Scientific Calculator Comparison Vol I

User Guide and Operational Comparison

Casio fx-115ES and Sharp EL-W516

by Jim Cullen

Scientific Calculator Comparison Vol I

User Guide and Operational Comparison
Casio fx-115ES and Sharp EL-W516
Jim Cullen
Document Version: v1.0

Dated: May 10, 2010

To my family,
for their patience.

Foreword
For starters, the CASIO fx-115ES and the SHARP EL-W516 are both excellent
scientific calculators and I do not promote one over the other. Both have relative strengths
and weaknesses and either one could be a selected as a better match for an intended
purpose. As the title implies, this manual is intended as a users guide for both models. The
operational comparison is meant to illustrate the strong points of one or the other to provide
some basis for deciding which model might be better suited to a particular application. Within
this manual, the CASIO fx-115ES will be referred to as the 115ES and the SHARP EL-W516
will be referred to as the W516 for brevity.
To put the CASIO fx-115ES and the SHARP EL-W516 into perspective, they are part of
the 'middle-tier' of a three-tiered hierarchy of calculator types. The three tiers are outlined
below with their fundamental differences outlined.
Basic calculators, sometimes referred to as 'four-bangers', are typically 8 to 12
decimal digit standard function varieties. They will typically perform addition, subtraction,
multiplication, division, square root, and percentage. The price of such calculators run
anywhere from US $1 to about US $15 depending on the specific brand and model and can
be found in just about any store.
Scientific calculators are differentiated from the usual 'four-banger' variety by the
availability of functions well beyond standard math operations. These functions customarily
include the power/root and logarithmic functions, trigonometric/hyperbolic functions, simple
statistics, polar/rectangular coordinate conversion, and number base operations. Numeric
representations will often be made available in fractional or exponential format. Calculations
with floating-point arithmetic will have 10 to 15 decimal digits in the mantissa and can have
exponents within the range of -99 to +99. The price of scientific calculators may run from US
$15 to US $40 or more, again depending on the brand and model. Usually you can only find
the scientific calculators in the larger retail outlets but some may have to be purchased
through the mail or online.
Graphing calculators, in the US $50 to US $150 plus range, perform all of the above
mentioned functions plus many many more. They are able to graph functions and plot data
points in a variety of formats including 3D graphing, scatter plots, bar and pie charts, or data
presented in tabular form. They also have larger screens for graphing purposes, faster and
more powerful CPU's, and more internal memory. Because of this, they have larger power
requirements and so they will use more batteries that will need to be replaced more often.
Many data types can be represented and the graphing calculator can support enormous
mathematical constructs based on expressions, matrices, vectors, strings, or other forms of
data. The operating system will usually include CAS capability, which stands for Computer
Algebra System. The CAS capability allows the calculator to perform symbolic manipulation of
expressions. This may mean simple algebraic operations on expressions containing variables
but usually includes also differentiation, integration, and a myriad other operations involving

any of the available data types. Graphing calculators may also be ( and almost always are )
programmable in one or more types of programming languages to extend their capabilities.
Communication software, included with the calculator, allows connection to a home computer
for file transfers, program and variable storage, or operating system updates.
The SHARP EL-W516 and the CASIO fx-115ES represent the best examples of the
newest incarnation of the scientific calculator which falls somewhere between the capabilities
of the scientific and graphing models currently available. They can be thought of as high-end
scientific calculators with some of the capabilities normally found only in the graphing
models... without the price-tag typical of a graphing calculator. For a very small investment,
students now have the advantage of being able to easily explore the mathematical properties
of vectors, matrices, simultaneous equation solving, statistics, complex numbers, and
calculus operations such as integrals and derivatives.
Due to the EL-W516's ability to store user-defined functions, as small as that storage
actually is, the EL-W516 is not permitted by many standardized tests. The CASIO fx-115ES is
specifically mentioned as being allowed in most testing standards, evidently because it is not
capable of storing user-defined functions. Please check the allowed calculator list prior to your
test date to verify your particular model is permitted in the test room.
Choosing between one of these two calculators comes down to a question of intended
use. Some will say the W516 is better others will claim that the 115ES is superior. In all
honesty, it can only really be claimed that one or the other may be better for a particular
application, and this will depend on the person. A high school or freshman math student, who
will not rely heavily on matrix or vector math, may get more out of the 115ES for basic math
and calculus functions. Someone pursuing more advanced math, or those who use their
calculator at work in an engineering field, may appreciate the speedy W516 with its advanced
matrix and vector functions.
Availability is also a consideration. The 115ES is widely sold at the usual retail outlets.
A short drive to the store will put a 115ES in your hands. The W516 will very likely have to be
ordered through the mail or from one of the online shopping sites. Shipping charges may
apply and, if they do, they will be close to half the price of the calculator itself.
You should remember this brief description of key notation used throughout the
manual. Keystrokes will be in color-coded bold print and will be enclosed within brackets
except for numeric entries which will just be bold print (no brackets). The resulting function will
be given immediately afterward in {bold curly braces} with no color coding. The keystrokes
are color coded for easier identification of the key entries that pertain to your model of
calculator; BLUE for the CASIO fx-115ES and GREEN for the SHARP EL-W516. The keys
will be identified by the label on its face or by the key's primary function. As an example; the
setup menu on the 115ES can be accessed by pressing [SHIFT] [MODE] {SETUP}, and on
the W516 you press [2ndF] [MATH] {SET UP}. Turn on your 115ES by pressing [ON] and
power down the calculator by pressing [SHIFT] [AC] {OFF}. Turn on your W516 by pressing
[ON/C] and power it down by pressing [2ndF] [ON/C] {OFF}. The color-coded brackets
indicate the actual key presses and the black curly-braces describe the resulting function.
Lastly, in an effort to stimulate interest in mathematics and technology, this manual and
its supplement is to be provided free of charge to any interested parties, without any special
permissions required. I ask only that this manual be passed on, unedited, in its current format.
Jim Cullen
June 2010

Table of Contents
1. The Basics.......................................................................................1
2. Operation Modes and Calculator Setup..........................................5
3. Expression Editing.........................................................................11
4. Memory Management....................................................................17
5. Manual Calculations......................................................................21
6. Scientific / Physical Constants.......................................................31
7. Metric & Unit Conversions.............................................................35
8. Complex Number Calculations......................................................37
9. Base-N Calculations......................................................................41
10. Function Table ( 115ES only )......................................................49
11. The General Solver......................................................................51
12. The Equation Solver....................................................................59
13. Summations, Integrals, & Derivatives.........................................69
14. Basic Matrix Calculations............................................................79
15. Advanced Matrix Operations ( W516 only ).................................91
16. Vector Operations........................................................................94
17. Matrix / Vector Conversions ( W516 only )................................105
18. List Operations ( W516 only )....................................................109
19. Statistical Calculations...............................................................115

Chapter 1. The Basics

1. The Basics
At first glance the CASIO fx-115ES and the SHARP EL-W516 appear to be very
different but they are actually very similar in several ways. They have comparable size,
weight, power supply, display type, key layout, and menu function. Both operate in one of
several different modes and keep operations relevant to that mode tucked away in menus
appropriate for that mode of operation. Beyond these initial similarities though, details of
operation begin to differ between the two; differences that may or may not be suitable for your
intended use of the device. Here is a table containing the basic descriptions of both
calculators, after which there will be given some extra details concerning initial impressions
some users have commented on.

115ES

W516

12.2(H)80(W)161(D) mm

15.5(H)79.6(W)161.5(D) mm

105 grams

102 grams

Lt gray / Silver

Black / Lt gray

1LR44 + Solar Cell

1LR44 + Solar Cell

46 + 4way pad

47 + 4way pad

9632 pixel LCD

9632 pixel LCD

Screen Layout

2 line 16 character

4 line 16 character

Std Digit Size

59 pixels

57 pixels

Mini Digit Size

56 pixels

55 pixels

Internal Digits

14

15

Size
Weight
Color
Power Supply
Number of Keys
Display Type

Screen contrast and readability is comparable between the two though the W516 in
general appears to make better use of the display area. The 115ES tends to maintain a larger
easier to read font throughout, which at times requires some scrolling to read calculation
results. The W516 employs a slightly smaller font on some calculation results, as well as a

Scientific Calculator Comparison Vol I: User Guide and Operational Comparison

rearrangement of the output, to prevent the need to scroll in order to read calculation results.
SHARP wisely chose an easy to read font for this purpose. The result is an easy to read
display that does not require scrolling in most cases to read calculation results. Still, some
may prefer to tolerate some amount of scrolling to read calculation results in order to enjoy
the larger font that CASIO uses for most purposes on the 115ES. I personally prefer the
screen on the W516 since the organization of menus, entry fields, and calculation results
seems more thought through. The display area is used more efficiently and so less scrolling is
required overall to view calculation results. Also, since a single continuous pane of clear
plastic encloses the entire upper portion of the calculator, the task of 'dusting off' the display is
quicker and easier on the W516.
Although CASIO claims only a 2-line by 16-character display area while SHARP quotes
a 4-line by 16-character display, the actual differences between them are not as drastic as
may be imagined. I would tend to say that both have a 4-line by 16-character display and that
the W516 is more flexible and makes some better use of the available display area.
Both the 115ES and the W516 are durable enough in construction; neither feels
'cheap'. Though different in appearance, they are practically the same size and weight and
they handle about the same. Visually speaking, the 115ES is typical of any other calculator
produced by CASIO. Simple, easy to read key layout. Utilitarian in a classroom sort of way
and nothing out of the ordinary to mention. The 115ES is very clean but very capable as an
advanced scientific calculator. The W516 looks slightly more professional, probably because it
is visually similar to some of HP's calculators, both in the colors selected for the body and key
labels, as well as in the appearance of the character font in the display. The key labels on the
W516 appear a bit more busy than those on the 115ES, due mainly to the greater number of
functions available on the W516. Both calculators have hard case slide on covers to protect
the keys and the LCD display. The W516 slide on cover requires some fumbling in order to
remove it as compared to the 115ES. SHARP just didn't provide enough good grip area to
remove the hard cover as easily as you can on the 115ES. With the cover removed from the
W516, the area down around the keys is noticeably flat while the area around the display is
thicker. The 115ES, with the cover removed, is still a consistent thickness from one end to the
other. This is just an observation and of no real concern.
There has been some comment on the tactile feedback from the keys on both
calculators. CASIO has always produced durable keys with adequate feedback; a small
amount of pressure produces a light soft click at the end of key travel, validating key entry.
Such tactile response prevents inadvertent keystrokes and consequent errors in calculations.
There has been some comment that the W516 has too little tactile response to adequately
validate keyed entries and prevent inadvertent keystrokes. While it is possible for a stray
finger to produce inadvertent keystrokes, there is just enough tactile response in the keys to
get used to. Not a showstopper but definitely a noticeable effect. I have personally had very
few problems with the keys on the W516. If you have used SHARP calculators in the past
then you are already familiar with how the keys feel under your fingers; the W516 is about on
par with past offerings from SHARP in this respect.
The W516 has the useful ability to retain screen contents and current work when it is
turned off. When the W516 is later turned back on, you pick up exactly where you left off; if
you were entering an expression for use in a certain cell of a matrix, you can power up again
2

Chapter 1. The Basics

the following day and finish typing in the equation. When you power down the 115ES or it
powers itself down, all current work is lost and you will later power back up to a blank screen.
The W516 also has a nice little customization feature for shutting down the calculator. The
user can enter a string of characters, letters, or numbers for display below the SHARP logo as
the unit is turning off. This 'OFF-message' is retained until it is changed by the user or the
calculator is wiped clean with a reset.
Overall, the SHARP W516 is a snappier performer, with little delay or dead-time in the
display update. In raw calculation, covered in a later section, the W516 is faster than the
115ES in nearly every respect. The CASIO makes up for this by providing, slightly more often
than the W516 does, exact results. Exact results are rational expressions, expressions
containing surds (square roots or symbols), expressions involving , or a combination of the
above. There has been some discussion that the internals of the 115ES may, under certain
unusual conditions, provide an exact value when in fact the result is not exact, but a decimal
value very close to an exact value. The chance of this happening is likely very rare and could
be revealed by checking the problem again using pencil and paper.
The W516 is a very poor numerical integrator as compared to the 115ES. It is rare for
the W516 to be as fast at numerical integration as the 115ES is, and the W516 is hardly ever
as accurate. In worst cases, the W516 can be six times slower or more in performing
numerical integration. As powerful as the W516 is, the weakness of its numerical integration
stands out like a sore thumb. The root cause of this weakness is not computational power but
algorithmic power. The W516 uses Simpson's Method of numeric integration, a very simple
and poorly performing algorithm. The 115ES uses a vastly superior and sophisticated
algorithm called Gauss-Kronrod Quadrature. This method is shrewdly implemented on the
115ES to provide lightning fast and surprisingly accurate numeric integration, many times
returning exact results.
To balance its weakness as a numeric integrator, the W516 may boast a superior
collection of matrix and vector capabilities. It is able to work with matrices with dimensions up
to 4X4 and 4-dimensional vectors; the 115ES can only handle matrices up to dimensions of
3X3 and 3-dimensional vectors. The 115ES has three matrix or vector memories while the
W516 has four. The 115ES can calculate matrix determinants and transposes while the W516
also offers the production of identity matrices, cumulative matrices, and random matrices. The
W516 is also equipped with instructions to compose a matrix from a collection of vectors or,
alternatively, to decompose a matrix into individual vectors. Either of these operations can be
performed in two different ways. The W516 can also work with dimensioning of matrices,
filling elements of a matrix with specific values, or augmenting separate matrices into one
matrix. In short, for matrix or vector operations, the W516 is the vastly superior tool.
The W516 contains a catalog of all functions and commands in the calculator itself,
though it is of little use after you've gained some experience with the calculator. Though the
command can be executed directly from the catalog, the catalog contains only the name of
the command and does not provide the required syntax for the use of that function or
command. This would have been a very useful feature for some functions or commands.
The W516 provides four custom keys ( memories D1 through D4 ) which can be used
to store often used functions or commands, freeing the user from the task of having to dig

Scientific Calculator Comparison Vol I: User Guide and Operational Comparison

through menus to access required functions or commands. The W516 can also store up to
four user-defined functions ( in dedicated function memories F1 through F4 ) for later recall
and execution. The 115ES has no capacity for custom keys or stored equations. The W516
can use variables A through F, plus X, Y, and an Independent Memory M, for a total of nine
storage locations for numeric values. The 115ES is almost as capable; it can use variables A
through D, plus X, Y, and an Independent Memory M. The 115ES is not able to use E or F as
variables, so there are only seven storage locations available.

Chapter 2. Operation Modes and Calculator Setup

2. Operation Modes and Calculator Setup

The operation modes on the 115ES and the W516 are very similar, with just a few
minor differences between them. Both calculators require you to enter an operation mode
specific to a given set of tasks such as matrix calculations, complex numbers, or statistics. To
enter a desired mode, simply access the mode menu and key in the appropriate digit. Certain
functions and commands intended for that mode of operation will then become available. On
the 115ES the mode menu can be accessed by pressing [MODE]. The W516 also has a key
labeled [MODE] for accessing the mode menu. There are two pages in the mode menu on
the W516. Press [] to access page 2 which contains the last item 6: EQUATION.
For the remainder of this section, refer to the following tables of operation modes for
both calculators:

fx-115ES
1: COMP
3: STAT
5: EQN
7: TABLE

2: CMPLX
4: BASE-N
6: MATRIX
8: VECTOR

Key: [AC] [MODE]

EL-W516 page 1
<MODE-1>
0: NORMAL 1: STAT
2: DRILL
3: CPLX
4: MATRIX
5:LIST
Key: [ON/C] [MODE]

fx-115ES Operation Modes

1. COMP

2. CMPLX

3. STAT

4. BASE-N

5. EQN

6. MATRIX

7. TABLE

8. VECTOR

EL-W516 page 2
<MODE-2>
6: EQUATION

Key: [ON/C] [MODE][]

Scientific Calculator Comparison Vol I: User Guide and Operational Comparison

EL-W516 Operation Modes

0. NORMAL

1. STAT

2. DRILL

3. CPLX

4. MATRIX

5. LIST

6. EQUATION

The default mode for both calculators is normal calculation mode. On the 115ES you
can access this mode by pressing [MODE][1]{COMP}, and on the W516 it is [MODE][0]
{NORMAL}. This is the mode that will be used most often for common calculations. This is
also the mode that results when the calculator is reset to factory defaults.
The current mode on the 115ES is shown by a small icon above the display area;
STAT, CMPLX, MAT, or VCT. For Base-N mode, the current number base is shown in the
middle-right area of the display. For the remaining modes, you are confined to a menu type
display system so there is no question about which mode you are in.
The current mode on the W516 is clearly printed across the top of the display area but
is conveniently removed while you are entering data or performing calculations. As with the
115ES, you are inobtrusively notified of your current mode of operation.
At first glance it seems as if the W516 is missing the BASE-N and VECTOR modes
and instead offers the DRILL and LIST modes. The W516 is able to perform Base-N
operations but does not require a separate mode of operation to do so; Base-N mode is
entered automatically, as soon as a number is entered in any base besides base-10. The only
catch is that you must be in normal calculation mode [MODE][0] for Base-N mode to activate.
In addition to Binary, Octal, Decimal, and Hexadecimal, the W516 is also equipped to handle
the Pental ( Base 5 ) number system.
The DRILL mode on the W516 is used to practice your basic math skills. You can
specify the operation and the limit to the sizes of the numbers during the quiz. Some
customizing is possible and you are graded at the end of a round of questions. It's not Tetris
but, in a pinch, could possibly kill a few idle moments. It could also give the young ones some
amusement.
The LIST mode on the W516 is a dual purpose mode, serving as both a list and a
vector capable mode of operation. Vector and list specific operations are available through the
[MATH] menu when this mode is selected. Though this is a dual purpose mode, the vector
capability of the W516 is more powerful than that of the 115ES. The matrix and list modes on
the W516 are also interconnected, allowing matrix and vector data to be transferred from one
mode to the other. In addition, the list operations available on the W516 have no counterparts
on the 115ES.

Chapter 2. Operation Modes and Calculator Setup

Calculator Setup
The 115ES and W516 both contain a calculator setup menu for configuring various
settings to your liking. The menu includes access to settings such as default unit of angular
measure, numerical display notation, complex number format, display contrast, and so on.
Typically, these settings apply across the board, in all modes of operation ( as applicable ).
Your setup configuration is retained until you change the configuration, the unit is reset, or the
memory is cleared. The setup menu on the 115ES is two pages of menus but each is very
simple. The setup menu on the W516 is only six items but each has sub-menus to choose
from. The 115ES is actually more configurable overall. You can access the setup menu on the
115ES by pressing [SHIFT][MODE]{SETUP}. To switch between page one and page two of
the setup menu, use the [] and [] directional keys. The setup menu on the W516 is
accessed by pressing [2ndF][MATH]{SET UP}. Refer to the following tables:

fx-115ES page 1
1: MthIO
3: Deg
5: Gra
7: Sci

2: LineIO
4: Rad
6: Fix
8: Norm

Key: [SHIFT][MODE]
{SETUP page 1}

fx-115ES page 2
1: a b / c
3: CMPLX
5: Disp

EL-W516

2: d / c
4: STAT
6: Cont

Key: [SHIFT][MODE][]
{SETUP page 2}

<SET UP>
0: DRG
1: FSE
2: EDITOR 3:CTRST
4: -----5: NAME
Key: [2ndF][MATH]{SET UP}

fx-115ES Setup Menu: Page One

1. MthIO

Set display output to 'textbook' ( default )

2. LineIO

Set display output to 'linear'

3. Deg

Set angular units to degrees ( default )

4. Rad

Set angular units to radians

5. Gra

Set angular units to gradians

6. Fix

Set numeric display to fixed with 0-9 decimal places

7. Sci

Set numeric display to scientific with 1-10 sig digits

8. Norm

Set numeric display to normal ( Norm1 or Norm2 )

Scientific Calculator Comparison Vol I: User Guide and Operational Comparison

fx-115ES Setup Menu: Page Two

1. a b/c

Set fraction display format to mixed

2. d/c

Set fraction display format to improper

3. CMPLX

Set complex display format to a+bi or r,

4. STAT

Frequency column display ON or OFF

5. Disp

Set decimal mark to DOT or COMMA

6. CONT

Adjust the display contrast

MathIO/LineIO: Sets the display mode for fractions and expressions. MthIO is the
default and causes expressions to be input and displayed just as they would appear in a
textbook. MthIO is short for 'Math Input / Output'. LineIO causes expressions to be printed on
a single line. Press [SHIFT][MODE]{SETUP} followed by [1] for MthIO or [2] for LineIO.
Deg/Rad/Gra: Sets the units of angular measurement; degrees, radians, or gradians.
Press [SHIFT][MODE]{SETUP} followed by [3] for degrees, [4] for radians, or [5] for
gradians. The default is degrees. Note that 360 degrees = 2 radians = 400 gradians.
FIX/SCI: Sets the numerical format to 'fixed' or 'scientific notation'. Fixed format sets
the numerical display to a fixed number of digits past the decimal, properly rounded. You may
enter anything from 0 to 9 digits. Scientific format always displays numbers in scientific or
exponential notation, and you may enter 1-9 to specify anything from 1 to 9 significant digits in
the mantissa; enter 0 to specify 10 significant digits. The value of is approximately
3.14159265 If FIX 0 is selected ( [SHIFT][MODE]{SETUP} [6][0] ) then the 115ES would
show this as 3. If FIX 4 is selected ( [SHIFT][MODE]{SETUP} [6][4] ) then the 115ES would
show this as 3.1416. If SCI 3 is selected ( [SHIFT][MODE]{SETUP} [7][3] ) then the 115ES
would show this as 3.142X100. If SCI 6 is selected ( [SHIFT][MODE]{SETUP} [7][6] ) then
the 115ES would show this as 3.141593X100. The current numeric display format is indicated
by an icon, FIX or SCI,
NORM: Sets the numerical format to 'normal'. What this means is that numbers are
displayed in a format equivalent to FIX 10 within a certain range. Outside of this range, the
number is displayed in a format equivalent to SCI 10. There are two normal formats; NORM1
and NORM2. Norm1, the default, displays a number in scientific format if it is less than
0.000000001 or greater than 9,999,999,999. Norm2 displays a number in scientific format if it
is less than 0.01 or greater than 9,999,999,999. Press [SHIFT][MODE]{SETUP} [8][1] for
Norm1 or [SHIFT][MODE]{SETUP} [8][2] for Norm2.
The setup options on page two of the 115ES setup menu are all self-explanatory and
straightforward. Select your fraction display format, complex display format, stat frequency
display ON or OFF, decimal mark to DOT or COMMA, and LCD display contrast. Simply
select your option at the prompt.

Chapter 2. Operation Modes and Calculator Setup

Here is the setup menu on the W516, which you can access by pressing [2ndF]
[MATH]{SETUP}:
EL-W516 Setup Menu
0. DRG

Set angular units to 0=Deg, 1=Rad, 2=Grad

1. FSE

Set 0=Fixed, 1=Scientific, 2=Engineering,

3=Norm1 (default), 4=Norm2

2. EDITOR

Set 0='Write-View' or 1='Line' for fraction and

expression display format

3. CTRST

Adjust LCD display contrast

4. ENTRY

0=Insert, 1=Overwrite. 'Line' mode only, otherwise ------

5. NAME

Compose custom 'OFF-Message'

Most of the setup options in the W516 are also available on the 115ES. The numerical
display formats have the same definitions and options. Engineering format, also available on
the 115ES as a function, is a format identical to scientific format except that the exponent is
always a multiple of +/-3. Engineering format is a sensible approach for engineers who prefer
the use of units with prefixes such as 'milli-', 'kilo-', 'micro-', etc.; all powers of 10 3. The WriteView and Line display for fractions and expressions is the same as the MthIO / LineIO option
on the 115ES. The W516 does not provide a means for you to select fraction display format,
complex format, or decimal mark option. The W516 does, by default, provide thousands
separators for displayed numbers. These are comma-shaped marks just above the digits,
separating them into groups of three. If you like this option then you're in luck; if you don't
then your stuck with it... there is no option to remove or change the thousands separator on
the W516. If it's any consolation, you quickly learn how to ignore them if you don't care to
make use of them.
DRG 'Deg/Rad/Grad': Degrees is the default unit of angular measure on the W516. To
select another unit, press [2ndF][MATH]{SETUP}[0] followed by [0]{DEG}, [1]{RAD}, or [2]
{GRAD}. The current unit of angular measure, DEG, RAD, or GRAD, is indicated by an icon
along the top center of the display.
FSE 'Floating/Scientific/Engineering': Used to configure the numeric display format.
Press [2ndF][MATH]{SETUP} [0]{FSE} followed by [0]{FIX}, [1]{SCI}, [2]{ENG}, [3]
{NORM1} (default), or [4]{NORM2}. After selecting fixed format, you will be prompted to enter
a digit 0-9 specifying the number of digits to round to and display after the decimal point. After
selecting scientific or engineering format, you will be prompted to enter a digit 0-9 specifying
the number of significant digits to round to and display in the mantissa. Entering 0 in this case
specifies 10 significant digits. Normal 1 and Normal 2 on the W516 have the same ranges for

Scientific Calculator Comparison Vol I: User Guide and Operational Comparison

exponential notation display as the 115ES. Normal 1, the default, displays a number in
scientific format if it is less than 0.000000001 or greater than 9,999,999,999. Normal 2
displays a number in scientific format if it is less than 0.01 or greater than 9,999,999,999. The
current numeric display format, FIX, SCI, ENG, N1, or N2, is indicated by an icon at the left
edge of the display.
Option 4: Entry method, an option available only if you are in 'Line' mode. The
available options are 0 for insert and 1 for overwrite. This setting controls how expressions
are edited in 'Line' mode. In 'W-View' mode, this option in the setup menu is not available and
is displayed as ------.
NAME: Option 5 on the W516 setup menu is a nice touch, allowing you to customize
the message that appears when the calculator is powered down. The message can be up to
32 characters ( two full lines of characters ) and may consist of letters ( capitals only ), digits,
and a handful of punctuation marks. To create a custom 'OFF-message', press [2ndF][MATH]
{SETUP} [5]. Use the [] and [] keys to select a character, the [] and [] keys to move
from one position to the next, and the [=] key to finalize your 'OFF-message'. Now whenever
the calculator is turned off, the SHARP logo and your customized 'OFF-message' appear on
the display for a few brief seconds.

10

Chapter 3. Expression Editing

3. Expression Editing
For the rest of this manual, it is important to note the configuration while keying in the
examples provided. Unless otherwise indicated, both calculators will be in 'normal calculation'
'textbook' mode with Normal 1 numeric display and radians as the angular unit of
measurement. Here are the keystrokes to get both calculators into that state:

CASIO fx-115ES
'Norm Calc'

SHARP EL-W516

[MODE][1]

[MODE][0]

'textbook'

[SHIFT][MODE][1]

[2ndF][MATH][2][0]

'Normal 1'

[SHIFT][MODE][8][1]

[2ndF][MATH][1][3]

[SHIFT][MODE][4]

[2ndF][MATH][0][1]

radians

The Write-View feature on the SHARP EL-W516 is the same feature that CASIO calls
V.P.A.M., or Visually Perfect Algebraic Method. On the 115ES, this same feature is called
Natural Display or Natural Textbook Display. For either calculator, I will call this 'textbook'
mode, since both VPAM and W-VIEW display math expressions as you would see them in a
textbook. This is the default mode on both calculators and the mode that I will always use in
the examples provided in this manual unless otherwise indicated. There are subtle differences
in how 'textbook' mode is implemented on the two calculators but they make a big difference
in how 'easy' or 'hard' the user finds it to work with the calculator. Both the 115ES and the
W516 offer an alternative mode, called 'Line' or 'LineIO' mode, where all expressions are
input and displayed on a single line using parentheses and simplified operator characters.
To select 'MthIO' mode on the 115ES, press [SHIFT][MODE]{SETUP} [1]{MthIO}.
The icon 'MATH' will appear at the upper right edge of the display. If the icon 'MATH' does not
appear, then you are in 'LineIO' mode. On the W516 press [2ndF][MATH]{SETUP} [2]
{EDITOR} [0]{W-VIEW}. The icon 'W-VIEW' will appear at the upper right edge of the display.
If the icon 'W-VIEW' does not appear, then you are in 'Line' mode.
The directional arrow keys on the 115ES and the W516 serve several purposes. Their

11

Scientific Calculator Comparison Vol I: User Guide and Operational Comparison

main function is for navigating your way around an expression during editing but, they are
also used for other purposes such as moving within menus or recalling calculation history.
The directional keypad is never the most popular feature on a calculator but, more often in
recent years than in the past, it has become an 'obligatory' feature. Since the numeric keys
are at the bottom of the calculator and the arrow keys at the top, it is generally less annoying
to the user if the calculator designers try to minimize the number of times the user has to be
interrupted, during routine calculations, to reach up and press an arrow key. The designers of
the 115ES and the W516 have done a pretty good job in this respect. The directional keypad
will be used often in the examples given but it's not often enough to be thought of as
'annoying'.
Fractions: Numerators and Denominators
Here's an example calculation where we are adding simple fractions:
7
3
13
5

=
= 1 = 1.625
8
4
8
8
On the 115ES there is a key for entering a template for entering fractions in 'textbook'
format. It is at the upper left of the keypad, just below the key labeled CALC. This key will be
shown as [/]. On the 115ES, the template must be entered before keying in the numerator
and denominator of the fraction. Here is how the calculation is performed: [AC] [/][7][][8]
[][+] [/][3][][4] [=], which is 12 keystrokes. The answer is shown as 13/8 or as 1 5/8,
depending on which format you have set for fractional display. You can press the [SD] key
to toggle between decimal and fractional display of the result. The decimal equivalent of 13/8
is 1.625. You can also press [SHIFT][SD] to toggle between improper and mixed fraction
display. Having two different methods of conversion for fractional display format is nice. The
W516 has only one key that rotates between the three possible formats.
On the W516 there is also a key for entering fractions in 'textbook' format. It is labeled
[a/b] and is just above the [8] key. This fraction template can be used just as it is on the
115ES. The nice thing about the W516 is that you can also enter the fraction template
midstream during a calculation, making the task of entering a fraction into the calculator
quicker and easier. Here is how the example is performed: [ON/C] [7][a/b][8] [][+] [3][a/b]
[4] [=], which is just 10 keystrokes. The answer is given as 1 5/8 which can be changed from
mixed fraction to improper fraction to decimal display by pressing the [CHANGE] key until the
desired format appears.
On either calculator, you can use a little algebra or parentheses if you like to achieve
the same result: [AC] [7][][8] [+] [3][][4] [=] on the 115ES or [ON/C] [7][][8] [+] [3][][4]
[=] on the W516, either of which only requires 9 keystrokes. The 'textbook' mode is very nice,
especially for displaying results but, you don't always need to use it for keying in the

12

Chapter 3. Expression Editing

calculation in the first place. The above example demonstrates that sometimes it's quicker
and easier to just key in an expression as you would in 'line' mode, especially for simpler
expressions. The 'textbook' mode of entry can come in especially handy though, when you
have a large expression with terms in both the numerator and denominator.
Answer Memory
The 115ES and W516 both maintain an Answer memory which retains the results of
the last calculation performed, even if the calculator is turned off. Answer memory is updated
when a calculation is ended with the [=] key or variables are altered using a store or +/operation. The 115ES maintains 15 significant digits for numeric values and the W516 retains
14 significant digits. The contents of the Answer memory can be accessed on the 115ES
using the [ANS] key for insertion of the value into a current expression. On the W516 press
[ALPHA][=]{ANS} or [RCL][=]{ANS}. Both calculators show this as the variable Ans in the
current expression being edited. I prefer using the [RCL] key on the W516 , for consistency,
since variables A-F can be recalled with the [RCL] key in many circumstances. For example;
[RCL][B] can be used to insert the variable B into an expression, unless it's the first thing in
the expression, then you will have to use the [ALPHA] key. The [RCL] key is just easier to
use and is not as far up the keypad as the [ALPHA] key is.
Calculation History
The 115ES and W516 both maintain a calculation history which holds both the input
expression and the results of past calculations. This feature is available in normal calculation
mode and complex mode. The 115ES also maintains a calculation history in Base-N mode.
When the calculator is turned off, calculation history is maintained on the W516 but not on the
115ES. There is limited memory allocated to calculation history so, as new calculations are
performed and entered into calculation history, the oldest calculations are rolled out of
memory and lost. The number of items maintained in the history depends on the size of the
expressions entered. The 115ES can handle up to 99 bytes (basically, characters ) for a
single expression. The W516 can handle up to 161. The manual for the W516 states that only
a 340-character memory is maintained for calculation history so storage is limited to just
about two items in the worst case.
Generally, to access calculation history, you must clear the display by pressing [AC] on
the 115ES or [ON/C] on the W516.
To navigate calculation history on the 115ES, use the directional arrow keys. Pressing
[] will step through calculation history backwards, from the newest items to the oldest.
Pressing [] will step through calculation history from the oldest items to the newest. When
you find the item you're looking for, pressing [] will scroll through the expression backwards
from the end; pressing [] will scroll through the expression forwards from the beginning.
There seems to be no method available to jump to the beginning or end of an expression.

13

Scientific Calculator Comparison Vol I: User Guide and Operational Comparison

Pressing [=] will re-evaluate the expression.

To navigate calculation history on the W516, you also use the directional arrow keys.
Pressing [] will step through calculation history backwards, from the newest items to the
oldest. [2ndF][] will jump to the oldest item in the calculation history. From there, you can
use the [] to step forward through the calculation history. When you find the expression
you're looking for, the [] and [] keys scroll backwards or forwards through the expression.
You can also jump to the beginning of the expression using [2ndF][] or jump to the end
using [2ndF][]. Pressing [=] will just re-evaluate the expression.
Locating Math or Syntax Errors
If you attempt to perform an illegal or improper calculation, or perform a calculation that
exceeds the calculators range for numeric results, you will get a Syntax Error or a Math Error.
To locate exactly where in the expression the error is located, press the [] or [] key, and
the expression will be recalled with the cursor at the location where the error occurred. From
there you can edit and re-execute the calculation, or clear the display altogether by pressing
the [AC] or [ON/C] key.
Implied Multiplication & Closing Parentheses
Both the 115ES and the W516 allow for implied multiplication. This means that
multiplication is understood to be the intended operation without the need to type in the actual
multiplication key. This is allowed before mathematical constants such as , i, or e, opening
parentheses or functions that require parentheses such as sin, , or log. Implied
multiplication also may be used immediately preceding a variable such as A or X, a physical
constant (115ES), or before a random number function. Some examples of valid expressions
using implied multiplication and closing parentheses are, on either calculator:
Both the 115ES and the W516 allow you to omit closing parentheses in an expression,
as long as they are at the very end of the expression. You cannot omit closing parentheses
after a function that is in the middle of an expression, though I've found there are many
exceptions to this. Just be careful of your order of operations. When in doubt, include the
parentheses or multiplication signs; with practice, you'll know when it's okay to leave them
out.
Interpretation of Delimiters
There are some differences in how the two calculators interpret delimiters in
expressions, either parentheses inserted by the user, or parentheses inserted automatically
by the keying in of certain functions. In general you'll find that the 115ES will automatically
insert opening parentheses for you when you key in many functions such as sin(, log(, and

14

Chapter 3. Expression Editing

so on. The W516 does not insert opening parentheses automatically, so you have to construct
your expressions with order of operations in mind. As an example of identical expressions
entered on both calculators returning differing results, consider the following:

fx-115ES
sin

EL-W516

2
3

sin

2
3

3
4
interpreted as

-0.947363561

2
sin
3

interpreted as sin

Key: [AC] [sin][/] [2][SHIFT]

[x10x]{}[][3] [] [)][x2] [=]

2
3

Key: [ON/C] [sin] [(] [2][2ndF][3]

{} [a/b] [3] [] [)] [x2] [=]

The 115ES automatically inserts the open parentheses for you as soon as you key in
the [sin] function, meaning that anything inside the parentheses will be interpreted as the
argument of the sine function. After the sine function is evaluated, since the square falls
outside the parentheses, the result is then squared. The W516 does not interpret delimiters
this way automatically. It instead will follow order of operations, with the items inside the
parentheses being evaluated first. This includes the square. The result is then evaluated as
the argument of the sine function. If in doubt, insert extra parentheses to force an order of
evaluation so that the result will be the one intended.
Multi-Statement Calculations
On the 115ES you may use the colon key, by pressing [ALPHA][], to separate a
calculation into two or more steps. This is useful for an extended calculation where the display
of intermediate results is desired. Enter the calculations in the order you'd like them to be
performed, separated by colons. When you execute the multi-statement, you'll have to press
the [=] key after the display of each intermediate result in order to continue the execution of
the multi-statement.

15

Chapter 4. Memory Management

4. Memory Management
There are enough types of memory or storage on the 115ES, and even more on the
W516, that it would be wise to learn how to manage memory contents properly. Managing
memory properly, for what it's worth on a calculator, means deleting unwanted items without
accidentally erasing data that you want to keep.
For both the 115ES and the W516 there are multiple methods of erasing the various
memories in the calculator; from simple display clearing to a total reset of the device. It is
recommended that you become familiar with the different memory areas and the methods of
erasure that affect them.
Display, Answer Memory, and Calculation History
The display itself is a type of memory that records the keystrokes you've entered into
the calculator. When the expression is evaluated, the result is stored into Answer Memory.
Answer Memory is just one item; the result of the last evaluation or other procedure. You can
access this memory for subsequent calculations with the [Ans] key on the 115ES. On the
W516 use [ALPHA][=]{ANS} or [RCL][=]{ANS}. Answer memory is replaced with the result
of the next evaluation. An amount of internal memory is also allocated to store previous
calculations; both the expressions and the results of their evaluations. This is your Calculation
History. When this allocated memory is filled, old items begin to drop out of the calculation
history as new ones are stored.
On the 115ES, the [AC] key is used to interrupt a calculation. This same key, pressed
again, will clear the display. On the W516, the [ON/C] key serves the same purpose. Answer
Memory and Calculation History are unaffected.
Going one step further, pressing [ON] on the 115ES will clear the display and the
Calculation History but, will leave the Answer Memory intact. This has the same effect as
turning the 115ES off and then back on again. The W516 has a 'Clear All' key, [2ndF][MODE]
{CA}, that will clear the display, the Calculation History, and the Answer Memory as well. In
addition, [2ndF][MODE]{CA} will clear Matrix Memories A-D, List Memories L1-L4, and Stat
Memory.
Turning the W516 off and then on again has no effect on the display, the Answer

17

Scientific Calculator Comparison Vol I: User Guide and Operational Comparison

Memory, Calculation History, or anything else for that matter. The calculator will power up in
the exact same state that it was in when the power was turned off.
Variable, Dependent, and Independent Memories
I recognize three types of variable memories on the 115ES and the W516. The first is
what I will call Variable Memories. On the 115ES you have A-D available. On the W516 you
get two additional memories so you have A-F available. Variable Memories are storage
locations that are free for you to use and they will not change unless you clear them or alter
their contents. STO and RCL are the operations provided for use with these memory
locations.
Both calculators have X and Y available for what I will call Dependent Memories. I call
them Dependent Memories because, though they are free for you to use, there are certain
functions within the calculator that are dependent on X and Y for their operation and so are
able to alter the contents of these two memories without asking your permission. Use them
with some caution. STO and RCL are provided for use with these two memory locations also.
Both calculators also have what is called an Independent Memory, M. The Independent
Memory is free for you to use and will not change unless you clear it or alter its contents.
Independent Memory M has extra functionality; you can store to it with STO, recall from it with
RCL, or alter its value through addition [M+] or subtraction [M-] functions specifically provided
for that purpose.
On the W516 you have access to all Variable, Dependent, and Independent Memories
in all modes of operation except for Complex Mode. While in Complex Mode, only the
Independent Memory M is available for you to use. On the 115ES, I have not found any
restrictions on memories available in any mode of operation.
For either calculator, your choices for clearing the Variable, Dependent, and
Independent Memories are somewhat limited. Both the manuals for the 115ES and the W516
actually suggest clearing their values individually, as required, by storing a zero into the
memory to be erased. Storing a zero to a memory location has the same effect as erasing the
memory by other methods.
fx-115ES 'Clear' Menu
Both the 115ES and the W516 have a menu available specifically for memory
management. The 115ES has a 'Clear' menu that can be accessed by pressing [SHIFT][9]
{CLR}. You may press [AC] to exit the 'Clear Menu' at any time if you change your mind.
There are three options in this menu:
1: Setup. Press [1] then [=]. If you select this option, the default setup configuration
will be loaded. Mode is set to normal calculation, angular units are set to degrees, numeric
display is set to Norm1, fraction display is set to d/c, complex format is set to a+bi, stat
frequency display is turned off, and the decimal point is set to dot. This option does not affect
memories.
18

Chapter 4. Memory Management

2: Memory. Press [2] then [=]. All Variable, Dependent, and Independent Memories
will be cleared. Answer Memory will also be cleared but Calculation History is left unaltered.
3: All. Press [3] then [=]. This will perform a complete device reset and is basically a
combination of options 1 and 2 along with wiping out calculation history.
EL-W516 'Clear' Menu
The W516 has a memory management menu called 'M-Clear' that can be accessed by
pressing [2ndF][ALPHA]{M-CLR}. The order and function of the menu items, though named
slightly differently, are very similar to that of the 115ES.
0: DISP. Press [0] then [0]. Resets display settings by setting angular units to degrees,
numeric display to Norm1, and N-Base to decimal. You remain in the current operation mode
and memories are not affected.
1: MEMORY. Press [1] then [0]. you will erase all Variable, Dependent, and
Independent Memories. Be aware however, that this option is just one step short of the
wholesale slaughter of all stored data on the W516. You will lose all Matrix and List Memories,
custom keys, user-defined functions, Statistics Memory, and Answer Memory. The only thing
left intact is your 'OFF-message'.
2: RESET. Press [2] then [0]. This option goes one extra step beyond option 1 and
wipes out your 'OFF-message' as well. The Reset option is the same as pressing the Reset
Button on the back of the calculator.

19

Chapter 5. Manual Calculations

5. Manual Calculations
This chapter covers the use of the basic categories of scientific functions, most of
which are standard and available directly on the keypad. Some examples of their use are
given in order to illustrate the differences that exist between the 115ES and the W516. Not all
functions and capabilities are available on both calculators.
Roots and Powers
The 115ES and W516 have a nearly identical set of power and root functions; the
difference is in the key layout for these functions. The 115ES has the [], [x2], and [x]
buttons on the front face ( no 2ndF or shift required ), which is a nice arrangement for general
work, especially if you use a lot of polynomials. These same keys shifted give you [SHIFT]
[]{3}, [SHIFT][x2]{x3}, and [SHIFT][x]{}.
The W516 is arranged a little differently. Only the [yx] and [x2] functions have front face
positions ( no 2nd'F or shift required ). Their 2'nd functions are [2ndF][yx]{x} and [2ndF][x2]
{}. You have to use [2ndF][RCL]{x3} and [2ndF][STO]{3} to get to the 'cubic' power and
root functions.
Note, from the examples given below, that you have to use the arrow keys to escape
from an exponent or from inside a radical if another term in the expression is required. Also,
when you use the n'th root function on the W516, the last term you typed is assumed to be
the root you want, otherwise the calculator will fill it in for you as the zero'th root.

4 81 = 3
91 = 4

115ES: [SHIFT][x]{} [4] [] [8][1] [=]

W516: [4] [2ndF][yx]{x} [] [8][1] [=]
115ES: [][9] [] [+] [1] [=]
W516: [2ndF][x2]{} [9] [] [+] [1] [=]

21

Scientific Calculator Comparison Vol I: User Guide and Operational Comparison

3225 = 41

3 25 51

= 4

115ES: [3][x2] [+] [2][x][5] [=]

W516: [3][x2] [+] [2][yx][5] [=]
115ES: [SHIFT][]{3} [2][x][5] [] [-][5] [] [+][1] [=]
W516: [2ndF][STO]{3} [2][yx][5] [] [-][5] [] [+][1] [=]

Exponents and Logarithms

The full range of exponential and logarithmic functions are provided on both the 115ES
and the W516, and those who use these functions often will appreciate how easy it is to enter
expressions containing exponents or logarithms.
Exponents and logarithms can be thought of as three pairs of functions. There are the
base 10 functions; log and 10x, the base e functions; ln and ex, and the base b functions;
logbx and bx. The bx function is actually the [yx] key on the W516 or the [x] on the 115ES.
The nice feature on both calculators is the logbx function. For years, if you wanted to take the
logarithm of x to base b on your scientific calculator, you had to use the clumsy workaround
ln(x)ln(b) or log(x)log(b). Not only is the logbx function faster, but the expressions are less
cluttered and easier to work with.
On both the 115ES and the W516, the [log] and [ln] functions are front face keys. The
inverse of these, the exponential functions 10x and ex, are the same keys shifted. On the
115ES they are [SHIFT][log]{10} and [SHIFT][ln]{e}. And on the W516 they are [2ndF]
[log]{10x} and [2ndF][ln]{ex}. The logbx function on the 115ES has a key of its own at the far
upper right, [log]. On the W516 it is the 2'nd function of the Exp key, [2ndF][Exp]{logax}.

ln 2 0.693

115ES: [ln][2] [=]

W516: [ln][2] [=]

log 7 0.845

115ES: [log][7] [=]

W516: [log][7] [=]

23.14

115ES: [SHIFT][ln]{e} [SHIFT][x10x]{} [=]

W516: [2ndF][ln]{ex} [2ndF][3]{} [=]

log 3 81 = 4

115ES: [log] [3] [] [8][1] [=]

W516: [2ndF][Exp]{logax} [3] [] [8][1] [=]

22

Chapter 5. Manual Calculations

Trigonometric functions and their inverses

All of the standard trigonometric functions and their inverses are present on both
calculators. The standard [sin], [cos], and [tan] functions are all front face keys ( no 2ndF or
shift required ), and the inverses are accessed by pressing [SHIFT] or [2ndF] before pressing
[sin], [cos], or [tan]. The 115ES automatically inserts opening parentheses with these
functions; the W516 does not.
A nice feature of the 115ES is the [SHIFT][Ans]{DRG} key. If you usually operate
with radians as your set angular units ( as is always assumed in this manual ), you can still
input an angle in degrees or gradians if you like. While radians angular units is set, the default
units when you enter an angle for a trig or other function is radians... unless you tell the
115ES differently. That's what the DRG function is for. Type in your trig function, type in
your angle, and then press the [SHIFT][Ans]{DRG} key. A menu pops up, asking what the
angular units are for the angle you're typing. Press [1] for degrees, [2] for radians, or [3] for
gradians. The angle will be converted for you internally and the correct value for the trig
function will be returned. An example would be [AC] [sin] [9][0] [SHIFT][Ans]{DRG} [1]
{deg} [=]. The result returned will be 1. To convert a result or a number to the set angular
units, use the same procedure. Press: [AC] [9][0] [SHIFT][Ans]{DRG} [1]{deg} [=]. and
the calculator responds with /2.
Note that, in the following examples and throughout the rest of this manual, the units of
angular measure are assumed to be set to radians.

1
2

115ES: [sin][/] [SHIFT][x10x]{} [] [6] [=]

W516: [sin][a/b] [2ndF][3]{} [] [6] [=]

sin 30o =

1
2

115ES: [sin][3][0] [SHIFT][Ans]{DRG}[1]{deg} [=]

W516: no equivalent

115ES: [SHIFT][cos]{cos-1} [/][1][][2] [=]

W516: [3][x2] [+] [2][yx][5] [=]

sin

cos1

1
2

tan

= 1

115ES: [sin][/] [SHIFT][x10x]{} [] [6] [=]

W516: [tan][a/b] [2ndF][3]{} [] [4] [=]

23

Scientific Calculator Comparison Vol I: User Guide and Operational Comparison

Hyperbolic functions and their inverses

The 115ES and the W516 both contain the full set of hyperbolic functions and their
inverses. The 115ES, when you press the [hyp] key, presents you with the following menu;

fx-115ES
1: sinh
3: tanh
5: cosh-1

2: cosh
4: sinh-1
6: tanh-1

Key: [hyp]
and you simply press the number key corresponding to the function you want access to.
While this might seem like a nice thing at first, you quickly learn that you need to stop, study
the table, and find the appropriate number for the function. The standard key combo of hyp
key and trig key, used on other calculators, will not work on the 115ES. You have to use the
menu.
The W516 approaches the use of the hyperbolic functions with the standard hyp and
trig key combo. Pressing [hyp][sin] gives you the hyperbolic sine function sinh, [hyp][cos] is
cosh, and [hyp][tan] is tanh. The inverses are gotten with the [2ndF] key. Pressing [2ndF]
[hyp][sin] is the inverse sinh, [2ndF][hyp][cos] is inverse cosh, and [2ndF][hyp][tan] is
inverse tanh. While the W516 does require one more button press than the 115ES to access
hyperbolic functions, you may find that the W516's approach is actually quicker since you just
press the key combos, without having to stop and look up the number for a function on a
menu.
If you use hyperbolics a lot, then you already know about the workaround for accessing
the other three hyperbolic functions and their inverses. These are the hyperbolic secant,
cosecant, and cotangent; sech, csch, and coth., defined by:

sech x =

1
cosh x

csch x =

1
sinh x

coth x =

1
tanh x

while the inverses of these are defined by:

sech1 = cosh1

1
x

csch 1 = sinh1

24

1
x

coth1 = tanh1

1
x

Chapter 5. Manual Calculations

tanh ln 3 =

tanh

3
5

17
8

= 4

115ES: [hyp][3]{tanh} [ln][3] [=]

W516: [hyp][tan] [ln][3] [=]
115ES: [SHIFT][ln]{e} [hyp][6]{tanh-1} [/][3][][5] [][)] [=]
W516: [2ndF][ln]{ex} [2ndF][hyp][tan]{tanh-1} [a/b][3][][5] [=]

= 2

sech ln 2 =
cosh

4
5

4
5

115ES: [/] [1] [] [hyp][2]{cosh} [ln][2] [)][)] [=]

W516: [1] [a/b] [hyp][cos] [ln][2] [=]
115ES: [SHIFT][ln]{e} [hyp][5]{cosh-1} [/][1][7][][8][][)] [=]
W516: [2ndF][ln]{ex} [2ndF][hyp][cos]{tanh-1} [a/b][1][7][][8] [=]

For more information, refer to Appendix K for a list of useful formulas and identities for
the hyperbolic functions and their inverses.

Rectangular / Polar Conversions

The rectangular / polar conversion function on the 115ES and the W516 is very direct
and easy to use, with a few catches. First, these conversions are one of the reasons I call the
variables X and Y dependent variables. The rectangular / polar conversions depend on these
variables to perform the calculation. If you have any data stored in variables X and Y, they will
be erased, overwritten by the results of the conversion. In polar conversions, r is stored in
variable X and is stored in variable Y. The angular units of are the same as the current set
angular units, assumed to be radians in this manual. In rectangular conversions, x and y are
stored in the respective variable memories X and Y.
Rectangular / Polar conversions are also line functions; they will not display or in
the results. You can get or in results though, especially on the 115ES, if you afterward
recall the memory contents X and Y; at that point you are back in 'textbook' mode. The W516
sometimes has problems recognizing exact answers in the results of these conversions,
failing at times to recognize exact radicals.
On the 115ES, the conversions are actually functions, and so they can be used in
expressions. The two conversions are Pol(x,y) and Rec(r,), and they are available on the
keyboard as [SHIFT][+]{Pol(} and [SHIFT][-]{Rec(}. When used in an expression, they are
evaluated to be equal to the first parameter in the conversion ( x or r ), which is stored in
variable memory X as well as in Answer Memory. The 115ES has a funny way of displaying
conversion results. For some reason, the 115ES tries to cram the full result of the conversion
onto a single line of the display, so you will likely have to use the arrow keys to scroll the
display over so you can read the full results.
On the W516, the rectangular / polar conversions would more properly be called

25

Scientific Calculator Comparison Vol I: User Guide and Operational Comparison

commands, since the conversion can only be performed on a keyed in pair of parameters,
separated by a comma. Even enclosing the parameter pair in parentheses results in an error.
The conversions can not be included into an expression. On the keypad, the conversions are
available as [2ndF][8]{r} and [2ndF][9]{xy}. As on the 115ES, the first parameter in the
conversion ( x or r ) is stored in Answer Memory. The W516 does make very good use of the
display, using three lines on the screen for the results. Each parameter is on a separate line
and so no scrolling is necessary to read the results.

fx-115ES

EL-W516

Pol(3,4)

3,4r

r=5 , =0.9272952

r:
:

Key: [SHIFT][+]{Pol} [3]

[SHIFT][)]{,} [4] [)] [=]

5.
0.927295218

Key: [3][x,y][4] [2ndF][8]{r}

Sexagesimal Conversion
The sexagesimal conversions on the 115ES and W516 are typically used to convert
back and forth between angles measured in decimal degrees and angles measured in
degrees, minutes, and seconds of arc. This conversion can also convert between time
measured in decimal hours and time measured in hours, minutes, and seconds. The
sexagesimal conversions, in general, work with a 3-digit number base 60, where each digit is
a base 10 representation of a base 60 digit. These sexagesimal values can be added,
subtracted, multiplied, and divided. They can also be stored into variable memories in
sexagesimal format. When applicable, they can be displayed in fractional format as well.
The 115ES has a peculiar 'memory' feature associated with sexagesimal numbers.
Even when converted to decimal format and stored to a variable memory, the calculator
seems to remember that the number used to be sexagesimal and it will revert back. To get
around this, take the reciprocal of the number twice while it's in decimal format, or try adding
one to it and then subtracting one back off again. When you want the number to stay in
decimal format, these operations seem to shake off the sexagesimal stigma. The W516 tends
to better respect your decision to convert the sexagesimal number to a decimal format.
There are two keys important to operations on sexagesimal numbers; the Deg/Min/Sec
entry key and the conversion key that toggles back and forth between decimal and
sexagesimal. On the 115ES, you key in sexagesimal numbers with the ['] key and the
conversion key is [SHIFT][']{}. On the W516 use the [DM'S] key to enter sexagesimal
numbers, and [2ndF][DM'S]{DEG} to convert back and forth.
We'll do one example that also illustrates another difference between the 115ES and
the W516. Consider an angle that is measured as 13 5' 27". Enter this angle on your

26

Chapter 5. Manual Calculations

calculator. Note that, as you type this in, it will appear as 13527 until you hit the [=] key:
On the 115ES: [1][3] ['] [5] ['] [2][7] ['] [=]
On the W516: [1][3] [DM'S] [5] [DM'S] [2][7] [DM'S] [=]
You'll notice that, after pressing [=], the angle now appears as 13 5' 27" at the bottom
of the display. Now convert this to decimal degrees by pressing [SHIFT][']{} on the
115ES or [2ndF][DM'S]{DEG} on the W516. The 115ES displays a decimal number that
is approximately 5.224166667. The W516 recognizes the number as a fraction and displays it
as 15709 / 1200. There are some cases where the W516 will recognize a number as a
fraction when the 115ES can not.
Store this number into variable memory A. On the 115ES press [SHIFT][RCL][(-)]{A}.
On the W516 press [STO][CNST]{A}. The W516 does so, storing the fractional value to
variable memory A. The 115ES has stored the number but, the display now informs you that
the value in memory A is equal to 13 5' 27". It has reverted back to sexagesimal format. The
workaround for this behavior is to recall variable memory A, add one to it, and then subtract
one from it before storing it back into memory A. It is converted to decimal format in the
process so you do not have to use the [SHIFT][']{} key combo to convert it again.
The W516 has two other sexagesimal conversions in tucked away in the math menu.
[MATH][4]{sec} will convert a sexagesimal result to seconds of arc, and [MATH][5]{min}
will convert a sexagesimal result to minutes of arc.
On both calculators, recall memory A and convert the number back to sexagesimal
format. On the 115ES press [RCL][(-)]{A} [SHIFT][']{}. On the W516 press [RCL][CNST]
{A} [2ndF][DM'S]{DEG}. Both displays will again read 13 5' 27".
Last example. Add 2hr 14min 23sec to 6hr 59min 48sec.
On the 115ES:
On the W516:

[2]['] [1][4]['] [2][3]['] [+] [6]['] [5][9]['] [4][8]['] [=]

[2][DM'S] [1][4][DM'S] [2][3][DM'S] [+] [6][DM'S] [5][9]
[DM'S] [4][8][DM'S] [=]

The answer you should get is 9hr 14min 11sec., displayed as 9 14' 11".
Combinatorial functions
There are three combinatorial functions on the 115ES and W516. The first is the
factorial function x!. The maximum allowed integer argument of the factorial function is 69;
anything larger causes an overflow. The other two are the Combination nCr and Permutation
nPr functions, which are defined in terms of the factorials of their arguments. Technically, you
could consider the power function [yx] on the W516 ( or the [x] on the 115ES ) to also be a
combinatorial function. Here are the standard Permutation and Combination formulas:

27

Scientific Calculator Comparison Vol I: User Guide and Operational Comparison

Without repetition:

Pr =

n!
nr !

Cr =

n!
= a
r ! nr !
b

Remember that, with permutations, order does matter. With combinations, order does
not matter. The nCr function is also sometimes called the 'binomial coefficient'. What these
particular formulas do not allow is repetition among the r selections chosen out of n objects.
When you allow repetition, the formulas become:
With repetition:

Pr = nr

Cr =

nr1!
r ! nr !

The 115ES and W516 both calculate permutations and combinations without repetition
so, if you want to use the formulas with repetition, you'll need to key them in manually.
If you allow the number of objects chosen r to be equal to the number of objects you
can choose from n, then the number of permutations and combinations are simplified. If you
do not allow repetition, nPr=n! And nCr=1. Using these definitions, and some logic, just about
any combinatorial problem can be described mathematically and solved on your calculator.
On the 115ES, the permutation function is accessed by pressing [SHIFT][X]{nPr} and
the combination function is accessed by pressing [SHIFT][]{nCr}. The factorial is the inverse
key shifted; [SHIFT][x-1]{x!}.
On the W516, the permutation function is accessed by pressing [2ndF][6]{nPr} and
the combination function is accessed by pressing [2ndF][5]{nCr}. The factorial is accessed
by pressing [2ndF][4]{x!}.

6P3 = 120

115ES: [6] [SHIFT][X]{nPr} [3] [=]

W516: [6] [2ndF][6]{nPr} [3] [=]

6C3 = 20

115ES: [6] [SHIFT][]{nCr} [3] [=]

W516: [6] [2ndF][5]{nCr} [3] [=]

6! = 720

115ES: [6] [SHIFT][x-1]{x!} [=]

W516: [6] [2ndF][4]{x!} [=]

Random Numbers
The 115ES offers just one type of random number; a three significant digit decimal
number from 0.000 to 0.999, accessed by pressing [SHIFT][.]{Ran#}. The W516 offers four
varieties of random numbers, accessed through the random menu, [2ndF][7]{RANDOM}.
The selections available are: 0:RAND, 1:R-DICE, 2:R-COIN, and 3:R-INT. Selection 0,

28

Chapter 5. Manual Calculations

RAND, is a three digit decimal from 0.000 to 0.999. Selection 1, R-DICE, is a random integer
from 1 to 6. Selection 2, R-COIN, is a random integer from 0 to 1, and selection 3, R-INT, is a
random integer from 0 to 99. These ranges are all inclusive; the possible random values are
greater than or equal to the stated lower limit, and less than or equal to the stated upper limit.
The manual for the 115ES states that Ran# random numbers are three-digit numbers
less than one. The manual for the W516 is more concise and states that the RAND random
numbers are three-digit numbers from 0 to 0.999. Just to be sure, some tests were run to
verify the range of the random numbers. The generated random numbers have been verified,
on both calculators, to be in the range [0,1), which means the possible values run from 0.000
to 0.999 inclusive. This was done by performing a summation of the reciprocals of 5000
random numbers; occasionally, a math error resulted due to division by zero, meaning that a
random number could assume the value of zero. A second test was done by performing a
summation of 10000 terms of 1 / (1Ran#); no math errors due to division by zero have ever
occurred, meaning that a random number never assumed the value of 1. Similar tests were
run on the other varieties of random numbers on the W516, verifying their ranges as well.
The W516 also offers one other kind of random number function, and that is in Matrix
mode. This function is used for the production of random matrices of dimensions up to 4x4.
Unfortunately, you can not specify what kind of random number you want; the only choice is
the default three digit random number, like RAND, from 0.000 to 0.999.
Neither calculator offers an integer Int(x) function, which would come in handy for
creating your own ranges of random numbers or for other uses.
CALC / ALGB Feature
Both the 115ES and the W516 include a repetetive calculation function. On the 115ES,
this feature is called CALC and on the W516 it is called ALGB.
The W516 implementation is simple. In normal calculation mode, enter an expression
with at least one variable. Allowed variables include A-F, M, X, and Y. Select option number 1
from the Math menu by pressing [MATH][1]{ALGB} and you will be prompted for the values
of any variables that are in your expression. Enter a value and press the [=] key for each
variable in your expression. After all the variables have been entered, a calculation result will
be displayed. Press [MATH][1]{ALGB} again to repeat the calculation with new values for
your variables.
The 115ES implementation is identical to that of the W516 except that the 115ES has a
key dedicated to the CALC function. It is the [CALC] key, found in the upper left area of the
keypad. The 115ES also allows you to enter expressions containing an equal sign, as long as
the expression to the left of the equal sign is a single variable. This allows you to perform
some iterative calculations since the values of the variables are updated with each successive
calculation.

29

Scientific Calculator Comparison Vol I: User Guide and Operational Comparison

Miscellaneous
Absolute Value function
The absolute value function on the 115ES [SHIFT][hyp]{Abs} is a dual purpose
function. When applied to a real number x in any mode, it returns |x|. When applied to a
vector in Vector mode, it returns the norm or magnitude of that vector. If a=[a1,a2,a3], then:
Abs
a =
a =

a a a
2
1

2
2

2
3

The W516 also has an absolute value function [2ndF][(-)]{Abs} but, it may only be
applied to real numbers. When applied to a real number x, it returns |x|. The absolute value,
as applied to a vector in determining that vector's norm or magnitude, is a separate function
buried in the vector menu. It is found by pressing [MATH][6][CNST]{abs_list} and may only
be applied to vectors in Vector mode ( or to vectors/lists in Vector/List mode ).
Percentage Calculations
The usual percentage calculations can be performed on either calculator. The [%] key
on the 115ES is [SHIFT][(]{%} and on the W516 it is [2ndF][1]{%}. The only real difference
between the two calculators is that you have to press the [=] key to finish a percentage
calculation on the 115ES while the W516 finishes as soon as you press the [%] key.
To add 25% to 8, calculate 8 + 25% to get the answer 10. To subtract 25% from 12,
calculate 12 25% to get the answer 9. To find 60% of 25, calculate 25 X 60% to get the
answer 15. To find what number 12 is 40% of, calculate 12 40% to get the answer 30.
Constant (K) Calculations on the W516
Whenever you perform a simple operation ( addition, subtraction, multiplication or
division ), part of the calculation is saved and can be used again as a constant operation. The
part of the calculation that is saved depends on which operation you performed. For example,
if you calculate 2 + 3 =5, the +3 part of the expression is saved. Now just type any other
number ( say 8 ) and press the [=] key. The display will show the 8+K calculation with 11
displayed as the result at the bottom. This can be repeated as often as necessary.
Subtraction and division can be used to form similar constant operations. If you
calculate 8 3 =5, the 3 part is saved and can be reused as a constant calculation. If you
calculate 10 2 =5, the 2 part is saved and can be reused as a constant calculation.
Multiplication is a little different. If you calculate 7X 2 =14, the 7X part is saved and can be
reused as a constant calculation. You may also specify a variable memory and the number
stored in the variable is used in the constant calculation.

30

Chapter 6. Scientific / Physical Constants

6. Scientific / Physical Constants

Both the 115ES and the W516 offer a large collection of scientific or physical constants
for use in calculations or as a reference. The 115ES refers to them as scientific constants
while the W516 refers to them as physical constants. In this manual I will use the term
physical constant for both calculators since NIST ( National Institute of Standards and
Technology ) uses the term Fundamental Physical Constants for their collection of
recommended values for these constants. The 115ES contains 40 physical constants while
the W516 contains 52. All 40 physical constants found on the 115ES are also found on the
W516 but the W516 also contains an additional 12 physical constants not found on the
115ES, for a total of 52 constants.
A complete list of all physical constants for both calculators may be found in the
appendices. The serial number, symbol, name, and numeric value ( including units ) are
included with each entry. The bold green entries in the physical constant table for the W516
are the 12 extra constants that are not found on the 115ES.
Being relatively new, the W516 contains the latest (2006) CODATA recommended
values for the physical constants while the 115ES still uses the outdated (1998)
recommended values. For anyone who may notice that the values for the physical constants
supplied by the 115ES do not agree with those on the W516, this is the reason why. The
differences are confined to the last couple digits and so should not pose any real problem for
the average user who will round off calculation results anyways. Those who wish to replicate
calculations from some cutting edge physics experiment will appreciate the availability of
current accepted values of the constants at their fingertips. It will be sufficient to note that the
W516 values will agree with the NIST website 2006 recommended values while those on the
115ES will not.
The physical constants on both calculators may be recalled and used in every mode of
operation except for Base-N.
On the 115ES, you access the physical constants by pressing [SHIFT] [7] {CONST}
and you are presented with a form for entering the two-digit serial number of the required
constant. You must always use two digits for the serial number. For example; constant 7
would be identified as 07. You can locate the required serial number in the table of physical
constants printed on the inside of the hard case cover. Type the serial number and, after a

31

Scientific Calculator Comparison Vol I: User Guide and Operational Comparison

short delay, you'll notice that the symbol for the constant has been transferred to the display.
The symbol appears as a stand-in for the actual value of the constant, just as would be done
with pencil and paper, and this symbol can be used as part of an expression. If you now press
the [=] key, the 115ES evaluates the symbol and the result is the actual numerical value of the
constant.
The physical constants on the W516 can be accessed by pressing the [CNST] key.
Unlike the 115ES, the physical constant table on the W516 is contained within the physical
constants menu itself. You can navigate through the constants, three at a time, by paging up
or down using the [] and [] directional keys. Notice that the index number, the symbol, and
the units for that constant, are all contained in the constants menu. Units are an important
consideration with all physical constants; the W516 supplies the units while the 115ES does
not. The 115ES does supply the units in the manual supplement. [2ndF] [] will take you to
the beginning of the list and [2ndF] [] will take you to the end. You can also type a single
digit to jump to the constants in the list beginning with that digit. Press backspace [BS] to
clear the entered digit to try again and jump to another section of the list with a different
starting digit. When you find the required constant, type in the two-digit index number for that
constant. As with the 115ES, the index number must contain two digits. Automatically, the
numeric value for that physical constant is transferred to the display. Unfortunately, the
symbol is not able to be inserted into an expression. The W516 always sends the numeric
value to the display, so expressions on the W516 that contain physical constants are larger
and more difficult to read than the compact symbolic equivalent expressions on the 115ES.
As an example, we'll calculate the speed of light (c0) from the magnetic (0) and electric
(0) constants, using an expression derived from Maxwell's equations:
c0 =

1
0 0

On the 115ES, clear the display by pressing [AC]. The electric constant (0) is constant
number 32, and the magnetic constant (0) is constant number 33. Evaluate the speed of
light, using the above expression, as follows: [/] [1] [] [] [SHIFT] [7]{CONST} [3][2]
[SHIFT] [7]{CONST} [3][3] and do not press the [=] key just yet. Notice the expression in the
display and how it looks just like the right side of the given expression. Now press the [=] key
and you'll see that the expression evaluates to 299792458, which is just the speed of light c0.
Actually, the full result, if you pull out the digits that are not displayed, is 299792458.014788,
which is still accurate to ten decimal digits. Compare this result to the value for the speed of
light c0, which is constant number 28 on the 115ES.
Now we'll evaluate the same expression on the W516, so clear the display by pressing
[ON/C]. Evaluate the above expression for the speed of light as follows: [1] [a/b] [2nd] [x2]
{} [CNST] [1][3] [X] [CNST] [1][2] and do not press the [=] key just yet. Notice what you
see in the display, or rather what you don't see:
c0 =

1
8.854187817 E121.256637061 E06

32

Chapter 6. Scientific / Physical Constants

There are no symbols for the constants on the W516, so full numeric values for the constants
you've selected are substituted into the expression. In this form, the expression we've entered
is so large that it cannot fit into the display; only the right half of it is visible. This is the one
frustrating thing about the W516 when working with the built in physical constants. Any
expressions containing physical constants are too large to see in their entirety, so getting an
intuitive feel for the form of these expressions as you're entering them is difficult. Now press
the [=] key. The result is not quite the same as that of the 115ES. The W516 evaluates this
expression to 299792458.1 which is a bit different than the accepted value for the speed of
light c0. The full result, which includes the digits not visible in the display, is 299792458.06251.
Compare this result to the value for the speed of light c0, which is constant number 01 on the
W516.
There isn't much more to add regarding the use of physical constants on either
calculator. The expressions involving physical constants may be as simple or as complex as
required, or you may just need to recall the value for reference. In any case, the use of
physical constants is very simple. There are several entries in the Appendices that involve the
use of physical constants if you wish to see more examples.

33

Chapter 7. Metric & Unit Conversions

7. Metric & Unit Conversions

Both the 115ES and the W516 offer a large collection of metric and unit conversions.
The 115ES contains 40 conversions and the W516 contains 44 conversions. There are some
differences between the two collections of conversions. When both calculators contain a
given conversion, the conversion factors are the same.
The 115ES contains six conversions that are not found on the W516. These have to do
with speed, astronomical distance, and pressure conversion.
The W516 contains ten conversions that are not found on the 115ES. These have to do
with fluid measures, calorie, and power conversions. Since the W516 offers ten extra
conversions versus the 115's six extra, the W516 overall offers four more conversions total
over the 115ES.
A complete list of all metric and unit conversions for both calculators may be found in
the appendices. The serial number, conversion entry, and conversion factor are included with
each entry. The bold blue entries are conversions found on the 115ES but not on the W516.
The bold green entries are conversions found on the W516 but not on the 115ES. There are
two red entries. These are the conversions back and forth between Watts and Hp. The W516
uses a conversion factor for going back and forth between Watts and Horsepower. The 115ES
uses a conversion factor for going back and forth between KiloWatts and Horsepower. When
the factor of one thousand between Watts and Kilowatts is taken into account, the conversion
factors are completely equivalent.
The conversions offered by both calculators are fairly standard ones. There are metric
conversions for length and area in size from inches and centimeters up to miles and
kilometers. The area conversion is for square meters and acres. The 115ES throws in parsecs
and speed conversions between m/sec and km/hr. There are fluid measures for liters and
gallons with the W516 including floz and mL. The US and UK standards are both provided for.
Weight conversions include oz, lb, grams, and kilograms. Both calculators provide exhaustive
pressure conversions for pascals, mm Hg, atm, and kgf/cm 2, though neither calculator
contains conversions that include psi. There are many conversions for energy and power in
units of W, J, Nm, kgfm, and various forms of calories. Temperature conversions are for
degrees Celsius and Fahrenheit only.
The metric and unit conversions on both calculators may be recalled and used in every
mode of operation except for Base-N.

35

Scientific Calculator Comparison Vol I: User Guide and Operational Comparison

On the 115ES, you access the matric and unit conversions by pressing [SHIFT] [8]
{CONV} and you are presented with a form for entering the two-digit serial number of the
required conversion. You must always use two digits for the serial number. For example;
constant 7 would be identified as 07. You can locate the required serial number in the table of
metric and unit conversions printed on the inside of the hard case cover. Type the serial
number and a symbol showing the conversion being executed is transferred to the display. If
you now press the [=] key, the 115ES evaluates the conversion and the result is displayed.
The units are never actually attached to the numerical result.
The metric and unit conversions on the W516 can be accessed by pressing the key
combination [2nd][CNST]{CONV}. Just as with the physical constants, the metric and
conversion table on the W516 is contained within the conversion menu itself. You can
navigate through the conversions, three at a time, by paging up or down using the [] and
[] directional keys. Notice that the index number and the conversion are contained in the
constants menu. [2ndF] [] will take you to the beginning of the list and [2ndF] [] will take
you to the end. You can also type a single digit to jump to the conversions in the list beginning
with that digit. Press backspace [BS] to clear the entered digit to try again and jump to
another section of the list with a different starting digit. When you find the required conversion,
type in the two-digit index number for that conversion. As with the 115ES, the index number
must contain two digits. A code for that conversion is transferred to the display. Unfortunately,
the code is simply the letters cv followed by the serial number of the conversion. Pressing the
[=] key executes the conversion and the result is displayed. As with the 115ES, the units are
never actually attached to the numerical result.
The following shows what the display looks like when 5 miles is converted to kilometers
on both calculators:
fx-115ES

EL-W516

5mikm

5cv07=
8.04672

8.04672

Key: [5] [SHIFT][8]{CONV} [0]

[7]{mikm} [=]

Key: [5] [2nd][CNST]{CONV}

[0][7]{mikm} [=]

It just so happens the the conversion from miles to kilometers on both calculators is serial
number 07. Note that the result is displayed as a rational number and so you'll have to
change the format to get the decimal result. On the 115ES press the [SD] key and on the
W516 press the [CHANGE] key.

36

Chapter 8. Complex Number Calculations

8. Complex Number Calculations

The 115ES and W516 perform the basic

complex number operations of addition, subtraction,
multiplication, and division. Complex numbers may
also be squared, cubed, and inverted. There are two
methods of representation; rectangular (x+yi) and
polar (r,), and you may convert back and forth
between the two. The only strictly complex operation
is the complex conjugate, which is just the same
complex number (x+yi) with the complex component
y negated. There is also the modulus or magnitude
which is computed using the absolute value function.
Complex numbers contain a real and an imaginary component, and can be expressed
in the form x+yi, where i is the imaginary unit with the property i 2 = 1. As represented on the
calculator, x is the magnitude of the real component and y is the magnitude of the imaginary
component. As shown in the illustration, a complex number Z can be plotted on the complex
plane with the position along the x-axis representing the real component of Z, and the position
along the y-axis representing the imaginary component of Z. The modulus or magnitude of Z
is the length of the hypotenuse of the right triangle formed by the real and imaginary
components of Z. This is shown as r in the illustration. The argument or phase of Z, which the
115ES will compute for you, is the angle between the real axis and the modulus r.
Complex number calculations can only be performed in complex mode. On both the
115ES and the W516, you access complex mode using the mode menu. On the 115ES,
complex mode is [MODE][2], and on the W516 it is [MODE][3].

37

Scientific Calculator Comparison Vol I: User Guide and Operational Comparison

fx-115ES
1: COMP
3: STAT
5: EQN
7: TABLE

EL-W516 ( page 1 )

2: CMPLX
4: BASE-N
6: MATRIX
8: VECTOR

<MODE-1>
0: NORMAL 1: STAT
2: DRILL
3: CPLX
4: MATRIX
5:LIST

Key: [AC] [MODE][2]

Key: [ON/C] [MODE][3]

One thing to keep in mind regarding complex mode on the W516 is that there is only
one variable memory that you are able to access, and that is the independent memory M. On
the 115ES, you may use variables A through D and the independent memory M. On both
calculators, the dependent variables X and Y are unavailable since they are used by the
operating system to perform complex number calculations. If this is a concern, you may want
to consider vector mode, which has more memory available for storage. Complex number
calculations are very much like vector calculations and, with a little ingenuity, vector mode will
be suitable for just about any complex number calculation you may need to do.
When you eventually exit complex mode, the values stored in the variable memories
are retained but, the imaginary components are cleared, leaving the real components
available for recall in another mode.
To the left is an illustration of the example we
will be using for complex number calculations. The
complex number Z will be entered using rectangular
coordinates and stored in variable memory M. The
real component x, is equal to the square root of three,
and the imaginary component y, is equal to 1i or just
the value i. The polar representation of Z is also
given in the illustration. The magnitude of Z is equal
to two, and the phase of Z is equal to /6. This value
of Z was chosen for its simplicity and for the fact that
it contains a radical and a nice fraction of .
Remember that, on the 115ES, you select your default complex number representation
in the setup menu, and it can be set to rectangular (a+bi) or polar (r). To access this
setting, press; [SHIFT][MODE]{SETUP} [] [3]{CMPLX}. Your choices are 1:a+bi or 2:r.
For now, select rectangular mode, option one.
While in complex mode on the 115ES, press the [ENG] key to place the imaginary unit
i, into your expressions. To enter complex number Z into the 115ES, enter it just as you would
write it on paper; [AC] [][3][] [+] [ENG]{i }, and store it into memory M by pressing;
[SHIFT][RCL]{STO} [M+]{M}. Recall it by pressing; [AC] [RCL][M+]{M}.
On the W516, entering Z is similar. Press [ON/C] [2ndF][x2]{} [3] [+] [dx]{i } and
store the value into variable M by pressing; [STO][M+]{M}. Recall the value to the display

38

Chapter 8. Complex Number Calculations

again by pressing; [ON/C] [RCL][M+]{M}. Now look to the left edge of your display. You
should see a small icon, either xy or r. There is no 'default' complex representation but you
can freely and easily switch back and forth between rectangular and polar form. To switch to
rectangular form, press; [2ndF][9]{xy}, and to switch to polar form, press; [2ndF][8]{r}.
In this case, we want rectangular form.
Notice the difference between the displays of the 115ES and the W516, when the value
stored in M is recalled. The 115ES maintains radical form but the W516 loses it and presents
the value as a decimal equivalent.
fx-115ES

EL-W516

M=
1.732050808
+1.i

3 + i
Key: contents of M

Key: contents of M

This is one of the advantages of the 115ES in complex number calculations; radicals
are not converted to their decimal equivalent... unless you want that to happen; just hit the
[SD] key to toggle back and forth between radical form and decimal form. Very nice. The
[CHANGE] key on the W516 has no effect on complex calculation results.
Suppose we wish to see the value of M in polar form. On the 115ES, the conversions
between polar and rectangular form are found in the complex menu. You can access the
complex menu by pressing [SHIFT][2]{CMPLX}. Item 3:r will convert the argument to
polar form, and item 4:a+bi will convert the argument to rectangular form. To see complex
number Z in polar form, press; [AC] [RCL][M+]{M} [SHIFT][2]{CMPLX} [3]{r} [=]. On
the W516, you only need to change the display to polar form by pressing [2ndF][8]{r}.
Now compare the displays from the two calculators.
fx-115ES

EL-W516
M=

Mr

2.
0.523598775

2 1/6
Key: contents of M

Key: contents of M

Now we see that, not only does the 115ES preserve radicals in rectangular form, but it
also preserves the value of in angular measures in polar form. You may also press the
[SD] key to toggle between the decimal form and an exact form involving . The W516
again returns only the decimal equivalent. This is true even if you perform operations such as
M2 or M3, which is calculated in the same way as with real numbers. The value of M3 for
example, should equal 81/2 . This is the value returned by the 115ES; the W516 returns
39

Scientific Calculator Comparison Vol I: User Guide and Operational Comparison

the decimal equvalent.

Instead of converting complex numbers from rectangular to polar form, you may enter
them in polar form in the first place. This is what the {} key is for. On the 115ES it is the key
combo [SHIFT][()]{}. On the W516 just press the [DM'S]{} key. For example, we'll
repeat the above calculation by entering the complex number Z in polar form and cubing it.
On the 115ES, press; [AC] [(] [2] [SHIFT][()]{} [/] [SHIFT][x10x]{} [] [6] [] [)]
[SHIFT][x2]{x3} [=]. You should get the result 8i. On the W516 press; [ON/C] [(] [2] [DM'S]
{} [(] [2ndF][3]{} [] [6] [)][)] [2ndF][RCL]{x3} [=]. You should get the result 8.i.
Remember that, on the W516, you may have to hit [2ndF][9]{xy} to get the rectangular
form of the result.
When expressed in polar form, a complex number Z has the representation r, where
the value of r is the magnitude or modulus of the number. This is the distance in the complex
plane, between the origin and the point Z. To isolate this value, on either calculator, you can
use the absolute value function. In the example where we calculated M3, the magnitude
should be equal to 8. To calculate this on the 115ES, press; [AC] [SHIFT][hyp]{Abs} [RCL]
[M+]{M} [SHIFT][x2]{x3} [=]. This returns just the value 8, now separate from its phase . On
the W516, perform this same calculation by pressing; [ON/C] [2ndF][(-)]{abs} [RCL][M+]{M}
[2ndF][RCL]{x3} [=].
Alternatively, you can isolate the value of the phase angle with Z in either polar or
rectangular form. The phase is also known as the argument and is easily calculated on the
115ES by using the argument function, which is option 1:arg in the complex menu. Calculate
the phase of M2 on the 115ES by pressing; [AC] [SHIFT][2]{CMPLX} [1]{arg} [RCL][M+]{M}
[x2] [=]. The result should be 1/3 . There is no easy way to isolate the phase of a complex
number on the W516 except visually verifying its value in polar form and then retyping that
value into another calculation.
One complex number function that both calculators perform is the complex conjugate.
This is a very simple idea but is very useful in many complex number calculations. When a
complex number Z is expressed in rectangular form a+bi, then the complex conjugate is just
abi. To access this function in the complex menu on the 115ES, press [SHIFT][2]{CMPLX}
[2]{Conjg(}. On the W516 it is item 1:CONJ in the math menu.
The usual math operations on complex numbers, on either calculator, are fairly simple.
If you have two complex numbers, Z=a+bi and Z2=c+di, then the rules are the same as those
that apply performing algebra on real numbers. For example, if we add the values of Z and Z2,
we get; Z + Z2 = a+b+ci+di = (a+b) + (c+d)i. Another example, if we calculate the product of
Z and Z2, we get; ZZ2 = ac + adi + bci + bdi2 = (acbd) + (bc+ad)i. Both calculators will
easily handle either of these operations, as well as subtraction and division. Just watch your
parentheses and order of operations or you may end up inadvertently splitting your complex
number during a calculation. This applies to polar or rectangular forms. The best bet is to
always place your complex numbers within a set of parenthese to be sure the real and
imaginary components are operated on together.

40

Chapter 9. Base-N Calculations

9. Base-N Calculations
The W516 does not have an actual Base-N mode; instead, Base-N mode is entered
automatically when a number is converted to a base other than base-10. There are five
different number bases available on the W516; base-2 (binary), base-5 (pental), base-8
(octal), base-10 (decimal), and base-16 (hexadecimal). The base conversions, math, and
logic functions are all available directly on the keypad itself.
The 115ES has a mode specifically for Base-N calculations, and you can access BaseN mode through the mode menu by pressing [MODE][4]{BASE-N}. The 115ES has four
different number bases available; base-2 (binary), base-8 (octal), base-10 (decimal), and
base-16 (hexadecimal). The base conversion functions are available directly on the keypad.
Base-N math and logic functions are then accessed through the base-N menu by pressing
[SHIFT][3]{BASE}.

fx-115ES ( page 1 )

fx-115ES ( page 2 )

1: and
3: xor
5: Not

1: d
3: b

2: or
4: xnor
6: Neg

Key: [AC] [SHIFT][3]

{BASE}

2: h
4: o

Key: [AC] [SHIFT][3]

{BASE} []

On page 2 of the 115ES base menu are the base over-rides. These are prefixes that
allow you to enter numbers from any of the bases available regardless of the current set
number base. It's a very handy feature. For example, you may enter hFF+o36 while binary is
the set base and get the answer in binary, which is b100011101 ( d255+d30=d285 ). The
W516 does not have this feature; you have to change number bases as you proceed through
a calculation to get the same results.

41

Scientific Calculator Comparison Vol I: User Guide and Operational Comparison

Calculation Ranges
Base-N values on the 115ES are limited internally to a maximum of 32 binary bits,
except for binary numbers, which are limited to just 16 binary bits. The base-N value ranges
on the 115ES reflect these internal limits, taking into account the fact that these are integers
with roughly half the values positive and half negative in the representation of signed numbers
in two's-complement format. Since binary bits determine the maximum range of the numbers
represented, you'll find varying maximum numbers of digits in the bases: 16 for binary, 11 for
octal, 10 for decimal, and 8 for hexadecimal.
There is the question of why there is a 16-bit limit on binary numbers. The width of the
display is exactly 16 characters so a 16-bit binary number is the largest that can be displayed
without scrolling. If you're working on a problem with more than 16 binary bits, hexadecimal
would be a more convenient number base to work in anyway, and you get the full 32 bits.

Base-N Value Ranges: fx-115ES

Base
Binary
Octal
Decimal
Hexadecimal

Value Ranges

Decimal Equivalent

Positive: 0000 0000 0000 0000 x 0111 1111 1111 1111

0 x +32767 (215-1)

Negative: 1000 0000 0000 0000 x 1111 1111 1111 1111

-32768 (-215) x -1

Positive: 000 0000 0000 x 177 7777 7777

0 x +2147483647 (231-1)

Negative: 200 0000 0000 x 377 7777 7777

-2147483648 (-231) x -1

Range: -2147483648 x 2147483647

(-231) x (231-1)

Positive: 0000 0000 x 7FFF FFFF

0 x +2147483647 (231-1)

Negative: 8000 0000 x FFFF FFFF

-2147483648 (-231) x -1

Base-N value ranges on the W516 can be a little confusing. Instead of basing the limits
of internal representation of numbers on binary bits, the W516 bases them on digits. For all
intents and purposes, each number base is allowed to use ten digits; ten binary digits, ten
octal digits, etc., limited by the maximum representable value, which is 9,999,999,999
decimal. Binary, pental, and octal are not limited by this maximum value so they are allowed
the full ten digits. Hexadecimal is limited since the ten-digit hex value 7FFFFFFFFF is equal
to decimal 549,755,813,887, which is greater than the allowed maximum of 9,999,999,999.
For this reason, the largest hexadecimal value allowed is 2540BE3FF which equals decimal
9,999,999,999. This awkward condition exists also for negative hexadecimal numbers.

42

Chapter 9. Base-N Calculations

Base-N Value Ranges: EL-W516

Base
Binary
Pental
Octal
Decimal
Hexadecimal

Value Ranges

Decimal Equivalent

Positive: 00 0000 0000 x 01 1111 1111

0 x +511 (29-1)

Negative: 10 0000 0000 x 11 1111 1111

-512 (-29) x -1

Positive: 00 0000 0000 x 22 2222 2222

0 x 4882812

Negative: 22 2222 2223 x 44 4444 4444

-4882812 x -1

Positive: 00 0000 0000 x 37 7777 7777

0 x +536870911 (229-1)

Negative: 40 0000 0000 x 77 7777 7777

- 536870912 (-229) x -1

Range: -9,999,999,999 x 9,999,999,999

(-231) x (231-1)

Positive: 00 0000 0000 x 2 540B E3FE

0 x +9,999,999,998

Negative: FD ABF4 1C01 x FF FFFF FFFF

-9,999,999,999 x -1

For the rest of this chapter, each number given in an example will be preceeded by
lower case letter, indicating the base; b for binary, p for pental, o for octal, d for decimal, and
h for hexadecimal.
Base Conversion
When you enter base-N mode on the 115ES by pressing [AC][MODE][4]{BASE-N},
decimal or base-10 is the current set base. This means you may only key in values which
make sense in base-10, and all calculation results will be returned in base-10 representation.
As an example, we'll key in the current year in base-10. On the W516, make sure you are in
normal calculation mode, the only mode that supports base-N operations, by pressing [ON/C]
[MODE][0]{NORMAL}. On either calculator, key in the current year [2][0][1][0], and press the
[=] key. To convert the year to hexadecimal on the 115ES, press [x]{HEX}. To do the same
on the W516, press [2ndF][X]{HEX}.

fx-115ES

EL-W516

2010

AnsHEX
Hex
000007DA

HEX

Key: [x ]{HEX}

7DA

Key: [2ndF][X]{HEX}

43

Scientific Calculator Comparison Vol I: User Guide and Operational Comparison

Notice that the conversion changes the current set base and converts any result in the
display to the new base. The current set base is always displayed along with the results. To
convert the year to octal (base-8) on the 115ES, press [ln]{OCT}. On the W516, press [2ndF]
[]{OCT}. On either calculator the result will be o3732. For binary representation, press
[log]{BIN} on the 115ES to get the result b0000011111011010. On the W516 you press
[2ndF][]{BIN} and get a nasty calculation error. Why? You've exceeded the maximum
allowed digits. You're only allowed ten digits in binary, which is a range of -d512 to +d511. No
problem; us old timers use hex anyway and interpret the bits from h7DA. The h7=b0111,
hD=b1101, and hA=b1010. Put them all together and you get b0111 1101 1010. The W516 is
able to give a pental (base-5) representation; d2010=p31020. The 115ES does not offer the
pental number system in base-N mode.
To change the base at any time, without regard to the last calclulated result, clear the
display by pressing [AC] on the 115ES or [ON/C] on the W516; then press the appropriate
keys to convert to another number base:

Binary

Pental

Octal

Decimal
2

Hexadecimal

fx-115ES

[log]

--------

[ln]

[x ]

[x]

EL-W516

[2ndF][]

[2ndF][=]

[2ndF][]

[2ndF][+]

[2ndF][X]

2's Complement: Negative Numbers in Base-N

Neither the 115ES nor the W516 allow you to choose between signed and unsigned
values for base-N calculations. All values are integers and all representations of numbers are
signed. The method used is 2's complement, which is a natural way to represent signed
integers in base-N since negative and positive numbers do not require any special treatment
during calculation.
In binary, with an N-bit representation, the 2's complement is of a number K, is the
binary representation of K. It is formally calculated as 2NK but, since we are working only
with integers, the shortcut is to calculate NOT(K)+1. The NOT function, in binary, will reverse
the bits of K; all the 1's become 0's and all the 0's become 1's. Take for example the number
d26, which is equal to b11010. We will use 8-bit representation such that d26 is now equal to
b00011010. To calculate NOT(K)+1, we first reverse the bits, resulting in b11100101, then we
add one to the result which gives us b11100110. This is the 2's complement representation of
the decimal number 26. Now take another number, say d53. In 8-bit binary representation,
this is equal to b00110101. Now suppose we add d53 and d26.
d53 =

00110101

+ d26 =

11100110

d27

b 100011011

44

Chapter 9. Base-N Calculations

The carry bit can be neglected in this operation, so the result is b00011011, which is
equal to d27. The calculation is correct.
The generalized 2's complement can be used in any number base to allow for math on
negative numbers in base-N. On the 115ES and W516, the NEG operation calculates the 2's
complement for us in any of the number bases available. The internal algorithms that carry
out arithmetic in base-N will keep track of the signs and carries for us and will generate an
error in the case of an overflow, indicating that the operation performed resulted in a number
that is outside of the allowed range for that base.
Arithmetic in Base-N
All the available number bases can be represented using the number keys for the
digits, except for hexadecimal. Hex digits 10 through 15 are represented by the letters A
through F. These are the same keys that are used for the variable memories, and are marked
above the appropriate key.
The usual operations are allowed; addition, subtraction, multiplication, and division,
using the normal [+], [], [X], and [] keys on the keypad. When an arithmetic operation
results in a fraction, the decimal portion of the result is truncated ( not rounded ) to leave an
integer result. As an example, we'll do a division in hexadecimal. To change to base-16 on the
115ES, press [AC][x]{HEX}. On the W516, press [ON/C][2ndF][X]{HEX}. Calculate h1E
divided by h8 by pressing, on either calculator, [1][E] [] [8] [=]. The answer should be 3.
This same problem, in normal calculation mode, would be 30 8 = 3.75. Notice that the
decimal portion of the result, 0.75, is truncated or dropped.
Here is an example that makes use of multiple number bases, and converts the result
of the operation to another base. What is o33 + hAB in binary and decimal? On the 115ES,
set binary as the base by pressing [AC][log]{BIN}. On the W516 set octal as the base by
pressing [ON/C][2ndF][]{OCT}. Then to perform the calculation:
On the 115ES, press: [SHIFT][3]{BASE}[][4]{o} [3][3] [+] [SHIFT][3]{BASE}[][2]
{h} [A][B] [=]. Result in binary is b11000110. To see this in decimal, press: [x2]{DEC}. Result
in decimal is d198.
On the W516, press: [3][3] [2ndF][X]{HEX} [+] [A][B] [=] [2ndF][]{BIN}. Result
in binary is b11000110. To see this in decimal, press: [2ndF][+]{DEC}. Result in decimal is
d198.
Results may be stored in the variable memories, if you like. When you leave base-N
mode, the results will still be stored in the memories but they will be converted to base-10.
Likewise, values calculated and stored into the variable memories will still be available when
you go into base-N mode, but they will be truncated to integer values or result in a math error
if the numbers stored in a recalled memory are out of range for the base you are working in.

45

Scientific Calculator Comparison Vol I: User Guide and Operational Comparison

Logic Operations in Base-N

The following table contains the truth tables for all of the built-in logic functions on the
115ES and the W516. I've also included two common compound logic functions that can be
built from the ones that are native to the calculator. These logic functions can be applied to
any value in any base, but are best understood in binary. If A and B are two binary numbers,
then the functions below are applied bitwise to the two arguments, A and B.

Input
A B NOT A A AND B A OR B A XOR B A XNOR B

NOR =
NAND =
NOT(A OR B) NOT(A AND B)

The logic functions on the 115ES are found on the first page of the base menu. On the
W516, the logic functions are the row of keys containing the hyp, trig, and integral key. The
logic function for each key is printed above the key in a light gray color.
We'll do two examples. What is h73 AND hAA in decimal? We'll set hex as the base on
both calculators. On the 115ES, press [AC][x]{HEX}; on the W516 press [ON/C][2ndF][X]
{HEX}. Then, to perform the calculation:
On the 115ES, press: [7][3] [SHIFT][3]{BASE} [1]{AND} [A][A] [=]. The result in hex
is equal to h22. Press [x2]{DEC} to see the result in decimal, which is d34.
On the W516, press: [7][3] [sin]{AND} [A][A] [=]. The result in hex is equal to h22.
Press [2ndF][+]{DEC} to see the result in decimal, which is d34.
Here is this same problem in binary. In this form, it's easier to see how the AND
function works with the bits of the two input numbers:
h73 =

01110011

AND hAA =

10101010

00100010

d34

Second example. What is o221 NOR o243 in decimal? Since NOR is not one of our

46

Chapter 9. Base-N Calculations

built-in logic functions, we'll have to reformulate the problem using the OR function and
calculate instead: NOT(o221 OR o243). We'll set octal as the base on both calculators. On
the 115ES, press [AC][ln]{OCT}; on the W516 press [ON/C][2ndF][]{OCT}. Then, to
perform the calculation:
On the 115ES, press: [SHIFT][3][5]{NOT} [2][2][1] [SHIFT][3]{BASE} [2]{OR} [2][4]
[3] [=]. The result in octal is equal to o37777777514. Press [x2]{DEC} to see the result in
decimal, which is d180.
On the W516, press: [hyp]{NOT} [(] [2][2][1] [cos]{OR} [2][4][3] [)] [=]. The result in
octal is equal to o7777777514. Press [2ndF][+]{DEC} to see the result in decimal, which is
d180.
Here is this same problem in binary. In this form, it's easier to see how the NOR
function works with the bits of the two input numbers:
d145 =

10010001

NOR d163 =

10100011

01001100

d180

How is it that the answer is d180 when b1001100 is equal to d76? We've neglected to
show the leading zeros in our problem. Since the result bit will be a 1 whenever the bits of
both arguments are equal to 0, our answer should have a string of leading 1's. This means
that our number is a negative number. The most significant bit of a 2's complement binary
number acts as a sign bit as well as contributes to the value of the number. If the most
significant bit is a 0, the number is positive. If it's a 1, the number is negative. Since we're
working with N=8 bits, we'll use the formula for calculating the two's complement of K=76, and
we'll do the calculation in decimal. The value of 2N-K in this case is 2876, or 25676, which is
equal to 180, the same result that the 115ES and W516 calculated.

47

Chapter 10. Function Table ( 115ES only )

10. Function Table ( 115ES only )

The 115ES has a separate mode for the generation of a function table, a capability that
is missing on the W516. Given an equation y=f(x), the function table application will generate
a table of x values from a specified Start value to a specifiec End value, in increments
specified by the Step value. For each x value in the table, a y value is calculated based on
the expression you defined for y=f(x). The Start, End, and Step values specified may not
result in a table that exceeds thirty x values. The Start, End, and Step values may also be
expressions that evaluate to a real number, as long as those expressions do not contain
summations, integrals, derivatives, or Rec/Pol conversions.
The function table is read only; no entries can be edited. As usual, x is the only allowed
variate; any other variable is treated as a constant. Dependent variable memory X is modified
by the generation of the function table so don't store any important values in memory X. The
function table and the entered expression for f(x) is cleared if you leave function table mode
or turn the calculator off.
The function table is useful for displaying a range of values for f(x); either for the
purpose of plotting the function or for getting a general idea of where the roots are located if
you want to find where f(x)=0 using the General Solver. The function table is a simple yet
powerful tool and therefore fast and simple to use, as it should be.
To enter function table mode, press; [AC] [MODE][7]{TABLE}. You are immediately
prompted for the expression to be used as f(x). As an example, we'll enter X24X+1 by
pressing; [RCL][)]{X} [x2] [] [4][RCL][)]{X} [+] [1] [=]. You are then prompted for a Start
value; press [][5][=] for this example. You are then prompted for a Stop value; press [5][=]
for this example. You are then prompted for a Step value; press [0][.][5][=] for this example.
The function table should now appear; there will be a total of twenty-one x values,
beginning with 5 and ending with 5, and every 0.5 increment in between. The initial view of
the function table should be something like the following:

49

Scientific Calculator Comparison Vol I: User Guide and Operational Comparison

fx-115ES
X F(X)
1
5
46
2 4.5 39.25
3
4
33
5

Key: sample function table

You can use the arrow keys to navigate the table and inspect the values. The current
highlighted cell's contents are displayed on the bottom line of the screen. You may use the
[SD] key to switch between fractional and decimal displays of cell contents on the bottom
line. If you scroll down through the table, you'll see that f(x) changes from a positive to a
negative value somewhere between x=0,y=1 and x=0.5,y=0.75. The sign changes again
back to positive somewhere between x=3.5,y=0.75 and x=4,y=1. You can estimate that the
real roots of f(x)=x24x+1 are are located at about x=0.25 and x=3.75.
When the equation f(x)=x24x+1 is entered into the Equation Solver ( for details, refer
to Chapter 12 ), the solutions returned are 0.2679491924 and 3.732050808, which agrees
with what we've estimated, within a reasonable tolerance.
After you're finished inspecting the function table, press [AC] to return to the prompt for
f(x). You can run the same equation again by just pressing [=]. You are prompted again for
Start, Stop, and Step values, with the current values displayed for you to either accept or edit.
This time I entered zero for the Start value, 0.5 for the Stop value, and 0.02 for the Step
value. According to the table, the first real root is somewhere between x=0.26,y=0.0276 and
x=0.28,y=0.28,y=0.0416. This puts the root at about 0.27. You can continue paring down the
range of the root using this method, for as many digits as you feel is necessary. If you go one
more step using Start=0.26, Stop=0.28, and Step=0.001, you can pare down the root as
being just about 0.2675. The General Solver would be able to take it from there, provided any
other possible roots are sufficiently separated from your estimate. If your f(x) is a difficult one
to find roots for using the General Solver, this is an excellent way to ensure that the solver will
be able to converge to the proper root.
To run the function table again with a different expression for f(x), press [AC] to clear
the expression while at the prompt for f(x). If you're finished with the function table altogether,
you can exit the application by using the mode menu.

50

Chapter 11. The General Solver

11. The General Solver

Next to manual calculations and physical constants/conversions, probably the most
used applications on the scientific calculator are the solvers. In the 115ES and the W516, this
capability is divided between the General Solver and the Equation Solver. The General Solver
is only available in normal calculation mode. The Equation Solver is a separate mode on both
calculators.
The General Solver calculates a numerical solution to an equation containing just
about any combination of functions you can think of. The algorithm used on both calculators is
known as the Newton-Raphson method and is able to iteratively calculate roots of real-valued
functions. The algorithm uses the numerical derivative of a function at an estimated solution
point (x0) to calculate the x-intercept of a line tangent to the curve at the initial estimate. This
x-intercept is a refined estimate (x1) of the desired root of the function. The process is then
repeated with x1 and the iterative process continues until a stable estimate for the solution is
found. That happens when xn-xn-1 is less than or equal to a very small preset value. The
General Solver on the W516 is very fast and clean. The implementation on the 115ES may be
slower but it is a very flexible solver that many may prefer. The Newton-Raphson algorithm
can be very fast and efficient, especially if used wisely, so the speed hit on the 115ES
implementation is not that big of a deal but definitely noticeable.
There are trade-offs for using such a handy solver. One is that, should there be
multiple solutions to an equation, the General Solver can only provide one solution at a time.
The solver has to be run again, with another initial estimate and tolerance, to find any other
roots. If you are not very familiar with the equation you are solving, you may have to make
several guesses and do some fiddling to locate the roots. In most cases however, even a few
widely spaced blind guesses will be sufficient for the algorithm to calculate the desired roots.
There are cases, especially with periodic functions, where no solution is found, even if they
exist. The solution is to make another guess for the initial estimate or enter a more precise
initial estimate by inspecting a rough graph of the equation.
The General Solver can not contain integrals, derivatives, summations, or polar or
rectangular conversions.

51

Scientific Calculator Comparison Vol I: User Guide and Operational Comparison

General Solver Example #1

Here is an example where a numerical solver can be very useful. It is derived from
calculations in orbital mechanics concerned with true and mean orbital anomalies. Due to the
appearance of the transcendental function (sine) within the equation, it cannot be solved
algebraically:
M sin M = 1
The choice of setting the expression equal to 1, beyond it being a non-zero value, was
arbitrary. From other sources, we know that the equation M+sin(M) will equal 1 if M is equal
to 0.51097342938856910952, to a precision of twenty significant digits.
The 115ES, being a very flexible solver, is able to process this equation directly. The
General Solver on the 115ES is able to solve for any variable, and we happen to have a
variable M at our disposal; it is our Independent Memory M., so we will go ahead and use it.
The 115ES is also able to solve an expression that is set equal to another value or another
expression. The W516 can only solve an expression that contains the variable X, and the
expression may not contain an equal sign. For the W516 to process this equation, it will have
to be modified. In most cases this is not difficult at all. In this case, we subtract one from both
sides and change the M to X. Our expression, for the W516, will look like this:
X sin X 1
Now we can enter our expression into the 115ES and the W516, side by side, and
watch the solver in action:
fx-115ES

EL-W516

M + sin(M) = 1 , M

X + sinX - 1

Key: [AC] [ALPHA][M+]{M} [+]

[sin][RCL][M+]{M} [)] [ALPHA]
[CALC]{=} [1] [SHIFT][)]{,}
[RCL][M+]{M}

Key: [ON/C] [RCL][RCL]{X} [+]

[sin][RCL][RCL]{X} [-] [1]

A few things to note. On the 115ES, if you are solving for some variable besides X, you
have to let the solver know by placing the variable you want to solve for after the expression,
separated by a comma. If you want to solve an expression for X, then you don't have to add
the comma and variable after the expression since X is the default variable to solve for. Also
on the 115ES, the first variable in the expression cannot use the shortcut [RCL] in place of
[ALPHA] for typing a variable; if you try, it will pop up in the expression as Ans.

52

Chapter 11. The General Solver

To launch the solver on the 115ES, key in [SHIFT][CALC]{SOLVE}. On the W516, you
key in [MATH][2]{SOLVER}.
fx-115ES

EL-W516

Solve for M

X + sinX - 1
Start?
some number is here

Key: [SHIFT][CALC]{SOLVE}

0.
Key: [MATH][2]{SOLVER}

This is the point in the algorithm where our initial guess is typed in. On the W516, you
are prompted for an initial estimate by the Start? that appears at the middle left portion of the
display. The zero at the lower right is just the current value of X, which is now zero since the
solver will always clear X and use zero as an initial estimate if you do not specify one. On the
115ES, you have no prompt telling you it's waiting for your initial estimate. There is just a
number at the lower right corner of the screen which means nothing; it's just the last value
that was stored into the M variable.
For the 115ES, press [1][=], which specifies the value 1 as the initial estimate and
launches the solver directly. The display goes blank while the calculation is carried out. On the
W516, press [1][=]. This specifies the initial estimate as the value 1. You will be greeted by
another prompt which asks dx? Just leave it at the default value of 0.00001 and press [=]
again, launching the solver. The display will blank briefly while the calculation is performed.
After about 4 seconds on the 115ES, or about 2 seconds on the W516, the results will
appear as shown below.
fx-115ES

EL-W516

M + sin(M) = 1 , M
M=
0.5109734294
L-R=
0

X + sinX - 1
X=
0.510973429

Key: [SHIFT][CALC]{SOLVE}

Key: [MATH][2]{SOLVER}

The result on the W516, including the hidden digits, is 0.51097342938858 which is
accurate to 13 significant digits. The last digit is in error. The results are stored in the X
memory for later use if needed. The result on the 115ES, including the hidden digits, is
0.510973429388571 which is also accurate to 13 significant digits. The result is stored in the
M memory for later use if needed. Besides the numerical result, the 115ES also gives a figure
for L-R which will usually have a value of zero. This figure gives an estimate of the accuracy
of the result, which is gotten by plugging the calculated result back into the original equation
and checking to see how well the equation has balanced. A result of L-R should be zero or a
very very small number for your result to be considered accurate.

53

Scientific Calculator Comparison Vol I: User Guide and Operational Comparison

General Solver Example #2

The General Solver on the 115ES can handle expressions that contain multiple
variables and it's not necessary for any of them to be the variable X. In such cases, specify
the variable to be solved for by stating the variable after the expression, separated from it by
a comma. You will be prompted for the value of any other variables present in the expression.
In the next example, specifically for the 115ES, we will investigate the following equation:
For the 115ES. Solve
For the W516. Solve

2B
= AB2
A

for A when B = {1, 2, 3, 4, 5}

X 22X BX 2B = 0

for X when B = {1, 2, 3, 4, 5}

The expression is a quadratic and will have two real roots. Hint: one of the roots will
always be -2 but we want to find the other real root. You may have to change your initial
estimate a few times to locate the real root that is not equal to -2.
Solution on the W516:
The W516 can not solve for any variable besides X, and any other variables present in
the expression are treated as constants. The above equation, to satisfy the requirements of
the W516 General Solver, will have to be rearranged to form an expression that is equal to
zero and then X must be substituted in for the variable to be solved for; A in this case. To do
this, multiply both sides by A and then subtract 2B from both sides. After substituting X for A in
the expression, we have X2+2X-BX-2B , or X(X+2-B)-2B if you prefer. Since B is treated as
a constant on the W516, you'll have to run the General Solver five different times, each time
storing the required value into memory B before running the solver. You may want to save
yourself some typing by saving the expression into a function memory for easier recall. Let's
store the expression X(X+2-B)-2B into function memory F4 by pressing: [ON/C] [ALPHA]
[RCL]{X} [(] [RCL][RCL]{X} [+] [2] [-] [RCL][yx]{B} [)] [-] [2][RCL][yx]{B} [STO][]{F4}.
EL-W516
X ( X + 2 - B ) - 2B
F4
Key: Expression stored in F4

To find X when B={1, 2, 3, 4, 5} on the W516, store a value into memory B and then
recall the expression from function memory F4: [ON/C] [1] [STO][yx]{B} [RCL][]{F4}. Now

54

Chapter 11. The General Solver

launch the General Solver: [MATH][2]{SOLVER}. Type in a positive value when prompted for
an initial estimate by Start?. The value 10 usually works.
EL-W516
X ( X + 2 - B ) - 2B
Start?

0.

Key: [ON/C] [1] [STO][yx]{B}

[RCL][]{F4}
[MATH][2]{SOLVER}
Press [=] and then press [=] again when prompted for dx? since the default works just
fine. If the solution x= -2 appears ( this is the root we don't want to find ), press [=] to re-run
the solver and try a different initial estimate until you find the other root. When you find it,
press [ON/C] to terminate the solver and start again by saving the next value from the list
B={1, 2, 3, 4, 5} into memory B: [2][STO][yx]{B}. Continue until all the roots for all values of B
are found.
Solution on the 115ES:
The 115ES does not have function memories available for you to use but that isn't a
problem in this particular application. The equation again is:
Solve

2B
= AB2
A

for A when B = {1, 2, 3, 4, 5}

We only need to key the equation in once on the 115ES and notify the General Solver
of which variable to solve for. We will be prompted by the solver to supply the values of any
other variables present in the equation. In this case we want to solve for A when B is equal to
one of the values from the list B={1, 2, 3, 4, 5}. Key in: [AC][/][2][RCL]['"]{B}[][RCL]
[(-)]{A} [] [ALPHA][CALC]{=} [RCL][(-)]{A} [-] [RCL]['"]{B} [+] [2] [SHIFT][)]{,} [RCL][(-)]
{A}.
fx-115ES
2B
= AB2 , A
A

Key: Entered expression

55

Scientific Calculator Comparison Vol I: User Guide and Operational Comparison

You can see the expression in the display with the A at the end, separated from the
expression by a comma, letting the solver know we want to solve for A. Launch the General
Solver by pressing [SHIFT][CALC]{SOLVE}.
fx-115ES
B?
some number here
Key: [SHIFT][CALC]{SOLVE}
The appearance of the variable B in the expression causes the solver to prompt us for
the value of B. We'll begin with B={1, 2, 3, 4, 5} so key in the first value: [1][=].
fx-115ES
Solve for A
some number here
Key: [1][=]
Now, since there are no other variables present in the expression, the solver is
satisfied and is ready to solve for A. There is no prompt for an initial estimate on the 115ES
but the Solve for A that appears in the display means that it's time to key it in. We'll use zero
as an initial estimate so key in: [0][=].
fx-115ES
2B
= AB2 , A
A
A=
L-R=

1
0

Key: [SHIFT][CALC]{SOLVE}
After a few seconds with a blank display as the solver does its job, the display appears
as above meaning the solver has located the desired root. When B=1, then A=1. Press [=] to
rerun the solver and you will be prompted again for a value of B. This time key in [2][=]. Enter
an initial estimate to locate the root when B=2. Continue the process until all the roots for the
list of values of B are located. Press [AC] to terminate the General Solver.
There is a very obvious pattern to the roots of the quadratic equation as the value of B

56

Chapter 11. The General Solver

is changed. Very obvious! We've tried integer values for B, from 1 to 5. How about noninteger values? How about negative values or other values for B outside the given list? What
happens if B is set equal to -2 and why?
Tips for the General Solver
There may be times when the 115ES will have a hard time locating a solution to an
equation. After a preset number of iterations, the solver may ask you if you want to continue
attempts at finding the solution. A prompt will appear: Continue: [=]. Press [=] to continue
with the search for a solution or press [AC] to cancel the operation.
If the W516 is taking a long time finding a solution, or you get an Error 2: calculation
error, there is another option besides trying a different initial estimate of the solution. Try using
a custom value for dx. A smaller value will calculate the iterations more precisely, perhaps
steering the algorithm closer to the solution.
The solution to an equation, after the General Solver is terminated, will be stored in the
variable that was solved for. This value can be accessed as usual and used for further
calculation if required.
There are an almost unlimited number of tasks that the General Solver can be put to
work on so learn to use it and use it well. It won't be long before you realize that it is one of
the most powerful and useful tools on the 115ES and W516.

57

Chapter 12. The Equation Solver

12. The Equation Solver

The Equation Solver is much more specific than the General Solver. It can give
algebraic solutions to equations, even multiple and/or imaginary solutions all at once, if they
exist. It is very fast and no initial guesses or tolerances are required as input from the user.
The trade-off here is that only certain forms of equations are recognized and accepted by the
Equation Solver. For the Equation Solvers in the 115ES and the W516 you may specify a
quadratic polynomial, a cubic polynomial, simultaneous solutions to two linear equations with
two unknowns, or simultaneous solutions to three linear equations with three unknowns.
There are definite advantages to using the Equation Solver vs the General Solver. For
quadratic or cubic polynomials, all solutions are listed together instantly, including any
imaginary roots, without the user having to guide a numerical process with initial estimates or
tolerances. The solution of simultaneous linear equations ( two equations with two unknowns
or three equations with three unknowns ) can only be accomplished using the Equation
Solver; the General Solver is not capable of solving linear systems.
An important reminder here is that equations are not always recognizable as a form
that the Equation Solver can handle. You may have to do a little homework first. As an
example, suppose you are looking for a real number, x, such that:
2
= x2
x

This is perfectly acceptable as an input to the General Solver, which uses a numerical
approach to finding a solution. You can type it into the General Solver, get your solution, and
move on with your life. You might miss out on something though, such as... what if there is
more than one solution?. The Equation Solver is not able to work with the problem in this form
but you can do some rearranging. Multiply through the equation by x, and then subtract 2
from both sides:
2 = x 22x

then

0 = x 22x2

or

x 22x2 = 0

There are then two solutions to the original problem since it can be restated as a
quadratic polynomial set equal to zero. The roots of the quadratic, and there are two of them,

59

Scientific Calculator Comparison Vol I: User Guide and Operational Comparison

will both satisfy the original equation. A little work with pencil and paper, before turning to an
electronic solution, is good practice and will give you a little more insight to the nature of the
problem you are attempting to solve.
By the way, the solution to the above equation is

1 3 .

While the General Solver is available only in normal calculation mode, the Equation
Solver has its own mode and is a menu driven application specifically designed to solve the
four types of equations mentioned above. To access the Equation Solver on the 115ES, press
[MODE][5]{EQN}. On the W516 press [MODE][][6]{EQUATION} or just [MODE][6], since
it's not really necessary to arrow down to page two of the mode menu before pressing [6] to
access the Equation Solver.
fx-115ES

EL-W516

1: anX + bnY = cn
2: anX + bnY + cnZ = dn
3: aX2 + bX + c = 0
4: aX3 + bX2 + cX + d = 0

<< EQUATION >>

0: 2-VLE
1: 3-VLE
2: QUAD
3: CUBIC

Key: [MODE][5]{EQN}

Key: [MODE][6]{EQUATION}

Options 1 and 2 on the 115ES ( or options 0 and 1 on the W516 ) are the linear system
solvers, which will calculate the simultaneous solution to a system of linear equations. In
linear equations, each term contains a single variable X and Y ( for 2-value systems ) or X, Y,
and Z ( for 3-value systems ), and a single coefficient. The coefficients can be any real
number, including zero. Each linear equation in the system is set equal to a constant term on
the right side of the equation. The required form of 2 equations in 2 unknowns is as follows:

a1 X b1 Y = c1
a 2 X b2 Y = c 2

The required form of 3 equations in 3 unknowns is as follows:

a 1 X b1 Y c 1 Z = d 1
a 2 X b2 Y c 2 Z = d 2
a3 X b3 Y c 3 Z = d 3

The solver will calculate the values of X and Y ( for 2-value systems ) or X, Y, and Z

60

Chapter 12. The Equation Solver

( for 3-value systems ) that will satisfy all the linear equations in the system simultaneously, if
such solutions exist. Both the 115ES and the W516 use an internal algorithm based on
matrices and so it follows that only matrices with non-zero determinants will result in a
solution to the system of equations. Singular matrices ( determinants of which are equal to
zero ) result in a math error. If this happens, there are two possibilities; either no solution to
the system exists, or an infinite number of solutions exist parameterized by one of the three
variables. The Equation Solver can say no more than that in such cases, and you'll have to
resort to pencil and paper in order to determine if the solution set is infinite or null.
Options 3 and 4 on the 115ES ( or options 2 and 3 on the W516 ) are the polynomial
solvers. These are polynomials of a single variable, X for either calculator, and the maximum
degree for any term is 2 for quadratic polynomials or 3 for cubic polynomials. Each term has a
coefficient ( a-c or a-d ) which can be any real number, including zero. The solver will
calculate the roots of the polynomial, or the value of X which causes the polynomial to equal
zero. These solutions may be real or imaginary ( complex ), or both.
The required form of a 2'nd degree polynomial is as follows:
a X 2b X c = 0
The required form of a 3'rd degree polynomial is as follows:
a X 3b X 2cX d = 0
The 115ES and the W516 only require you to enter the coefficients ( a, b, c, and d ) in
the Equation Solver since the variables ( X, Y, and Z ) are fixed.
The system for entering the coefficients on the 115ES uses a form that resembles a
table or matrix and you can navigate around the table using the directional keys. The table
entries are 'tiny print' truncated versions of entered coefficients but larger print full digit
versions of the coefficients are shown on the bottom line of the display, revealing the full
contents of the coefficient that the cursor is currently positioned over. This is very intuitive,
reflects the form of the equations you are keying in, and you can see a good portion of
everything you've entered.
On the W516, the input form is just one long vertical table that you have to scroll up
and down through in order to see what you've typed in. There's no clear separation between
one equation and the next in the linear equation modes, so you have to read the coefficient
subscripts to be certain. The one positive comment is that all the digits of each coefficient are
displayed in full and in large type. I'd give this up in a minute though, to have to intuitive table
format of the 115ES.

61

Scientific Calculator Comparison Vol I: User Guide and Operational Comparison

Polynomial Solver Example #1

I will give two examples of the use of the polynomial Equation Solvers. One quadratic
which will have two real roots, and one cubic which will have one real root and two complex
roots. We'll begin with the quadratic equation. Suppose we want to find the values of x such
that 2x2+6x=20. We must first put this into a form accepted by the solver so we subtract 20
from both sides.
2x 26x20 = 0
We now have a quadratic polynomial in x with descending degrees each term, and the
polynomial is set equal to zero. This particular problem could easily be solved by the General
Solver as well but we'll use the Equation Solver instead for this example. On the 115ES key in
[MODE][5][3]{Quad Solver} and on the W516 key in [MODE][6][2]{Quad Solver}:
fx-115ES
a
[

b
0

EL-W516
c

a:
b:
c:

0]

0.
0.
0.

0
Key: [MODE][5][3]{Quad
Solve}

Key: [MODE][6][2]{Quad
Solve}

Remember that the Equation Solver only prompts for the entry of the coefficients. In
the case of the quadratic polynomial in the example, the coefficients are: 2, 6, and -20. So
a=2, b=6, and c=-20. To enter the coefficients on the 115ES, press [2][=] [6][=] [(-)][2][0][=]
and notice the cursor advances to the next coefficient automatically. You can also use the
directional arrow keys to navigate back and forth between the coefficients or press [AC] to
clear all the coefficients in case you need to start over. Press [=] when you are ready for the
solution to be calculated. To enter the coefficients on the W516, press [2][=] [6][=] [(-)][2][0]
[=] and notice that the cursor moves down to the next coefficient automatically. You can also
use the directional arrow keys to navigate up and down between the coefficients or press
[2ndF][MODE]{CA} to clear all the coefficients in case you need to start over. Press [=] when
you are ready for the solution to be calculated.
fx-115ES

EL-W516

X1 =

X=
1:
2:

2.
-5.

2
Key: [=] solves the equation

Key: [=] solves the equation

62

Chapter 12. The Equation Solver

On the 115ES, each solution has its own screen so use the [] and [] keys to
navigate through the solution screens to see all the solutions. The W516 will show all of the
solutions on one screen. The solutions are x=2 and x=-5., the roots of the original quadratic
2x2+6x=20. Either value for x satisfies the equation which can be easily verified.

Polynomial Solver Example #2

This example is a little more interesting in that the General Solver will only be able to
find one of the three solutions. The numerical algorithm the General Solver uses will have
little trouble finding the real solution to the equation but it won't be able to find the other two
complex solutions. Here is the polynomial we will be considering:
x 38x 214x68 = 0
This is a cubic polynomial ( the term with the highest degree is x3 or third degree ), and
this time there are four coefficients: 1, -8, 14, and 68. On the 115ES key in [MODE][5][4]
{Cubic Solver} and on the W516 key in [MODE][6][3]{Cubic Solver}:
fx-115ES
a
[

b
0

EL-W516
c

a:
b:
c:
d:

0
0

Key: [MODE][5][4]{Cubic}

0.
0.
0.
0.

Key: [MODE][6][3]{Cubic}

Notice that, on the 115ES display, the coefficient d is not visible. It is located off the
right side of the screen but will become visible as you enter your coefficients and the cursor
advances to the right. Enter your coefficients just as you did with the quadratic solver: [1][=]
[(-)][8][=] [1][4][=] [6][8][=], the same key sequence on either calculator. Press [=] when you
are done entering the coefficients. There is a brief pause while the cubic Equation Solver
calculates the solution.
fx-115ES

EL-W516

X1 =

X=
1:
2:
-2

Key: [=] solves the equation

-2.
5.
3.i

Key: [=] solves the equation

63

Scientific Calculator Comparison Vol I: User Guide and Operational Comparison

Again, the 115ES shows one solution per page, so use the [] and [] keys to
navigate through the solution screens to see all the solutions. The solutions the 115ES give
are: X1=-2, X2=5+3i, and X3=5-3i. The W516 displays all solutions on one page. The real
solution is X1=-2, and the complex solution is X2=53i.
The real solution -2, if you plot the polynomial x3-8x2+14x+68, is the x-coordinate
where the graph of the equation crosses the x-axis. You can not see the complex solutions on
a real number graph of the equation, but you can show, mathematically, that the complex
solutions will satisfy the equation. Take for example one of the complex solutions, x=5+3i :
x 38x 214x68

with

x = 53 i

= 53 i 3853 i 21453 i68 = 10198 i8 1630 i 1453 i68

= 10198 i128240 i7042 i 68 = 10198 i128240 i7042 i 68
= 101287068198 i 240 i42 i = 0198 i240 i42 i = 00 i
To check this on your calculator, you could go into Complex Mode and store 5+3i into a
memory, say M. Then calculate M3-8M2+14M+68. The answer should be simply zero. See
Chapter 8, Complex Mode calculations, for more detail.

Linear System Solver Example

The 115ES and the W516 are both capable of working with linear systems of
equations. Two of the four items in the Equation Solver menu have to do with systems of
linear equations; one for two equations with two unknowns and one for three equations with
three unknowns. In the 115ES Equation Solver ( [MODE][5]{EQN} ), item one is labelled as:
anX + bnY = cn and item two is labelled as: anX + bnY + cnZ = dn, the general form of the linear
equations in the system. The index n runs from one to two or from one to three, respectively,
incremented for each equation in the system. On the W516, these linear systems are items 0:
2-VLE and 1: 3-VLE in the Equation Solver menu ( [MODE][][6]{EQUATION} ).
Here again are the Equation Solver menus on the 115ES and the W516:
fx-115ES

EL-W516

1: anX + bnY = cn
2: anX + bnY + cnZ = dn
3: aX2 + bX + c = 0
4: aX3 + bX2 + cX + d = 0

<< EQUATION >>

0: 2-VLE
1: 3-VLE
2: QUAD
3: CUBIC

Key: [MODE][5]{EQN}

Key: [MODE][6]{EQUATION}

64

Chapter 12. The Equation Solver

One example should be sufficient for demonstrating how the linear system solver on
both calculators function, so we will choose a three equation system with three unknowns; X,
Y, and Z. Recall that earlier the general form of a three equation system with three unknowns
was given as:

a 1 X b1 Y c 1 Z = d 1
a 2 X b2 Y c 2 Z = d 2
a3 X b3 Y c 3 Z = d 3

and you can see that the coefficients here are an, bn, cn, and dn. Notice that the coefficient
subscripts (n) run from one to three for the three equations. The unknowns are X, Y, and Z.
Here then is the linear system that we will be solving:

7 X 2 Y 1 Z = 37
0 X 3Y 4 Z = 9
2 X 1 Y 4 Z = 23

Because of the way you will be keying the equations into the Equation Solver, it will be
helpful if you think of this linear system as a matrix of coefficients:

a1=7 b1=2 c 1=1 d 1 =37

a 2=0 b 2=3
c 2=4
d 2 =9
a 3=2 b3=1 c 3=4 d 3=23

To enter the linear Equation Solver for three equations with three unknowns, press the
key sequences show in the graphics below. Once you enter the key sequence, you will be
dropped immediately into the coefficient input form.
fx-115ES
a
1
2
3

b
0
0
0

EL-W516
c

0
0
0

a1
=
b1
=

0
0
0

0.
0.

0
Key: [MODE][5][2]{3-L.Eqn}

Key: [MODE][6][1]{3-L.Eqn}

On the 115ES, type in the coefficients, one row after another just as you would when
entering matrix data, pressing [=] for each entry. Note that, initially, you can not see the matrix
column for coefficient d. It is off of the right side of the screen but will scroll into view as you

65

Scientific Calculator Comparison Vol I: User Guide and Operational Comparison

move through the matrix. The cursor will advance through the coefficient matrix for you
automatically as you key in values. You can also use the [][][][] keys to navigate your
way around the matrix at will to make any edits. At any time press [AC] to set all of the
coefficients equal to zero. When you press the [=] key without making any edits, the 115ES
leaves the coefficient input form and begins solving the linear system.
On the W516, the coefficient input form is one long continuous list, and it can get
confusing at times, trying to keep straight which coefficient it is you're keying in. Enter the
coefficients in order, pressing the [=] key for each entry. The cursor will advance down the list
for you automatically. You can also use the [][] keys to move up and down the coefficient
list to make any changes. At any time press [2ndF][MODE]{CA} to set all of the coefficients
equal to zero. When you press the [=] key after editing coefficient d3, the W516 leaves the
coefficient input form and begins solving the linear system.
The W516 is definitely much faster at solving linear systems but the process is quick
enough on either calculator that this makes little difference for normal use. After a brief period
of calculation ( screen is blank ), the solution screen for the linear system appears.

fx-115ES

EL-W516

X=

X:
Y:
Z:
D:

3
Key: solution to linear system

3.
-5.
6.
-134.

Key: solution to linear system

The 115ES displays the result that X=3. Press [=] or use the [][] keys to go back
and forth through the solution screens to find that Y=-5 and Z=6. When you press the [=] key
while on the Z solution screen, you will be taken back to the coefficient input form where you
can run the solver again with a new set of coefficients ( or edit the old ones ). If you are done
with your work, you can exit the solver using the Mode menu.
The W516 shows the complete solution set on one screen. You can see that the W516
also solved the system; X=3, Y=-5, and Z=6. It also displays something called D: which has a
value of -134. This is the determinant of the coefficient matrix for the terms anX, bnY, and cnZ.
This excludes the constant terms dn. In other words, the matrix:

7 2 1
A = 0 3
4
2 1 4

has a determinant of -134, which is useful information to have for some applications. The
W516 has just this one page for displaying the solution to the system so, if you press the [=]
key while on the solution screen, you will be taken back to the coefficient input form where

66

Chapter 12. The Equation Solver

you can run the solver again with a new set of coefficients ( or edit the old ones ). If you are
done with your work, you can exit the solver using the Mode menu.
How can we verify a solution to a system of linear equations? If we look at our original
linear system and the solution the solver calculated:

7 X 2 Y 1 Z = 37
0 X 3Y 4 Z = 9
2 X 1 Y 4 Z = 23

with

{ }
X = 3
Y = 5
Z = 6

one way that we can verify it is to substitute in the values calculated for the unknowns to
verify that each of the linear equations in the system has been balanced.

7 32516 = 37
0 33 546 = 9
2 31 546 = 23

} {
or

21106 = 37
01524 = 9
6524 = 23

} {
or

37 = 37
9 = 9
23 = 23

Since all three linear equations are balanced, we have found the values of X, Y, and Z
such that all three equations in the system are simultaneously solved. This can take a fair
amount of typing, even for this simple system.
Perhaps a quicker way would be to verify the matrix equation for the system in Matrix
mode. Here is the matrix equation for the system we need to verify:

7 2 1
A = 0 3
4
2 1 4

Ax = b where

[]

3
x = 5
6

[ ]

37
b =
9
23

where A is the coefficient matrix, x is the solution vector, and b is the vector for the constant
term from our linear system.
When working with the matrix form of the system, we only need to be concerned with
the coefficients. We key in the equation above, finding that:

7 2 1
0 3
4
2 1 4

] [] [ ]
X

67

3
5
6

37
9
23

Scientific Calculator Comparison Vol I: User Guide and Operational Comparison

which gives a solution that is the same as the b vector that came from the constant terms in
our linear system. The solution to the matrix equation agrees when we use the solution the
Equation Solver calculated. The solution to the system is verified. For more information and
examples on matrices, refer to Chapter 14.
Note: The W516 is capable of working with 4X4 matrices and so is capable of solving
linear systems of four equations with four unknowns. For some reason, this capability was not
built into the Equation Solver but, you can certainly work with such linear systems on your
own in Matrix mode. See Appendix G for more information.

68

Chapter 13. Summations, Integrals, & Derivatives

13. Summations, Integrals, & Derivatives

The 115ES and the W516 both contain functions for performing numerical integration
and numerical differentiation. Both calculators also offer numerical summation. As far as
actually entering the expressions and parameters for these functions, there are only slight
differences between the 115ES and the W516. A greater difference will be noted in
performance while evaluating the functions. In brief, the 115ES is a faster and more powerful
integrator, the W516 is faster at summation, and both calculators are about equally adept at
differentiation. The W516 has the added benefit of being able to perform integration,
summation, and differentiation on stored expressions in function memories F1-F4.
When the key for summation or integration is pressed on either calculator, a template
appears that is required to be filled in before its evaluation can be performed. You must
specify a function to be summed or integrated, a lower limit (LL), and an upper limit (UL). The
W516 also offers some useful optional parameters. In the table below, numbers in [brackets]
indicate position within the template. Position [1] is the default location of the blinking cursor
when the integration or summation key is first pressed. You must use the arrow keys to
navigate through the expression while entering the required parameters. Pressing [] will
move the cursor to higher numbered positions within the template, and pressing [] will move
the cursor to lower numbered positions within the template. The template positions for both
calculators are given in the table below.
Form Templates: Summation & Integration
fx-115ES

EL-W516

[3]

[0]

[1] [4]

[2]

[0]

[2]

[0]

[3]

[2] [1] dx

[ 4]

[3],step [4]

UL

prev

[1]

[0]

[2]

[1] [3],subs dx

[4]

func

next

LL

prev

UL

LL func dx

next

The default position of the cursor on the 115ES upon entering a summation or

69

Scientific Calculator Comparison Vol I: User Guide and Operational Comparison

integration template is the function (func)position.. The W516 default cursor position upon
entering a summation or integration template is the lower limit (LL). Navigation through the
template on the 115ES can be a bit clumsy since the template positions jump around through
the expression in a sort of loop. The template positions on the W516 are pretty much straight
through so navigation is easier and more natural. The previous (prev) and next (next)
positions are for navigation outside of the integration or summation, if it is part of a larger
expression.
For both calculators, X is the only allowed variable of summation or integration. Other
variables may be contained within the summation or integration but they are treated as
constants. The W516 does allow some modification of the summation and integration
commands. While in the function position of the summation template, you may specify a
stepsize (step) after the function, separated by a comma. The default stepsize is one. While
in the function position of the integration template, you may specify the number of subintervals (subs) after the function, separated by a comma. The default number of subintervals is 100. The 115ES does offer modification of the integration command but, this is
only permitted in Line Mode. Consult your manual for details since Line Mode is not covered
here.

Numerical Summation Examples

Summation is the process of adding together the numeric evaluations of a function for
all values of an index variable, as the index variable is incremented by a stepsize from a lower
limit to an upper limit. For the 115ES and W516, the index variable is internally set to use X
and the index variable is incremented by one (the default stepsize) for each evaluation.
The stepsize on the 115ES is fixed at one. The W516 allows you to specify the
stepsize to any value, positive or negative, and not necessarily an integer.
To get an idea for how numeric summation is accomplished on the 115ES and the
W516, here are a few examples to try on your own. Key sequences are given for all
examples. The first example is the sum of all the squares from one to a hundred:
100

X2

= 338350

X=1

To evaluate on the 115ES: [AC] [SHIFT][log]{} [RCL][)]{X} [x2] [][1] [][1][0][0]

[=].
To evaluate on the W516: [ON/C] [2ndF][(x,y)]{} [1] [][1][0][0] [][RCL][RCL]{X}
[x2] [=].
The second example is the sum of all the squares of the even numbers from 2 to 100.
This example may take advantage of the stepsize option the W516 makes available for
summation. The 115ES will have to modify the function in order to achieve the same result:

70

Chapter 13. Summations, Integrals, & Derivatives

fx-115ES

EL-W516

50

X 2

100

X 2, 2

= 171700

X=1

= 171700

X=2

To evaluate on the 115ES: [AC] [4] [SHIFT][log]{} [RCL][)]{X} [x2] [][1] [][5]
[0] [=].
To evaluate on the W516: [ON/C] [2ndF][(x,y)]{} [2] [][1][0][0] [][RCL][RCL]{X}
[x ][(x,y)][2] [=].
2

The evaluation of this last example on the 115ES requires some thought before it can
be keyed in. The stepsize for summation on the 115ES is set at one and can not be modified.
If we set the limits of summation from 1 to 50 and double the value during the summation,
then we are summing x2 for the terms {2,4,6, ... 100} instead of {1,2,3, ... 50}. This means our
summed expression must be changed from x2 to (2x)2 for the terms. This is equal to 4x2 and
the 4 is a constant that can be pulled to the outside of the summation. The expression is then
4 times the summation of x2 as x goes from 1 to 50.
Here are a few more summation examples you may try on your own. These are
summations from a speed comparison of the 115ES and W516. The W516 is the faster of the
two calculators overall so it should be no surprise that it's faster at numerical summation than
the 115ES is by about a factor of two. The following table gives the timing results of the
various summations performed.
Numerical Summation Timing Comparison
fx-115ES

EL-W516

Execution
Time (sec)

Result

Execution
Time (sec)

Result

8.5

1.6349839

5.7

1.6349839

10.1

-0.09912279951

5.9

-0.09912279951

148.477767

3.6

148.477767

12.7

91.85793086

6.8

91.85793086

100

x12
x=1

50

sin x
x=1
50

ln x
x=1

50

5 x
x=1

71

Scientific Calculator Comparison Vol I: User Guide and Operational Comparison

Numerical Integration Examples

The 115ES has, by far, the best numerical integrator I have ever seen on a scientific
calculator. The numerical integration on the W516 is good but, when compared to what the
115ES is armed with, the W516 seems so much slower and weaker. The W516 is an amazing
machine in so many other ways that it's hard to understand why the numerical integration
would be left so inadequate.
Integration will evaluate a smooth continuous function over an interval, dividing the
area under the curve into many thin vertical strips. The areas of the strips are interpolated,
usually by an easily integrated polynomial, and summed up to closely approximate the area
under the curve over the interval.
The integration algorithm on the W516 is Simpson's Method. This method splits the
area under a curve into a number of equally spaced subintervals (100 by default ) and fits
each one with a parabolic function. This is a very elementary method of numerical integration
and does not perform well for many types of functions. The Gauss-Kronrod Quadrature
algorithm on the 115ES is a much more efficient and advanced method of numerical
integration. This algorithm varies the width of the subintervals and fits a more tailored
polynomial to each one. This method gets the job done with fewer calculations overall and so
performs numerical integration much faster and more accurately.
In the below table is a timing comparison for numerical integration on the 115ES and
the W516. Depending on the expression being integrated, the 115ES can be 2 to 6 times
faster than the W516 and is nearly always more accurate.
Numerical Integration Timing Comparison
fx-115ES

0 sin x dx
e

1 ln x dx
8

1 x 2 dx
4

0 x1 dx

EL-W516

Execution
Time (sec)

Result

Execution
Time (sec)

Result

3.6

2 exactly

22.4

2.000000001

2.9

1 exactly

14.2

0.999999999

6.5

7/8

11.8

0.875000199

0.8

12 exactly

2.5

12 exactly

The W516 offers an optional parameter for its numerical integration; the number of
subintervals between the upper and lower limit of integration. This parameter is entered

72

Chapter 13. Summations, Integrals, & Derivatives

immediately after the integrand, separated from it by a comma. The default value of this
parameter is 100 subintervals if you choose not to specify a custom value. A greater number
of subintervals means the results will be more accurate but, it also means that the integration
will take longer to evaluate.
As an example of how the use of the optional subintervals parameter affects accuracy
and calculation time, we'll use the integral of the natural exponential function over the interval
of -2 to 2. From other sources, we know that the value of this integral should equal
7.2537208156940375353 to 20 significant digits. The 115ES evaluates this integral in 3
seconds flat with the returned decimal result of 7.25372081569396 and is accurate to 12
significant digits. To achieve this level of accuracy, the W516 would have to perform a
numerical integration with a custom subintervals parameter set to about 500 and a resulting
calculation time of about a full minute.
The below table illustrates the trade-off between the number of subintervals specified,
the execution time, and the number of accurate significant digits in the results obtained from
the W516. The red digits in the results are incorrect digits.
W516 Numerical Integration Sub-intervals Parameter

2 e X , 4 dx
2

2 e X , 20 dx
2

2 e X , 100 dx
2

2 e X , 500 dx

Sub-intervals

Execution
Time (sec)

Result

Sig Digits
Accuracy

1.0

7.256166427503

20

3.1

7.2537248407443

100

13.4

7.253720822136

500

64.0

7.253720815692

12

Integration Example #1: Cubic Polynomial

We'll inspect a cubic polynomial function for this first example.
f x = x 33x 24x12
The function has three real roots, values of x where the function evaluates to zero.
These occur at x=-2, x=2, and x=3. Our interest will be to evaluate the function's integral from
x=-2 to x=2, where the function is positive such that the integral over the interval will also be
positive. The area under the curve, or the integral from x=-2 to x=2, is already known to be
exactly 32. To verify this, we need to evaluate the following integral.

73

Scientific Calculator Comparison Vol I: User Guide and Operational Comparison

2 x3 3x 24x12 dx

= 32

To evaluate on the 115ES: [AC] [] [RCL][)]{X}[SHIFT][x2]{x3} [-] [3][RCL][)]{X}[x2]

[-] [4][RCL][)]{X} [+][1][2] [][(-)][2] [] [2] [=].
To evaluate on the W516: [ON/C] [dx] [(-)][2][] [2][] [RCL][RCL]{X}[2ndF][RCL]
{x } [-] [3][RCL][RCL]{X}[x2] [-] [4][RCL][RCL]{X} [+][1][2] [=].
3

The 115ES evaluates the integral in 1.9 seconds, returning an exact integer value of
32. The W516 evaluates the integral in 9.4 seconds and also obtains the exact answer of 32.
In this example, both calculators performed well as far as obtaining correct results. The
115ES was faster as expected, performing the integration nearly five times faster than the
W516. Low-order polynomials are not very good tests of numerical integration though; the
internal algorithm uses polynomials to interpolate over the interval so exact answers may be
expected from both calculators. We'll try something a little more difficult.
Integration Example #2: True RMS Equation
One of the more surprising results of numerical integration on the 115ES that I've run
across is the following expression:
RMS =

sin x dx

2
2

which is the square root of the mean squared value over a half-cycle sine wave. This is the
true RMS or 'root mean square' value often encountered in electricity, which states that a
proportion of the peak voltage value of an alternating current source represents an equivalent
power voltage of a direct current power source. The value is a constant and is one-half of the
square root of two, or 0.7071 to four significant digits. The 115ES finishes the integral
evaluation in just 4.2 seconds and returns the exact symbolic result above, not a decimal
approximation. The 115ES, in this superb example of its fast and powerful numerical
integration algorithm, provides an exact answer the square root of two over two. Previous to
the 115ES, a high end graphing calculator would have been required to obtain an exact
evaluation of the RMS equation. The W516 takes 25.1 seconds to finish evaluating the
expression and comes up with just a decimal approximation of 0.707106781 to nine
significant digits, though internally the value is correct to eleven significant digits.

74

Chapter 13. Summations, Integrals, & Derivatives

To evaluate the RMS equation on the 115ES: [AC] [][/][] [sin] [RCL][)]{X} [)]
[x2] [][0] [][SHIFT][x10x]{} [][] [SHIFT][x10x]{} [=].
To evaluate the RMS equation on the W516: [ON/C] [2ndF][x2]{} [dx] [0][]
[2ndF][3]{}[] [(] [sin][RCL][RCL]{X} [)][x2] [][a/b] [2ndF][3]{} [=].
In this case, the 115ES performed the integration nearly six times faster than the
W516, and the 115ES obtained an exact result.
Integration Example #3: Standard Normal Distribution
One of the more difficult equations to integrate numerically is the standard normal
probability distribution. In this example we will calculate the probability that a normal random
variable will fall within one standard deviation of the mean if the distribution has a mean of
zero and a standard deviation of one. This means that the normal probability distribution will
need to be integrated from -1 to 1 numerically. We know already, from other sources, that the
result should be equal to 0.68268949213708589717 to twenty significant digits. The
integration required to calculate this value is:

e
2

1
2

x
2

The 115ES completes evaluation of the integral in 4.5 seconds, providing a decimal
approximation of 0.682689492137081 when internal as well as displayed digits are accounted
for. Amazingly, all 15 digits, except the last, are correct. Only high end graphing calculators
are faster or more accurate, and then by just a margin. The W516 completes the task in 22.8
seconds and provides the decimal answer 0.68268949219058 which is accurate to just 10
significant digits. We can compare these results to those obtained from two high end graphing
models. The TI Voyage 200 calculator completes the integration in 1.4 seconds with the result
0.68268949213709 in which all 14 digits are correct (the last digit is rounded). The HP50g
completes the integration in 9.2 seconds with the numeric result of 0.682689492136 which is
accurate to just 11 significant digits (last digit is in error).
To key in this example on the 115ES: [AC] [/] [] [SHIFT][ln]{e} [/] [RCL][)]{X}
[x2] [][(-)][2] [][][][(-)][1] [][1] [][] [][2] [SHIFT][x10x]{} [=].
To key in this example on the W516: [ON/C] [dx] [(-)][1] [][1] [] [2ndF][ln]{ex}
[RCL][RCL]{X} [x2] [] [a/b] [(-)][2] [][] [a/b] [2ndF][x2]{} [2][2ndF][3]{}[=].
In this example also, the 115ES finishes the integral five times faster than the W516
and returns a result which was more accurate.

75

Scientific Calculator Comparison Vol I: User Guide and Operational Comparison

Numerical Differentiation Examples

Differentiation is the process of calculating the slope of a smooth continuous function
f(x) for a specified value of x. It is the rate of change of f(x) at the specified value of x. No
special methods are required in the case of numerical differentiation. The function f(x) will be
evaluated at a small interval dx above and below the specified value of x. These are
subtracted and divided by 2dx to approximate the slope ( rise divided by run ) or the
differential of f(x) at x. Mathematically, as dx is taken to the limit of being infinitely small ,this
can be stated as:
f xdx f xdx
d
f x =
dx
2dx
The default value of dx on the W516 is 1x10-5 if the specified value of x is equal to
zero; otherwise dx is calculated to be |x| times 10-5. You may specify the value of dx in your
calculation on the W516. The 115ES also allows you to specify the tolerance of differential
calculations but, only in Line Mode. Consult your manual for details since Line Mode is not
covered here.
For a simple example of calculating the differential of a function manually, we'll use a
simple function, f(x)=x2 at x=2, with dx=0.001. We can then calculate the differential as:
2

20.001 20.001
d 2
4.0040013.996001
0.008
x =
=
=
= 4
dx
0.002
0.002
2 0.001
On the 115ES, when you press the differential key [SHIFT][]{d/dx}, the differential
template appears
d
()|x=
dx
with the cursor at the first position. Type in the function x2 from the above example. After that,
press the [] and type 2 for the specified value of x at the second template position. After
that, press [=] to evaluate the differential.
(X2)|X=2
4
The procedure for the W516 is identical but the display is just slightly different. The key
sequence for the above example is: [ON/C] [SHIFT][dx]{d/dx} [RCL][RCL]{X} [x2] [] [2]
[=].

76

Chapter 13. Summations, Integrals, & Derivatives

|X=2
4.
These next examples use another function, the natural logarithm function ln(x) at x=e,
where e is a mathematical constant equal to 2.718281828459 to thirteen significant digits.
The derivative of ex at x=e is equal to e-1 or 1/e which is 0.3678794411714 to thirteen
significant digits. This example will be performed on the W516 and will use different values for
the small interval dx to illustrate the effect of dx on the accuracy of the results. The red digits
in the results column of the below table are the incorrect digits.
W516 Numerical Differentiation with Varying dx
Numeric Results

|x=e , 0.1

0.36792093882

|x=e , 0.01

0.36787985606

|x=e , 0.001

0.36787944530

|x=e , 0.0001

0.36787944100

The examples in the above table, which use varied values for dx, work only on the
W516. The key sequence to enter these examples is: [ON/C] [SHIFT][dx]{d/dx} [ln][RCL]
[RCL]{X} [] [RCL][Exp]{e} [(x,y)]{,} tolerance [=]. The tolerance is entered as [.][0][0][1]
and so on, according to the example given.

Handy Tip for the W516!

There is a useful loophole in the W516's handling of formulas saved in the dedicated
function keys ( F1 to F4 ) that can save plenty of keystrokes when working with saved
formulas. To make use of the loophole, remember to always enclose your entire formula
within parentheses before storing it into a function memory. As an example, suppose you are
working with this polynomial ( note the enclosing parentheses ):

77

Scientific Calculator Comparison Vol I: User Guide and Operational Comparison

y =

2x 34x 23x5

and you'd like to store the function into function memory F1. Here are the keystrokes to
perform that operation. [ON/C] [(] [2][RCL][RCL]{X}[2ndF][RCL]{x3} [-] [4][RCL][RCL]{X}
[x2] [+] [3][RCL][RCL]{X} [-] [5] [)] [STO] []{F1}. Recall that [RCL][RCL]{X} is the same as
[ALPHA][RCL]{X} on the W516; using the [RCL] key instead of the [ALPHA] key is just a
little faster and a more convenient way to access the variable X.
This is the standard solution to the problem of saving a formula for repeated evaluation
with various values of X for graphing the function or some other purpose. Now suppose you
decide that you'd like to take the integral of this function from zero to one. Formulas recalled
from the function memories can not be inserted into an expression since, according to the
manual, 'Any recalled expressions will overwrite any expressions that are currently being
entered.' Well then, we'll just have to enter the function memory F1 contents first.
Here's how. Press [ON/C] to clear the display and press [RCL][]{F1} to recall
function F1 to the display. Notice that the polynomial is recalled, complete with enclosing
parentheses, and the cursor is blinking at the end of the formula. Now press [2ndF][] to
bring that blinking cursor to the very beginning of the formula, just before the opening
parentheses. Now you are free to press the [dx] to enter the numerical integration function...
but look where the recalled polynomial ended up. It wasn't cleared but inserted into the
numerical integration command:

2x 34x2 3x5

dx

All that's left to do is enter the limits of integration and press [=] to calculate the result.
This trick can be used in the exact same way with the numerical differentiation or summation.
Actually, it will work with just about any function on the W516. Some commands, such as
summation, will insert an extra set of parentheses around the expression. This is not a
problem and the command will execute normally. Parentheses are still required in your stored
formula since there is a chance that only part of the expression will be evaluated if you leave
them out. Your formula must consist of only one term and parentheses will ensure this
happens. There is a catch; you can perform this formula insertion trick only once per
expression. If you try it a second time in the same expression, the entire expression will be
cleared and replaced with your stored formula.

78

Chapter 14. Basic Matrix Calculations

14. Basic Matrix Calculations

The ability to perform calculations on matrices is just one of the advantages the 115ES
and W516 have over their competitors. For about US $20 ( current prices ), students can
quickly and easily perform basic matrix operations, gaining confidence and familiarity with the
many properties and applications of matrices. To get features like these from the competitors,
you would likely have to spend three times as much. Matrices and vectors are well worth
taking the time to learn since they are stepping stones between conventional algebra/calculus
and areas of advanced number theory and theoretical physics.
A matrix is a rectangular array of numbers, specified by dimension using a standard
mxn ( or rows x columns ) format. The individual elements may be any real numbers. Here is
a typical 2x3 matrix A, a row vector v, and a column vector u:

A = 7 3 4
0 1 5

v = [ 1 8 2 ]

u =

[]
3
0
1

A matrix with m=1 ( only one row ) or n=1 ( only one column ) is more properly called a
vector. In this way then, a matrix can be described as a list or a vector of vectors. Vectors can
be row vectors, where the elements are listed horizontally in a row. Vectors can also be
column vectors, where the elements are listed vertically in a column.
So one-dimensional matrices are vectors. In a majority of applications you will see the
use of typical two-dimensional matrices. Three-dimensional matrices are possible, as are four,
five, six, or any number of dimensions. Matrices of dimensions higher than two or three are
usually called tensors ( depending on their application ) but, a tensor is a generalized concept
that includes all vectors and matrices of any dimension. At the very root of tensors is the
concept of operations with, and properties of, n-dimensional vectors. The mathematics quickly
becomes very deep and very difficult theoretically for dimensions of three or more so tensor is
a term usually found employed in those cases. Even Einstein, when he made use of tensors
in his theory of general relativity, had a very difficult time learning to use the operations of
tensors correctly.

79

Scientific Calculator Comparison Vol I: User Guide and Operational Comparison

Matrix mode overview on the 115ES and W516

Matrix calculation is one area where the W516 has a clear advantage over the 115ES.
The 115ES is limited to matrices of dimensions up to 3x3 and has only three matrix
memories; MatA, MatB, and MatC. The W516 can perform operations on matrices up to
dimension 4x4 and there are four matrix memories available ( MatA MatD ). There are
many applications where 4x4 matrices are very much preferred, one of these being equations
in general relativity where calculations using matrices will deal with four dimensions ( three of
space and one of time ). The W516 is faster performing these operations, performs more
functions, and is able to decompose a matrix into into vectors or vice versa.
Both calculators have an additional matrix memory which is the result of the last matrix
operation or edit. This is the Matrix Answer memory. The 115ES can actually recall this matrix,
just as you would recall MatA-MatC, for use in an expression by selecting MatAns from the
matrix menu. The W516 is not able to do this; its Matrix Answer memory can only be recalled
to the editor and must be stored to one of the four matrix memories ( MatA-MatD) if you want
to use it in an expression.
Matrix calculations have to be performed in matrix mode. On both the 115ES and the
W516, you access matrix mode using the mode menu. On the 115ES, matrix mode is
[MODE][6], and on the W516 it is [MODE][4].
fx-115ES
1: COMP
3: STAT
5: EQN
7: TABLE

EL-W516 ( page 1 )

2: CMPLX
4: BASE-N
6: MATRIX
8: VECTOR

<MODE-1>
0: NORMAL 1: STAT
2: DRILL
3: CPLX
4: MATRIX
5:LIST

Key: [AC] [MODE][6]

Key: [ON/C] [MODE][4]

We'll compare the function of the matrix editors on the two calculators by taking a look
at the contents of the matrix menu. Once you are in matrix mode on the 115ES, you can
access the matrix menu by pressing: [AC] [SHIFT][4]{MATRIX}. On the W516, you should
press: [ON/C] [MATH].Take a look at the matrix menus below for both calculators.
fx-115ES
1: Dim
3: MatA
5: MatC
7: det

2: Data
4: MatB
6: MatAns
8: Trn

Key: [AC] [SHIFT][4]

{MATRIX}

EL-W516 ( page 1 )
<Math-1>
0: CTLG
1: MATRIX
2: EDIT
3: RECALL
4: STORE 5: OPE
Key: [ON/C] [MATH]

80

EL-W516 ( page 2 )
<Math-2>
6: CALC
7: matlist
8: matAlist
Key: [ON/C] [MATH][]

Chapter 14. Basic Matrix Calculations

The matrix editor on the 115ES is central to the matrix mode and is programmed to be
entered for a specific purpose, usually for editing one of the three matrix memories, and the
matrix menu is set up with this idea in mind. This approach ends up being very efficient in the
long run. Item 1:Dim in the matrix menu allows you to change the dimensions of one of the
matrix memories. Item 2:Data in the matrix menu allows direct access to the contents of a
selected matrix memory for editing of individual elements of that matrix. You are prompted to
select one of MatA-MatC for editing and, as you complete your edits, the edited matrix is
automatically updated in the matrix memory. You may also decide to copy the edited ( or
unedited ) matrix to another matrix memory by pressing [SHIFT][RCL]{STO}, followed by [(-)]
{MatA}, [']{MatB}, or [hyp]{MatC}. The editor follows to the new matrix memory and you
may continue edits from there. Items 3:MatA through 6:MatAns in the matrix menu are used
primarily for recalling the matrix memories ( MatA-MatC and MatAns ), as variables, to the
home screen for entry into expressions for the purpose of calculation.
The little things that the 115ES does for you automatically can not be done for you on
the W516 since the editor and matrix menu have not been programmed to anticipate what it is
you want to do. You have to tell it what you want to do every step of the way, even if it seems
obvious. This means that, in general, it takes more keystrokes to perform a matrix task on the
W516 than it does on the 115ES. The reason is that the the Matrix Answer memory is central
to the W516 matrix mode, instead of the matrix editor. Anything you want to do you have to do
via the Matrix Answer memory which means eventually exiting the matrix menu and then reentering the matrix menu to decide where to store the edited Matrix Answer memory.
Though the Matrix Answer memory is central to the W516 matrix mode, the W516 does
not allow you access to the Matrix Answer memory for use in an expression. You must go to
the matrix editor and store the Matrix Answer memory to one of the four matrix memories in
order to write an expression that includes the result from a previous matrix calculation.
Item 1:MATRIX on the W516 matrix menu is primarily for recalling one of the four
matrix memories ( matA-matD ), as variables, to an expression on the home screen for the
purpose of calculation. You will be prompted to select one of the four matrix memories. Item
2:EDIT takes you to the matrix editor where you may edit the contents or dimensions of the
Matrix Answer memory. Item 3:RECALL will recall the contents of one of the four matrix
memories to the editor. Any edits you perform however, are stored to the Matrix Answer
memory, not to the matrix memory itself. You have to exit the editor then return again to store
the contents of the Matrix Answer memory to one of the four matrix memories ( matA-matD ).
Item 4:STORE is provided specifically to store the contents of Matrix Answer memory to one
of the four matrix memories ( matA-matD ). Yes, it is very awkward.
Entering Matrix mode on the 115ES and storing a matrix
When you first enter Matrix mode from some other mode on the 115ES, it is assumed
that you will want to enter data into one of the three matrix memories, and so you are met with
a screen prompting you to choose a matrix memory to work with. Press [1]-[3] for matrix
memories MatA-MatC. We'll select MatA, so press [1]. You are then prompted for the
81

Scientific Calculator Comparison Vol I: User Guide and Operational Comparison

dimension of matrix memory A, using mxn format where m is the number of rows and n is the
number of colums. Find your choice in the dimension menu and press the corresponding
number key. There are two pages in the dimension menu so use the [] and [] keys to flip
back and forth between the two pages. We'll be entering the matrix from the introduction to
this chapter, and so we'll press [4] for a 2x3 matrix.

fx-115ES
Matrix?
1: MatA
3: MatC

2: MatB

Key: [1]

fx-115ES dim page 1

fx-115ES dim page 2

MatA(mxn)
1: 3x3
3: 3x1
5: 2x2

MatA(mxn) mxn?
1: 1x3
2: 1x2
3: 1x1

mxn?
2: 3x2
4:2x3
6:2x1

Key: [4]

After pressing [4] for a 2x3 matrix, you are taken to the editor screen where you are
presented with a blank ( all zeros ) 2x3 matrix template, and an A at the top left of the display
indicating that matrix memory MatA will be edited. Any time you enter the editor on the
115ES, there will be an indicator at the top of the screen informing you of exactly what it is
you are editing. You will see A for MatA, B for MatB, C for MatC, or Ans for MatAns, which is
the special matrix Answer Memory that is the result of the last operation or edit. Type in the
elements, one row after another, pressing [=] for each entry. The cursor will advance through
the matrix for you automatically. Key in the elements: [7][=] [(-)][3][=] [4][=] [0][=] [1][=] [(-)]
[5][=]. Again, the A at the top left of the display indicates that the data you are typing in is
being saved to matrix memory MatA. When you are done, press [AC] to exit the editor screen
( your data is already saved ) and return to the home screen.

fx-115ES
A

|
|

A = 7 3 4
0 1 5

7
0

-3
1

4|
-5 |
-5

Entering Matrix mode on the W516 and storing a matrix

To enter Matrix mode on the W516, press [ON/C][MODE][4]. When you first enter
82

Chapter 14. Basic Matrix Calculations

Matrix mode on the W516, nothing is assumed about what it is you want to do, so you simply
end up with the Matrix mode indicator at the top of the display. Much of the matrix functionality
is available through the math menu by pressing the [MATH] key. We need to access the
matrix editor so key in [MATH][2]{EDIT}. The default matrix dimensions are 2x2 but press [2]
[3] to change this to a 2x3 matrix. Press [] and [] to correct any mistakes, then [=] to
accept and the cursor moves down to the blank matrix template. Type in the matrix elements
from the example at the beginning of this appendix: [7][=] [(-)][3][=] [4][=] [0][=] [1][=] [(-)][5]
[=]. When you are finished, press [ON/C] to exit the editor.

EL-W516 ( page 1 )
<Math-1>
0: CTLG
1: MATRIX
2: EDIT
3: RECALL
4: STORE 5: OPE
Key: [ON/C] [MATH][2]

EL-W516
matrix : 2 x 2
|0
0
|0
0

EL-W516
|
|

Key: [2] [3] [=]

matrix : 2 x 3
|7
-3
|0
1

4
-5

|
|
-5.

Key: [ON/C]

Note that you have not yet stored the data; you have only edited the Matrix Answer
memory. You must return to the matrix menu to store the edited data to one of the four matrix
memories. Let's store the edited matrix data into matrix memory matA. To do this, enter the
matrix menu again by pressing [ON/C][MATH], then press [4] to select the matrix store menu.
Press the key corresponding to the matrix memory you want the data stored to. In this case,
press [0] to select matrix memory matA. You are alerted that the storage was successful and
the display returns to the home screen.

EL-W516 ( page 1 )
<Math-1>
0: CTLG
1: MATRIX
2: EDIT
3: RECALL
4: STORE 5: OPE
Key: [ON/C] [MATH][4]

EL-W516

EL-W516

<<STORE>>
0: matA
1: matB
2: matC
3: matC

STORED!

Key: [0]

Basic Matrix Operations: Scalar Multiplication

Scalar multiplication is the multiplication of a matrix by a real number we'll call k., which
we'll assign a value of three. We'll use the matrix A we stored earlier in this appendix. Scalar
multiplication will simply multiply each element in the matrix by the value of k.

83

Scientific Calculator Comparison Vol I: User Guide and Operational Comparison

A =

7 3 4
0 1 5

and k=3

kA =

21 9 12
0
3 15

On the 115ES, key in: [AC] [3] [X] [SHIFT][4]{MATRIX} [3]{MatA} [=].
On the W516, key in: [ON/C] [3] [X] [MATH][1]{MATRIX} [0]{matA} [=].
We'll repeat this operation with a new value for k, to illustrate one of the special
capabilities the 115ES has built into the matrix mode. This time k will be set equal to the
fraction 1/3. In this case we'll simply divide by 3, just to show that division of a matrix by a
scalar is a legal operation.

A =

7 3 4
0 1 5

and k=

1
3

kA =

[ ]
7
4
1
3
3
1
5
0

3
3

On the 115ES, key in: [AC] [SHIFT][4]{MATRIX} [3]{MatA} [] [3] [=].

On the W516, key in: [ON/C] [MATH][1]{MATRIX} [0]{matA} [] [3] [=].
Notice that both calculators will show the correct decimal equivalent of the above
solution matrix. Matrix results in the matrix display are not shown in fractional form but in their
decimal equivalent. On the 115ES though, we can navigate to the desired matrix element
using the four arrow keys [][][][] and press [SD]. This will transform the decimal
result of the current highlighted element in the matrix to fractional display format:
fx-115ES

EL-W516

Ans
2.3333
-1 1.3333

0 0.3333 -1.666

matrix : 2x3
2.33333
-1 1.33333

0 0.33333 -1.6666

213

Key: [SD]

2.333333333

Key: no equivalent

The only way to replicate this behavior on the W516 is to highlight an element and
press [ON/C] to exit to the home screen. If we now press [ALPHA][=]{ANS}, we recall this
matrix element, then pressing [CHANGE] will give us the fractional form of the result.
Unfortunately, a trip back into the matrix display and back out again, for each element of the
matrix, is required to see all the elements in fractional form.

84

Chapter 14. Basic Matrix Calculations

Basic Matrix Operations: Matrix Addition / Subtraction

The kind of matrix addition that the 115ES and the W516 performs is the simple
entrywise matrix sum. To perform this kind of matrix addition, both matrices must be of the
same dimension. The sum is a matrix, with the same dimensions as the summands, and each
element is the sum of the corresponding elements from each summand. Subtraction would be
defined the same way; corresponding elements are subtracted to obtain the result matrix.
Let's define a new matrix, matrix B, and store it into matrix memory MatB on both
calculators. Define matrix B as:
B =

2 0 6
8 1 9

On the 115ES, key in: [AC] [SHIFT][4]{MATRIX} [1]{Dim} [2]{MatB} [4]{2x3 matrix}
to set the dimension of MatB. It's not absolutely necessary to check the matrix dimension first
before you proceed to the editor because, once you've entered the editor and find there is
another matrix already stored there with the wrong dimension, you may still press [SHIFT][4]
{MATRIX} [1]{Dim} to change the dimension of the matrix. You will be brought right back to
the editor to fill in the now blank template. You may now enter the elements: [(-)][2][=] [0][=]
[6][=] [8][=] [(-)][1][=] [9][=]. Press [AC] to exit the editor; your data has been saved in MatB.
On the W516, key in: [ON/C] [MATH][2]{EDIT} [2][3]{2x3 matrix} [=] to set the
dimension of matB and enter the editor. You may now enter the elements of the matrix: [(-)]
[2][=] [0][=] [6][=] [8][=] [(-)][1][=] [9][=]. Press [ON/C] to exit the editor. To store the edited
matrix to matB, press: [MATH][4]{STORE} [1]{matB}.
By now you should be catching on to the flavor of matrix math on each of these
calculators. Each requires a unique line of thinking in order to edit, store, and work with
matrices. It's hard to say which is faster and easier; the 115ES or the W516. What can be said
is that you will quickly get used to the way your calculator works and the advantages or
disadvantages of one over the other does not seem all that big of a deal in the end.
Back to our problem. We now have matrices A and B stored on our calculators. We
want to find the sum of these two matrices: A+B.

A = 7 3 4
0 1 5

B = 2 0 6
8 1 9

AB = 5 3 10
8 0
4

On the 115ES, key in: [AC] [SHIFT][4]{MATRIX} [3]{MatA} [+] [SHIFT][4]{MATRIX}

[4]{MatB} [=].
On the W516, key in: [ON/C] [MATH] [1]{MATRIX} [0]{matA} [+] [MATH] [1]
{MATRIX} [1]{matB} [=].

85

Scientific Calculator Comparison Vol I: User Guide and Operational Comparison

Matrix subtraction, again entrywise, is very similar to the above example of matrix
addition. Try it for yourself, the answer you should get is:
AB =

9 3 2
8 2 14

Basic Matrix Operations: Matrix Multiplication

The multiplication of matrices is a little more difficult than simple addition or subtraction.
Not all matrices A and B can be multiplied together and, even if they can, it's not very often
found that AXB = BXA. Matrix multiplication does not commute. There are rules though, for
determining if two matrices can be multiplied together. If the matrices are both square, they
must be the same dimension. If the matrices have different dimensions, then the number of
rows in the first matrix must be the same as the number of columns in the second matrix.
In order to give an example of matrix multiplication, we'll enter a third matrix C into
MatC of our calculators. Matrix C is a 3X2 matrix, and is defined as:

[ ]

3
6
C = 4 1
5 2

Enter this matrix into MatC on your calculator. Now find the product (multiply) AXC and
the product CXA, and verify that your results agree with the following:

AC =

53 31
29 11

21 3 18
C A = 28 13 21
35 17 30

You will find that the dimension of the product matrix depends on the dimensions of the
multiplicand matrices. The product matrix will have the same number of rows as the first
matrix and the same number of columns as the second matrix.
Basic Matrix Operations: Determinants and Inverses
The inverse of a square matrix A, notated as A-1, is that matrix such that AA-1 = Im.,
where Im stands for identity matrix of dimension m. The identity is the matrix equivalent of a
one, such that, for a matrix of dimension mxn, ImA = AIn = A. Here are some examples of
identity matrices for m=2,3,4. Identity matrices are square matrices with all elements equal to
zero, except those elements along the main diagonal, which are all equal to one.

86

Chapter 14. Basic Matrix Calculations

I2 =

[ ]
1 0
0 1

[ ]

1
I4 = 0
0
0

[ ]

1 0 0
I3 = 0 1 0
0 0 1

0
1
0
0

0
0
1
0

0
0
0
1

With some technical exceptions, only square matrices are invertible, and then only if
the determinant of the matrix is non-zero. Any square matrix with a determinant equal to zero
is singular ( or degenerate ) and does not have an inverse. Both the 115ES and the W516 are
able to calculate matrix inverses and determinants. The W516 is also able to produce identity
matrices of dimensions 1 through 4. Here is an example of a square matrix, taken from the
Sample Matrices in Appendix E. Store this matrix into matrix memory MatA.

A =

3 2 1
3
1
0
5 1
1

Is this matrix invertible? We can easily calculate its determinant to find out.
On the 115ES, key in: [AC] [SHIFT][4]{MATRIX} [7]{det} [SHIFT][4]{MATRIX} [3]
{MatA} [=].
On the W516, key in: [ON/C] [MATH] [6]{CALC} [0]{det} [MATH] [1]{MATRIX} [0]
{matA} [=].
Of course we already knew that the determinant was equal to one since we took our
example from the Sample Matrices in Appendix E. Since our determinant is non-zero, matrix
A is non-singular and does have an inverse. Now calculate the matrix inverse A-1.
On the 115ES, key in: [AC] [SHIFT][4]{MATRIX} [3]{MatA} [x-1] [=].
On the W516, key in: [ON/C] [MATH] [1]{MATRIX} [0]{matA} [2ndF][2]{x-1} [=].
and we find that the inverse of matrix A, notated as A-1, is:

1
1
1
A1 = 3 2 3
8
7
9

Let's store this result into matrix memory MatB.

On the 115ES, key in: [SHIFT][RCL]{STO} [']{MatB}.

87

Scientific Calculator Comparison Vol I: User Guide and Operational Comparison

On the W516, key in: [ON/C] [MATH][4]{STORE} [1]{matB}.

Finally, verify that the product AB ( or BA ) is equal to the identity matrix I3.
On the 115ES, key in: [AC] [SHIFT][4]{MATRIX} [3]{MatA} [X] [SHIFT][4]{MATRIX}
[4]{MatB} [=].
On the W516, key in: [ON/C] [MATH] [1]{MATRIX} [0]{matA} [X] [MATH] [1]
{MATRIX} [1]{matB} [=].
Basic Matrix Operations: Matrix Transpose operation
The matrix transpose is a very simple yet mathematically useful concept. The
transpose of matrix A ( notated AT ) has all the same elements of the original matrix but the
columns have been transposed into rows and the rows have been transposed into columns.
The effect is to flip the matrix along the main diagonal, such that the elements along the main
diagonal remain in place.

3 2 1
A = 3
1
0
5 1
1

3 3 5
A = 2 1 1
1 0 1
T

On the 115ES, key in: [AC] [SHIFT][4]{MATRIX} [8]{Trn} [SHIFT][4]{MATRIX} [3]

{MatA} [=].
On the W516, key in: [ON/C] [MATH] [6]{CALC} [1]{trans} [MATH] [1]{MATRIX} [0]
{matA} [=].
Info: Orthogonal Matrices
Having introduced both matrix inverses and transposes, we'll go one step further to
mention the interesting things that begin to happen when matrix inverses and transposes are
compared. What if, for instance, a matrix A could be found such that its inverse A-1 were equal
to its transpose AT? If A-1=AT, then this would imply that A-1AT=Im where m is the dimension of
the square matrix. One immediate example of this would be the identity matrix:

[ ]

1 0 0
A = 0 1 0
0 0 1

[ ]

1 0 0
= 0 1 0
0 0 1

88

[ ]

1 0 0
A = 0 1 0
0 0 1
T

Chapter 14. Basic Matrix Calculations

Intuition may tell you that the determinant of an orthogonal matrix would be 1, and you
would be correct but, this property alone is not enough to determine whether or not a matrix is
orthogonal. In other words, not all matrices with a determinant equal to 1 have the property
such that AT=A-1. It would seem that a matrix would have to be symmetrical in order to be
orthogonal but, this is not always true. Consider the following matrices:

0 1 0
A = 1 0
0
0 0 1

A =

[ ] [ ]
2
2
1

3
3
3
1
2
2
3
3
3
2
1
2

3
3
3

A =

4
5
3
5

4
3

5
5
9
12

25
25
12
16

25
25

While it is true that the determinants of the above matrices are equal to 1, they
certainly aren't symmetric matrices. They are symmetric in another sense though. Each row
( or column ) in one of the above matrices is a vector. and all are mutually perpendicular, that
is, they form an orthogonal basis. Further, the length of each of these vectors is also equal to
1, so they also form an orthonormal basis. The dimension of all the matrices given above is
3x3. The reason for this was to demonstrate that, as different as they may appear at first
glance, they are all related. Each of the three matrices represent a set of three mutually
perpendicular vectors of length 1 and form an orthonormal basis. The three dimesions of the
matrix could represent three-dimensional space. This means that all three matrices are
identical but are rotated about a three-dimensional axis. In fact, they form a group known as a
compact Lie group. Such groups and symmetries have been widely studied and used in
number theory and other fields. The Pauli spin group is one example of the application of
these kinds of groups to theoretical physics.
Info: Powers of matrices
We've seen that some matrix operations are available by using the ordinary [+], [-], [X],
[(-)], and [x-1] keys. The 115ES is also able to take the absolute value (Abs) of a matrix, which
just returns the absolute value of all the elements in the matrix. There is also some limited
ability for both calculators to raise square matrices to integer powers. You may square or cube
a square matrix by using the [x2] or [x3] keys. With a little ingenuity ( and patience ) you can
achieve integer powers higher than three. If you square a matrix twice (matA2)2, you get
matA4. A matrix squared times a matrix cubed matA2XmatA3 gives you matA5. A matrix
cubed times a matrix cubed matA3XmatA3 gives you matA6. A matrix squared twice times a
matrix cubed (matA2)2XmatA3 gives you matA7. A matrix squared three times ((matA2)2)2
gives you matA8. There are various ways to achieve these powers but some patience is
required to type in the expressions and wait for the calculator to chug through the numbers to
produce a result.

89

Scientific Calculator Comparison Vol I: User Guide and Operational Comparison

Info: Advanced Matrix Operations on the W516

The W516 has a collection of advanced matrix operations that are accessed from the
matrix menu. These included the ability to produce special matrices such as identity matrices,
random matrices, and cumulative matrices. There is also a fill operation that will produce a
matrix of specified dimensions, filled with a specified real number. You can redimension any of
the matrix memories or take any two matrices and augment them to produce a single matrix
as a result. To access the advanced matrix functions, select item 5: OPE in the math menu.
The OPE label is an abbreviation for Operation. See the next chapter, Chapter 15, for more
detailed instructions on how to use the W516 advanced matrix operations.
The W516 is also capable of working with 4X4 matrices and so is capable of solving
linear systems of four equations with four unknowns. For some reason, this capability was not
built into the Equation Solver but, you can certainly work with such linear systems on your
own in Matrix mode. See Appendix G for more information and an example to get you started.

90

Chapter 15. Advanced Matrix Operations ( W516 only )

15. Advanced Matrix Operations ( W516 only )

This brief chapter will cover the extra matrix operations that are only available on the
W516. These advanced matrix operations can be found in the matrix menu. While in Matrix
mode, the matrix menu can be accessed by pressing [MATH][5]{OPE}.

EL-W516 ( page 1 )

EL-W516 ( page 2 )

<<OPERATION-1>>
0: dim(
1: fill(
2: cumul
3: aug(
4: identity

<<OPERATION-2>>
5: rnd_mat(

Key: [MATH][5]{OPE}

[MATH][5][0]

Key: [MATH][5]{OPE}[]

dim(

Format: dim(matrixname, #rows, #cols). This is a function that acts on the contents of a
stored matrix ( matA-matD ). The function returns the specified matrix with the dimensions
trimmed or expanded as necessary according to the #rows and #cols specified. The matrix
memory itself is not altered. The parameters of this function are separated by commas, which
is the [(x,y)] key. The #rows or #cols may not exceed four.
Example: dim(matA,2,3). [ON/C] [MATH][5]{OPE} [0]{dim(} [MATH][1]{MATRIX} [0]
{matA} [(x,y)] [2] [(x,y)] [3] [)] [=].

0
1 1
matA = 3 2 5
1 0 2

dim matA , 2,3 =

91

0
1 1
3 2 5

Scientific Calculator Comparison Vol I: User Guide and Operational Comparison

[ ]

0
1
3 2
dim matA , 4,2 =
1 0
0
0

0
1 1 0
3 2 5 0
dim matA , 4,4 =
1 0 2 0
0
0
0 0

If the specified dimensions are larger than the original matrix, then zeroes are padded
into the matrix. If the specified dimensions are smaller than the original matrix, then the
elements are trimmed from the original matrix. The specified matrix memory is not altered.
[MATH][5][1]

fill(

Format: fill(value, #rows, #cols). This is a function that returns a matrix of the specified
dimension (#rows x #cols), with each element equal to the specified value. The parameters of
this function are separated by commas, which is the [(x,y)] key. The #rows or #cols may not
exceed four.
Example: fill(7,2,4). [ON/C] [MATH][5]{OPE} [1]{fill(} [7] [(x,y)] [2] [(x,y)] [4] [)] [=].
fill 7,2,4 =

7 7 7 7
7 7 7 7

This function could be useful for creating a matrix that happened to have many of its
elements equal to a certain value. The result of this function could be stored to a matrix
memory ( matA-matD ) and then edited for the few values that differ from that specified.
[MATH][5][2]

cumul

Format: cumul matrixname. This is a function that returns a specified matrix with the
elements transformed to a matrix of columnwise running sums.
Example: cumul matA. [ON/C] [MATH][5]{OPE} [2]{cumul} [MATH][1]{MATRIX} [0]
{matA} [=].

matA =

3 2 1
3
1
0
5 1
1

3 2 1
cumul matA = 6 1 1
1 0
0

First column: 3=3, 3+3=6, 3+3-5=1. Second column: -2=-2, -2+1=-1, -2+1+1=0. Third
column: -1=-1, -1+0=-1, -1+0+1=0. The specified matrix memory is not altered.

92

Chapter 15. Advanced Matrix Operations ( W516 only )

[MATH][5][3]

aug(

Format: aug(matrixname1,matrixname2). This function appends the second matrix to

the first matrix as new columns. The two matrix names must be separated by a comma, which
is the [(x,y)] key. matrixname1 and matrixname2 may be the same name.
Example: aug(matA,matB). [ON/C] [MATH][5]{OPE} [3]{aug(} [MATH][1]{MATRIX}
[0]{matA} [(x,y)] [MATH][1]{MATRIX} [1]{matB} [=].
matA =

[ ]
1 2
3 4

matB =

[ ]
5 6
7 8

aug matA , matB =

1 2 5 6
3 4 7 8

Both matrix memories must have the same number of rows. The matrix memories do
not need to have the same number of columns but, the total number of columns in the
resulting augmented matrix must not exceed four or an error will occur.

93

Chapter 16. Vector Operations

16. Vector Operations

Both the 115ES and the W516 have a Vector mode of operation. The 115ES is capable
of operations on vectors of two or three dimensions and has three vector memories while the
W516 can handle vectors from two to four dimensions and has four vector memories. The
Vector mode on the W516 also doubles as List mode, something not found on the 115ES, but
this chapter will only discuss the Vector operations. The list operations on the W516 are
covered in Chapter 18.
Vectors can be thought of as geometric objects that extend the definition of a quantity
by posessing both a magnitude and a direction. The magnitude or length of a vector a is
sometimes referred to as the norm, indicated by |a|, which refers to the similar operation of
the absolute value on a scalar quantity. If represented in rectangular ( Cartesian ) form, a
vector is that set of coordinates equal to the coordinates of the endpoint minus the
coordinates of the initial point, with the order of this subtraction of coordinates defining the
direction. Since the initial point is usually taken to be the origin of the system of coordinates,
the vector is then more conveniently defined by the coordinates of the endpoint, and so the
vector a can be represented as a=[a1,a2,a3]. In this case then, the norm of the vector a is
defined by:

a =

2
1

a 22a 32

where a1, a2, and a3 are the rectangular coordinates of the vector a. In practive, vectors can
be represented in rectangular or polar coordinates, but since the 115ES and the W516 use
rectangular coordinates to represent vectors in memory, that is the representation that will be
used here.
Vector mode overview on the 115ES and W516
Vector operations on the 115ES are fairly simple, with all functionality stored in just one
menu in Vector mode. On the W516, Vector operations are shared with List operations in List
mode so that all the functions for lists and vectors are found in one set of menus. This can be
a little confusing until you get used to it. The 115ES is limited to two or three dimensional

94

Scientific Calculator Comparison Vol I: User Guide and Operational Comparison

vectors and has only three three memories; VctA, VctB, and VctC. The W516 can work with
two, three, or four dimensional vectors, and there are four vector memories available ( L1
L4 ). Both calculators have an additional vector memory which is the result of the last vector
operation or edit. This is the Vector Answer memory. The 115ES can actually recall this
vector, just as you would recall VctA-VctC, for use in an expression by selecting VctAns from
the vector menu. The W516 is not able to do this; its Vector Answer memory can only be
recalled to the editor and must be stored to one of the four vector memories ( L1 - L4) if you
want to use it in an expression.
Vector calculations have to be performed in Vector mode. On both the 115ES and the
W516, you access Vector mode using the mode menu. On the 115ES, Vector mode is
accessed by pressing [AC][MODE][8]{VECTOR}, and on the W516 it is [ON/C][MODE][5]
{LIST}. Note that, on the W516, this is called List Mode, which is the W516's dual purpose
List/Vector mode. Even though this is the case, I will refer to this mode of operation as vector
mode while covering vector operations, and as list mode when discussing list operations.

fx-115ES
1: COMP
3: STAT
5: EQN
7: TABLE

EL-W516

2: CMPLX
4: BASE-N
6: MATRIX
8: VECTOR

<MODE-1>
0: NORMAL 1: STAT
2: DRILL
3: CPLX
4: MATRIX
5:LIST

Key: [AC][MODE][8]{VECTOR}

Key: [ON/C][MODE][5]{LIST}

We'll compare the function of the vector editors on the two calculators by taking a look
at the contents of the vector menu. Once you are in vector mode on the 115ES, you can
access the vector menu by pressing: [AC] [SHIFT][8]{VECTOR}. On the W516, you should
press: [ON/C] [MATH].Take a look at the vector menus below for both calculators.

fx-115ES
1: Dim
3: VctA
5: VcC
7: Dot

2: Data
4: VctB
6: VctAns

Key: [AC] [SHIFT][5]

{VECTOR}

EL-W516 ( page 1 )
<Math-1>
0: CTLG
1: LIST
2: EDIT
3: RECALL
4: STORE 5: OPE
Key: [ON/C] [MATH]

EL-W516 ( page 2 )
<Math-2>
6: CALC
7: listmat
8: listmatA
Key: [ON/C] [MATH][]

The vector editor on the 115ES is central to the Vector mode and is programmed to be
entered for a specific purpose, usually for editing one of the three vector memories, and the

96

Chapter 16. Vector Operations

vector menu is set up with this idea in mind. This approach ends up being very efficient in the
long run. Item 1:Dim in the vector menu allows you to change the dimensions of one of the
vector memories. Item 2:Data in the vector menu allows direct access to the contents of a
selected vector memory for editing of individual elements of that vector. You are prompted to
select one of VctA-VctC for editing and, as you complete your edits, the edited vector is
automatically updated in the vector memory. You may also decide to copy the edited ( or
unedited ) vector to another vector memory by pressing [SHIFT][RCL]{STO}, followed by [(-)]
{VctA}, [']{VctB}, or [hyp]{VctC}. The editor follows to the new vector memory and you
may continue edits from there. Items 3:VctA through 6:VctAns in the vector menu are used
primarily for recalling the vector memories ( VctA-VctC and VctAns ), as variables, to the
home screen for entry into expressions for the purpose of calculation.
The Vector Answer memory is central to the W516 vector mode, instead of the vector
editor. Anything you want to do you have to do via the Vector Answer memory which means
eventually exiting the vector menu and then re-entering the vector menu to decide where to
store the edited Vector Answer memory.
Though the Vector Answer memory is central to the W516 vector mode, the W516
does not allow you access to the Vector Answer memory for use in an expression. You must
go to the vector editor and store the Vector Answer memory to one of the four vector
memories in order to write an expression that includes the result from a previous vector
calculation.
Item 1:LIST on the W516 vector menu is primarily for recalling one of the four vector
memories ( L1 - L4 ), as variables, to an expression on the home screen for the purpose of
calculation. You will be prompted to select one of the four vector memories. Item 2:EDIT takes
you to the vector editor where you may edit the contents or dimensions of the Vector Answer
memory. Item 3:RECALL will recall the contents of one of the four vector memories to the
editor. Any edits you perform however, are stored to the Vector Answer memory, not to the
vector memory itself. You have to exit the editor then return again to store the contents of the
Vector Answer memory to one of the four vector memories ( L1 - L4 ). Item 4:STORE is
provided specifically to store the contents of Vector Answer memory to one of the four vector
memories ( L1 - L4 ). Just as in matrix mode... yes, it is very awkward.
Entering Vector mode on the 115ES and storing a vector
When you first enter Vector mode from some other mode on the 115ES, it is assumed
that you will want to enter data into one of the three vector memories, and so you are met with
a screen prompting you to choose a vector memory to work with. Press [1]-[3] for vector
memories VctA-VctC. We'll select VctA, so press [1]. You are then prompted for the
dimension of vector memory A, using m as the number of elements. Choose either two or
three dimensions and press the corresponding number key. In this case, press [1] for a three
dimensional vector.

97

Scientific Calculator Comparison Vol I: User Guide and Operational Comparison

fx-115ES
Vector?
1: VctA
3: VctC

2: VctB

fx-115ES dim page 1

VctA (m) m?
1: 3
2: 2

fx-115ES dim page 2

A
|

0|
0

Key: [1]

Key: [1]

Key:

After pressing [1] for a three dimensional vector, you are taken to the editor screen
where you are presented with a blank ( all zeros ) vector template, and an A at the top left of
the display indicating that vector memory VctA will be edited. Any time you enter the editor on
the 115ES, there will be an indicator at the top of the screen informing you of exactly what it is
you are editing. You will see A for VctA, B for VctB, C for VctC, or Ans for VctAns, which is
the special Vector Answer memory that is the result of the last operation or edit. Type in the
elements, one after another, pressing [=] for each entry. The cursor will advance through the
vector for you automatically. When you are done, press [AC] to exit the editor screen ( your
data is already saved ) and return to the home screen.
Entering Vector mode on the W516 and storing a vector
To enter Vector mode on the W516, press [ON/C][MODE][5]. When you first enter
Vector mode on the W516, nothing is assumed about what it is you want to do, so you simply
end up with the Vector mode indicator at the top of the display. Much of the vector
functionality is available through the math menu by pressing the [MATH] key. We need to
access the vector editor so key in [MATH][2]{EDIT}. The default vector dimension is 1 but
press [3] to change this to a 3 dimensional vector. Press [=] to accept and the cursor moves
down to the blank vector template. Type in the elements, one after another, pressing [=] for
each entry. The cursor will advance through the vector for you automatically. When you are
finished, press [ON/C] to exit the editor.

EL-W516 ( page 1 )
<Math-1>
0: CTLG
1: LIST
2: EDIT
3: RECALL
4: STORE 5: OPE
Key: [ON/C] [MATH][2]

EL-W516
LIST SIZE : 1
| 1:0
|

Key: [3] [=]

EL-W516
LIST SIZE : 3
| 1:0
2:0
| 3:0

Key: [ON/C]

98

|
|
0.

Chapter 16. Vector Operations

Note that you have not yet stored the data; you have only edited the Vector Answer
memory. You must return to the vector menu to store the edited data to one of the four vector
memories. Let's store the edited vector data into vector memory L1. To do this, enter the
vector menu again by pressing [ON/C][MATH], then press [4] to select the vector store menu.
Press the key corresponding to the vector memory you want the data stored to. In this case,
press [0] to select vector memory L1. You are alerted that the storage was successful and the
display returns to the home screen.

EL-W516 ( page 1 )
<Math-1>
0: CTLG
1: LIST
2: EDIT
3: RECALL
4: STORE 5: OPE
Key: [ON/C] [MATH][4]

EL-W516
<<STORE>>
0: L1
1: L2
2: L3
3: L4

EL-W516
STORED!

Key: [0]

For the rest of this chapter, I will use the following notation with regard to vectors. I will
use bold italic lower-case letters for vectors, such as a, b, or c. The rectangular components
of the vectors will be enclosed in brackets, such as a=[a1,a2,a3].

99

Scientific Calculator Comparison Vol I: User Guide and Operational Comparison

Basic Vector Operations: Vector Addition / Subtraction

Vector addition, geometrically, is

performed by arranging the vectors, head to
tail, with the vector sum found to be the vector
drawn from the initial point of the first vector to
the endpoint of the last vector. This is true
whether you are adding just two vectors or a
hundred vectors. When you look at the vector
components, the i'th component of the vector
sum is the sum of the i'th components of all
the summand vectors.

Here are the example vectors we will be working with. On the 115ES, store the vectors
a, b, and c in vector memories VctA-VctC. One the W516, store them into vector memories
L1-L3:
a = [4 ,0,0]

b = [ 0 , 3 ,0 ]

c = [ 0 , 0 , 2 ]

To get you started, here are the key sequences, for both calculators, for storing vector
a into a vector memory:
On the 115ES, press: [AC][SHIFT][5]{VECTOR} [1]{Dim} [1]{VctA} [3]{Dim=3} [4][=]
[0][=] [0][=] [AC].
On the W516, press: [ON/C] [MATH][2]{EDIT} [3][=]{Dim=3} [4][=] [0][=] [0][=]
[ON/C] [MATH][4]{STORE} [0]{L1}.
Adding vectors is pretty simple on either calculator; recall the vectors and use the [+]
key to add them. Here are the key sequences for adding vectors a and b:
ON the 115ES, press: [AC][SHIFT][5]{VECTOR} [3]{VctA} [+] [SHIFT][5]{VECTOR}
[3]{VctA} [=]
On the W516, press: [ON/C] [MATH][1][0]{L1} [+] [MATH][1][1]{L2} [=] [ON/C]
[MATH][4]{STORE} [3]{L4} [MATH][1][3]{L4} [=].
Verify that the answer you get is:

100

Chapter 16. Vector Operations

a b = [ 4 , 3 , 0 ]

For the W516 key sequence, the vector sum was also stored into vector memory L4.
This was done so that vector subtraction could be demonstrated using this sum. This step is
not necessary on the 115ES since we have access to VctAns as a variable.
From the result of the last calculation, subtract vector b. The result should be the value
of vector a. Here are the key sequences:
ON the 115ES, press: [AC][SHIFT][6]{VctAns} [-] [SHIFT][5]{VECTOR} [4]{VctB}
[=].
On the W516, press: [ON/C] [MATH][1][3]{L4} [-] [MATH][1][1]{L2} [=].
The norm of vector a, notated by |a|, is equal to 4. The norm of vector b, notated by |b|,
is equal to 3. What is the norm of a+b? On the 115ES, we simply use [SHIFT][hyp]{Abs} for
this function. On the W516, the norm function is hidden away: [MATH][6][CNST]{abs_list}.
The key that is usually used for abs ( [2ndF][(-)]{abs} ) on the W516 keyboard will not work
for calculating the norm of a vector. Calculate the norm of a+b. Note that we have to
recalculate the sum on the 115ES but, on the W516, we have the sum stored in vector
memory L4:
ON the 115ES, press: [AC][SHIFT][hyp]{Abs} [(] [SHIFT][5][3]{VctA} [+] [SHIFT][5]
[4]{VctB} [=].
On the W516, press: [ON/C] [MATH][6][CNST]{abs_list} [MATH][1][3]{L4} [=].
The result you should get is that |a+b|=5.
Basic Vector Operations: Scalar Multiplication
A scalar quantity, for our purposes, is just a real number. When you multiply a vector by
a scalar, the result is a vector. The vector's direction is not changed, but the magnitude is
multiplied by a factor equal to the scalar quantity. Multiply vector sum a+b by 2 and you'll find
that the norm of the resulting vector has doubled to 10. You'll also find that this happens
because each component of the vector is multiplied by the scalar:
ON the 115ES, press: [AC] [2] [X] [(] [SHIFT][5][3]{VctA} [+] [SHIFT][5][4]{VctB} [=].
[AC] [SHIFT][hyp]{Abs} [SHIFT][5][6]{VctAns} [=].
On the W516, press: [ON/C] [2] [X] [MATH][1][3]{L4} [=].
[ON/C] [MATH][6][CNST]{abs_list} [(] [2] [X] [MATH][1][3]{L4} [)] [=].

101

Scientific Calculator Comparison Vol I: User Guide and Operational Comparison

Basic Vector Operations: Dot ( Scalar ) Product

The dot product of two vectors results in
a scalar quantity, which is why the dot product
is often called the scalar product. You may
also see it referred to as the inner product
since the concept of the dot product, in more
generalized applications, is known as the inner
product.
The dot product of two vectors a and b
is equal to the magnitude of vector a, times the
magnitude of the projection of vector b onto a,
or vice versa. The dot product commutes.
The dot product of a vector with itself
results in |a|2. The dot product of two vectors
that are perpendicular to each other is zero.
If you define the dot product using the
components of the two vectors a=[a1,a2,a3] and b=[b1,b2,b3], then the dot product is defined as
the sum of the products of the respective components, or:
ab = a1 b1a 2 b 2a3 b3 . . .a n b n

Here are two example vectors, a and b, which should be stored into vector memories
VctA and VctB on the 115ES, or into vector memories L1 and L2 on the W516:

b = [6 , 5 , 4 ]

a = [1 , 2 , 3]

The dot product on the 115ES is available as item 7:Dot in the vector menu. On the
W516, it is item 9:i_prod( within item 6:CALC of the MATH menu ( not quite as handy ). Now
we can calculate the dot product of vectors a and b:
ON the 115ES, press: [AC] [SHIFT][5][3]{VctA} [SHIFT][5][7]{Dot} [SHIFT][5][4]
{VctB} [=].
On the W516, press: [ON/C] [MATH][6][9]{i_prod(} [MATH][1][0]{L1} [(x,y)]{,}
[MATH][1][1]{L2} [)] [=].
The result should be a real number, in this case an integer, 28. You can verify this with
the components of the two vectors: a dot b = (1)(6)+(2)(5)+(3)(4) = 6+10+12 = 28. You can
also verify that the dot product commutes by calculating the result of b dot a.

102

Chapter 16. Vector Operations

Basic Vector Operations: Cross ( Vector ) Product

The cross product of two vectors results
in a vector quantity, which is why the cross
product is often called the vector product. You
may also see it referred to as the outer or
exterior product since the concept of the cross
product, in more generalized applications, is
known as the outer product.
The cross product of two vectors a and
b results in a vector with a magnitude equal to
the area of a parallelogram spanned by the
two vectors. The cross product, perpendicular
to both vectors a and b, points in the direction
defined by the right hand rule; the fingers of
the right hand curl from vector a to b, and the
extended thumb points in the direction of the
cross product.
If you define the cross product using the components of the two vectors a=[a1,a2,a3]
and b=[b1,b2,b3], then the dot product is defined as:
a b = [ a 2 b3a 3 b 2 , a 3 b1a 1 b3 , a 1 b2a 2 b1 ]

We'll use the same two example vectors, a and b, which should be stored into vector
memories VctA and VctB on the 115ES, or into vector memories L1 and L2 on the W516:

b = [6 , 5 , 4 ]

a = [1 , 2 , 3]

The cross product on the 115ES is simply the [X] key on the keypad. On the W516, it is
item 8:o_prod( within item 6:CALC of the MATH menu ( again not quite as handy ). Now we
can calculate the cross product of vectors a and b:
ON the 115ES, press: [AC] [SHIFT][5][3]{VctA} [X] [SHIFT][5][4]{VctB} [=].
On the W516, press: [ON/C] [MATH][6][8]{o_prod(} [MATH][1][0]{L1} [(x,y)]{,}
[MATH][1][1]{L2} [)] [=].
The result should be the three dimensional vector [-7,14,-7]. You can verify this with the
components of the two vectors: a cross b = [ (2)(4)-(3)(5) , (3)(6)-(1)(4) , (1)(5)-(2)(6) ] = [ (8)(15) , (18)-(4) , (5)-(12) ] = [ -7 , 14 , -7 ].

103

Scientific Calculator Comparison Vol I: User Guide and Operational Comparison

[MATH][5][4]

identity

Format: identity value. This function returns the square identity matrix with dimension
equal to value. The dimension value may not exceed four.
Example: identity 3. [ON/C] [MATH][5]{OPE} [4]{identity} [3] [=].

[ ]

1 0 0
identity 3 = I 3 = 0 1 0
0 0 1

The identity matrix is a square matrix with all elements equal to zero, except for those
elements along the main diagonal, which are equal to one.
[MATH][5][5]

rnd_mat

Format: rnd_mat(#rows,#cols). This function returns a matrix with the specified #rows
and #cols, with all elements equal to a random decimal number between zero and one,
excluding one [0.000-0.999]. The #rows or #cols may not exceed four.
Example: rnd_mat(2,3). [ON/C] [MATH][5]{OPE} [5]{rnd_mat(} [2][(x,y)][3] [=].
rnd mat 2,3 =

0.619 0.937 0.05

0.436 0.972 0.163

Since the elements of the matrix resulting from the rnd_mat function are random, your
results will most likely not agree with those shown above.
Handy Tips
You can create an augmented matrix by appending the second matrix B to the first
matrix A as new rows instead of as new columns by using transpose operations. Transpose
each matrix, perform the augment, then transpose the result. The resulting operation on the
W516 would look like: trans aug( trans matA , trans matB ). The number of rows in the
augmented matrix may not exceed four or an error will occur.
You can create random matrices with ranges of anything you like by multiplying by a
constant R and adding an offset constant K by using a fill operation. If you want random
numbers from 5 to 14, for example, set R=10 and K= 3x3 matrix filled with 5's. An example
calculation could be: fill(5,3,3)+10 X rnd_mat(3,3). The results will have to be truncated by
hand. Store this expression to a function memory to execute it multiple times.

104

Chapter 17. Matrix / Vector Conversions ( W516 only )

17. Matrix / Vector Conversions ( W516 only )

One of the more surprising features of the W516 Matrix mode is the ability to convert a
matrix into a collection of vectors or lists. The W516 is also capable of performing the reverse
operation; converting a a collection of lists or vectors into a matrix. This can be accomplished
in one of two different ways, so there are four functions for converting back and forth between
matrices and vectors; two functions in Matrix mode and two in Vector mode.
I will describe the converted matrix as a collection of 'vectors' since this makes more
sense mathematically, given that a matrix actually is considered to be a list of vectors. You
can consider a matrix as a list of row vectors or as a list of column vectors. There are uses for
either viewpoint mathematically. When working with the W516 though, remember that the
Matrix / Vector conversions will operate with column vectors. If you wish to work with row
vectors instead, you have the matrix traspose function at your disposal to enable the
conversion to operate with row vectors.

Parallel MatrixVector Conversion: [MATH][7]{matlist}

Matrix Mode. This conversion works in parallel; taking the left column of each of the
matrix memories and storing them to separate vector memories. matAL1, matBL2,
matCL3, and matDL4. The mode is then switched to Vector Mode.

[ ]

matA = 1 2
3 4

[ ]

matB = 5 6
7 8

matC =

9 10
11 12

matD = 13 14
15 16

Matrix mode: [MATH][7]{matlist}

L1 = [ 1 3 ]

L2 = [ 5 7 ]

L3 = [ 9 11 ]

105

L4 = [ 13 15 ]

Scientific Calculator Comparison Vol I: User Guide and Operational Comparison

Parallel VectorMatrix Conversion: [MATH][7]{listmat}

Vector/List Mode. This conversion works in parallel; taking each of the vector memories
and storing them into separate matrix memories as column matrices. L1matA, L2matB,
L3matC, and L4matD. The mode is then switched to Matrix Mode.
The maximum dimension of a matrix is 4X4, so the maximum size of a column vector
in this matrix is also 4. If a list is longer than four items, the conversion still works but the list is
truncated to four items.
L1 = [ 1 3 ]

L2 = [ 5 7 ]

L3 = [ 9 11 ]

L4 = [ 13 15 ]

Vector mode: [MATH][7]{listmat}

matA =

[]
1
3

matB =

[]
5
7

matC =

[ ]
9
11

matD =

[ ]
13
15

Point MatrixVector Conversion: [MATH][8]{matAlist}

Matrix Mode. This conversion works from a single matrix; matrix A. Each column vector
of matrix memory matA is stored to separate vector memories. matA(col#1)L1,
matA(col#2)L2, matA(col#3)L3, and matA(col#4)L4. The mode is then switched to
Vector Mode.

[ ]

1 2 3
A = 4 5 6
7 8 9
L1 = [ 1 2 3 ]

Matrix mode: [MATH][8]{matAlist}

L2 = [ 2 5 8 ]

L3 = [ 3 6 9 ]

L4 = undefined

Point VectorMatrix Conversion: [MATH][8]{listmatA}

Vector Mode. This conversion creates a single matrix; matrix A. Each vector memory is
stored to a separate column of matrix memory matA. L1matA(col#1), L2matA(col#2),
L3matA(col#3), and L4matA(col#4). The mode is then switched to Matrix Mode.
L1 = [ 1 2 3 ]

L2 = [ 2 5 8 ]

L3 = [ 3 6 9 ]

106

L4 = undefined

Chapter 17. Matrix / Vector Conversions ( W516 only )

Vector mode: [MATH][8]{listmatA}

[ ]

1 2 3
A = 4 5 6
7 8 9

These last two examples show that it is not necessary to have all vectors defined in
order for a VectorMatrix conversion to work. The only real limit is the number of elements in
a vector to be converted to a matrix column vector. Any elements in a vector memory beyond
four will be truncated since matrices are limited in dimension to 4X4.

107

Chapter 18. List Operations ( W516 only )

18. List Operations ( W516 only )

The W516 has a dual mode for vector and list calculations, and it is formally described
as List mode in the W516 documentation. Since the 115ES does not have list capabilities, this
chapter will cover only the List mode capabilities of the W516; the vector capabilities of the
115ES and W516 are covered in Chapter 16.
It wouldn't seem that list processing capability would be required very often on a
scientific calculator. List operations the W516 are fairly simple and consist of sorting, taking
sums, differences, or products, or providing the basic statistics such as mean, median,
variance, etc. The amount of data that would require electronic assistance for processing, I
would think, be very tedious to key into a calculator. That's what spreadsheets on home
computers are for, right? At any rate, when you need it, list processing is at your fingertips on
the W516, and this mode is very easy to use. For those who's work tends to lean more
towards data processing , List mode on the W516 might be useful. Take note also that List
processing is a capability usually only found on the more advanced ( and expensive )
graphing calculators.
Being a fairly simple mode, it should be sufficient to list the commands and operations
available in List mode, along with their required syntax. First, to enter List mode on the W516,
press [ON/C][MODE][5]{LIST}. Most all of the commands described will be found in either
the operations (5:OPE) menu or the calculation (6:CALC) menu of the math menu.

EL-W516
<MODE-1>
0: NORMAL 1: STAT
2: DRILL
3: CPLX
4: MATRIX
5:LIST
Key: [ON/C][MODE][5]
{LIST}

EL-W516 ( page 1 )
<Math-1>
0: CTLG
1: LIST
2: EDIT
3: RECALL
4: STORE 5: OPE
Key: [ON/C] [MATH]

EL-W516 ( page 2 )
<Math-2>
6: CALC
7: listmat
8: listmatA
Key: [ON/C] [MATH][]

Besides the operations and commands found in the math menu, lists may be added,

109

Scientific Calculator Comparison Vol I: User Guide and Operational Comparison

subtracted, multiplied, or divided using the usual keys on the keypad. These operations are
performed element-wise so the lists must be the same length or an error will occur. Lists may
also be squared [x2] or cubed [2ndF][RCL]{x3} or inverted [2ndF][2]{x-1}. Lists may be a
maximum of 16 elements long. The editing, recalling, and storing of lists are performed in
exactly the same manner as in Vector mode. You have four list memories available for use;
L1, L2, L3, and L4. With only 16 elements allowed per list, this is a grand total of 64 elements
capable of being stored in list memory.

List Operations Menu

EL-W516 List OPE ( page 1 )

EL-W516 List OPE ( page 2 )

<< OPERATION - 1 >>

0: sortA
1: sortD
2: dim(
3: fill(
4: cumul

<< OPERATION - 2 >>

5: df_list
6: aug(

Key: [ON/C][MATH][5]{OPE}

[MATH][5][0]

Key: [ON/C][MATH][5][]

sortA

Format: sortA listname. This is a function that returns the stored list ( L1 - L4 ) with the
elements sorted into ascending order. The list memory itself is not altered.
Example: sortA L1. [ON/C] [MATH][5]{OPE} [0]{sortA}[MATH][1]{LIST}[0]{L1}.
L1 = [ 5 1 3 0 6 1 ]

[MATH][5][1]

sortA L1 = [ 1 0 1 3 5 6 ]

sortD

Format: sortD listname. This is a function that returns the stored list ( L1 - L4 ) with the
elements sorted into descending order. The list memory itself is not altered.
Example: sortD L1. [ON/C] [MATH][5]{OPE} [1]{sortD}[MATH][1]{LIST}[0]{L1}.
L1 = [ 5 1 3 0 6 1 ]

sortD L1 = [ 6 5 3 1 0 1 ]

110

Chapter 18. List Operations ( W516 only )

[MATH][5][2]

dim(

Format: dim( listname, #size ). This is a function that acts on the contents of a stored
list ( L1 - L4 ). The function returns the specified list with the size trimmed or expanded as
necessary according to the #size specified. The list memory itself is not altered. The
parameters of this function are separated by a comma, which is the [(x,y)] key. The #size may
not exceed sixteen.
Example: dim(L1,3). [ON/C] [MATH][5]{OPE} [2]{dim(} [MATH][1]{LIST} [0]{L1}
[(x,y)] [3] [)] [=].
L1 = [ 5 1 3 0 6 1 ]

dim L1 , 3 = [ 5 1 3 ]

dim L1 ,10 = [ 5 1 3 0 6 1 0 0 0 0 ]

[MATH][5][3]

fill(

Format: fill(value, #size). This is a function that returns a list of the specified size (
#size ), with each element equal to the specified value. The parameters of this function are
separated by a comma, which is the [(x,y)] key. The list #size may not exceed sixteen.
Example: fill(3,5). [ON/C] [MATH][5]{OPE} [3]{fill(} [5] [(x,y)] [7] [)] [=].
fill 3,5 = [ 3 3 3 3 3 ]

[MATH][5][4]

cumul

Format: cumul listname. This is a function that returns a specified list with the elements
transformed to a running sum.
Example: cumul L1. [ON/C] [MATH][5]{OPE} [4]{cumul} [MATH][1]{LIST} [0]{L1} [=].
L1 = [ 5 1 3 0 6 1 ]

cumul L1 = [ 5 4 7 7 13 14 ]

The cumulative elements are: 5=5, 5-1=4, 5-1+3=7, 5-1+3+0=7, 5-1+3+0+6=13, 51+3+0+6+1=14. The specified list memory is not altered.

111

Scientific Calculator Comparison Vol I: User Guide and Operational Comparison

[MATH][5][5]

df_list

Format: df_list listname. This is a function that returns a specified list with the elements
transformed to the difference between the elements.
Example: df_list L1. [ON/C] [MATH][5]{OPE} [5]{df_list} [MATH][1]{LIST} [0]{L1} [=].
L1 = [ 5 1 3 0 6 1 ]

cumul L1 = [ 6 4 3 6 5 ]

The element differences are: -1-5=-6, 3-(-1)=4, 0-3=-3, 6-0=6, 1-6=-5. The specified list
memory is not altered.
[MATH][5][6]

aug(

Format: aug(listname1,listname2). This function appends the second list to the first list.
The two list names must be separated by a comma, which is the [(x,y)] key. listname1 and
listname2 may be the same name.
Example: aug(L1,L2). [ON/C] [MATH][5]{OPE} [6]{aug(} [MATH][1]{LIST} [0]{L1}
[(x,y)] [MATH][1]{LIST} [1]{L2} [=].
L1 = [ 1 2 3 ]

L2 = [ 4 5 6 7 ]

aug L1 , L2 = [ 1 2 3 4 5 6 7 ]

The total number of elements in the resulting augmented list must not exceed sixteen
or an error will occur.

List Calculations Menu

EL-W516 Calc ( Page 1 )
<CALC-1>
0: min
1: max
2: mean
3: med
4: sum
5:prod
Key: [ON/C][MATH][6]

EL-W516 Calc ( Page 2 )

<CALC-2>
6: stDv
7: vari
8: o_prod(
9: i_prod(
Key: []

EL-W516 Calc ( Page 3 )

<CALC-3>
A: abs_list

Key: []

112

Chapter 18. List Operations ( W516 only )

[MATH][6][0]

min

Format: min listname. This is a function that returns the minimum value in a stored list (
L1 - L4 ).
Example: min L1. [ON/C] [MATH][6]{CALC} [0]{min}[MATH][1]{LIST}[0]{L1}.
L1 = [ 5 1 3 0 6 1 ]
[MATH][6][1]

min L1 = 1

max

Format: max listname. This is a function that returns the minimum value in a stored list
( L1 - L4 ).
Example: max L1. [ON/C] [MATH][6]{CALC} [1]{max}[MATH][1]{LIST}[0]{L1}.
L1 = [ 5 1 3 0 6 1 ]
[MATH][6][2]

max L1 = 6

mean

Format: mean listname. This is a function that returns the average or mean value of all
the elements in a stored list ( L1 - L4 ).
Example: mean L1. [ON/C] [MATH][6]{CALC} [2]{mean}[MATH][1]{LIST}[0]{L1}.

L1 = [ 5 1 3 0 6 1 ]

[MATH][6][3]

mean L1 = 2.333333333 = 2

1
7
=
3
3

med

Format: med listname. This is a function that returns the median value of all the
elements in a stored list ( L1 - L4 ).
Example: med L1. [ON/C] [MATH][6]{CALC} [3]{med}[MATH][1]{LIST}[0]{L1}.
L1 = [ 5 1 3 0 6 1 ]

113

med L1 = 2

Scientific Calculator Comparison Vol I: User Guide and Operational Comparison

[MATH][6][4]

sum

Format: sum listname. This is a function that returns the sum of all the elements in a
stored list ( L1 - L4 ).
Example: sum L1. [ON/C] [MATH][6]{CALC} [4]{sum}[MATH][1]{LIST}[0]{L1}.
L1 = [ 5 1 3 0 6 1 ]
[MATH][6][5]

sum L1 = 14

prod

Format: prod listname. This is a function that returns the product of all the elements in
a stored list ( L1 - L4 ).
Example: prod L1. [ON/C] [MATH][6]{CALC} [5]{prod}[MATH][1]{LIST}[1]{L2}.
L2 = [ 4 5 6 7 ]
[MATH][6][6]

prod L2 = 840

stDv

Format: stDv listname. This is a function that returns the standard deviation of all the
elements in a stored list ( L1 - L4 ).
Example: stDv L1. [ON/C] [MATH][6]{CALC} [6]{stDv}[MATH][1]{LIST}[0]{L1}.
L1 = [ 5 1 3 0 6 1 ]
[MATH][6][7]

stDv L1 2.804757862

vari

Format: vari listname. This is a function that returns the variance of all the elements in
a stored list ( L1 - L4 ).
Example: vari L1. [ON/C] [MATH][6]{CALC} [6]{vari}[MATH][1]{LIST}[0]{L1}.
L1 = [ 5 1 3 0 6 1 ]

vari L1 7.866666667

Items 8:i_prod, 9:o_prod, and A:abs_list are vector functions. See chapter 16 for
details and examples of vector operations.

114

Chapter 19. Statistical Calculations

19. Statistical Calculations

The 115ES and W516 have comparable statistical capabilities. The W516 is, as usual,
a bit quicker providing calculation results. The 115ES, though a little slower at calculation, is,
also as usual, a bit more organized and flexible in the layout and execution of it's statistical
functions. Of particular note is the ability of the 115ES to compare the fit of several types of
regression to paired variable data, without having to retype all the data for each regression as
you have to do on the W516. Both calculators offer a wide array of statistical functions which
can be divided into three basic categories:
Single-Variable Statistics: Provides the usual descriptive statistics, for a singlevariable data set, such as; mean, standard deviation, variance, sums of data, sums of data
squared, etc. The 115ES supports a maximum of 80 data points; the W516 supports a
maximum of 100. These figures are halved if you use frequency figures for your data.
Standard Normal CDF: Calculates the usual transformations of the standard normal
cdf ( cumulative distribution function ). These are P(t), Q(t), and R(t); where t is the standard
normal random variable. The W516 calculates these probabilities to six decimal places,
though the last digit does not seem to be rounded properly at all times. The 115ES does not
state the precision of it's calculation but it does seem, from my own experimentation, to be
five decimal places and always properly rounded at the last digit.
Paired-Variable Regression: Performs seven different types of regression on pairedvariable data sets. These are; Linear, Quadratic, Exponential, General Exponential,
Logarithmic, Power, and Inverse Regression. The 115ES supports a maximum of 40 paired
data points; the W516 supports a maximum of 50. These figures are reduced to 26 (115ES) or
33 (W516) if you use frequency with your data pairs.
On both the 115ES and the W516, you access statistics mode using the mode menu.
On the 115ES, statistics mode is [MODE][3], and on the W516 it is [MODE][1].

115

Scientific Calculator Comparison Vol I: User Guide and Operational Comparison

fx-115ES
1: COMP
3: STAT
5: EQN
7: TABLE

EL-W516 ( page 1 )

2: CMPLX
4: BASE-N
6: MATRIX
8: VECTOR

<MODE-1>
0: NORMAL 1: STAT
2: DRILL
3: CPLX
4: MATRIX
5:LIST

Key: [AC] [MODE][3]

Key: [ON/C] [MODE][1]

When statistics mode is selected from the mode menu, you are presented with a 'type'
menu that prompts you for the type of statistical calculations you want to do. This menu is
shown below for both the 115ES and the W516:

fx-115ES
1: 1-VAR
3: _+CX2
5: e^X
7: AX^B

2: A+BX
4: ln X
6: AB^X
8: 1/X

Key: [1]{1-VAR}

EL-W516 ( page 1 )

EL-W516 ( page 2 )

<STAT-1>
0: SD
1: LINE
2: QUAD 3: E_EXP
4: LOG
5: POWER

<STAT-2>
6: INV
7: G_EXP

Key: [0] {SD}

Key: [][]

Single-Variable Statistics
Since we're going to begin with single variable statistics, choose type [1]{1-VAR} on
the 115ES or [0]{SD} on the W516.
Here is the where the key difference between the two calculators is regarding statistical
type. On the 115ES the statistics 'type' menu is accessible from inside statistics mode by way
of the stat menu, which means you can change the regression type at any time in statistics
mode without having to leave statistics mode or retype in any of your data. This is nice if you
want to compare how well several types of regression fit your data. On the W516, the only
way to change the regression type is to exit statistics mode and come back in again through
the mode menu to get to the 'type' menu, and this means that all your statistical data is
cleared. Once you select your statistics type on the W516, you are locked into that type for as
long as you are in statistics mode.
If you recall the setup menu for the 115ES, here is where one of the setup options
takes effect. On the 115ES, you will see one of the following data entry forms when you enter
single-variable statistics mode:

116

Chapter 19. Statistical Calculations

fx-115ES

1
2
3

fx-115ES

1
2
3

Key: data entry w/o freq

FREQ

Key: data entry w/ freq

Whether or not you see a 'frequency' column in the data entry form depends on what
the setting is for statistical frequency in the setup menu. It can be set to ON or OFF. This can
be changed, if you wish, even though you're already in statistics mode. Here's how: press
[AC][SHIFT][MODE]{SETUP} and press [] to get to page two of the setup menu. Select
item 4:STAT. Press [1] for frequency ON or [2] for frequency OFF. Press [AC][SHIFT][1]
{STAT} [2]{Data} and you'll be right back to the data input form with the frequency column
turned on or off, just as you selected. You need to be sure to do this before you start entering
data since toggling the frequency column on or off will erase all data entered up to that point.
For those of you not familiar with the use of frequency, it's a quick way of entering
multiple data points at once. If for instance you have three data points equal to 2.3, then you
would enter 2.3 as the x data and enter 3 as the frequency. Since memory for statistical data
is limited on either calculator, the use of frequency may help when keying in large data sets. If
you have a large number of data points but very few of them are the same, then do not use
frequency with your data on the 115ES since you can only store half as many data points if
you do.
The 115ES supports a maximum of 80 data points with frequency turned OFF but only
40 data points with frequency turned ON. The W516 supports a maximum of 100 data points
without frequency data and only 50 data points with frequency data. A fine distinction exists
between the 115ES and the W516 regarding the use of frequency. If you have just one data
point with an associated frequency on the 115ES, then you must have frequency turned ON
and you automatically drop to 40 available data points. The W516 works incrementally; if you
have just one data point with an associated frequency, then you only use that one extra
memory spot, so you'll still be able to fit 99 data points total. This is a very flexible approach
and wise use of memory. There is also no reason then to have to choose whether or not you
want to use frequency with your data on the W516.
Entering and Editing Statistical Data
The 115ES provides a special form for the entry of statistical data. On the W516,
though there is a data screen, the actual entry of statistical data is done directly from the
keypad. For an example, we will use the following data set:

117

Scientific Calculator Comparison Vol I: User Guide and Operational Comparison

x = [ 2 , 2 ,3 , 4 , 4 , 4 , 5 , 5 , 7 ,8 ]

fx-115ES

1
2
3
3

fx-115ES

2
2

1
2
3
3

Key: data entry w/o freq

FREQ
2
2
3
1
4

Key: data entry w/ freq

While in the 115ES stat data entry form, just type your data points and press [=] to
enter a data point. The cursor moves down when you press [=] so, if you are using frequency
with your data, you'll have to use the arrow keys to navigate around the table. Your data
points are automatically numbered and there's always a spot for you to enter more data at the
end of the data list. You can delete an entry entirely by pressing the [DEL] key. If you need to
edit the value of a certain data point, you need to put your cursor on that item and retype the
entire number; you can't just edit specific digits. Press [=] to accept the edit.
There are two other data editing functions available for you to use. You can insert new
statistical data anywhere in the list or you can delete all your data and start over. These two
functions are located in the stat edit menu, which you can access by pressing [SHIFT][1]
{STAT} [3]{Edit} and selecting 1:Ins for the insert data function or 2:Del-A for the delete all
function. You will then be returned to the data entry form. The insert function is a standard
one; put your cursor on a data item in the entry form then call up the insert function. That data
item, and all the rest of the data items below it, will be pushed down to make room for the
data to be inserted.
When you are done entering and editing your stat data, press [AC] to exit the data
entry form. You're ready to begin the analysis of the data. Note that, unless you edit your data
points manually, your statistical data will be retained as long as you are in single-variable
statistics mode, even if the calculator is turned off.
Entering data on the W516 is performed right from the keypad. Press [ON/C] to be
sure you are in statistics home screen. You'll see Stat 0 [SD] displayed at the top of the
screen. Type in the first data point, [2], and press DATA, which is the same as the [CHANGE]
key. Your first data point is accepted and repeated back to you, along with an update on the
number of data points you've entered so far.

118

Chapter 19. Statistical Calculations

EL-W516

EL-W516

Stat 0 [ SD ]

2DATA

0.

DATA SET=

Key: [AC]

1.

Key: [2][CHANGE]{DATA}

Continue entering data points: [2][DATA] [3][DATA] [4][DATA] [4][DATA] and so on,
until all the data points have been entered. If you want to use frequency instead to enter your
data, then just use the format data , frequency [DATA]: enter the data point, a comma
[(x,y)], the frequency, then press [DATA]. The data set would be entered as [2][(,)][2][DATA]
[3][DATA] [4][(,)][3][DATA] [5][(,)][2][DATA] [7][DATA] [8][DATA]. When you're finished,
press [ON/C] which just gives you a clean screen to begin your data analysis.
To edit data on the W516, press the [] to scroll through your data from the first item,
or press the [] to browse your data from the last item.

EL-W516
DATA SET=3
X:
F:

4.
3.

Key: [][][][][]

You can jump to the beginning or end of your data set by pressing [2ndF][] or [2ndF][].
To edit a data point or a frequency, move your cursor to the item to be edited, type in
the correct data and press [=]. To delete a data entry, press [2ndF][CHANGE]{CD}. This
clears the data item and its associated frequency from your data set. When you're finished
editing your data points, press [ON/C] to get back to the calculation screen. There is no
'insert' function as found on the 115ES. You can add more items to your data set at any time
from the calculation screen and the new data will be tacked onto the end of your data set.
Summary Statistics for Single-Variable Data
On the 115ES, the summary statistics for the data set are available in the STAT menu,
which is accessed by pressing [SHIFT][1]{STAT}. There are three sub-menus that hold all the
summary statistics. Sub-menu 4:Sum, 5:Var, and 6:MinMax:

119

Scientific Calculator Comparison Vol I: User Guide and Operational Comparison

fx-115ES
1: Type
3: Edit
5: Var
7: Distr

2: Data
4: Sum
6: MinMax

Key: [SHIFT][1]{STAT}

fx-115ES STAT Sum

Fx-115ES STAT Var

1:x2

1: n
3:xn

2:x

Key:[4]{Sum}

2:x
4: xn-1

Key: [5]{Var}

These summary statistics can be displayed for reference or may be recalled to the
calculation screen for use in expressions for further calculation. The sum sub-menu provides
the sum of the data points and the sum of the squares of the data points. The var sub-menu
provides the number of data items n, the mean of the data items x, the population standard
deviation xn, and the sample standard deviation xn-1. The population and sample
variances can be calculated by squaring the respective standard deviations:

Population (n)
Standard Deviation

Variance

xn =

x n =

x x

Sample (n-1)
2

x x 2
n

x n1 =

x n1 =

x x
n1

xx 2
n1

Variance is a measure of the amount of variation in the data. It is the expected or

average squared distance of a data point from the mean value of all the data points in the
data set. About 68% of the data points will be within 1 standard deviation of the mean, if the
data is distributed roughly normal. The standard deviation is the square root of the variance
and so could be said to be the root mean square value of the deviation of a data point from
the mean of all the data points in the data set.
The sub-menu 6:MinMax on the 115ES contains two items; 1:minX and 2:maxX.
These will provide you with the minimum and maximum values found in your data set. The
W516 does not provide this set of functions.
Summary statistics on the W516 are accessed directly on the keypad using [ALPHA]
shifted keys; the names of the summary statistics are printed in green above the appropriate
keys. The x is the population standard deviation and sx is the sample standard deviation:

120

Chapter 19. Statistical Calculations

[ALPHA][4] = x

[ALPHA][5] = sx

[ALPHA][6] = x

[ALPHA][0] = n

[ALPHA][.] = x

[ALPHA][(-)] = x2

Just as on the 115ES, the summary statistics can be simply displayed for reference or
may be recalled to the calculation screen for use in expressions for further calculation.
Using the information given for your calculator, verify the following summary statistics
for the sample data set:

x=[2,2,3,4,4,4,5,5,7,8]
n = 10

x = 4.4

x = 44

x2 = 228

x = xn 1.854723699

sx = xn-1 1.95505044

Here's an exercise that illustrates the use of the built-in summary statistics as part of a
larger expression. Using the summary statistics, can you verify that the following formula for
the population standard deviation gives the same figure as x on the W516 or xn on the
115ES?

x = xn =

E x E x
2

x
x
2

Remember that, while in statistics mode, you may only use line mode for your math
expressions. This is true for other modes as well. The expression for verifying the above
formula for population standard deviation should look like: ((x2 (x)2n)n)= where the
built-in statistics are in bold red.

Standard Normal Cumulative Distribution Function

Both the 115ES and the W516 calculate probabilities associated with the standard
normal distribution. These are variations on the standard normal cdf or cumulative distribution
function, P(t)., where t is the standardized normal variable. Given the value of t, the function
P(t) will return the probability that a random variable from the same distribution will be less

121

Scientific Calculator Comparison Vol I: User Guide and Operational Comparison

than t. This basically involves numerical algorithms that approximate the area under the
standard normal probability density function:
P t =

1
2

t2
2

dt

where

t =

xx
x

The variate t is the standardized normal variate. It represents a deviation from the
mean value or peak of the standard normal distribution, measured in units of the standard
deviation of the distribution. Both calculators can convert a value x to a standardized normal
variate t, using the parameters calculated from the current data set as shown in the equation
above. The conversion is found in the MATH menu on the W516 and in the stat menu on the
115ES.
Since the standard normal distribution is a pdf (probability density function), the area
under the curve from negative infinity to positive infinity will be exactly equal to one. This
means that P(t) will return a decimal value between zero and one to represent a probability
from 0% to 100%. The 115ES and W516 provide this function and two common variations;
Q(t) and R(t). Illustrations of the areas under the standard normal curve and the probabilities
associated with those areas are pictured below.

As an example, consider a value of t=1, which is one standard deviation from the mean
of the distribution. We know that about 68.3% of all random data will fall within 1 standard
deviation of the mean. From the illustration above for Q(t), we can verify this by finding the
area under the curve from 1 to +1. Since the standard normal curve is symmetric about the
mean, we just need to calculate 2Q(1).
On the 115ES press: [AC] [2] [SHIFT][1]{STAT} [7]{Distr} [2]{Q(} [1][=]. The 115ES
returns the value 0.68268 or 68.268%.
On the W516 press: [ON/C] [2] [MATH][3]{Q(} [1][=]. The returned value is 0.68269 or
68.269%.
The correct value is about 68.2689492%. The 115ES does provide five decimal digits
as advertised but the last digit is incorrectly rounded. The W516 also provides five decimal
digits but the last digit is accurately rounded. Why only five digits if the W516 manual states
that six decimal digits are calculated? The reason lies in the value of Q(1). It should equal
about 0.341344746. On the 115ES, Q(1) returns 0.34134, an incorrectly rounded five digit

122

Chapter 19. Statistical Calculations

result. The calculated value for Q(1) on the W516 is 0.341345, which is six digits, correctly
rounded. Since the last digit is a five, twice this value results in a zero in the final digit. So the
W516 is actually returning a six digit result with roundoff error in the last digit.
P(t) returns the lower-tail probability or the probability that a random variable is less
than t, the integral of the standard normal pdf from negative infinity to t. R(t) returns the
upper-tail probability or the probability that a random variable is greater than t, the integral of
the standard normal pdf from t to positive infinity. All returned probabilities are positive
decimals.
Note that the functions P(t), Q(t), and R(t) are available on the 115ES stat menu only
while in single-variable statistics mode. The W516 keeps these funtions available in the MATH
menu at all times, regardless if the type is single or paired-variable statistics.
Since the 115ES and W516 are both capable of numerical integration, you may want to
calculate the integrals of the standard normal pdf for yourself to obtain higher precision values
for P(t), Q(t), and R(t). For more information on these calculations, refer to Appendix XX.

Paired-Variable Statistics & Regression

The 115ES and the W516 are able to perform seven kinds of regression on pairedvariable statistics. Paired-variable statistics, as treated on the calculator, consist of a
independent variables x, and associated dependent variables y. If there is a correlation
between the independent and dependent variables, regression calculations will attempt to find
the best equation to 'fit' the given values of the x,y pairs. Both calculators will calculate a
correlation coefficient for all regressions except quadratic regression. This figure summarizes
how well correlated the data pairs are. Even so, judging whether or not a calculated curve
explains the nature of the relationship between the data pairs is a very difficult task; you
cannot judge a curve fit by correlation figures alone. Given that fact, use paired-variable
statistics on the calculator with caution.
Here are the seven types of regression available on the 115ES and the W516, their
model equations, and the stat menu entries for both calculators.

Regression Type
Linear

Model Equation

115ES stat menu

Y = A + BX

2: A+BX
2

W516 stat menu

1: LINE

Quadratic

Y = A + BX + Cx

3: _+CX

Logarithmic

Y = A + B ln(X)

4: ln X

4: LOG

Exponential

Y = A eBX

5: e^X

3: E_EXP

Power

Y = A XB

7: AX^B

5: POWER

Inverse

Y=A+B/X

8: 1/X

6: INV

General Exponential

Y = A BX

6: AB^X

7: G_EXP

123

2: QUAD

Scientific Calculator Comparison Vol I: User Guide and Operational Comparison

Notice that there are equation parameters A, B, and C that adjust the shape of the
curve of the selected equation. The values of these parameters are calculated by the
regression algorithm to best fit the entered data set. The regression types are not the only
ones possible and so not all data sets will find a good fit with the included regressions. Also,
more than one type of regression may give a good fit to your data and there's no guarantee
that the regression type with the best fit will be the proper form of equation to explain the data
correlation. In short, know your data, where it came from, and the nature of the process that
produced that data. You should probably have a good idea ahead of time what kind of data
you're working with since depending on the regression results alone may not be of much help
to you at all.
Here's a sample data set that we'll use for our first example. You could, if you wanted
to, use the x,y data to plot these points on a graph, which is what is shown below. The blue
x

0.7

1.2

1.2

2.8

1.8

7.3

2.3

11.9

3.0

22.1

4.7

58.0

5.2

71.5

curve on the graph is an estimate of the curve that regression might calculate to fit the data
points. It could be logarithmic, quadratic, or exponential. It could be anything really; the points
are not exactly accurate since there was some imprecision in the measurement. I will give you
a hint though... the curve is quadratic.
With that hint, let's enter the data into our calculators. On the 115ES, even if you're
already in stat mode, press [MODE][3]{STAT} [3]{_CX2} for quadratic regression. Using the
mode menu instead of just changing type in the stat menu will also clear your data set, which
is what we want for this new example. To enter your data, press [SHIFT][1]{STAT} [2]{Data}.
You'll see the X and Y columns where you'll enter your data. You'll also see a FREQ column if
you have data frequency turned ON. Using your arrow keys, navigate through the table,
entering your X and Y data just as they appear in the table above. See Entering and Editing
Statistical Data in the Single Variable Statistics section earlier in this chapter for details on
entering your data.
On the W516, press [MODE][1]{STAT} [2]{QUAD} for quadratic regression. Your data
set is cleared; you have no choice since the W516 clears your data set anytime you re-enter
stat mode. You can enter your data in the format {xdata}[(,)]{ydata}[DATA]. Enter all seven
data pairs, one at a time, then press [ON/C] to give yourself a clear display.

124

Chapter 19. Statistical Calculations

Next we'll take a look at the results of the quadratic regression. The calculator will find
the values of the parameters A, B, and C in the following equation, such that the resulting
quadratic function will pass as close as possible through all your data pairs:
y = A + Bx + Cx2
In other words, the coefficients of the terms of the quadratic function are those that describe a
curve of this type that is the best fit through all of the data points in question.
To see the values of these coefficients on the 115ES, you'll want to go to the regression
menu. Press [SHIFT][1]{STAT} [7]{Reg}. Items 1:A, 2:B, and 3:C are the ones you're looking
for, and their values are: A=1.068661827, B=2.055774952, and C=3.005882983. So we find
that the quadratic equation that best fits our data set is approximately y=3x2-2x+1.
To see the coefficients of the quadratic on the W516, press [ALPHA][(]{A}, [ALPHA][)]
{B}, or [ALPHA][X]{C}. You'll find that the values for the coefficients agree with the 115ES;
A=1.068661827, B=2.055774952, and C=3.005882983.
Neither the 115ES nor W516 provide a correlation coefficient for quadratic regression
but, there is a way to see how well the calculated curve fits your data. On the 115ES, there
are a couple other items in the regression menu. They are 4:x1, 5:x2, and 6:y. These items
will calculate values from the quadratic y=Cx2+Bx+A using the coefficients determined by the
regression algorithm. For example, the fourth data pair in the data set is x=2.3,y=11.9. Press
[AC] [2][.][3] [SHIFT][1]{STAT} [7]{Reg} [6]{y} [=] and the result returned is about 12.24. It
doesn't quite match our actual value of 11.9 but it is close. We can also supply the y value
and have the calculator find the x value from the quadratic. We'll try the 11.9 from the fourth
data point again. Press [AC] [1][1][.][9] [SHIFT][1]{STAT} [7]{Reg} [4]{x1} [=] and the result
returned is about 2.27, which is close to 2.3. Item 5:x2 from the regression menu, since we're
working with a quadratic, returns the x value from the mirror image half of the quadratic; in
this case returning an x value of about 1.59.
The W516 will also calculate estimates from regressions. These are available right on
the keypad and are; [2ndF][(]{x} for estimates of the x values, and [2ndF][)]{y} for
estimates of the y values. Press [ON/C] [2][.][3] [2ndF][)]{y} and you get about 12.24. For
an estimate of x values try [ON/C] [1][1][.][9] [2ndF][(]{x} and you get two answers which
are from the two mirror image halves of the quadratic. They are about 2.27 and 1.59.
While we're on this problem, check out the sum menu on the 115ES; there are eight
calculated sums derived from your data set. The W516 only provides five: x, y, x2, y2,
and xy. The 115ES provides these sums along with three others: x3, x4, and x2y. The
sum menu on the 115ES can be accessed by pressing [AC] [SHIFT][1]{STAT} [4]{Sum}.
This is about as far as you can go with your data set on the W516. If you want to see
the results of other types of regression on your data, you'll have to leave stat mode and reenter it again, selecting a different type of regression. This also means that your data set has
been cleared and you'll have to retype all your data pairs. The 115ES does not clear your data
when you change regression types, so you're able to compare the fits of several types of
regression to your data set. Here is a table of correlation coefficients, for the data set in our
example, for all of the regression types available on the 115ES.
125

Scientific Calculator Comparison Vol I: User Guide and Operational Comparison

Regression Type
Linear

Model Equation

115ES Stat Menu

Y = A + BX

2: A+BX
2

3: _+CX

Correlation Coefficient (r)

0.9759

Quadratic

Y = A + BX + Cx

Logarithmic

Y = A + B ln(X)

4: ln X

0.8789

Exponential

Y = A eBX

5: e^X

0.9695

Power

Y = A XB

7: AX^B

0.9983

Inverse

Y=A+B/X

8: 1/X

-0.7116

General Exponential

Y = A BX

6: AB^X

0.9695

To change regression type on the 115ES, using the current data set, you simply call up
the stat menu [SHIFT][1]{STAT} and select 1:Type. You then have seven types of paired
variable regression to choose from, items 2 through 8. If you want to preserve your data, do
NOT choose item 1:1-VAR. This is not a regression type but the entry point for single variable
statistics; if you choose single variable statistics, your data will be cleared. When you press a
number key for the type of regression you want, you'll be taken immediately to the data entry
form, showing you the current data set and offering you the chance to edit the data. We don't
want to do this so just press [AC]. For example, I've just selected item 7: AX^B, which is the
entry for Power Regression, and then pressed [AC]. To see the correlation coefficient for
power regression on your data, press [SHIFT][1]{STAT} [7]{Reg} [3]{r} [=] and you'll see the
result, after a few seconds of calculation, 0.9983421711. Repeat the above procedure for as
many types of regression as you like.

126