Sorry your browser is not supported!

You are using an outdated browser that does not support modern web technologies, in order to use this site please update to a new browser.

Browsers supported include Chrome, FireFox, Safari, Opera, Internet Explorer 10+ or Microsoft Edge.

Geek Culture / Is 10 = 9.9999999... ??

Author
Message
Chris K
21
Years of Service
User Offline
Joined: 7th Oct 2003
Location: Lake Hylia
Posted: 29th Aug 2008 17:32 Edited at: 29th Aug 2008 17:33
Heh, that's a pretty tough problem to even understand fully.

If you fancy trying your luck at a great unsolved problem that is easy to understand, try the Goldbach Conjecture:

Quote: "Prove that every even number (except 2) can be written as the sum of two primes."


So simple; so, so, so hard.

If you prove that, you will be world famous.

-= Out here in the fields, I fight for my meals =-
mamaji4
21
Years of Service
User Offline
Joined: 24th Nov 2002
Location:
Posted: 29th Aug 2008 17:54
Lol. That's going to keep me awake a lot of nights. Is there any moolah attached to it?
Chris K
21
Years of Service
User Offline
Joined: 7th Oct 2003
Location: Lake Hylia
Posted: 29th Aug 2008 17:56
There used to be a $1,000,000 prize but I think it has expired.

You would earn a lot of money from proving it though, from interview fees etc.

-= Out here in the fields, I fight for my meals =-
Hobgoblin Lord
19
Years of Service
User Offline
Joined: 29th Oct 2005
Location: Fall River, MA USA
Posted: 29th Aug 2008 23:46
Quote: "1/3 = 0.33333...
3 x 1/3 = 1
3 x 0.33333... = 0.99999...
0.99999... = 1"


absolutely interesting, however this is why a grocery store that sells things 3 for a dollar sells you the first one for 34 cents and the other 2 for 33 cents. 1/3 being an infinite decimal we have no way to accurately notate 1/3 in decimal form, we will never reach an end. It is sort of like if you had a doorway 10 feet away and walked a third of the way there and then a third of the remaing distance and so on you would never pass through the doorway you would keep inching closer but never arrive

Benjamin
21
Years of Service
User Offline
Joined: 24th Nov 2002
Location: France
Posted: 29th Aug 2008 23:51
I'm not 99.999...% sure but I think we covered something similar here already.

Mahoney
16
Years of Service
User Offline
Joined: 14th Apr 2008
Location: The Interwebs
Posted: 30th Aug 2008 00:34
Quote: "1/3 being an infinite decimal we have no way to accurately notate 1/3 in decimal form, we will never reach an end."


Quote: "I'm not 99.999...% sure but I think we covered something similar here already."


We did. But, in response to H. Lord, you have to realize that the decimal equivalent of 1/3 being infinite has no effect on being able to calculate 0.333... x 3

0.333... x 3 would indeed equal 0.999... because, well, let me represent this in C++.



While far-fetched for C, I hope you get my point. The array "result" would (assuming an infinite array) be nothing but 9's. My point being, 0.333... x 3 = 0.999...

Windows Vista Home Premium Intel Pentium Dual-Core 1.6 Ghz 1GB DDR2 RAM GeForce 8600GT Twin Turbo
Hobgoblin Lord
19
Years of Service
User Offline
Joined: 29th Oct 2005
Location: Fall River, MA USA
Posted: 30th Aug 2008 01:03
sorry missed that

Seppuku Arts
Moderator
20
Years of Service
User Offline
Joined: 18th Aug 2004
Location: Cambridgeshire, England
Posted: 30th Aug 2008 01:38
Wasn't the same person that deduce that 1=0.999~ the inventor of the decimal point? Sounds like to me that he's just excusing the fact he could make the equivalent of 1/3 properly and had to make up any old rubbish to promote the decimal points. That's what I think.




Of course for anybody who jumps on that as idioticness fails to recognise my twisted humour.

You sir have the moral ambivalence of a mutated shrimp!
Aaron Miller
18
Years of Service
User Offline
Joined: 25th Feb 2006
Playing: osu!
Posted: 31st Aug 2008 01:28
Quote: "Prove that every even number (except 2) can be written as the sum of two primes."

What's so hard about that?

From the wikipedia page:
Quote: " 4 = 2 + 2
6 = 3 + 3
8 = 3 + 5
10 = 3 + 7 = 5 + 5"


Cheers,

-naota

I'm not a dictator to those that do stuff for me by will. Only those who don't.
dark coder
22
Years of Service
User Offline
Joined: 6th Oct 2002
Location: Japan
Posted: 31st Aug 2008 01:55 Edited at: 31st Aug 2008 01:55
Quote: "What's so hard about that?"


Maybe because 6, 8 and 10 aren't every number that fits this criteria? You can't just list 3 and claim this to be true, you'd need a formula that works for any arbitrary even number other than 2.

Aaron Miller
18
Years of Service
User Offline
Joined: 25th Feb 2006
Playing: osu!
Posted: 31st Aug 2008 03:50
I listed 4. And I listed them from Wikipedia. I didn't realize a FORMULA was what was meant.

Cheers,

-naota

I'm not a dictator to those that do stuff for me by will. Only those who don't.
Green Gandalf
VIP Member
19
Years of Service
User Offline
Joined: 3rd Jan 2005
Playing: Malevolence:Sword of Ahkranox, Skyrim, Civ6.
Posted: 31st Aug 2008 12:25
Doesn't have to be a formula. "Just" logic.

Quote: "Quote: "Prove that every even number (except 2) can be written as the sum of two primes.""


Where in your post have you shown that the result is true for every even number greater than 2? A lack of knowledge of maths is not an excuse for failing to understand plain English.

This is certainly an intriguing result if true. I wonder if there are extensions to numbers divisible by 3, or by 5, etc?
Diggsey
18
Years of Service
User Offline
Joined: 24th Apr 2006
Location: On this web page.
Posted: 31st Aug 2008 12:33 Edited at: 31st Aug 2008 13:25
9.9999... IS 10

All the people who have tried to disprove it have treated it as though there was a finite amount of 9s after the point. There are an infinite number, and so there will always be another 9 to match the one before:

Math89 said:
Let a = 9.99
10*a = 99.9
10*a-a = 89.91 and not 90

When actually, it would be more like this (collapsing infinite to two decimal places:
Let a = 9.99...
10*a = 99.99...
10*a-a = 90

Another more common sense proof:

What is 10/3?
Answer: 3.333...

What is 9.999.../3?
Answer: 3.333...

So:
10/3 = 9.999.../3
10 = 9.999...


And another:
1/9 = 0.111...
2/9 = 0.222...
3/9 = 0.333...
4/9 = 0.444...
5/9 = 0.555...
6/9 = 0.666...
7/9 = 0.777...
8/9 = 0.888...
9/9 = 1 (or 0.999... they are the same)

And another:
0.000... is obviously equal to 0 (however many 0s you put on the end, it's still 0)

1-9.999... = 0.000...
1-9.999... = 0
1 = 9.999...


Etc. (there are loads more proofs)

[b]Yuor signutare was aresed by a deslyxic mud...
BOX2D V2 HAS HELP FILES! AND A WIKI!
Chris K
21
Years of Service
User Offline
Joined: 7th Oct 2003
Location: Lake Hylia
Posted: 31st Aug 2008 12:55
Quote: "This is certainly an intriguing result if true"


It's true for the first few billion, billion numbers... it's just those remaining infinity they are still working on

-= Out here in the fields, I fight for my meals =-
Chris K
21
Years of Service
User Offline
Joined: 7th Oct 2003
Location: Lake Hylia
Posted: 31st Aug 2008 12:56
@ Diggsey

None of those proofs are valid.
I mean, they are just a matter of definition.

-= Out here in the fields, I fight for my meals =-
Diggsey
18
Years of Service
User Offline
Joined: 24th Apr 2006
Location: On this web page.
Posted: 31st Aug 2008 14:07 Edited at: 31st Aug 2008 14:09
And what's that supposed to mean?

Here is another proof, that is unarguably true:
You have two cars driving towards each other at 10km/h

When they are exactly 20km apart, a fly flies in front of one of the cars. To escape getting hit by the car, it flies away from the car (towards the other) at 40km/h. When it meets the other car, it instantly turns round and flies away from that car instead, and so on until the cars meet and it is squashed.

If each car is travelling at 10km/h, from 20km away, it will take exactly 1 hour for them to collide, squashing the fly in between.

The question is, how far did the fly travel?

FIRST SOLUTION:
The relative velocity of the fly and the opposite car is 50km/h:
40km/h + 10km/h = 50km/h, so the fly met the car after 24 minutes.

In 24 minutes, the fly would travel 16km
The cars would by then be only 12km apart

The fly would turn, and travel back towards the first car, and the two would have a relative velocity of 50km/h just like with the other car. This time, the fly would meet the car after only 14.4 minutes, and travel only 9.6km

As you can see, each time, the fly only travels 0.6 of its previous distance: 16, 9.6, 5.76, 3.456, etc.

This sequence will go on forever, because you will never reach 0.
If you add up the first few distances, you get 34.816.

2.0736
1.24416
0.746496
0.4478976
0.26873856
0.161243136
0.0967458816

With these next ones, you get 39.8548811776.
As you can see, it approaches 40. If you continue the sequence to infinity, the total will be 39.999... I'm sure you will agree with that.

But wait, what about the second solution???


SECOND SOLUTION:
As we already worked out, the cars will collide after exactly 1 hour. The fly is travelling at exactly 40km/h the whole time (albeit in different directions). 40km/h for one hour means the fly travelled exactly 40km


The two solutions: 39.999...km and 40km. Both are correct. 39.999... has the same value as 40


An easy to understand way of how they can be equal:
The decimal number system is created by us to represent actual numbers. These numbers were not invented by us, but are fundamental laws in the universe. The ACTUAL number which we represent as both 40 and 39.999... is a single number, it's just that the decimal number system provides two ways of representing it.

[b]Yuor signutare was aresed by a deslyxic mud...
BOX2D V2 HAS HELP FILES! AND A WIKI!
Little Bill
16
Years of Service
User Offline
Joined: 18th Jun 2008
Location: UK
Posted: 31st Aug 2008 16:12
Surely 10 is 10 and 9.999 recurring is just a number incredibly close to 10.

Talking to someone about this they said if you had 9 and 10 and "keep chopping the pieces up there will eventually be nothing left". But we a talking about numbers, not chocolate bars, so it would go forever. You can not make 10 from 9.999 recurring.

Windows XP Home Edition SP3 - Intel(R) Pentium(R) D CPU 2.80GHz (2 CPUs) - 2GB RAM - NVIDIA GeForce 7300LE
Alucard94
17
Years of Service
User Offline
Joined: 9th Jul 2007
Location: Stockholm, Sweden.
Posted: 31st Aug 2008 16:21
Is there even any point to this thread? All the people in this thread won't ever just decide on the same side because of this thread, so what's the point of trying to prove it?


Cash Curtis II
19
Years of Service
User Offline
Joined: 8th Apr 2005
Location: Corpus Christi Texas
Posted: 31st Aug 2008 16:51
It had a point. Some people learned something from it. Others are continuing to post the same mathematical opinions in an effort to disprove something that they aren't comfortable with.

The mathematical proof is absolutely conclusive. Those who disagree with it are saying that although there is a sound logical proof that explains it they are rejecting it on the basis that they don't feel it is right. Or, that they simply haven't looked at it.

Then you've assumed that you are smarter than every mathematician and professor and teacher that has come before you and you have so easily found the answer to a centuries old problem that has already been solved. You shouldn't accept something at face value because countless experts have agreed. It does however lend enough credibility to any concept to be open to that same possibility.

The bottom line is that math isn't about feelings, it's about logic and proofs. You have to open your mind to the concept of infinity which is very different from 9.999...


Come see the WIP!
Mahoney
16
Years of Service
User Offline
Joined: 14th Apr 2008
Location: The Interwebs
Posted: 31st Aug 2008 19:06
Quote: "@ Diggsey

None of those proofs are valid.
I mean, they are just a matter of definition."


Quote: "And what's that supposed to mean?"


Diggsey, if you scroll up, he's already agreed that 0.999... = 10. He's just saying that those proofs aren't valid.

Quote: "For the record, you can't do normal arithmetic with recurring decimals, so the intuitive 'proofs' which are normally produced to settle this debate are in fact, not valid.
They do however often convince people of the right answer, even if in the wrong way."


Windows Vista Home Premium Intel Pentium Dual-Core 1.6 Ghz 1GB DDR2 RAM GeForce 8600GT Twin Turbo
Chris K
21
Years of Service
User Offline
Joined: 7th Oct 2003
Location: Lake Hylia
Posted: 31st Aug 2008 20:16
Cheers, Mahoney.

The crux of it is, you would have to strictly define what you mean when you manipulate recurring decimals.

What's 0.666... + 0.666... ? Well, our normal rules of adding decimals don't apply, because we can't start 'on the right'.

You could argue, well clearly it is 1.333..., but that is not Maths, unfortunately.

-= Out here in the fields, I fight for my meals =-
Aaron Miller
18
Years of Service
User Offline
Joined: 25th Feb 2006
Playing: osu!
Posted: 31st Aug 2008 22:17 Edited at: 31st Aug 2008 22:22
@Chris K
I'm not arguing, but if I add 0.6666666666 (It's finite, I understand) and 0.6666666666 I get 1.3333333332. (I did so in the Calculator program that comes with Windows). Why is this? EDIT Someone argued that there could not be an infinite series of decimals and an ending decimal, but if 0.6~ was added to 0.6~ then logically wouldn't the correct answer be 1.3~2? I mean, it's not, but wouldn't it be, logically? EDIT 2 What I mean by that is wouldn't there have to be an accepted last digit at the end of the infinity (I see the flaws in that statement). For example, if you add it on paper and keep on adding until finally you decide to stop, where you stop wouldn't that be that 2 (the last digit)? I just want to clarify this since you clearly know more math than I do.

@Diggory
From what I understand the universe doesn't use numbers or any form of mathematics. I thought that mathematics were used to represent the workings of the universe. I don't think your comparison to the universe is valid, but I'm not going to argue with it either. I'm just wondering if you meant it like that or not?

@Cash Curtis
If the media accepts only one president does that mean I should accept that president too? If a large amount of scientists and mathematicians support only one answer I should support it too? Well, I do support that 0.999~, as a decimal, can be used to represent 1.0 - but only for lack of better representation. I think the system is smart indeed, but if two numbers have to be used to represent one, I don't think that's smart or efficient. Do you?

(I'm fairly certain I just pissed off a lot of people, however I didn't actually argue with anything, I only asked questions).

Cheers,

-naota

I'm not a dictator to those that do stuff for me by will. Only those who don't.
Cash Curtis II
19
Years of Service
User Offline
Joined: 8th Apr 2005
Location: Corpus Christi Texas
Posted: 31st Aug 2008 23:13 Edited at: 31st Aug 2008 23:18
Quote: "If the media accepts only one president does that mean I should accept that president too?"

That couldn't be more off topic if you'd planned it out all night. Accepting a president is a matter of preference and opinion. Accepting mathematical concepts is not. Once again you've confused logic and feeling. Disprove the proof on the Wikipedia article and then we'll talk.

Quote: "but if I add 0.6666666666 (It's finite, I understand) and 0.6666666666 I get 1.3333333332. (I did so in the Calculator program that comes with Windows). Why is this?"

Windows calculator is irrelevant. .6666666666~+.6666666666~ is not 1.3333333332, it's 1.3333333333~. Why? Because (2/3)+(2/3)=1 1/3, which is 1 followed by a decimal point and infinite 3s. Windows Calculator can no more influence this discussion than Windows Notepad choose your presidential candidate.

I'll simplify this for you. 1/3 = 0.333333333333~. 1/3 * 3 = 1. Therefore .33333333333~ * 3=1, not 0.999999999~


Come see the WIP!
Chris K
21
Years of Service
User Offline
Joined: 7th Oct 2003
Location: Lake Hylia
Posted: 1st Sep 2008 00:47 Edited at: 1st Sep 2008 00:48
I think the clearest I can explain, is that 'dot dot dot' is not a mathematical construct.

It doesn't mean anything. What does it mean to say a zero followed by a dot followed by an infinite amount of nines.

These are questions a human cannot answer.

It is just a convenient stop gap that people are taught when they are teenagers because it is quite easy to understand.

The concept of 0.9dotdotdot is not mathematically sound. This is why when mathematicians ACTUALLY talk about infinite sequences it is properly defined and absolutely clear.

-----

It is not pretty, but when someone says 0.999... = 1, what they mean to say is (deep breath)

Quote: "
For ANY tiny positive real number you give me, I can find a point in the sequence S so that after that point, EVERY term is at most your tiny number away from 1.

Where the nth term of the sequence S is found by adding up the first n terms of the sequence T

Where the nth term of the sequence T is 9/10^n."


If you think you are not claiming that, and you are actually claiming 'a zero then infinite nines' then you are not discussing Maths.

-= Out here in the fields, I fight for my meals =-
ionstream
20
Years of Service
User Offline
Joined: 4th Jul 2004
Location: Overweb
Posted: 1st Sep 2008 01:00
0.9~ is the limit statement though. The limit as n approaches infinity of the sum of k=1 to n of 9/(10^k) equals 1, and that's all that 0.9~ is saying.

Chris K
21
Years of Service
User Offline
Joined: 7th Oct 2003
Location: Lake Hylia
Posted: 1st Sep 2008 01:17
Exactly, so there is no reason to think you can perform arithmetic with it like a number.

-= Out here in the fields, I fight for my meals =-
Seppuku Arts
Moderator
20
Years of Service
User Offline
Joined: 18th Aug 2004
Location: Cambridgeshire, England
Posted: 1st Sep 2008 01:56
I guess it's a matter of dealing with infinity - a concept our finite minds cannot understand wholly. And of course 0.999~ appears infinitely smaller, but the difference between 0.999~ and 1.0 cannot happen because if you subtract 0.999~ from 1.0 you get 0.000~ with an infinite number of zeros and that '1' doesn't exist because you can never reach infinity so 0.999~ has to logically be the same as 1.0? This is the way my feeble un-mathematic mind seems to understand this, but am I right or wrong?

You sir have the moral ambivalence of a mutated shrimp!
mamaji4
21
Years of Service
User Offline
Joined: 24th Nov 2002
Location:
Posted: 1st Sep 2008 02:33 Edited at: 1st Sep 2008 02:56
For any natural number x, x > 0, the multiplicative inverse of x is defined as 1/x
i.e. If you multiply x by 1/x you get the identity over the multiplication operation
Let us use an accepted notation for representing a recurring decimal as 0.9~ where ~ is the bar notation for recurring decimal representation.
We further need to work with the number 0.9~ to be able to study its mathematical properties. We can't do that with the bar notation, so we represent it as 0.999...
This helps us visualise the recurring decimal as the sum of a geometric series, and using results from the theory of limits we arrive at some conclusion about the value of this number.
So it does appear that the representation 0.999... does have its practical merits which the bar notation does not have.

Now if we multiply 0.999... by the multiplicative inverse of 3
i.e. 1/3 * 0.999...
we have multiplied the number 0.999... without using the conventional high school right to left method, but instead using a left to right method. Imagine. We have now changed the convention for multiplying a number by starting from the left because multiplication by the multiplicative inverse of a number is essentially division by that same number.
So if we do this we see that
1/3 * 0.999...
= 0.999.../3
and since the operation for division moves from left to right we can obtain the value of this as 0.333...
i.e. 0.999.../3 = 0.333...
0.999... = 3*0.333... --- EQUN. 1
But we know that
1/3 = 0.333...
3 * 1/3 = 1
3 * 0.333... = 0.999... from EQUN. 1
0.99999... = 1

Hey! Guess what. We didn't do anything mathematically unsound, we did not use the theory of limits(well established) on the recurring decimal 0.9~ and we actually proved 0.9~ = 1 --- EQUN.2

Now whether EQUN.2 is true using the well established limit of a sequence defintion, is a story for another day.

I'm sorry Chris. I just had to do it.
Actually I returned to this thread to tell you that I had reduced the 'strong' Goldbach Conjecture. However the reduced problem still needs to be proved for all n.
It however does give an algorithm for finding the prime q for any prime p and any even number, such that the even number is the sum of the two primes p,q
I wanted your opinion on whether this is of any utilitarian value, since I have seen people using the twin prime constant to show this for large prime numbers.
I also haven't studied all the properties of prime numbers to claim that I can even attempt to prove the reduced problem.
In any case there seem to be several proofs floating around so I don't have much incentive to carry on.
Let me know what you think.
Aaron Miller
18
Years of Service
User Offline
Joined: 25th Feb 2006
Playing: osu!
Posted: 1st Sep 2008 02:50
Quote: "Windows calculator is irrelevant. .6666666666~+.6666666666~ is not 1.3333333332, it's 1.3333333333~. Why? Because (2/3)+(2/3)=1 1/3, which is 1 followed by a decimal point and infinite 3s. Windows Calculator can no more influence this discussion than Windows Notepad choose your presidential candidate.

I'll simplify this for you. 1/3 = 0.333333333333~. 1/3 * 3 = 1. Therefore .33333333333~ * 3=1, not 0.999999999~"

Wasn't directed at you Cash. I will thank you to read and respond only to what I direct towards you or towards every one in general. That is why I directed it at Chris K.

Quote: "That couldn't be more off topic if you'd planned it out all night. Accepting a president is a matter of preference and opinion. Accepting mathematical concepts is not. Once again you've confused logic and feeling. Disprove the proof on the Wikipedia article and then we'll talk."

For lack of better example I used something many people accept because other people accept it. Science is supposed to be continuously questioned, is it not? That's what makes it so valid. All I was implying is that the decimal system does not seem accurate, in my opinion. If it seems accurate to everyone else, that's fine. But in the same situation, people once thought the world was flat too. You didn't answer my question though, does it seem efficient or smart to represent one number using two numbers?


Cheers,

-naota

I'm not a dictator to those that do stuff for me by will. Only those who don't.
Cash Curtis II
19
Years of Service
User Offline
Joined: 8th Apr 2005
Location: Corpus Christi Texas
Posted: 1st Sep 2008 03:18 Edited at: 1st Sep 2008 03:23
Quote: "I will thank you to read and respond only to what I direct towards you or towards every one in general."

I'll respond to whatever I want to.

If you disagree with this concept then disprove the proof. Otherwise there is no valid argument.

Quote: "people once thought the world was flat too"

Another sweeping example that has nothing to do with this.

Quote: "does it seem"

What I feel has no place in math.


Come see the WIP!

Attachments

Login to view attachments
Mahoney
16
Years of Service
User Offline
Joined: 14th Apr 2008
Location: The Interwebs
Posted: 1st Sep 2008 05:23
Quote: "You didn't answer my question though, does it seem efficient or smart to represent one number using two numbers?"


You obviously didn't understand what he said.

Quote: "I think the clearest I can explain, is that 'dot dot dot' is not a mathematical construct.

It doesn't mean anything."


Quote: "It is just a convenient stop gap that people are taught when they are teenagers because it is quite easy to understand.

The concept of 0.9dotdotdot is not mathematically sound. This is why when mathematicians ACTUALLY talk about infinite sequences it is properly defined and absolutely clear."


Read: 0.999... is just the dumbed-down representation of a much deeper concept.

Windows Vista Home Premium Intel Pentium Dual-Core 1.6 Ghz 1GB DDR2 RAM GeForce 8600GT Twin Turbo
Chris K
21
Years of Service
User Offline
Joined: 7th Oct 2003
Location: Lake Hylia
Posted: 1st Sep 2008 12:07
@ Seppuku

Quote: "if you subtract 0.999~ from 1.0 you get 0.000~"


So you claim. How are you meant to 'subtract' this weird thing from a number?

I claim if you subtract a hexagon from three you get a simultaneous equation. It doesn't make me right.

When I started my course on Analysis the lecturer was just quiet for a second and then he said "If there is one thing you will learn in this course it is that INFINITY IS WEIRD".

That is why you must strive, strive, strive against the urge to apply anything that is remotely human. You can't go, "so clearly". EVERYTHING has to be VERY strictly defined or you will go wrong very quickly.

Let me give you an example...

Consider this infinite sum:
1 - 1/2 + 1/3 - 1/4 + 1/5 - 1/6 + 1/7 - 1/8 + 1/9 - 1/10 + ...

It converges to ln2, if you don't know what ln2 is it is just a fancy number (about 0.7).

We can rearrange that sum, so that we are adding up exactly the same numbers - ie. every term from the first series appears once and only once in the following series...

1 - 1/2 - 1/4 + 1/3 - 1/6 - 1/8 + 1/5 - 1/10 - 1/12 + 1/7 + ...

You can see this is still taking 'one over the even numbers' from 'one over the odd numbers'. But does it converge to about 0.7?

No!
I converges to (ln2)/2 or about 0.35...

ie adding in a different order produces a different answer. Why? Because this is INFINITY and we are HUMANS and we don't have a freaking clue about infinity!

-= Out here in the fields, I fight for my meals =-
Chris K
21
Years of Service
User Offline
Joined: 7th Oct 2003
Location: Lake Hylia
Posted: 1st Sep 2008 12:16
Quote: "It however does give an algorithm for finding the prime q for any prime p and any even number, such that the even number is the sum of the two primes p,q"


Um, that's impossible unfortunately.

For example, what about...

n = 26
p = 17

?

-= Out here in the fields, I fight for my meals =-
Green Gandalf
VIP Member
19
Years of Service
User Offline
Joined: 3rd Jan 2005
Playing: Malevolence:Sword of Ahkranox, Skyrim, Civ6.
Posted: 1st Sep 2008 12:48
I love counter-examples - they save so much time when dealing with plausible, but wrong, conjectures.

Trouble is they can take a long time to find (even if the conjecture is false ), your "sum of primes" conjecture being a prime example.
Green Gandalf
VIP Member
19
Years of Service
User Offline
Joined: 3rd Jan 2005
Playing: Malevolence:Sword of Ahkranox, Skyrim, Civ6.
Posted: 1st Sep 2008 12:51
Quote: "Another sweeping example that has nothing to do with this."


Just seen your image - have I just entered a parallel universe?
Seppuku Arts
Moderator
20
Years of Service
User Offline
Joined: 18th Aug 2004
Location: Cambridgeshire, England
Posted: 1st Sep 2008 13:23 Edited at: 1st Sep 2008 13:24
Quote: "So you claim. How are you meant to 'subtract' this weird thing from a number?

I claim if you subtract a hexagon from three you get a simultaneous equation. It doesn't make me right."




You're not playing on the semantics on this squiggly little thing:, '~', are you? Because I am using it to represent a recurring number. (which I would have hoped by now is obvious)


I agree with your infinite sentiment, as I said, it's difficult for humans to understand, especially when you deal with the finite.

But is one reason why 0.999 recurring = 1 is because if you subtract it from '1' you get the same result? (0 - with an infinite number of zeros behind a decimal point) If yes, then I believe I can understand the argument, if not, then it's not with in my mathematical understanding.

You sir have the moral ambivalence of a mutated shrimp!
RalphY
20
Years of Service
User Offline
Joined: 6th Sep 2004
Location: 404 (UK)
Posted: 1st Sep 2008 15:16
I think he was trying to say Seppuku that you can't apply normal subtraction to an infinite number as it's not defined (or something like that).

Quote: "How are you meant to 'subtract' this weird thing from a number?"

i.e. 1 is a number whereas infinity is a "weird thing" thus you cannot subtract infinity from a number. As said earlier when you say 0.999~ you don't mean a number where its 0.9(followed by an infinite number of 9s) but the statement given by Chris earlier in the thread.

Or at least that's my understanding of it.

Oh boy! Sleep! That's when I'm a Viking! | Super Nintendo Chalmers!
mamaji4
21
Years of Service
User Offline
Joined: 24th Nov 2002
Location:
Posted: 1st Sep 2008 15:45
Quote: "Um, that's impossible unfortunately.

For example, what about...

n = 26
p = 17

?
"


Yeah. I gave a very 'loose' formal definition of the reduced problem, because I don't want to go into the details of the correlation I have established between prime p and the even number.
"any even number" is strictly true.
"any prime p with a specific property", is strictly true, the property being what I discovered. I can't talk about it on the forums.

Do you think there might be some utilitarian value for an algorithm to COMPUTE q, given p and the even number?
Could you ask your professor, maybe, and get back to me?
Seppuku Arts
Moderator
20
Years of Service
User Offline
Joined: 18th Aug 2004
Location: Cambridgeshire, England
Posted: 1st Sep 2008 15:50
Quote: "I think he was trying to say Seppuku that you can't apply normal subtraction to an infinite number as it's not defined (or something like that)."


I see. That only makes matters more confusing.

You sir have the moral ambivalence of a mutated shrimp!
mamaji4
21
Years of Service
User Offline
Joined: 24th Nov 2002
Location:
Posted: 1st Sep 2008 16:14 Edited at: 1st Sep 2008 17:13
.
Cash Curtis II
19
Years of Service
User Offline
Joined: 8th Apr 2005
Location: Corpus Christi Texas
Posted: 1st Sep 2008 16:20 Edited at: 1st Sep 2008 23:46
.99999999999~ doesn't even exist. You can't express it as a fraction, while other repeating decimals can be. For example...

1/9=.1111111~
2/9=.2222222~
8/9=.8888888~

But how do you get .9999999~? You can't. You can come close, 8.9999999999/9. The only way it occurs is through rounding errors when dealing with infinitely repeating decimals. This is why the proof works like it does. At first I was uncomfortable with it, but when I realized that it's not an approximation, or another way of representing it, it's just that .999999999~ doesn't exist. It's the result of an intermediate math step that didn't quite go right. It's like doing algebra with square roots, you don't figure them before you cancel out like terms because you'll be left with decimal inaccuracies.

@Green Gandalf -
Quote: "Just seen your image - have I just entered a parallel universe? "

That was just to drive my point home...

That's actually a scene from Geisha House. I posted some new pictures in the WIP thread. I just like attaching them to posts that way in 100 years when my website has turned into stone all of the pictures will still show up.


Come see the WIP!

Attachments

Login to view attachments
mamaji4
21
Years of Service
User Offline
Joined: 24th Nov 2002
Location:
Posted: 1st Sep 2008 17:07
After re-reading my own post I just concluded that the Theory of Limits gives the EXACT value of the limit of a sequence.
Well that's just me. Just punched a hole in my own analysis, realised it is flawed and therefore arrived at different conclusion. When I said question everything, I menat it. Even your own conclusions.

So here is what I said wrong.
IF the guy had written
1/3 ~= 0.333... he would have actually been wrong, because
1/3 ~= 0.333...n for any finite number of n
But 1/3 = 0.333...n , n -> infinity is the EXACT value of 1/3

The limit theory had to be developed so we could work with 0.333...
Using the theory if we can show that 0.333... = 1/3 we have actually verified using the above empirical evidence that the theory of limits is valid and gives the EXACT value of the limit, not just an approximation.


I'm never satisfied. Not with my own proofs or with others.
Chris K
21
Years of Service
User Offline
Joined: 7th Oct 2003
Location: Lake Hylia
Posted: 1st Sep 2008 17:43
Quote: "Do you think there might be some utilitarian value for an algorithm to COMPUTE q, given p and the even number?
Could you ask your professor, maybe, and get back to me?"


Well, no the reason the problem is hard is because you don't know what either of the primes are.

If you knew what one of them was, then the other one would just be the even number minus the first prime!

Remember, to prove the conjecture you don't need to actually come up with a way of finding the two primes for every even number, you just have to show that these two primes exist.

-= Out here in the fields, I fight for my meals =-
mamaji4
21
Years of Service
User Offline
Joined: 24th Nov 2002
Location:
Posted: 1st Sep 2008 18:55 Edited at: 1st Sep 2008 19:14
Quote: "If you knew what one of them was, then the other one would just be the even number minus the first prime!
"


If I tell you the algorithm I developed does just that, would it be of utilitarian value, say in cryptanalysis?

Quote: "Remember, to prove the conjecture you don't need to actually come up with a way of finding the two primes for every even number, you just have to show that these two primes exist."

I read through some of the proofs and they do just that. Is it that none of them have been accepted by the mathematical community?
I was also wondering about the utilitarian value of establishing such a proof.

I wonder why people pay so much money for proving a conjecture and never any for developing algorithms related to those conjectures that can have some practical value?
Chris K
21
Years of Service
User Offline
Joined: 7th Oct 2003
Location: Lake Hylia
Posted: 1st Sep 2008 21:04 Edited at: 1st Sep 2008 21:06
Quote: "If I tell you the algorithm I developed does just that, would it be of utilitarian value, say in cryptanalysis?"


If you have an algorithm that can input an even number and output two prime numbers that sum to the input, then it is one of the most impressive discoveries of the last 100 years.

But you seem to be saying, you also need a prime. If you mean you need to know before hand one of the primes from the results, then that's, er, cheating.

Quote: "I read through some of the proofs and they do just that. Is it that none of them have been accepted by the mathematical community?"


Er... I've never seen a proof even submitted for review...? Link?

Quote: "I wonder why people pay so much money for proving a conjecture and never any for developing algorithms related to those conjectures that can have some practical value?"


Perhaps you should watch the lecture I provided earlier...
http://www.youtube.com/watch?v=BsIJN4YMZZo

It discusses why people should fund Maths even when it has no economics benefit.

-= Out here in the fields, I fight for my meals =-
Cash Curtis II
19
Years of Service
User Offline
Joined: 8th Apr 2005
Location: Corpus Christi Texas
Posted: 1st Sep 2008 23:46 Edited at: 1st Sep 2008 23:48
Quote: "If you have an algorithm that can input an even number and output two prime numbers that sum to the input, then it is one of the most impressive discoveries of the last 100 years."


Okay, if I'm understanding this right, that sounds pretty easy. Enter an even number greater than 4 and it returns two prime numbers that add up to the original number?

That's so easy that I just made a program that does it. I've attached it to this post. It's a DBP program so just un-rar it and run it, it's completely safe.

https://forumfiles.thegamecreators.com/download/1583979



So, I've determined this can't be the problem, that was very easy to do.


Come see the WIP!

Attachments

Login to view attachments
mamaji4
21
Years of Service
User Offline
Joined: 24th Nov 2002
Location:
Posted: 2nd Sep 2008 00:46
Quote: "So, I've determined this can't be the problem, that was very easy to do.
"


You've just used the seive to generate the primes and then iterated through the sequence of primes to find the two numbers that add up to the even number that was input.
That's fine for small numbers.
It doesn't work for large prime numbers of the order 10^14
That's the problem researchers are trying to address.
Chris K
21
Years of Service
User Offline
Joined: 7th Oct 2003
Location: Lake Hylia
Posted: 2nd Sep 2008 00:52
@ Cash Curtis

Mamaji4 got it; how do you know that your program works for every even number?

-= Out here in the fields, I fight for my meals =-
mamaji4
21
Years of Service
User Offline
Joined: 24th Nov 2002
Location:
Posted: 2nd Sep 2008 00:56
Quote: "Er... I've never seen a proof even submitted for review...? Link?
"


http://www.google.co.in/search?hl=en&q=goldbach+conjecture+proof&meta=

Quote: "But you seem to be saying, you also need a prime."

I have identified a set of primes p with specific characteristics, which would suffice to generate the prime q for any even number input.
You don't need the seive to generate any primes except the ones with the characteristic, so that reduces the computational time problem to a large extent, for very large primes.
Cash Curtis II
19
Years of Service
User Offline
Joined: 8th Apr 2005
Location: Corpus Christi Texas
Posted: 2nd Sep 2008 01:17 Edited at: 2nd Sep 2008 01:19
More or less, though not exactly.

What I'm doing is starting at the bottom then checking the primality of the second number. If it's not I continue my march up the list. The dependency of the test depends on the speed of the primality test. Is there a good way to test a huge number? If there isn't then it's impossible to do.

This works fairly well because the two primes are a low prime and a high prime, but how low and high doesn't follow any pattern.

This test works for every even number in the data type range. Try it yourself and see.

@mamaji4 -
Post a working example.


Come see the WIP!

Login to post a reply

Server time is: 2024-11-20 16:13:19
Your offset time is: 2024-11-20 16:13:19