#help-49
1 messages · Page 220 of 1
and try to solve this :p
tell you what it is
its right there in the picture
thats called "chain rule"
say youre trying to take the derivative of one function inside another
usually youre taking derivatives of functions like f(x) * g(x), or f(x) + g(x)
yea thing with me tho ill prolly ask u why that works now 😔
mhmm
those are functions that are multiplied or added
now consider composite functions, f(g(x))
do you know what composite functions are
yes
now composite functions have their own derivative
if you want to get a sense of this, consider differentiating f(2x)
and think of this conceptually
how does that work
I want you to focus on specifically finding the derivative of f(2x)
you should be able to see a visual pattern here
so in red here is f(x)
and in green is f(2x)
youll notice f(2x) is just f(x) but horizontally squished, right?
so lets say we were looking at f(x) at x = 0.9
the slope of f(x) at x = 0.9 would be f'(0.9), right?
mhm
now take a look at the green function
its marking a particular point
to have a particular derivative
for starters, where is this point at
are u referring to 0.9/2 btw
OH WAIT
IK THAT
thats not going to explain what the question really wants though
which is why the chain rule has to work
but i never learned it properly i legit js thought of du as variables that cancel out 💀
right we'll need to get to that later
yeah pls keep going tho
f(0.9/2)
are you sure?
you could just say f(0.9) you know
yea ok
that you said this though is interesting, its also correct
yea it kinda clarifies what function ur using :p
so the two coordinates fall on a horizontal straight line?
cuz the output is the same
alr
at 0.9, the gradient is f'(0.9)
now compare that to the green point
what do you think the gradient would comparatively be?
we know its steeper for one
yea
itd be larger but
if ur looking for me to say the exact value i will probably disappoint u
you already know what this value is
consider this: if it takes 30mph to get from A to B
and you see someone else travel from A to B in half the time you did,
they went 60mph didnt they
no you didnt
you said cos(x)
you forgot a key factor here
you didnt say cos(something else)
you forgot the "something else" part of it
hmm
youre saying they went 60mph but took as long as you did
thats not correct
this does mean though you have a good sense of what this slope would be
given the red point's slope is f'(0.9),
what would the green point's slope be
no it isnt, you need to stick with me before you talk
yeah sorry
notice here youre taking derivatives at two different points
you found the slope of a different point at x = 0.9/2 instead of x = 0.9
we know the slope must be double that of the slope at x = 0.9
so if you know the slope f'(a),
and h(x) = f(2x),
then we're saying h'(a/2) = 2 f'(a)
so since h'(x/2) = 2 f'(x),
but algebraically im lost
we js rewrote the same thing in a diff way?
you say that like this is the same thing, I havent shown you that thing yet
let me continue
you can see this fact shouldnt really depend on a
it should be true for any number in general
so h'(x/2) = 2 f'(x) for any x
right?
nah nowhere ignore that part i was trying to link them wrong...
i thought you introduced a new function
but you were
describing
an existing one
wed been talking abt
anyways
and people wonder why wearing glasses counts as a disguise
anyways since this is true for any x,
doubling x wouldnt change how true it is
so we also know h'(2x/2) = 2 f'(2x) for any x
does that make sense
yes it does
and so h'(x) = 2 f'(2x)
previoulsy you were just doing h'(x) = 2 f'(x)
you didnt write down the 2x part of it
here,
h'(0.45) = 2 f'(0.9)
if you had h'(0.45) = 2 f'(0.45) instead thatd be this slope instead, which is not the correct one to double
ohh i just understood it rn
when you put it like that
ok yes i got it (ill still have to teach myself it again 🥀 )
in general,
if you have h(x) = f(g(x)),
and you want to find its derivative,
its f'(g(x)) g'(x)
ur a patient person
sure
if you think about it, g(x) should be differentiable if you zoom in on any section and it looks like a line
so that g'(x) would then be the slope of that line
so if you want the derivative of f(g(x)), and g(x) has a slope of g'(x) at x,
we use the same pattern as with 2x
f(2x) has derivative 2 f'(2x)
f(g(x)) has derivative g'(x) f'(g(x))
this is where the f'(g(x)) g'(x) formula comes from
okay that js made a lot more sense.. in the case u were talking about btw like f(x) and f(2x) is g(x) essentially =2x and so the derivative would be f'(2x)*2?
checks out :o
yes
now for the formula you did learn, which is dy/du du/dx
this is the same idea as before, but you missed a key area
seems drastically different than this one idk why..
this formula hides how you plug in each variable
its $\frac{dy}{dx}(x)=\frac{dy}{du}(u)\cdot\frac{du}{dx}(x)$
mtt
you plugged in x instead of u into the function dy/du
now how does this link with the original definition we wrote down? we have to carefully change each part so that they line up as expected
i also get how it was wrong to say the derivative of sin(pi*x/180) would be (pi/180)(cos(x)) btw cuz that makes no sense rn and i see how itd mess up the whole graph
right, but you still need to use this formula correctly so that you dont make this mistake again
we can do this by matching it up with the other formula d/dx f(g(x)) = f'(g(x)) g'(x)
first, Ill change the letters to match closer with the definition:
d/dx y(u(x)) = y'(u(x)) u'(x)
wait this isnt correct
this isnt going to be easy
its okk
i can proceed here on my own
maybe...
you sort of cant
lets see
this is a very precise kind of difference here
first, do you know how to do $\frac{d}{d(2x)}f(2x)$?
mtt
mtt
what would the derivative be
ok in that case i have a sense of it
wait ur not supposed to ask me that 😰
this brings my middle school math anxiety back
I just replaced f with sin...
its the exact same question but you know what f is now
and you already know d/dx sin(x) = cos(x)
do the exact same thing again
mb if i get it wrong after all that we did but is it cos(2x)
yep its cos(2x)
d/d(2x) sin(2x) = cos(2x)
now lets replace 2x with u
d/du sin(u) = cos(u), it still lines up
thats certainly one of the things to say of all time
we're not going to mention that until later
thats how i thought of it too
😭 that bad?
yes, but when you did d/dx sin(pi x/180), you put in x instead of u into cos(x)
(for what it's worth, this should be a safe space where we won't be too hard on you or judge you if you get the wrong answer, we want you to learn as much as possible
)
it wouldve been pi/180 cos(u), or pi/180 cos(pi x/180)
so here its no more than a typo, but its a very easy to forget typo
so you can see here we have a reason for going about this chain rule idea more literally so that we can avoid this mistake with more intuition
yea i understood why that was wrong like 20 messages back but it makes more sense now that you put it this way ^o^ (i was thinking in terms of graphs but i need the algebra to check out too)
its characteristically good but a major "abuse of notation"
thats so sweet ☹️ tysm 💛
that means youre using the words in unintentional ways which wont work in general
when i first said it, it was like a deep error in my foundations i think
i get it now to some extent
yea i js hear of it when it comes to integration
WHICH I HAVENT DONE YET
i think my original question is answered too in terms of why the red graph on the right looks scaled down (while still holding the features of the red graph on the left)
tysm @twilit jetty
back
@ebon atlas yea by this point the original question has long been answered, but we still gotta make sure the mistake wont come up again
so thatll come in terms of making extra sure that we get dy/du du/dx
knowing that d/d(2x) sin(2x) = cos(2x) is what we expect
i probably wont be making the same mistake anytime soon :p
alr
sin(2x)
= dsin(2x)/d(2x) * d(2x)/dx
= cos(2x) * 2
very similar to the
sin(2x) = cos(2x) 2 we'd get anyway for f'(g(x)) g'(x)
junior year
3rd out of the 4 years we stay in uni
I mean you pieced together what you got wrong and got correct almost every big leap we did
hmm idk i could be a faster learner
OK WAIT if im entering 11th grade in liek a month do you think i could keep up with uni math by the time i finish hs
ikk u probably couldnt give a judgement of that but meh
if you know calculus when you come out of hs youll already be better than most people
so youll be good for uni
ok tyy ^^
np
.close
Closed by @ebon atlas
Use .reopen if this was a mistake.
Send your question here to claim the channel.
Remember:
• Ask your math question in a clear, concise manner.
• Show your work, and if possible, explain where you are stuck.
• After 15 minutes, feel free to ping <@&286206848099549185>.
• Type the command .close to free the channel when you're done.
• Be polite and have a nice day!
Read #❓how-to-get-help for further information on how to ask a good question, and about conduct in the question channels.
any ideas with this, i was trying to think with examples but theres too many of them,
maybe something with gcd or taking 7 common from the expression but i cant seem to figure that out
no its the product
of abc
i mean
yes
its over all the values
mb, i read your question wrong
like 1,1,1
thatll be 1
then 1,1,2
thatll be 2
so then 1 + 2
like that
yes its a sum over all $(a,b,c)\in T$, or $S=\sum_{(a,b,c)\in T}abc$
mtt
i.e. your condition for a, b, c being natural numbers should be inside the set defined as T
yes
consider the action of (7 - a, 7 - b, 7 -c) on T
wont i get the same set T then?
assuming you sort the triples in non-decreasing order yes it is a simply map that sends T to T
the point is the properties of this map will allow you to show 7 | S
so then i replace abc with (7-a)(7-b)(7-c), and then that product will supposedly give me a 7 to factor out right?
from which i can say its true?
can you spell out more explicitly what you think the argument is
like, wdym
write it out very clearly
since (7-a),(7-b),(7-c) gives me the same set(kinda) if i show that the sum of (7-a)(7-b)(7-c) gives me a result which is divisible by 7, it should prove the original one too, but trying it out now, i think im going in the wrong direction
i see
i cant figure out more
so you agree
$S = \sum_{(a, b, c) \in T} abc = \sum_{(a, b, c) \in T} (7 - a)(7 - b)(7 - c)$
Mqnic_
this follows from the fact: the action from T to T is a bijection
right
now if you take mod 7...
okay, really sorry to interrupt, i still havent worked with mods, its right after this tho, so, i do have a hard time understanding them
hm
okay well that is fine
for simple purposes you do not need to know much aside from
(a + b) mod n = a mod n + b mod n
ab mod n = (a mod n)(b mod n)
so taking this mod 7 something important happens
-abc?
yes
yes, so 7 | S because 7 is odd, to clarify
Closed by @grave kernel
Use .reopen if this was a mistake.
Send your question here to claim the channel.
Remember:
• Ask your math question in a clear, concise manner.
• Show your work, and if possible, explain where you are stuck.
• After 15 minutes, feel free to ping <@&286206848099549185>.
• Type the command .close to free the channel when you're done.
• Be polite and have a nice day!
Read #❓how-to-get-help for further information on how to ask a good question, and about conduct in the question channels.
help
What step are you on?
1. I don't know where to begin.
2. I have begun but got stuck midway.
3. I got an answer but I was told that it's wrong.
4. I got an answer and would like my work checked.
5. I have a question about someone else's work/solution.
6. I have completed the problem and don't need help anymore. Thank you.
7. None of the above
Your inequality is the wrong way around
Also put parentheses around your numerator for clarity
But you can't solve this inequality algebraically (at least with "standard" tools)
your best bet is to use trial and error
the solution is quite small
how'd you determine this because i was trying to solve it algebraically
i thought there was a way with logs or something
This is a standard case of using the lambert W function
which I doubt is assumed knowledge (also you'd still have to estimate the solution after)
complex functions normally require 4 dimensions to graph
we don't draw in 4 dimensions
so that's the work around
you smart as hell
u studying string theory or something
the answer is first term btw thanks
eh it's one of the weird internet math things that people like
idk why they picked this specifically

also had this show up in my exams lol so
yup it's just the first term
hence why I said the solution is small (and thus guessable by trial and error)
If you are done with this channel, please mark your problem as solved by typing .close
wait wait
im getting tripped up
ugh
why does the question say how many terms to sum together
the error bound is just the next term right
no addition needed right
or am i over thinking it
I assume you're referring to this
This says that the error of approximating a sum by adding the first n terms is the (n+1)st term
yeah
which is what I assume this is about
now back to this
This says that the error of approximating a sum by adding the first n terms is the (n+1)st term
so let's say we got something like n > 2.85 and we answered n=3
this means you would have to add the first 3 terms to get within your desired error
here the answer was n=1
i thought n+1 meant you just add 1 to the term
so you're just "adding" the first term to get a sufficient estimate
aka just using the first term and nothing else
which is why the idea of "adding/how many terms" is a bit odd here
no - it means that you calculate the (n+1)st term
Ex. If a(n) = 1/n and we had n = 3, then a_(n+1) would be a(4) = 1/4
ok i kinda get it
yeah tripped me up

ahhh
hold on lmk if i get this wrong
let's say u solve for n given a remainder of .01
and n = like 20.54
u round up to 21
oh yeaj
,w (n+1)/6^(n+1) \leq 0.01
so you can just trust the inequality sign
it says n is at least 2.22343
the smallest integer n satisfying that is 3
done
ok
and wait
u don't need to add one to that cuz u already did the n+1 when setting up the intequality right
and this means that when you add n amount of terms in the series you get a remainder of what ever you were given
yes as per the statement of the test
alright thanks
you get a bound on the remainder
and this right?
not the exact remainder
because if you could get the exact remainder
oh yeah i keep forgetting mb
then calculating any infinite series would be very easy
u right it would be too OP

@small jasper i need a sanity check
the question said no more than .1
so it's .1 >= 1/6
right
yes
,w |-\frac{(-1)^1 * 1}{6^1}+\sum^{\infty}_{n=1} \frac{(-1)^n * n}{6^n}|
what
oh that's the partial sum
dude idk if my brain is really fatigued but
how is .1 greater than 1/6
,calc 1/6
Result:
0.16666666666667
u change the signs when inverse?
i thought only when u multiply or divide by negative
You say "only" probably cause you weren't taught what happens if you take the reciprocal of both sides
only what happens when you add/subtract/multiply/subtracft
I assume you're American so this sounds reasonable
idk
so reciprocal changes signs
whatever
anyway a < b implies 1/a > 1/b if a and b have the same sign
and if a > b, then 1/a < 1/b
this comes from dividing both sides of a < b by ab (and similar for a > b)
so in essence it is dividing both sides by a (positive) number ab
but having the shortcut offhand is nice
naw i must be brain fatigued if i can't even understand this
how long can u do math for without having a headache
depends
I mean I've had six hour exams before lol
which would take even longer if you tried to finish
maybe the power of negative 1 flips the sign and the negative there makes it flip idk
but when actually learning math
eh depends
is it a reading comp exercise or does it actually require thought
it's all practice problems then i dug into rabit holes
this comes from dividing both sides of a < b by ab (and similar for a > b)
rabbit holes for me may not seem like rabbit holes for u btw
eh imma count this under "learning" since it's new, so again it comes back to "is this reading comprehension/direct application or does it actually require thought"
which ofc changes overtime
why did this guy not flip it
,w 1000 \leq (n+1)^2
the solutions aren't the same
yeah that's what being a muppet means lol
why it say this
Also if it was this
this would be useless since the remainder could go off to infinity
okayy
why in the problem it's .1 >= a(n+1)
is it cuz u want an error of no more than
$|s-s_{n}|<|s_{n+1}| \leq 0.1$ is the usual way you write the string of inequalities
Civil Service Pigeon
and then just take the last bit
yeah i don't think im truly understanding the theorem like you are im just memorizing it which is a mistake
uh this is what the alternating series error bound is for
so
sure ig
tldr the absolute difference between a sum and the nth partial sum is at most the absolute value of the (n+1)st term
ur hurting my brain
yeah i get that
wtf
chill yo i get it
like
u have the whole series with all it's terms
if u subtract Sn from it ur gonna have a(n+1)
that's the bound right
There is a nice result for approximating the remainder of convergent alternating
series.
$|s-s_n|=|a_{n+1}+a_{n+2}+\cdots| \leq |a_{n+1}|$
Civil Service Pigeon
$|s-s_{n}|=R_n \leq |s_{n+1}| \leq 0.1$
Civil Service Pigeon
if you want to throw that in go ahead
but usually we define R_n to be the absolute difference
aka $|s-s_n|$, not necessarily $s-s_n$
Civil Service Pigeon
ofc this applies for when said difference is negative
if the problem says u want an error of let's say .0001
would u write it as |an+1| <= .0001
sry if im being a dumbass
now it's good
i was doing math the whole day
what if the question was an error no more than .0001
still the same?
this is what I assumed you meant in the first place
because again, finding the exact error would mean you found the exact value of the series itself
yeah im being dumb my fauly
which defeats the entire purpose of bounding the error
ok i'll take ur advice of
Rn <= |an+1| <= (given remainder)
like i'll remember that for ever
cuz that helps a lot
fix the first less than sign to a $\leq$
Civil Service Pigeon
wait why'd u write s
typo
oh
alrighty dude thanks
i'll see if i can do another problem that's similar real quick

alright GN dude thanks
naw man i've been doing math for so long
my brain hurts
wish me luck on my exam
thanks a lot
.clos
.solved
Closed by @chrome swift
Use .reopen if this was a mistake.
Send your question here to claim the channel.
Remember:
• Ask your math question in a clear, concise manner.
• Show your work, and if possible, explain where you are stuck.
• After 15 minutes, feel free to ping <@&286206848099549185>.
• Type the command .close to free the channel when you're done.
• Be polite and have a nice day!
Read #❓how-to-get-help for further information on how to ask a good question, and about conduct in the question channels.
what happened to the first t inside the integral?
(not cheating btw i’m just trying to learn how to integrate)
when you perform the u-substitution the dt has to be replaced with something involving du
so how did they eliminate the t to make it du instead of dt?
what do you get when you take the derivative of u = t^2 - 1 with respect to t?
2(du/dt)?
no, it should be du/dt = ...
i kinda see it
@fringe pine Has your question been resolved?
Closed due to timeout
Use .reopen if this was a mistake.
Send your question here to claim the channel.
Remember:
• Ask your math question in a clear, concise manner.
• Show your work, and if possible, explain where you are stuck.
• After 15 minutes, feel free to ping <@&286206848099549185>.
• Type the command .close to free the channel when you're done.
• Be polite and have a nice day!
Read #❓how-to-get-help for further information on how to ask a good question, and about conduct in the question channels.
find all irrational $x$ such that $x^3-15x$ and $x^4-56x$ are both rational
ihave<skissue>
err whar?
Damn
some sort of factoring maybe?
the irrationals must be of the form $a+b \sqrt{c}$ and the square root terms must cancel
cytos
proof?
no need. it’s all in my head
?? 😭
a = x^3 - 15x, b = x^4 - 56x
continue

might’ve been wrong icl tbh ong
what do you want me to do?
reduction
yeah thats kinda what i was guessing, but how?
i mean im guessing you want rational=(something in x)
the proof is cbrt2 is not of that form and therefore its wrong
something in x thats easier to solve for rationals (like linear?)
x^4 = x * x^3
i think he was going for irrationals that might solve the problem are in that form?
oh fair
you want to multiply a by x?
oh wait
b-ax=15x^2-56x
then solve for x using quad eqn?
$x=\frac{56+a\pm\sqrt{a^2-112a+3136-b^2}}{30}$
ihave<skissue>
$x^3-15x=a,a \in \mathbb{Q} \ x^3-15x-a=0 \ (x-\frac{a}{5})(x+\frac{a}{10})^2=0$
gross
would this then just not have any irrationals that produce a rational
there will be an a^3 term in the expansion no?
,w expand (x-a/5)(x-a/10)^2
eh?
did wikipedia lie to me
did you try applying cardanos
you have proven
oh no I’m stupid
wait wait
,w expand (x+a/5)(x-a/10)^2
sus
aight
cytos
surely I got it this time
you can tell just by the coefficients this is gonna be wrong
using this the idea is to get equations in p and q
then solve those
again use the given conditions x^3 - 15x = a rational, etc
p and q are rationals? right
a=x^3-15x=px^2+qx-15x=p^2x+qx-15x+pq
0=p^2x+qx-15x+pq-15p-56
i was too obsessed with doing some ratio thing but it turns out it was adding
Am I the only one who thinks you guys are massively overcomplicating things? Unless you guys are done and I should shut up
is there a simpler solution?
don't think it's p^2q at the end but sure this fine
15p+56=a?
don't go back to a and b
there is something fairly important that equation you have tells you
wait wrong variable
- Find all irrational x such that x³-15x is rational
- Find all irrational x such that x⁴-56x is rational
- Take the intersection
hm.
oh its pq right
no the point of this is to get rid of a and b
a and b are very inexplicit things that don't help you when you're trying to find x explicitly
and to do this you want to find p and q explicitly
like this?
is that true
well wtv the point is
an irrational + a rational can never be rational
which are
how do you sppose we do that?
-p^2=q
oh wait
exception at p^2+q-15=0, which we can basically isolate q=15-p^2 then sub back?
i wouldn't do that
do the same process for the other initial condition you are given
mk
b=x^4-56x=x^2(px+q)-56x=px(px+q)+q(px+q)-56x=p^2(px+q)+pqx+pqx+q^2-56x
coefficient of x is p^3+2pq-56=0
this ones ugly :(
oh yeha
uhh how do you solve it exactly?
exactly how you'd expect (substitution to remove one of the variables)
@viral dagger Has your question been resolved?
p=-cbrt(56), which gets a=-15cbrt(56)+56
wait -cbrt(56)
that doesn't make sense either
try the substitution again and recheck very carefully
remember your equations are
p^2 + q - 15 = 0
and
p^3 + 2pq - 56 = 0
hm
=0
looks good
,w solve x^3-30x+56=0
so p=4
correct
sorry smth came up, back
nw
so, we got 15p+56=a <=> a=116 <=> x^3-15x-116=0
uhh did i plug it wrong
tbh i'm not sure where 15p + 56 = a came from to begin with
anyway you should find q not a
q=-1
oop wrong reply
oh wait and the constant term (aka q) here is b, so -1=q=b
remember what the definition of p and q are
but does finding a and b really help you
this entire time the point was finding p and q
.
yes so you are done
🤔
i underatand why would somebody try this up to here, whats the (incentive? encouragement? what makes you think to do that?) to make it do the x^2=px+q and get equations interms of p and q
and after getting the equations i think its reasonable on how one would think to complete it
tbh the stuff with a and b is just too messy
you want to make use of 15x^2 - 56x + ax - b = 0 somehow but this form takes too long to do calculations with
so supposing x^2 = px + q is just as well
and solving for x before knowing p and q (or a and b) is no good either, since then you can only proceed by plugging x into x^3 - 15x, which is completely terrible
even if you guessed x = (a + bsqrt c) you would still not have such an easy way to proceed
so after having x^2 = px + q you should think to make further reductions
real quick skish can you do me afavor and just make sure you transcribed the problem properly
yeah gimme a sec
as far as i understand the problem, yes
ok thanks
why btw? im curious
ohh ok, ill try to keep that in mind
and in this case you did that via 0 = sx + t where s and t are rational so s = 0
seems like a lot of work
really? there is not much
youve been here for an hour already
so? i could probably write out this solution in full in 5 minutes
ok fair
yeah mb
.solved ty guys!
Closed by @viral dagger
Use .reopen if this was a mistake.
no i was just saying
i do not think this problem is particularly easy
i mean like 2/3rds of it was me not responding
oh ok
Send your question here to claim the channel.
Remember:
• Ask your math question in a clear, concise manner.
• Show your work, and if possible, explain where you are stuck.
• After 15 minutes, feel free to ping <@&286206848099549185>.
• Type the command .close to free the channel when you're done.
• Be polite and have a nice day!
Read #❓how-to-get-help for further information on how to ask a good question, and about conduct in the question channels.
im guesing [.] denotes gif
Yess
not a good idea
Sorry forgot to mention
All of the subsequent questions were done in the same manner
split the fraction into $\frac{\sin x}x\cdot\frac{\tan x}x$
mtt
what do you think
do sin x / x and tan x / x remind you of anything
remember this is a limit as x approaches 0
yes
never just adopt to a pattern , i see what you are trying to do with that but thats not needed
Wrong
sigh
so this limit is?
Actually the second limit is not 1
1
yeah we know
How tho
,,\frac{x^2}{\sin x\tan x}=\frac1{\frac{\sin x\tan x}{x^2}}
mtt
This is 0
no
i think you missed the gif function
what was the limit of this, again?
it was not mentioned in the question
1
What about the gif of this expression
What would be the limit of that is the question
answer the question
1
ah ok mb
and since these two are equal,
Yes
what would this limit be
Idk
you dont know?
well "Dude,"
Stop and observe for a sec
these expressions are exactly the same
floor function-
first off
@twilit jetty
whats the floor of 1?
so its either 1 or its DNE
Limit of sinx/x means that it approaches to 1
Stop and listen for a sec
zyro you
listen to me
we do this one step at a time
and the first step is assuming that the limit exists
then the second step is verifying that the limit exists
the first step, you havent done yet
U are not understanding the question
zyro you seem to be seeing some sort of idiot in me that isnt real
just say the number 1 ok?
Ok
what point do you think I am making?
Go ahead
you keep typing over me like you know
take the reins: what was I going to do with this?
I think that u are saying that because the limit of sinx/x is 1 and tax/x is 1 the value of the question would be Q
wrong
1
but we've already verified that the number inside the limit would be 1 regardless
if its 0 and its 1,
thatd be DNE
no I see what youre getting at
yes
Yes
I was thinking itd either be 1 or DNE
Thats what I am trying to say this whole time
but sin(x)/x approaches from below
so the limit would be 1 but a lil shorter than 1
zyro was trying to compare functions with sinxtanx - x^2
And since tanx>x the limit of its floor would be 1
so he can conclude if sinxtanx > x^2 or the other way arround
Yes
yeah?
Thats where I am having problem
you can just check whoese rate is more , tan or sin
lets not use derivatives on a question that doesnt require them unless absolutely necessary
especially with floor
not really you would be comparing
Tanx /x and sinx/ x just check who rate more , so you know which one dominates in the gif function
i think taking f(x) = sinxtanx-x^2 and checking the signs of the function would be better
thats works too but you are effectively doing the same , if its subjective paper and step marking is there go for it
but its not required to do all this when you can judge the whoese rate would be more
the only answers possible are 0 or 1 for each limit
so theres in total 4 possible answers
so it being a multiple a choice wont belp us
if one of the answers is a DNE then we can use that to eliminate up to 2 of the answers
you know the range na , thats all thats required to determine the rate tbh
since the limits are 0 and 1 or 1 and 0
how do i compare the rates
you can do that roughly too ig , we alr know tanx is increasing in x belonging to 0 to pi/2 , same for sinx , so just assume value at end points and yeah rates
if you wanna be precise and sure , derivative is the way
average would work too tho tbh
average wont tell you the behavior near 0
yeah i gotchu jee
so i want to try to find the most precise way
theres a precise way I want to avoid as a last resort
taylor can be used
you get to do that?
but the thing is that is not in our syllabus
oh that wouldve been powerful
it is? tf
so we cant relly use them
you can consider taking the double derivative of (sin x)(tan x) - x^2
i did that
did i get which exam you prep for wrong?
was it positive?
oh right itd be 0
but the thing is taylor hsnt been taught
yeah taylor is very much relevant
rn
so i cant use it
tanx > x > sinx , its a very comon inequality
youd have to take 4 derivatives then
i wanna do it by analyzing the function
no
i can just take 2 to check the behaviour
thats not enough
show your work
oh heres something easier
you can take two derivatives of (sin x)/x and of (tan x)/x
(tan x)/x has a larger second derivative than (sin x)/x
problem is thatll tell you what happens for (tan x)/x - (sin x)/x, not (tan x)/x * (sin x)/x
@last slate I found a function you can d/dx twice for
we need to find out whether this is true or not in the neighborhood of x=0
now if you write it like this,
that just means we check the same thing for (sin x)/x - sqrt(cos x)
due to the square root and dividing by x, this is a function with a nonzero second derivative (as you can check if you d/dx it twice)
@last slate Has your question been resolved?
however it takes quite a while before you get somewhere useful
each of these are either special limits or you can plug in x=0 into it
this though, you need l'hopital for this
Closed due to timeout
Use .reopen if this was a mistake.
Send your question here to claim the channel.
Remember:
• Ask your math question in a clear, concise manner.
• Show your work, and if possible, explain where you are stuck.
• After 15 minutes, feel free to ping <@&286206848099549185>.
• Type the command .close to free the channel when you're done.
• Be polite and have a nice day!
Read #❓how-to-get-help for further information on how to ask a good question, and about conduct in the question channels.
the proof? or the choice of weights?
mostly the proof
so to be able to get every distinct weight you need to have the weights for (2^n)-1 of 2^0, 2^1, 2^2 etc up to 2^(n-1)
first you say mostly the proof then you talk about the other part of the question...
sorry idk
i feel like its both
anyway, with n weights you have n binary choices to make for weighing: either you put the weight on the scales or you dont
which means 2^n different configurations
you cannot have more than this many different achievable weights
(but you can certainly have less)
oh yeah thats true
yeah ok
i forgot you can also include 0 as a configuration
so ig i felt confused because you could only go up to (2^n)-1 but i forgot 0 is included
also this is something but then how do we actually prove this
because writing a few sentences isnt a proof but i honestly dk how to do a proof for this either
writing a few sentences isnt a proof
i'd beg to differ depending on what's contained in these sentences
you can write a formal proof by induction if you wanna spend time on that
that n (distinguishable) weights yield 2^n configurations
oh so you can just write a proof like that?
I always preferred like formal mathematical proofs rather than just writing/explaining one
I mean if I don't have to for the question then it's fine
what do you think of as a "formal mathematical proof"
do you think it's something completely devoid of words and is instead just symbol soup
im gonna be honest
kind of yes
or at least very algebraically
well then you have the wrong idea about proofs
real mathematicians write "prose proofs" all the time
Yeah probably
I guess it's something I need to get used to
because idk for some reason I wasn't a huge fan of prose proofs
whenever i write one it's very hit or miss
heres what a proof from my own master's thesis looks like. it's not important what it's talking about but i just wanted to show you what a theorem & proof in real, serious research looks like.
don't mind the "through joining" 
huh interesting...
where did you do your masters btw
just curious
yeah fe
I'm going to probably try the last part now
this time there can only be one weight on one side, one weight off or one weight on the other side unlike last time where it was a binary true or false statement so that's why you can't weigh more than 3^n different weights
Closed by @jade magnet
Use .reopen if this was a mistake.
Send your question here to claim the channel.
Remember:
• Ask your math question in a clear, concise manner.
• Show your work, and if possible, explain where you are stuck.
• After 15 minutes, feel free to ping <@&286206848099549185>.
• Type the command .close to free the channel when you're done.
• Be polite and have a nice day!
Read #❓how-to-get-help for further information on how to ask a good question, and about conduct in the question channels.
I dont understand this explanation. I understand everything up to where alpha is solved
@hoary hamlet Has your question been resolved?
Closed due to timeout
Use .reopen if this was a mistake.
Send your question here to claim the channel.
Remember:
• Ask your math question in a clear, concise manner.
• Show your work, and if possible, explain where you are stuck.
• After 15 minutes, feel free to ping <@&286206848099549185>.
• Type the command .close to free the channel when you're done.
• Be polite and have a nice day!
Read #❓how-to-get-help for further information on how to ask a good question, and about conduct in the question channels.
what am i doing wrong here
using percentages in the equations by themselves
not using variables
its unclear what they represent and also makes the equations invalid
start with
clearly introducing variables and what they represent
here to start a good idea would be to have two variables
to represent:
amount invested at 15%
amount invested at 12%
i think i might know what's happening
wizard might be attempting to reproduce an algebra-free solution to this exact question that i gave some time ago
only he's decided to completely strip away every single word of it
.
this is what i wrote 5 days earlier
you can follow this method, it's fine -- but YOU HAVE TO SAY ALL THE WORDS AND TALK THROUGH ALL THE REASONING. absolute, non-negotiable, hard requirement.
otherwise you suffer consequences.
the consequences are:
- exactly zero people understand your work -- this means even you do not understand it.
- you get things wrong.
3 years,
the interest is get by multiplying 12000 by 12% and 3,no?
