The Student Room Group

0.99999 recurring does not equal 1

Scroll to see replies

1/3=0.3333 recurring 3/3=0.9999recurring if you had 3 thirds of something you would have 1 something the fact 0.9999recurring=1 is simply a fault in our number system but it is true.
Reply 21
Original post by Person1001
All solutions that I have seen only demonstrate a 'limit towards 1 answer'. They do not demonstrate that 0.9999 recurring is exactly 1.


If you don't accept limits as a reasonable answer then why exactly do you accept that 0.99… exists as a number at all?

Original post by Stinkum
How can they be the same if they're different?


In the same way that 1.000000 is the same as 1, they're just different representations of the same thing.
Original post by Person1001
Only by 'definition'.

But definition is all that matters, if nothing is defined then you can't categorically say that anything is true
Original post by Arithmeticae
0. \dot{9} = \displaystyle\sum_{r=1}^{r= \infty} 0.9 \cdot 0.1^{r-1}
and from there on it's just a matter off using the standard GP formula

Posted from TSR Mobile

As I have mentioned earlier. The method relies on 'limits' and not equivalence.
Original post by Stinkum
Actually...I'll concede that I'm not qualified to comment on this topic because my knowledge of maths doesn't go beyond A2 Level, and it's not something I've ever encountered.


Ah-ha so if you have studied A Level maths you will know how to find the sum to infinity of a geometric series.

If we wanted to sum 910+9100+91000...\frac{9}{10} + \frac{9}{100} + \frac{9}{1000} ... we would note that this is a geometric series with first term a=910a=\frac{9}{10} and common ratio r=110r=\frac{1}{10}.

Now substitute these numbers into the formula S=a1rS_{\infty} = \frac{a}{1-r}.
Original post by Stevo F
But definition is all that matters, if nothing is defined then you can't categorically say that anything is true


I'm okay with people saying that they are equivalent by definition.

If however, from other agreed axioms they say they are equivalent then I an issue with that.
Original post by Person1001
As I have mentioned earlier. The method relies on 'limits' and not equivalence.


What's wrong with taking limits? We can define 0.999999 is by taking limits, so it makes sense that limits should come up in your proof

We can also prove it by using the construction of the real numbers (slightly beyond my level so I probably won't be able to explain, but IIRC it involves taking Dedekind cuts)

Posted from TSR Mobile
(edited 9 years ago)
Original post by Arithmeticae
0. \dot{9} = \displaystyle\sum_{r=1}^{r= \infty} 0.9 \cdot 0.1^{r-1}
and from there on it's just a matter off using the standard GP formula

\displaystyle\sum_{r=1}^{r= \infty} 0.9 \cdot 0.1^{r-1} = \dfrac{0.9}{1-0.1} = \dfrac{0.9}{0.9} = 1
That's clever, I've only ever seen the multiply-by-ten-and-subtract proof before.
Reply 28
Original post by Mr M
Ah-ha so if you have studied A Level maths you will know how to find the sum to infinity of a geometric series.

If we wanted to sum 910+9100+91000...\frac{9}{10} + \frac{9}{100} + \frac{9}{1000} ... we would note that this is a geometric series with first term a=910a=\frac{9}{10} and common ratio r=110r=\frac{1}{10}.

Now substitute these numbers into the formula S=a1rS_{\infty} = \frac{a}{1-r}.


Had you asked me this question in 2008/2009, I would have answered it without hesitation...unfortunately, I've completely forgotten that part of the curriculum, I have no idea how to solve that! I would be interested in the answer.

I had a brief look at the Wikipedia page for this topic and there are some very basic algebraic and arithmetic summaries which clearly show that the two numbers are equal, so I'm convinced that that's the case (it's impossible to dispute, therefore it must be true). It's so strange though...my immediate instinct was to reject that, but as you said, there is no number between 0.9 recurring and 1...it's all quite odd.
(edited 9 years ago)
Original post by Arithmeticae
What's wrong with taking limits? We can define 0.999999 is by taking limits, so it makes sense that limits should come up in your proof

We can also prove it by using the construction of the real numbers (slightly beyond my level so I probably won't be able to explain, but IIRC it involves taking Dedekind cuts)

Posted from TSR Mobile


Limit methods don't prove equivalence. A limit to a number as used all over the place in maths is as good as saying "workably equivalent".
Original post by Person1001
As I have mentioned earlier. The method relies on 'limits' and not equivalence.


Given that the way any real number (Such as your 0.9 recurring) is defined is as the limit of a sequence of rationals, your objection doesn't really make sense. As has been mentioned, your assumption that real numbers exist relies on limits.

However, I have a method which (whilst implicitly using limits) does not require taking one. Let's turn the question on its head: What is the absolute difference between the two? More specifically, what is:

10.999 |1-0.999 \cdots| ?

As this is an absolute difference, we clearly have 10.9990 |1-0.999 \cdots| \geq 0 . It cannot be greater than 0; suppose the difference is some ϵ>0 \epsilon > 0 , then truncating the recurring decimal at the nth decimal digit leaves a difference of 10n 10^{-n} which for sufficiently large n we can take smaller than ϵ \epsilon . Thus the only possibility is the absolute difference is 0, thus the two are the same.

Couple of addendums:

1) In particular, take n>logϵlog10 n>\frac{-log\epsilon}{log10}

2) The "implicitly uses limits" part is that we can take 10n 10^{-n} as close to 0 as we like, giving it limit 0. You never actually need to take a limit though; you just see the difference between the two is 0 \geq 0 and any non-zero difference fails, leaving 0 as the only option.
(edited 9 years ago)
Original post by Stinkum
Had you asked me this question in 2008/2009, I would have answered it without hesitation...unfortunately, I've completely forgotten that part of the curriculum, I have no idea how to solve that! I would be interested in the answer.

I had a brief look at the Wikipedia page for this topic and there are some very basic algebraic and arithmetic summaries which clearly show that the two numbers are equal, so I'm convinced that that's the case (it's impossible to dispute, therefore it must be true). It's so strange though...my immediate instinct was to reject that, but as you said, there is no number between 0.9 recurring and 1...it's all quite odd.


You just need to substitute into the formula I provided! I'm sure you can still do that.

Anyway here's an easier one.

x=0.9999999...x = 0.9999999 ...

Multiply by 10:

10x=9.9999999...10x=9.9999999 ...

Subtract x:

9x=99x = 9

Divide by 9:

x=1x=1

It's not a trick!
Original post by Stinkum
Actually...I'll concede that I'm not qualified to comment on this topic because my knowledge of maths doesn't go beyond A2 Level, and it's not something I've ever encountered.


Basically:

1 - 1 = 0

We know this. It makes sense.

1 - 0.999 = 0.0001

Also simple and makes sense.

1 - 0.999... = 0.000...

You end up with recurring zeroes for infinity as you have recurring nines to infinity.
Original post by Stinkum
Actually...I'll concede that I'm not qualified to comment on this topic because my knowledge of maths doesn't go beyond A2 Level, and it's not something I've ever encountered.


It's not hard to understand. If you want some intuition think about what value you would give to 33 \dfrac {3}{3} ?

Considering that 13=0.333.... \dfrac {1}{3}=0.333....
Original post by ThatPerson
It's not hard to understand. If you want some intuition think about what value you would give to 33 \dfrac {3}{3} ?

Considering that 13=0.333.... \dfrac {1}{3}=0.333....


I suspect the OP will argue that three thirds of a cake isn't the same as a whole one because some of the filling always gets stuck to the blade of the knife when you cut it.
Original post by Person1001
I am nearing the end of my physics degree at a top university and I have not seen an error free proof that that 0.9999 recurring and 1.0 are one and the same.


﴾͡๏̯͡๏﴿ O'RLY?
ahh i remember when my maths teacher showed this to the class...*mindblown*
Original post by DJMayes
Given that the way any real number (Such as your 0.9 recurring) is defined is as the limit of a sequence of rationals, your objection doesn't really make sense. As has been mentioned, your assumption that real numbers exist relies on limits.

However, I have a method which (whilst implicitly using limits) does not require taking one. Let's turn the question on its head: What is the absolute difference between the two? More specifically, what is:

10.999 |1-0.999 \cdots| ?

As this is an absolute difference, we clearly have 10.9990 |1-0.999 \cdots| \geq 0 . It cannot be greater than 0; suppose the difference is some ϵ>0 \epsilon > 0 , then truncating the recurring decimal at the nth decimal digit leaves a difference of 10n 10^{-n} which for sufficiently large n we can take smaller than ϵ \epsilon . Thus the only possibility is the absolute difference is 0, thus the two are the same.

Couple of addendums:

1) In particular, take n>logϵlog10 n>\frac{-log\epsilon}{log10}

2) The "implicitly uses limits" part is that we can take 10n 10^{-n} as close to 0 as we like, giving it limit 0. You never actually need to take a limit though; you just see the difference between the two is 0 \geq 0 and any non-zero difference fails, leaving 0 as the only option.


Not satisfied for the same reasons. We can take n sufficiently large but not ever large enough to get it to equal zero.
Original post by Learner4
﴾͡๏̯͡๏﴿ O'RLY?


Yup :smile:
Original post by Mr M
I suspect the OP will argue that three thirds of a cake isn't the same as a whole one because some of the filling always gets stuck to the blade of the knife when you cut it.


Haha. Just to be clear, despite the thread title, I am not arguing that the two are not equivalent. Rather, I have not met a proof that they are equivalent.

Quick Reply

Latest

Trending

Trending