# Lesson 1 --- Introduction Abacus Chinese abacus and modern mathematical computing Introduction

Abacus or 算盘 (suan4 pan2) in Chinese was a widely used computing machine before the invention of modern computers. The mathematics of abacus is called 珠算 (zhu1 suan4) in Chinese. Although it has been used for thousands of years in businesses all around the world, abacus is losing to computers as a practical computation tool. Then why should we learn abacus? First, it is fun. With some practice, you can impress friends with amazing computation skills. Second, abacus is a rudimentary computer. One can visualize the computer at work by pushing beads and learn some very basic concepts of modern computers. We�ll teach here the Chinese abacus. There are other forms of abacus, but the principles are all the same. For more information, please Google for abacus. A Chinese abacus instrument can be purchased from Amazon: Chinese abacus

Counting

1. Heaven and earth beads The Chinese abacus has 2 beads in the upper deck (heaven beads) and 5 beads in the lower deck (earth beads).

Each heaven bead in the upper deck has a value of 5; each earth bead in the lower deck has a value of 1. Beads are considered counted, when moved towards the central bar that separates the two decks.

2. Five There are two ways to count 5: one is to move 5 earth beads up to the center; the other is to move 1 heaven beads down to the center.

3. Ten There are also two ways to count 10: one is to move 2 heaven beads down to the center; the other is to move 1 earth beads in the column immediate left up to the center.

4. Any number Once we understand how to count to 10, it is straight forward to count any integer in abacus. Just remember that the value for a bead is 10 times the value of the bead on its right; or is 1/10th the value of the bead on its left.

5. Unit and decimal point You must be asking, where do I place the decimal point? You can start anywhere on the abacus, i.e., chose any rod as the unit rod. The earth bead on the unit rod is exactly 1; earth bead on the left of the unit rod is 10 and earth bead on the right of the unit rod is 1/10th, and so on.

**Lesson 2 --- Addition Abacus Chinese abacus and modern mathematical computing
**

Addition on the abacus is straight forward: simply add the numbers in the left-to-right sequence. (One could add from right to left, the result is the same.) Example 1: 146 + 52

Example 2: 42 + 33

Example 3: 28 + 16

Example 4: 378 + 1659

Note when adding, we are not simply adding beads but sometimes removing beads. In mathematical expression, these operations are as follows: +2 = +5 -3 +2 = +10 - 8 +6 = +10 - 4 +6 = +1 - 5 + 10 ......

**Lesson 3 --- Subtraction Abacus Chinese abacus and modern mathematical computing
**

Subtraction on the abacus is performed by subtracting the numbers in the left-to-right sequence. Example 1: 146 - 25

Example 2: 47 - 34

Example 3: 128 - 36

Example 4: 623 - 424

Note when subtracting, we are not simply removing beads but sometimes adding beads. In mathematical expression, these operations are as follows: -2 = -5 +3 -2 = -10 + 8 -6 = -10 + 4 -6 = -1 + 5 - 10 ...... Experienced users can use two or three fingers (thumb, index, or middle finger) to move the beads. For example, 8 + 4 = 8 + 10 � 6 can be done in two moves. The first move adds 1 to the rod on the left using thumb or index finger; the second move subtracts 6 using both thumb and index finger (or middle finger) by pushing beads outward. Similarly +3 = +5 � 2 can be done in one move: pushing down one heaven bead with index (or middle) finger and 2 earth beads with thumb simultaneously. Alternatively +3 = +5 � 2 can be done in one continual move: using the index finger to push down one heaven bead and continue on to push down 2 earth beads.

**Lesson 4 --- Multiplication Abacus Chinese abacus and modern mathematical computing
**

Before we start, let�s explain what multiplier and multiplicand are. In the expression a*b, a is the multiplier and b is the multiplicand. The steps of multiplication on abacus are summarized in the figure below.

Step 1. Start from right to left for each digit of the multiplicand, calculate the multiplication product for the digit and a digit in the multiplier. The order for the multiplier is also from right to left. Step 2. Add the multiplication product number on the abacus. Step 3. As one moves from right to left for either multiplicand or multiplier, the product number should be added on the abacus one column to the left. It is confusing to read these sentences. Let�s go through some examples. The procedure is actually very simple. Example 1. 27 * 4 = 108 Step 1. The first digit from right for the multiplicand is 4. The first digit from right for the multiplier (27) is 7. 4 * 7 = 28. We add 28 on the abacus. Step 2. The first digit from right for the multiplicand is 4. The second digit from right for the multiplier (27) is 2. 4 * 2 = 8. Since we have moved one digit to the left for the multiplier, so the number 8 should be added one column left relative to the unit column. That is we need to add 80 on the abacus. The result is 108.

Example 2. 45 * 27 = 1215

Step 1. The first digit from right for the multiplicand is 7. The first digit from right for the multiplier (45) is 5. 7 * 5 = 35. We add 35 on the abacus. Step 2. The first digit from right for the multiplicand is 7. The second digit from right for the multiplier (45) is 4. 7 * 4 = 28. Since we have moved one digit to the right for the multiplier, so the number 28 should be added one column left relative to the unit column. That is we need to add 280 on the abacus. The result is 315. Step 3. Now we are done with the first digit from right for the multiplicand, we move on to the second digit 2 from 27. 2 times the first digit from right for the multiplier (45) is 10. Since we have moved one column to the left for the multiplicand, the number 10 should be added 1 column left relative to the unit column, i.e., 100 should be added. The result is 415. Step 4. The second digit from right for the multiplicand is 2. The second digit from right for the multiplier (45) is 4. 2 * 4 = 8. Since we have moved one digit to the left for the multiplier and one digit to the left for the multiplicand, so the number 8 should be added two columns left relative to the unit column. That is we need to add 800 on the abacus. The final result is 1215.

**Lesson 5 --- Division Abacus Chinese abacus and modern mathematical computing
**

A division of two numbers x and y can be written as follows: x/y. Here x is called the dividend and y is called the divisor. Division is the inverse operation of multiplication, so the abacus moves are almost the inverse. There is one important difference: the number on the abacus after each move is the remainder value but the result. Unlike addition, subtraction, and multiplication, the result of a division must be recorded separately.

Step 1. Start with the number for the dividend on the right half of the abacus. Work from left to right. Record the first digit (from left) of the result on the left half of the abacus. Step 2. Multiply the first digit of the result and the first digit (from left) of the divisor and subtract the number from the dividend. Step 3. Multiply the first digit of the result and the second digit (from left) of the divisor and subtract the number from the remainder. As one moves from left to right, the product number should be subtracted on the abacus one column to the right. Step 4. Repeat step3 until we reach the end of divisor. Step 5. Repeat steps 2-3 for the second digit of the result and so on. As one moves from left to right for the result, the product number should also be subtracted on the abacus one column to the right. Example 1: 373 / 6 Step 1. Add 373 on the right half of the abacus. Step 2. The first digit of the result from left is 6. Record this number on the left half of the abacus. Step 3. Multiply the first digit from left for the result (6) and the first digit of the divisor (6) and subtract the number 36 (6 x 6) from the dividend. Note the alignment should be on the left for both the dividend (373) and 36. That is 360 is subtracted from 373. In some cases, when subtracting, align to the left most digit of the dividend may not be possible (e.g., 12/8) then the multiplication product should be aligned to the second digit from left for the dividend. Step 4. The second digit from left for the result is 2. Record this number next to the first digit already on the abacus. Step 5. Multiply 2 and the divisor, and subtract the result (12) from the remainder number on the abacus. Note when subtracting, we need to shift one column to the right since we have shifted one column to the right for the result. So 12 are subtracted from the remainder.

Step 5. One may continue to get results after the decimal points.

Example 2: 3732 / 63 Step 1. Add 3732 to the right half of the abacus. Step 2. Start from left to right. Record the first digit of the result 5 on the left half of the abacus. Step 3. Multiply 5 and the first digit (6) of the divisor, subtract the result 30 from the dividend. Note 30 should be aligned to the left of the dividend; that is we actually subtract 3000 in this example. Step 4. Multiply 5 and the second digit (3) of the divisor, subtract the result 15 from the remainder on the abacus. Since we have shifted one column right for the divisor, we also need to shift one column right when subtracting. We actually subtract 150 in this example. Step 5. Now we repeat for the second digit for the result. Record 9 next to 5 already on the abacus. Step 6. Multiply 9 and the first digit (6) of the divisor, subtract the result 54 from the remainder. Note 54 should be aligned to the second digit from left of the dividend since we have shifted one column right for the result; that is we actually subtract 540 in this example. Step 7. Multiply 9 and the second digit (3) of the divisor, subtract the result 27 from the remainder. We need to shift right one more column since we shifted one column right for the divisor. In the example, we subtract exactly 27. Step 8. One may continue to get results after the decimal points.

**Lesson 6 --- Abacus and computer Abacus Chinese abacus and modern mathematical computing
**

Abacus, as one of the earliest computers, is not fundamentally different from the electronic computers. In the following, we shall explain how computers work using abacus as an example.

Base

We are familiar in counting in the numbering system 1,2,3,�10, 11,�100,� We add one more digit (column) once we reach 10 and another digit once we reach 100, and so on. Mathematically speaking, it is a numbering system of base 10. It is possible to construct numbering systems with a different base. For example, the number 17 in base 10 is represented by different numbers in different bases:

Base 10: 17base 10 = (1 * 10 + 7)base 10 Base 16: 17base 10 = (1 * 16 + 1)base 10 = 11base 16 Base 5: 17base 10 = (3 * 5 + 2)base 10 = 32base 5 Base 3: 17base 10 = (1 * 32 + 2 * 3 + 2) )base 10 = 122base 3 Base 2: 17base 10 = (1 * 24 + 0 * 23 + 0 * 22 + 0 * 21 + 1) base 10 = 10001base 2

The concept of base can be easily explained using abacus with different number of beads (see Figure). You may have noticed that the Chinese abacus can also be used as a calculator in a base 16 numbering system. You are absolutely correct: Chinese used to use a base 16 system for weight measures (1 斤 = 16 两), which is probably the reason why Chinese abacus used more beads than necessary for a base 10 numbering system.

The electronic computers can be imagined as an abacus with the base 2 numbering system.

Computation

Computation in abacus is done by moving beads. The principle is the same in electronic computer, only that the tiny beads in an electronic computer are chip transistors with two �positions�: �0� position and �1� position. The transistors are not moved by mechanical force. The �position� or state of a transistor is changed by electronic signals. The mathematics for an electronic chip computer and the mathematics for a base-2 abacus computer, however, are no different. For example: 10001 + 1011 = 11100 10001 � 1011 = 110 One can easily verify that in base 10 the above calculations are: 17 + 11 = 28 17 � 11 = 6

Algorithm

For most computations, there are different methods to get the same results. But not all methods are as efficient. For example, to get the result of 123456789 * 10 we can simply add a zero at the end of 123456789 or add 123456789 nine times. The first procedure is much more efficient. An algorithm is a mathematical procedure for certain type of computations. Mathematicians and computer engineers have developed thousands of algorithms for efficient calculations. Here we�ll use abacus to illustrate a simple principle in algorithm development: minimize the number of times �beads� are moved to speed up calculations. We have learned in elementary school that the easiest multiplications and divisions are multiplying or dividing by 10, 100, and other powers of 10. On base 10 abacus, the calculation can be easily achieved by shifting beads positions left for multiplication or right for division. This efficient procedure is also true for the base 2 numbering system: on a base 2 abacus, multiplying or dividing by 2 can be easily achieved by shifting beads positions left or right.

Because multiplying by the base is efficient, it is repeatedly used in multiplications to minimize the number of �beads� moving operations to increase computation speed.

In the figure above, we show how to do multiplication in the binary system (Fig. A). One can easily see the similarities between the multiplication algorithms used for

base 10 and base 2 numbering systems. Interestingly, if we rewrite Figure A in the base 10 system, we have an easy way to multiply two numbers without remembering the multiplication table. For the example of 54 * 11, the procedure is as follows: 1.Write down 11 and 54 on row 1. 2.Divide 11 by 2. The result is 5 and remainder 1. Write down 5. If the remainder was 0, cross out 54 in the preceding line. Since the reminder is 1, don�t cross out 54. 3.Time 54 by 2 and write down 108. 4.Repeat steps 2 and 3. Divide 5 by 2 to get 2 and remainder 1. Write down 2 and don�t cross out 108. Time 108 by 2, and write down 216. 5.Repeat steps 2 and 3. Divide 2 by 1 to get 1. Write down 1 and cross out 216 in the preceding line because the remainder is 0. Time 216 by 2 and write down 432. 6.Repeat steps 2 and 3 if necessary until the number on the left is 1. Add up all the numbers on the right that have not been crossed out, you get your answer. In our example, the result is 594. We�ll leave it to the readers to figure out why the simple procedure works. We�ll give you a hint: (110110)base 2 = (54)base 10, (1101100)base 2 = (108)base 10, and (110110000)base 2 = (432)base 10.

Overflow

With a smaller abacus computer, we can easily see why computers may make mistakes. Suppose we start with 123 and times 10 repeatedly. Sooner or later we�ll reach the left end of the abacus --- we got a data overflow problem. This is one of the major reasons why our computers may freeze. Sometimes it is even worse; the computer, when reaching the end of the abacus, may simply continue and discard data. We may get results similar to this: 123, 000, 000, 000, 000 * 10 = (1 discarded) 230, 000, 000, 000, 000. Similarly one may also reach the right end of an abacus. This problem is called data underflow.

**Lesson 7 --- Square root Abacus Chinese abacus and modern mathematical computing
**

Abacus can be used to solve more difficult computational problems such as square root and cubic root. The procedures used in solving these problems again remind us of algorithms for electronic computers: simple procedures are repeated many times to solve complex problems. Square root instructions For simplicity, we give instructions for integers only. It is easy to generalize to any positive real number. 1. Add the integer number on the abacus. We shall call this number the operand. 2. Group the digits into columns of two for right to left. For example, 509 should be grouped as 5 09. 94664 should be grouped as 9 46 64. 3. Start from the first group from left, add 1 to the result, and subtract one from the first group. 4. If the first group is greater than result plus 2, add 2 to the result and subtract the value of the result from the first group. Continue adding 2 to result and subtracting from the first group until the value for the first group is smaller than the result. 5. If there are more groups, shift one column to the right for the result and 1 group (2 columns) right for the operand. 6. Add 11 to the result and subtract the result from the operand. If after adding 11 the result is greater than the operand, then don�t add 11 but continue shifting for the result and the operand. Then add 101 or 1001 etc to the result and subtract the result from the operand. 7. Add 2 to the result and subtract the result from the operand. Continue adding 2 to result and subtracting from the operand until the value is smaller than the result. 8. Repeat steps 5-7 if necessary. 9. The final square root value is (result + 1)/2. Example 1:

Example 2:

There are many algorithms for calculating square root. For more information, please visit Wikipedia. The algorithm based on the abacus procedure can be found here.

The Chinese Abacus

The Abacus appears in many different guises in many countries. They have also been used for many different number bases too. The Chinese Abacus counts in base 10, and has been around for a long time. It can still be seen in use in Chinese stores to reckon the grocery bills, or taught its use in the classroom to the younger generations. With the advent of the pocket sized calculator, in the last couple of decades, the Chinese abacus has still got a place in modern commerce and arithmetic. It does not require any power source, other than human manipulation, and accuracy depends on the proficiency of the user. I am learning by practicing on my wooden abacus. I'd like to show you sone techniques for manipulating it for arithmetic. The little manual that is my guide has been hard to follow since its english is not very clear, and cramps many stages of a process together. I hope that my presentation here using pictorial examples with my own comments will put across the message. Below is a diagram of an abacus (a hybrid of the two I have).

It is best to leave any preconcieved notions about beads on sticks behind, as there are things which require a new way of thinking, which hopefully, will not bee too hard to comprehend.

**The Chinese Abacus 2
**

Here we define a few things which will ease our introduction into the use of the abacus.

The correct position of the abacus is as shown above in the diagram. The CROSSBAR is that which separates the two sets of beads. The TOPBEADs are the set of two beads above the CROSSBAR. The UNITBEADs are a set of five beads below the CROSSBAR. STACKing is the process where we place a bead against the CROSSBAR WITHDRAWing is the process where a bead is taken off a stack (of beads against the CROSSBAR) When a UNITBEAD is moved to rest against CROSSBAR, it represents the number one (1). When two UNITBEADS are moved to rest against a CROSSBAR it representw the number two (2), and so on.

When a TOPBEAD is made to rest against the CROSSBAR, it represent the number five (5). Since there are two, the other is rarely used, but if it is, then the value of 10 is expected. Note that the abacus works in base 10, as in normal counting. The rods indicate the various units of counting in base 10. That is, starting from an arbitary rod, units, and continuing left, tens, hundreds, thousands... However, to the right of the units rod, decimal fractions are represents, tenths, hundredths, thousandths etc. By moving the beads against the bar, they represent the number desired.

**The Chinese Abacus 3
**

Stacking and Withdrawing

To stack or withdraw a bead, any of the following finger motions can be used. 1. Using only the forefinger, 2. Using both the thumb and forefinger 3. Using the thumb, forefinger and middle finger. The second method is probably the easiest one to use.

**Manipulation of the beads
**

1. 2. 3. 4. 5. 6. Stacking of counters Withdrawing of counters Stacking topbeads (5) and withdrawing a unitbead (1) Stacking a untibead (1) and withdrawing off a topbead (5). Withdrawing unitbeads and stacking a bead to the left. Withdrawing a bead on the left and stacking unitbeads.

Something to remember

As we have said, certain things about abacus arithmetic are very different from that of writing down the numbers and then doing the arithmetic. It is mainly because of the small number of beads on the abacus which forces such differences, and that in some operations, we commence from the left rather than the right.

**The Chinese Abacus 4
**

Addition 1

This is the diagram which ended the last section. We show how to add 7 to 7 and give the answer 14 above. Firstly, seven is stacked by moving a topbead and two unitbeads to the crossbar. Next we have to think differently about the seven we are about to add to the original number seven. Seven can be thought of as 10 minus 3. ie, 7 = 10 - 3 Also, 3 can be thought of as 5 minus 2. i.e. 3 = 5 - 2

By going through two indirect routes, we arrive at the answer. ie. 7 = 10 - (5 -2) = 10 - 5 + 2 That is, put a bead in the tens column, remove a topbead (5) and then add two unit beads. Alternatively, the net result of this is that, 7 = 12 - 5, so we stack in 12, and remove five. Due to the constraints of the abacus' design, doing this route is by far the most natural way to the answer. It is a basic Chinese abacus manipulation technique, which we call SUBSTITUTION. It is modified later when we see how it is used in subtraction. Obviously if the two numbers add up to less than 10, so it fits on a rod, substitution is clearly not used.

Exercises

Try the following, bearing in mind the substitution techiques (10 - A, and 5 - B where A and B are numbers) above, for some of the excercises below:

• • • • • • • •

1+2=3 2+3=5 3+4=7 5 + 7 = 12 6 + 6 = 12 8 + 9 = 17 7 + 8 = 15 8 + 2 = 10

**Double Digit numbers
**

When there is one or more double digit numbers to add, put the smaller of the two on the abacus first, the put in the larger number from the left, so

•

8 + 12 = 20

will have 8 on the unit rod first, then the digit 1 (from the number 12) is placed into the tens column. The digit 2 is added to the units column, but we note it gives 10, so we carry a one over to the tens column and clear the units column. (ie., 2 = 10 - 8, so add a second bead to the tens rod, and then take away 8 from the units column)

The next page will deal with the addition of numbers with many more digits.

Exercises

Try the following, bearing in mind the substitution techiques (10 - A, and 5 - B where A and B are numbers) above, for some of the excercises below:

• • • • • • • •

3 + 19 = 22 7 + 58 = 65 9 + 42 = 51 11 + 72 = 83 21 + 13 = 34 43 + 49 = 92 15 + 27 = 42 69 + 26 = 95

**The Chinese Abacus 5
**

Addition 2

Larger Numbers

After doing the excercises in Page 4, you will have by now got a good grasp of the basic manipulation techniques for addition. For numbers with a longer string of digits, the processes are just the same, starting from the left and proceeding to the right.

86452 + 9621 = ?????

As usual if there is a smaller number to be added, then this should be the first number to be stacked into the abacus. In our example here, we will stack in 9621 first and then add 86452, ie, 9621 + 86452 = ?????

Other than noting that 8 is an order of magnitude greater, so it is stacked one rod to the left of the highest place of 9621, there are no surprises here.

Here we note that to enter the next digit in 86452, (i.e. 6), we have to think of 6 as being 10 - 4. That is stack in one 10's bead and withdraw 4 unitbeads.

The next digit in 86452, is 4, we have to think of this as being 10 - 6. That is, stack in one 10's bead and withdraw a topbead worth 5 and a unitbeads worth 1.

The next digit in 86452, is 5, we only have to stack in a topbead worth 5.

The last digit in 86452, is 2, only requires us to stack two unit beads towards the crossbar.

Notes

Notice that the numbers are constantly revised to take account of numbers greater than the current base 10 position. We say in the second move that adding the next number required the addition of a bead to the left of the current rod. This updated the 8.... to 95...

Exercises

Try adding 123456789 repeatedly to itself nine times. (You should get 1111111101 as an answer. You can ignore the advice of putting in the smaller number first, since that was just for getting you used to entering from the left to right, since the answer at each stage is larger than the number 123456789 to be added.) Just to check, the answer at each stage is : 1. 2. 3. 4. 5. 123456789 246913578 370370367 493827156 617283945

6. 7. 8. 9.

740740734 864197523 987654312 1111111101

**The Chinese Abacus 6
**

Subtraction

This also involves working from the lefthand side to the righthand side of the abacus, involving the use of a slightly different substution method. We outline it below. Numbers less than 10 do not require the substitution method since they can be handled on the rod itself. So 9 - 6 would just meant that a topbead and a unitbead would be removed from the abacus representation of 9, leaving three unitbeads. When subtracting, it is usual to subtract the smaller number from the larger. Since there is no sign to represent negative numbers on an abacus, the operator will have to know if the the answer is negative or not by him/herself.

12 - 8 = ??

From our substitution method, we note that 8 = 10 - 2. If we place this into 12 - 8 we get, 12 - (10 - 2) = 12 - 10 + 2 This means, we take off a unitbead in the tens column, and add two unitbeads into the units column.

Exercises

Try the following out:

1. 2. 3. 4. 5. 6. 7.

16 - 7 = 9 53 - 6 = 47 22 - 9 = 13 71 - 4 = 67 40 - 2 = 38 94 - 8 = 86 31 - 2 = 29

**Subtracting double digit numbers and higher.
**

The above principle is applied to the above, but we take it a digit at a time, starting again from the left.

52 - 39 = ???

Exercises

Try the following out: 1. 2. 3. 4. 5. 6. 7. 86 - 72 = 14 32 - 19 = 13 24 - 17 = 7 102 - 39 = 63 528 - 89 = 439 941 - 812 = 129 160 - 74 = 86

**Subtracting larger numbers
**

As you see in the examples, the substitution method is very powerful for addition and subtraction.

Exercises

Before, we asked you to add 123456789 successively to itself nine times to get 1111111101. It is now time to try subtracting 123456789 from 1111111101 successively. To help check your work, the numbers you should get at each stage is again given. 1. 2. 3. 4. 5. 6. 7. 8. 9. 1111111101 987654312 864197523 740740734 617283945 493827156 370370367 246913578 123456789

**The Chinese Abacus 7
**

Multiplication and Division

These processes are a little more complex than addition and subtraction. It requires the user to be familiar with the multiplication tables. In essence, the abacus becomes only a recording device where interim results are tabulated. It is only in the final movement of the last manipulated bead, that the answer is revealed. The MULTIPLICAND is the number we start off with. The MULTIPLIER is the number we multiply the multiplicand with. The PRODUCT is the result of the multiplication. 1 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 2 2 4 6 8 10 12 14 16 18 3 3 6 9 12 15 18 21 24 27 4 4 8 12 16 20 24 28 32 36 5 5 10 15 20 25 30 35 40 45 6 6 12 18 24 30 36 42 48 54 7 7 14 21 28 35 42 49 56 63 8 8 16 24 32 40 48 56 64 72 9 9 18 27 36 45 54 63 72 81

The multiplication tables shown above is colour coded to show the multiplier, multiplicand, and also those products which are single digit values, and double digit values.

When we tackle the idea of multiplication, the learner to needs to take into account the number and position of the rods at each stage. For the most part, two rods are must be used for each step of the manipulation, but occasionally, three rods may be involved. The process of multiplication and division is complicated by the position of the decimal point. We have to make room on the abacus for it, and there are three important cases which arise with mumbers. A whole number is known as an INTEGRAL. Therefore, DECIMALS are non-integral because they contain part of a whole number.

• • •

Both the multiplicand and multiplier are integral One of these numbers is integral whilst the other is a decimal Both numbers are decimals

Since decimals are base 10 numerals, the decimal point does not take too much manipulation. We shall first explore the method for integral multiplicand and multiplier first, then extend the idea towards the other two cases.

**Multiplication 713 x 289 = ?
**

The integer multiplicand 713 is multiplied by the multiplier 289. We must first set out space on the abacus for our calculation.

• • • • • • •

If a single digit number is multiplied by another single digit number, the maximum number of digits the product has is two. If a single digit number is mulitiplied by a double digit number, the maximim digits the product has is three. If both the multiplicand and multiplier are double digit, their product has a maximum of four digits.

The trend is that for an M digit multiplicand which is multiplied by an N digit multiplier, the product has a maximum of (M + N) digit result.

In our example, both 713 and 289 have three digits each so that makes a maximum number of 6 digits in the resultant product. Hence, for the calculation, six rods on the abacus is needed.

In this case, count three rods along from the right side of the abacus, then another three and place the multiplier in the latter three rods. We shall see that the production will eventually replace the multiplier as each of its digits are considered in the calculation. Some where away from the six rods, you place your multiplicand for reference. The process of multiplication requires the learner to start with the rightmost digit of the multiplier, and mulitiply the multiplicand digit by digit from the left of the multiplicand to the right. The second rightmost digit of the multiplier then repeats the process with thee multiplicand in the same manner until the leftmost digit of the multiplier is done. The product will be that which is shown at the end of the manipulation. So, the calculation follows the following route. 713 x 289 Step (1) The right most digit of the multiplier is "9". Multiply the leftmost digit of the multiplicand, "7". Step (2) Next, 9 multiplies the next digit of the multiplicand, "1". Step (3) Then, 9 multiplies the last digit of the multiplicand, "3". We then repeat this process with the second digit of the multiplier "8", and finally the leftmost digit "2" As we said, the manipulation of the beads on the abacus requires one to think about two rods at a time. For example, in step (1), the product of this minor calculation was 63. We note the position of digit nine of the multiplier and the rod to its right. This forms the tens and units place in which the numeral 63 is inserted, replacing the digit 9 by the 6 and placing 3 to the rod on its right. In step (2), the product is nine, a single digit. Again, we consider this as a double digit number 09, and insert increment 0 beads in the tens (where the digit 3 appears), and put in the value of 9 on the abacus to the next rod on the right. In step (3), the product of 9 by 3 is 27. This move requires us to consider more than two rods. This is because we must put in the value 2 to the position of 9 on the rods, but this would make 11, hence the a bead on the next left rod is incremented by one, and the rod containing 9 would be ammended to 1, and finally the digit 7 would be put into the rightmost column as the units column. The next step with the second digit of the multiplier, viz, "8", would give the products 56, 8 and 24. The first digit of 56, "5" replaces the the multiplier digit "8" and the second digit is added to the value on the rod immediately to the right, and so on. Finally, the products of the last digit of the multiplier, viz, "2", yields the products, 14, 2, 6. The first digit of 14, "1", replaces the first digit of the original multiplier, and the second digit is added to the rod immediately to its right. The other numbers are placed in in the same way described above.

**What we have is the following representation
**

*** are unused rods. --- are places which have not been considered yet, but are about to change because of carrying overflow. +++ are digits which are not part of the final product

713

289***

Abacus is loaded thus

Abacus reads 289*** ++

63 9 27

The product of 9 and 7 The product of 9 and 1 The product of 9 and 3

Abacus reads 2863** Abacus reads 28639* Abacus reads 286417 +

56 8 24

The product of 8 and 7 The product of 8 and 1 The product of 8 and 3

Abacus reads 25---- then 262417 Abacus reads 262--- then 263217 Abacus reads 263457

14 2 6

The product of 2 and 7 The product of 2 and 1 The product of 2 and 3

Abacus reads 1----- then 203457 Abacus reads 205457 Abacus reads 205--- then 206057

The answer is the product 206057.

**Manipulation of the decimal point
**

We present here a simple method for manipulating the decimal point which basically stems from arithmetic, rather from abacus tradition. Since we are using base10 it is just a matter of noting how many decimal significant places there are in both the multiplicand and multiplier. Our reason for doing this is itself simple. The abacus does not physically dictate where the last integer and first decimal digits are. At the end of the day, the calculations produce a bead representation of the final product or result. All that is needed if there were decimals involved would be ascertaining where the decimal point should be placed. A decimal number contains three parts, the integer part (I), the decimal point (.), and finally the decimal part (D) or I.D Supposing we had a number which can be represented by III.DDDDD (e.g. 123.12345 ). We see that the integer part contains three digits. The decimal part contains five digits. When placing the number onto the abacus, we cannot indicate the decimal point in anyway. We would have just loaded in 12312345 as in the example above.

**Integer multiplied by a Decimal
**

When multiplying a decimal with an integer, we notice what often happens is that there will be the same number of decimal places to the decimal significant figure in the product as the decimal multipliying number.

E.g. Integer Multiplicand 723 12 817 Decimal Multiplier x 123.12345 = x 1.333 = x 1.1 = Decimal Product 89018.25435 15.996 898.7

Five decimal significant places Three decimal significant places One decimal significant place

What we do is first work out the number of total digits there are, then count the number of rods to take account of the maximum number of digits there are as if the multiplication were done on two integers. Once the product is obtained, the number of decimal significant places is counted from the right of the abacus and you have the answer you seek.

**Decimal multiplied by another Decimal
**

When two decimals multiply, the number of decimal places to the decimal significant figure is indicated by how many decimal digits in total there are in both numbers.

E.g. Decimal Decimal Multiplicand Multiplier 723.3 x 123.12345 = 12.217 x 1.333 = 817.12378 x 1.1 = Decimal Product 89055.191385 16.285261 898.836158

Six decimal significant places Six decimal significant places Six decimal significant place

A similar process is used; counting all the digits and making space on the abacus, obtaining the product before noting how many decimal significant places in the original multipliers (the multiplicand and multiplier) before counting from the right of the abacus to find the correct positioning of the decimal point on the abacus representation.

Example

• •

27.42 x 3.91 Our example problem 2742 x 391 Consider these two numbers as integers. There are seven digits in total, four in 2742 and three in 391

Abacus is loaded thus The The The The product product product product of of of of 1 1 1 1 and and and and 2 7 4 2 Abacus reads 391**** ++ Abacus reads 3902*** Abacus reads 39027** Abacus reads 390274* Abacus reads 3902742

2742 391*** *2 *7 *4 *2

18 63 36 18 *6 21 12 *6

The The The The The The The The

product product product product product product product product

of of of of of of of of

9 9 9 9 3 3 3 3

and and and and and and and and

2 7 4 2 2 7 4 2

Abacus Abacus Abacus Abacus Abacus Abacus Abacus Abacus

reads reads reads reads reads reads reads reads

+ 3182742 3245742 3249342 3249522 0849522 1059522 1071522 1072122

The answer is the product 1072122. However, we must now put back in the number of decimal significant places in our original problem (27.42 x 3.91) We note that there are four decimal significant places in total, therefore, we must count four digits from the right to obtain the correct answer : 107.2122 = 27.42 x 3.91 There is one special case worth mentioning. That is when one or both numbers has zero as the integer part of the decimal. In this case, use only the significant numbers in the multiplication, and then follow the method for finding the correct position of the decimal point as usual.

0.028 x 0.11 28 x 11 = 308

There are five decimal significant places, therefore the answer is 0.00308 = 0.028 x 0.11 as the correct answer. That's all there is to decimal multiplication. The section on division can be found in Page 8.

**The Chinese Abacus 8
**

Division

Division manipulation on the abacus can also leave the matter of the decimal places until last. We introduce the method first for integral numbers, then show how the decimal division can be achieved.

Method

The DIVIDEND is the number which is to be divided. The DIVISOR is the number which dives the dividend.

The initial digit of the dividend (placed near the middle of the abacus) is compared with the initial digit of the divisor and the multiples are noted on the abacus someplace in the left most rods. This multiple is multiplied by the initial digit of the divisor and subtracted from the initial digit of the dividend, resulting in a partial dividend. The second digit of the divisor is then multiplied by the multiple and subtracted from the second digit of the partial dividend. (It is likely that this product of the multiplication is greater than the number represented by the first and second rods. If so, the initial multiple is reduced in value by one, and this new multiple multiplied by the initial digit of the divisor is added back to the initial digit of the dividend, and this forms our first revised partial dividend. This way, the revised multiple multiplied by the second digit of the divisor may be small enough to be subtracted from the the left most rods of the dividend. All will become apparent in the example below) This same process continues for each digit of the divisor until the last digit for the first multiple. If there are any remaining digits to the dividend, the whole process is repeated for the second digit of the multiple. Once there are no digits left in the dividend, the answer will be all the multiple digits found.

•

**3804 / 12 = ? Our first example.
**

3804 -3 0804 -6 204 The first digit of the divisor is 1, 1x3=3 .: take 3 from first digit is the first partial dividend The second digit of the divisor is 2, 1x3=6 .: take six from second digit is the second partial dividend

3

32

-2 The divisors's first digit yield a factor of 2 (the multiple) .: take 2 004 is the third partial dividend 32*********** -4*** The divisor's second digit is 2, multiply by the multiple 2 to yield 4 Problem!! Can't take 4 off so backtrack -1 104 Remove 1 from the multiple, and multiply it by the first divisor digit So we arrive at this revised partial dividend 31 -2 Therefore most recent digit of the multiple, 1, is multiplied by the second digit of the dividor 2, and taken away 84 is our next partial dividend. 318 -8 The first digit of the divisor, 1, has a multiple of 8 so subtract 8 04 is our next partial dividend.

318********** -16** The divisor's second digit is 2, so 8x2=16, subtract 16 Problem again! Backtrack -1 14 And recover 1x1 (first divisor digit * recovered multiple) 317 14 is our revised partial dividend -14 The second digit of the divisor is 2, therefore, 7x2=14, subtract 14 00 This leaves nothing on the abacus.

Our answer is therefore 317

**Manipulating decimal division
**

We can force a decimal divisor into an integral on just by scaling by moving the decimal point. However, the number and direction of places the decimal point moves to achieve this must be matched also by the other number to balance out the effect of this change. E.g.

( 462.1 / 3.9 ) = ( 4621 / 39 ) = ( 46210 / 390 ) = ( 46.21 / 0.39 ) = ( 4.621 / 0.039 )

And so on. We note that each differs by factors of 10 from each other, because of the way we have move the decimal point. Once one number has been changed, the other must follow in the same manner. This keeps the ratios constant and allows us to manipulate the numbers before we load them onto the abacus. To some, this is not very satisfactory, because we may have two numbers with virtually endless number of decimal significant places. To ascertain the correct location of the decimal requires one to think about the number of digits and their decimal point positions.

11 11 11 11 11 11 11 220 220 220 220 / / / / / / / / / / / 11000 1100 110 11 1.1 0.11 0.011 11000 1100 110 11 = 0.001 = 0.010 = 0.100 = 1.000 = 10.000 = 100.000 = 1000.000 = = = = 0.02 0.20 2.00 20.00

220 / 1.1 220 / 0.11 220 / 0.011

= 200.00 = 2000.00 =20000.00

Scientific notation.

A note about scientific notation also helps understand the manipulation of numbers. Scientists and engineers like to represent the number 3200 as 3.2 x 103, that is read three point two times ten to the power of three. Quite often, they will multiply two numbers in scientific notation as follows: 62.42 x 0.0000019 = ( 6.242 x 101 ) x ( 1.9 x 10-6 ) = ( 6.242 x 1.9 ) x ( 10 [1 + (-6) ] ) = ( 6.242 x 1.9 ) x ( 10 -5 ) = ( 11.8598 ) x ( 10 -5 ) = ( 1.18598 x 10 1) x ( 10 -5 ) = ( 1.18598 x 10 [1 + (-5) ] ) = ( 1.18598 x 10 -4) = 0.000118598 Basically, in this notation, you can deal with powers of 10 immediately whilst leaving the more fiddly decimals as another bit of the problem. This way you can do the estimates to the answer quickly by rounding up the decimal parts. The indices or powers of 10 that you see as superscript are added in the case of multiplication. Decimals which are numerically lower than 1.0 are shown with a negative power of ten. In division a similar trick is employed. However, in division, the indices are subtracted in the following way. Using the same numbers but changing to division: 62.42 / 0.0000019 = ( 6.242 x 101 ) / ( 1.9 x 10-6 ) = ( 6.242 / 1.9 ) x ( 10 [ 1 -(-6) ] ) = ( 6.242 / 1.9 ) x ( 10 7 ) =

( 3.285263157895 ) x ( 10 7 ) = 32852631.57895 Like scientific notation, you must have a close look at the powers of tens involved when you have divided by integer translations of the decimals. So if you had 22.111 / 0.024 and you entered the dividend in as 22111 and used 24 as the divisor, be sure to cound the number of powers of ten you deal with either side of the decimal point. The list of of examples of integers divided by various powers of ten of a divisor is above for reference.