button, which will move the process along at a quicker pace. If you want
to slow the process down again, simply click the pause button and then the
step button.
4. In the Selection Sort demonstration, the red bar represents the current
position that is being filled and the moving yellow bar represents the bar-by-
bar comparisons that are being made to determine the shortest remaining bar.
5. In the Merge Sort demonstration, bars 1 and 2 are merge-sorted, followed by
the merge-sorting of bars 3 and 4. Now those two ‘sorted lists’ are merged-
sorted. Thus we have an ordered list of bars 1 – 4. The next step the applet
takes is to merge-sort the 5th and 6th bars and then the 7th and 8th bars. These
two sorted lists of size two are now merged—forming another sorted list of
four bars (representing bars 5 – 8). The two lists of four bars are merge-sorted
into one list of eight bars. This same process is now applied to bars 9 – 16,
until they have been sorted from shortest to tallest. The last step is to merge
the two sorted lists of 8 bars, the first one consisting of bars 1 – 8 (in order
from shortest to tallest) and the last one consisting of bars 9 – 16 (in order
from shortest to tallest).
6. Now change the number of elements to be sorted and compare the algorithms
again.
Another piece of information we need to know is how many steps there are going to be in
the merge sort process. In our example of 1024 words, we can ask this question: How
many times will we decrease the number of sorted lists in half (recall we started with
1024 lists of size one, and then reduced these to 512 lists of size two, and so on) before
we get to one sorted list. The nature of this problem suggests we use the log function, as
log2(n) can be defined as the number of times the quantity n can be cut in half until it has
been reduced to the value 1. So log2(1024) = 10, and we know that there will be 10 steps
involved in reducing 1024 sorted lists of size one to one sorted list of size 1024. Recall
that at each step there are less than n comparisons which means we are guaranteed to be
able to sort the list of size 1024 by using fewer than 10,240 comparisons. This is
significantly less than 523,776, which is the number of comparisons required for
accomplishing the same task using the greedy selection sort method.
In general, if there are n words to be sorted, there will be log2(n) steps and no more than n
comparisons at each step. So we are guaranteed that there will be less than n·log2(n)
comparisons made over the course of the entire process. In other words the merge sort
algorithm is O(n·log n). As was mentioned earlier, the selection sort algorithm is O(n2).
51
The first exercise will give you a greater understanding of the difference between n2 and
n·log n.
Exercises: Answer each of the following questions and explain your reasoning as
necessary.
Q2: Fill in the following table for the given values of n.
n n2/2 n·log2 n
10
100
10,000
100,000
1,000,000
Q3: Use the merge sort algorithm to alphabetize the following names: Rachel, Sammy,
Wilma, Tyrone, Carlos, Olga, Bethany, Jasmine, Pedro, Mikhail.
52
Q4: This question concerns your experience with the applet. When the number of
elements is 16, which of the two sorting methods (selection sort and merge sort) is
fastest? What about when the number of elements is 75? Why is there a difference?
Q4: A king has 18 golden coins, one of which he knows is counterfeit. The counterfeit
gold coin weighs less than a regular coin. You have been assigned to use a balance
(scale) to determine which one of the coins is not pure gold. Explain how you would
accomplish this using the scale a minimum number of times. How many times did you
use the scale? Can you prove there is no method that requires fewer weighing steps?
53
 i   2  when n is odd.
n
n n 1
Challenge Question: Prove that
i1

54
CS 160 Name ________________________
Searching Methods Worksheet
Some situations present problems that have many solutions, but only one that is optimal.
In such a situation a searching strategy may be helpful in finding the optimal solution.
To illustrate problem solving by searching, we’ll consider the 8-puzzle.
The 8-puzzle is tile game consisting of nine squares—eight numbered tiles and one
‘empty’ square (see square below, left which was adapted from Zhu, S. C.,
http://www.stat.ucla.edu/~sczhu/Courses/UCLA/Stat_232B/Handouts/Introduction.pdf .).
The goal of the puzzle is to use the free space to move the tiles so that the numbers are in
order starting at the top left and proceeding around the square until the 8, which is right
below the 1 (see square below, right). When the numbers are in this order, we have
found a solution.
2 8 3 1 2 3
1 6 4 8 4
7 5 7 6 5
The puzzle on the left represents an 8-puzzle in a non-solution state, while the puzzle
on the right represents the solution, or objective, of an 8-puzzle.
An optimal solution consists of one that consists of the fewest slides. To track the
progress toward the optimal solution, we’ll implement a penalty rule system. Each time
we slide a tile we’ll be assessed one point. Further, for each number that is out of its
solution position, we will assess one penalty point. So if the 8 – puzzle we want to solve
is the one on the left above, the initial position has a score of 4 points—we’ve made no
slides, and there are four numbers (2, 8, 6, and 1) out of place.
First, we consider the three possible first moves and the number of points for each
of the resulting states, as shown below. We’ll continue to search from the puzzle with the
lowest point total, in this case the puzzle with 4 points.
55
2 8 3
1 6 4
7 5
1 + 5 = 6 pts 1 + 3 = 4 pts 1 + 5 = 6 pts

2 8 3 2 8 3 2 8 3
1 6 4 1 4 1 6 4
7 5 7 6 5 7 5
Diagram note: a shaded puzzle represents one whose ‘next steps’ have been expanded. Solid
arrows represent the current ‘puzzle path’ we are currently exploring. Dotted arrows represent
paths to puzzles we have not currently chosen, but may choose in the future if they become more
desirable than the current path.
Since two puzzles tie for the lowest number of points (i.e. two of them have only five
points) we’ll choose one of the two arbitrarily, the one with the empty square in the top
row.
2 8 3
1 6 4
7 5
1 + 5 = 6 pts 1 + 3 = 4 pts 1 + 5 = 6 pts

2 8 3 2 8 3 2 8 3
1 6 4 1 4 1 6 4
7 5 7 6 5 7 5
2 + 4 = 6 pts 2 + 3 = 5 pts 2 + 3 = 5 pts

2 8 3 2 3 2 8 3
1 4 1 8 4 1 4
7 6 5 7 6 5 7 6 5
56
2 8 3
1 6 4
7 5
1 + 5 = 6 pts 1 + 3 = 4 pts 1 + 5 = 6 pts

2 8 3 2 8 3 2 8 3
1 6 4 1 4 1 6 4
7 5 7 6 5 7 5
2 + 4 = 6 pts 2 + 3 = 5 pts 2 + 3 = 5 pts

2 8 3 2 3 2 8 3
1 4 1 8 4 1 4
7 6 5 7 6 5 7 6 5
3 + 3 = 6 pts 3 + 4 = 7 pts 3 + 3 = 6 pts

8 3 2 8 3 2 8 3
2 1 4 7 1 4 1 4
7 6 5 6 5 7 6 5
Searching this path leads to three puzzles with point totals of six (two have this
many points) and seven. At this point if we were using the greedy method, we would
follow the path of one of the puzzles having six points. However, according to the search
method we will take a step back and survey all possible remaining paths and then choose
from them the one that is most desirable. Currently there are seven puzzles that have not
yet been expanded. Each of these puzzles has five, six or seven points. There is only one
puzzle with five points and we’ll expand that puzzle next, as it is currently the most
desirable path.
As shown on the next page, by following this path and implementing the search
strategy, the optimal solution to the puzzle can be obtained in five slides, indicated in the
diagram by the solid black arrows.
57
2 8 3
1 6 4
7 5
1+5=6 1+3=4 1+5=6

pts
2 8 3 pts
2 8 3 pts
2 8 3
1 6 4 1 4 1 6 4
7 5 7 6 5 7 5
2 + 4 = 6 pts 2 + 3 = 5 pts 2 + 3 = 5 pts

2 8 3 2 3 2 8 3
1 4 1 8 4 1 4
7 6 5 7 6 5 7 6 5
3 + 3 = 6 pts. 3 + 4 = 7 pts. 3 + 3 = 6 pts. 3 + 3 = 6 pts 3 + 2 = 5 pts 3 + 4 = 7 pts

8 3 2 8 3 2 8 3 2 8 3 2 3 2 3
2 1 4 7 1 4 1 4 1 4 1 8 4 1 8 4
7 6 5 6 5 7 6 5 7 6 5 7 6 5 7 6 5
4 + 3 = 7 pts
4 + 1 = 5 pts
2 3 1 2 3
1 8 4 8 4
7 6 5 7 6 5
5 + 0 = 5 pts 5 + 2 = 7 pts
1 2 3 1 2 3
8 4 7 8 4
7 6 5 6 5
SOLUTION
!
The key principle of this searching method is always following the path that
appears to be leading to the lowest-cost solution. If the algorithm is following a
promising search path, but at some point the path seems to be inferior to another path
58
through the search tree, the algorithm should switch focus and pursue the path that now
seems to be better.
Exercises: Answer the following questions using the searching strategy outlined in this
worksheet.
P1: On a dark, cold and stormy night four friends find themselves on one side of a
bridge. There is only one lamp between them and because the bridge is so old, at most
two of them can cross the bridge at once. Whenever anyone crosses the bridge—either
alone or with a partner—the lamp must go with him.
One of the friends can cross the bridge in 15 minutes, the second in 10 minutes, the third
in 2 minutes and the last in 1 minute. Find the minimum amount of time needed before
they each reach the other side of the bridge.
59
P2: What is the minimum number of steps required to transform the eight-puzzle below
into its solution state?
1 3
6 2 4
8 7 5
60
P3: Two boys with a small boat agree to help three soldiers cross a river without a
bridge. The boat is so small it can support only one soldier or two boys. A soldier and a
boy can't be in the boat at the same time for fear of sinking it. What is the minimum
number of trips it will take to ferry all the soldiers across? (Adapted from Science News
for Kids, http://www.sciencenewsforkids.org/pages/puzzlezone/muse/muse0904.asp).
61
62
CS 160 Name ________________________
Monte Carlo Problem-Solving Worksheet
One way to incorporate technology into problem solving is to use a computer or other
machine to simulate an activity or experiment. This process can often save time and
money and can yield results very similar to what would have occurred if the experiment
had actually taken place. The Monte Carlo method is a problem solving strategy that
finds an approximate solution by sampling points in the solution space. The strategy gets
its name because the sample points are created using a random number generator, the
computer equivalent of throwing dice.
We will illustrate the Monte Carlo method by using a simple spreadsheet to provide a
(rough) estimate of . We begin with a circle of radius one, inscribed inside a square of
side length two, as shown below. The area of this circle is . The ratio of the area of the
circle to the area of the square is /4. Now look at the upper right quadrant (shaded in the
diagram below). The area of the shaded quarter circle to the area of the shared square is
also /4.
1 unit
2 units
1 unit
2 units
One way to apply the Monte Carlo method to finding an approximation of 
is to select
n random points in the shaded square and then compute the ratio of points that lie within
the circle to n, the total number of points randomly selected. The Law of Large Numbers
(sometimes called the Law of Averages) says that as n increases, the proportion of points
within the circle will approach /4. Multiplying this proportion by 4 will give us an
approximation to .

An illustration of this strategy is shown below. A spreadsheet is used to compute two
columns of 15 random numbers between 0 and 1. The first column represents the x-
coordinate of the point and the second column represents the y-coordinate. The
spreadsheet can determine which of the points are within the quarter circle. We can
assume that the center of the circle is at the origin (0, 0) so that the equation of the circle
of radius 1 centered at the origin is x2 + y2 = 1. With this in mind, the spreadsheet
63
identifies those points that are inside the quarter circle by determining which of them
satisfies the inequality x 2  y 2  1. In the simulation shown below, 12 of the 15 points are
inside the circle. Thus our approximation for /4 is 12/15, or 0.8, and our approximation
for is 3.2 
. 
(0 , 1 ) (1 , 1 )

(0 , 0 ) (1 , 0 )
Increasing the number of samples tends to increase the accuracy of the estimate, as
shown by the graph below.
3.19
3.18
Estim ate of pi = 3.14159
3.17
3.16
3.15
3.14 
3.13
3.12
3.11
3.1
0 2000 4000 6000 8000 10000 12000 14000 16000 18000 20000 22000 24000
Number of Points Chosen

Nobody would use the Monte Carlo method to compute  because we already know the
value of . We chose that as our example so we could evaluate the quality of the
estimate. In actual practice, the Monte Carlo algorithm is used in situations where it is
impossible to calculate the true answer analytically.
64
Exercise: Use Microsoft Excel or another spreadsheet program to solve the following
problem using the Monte Carlo method.
P1: A homeowner wants to improve the looks of his property by fertilizing the lawn, part
of which is adjacent to a river. He figures he will need a bag of fertilizer for every 10
square yards of grass. Based on the plot of his property below, how many bags of
fertilizer should he buy? You can assume that the entire property is lawn, except for the
building and the driveway.
H e i g h t o f c u r v e a b o v e d o t t e d li n e = 1 0 * s i n ( 0 . 0 7 8 5 4 * x )
w h e r e x is y a r d s f r o m le f t e d g e o f p r o p e r t y
2 0 y ard s
10 yards
30 yards
H ouse
1 0 y ard s
D r iv e w a y
5
4 0 y ard s
65

Worksheets For Computer Science Orientation: Daniel Balls

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Worksheets For Computer Science Orientation: Daniel Balls

Uploaded by

Copyright:

Available Formats

Worksheets for

Computer Science Orientation

This document is in the public domain.

Funding for this project was provided by the Hewlett Foundation

Logic Gates Worksheet

A unary operation (one input). A binary operation (two inputs).

The NOT gate returns the complement of the input.

The OR gate returns 1 when at least one of its inputs is 1.

1. Make sure the option 1 Gate is selected.

An example of a combinatorial circuit.

Use the applet

1. Select the option labeled 2 Gates.

Binary Number System Worksheet

2 4 7 = 2·100 + 4·10 + 7·1

The value of the decimal representation 247.

1101 = 1 1 0 1 = 1·8 + 1·4 + 0·2 + 1·1 = 13

Binary Decimal Binary Decimal Binary Decimal Binary Decimal

a. What is the binary representation of the decimal value 17?

Converting Between Binary and Decimal Representation

8÷2=4 R0 1 0 0 0 1 0 0 = 68 (base 10)

Critical Thinking: Answer the following questions.

1011000110 1100101101110 10000001111

Octal and Hexadecimal Integers Worksheet

An example of a hexadecimal number is 4C7. To compute the decimal value of this

4·162 + C·161 + 7·160 = 4·256 + 12·16 + 7·1 = 1024 + 192 + 70 = 1223.

4·82 + 5·81 + 2·80 = 4·64 + 5·8 + 2·1 = 256 + 40 + 2 = 298.

Decimal Octal Hex. Decimal Octal Hex.

Hexadecimal and Decimal Conversions

b. two octal digits?

c. two decimal digits?

d. two hexadecimal digits?

e. two digits from a base-k number system?

Two’s Complement Worksheet

001110 negation: ____________________________________

010010 negation: ____________________________________

110011 negation: ____________________________________

000000 negation: ____________________________________

Critical Thinking: Answer the following question.

Exercise: Convert these decimal integers to 6-bit, two’s complement form.

Binary and Two’s Complement Addition

Adding numbers written in two’s complement notation is equivalent to adding numbers

0+0=0 0+1=1 1+0=1 1 + 1 = 10 10 + 1 = 11

1. Make sure the option Demo Mode is selected.

Machine Language Worksheet

Computer programmers typically write programs in a high-level language, but in

Another example of an instruction is 87F2. The op-code 8 instructs the machine

CPU Main Memory

CPU Main Memory

CPU Main Memory

CPU Main Memory

CPU Main Memory

CPU Main Memory

a) Translate each of these instructions into machine language.

1. Load the contents of memory cell D2 into register 0.

1. Load the contents of memory cell C2 into register 1.

Problem-Solving By Enumeration Worksheet

The ability to solve problems is important in many disciplines, including

Dancer Singer Painter Writer Dancer Singer Painter Writer

Dancer Singer Painter Writer Dancer Singer Painter Writer

Dancer Singer Painter Writer Dancer Singer Painter Writer

Dancer Singer Painter Writer Dancer Singer Painter Writer

P1: In a certain bank three employees—Sonya, Kirk, and Dakota—work as a teller,

Problem-Solving by Greedy Methods Worksheet