ShmoopTube

00:00

Thank you We sneak And here's your shmoop du jour

00:05

brought to you by floating point variables there the point

00:09

Nine nine nine nine nine and an eight You can

00:11

count on what is the output of the following code

00:14

segment for a night of eating thai food Keep pressing

00:20

right And here your potential answers Okay true false got

00:25

it All right let's Go let's take a look at

00:27

what this code segment is trying to do First we

00:29

get two double variables The first being given the value

00:32

of ten and a second being given the value of

00:34

zero Then we start a loop The loop begins counting

00:37

with inger i zero halt when i stopped being less

00:41

than fifty and increases i buy one each time the

00:44

loop like once inside the loop will be adding point

00:48

to numb teo over and over and over again fifty

00:51

times total since that's how long our loop will run

00:55

afterwards we'll have three statements Print results of a few

00:58

operations were the first prints the tour false results of

01:01

whether numb one equals to the second print through her

01:04

false based on whether numb one divided by two equals

01:06

one same number and not zero And the third prince

01:11

true or false depending on whether numb one minus number

01:14

two equal zero again away to check if they're the

01:16

same number So let's try and predict the kinds of

01:19

results will get All right Back up to our loop

01:21

We've got numb to starting out equal to zero and

01:25

we add point two two it fifty times Well in

01:28

california fifty towns point two is ten so after the

01:30

loop is finished running canting treating tonight someone in them

01:35

too should both be tense And if we plug that

01:39

into our print statements we'd get numb One equals dumb

01:41

two or ten equals ten True can divide by ten

01:43

equals one true and minus ten equals zero True but

01:47

way Just hold your horses one sec If you got

01:50

your handy compiler you can pause the video here and

01:53

try running the code yourself and no seriously give the

01:55

shot will wait here's the code again and a picture

01:59

of a cold er Yesterday all my troubles seemed so

02:07

far ready All right So what did you find out

02:10

Well for those of you who didn't run the code

02:12

this is what happened Falls all falls which is our

02:16

answered the question by the way option he but wait

02:18

Why Tennis Ten right there's Not much That could possibly

02:22

go wrong with the statement Tennis Just ten even in

02:24

california with all you communists out there But i will

02:26

unless tennis somehow not ten Well let's add a line

02:30

to the end of the snippet to investigate a little

02:32

more closely We'll have it print the actual value of

02:35

numb to to make sure we're getting the ten that

02:37

we paid for system out print line none too And

02:41

go well ten is not ten it's Very very close

02:45

But it's just not in this case It's like nine

02:49

point nine nine nine nine nine six something that's why

02:52

our statements returned false and it's Not even a bug

02:55

issue here has to do with the very nature of

02:57

floating point arithmetic Those double variables we declared earlier and

03:01

more formally known as double precision floating point variables which

03:05

is to say is not quite as solid as an

03:07

imager it's More like an approximation of a real number

03:11

This has its uses when dealing with very complicated numbers

03:14

that don't quite need to be perfectly exact The distance

03:17

to the next galaxy over for example like that But

03:20

when you're dealing with the values that need to be

03:22

concrete like money do yourself a favor and compute using

03:25

interest or long otherwise you're going to be going people

03:29

One craw drill yin of a penny and your knives 00:03:32.22 --> [endTime] probably aren't that shark even against you

Full Transcript