MATH CHALLENGE

I see a lot of people here like to brag about their supposed intelligence and I want to put that to the test.

I have nothing fancy, nothing too niche or specific so that you don't need too much previous training to start. It is a problem completely about geometry and number theory, the literal pillars of elementary mathematics which means that if you can't solve this then you are not even close to being a genius. Lets go:

Lets set the stage. Consider the first quadrant of the plane (the one with only positive numbers) and consider just the integer points. Like (1,2) and (343,52) for example.

Definition 1: We say that a point is invisible if when you draw the line that goes from the origin to that point, that line crosses through at least another integer point.

Quick example: If you draw a line between the origin and the point (2,2) then you inevitably have to go through (1,1).

Theorem 1 (A little help you could use): If the numbers A and B are relatively prime then the point (A,B) is not invisible. And if a point (A,B) is not invisible then A and B are relatively prime.

Definition 2: We say that the point (A,B) forms an invisible square of length N if for each integer [math]i,j \in \{0,1,2,...,N \} [/math], the point (A+i,B+j) is invisible.

Quck example: The point (1308,1274) forms an invisible square of length 2. You can check for yourself that (1308,1274) is invisible, (1309,1274) is invisible, (1310,1274) is invisible, etc. if you want.

[math] \text{THE PROBLEM:} [/math]
If (A,B) is a point that forms an invisible square of length N, give your best lower bound for how big [math] min\{A,B\} [/math] has to be compared to N.

Easy example:
You can easily prove that [math] min\{A,B\} [/math] has to be at least bigger than N, by Bertrand's Postulate.

But you can give a much better lower bound. Go.

Other urls found in this thread:

mathworld.wolfram.com/VisiblePoint.html
en.wikipedia.org/wiki/Set-theoretic_topology
twitter.com/SFWRedditGifs

Bump

Where are the 500 IQ polymaths I keep hearing so much about?

Don't mind me, I will just post a scoreboard:

If you can't even get a better lowerbound than N: permanent brainlet

If your lower bound is of the form N + K where K is just a constant: Dumber than a /pol/fag.

If your lower bound is of the form N^K where K is just a constant: Brainlet in recovery

If your lower bound is of the form K^N, where K is just a constant: You have the intelligence of an slightly above average smart but lazy highschooler.

If your lower bound is of the form N^N: Hey, you are actually kinda smart.

If your lower bound is better than N^N: You are up there with Shinichi Mochizuki

They don't exist. 200+ IQ is considered "immeasurable genius" or at least that's what I read the last time I cared what an IQ test said.

I'm not doing your middle school homework, PIGGOT

If you see the structure of this problem you can see it is not homework. It is a contrived construction using only elementary terms so that it is hard but approachable.

If you google this then you will literally not find much about it other than some basic facts about invisible points.

If you really think it is middle school homework then why don't you give it a try?

Bruh, you can't measure intelligence with a problem that requires prerequisite knowledge. That's not how it works.

Guess what, they're actually brainlets.

>Piggot
Hahaha, that insult was never funny, stop it, faggot.

That is why I gave you all the definitions and even a theorem so that you know where to start looking for some assistance.

This problem is not hard if you have the right intuition and know where to look for some extra facts about invisible points. My rule is this: You can apply anything you find as long as the core argument is yours.

Here is the wolfram page about invisible points so that you can get started. This is a very interesting problem with a lot to say about the geometry of numbers

mathworld.wolfram.com/VisiblePoint.html

>piggot detected

sqrt(n)^sqrt(n)

You gave the definitions and theorems and assumed people were familiar with basic set theory. Not everyone is. Most occupants (engineers and engineering majors) of this board aren't. Knowledge =/= intelligence

That is not the best possible but it is a valid lower bound.

That said, post proof or you are just a brainlet.

Wow wow calm down. The only thing I set theory was to list the numbers from 0 to N.

They probably know that notation.

Rofl. Now we know you're looking for homework answers. Enjoy failing your intro to proofs, faggot.

>if you can't solve this novel problem I did you're a brainlet!!!
Nice one dimensional thinking, pinhead.

What the fuck?

Anyone who can code up a program that generates these squares can "realize" a lower bound. The trick is actually proving it is correct because it is not trivial to deal with relative primes in such broad and general terms.

I see now you can't even prove your little conjecture there. Don't worry about it. You are not the first one to leave an unproven conjecture behind. It shall be known as Retardo's conjecture

>number theory
you're at best measuring autism

Whatever you say asperger Andy

>He can't even handle the counting numbers

Lol how is middle school going man. What's the biggest number you know? Your mind is going to be blown when you reach fractions haha

(N*log^2(N)) ^ (N*log^2(N))
easy

Damn. That's pretty good. According to the scoreboard you are up there with Shinichi Mochizuki.

I had previously gotten N^N at best so I wasn't expecting someone to top that. Could you share a proof or the strategy you took? I am genuinely interested in your result.

no proof, just intuition. sorry I'm a brainlet

What the fuck, you are lying. Your post is too specific.

If you cheated at least share where you got the answer from.

I'm not lying. it just happens I research a distantly related problem. what's the best possible bound? I'm not sure about that square over log, feels like it could be 1 tbqh

Are you fucking retarded? How could it be anything else?

Well, I guess the problem remains unsolved.

Well, I don't know. It is obviously true, you can check it numerically but there is no proof and if there is no proof it's kinda moot.

Do you have a proof of that?

ok, fine. i looked at a few of the links on the wolfram article and managed to come up with a basic sketch of the proof. I'll write it up in the next hour.

Thank you for coming out my man. I guess I'll be here waiting for it.

so N^N is best possible? neat

Square root(2(n)^2)=a and b

No proof has been provided though so this really makes me think

what does numerical evidence say?

Makes you think what? That you're not going to finish your homework? Fuck off

That it does work.

Really makes me think that you keep claiming it is trivial and just "homework" yet you barely know where to start. Come on man.

I have a degree in math and I hate number theory.

>Mathematics is the queen of sciences and number theory is the queen of mathematics.

If after 4 years of doing mathematics you could not at least learn and understand this little fact of the universe then I must assume your degree was simply printed out of spite on toilet paper because the department wanted to get rid of your cancer.

> Definition 1: We say that a point is invisible if when you draw the line that goes from the origin to that point, that line crosses through at least another integer point.

No such point exists.
Proof:
Suppose such a point exists (A,B).
This forms a line with (0,)
y-B = B/A (x-A)
There is another such integer pair (x,y) that is not (0,0) or (A,B): (-A, -B)

B/A (-A-A) = -2B/A A = -2B = -B-B

Therefore N=infinity is a valid lower bound. Dude, you're a genius according to OP

...

Nice one smartass.

I meant segment.
Haha great job you sure got me.

Are you unironically this dumb?

Number theory is for nerds, set theoretic topology is where the cool guys hang.

Topology is pretty cancerous. It is a neat idea but then in the real world you see how hard you have to fucking beat that horse to get results.

Analysis and Algebra is where it is at. That is why there is both Analytic Number Theory, Algebraic Number Theory, and the closer thing to "Topological Number Theory" are a couple of little theorems I call the shitstains of number theory that are proved by using manifolds.

>set theoretic topology is where the cool guys hang.
is there even any research in general topology ?

I suppose you are not part of the kool kidz club then.

genuine question

TVSs and universal algebras are cool, but anything touched by number theory immediately becomes shit.

en.wikipedia.org/wiki/Set-theoretic_topology
It's not as active as algebraic geometry or PDEs, but yeah.

>the useless shit with zero practical usability that idiot mathematicians invent so they can have hobby

>says the fag whose hobby probably involves videogames or some other gay shit.

Nice projections. Universe is my witness that I dont have any video games on my computer.