#competition-math
70351 messages · Page 71 of 71 (latest)
You solved my previous problem right
Yeah
so would you like another one
Oh oh yeah my brain glitched lol
Yeah sure
ok
oh damn
I dont think a lotta ppl solved this
btw i only give questions which i have solved soif u have any q i can explain
Hmm I’ll solve it later in the evening cuz I’m about to shower rn
Yeah sure I’ll ask if i have any
okiee enjoy da problem
nuh uh
Thanks :DD
at least 12
what
r u referring to the number of ppl who solved it or the answer
hmmm
I'm like 67.89% sure it's perchance ||C||
I'm prolly stupid tho
hard to say I can't divide by 10 in my head
Lemme find my answer
hmm wait i've seen this
is it ||c||?
I should rly relearn how to do stuff in my head real fast
when u have final answertell meh
I did type it
also u shud blur ur answers to not spoil it forothers
lol cuz i sent wen it
yessirs corrects
how yall mental math so fast
u got any comp calc problems
it took me like five minutes
larpmaxxing
== I'm so bad at normal olympiad math I just do this bc I suck 
tis is not intuitively obvious
I can't load it
calculus?
ye it's not loading for me
@inland totem u shud watch takopis original sin
refresh
DON'T
DON'T DO IT
ok
u watched it?
yez
soooo peak
uhh i saw one this morning
but sooooo sad
it was cool
I had my review on it
dude I can't load da problem
idk if this quals as comp calc
wait ik how to do ts
I can't load this either
just refresh discord
hmm
just close and reopen the app
alr wait i havs smth to do rn
Your phone gets tired too yk
ok bye plens
am I stupid
is it ||B||
it's probably intuitive but I don't know series
ye ur right
da hint given in the Q
cuz like limit k->inf summation n=1 to k n^10/k^11, take the 1/k^11 out and u get like
uh
yeah not a pretty number
ye u can turn into a riemann sum
sounds made up
glad I never studied that(severe larp detected)
lol larping in math is the funniest
27>x>7
consider right triangle sides wherein the hypotenuse would be greater than
so we get x^2>10^2+17^2 and x^2+10^2>17^2 giving us x values of 19-27 and 9 giving us 208 AND it's not in the choices I'm so smart
I'm the goat
what is 17^2 anw
oh it's 289, not 189
289
🤦
I'm building numberthon (numberthon.com) — what’s the best part of competition math for you guys?
the geometry problems
larp part
wdym
The part where you win
lemme checks
oh, ok
ok so 289 389 20-27, x^2>189 so x>14 hm
looks weird
thats not even an answer
ah hypotenuse larger
yeah it's 17 squared
yessir
i love bashing
no
8-13, 20-26
8 9 10 11 12 13 20 21 22 23 24 25 26 should be ||E or 224||
289 isn't 17 squared?
split into cases where x is the longest side and then 17 is the longest side
i'm pretty sure it's ||e||
why can't 10 be the longest side
they never specified euclidean geometry

yeah I got the karma I deserved, I was walking bear foot in the dark and stepped on a Lego 🥀
Which set
I love lego
how would I know the set it just hurt like hell
Fair
yes
yes
17 squared just isnt correct
isn't 17 times 17 289
I'm confused
my answer was 208 I was wondering why I was wrong I got the wrong number and did 189-100 not 289-100
C?
lemme check
Uhh my working might be hard to read
Thanks 😄
For reading through that mess
Yeah I’m doing it now
theres like two main equations i gotta look for lol
Ohhhh
exactly
For specifying it’s..oh yeah right it can’t be non degenerate
Uhh okay i got…from the triangle inequalities it’s 323
Guys I didn’t use the first 2 parts tho…
Anyone wanna read my working?
ill read
Wait I’ll send
ok thats wrong but ill still read
||oh ye it said that the final triangle had to be obtuse||
It’s wrong?
uhmm yeah
nah
Uhh might not be readable. I wrote the last few steps on another page
Oh lol I’m totally wrong then
ok you forgot to consider another case from what i seeing
Oh…but…wait lemme think
hI
greetings
yess
Like my brain warped around “either x is opp acute angle or obtuse” and forgot to consider this
lols
want another one
Neither have i, ATLEAST not in the Olympiad way.. my coaching center has started calc but it’s like the basics
The last one was kinda annoying
o cool
u do math coaching?
No JEE coaching. It’s like an engineering entrance in India (yes many Indian parents have a very binary view on careers- either you become a doctor or an engineer)
lolsss funny
No not math specifically, but it’s included in the JEE syllabus, but like i do Olympiad math on my own
Engineering coaching is a new concept for me
kinda like my parents
o nice
So like this foundation course is like- super intensive so they’re spoon feeding us calculus with 0 proofs
xd so pure memorization?
Like mine
Basically, but i chose the Olympiad cycle, so now i have to do both- the memorisation and aops on my own so that i can give Olympiads
lol thats good
hey it will pay off in the future
I guess
btw TRY THESE OUT
all roads lead to one (collatz conjecture)
But like sometimes my parents try to make me abandon Olympiads…but I don’t want to so they said I’m gonna have to do both
OH DONT REMIND ME
bruh
at least 12
As a 10 year old i tried solving it. Gave me mental trauma
why are we trying to solve the collatz conjecture at 10 years old 😭
lmao
Ask 10 year old me who saw the Veritasium video
oh ye i did too
I was a very delusional kid
but didnt it explicitly say "do not try doing this problem, do some normal math for once" or smth
hmm
I genuinely don’t know what 10 year old me was thinking
lol
how did you try solving it
what would the first step even be
I tried using primes
Like seeing if there was a specific structure on how each of them come back to one
Youd need statisticsl analysis and a bit of probability theory minimum
circle rectangle thing, uhh probably ||D||, it's a bit shorter and there's no choice 3950
quarter circle's a nice one
don't think id's possible for the grasshopper to do that
hard to imagine on phone mm
lol
Uhhh I didn’t know the difference between LCM and HCF at that time btw
Yeah uhh that’s what happens when ur neurodivergent
Just a buncha number theory jargon you'll do fine
pfft
wrong
LMAO
oo i see
the grasshopper uses magic
Dude I obviously studied them later but like, yeah that was how versed i was at that time
the grasshopper is prolly ||E|| or smth
10 10sqrt5
triangle base is 10sqrt5-10
10sqrt5-20=base of really small triangle, 30-10sqrt 5 is base of the larger shaded so now u can do maht to do 100(14-6sqrt5)+50(9-4sqrt5) which I'm not calculating thx
Uhhh i think the answer is ||option d 4000|| but I’m not sure
For the first 3 circles one
I got 3950 cuz u get the radii 5, 15, 25, and then pythag on XZ and some stuff
Mine is wrong guys just found out
Will send sol later
||using approximation for root 15 as 38.7 (yes i used a calculator) I got this…||
is using a claculator
liek
like*
forbidden or sum
😭
you don’t need to write out approximate decimal places i assume
so all it is is just dividing a number by primes until you get all primes
like up to three digits
babylonian method
not mental math
also i dont usually mental math roots
i mean like with addition stuff
oof
lol
i kinda like doing it tho
like in my free time
anything thats not calc related tho

and trig
related
yessir
i only let myself do it after i do my textbook work
so it feels like a reward
noice
<@&268886789983436800>
what do you guys like abt competition math?
(I am asking bc I'm making a competitive math platform called numberthon.com)
Cuz there're many fun combi problems
Now get as many combi problem as possible into the platform, people will love it, trust me
i like how i dont immediately know the solution to them
they challenge me and force me to use new methods that might not be new to the world but is new to me
kewl
the feeling of success yk
They asked for closest, and we know it’s less than 4, so like, so basically we know its less than 4 (so less than 4000) but not like very less, so guessing 3950 without finding root is possible. But like there’s actually a long division method to find square root (digit by digit algorithm i think) which you can do in ur head with practice
idk i mean i like that it teaches you different stuffs than regular hs math
Uhhh the fact that it values deep thinking over rote speed and memorisation and I’m not good at speed but im better at depth…
i mean some comps like mathcounts also prioritize speed
Yeah but like, the speed in those competitions is developed by pattern recognition and like to recognise the pattern u gotta understand it deeply?
I guess?
hmm ye fair
3950 hm
Uhhh it’s wrong? I did pythogoras on XZ like someone else messaged…
I first got 4000 but then corrected it
i mean recognizing 5x12 is 60 doesnt require deep understanding just seeing it a lot engraves it in your head
Idk how I got 4000 first
Uhhh but like most Olympiad problems in speed based tests usually follow common Olympiad topics and methods
Yeah that’s what i got
Did u do the grasshopper q
No not yet i had a test in my class for the past 2 hrs
I just got out of the test (it was online)
o cool
yay how did u do
It was like moderately difficult, but the results come out tomorrow afternoon after the teachers check the results (like it was McQ but there’s like a check before they publish the results) so yeah
o math?
Math, physics and chem but like, it was mostly testing speed
sqrt15 and 5 aren't really 'nice'
I can do 2 digits but 3 digits is preferrable cuz of the amount of 10s
its the key to larp
my friend tries to solve as many things as he can with combinatorics to maximize how sophisticated he looks
combinatorics can't possibly be that hard there's only 3 formulas
3 formulas everyone refuses to learn and make sense of
Im trying to figure a way to do these without groking it out
I know /varnull is (a + b + c)^2- 3(ab + bc + ac), but not sure that helps
Elementary symmetric polynomials my beloved
Never heard of them
54 use pascal triangle
For the coefficients?
Yes
Im aware of the binomial theorem for the coefficient but this implies actually working it out entirely?
Yeah that’s how I did it; simplify both sides a bit and it’s easy to show
Just not very elegant
Although
For context it’s from an algebra book’s chapter review on polynomials and factoring, so it’s entirely possible it’s just meant to impart a ridiculous amount of computation. Just caught my eye and I was curious
3 formulas?
Like perms combs and binomial expansion u mean?
will do
I solved 51 in case anyone wanted?
Can u send it?
I'm quite puzzled
Idk how to solve it
Does anyone have any moderately easy number theory problem? I'm getting kinda bored
Prove that 2^n-1 is not divisible by n for any integer n≥2.
Yeah sure I’ll send it in some time my notebook’s not with me rn
Via contradiction orr?
Yep
||newton's sums probably||
Is the -1 in the exponent or not? If it is then it is divisible by n for n = 2
Of course it is this
my bad
I think this should be the solution to 51
One flipped sign and it’s over
Wait it’s wrong?
No no just in general
I was planning on splitting it between even and odd but then couldn’t do odd. I’m new to number theory and proofs in general, could you help me out
yep, this is kind of obvious, an odd number (2^n-1 is always odd if n>=2) can never be divisible by an even number.
I think u have a good idea but instead of multiplying by n make it so it’s 2^(n-1) -1/n = k
Progressing from there should be easy
i think thats correct
i did this with ||fermat's little theorem||
yep, I used it too
What do you guys think abt this competitive math website that I made?: https://numberthon.com/ Any suggestions on features to add (it aims to be the "chess.com of math")?
i quite like the ui and design tbh
Thanks!
Yeah so this is how I did it too; I’m just curious if there’s a way to derive the rhs from the lhs or vice versa
Also, for what it’s worth, 52 is pretty easy if you just multiply out the RHS
As far as I’ve tried, there doesn’t seem to be a way to ‘deduce’ the RHS from the LHS in 52. Best I can do is something like this (basically equates to: “oh! Notice this occurs, which I honestly think is BS when including on a problem set, but I guess it makes the authors feel cool)
Also, in my opinion, the only reason anyone would even bother trying to figure this out is because the factored form is already given. I highly doubt anyone would recognize/care for this “in the wild”
There are theorems in abstract algebra regarding symmetric polynomials, I just don’t see what the use of including them here is- especially when no mention or even reference is made to those notions/results elsewhere
Ty!
weren't lying when you said everyone then
In particular, this is exhibiting an instance of the fundamental theorem of symmetric polynomials (with no mention of even the term symmetric polynomial)
Yeah I mean it’s kind of upsetting to me when stuff like this comes up; like this is a result that is usually shown using galois theory; if you’re going to include these, say something about that so people aren’t rereading the chapter and rechecking their work because they’re under the impression they can’t solve it because of some fundamental misunderstanding
Why are you tagging me?
You seem to think because no one else is responding right now that no one else will respond at all
@silver forge pls dont make irrelevant posts in topic channels
And you can discontinue speaking for a longer time, considering your history...
Yeah this is one of those "classic" facts that's normally proven by expansion. Ig you could also say that
$$(x+y)^3=x^3+y^3+3xy(x+y) \implies x^3+y^3=(x+y)^3-3xy(x+y),$$
and so
\begin{align*}
a^3+b^3+c^3-3abc &= [(a+b)^3-3ab(a+b)]+c^3-3abc \
&= [(a+b)^3+c^3]-3ab(a+b+c) \
&= (a+b+c)((a+b)^2-(a+b)c+c^2)-3ab(a+b+c) \
&= (a+b+c)[(a+b)^2-(a+b)c+c^2-3ab] \
&= (a+b+c)(a^2+b^2+c^2-ab-bc-ca).
\end{align*}
You can also inspect that setting $a=-b-c$ forces the expression to be zero, so $a+b+c$ is a factor and then you can do division. (I've never done this, so it probably sucks.) \
Fwiw, this is precisely the determinant of a $3 \times 3$ circulant matrix:
$$\begin{vmatrix} a & b & c \ c & a & b \ b & c & a \end{vmatrix}=a^3+b^3+c^3-3abc,$$
So the factorisation makes it easier to find the eigenvalues. (Note that circulant matrices are common in digitial signal processing.) \
In a sense, you could also use this to prove the identity. Because
$$\begin{vmatrix} a & b & c \ c & a & b \ b & c & a \end{vmatrix}=\begin{vmatrix} a+b+c & b & c \ a+b+c & a & b \ a+b+c & c & a \end{vmatrix}=(a+b+c) \underbrace{\begin{vmatrix} 1 & b & c \ 1 & a & b \ 1 & c & a \end{vmatrix}}_{a^2+b^2+c^2-ab-bc-ca},$$
and so
$$a^3+b^3+c^3-3abc=(a+b+c)(a^2+b^2+c^2-ab-bc-ca).$$
Civil Service Pigeon
rough first draft, so you could probably try to refine these proofs as an exercise (especially the first one)
Yeah, great points- I missed the determinant equivalence!
I guess I was looking for a way to factor the LHSs to obtain the RHSs, which I don’t think is possible unless you’re aware of decomposing symmetric polynomials in terms of elementary symmetric polynomials- which no one reading a college algebra book to learn from would have any idea about or be able to deduce
why u spoilering it like it's an answer
define moderate
idk it leads to the answer
oh ye it's just fermat
wait
hmmm
yeah it is I'm prolly just dunv
🎁
Civil Service Pigeon
this looks like someone made it with hate
243 is 3^5
I know ur in there euler I just don't wanna think about it
||what I'm thinking is that either x or y must have a common factor with 243 so basically just all numbers divisible by 3 and stuff and then the gcd is any factor of 243 prolly||
but I'm prolly approaching wrong
yeah ||common factors|| is a relevant idea
247
hmm
||Like on rearranging we get that gcd(x,y) will divide 243 ( since gcd divides LCM always)
Idk whether this will be a relevant idea
So you can now look into possible gcds||
ts is a fun problem but id be cooked if i had no calculator
is the answer ||2668||?
yay
yeah

oh lmao
fr
i did it by
||letting 3^a = gcd and 3^b = xy/gcd²+1 and since a + b = 5 i solved for every possibility of x and y|| is there a faster way
You can find the sum without finding the actual pairs $(x,y)$ ig
Civil Service Pigeon
that's really the only way I can think of making this more efficient
how tho
Let $\gcd(x,y)=3^a$ and let $x=3^a u$, $y=3^a v$ for $\gcd(u,v)=1$. Then, $\operatorname{lcm}(x,y)=3^a uv$. Hence,
$$3^a uv-3^a=243 \implies 3^a (uv-1)=3^5.$$
So, $uv-1=3^{5-a}$. \
But since $u$, $v$ are coprime, every prime power in the factorisation of $3^{5-a}+1$ must go entirely to $u$ or $v$. This means the number of ordered coprime positive pairs $(u,v)$ is $2^{\omega(3^{5-a}+1)}$. At this point, you can simply run over $a=0$, $1$, $2$, $3$, $4$, $5$ by finding the pairs $(u,v)$ and multiplying them by $3^a$ since
$$x+y=3^a u+3^a v=3^a (u+v).$$
Civil Service Pigeon
noo training my reading comprehension
ok ima read
oh wait thats what i did basically i think
cool
what is $\omega$ though
aKube
Anyone wants a moderate geometry problem I’ve solved?
thats cool, it looks like theres lots of greek lettered number theory related functions
or i think so im not too sure
oh yeah there's a lot lol
If you want to share it, then I'm sure someone here would take interest 
a list of these and all but two are greek
sure
I might be busy for some time so I’ll not be replying, but i have solved it and i can help whoever needs help
oh god theres more https://en.wikipedia.org/wiki/Arithmetic_function
and it doesnt help that the sum and product symbol are both greek too
In number theory, an arithmetic, arithmetical, or number-theoretic function is generally any function whose domain is the set of positive integers and whose range is a subset of the complex numbers. Hardy & Wright include in their definition the requirement that an arithmetical function "expresses some arithmetical property of n". There is a lar...
number theory might as well have been made in greece atp
$$x+\frac{2\Delta}{x}=y+\frac{2\Delta}{y} \implies x-y=2\Delta \left(\frac{x-y}{xy} \right) \implies \Delta=\frac{1}{2}xy.$$
This implies that we have a right triangle with legs $60$, $63$, and so the longest side of the triangle has length $\sqrt{60^2+63^2}=3 \cdot 29=\boxed{87}$.
Civil Service Pigeon
Correct
Dude how’d you do it so fast lol it took me like 20 minutes


"