#help-36
1 messages · Page 184 of 1
i initially had it set at 1 in 4000 which gave me a secret normal probaility of 6hrs
that was at a 0.025% chance
the issue with splitting them up is that these times are asking for a secret of a given type
but your question, as i understand it, is just about getting a secret
do you care which type it is?
because if you do, then what type are you looking for? that changes things a lot
so you want either a fire or candy type
yes
so now this is a trickier problem to solve
that means all of the normal period calculations don't really matter to us
(unless your game has a pity system - if it has, please reveal it now)
none
ok cool
that i know of
so our calcullations are going to be a little trickier
but we are sure that during special events, the probability for rarities do not change?
Its only assumed. But based on my gameplay. it remains the same. It only becomes that much rarer to acquire the higher rarities because as you mentioned, the event only runs 15mins out of every 75. So the wait times are 5x longer just based off that?
i don't think we can naively make that assumption. let me look this through.
because the spawn behaviour is deterministic during the event period, which will affect calculations
so how would i turn the alternating spawns in the special event into math?
i'm looking to see if 50/50 is ok. i did say it was earlier because i didn't know you were looking for types as well
let me ask chat gpt lol
how long have you waited for a fire secret?
(or event secret of either type, i'm consolidating them both since they are both event-exclusive)
I have acquired only one after roughly 275 hours of gameplay. The game has a player time counter.
ok so, i might be very mistaken
but from the looks of the problem, the average wait time between event-exclusive secrets (not just fire or candy, but both combined) looks to be about 122 hours
so i'd say given the number of uncertainties we're working with, your experience could very well be normal
remember that we don't actually know the secret drop rate, the actual rarity behaviour during a special event, and the effects of the timers you mentioned the game having to guarantee certain types or rarities
Yes but those rarity timers only exist for the legendary, mythic and supreme rarities. Also i dont calculate them into any of the probabilites because they are guaranteed.
No timer exists for the secret type.
When it comes to the spin wheel it is guaranteed to change the pet in queue to the selected type but the type is still determined by the spinning of the wheel which is rng.
does the wheel spin include event-exclusive types?
yes
but you can only spin it once per hour
it even gives you the drop rates %
so are we taking this into account? because i didn't
Note: the secret listed there is for that specific pet. Not all secret pets.
No i didnt either. I am only discussing it cuz it was brought up
ok wait hold up. those rarity timers are for those rarities, but do those timers have an effect on events?
problem is, while those timers only guarantee that the next pet is one of that rarity, having a pet be guaranteed a different rarity can skew our chances away from secret pets
because every so often you're forcing a pet to be the rarity of the given timer
Yes, if the timer counts down during the event. It will spit out a pet of that rarity. BUT only of a event type or normal type.
It will skew the numbers off a little bit just like the normal conditions but they are removed from the probability calculations because they are guaranteed
they cannot be removed from calculations because they are not of the secret rarity
i mean yes that makes sense
if there was a secret timer then of course it can't be removed, but imagine that every so often you stop rolling the dice and immediately announce the result of the dice
that's exactly what these timers are doing
this will increase the average wait time by a little and i don't have the mental capacity to go through one more round of calculations with the timers, esp when i don't know how long the timers last
will the event-normal type rule still be adhered to if the next pet's rarity is guaranteed by the timer?
for instance, if the previous pet was normal, and a timer expires, say the supreme one, will the next pet be guaranteed a fire supreme?
I said it earlier but only 5 pets are unaccounted for during the 1hour period. Around 25 seconds of unaccounted time. So overall i think the base probability chart would still be 99% accurate
5 pets unaccounted for per hour over 122 hours or more is quite a lot
but like i said, the average wait time will increase probably by an hour or so, by a quick estimation
but it would also screw up my probability if i factor them in because they are guaranteed? Thats not probability it is certainty.
well again, yes, but you have to include them in regardless if you want accuracy
like it would instantly double the supreme rate
which rule?
the event-normal alternating type rule
it won't double the rate per se over a long period of time, but the rate will be slightly higher than at first apparent
Yes it would be forced to be supreme because of the timer, and forced to be fire because of the alternating rule
ok, fair
same for legendary and mythic
yes
either way, if we do ignore all of these effects, i suppose 122 hours is the value i found for the average wait time
so if you want to factor in all of those other factors, i can only estimate for now
Like for example there is currently an exploit with the spin wheel that uses the timer and pet in queue mentioned above. If you spin the timer at just the right time you can force the pet in limbo waiting to be dumped to be of the legendary, mythic or supreme variant because of the timers. and of the types on the wheel.
that would complicate things even more, and i think i'll throw in the towel for this
sorry
Yeah i know but the spin wheel is optional so i dont calculate it . The timers on the other hand are forced
i think i'm still throwing in the towel for that
So would you reckon i gather data including the timer pets?
i suggest you gather data from multiple players, then keep a consolidated view of the overall probability
this helps you scale your data gathering faster
and gather data on every single forced event
timers, events, whatever else
yeah no one is going to help me lol thats why i only did 2 hours
maybe you can ask in a community server for the game
see if anyone's willing to help you out
or heck, maybe someone already figured out the exact drop rates
but yeah as a solo player, data collection is hard
unless you multiacc and somehow collect data off every account at once i suppose
They banned me because i already proved the data for their other game wrong
they were really upset about that
oh.
I will atleast gather the rarity of pets independently on the event and see if they are conclusive with the probability during the normal condition.
all the best
I would have to do it for the same period as the normal conditions i gathered so 8 events worth.
how do i tell the bot we give up
.close
.close
Closed by @nocturne agate
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.
hii kinda curious on why i cant rewrite (x^2 + y^2 -1)^3 = y by moving the cube power to the other side so that y becomes y^(1/3) and then solve for dy/dx. ik the typical and valid approach is to solve this via chain rule but js wondering since (x^2 + y^2 -1)^3 = y wouldnt x^2 + y^2 -1 = y^(1/3) be true as well?
why would they not be equivalent?
uhhh not sure i differentiated x^2 + y^2 -1 = y^(1/3) and got a different answer
then you did something wrong (most likely)
Show us what you did
it’s correct ig
when i apply chain rule the final answer looks different
so i thought it’s wrong
.close
Closed by @loud moss
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.
Can someone help me out wiht this question
I'm not sure where to begin for this problem
wait i got it
im sorry
u can stop typing
.close
Closed by @vernal bloom
Use .reopen if this was a mistake.
does the book have an example of what they mean by "use differentials"
because im curious
i think they are using $\Delta$ and calling it differential
Wumpus Man
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.
Is [ ] notation used for floor or just () ?
@severe elbow Has your question been resolved?
Consider rationalising the denominator
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.
Hi!! I'm here for my friend whos in a probabilities and statistics class and hes struggling with this question
could someone explain all the steps
I am not in this class notably. I am a social work major lol.
i can add him!!
Please do :)
i genuinely cannot wrap my head around how to do standard deviation for probability distribution
and i have to in order to do an exam thats due by 11:59 tonight

chiming in to say that the channel still belongs to andy, so it will close unless he also continues to talk here
thank you i didnt know that one moment
Did you find the mean?
I’ll let him make a new one :3 tysm
Ah alright
Closed by @cedar tendon
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.
How would I do this question? I'm in ap calc ab and we haven't learned L'hopitals yet.
can you interpret this as a derivative?
@little cobalt Has your question been resolved?
The derivative idea is very nice
a different approach is noticing that your numerator is of the form $\sqrt[3]{a} - \sqrt[3]{b}$ which, when multiplied with $\sqrt[3]{a^2} + \sqrt[3]{ab} + \sqrt[3]{b^2}$ yields $a-b$, which will take care of that pesky $x$ term in the denominator
Peter
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.
Hi, is it possible to model the volume of ANY kind of shape? I know volumes of revolution but what about things that don't revolve? Like if I want to find the volume of something wonky like a plushie or a wine aerator, how can I do that?
Use water and measure the displacement maybe
if you are able to describe the region of the 3d shape then yeah you can find the volume as a triple integral
Neat
But I'd need the equation... right?
yes for the exact volume
im not too sure but have you used solidworks before?
they directly calculate the volume for you once you build the object in their system
@rose agate Has your question been resolved?
i believe the most common is monte carlo integration, which never fails to make me laugh:
- bound your shape in a cube of some kind
- pick a random point inside the cube, and check if it is inside or outside the shape
- continue doing this until you feel confident enough
- find the percentage of points that are inside the shape, compared to the total (let's say x%)
- the volume of the shape is then x% of the volume of the cube
I haven't no
How does that work?? I am so confused rn lmao
What if I had too little points in the cube
Wouldn't I need an infinite amount of points?
the idea is: if 3/5 of the points are in the shape, then 3/5 of the shape is probably in the cube
pretty much! from this graph, 10^4 points are needed to be within 0.01 of the true volume
yeah, but you only need to go so far for, say, a metalworks project
lmao 💀
I see
this is why we try to integrate algebraically
normally, this means we break the shape up into easily integrable shapes, then add it all together
oh i didnt realize this was for a project ._. most 3d modelling programs will have some kind of volume thing, like dj said (based on this algorithm prolly)
hope that helps ^-^
if this is for your IA, id recommend modelling a far simpler shape in that case, not a plushie
lmao u were in IB too that's awesome
I am doing it for my iA yeah
I know, I was thinking if the math were less complex I could do that and it would be really cool
@rose agate Has your question been resolved?
@rose agate Has your question been resolved?
@rose agate 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.
Let the two numbers be a and b. If their HCF is 81, what does that tell you about the numbers?
Both are divisible by 81
yes, but mathematically, can you write that?
Yea
Alr 1 min
Is it really necessary to write all that if it's given in the questions already
thats what I wanted you to write mathematically
How do I write it mathematically
no I dont want you to write the alphabets
Idk how to use that maths bot if that's what you mean
if some number n is divisible by say 10, then you can write n=10*m for some number m
thats what I want you to do here
No let the no. Be 81a and 81b
uhh, ok if you want it like that
so, can you use the other bit of information in the problem and use it with this?
81a + 81b = 1215
81(a+b) = 1215
yes
A+b = 1215/81
mhm, go on
What will I do after this
,calc 1215/81
Result:
15
a+b = 15
now can you see what all natural numbers can a,b take?
can you list them?
(1,14) , (2,13) , (3,12) , (4,11) and so on
are you sure all of these work?
Yea
ok, then lets try 3,12 shall we?
Yess
what would be the original numbers in that case?
243 and....
,calc 12*81
Result:
972
Yes
243 and 972 ok?
Ye
and whats their HCF
81
Result:
4
seems to me its 243
why do you think 3,12 did not work?
OOOOOHHHH the numbers should be coprime
yep
How will I write it in exam can you guide me
Like once I get to a+b = 15
What after that
you form all the possible pairs, and eliminate the ones that are not coprime
No statement to be written?
theres only (1,14) to (7,8)
the statement necessary would need to be written earlier
right here
..... Where a and b are co prime?
There you should write Let the numbers be 81a and 81b such that a and b are coprime natural numbers
Alright thanks for the help 🙏
in fact you can also add WLOG Let a<b
What's that (haven't studied log yet)
coz it doesnt really matter if a is less than b or not, but specifying the condition lets you list the numbers properly
so uhh, back to this
So the no. We will get is 1,14, 2,13 4,11 7,8
yes, those are all the possible pairs
Alright, thanks alot
If you are done with this channel, please mark your problem as solved by typing .close
.close
Closed by @lost axle
Use .reopen if this was a mistake.
.reopen
✅
@odd seal will this question also require a similar approach?
Like let a = 12x, b = 12y
12(xy) = 1152
xy = 1152/12
,calc 1152/12
Result:
96
Closed by @lost axle
Use .reopen if this was a mistake.
.reopen
✅
this is not how a product would be
12x * 12y = 1152 means 144xy = 1152
so xy would be 1152/144
,calc 1152/144
Result:
8
and the LCM would be 12xy or 96
Oh
But you might wanna also know that HCF * LCM = product of the numbers = 1152, so you can directly do 1152/12 = 96 to get the answer
Oh yea I forgot that
.close
Closed by @odd seal
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 😭
start small, do you know triangle inequality?
In any triangle, the sum of two sides is always greater than the third side.
Yes, thats one way to do it
yea, both are suitable triangles
its up to you to decide which one you wanna choose to solve the problem
okay the ab+bd>ad
alr
Take 2 triangles you'll get twice of everything then cancel 2 you'll get the needed I equality
(tho keep the side names capitalized, like AB + BD > AD)
ooo okay my bad
so, based on what the question wants you to solve, you have an extra BD and a missing side pair BC and CD
can you resolve that?
no need to construct anything
oh okay
can you check the figure to see if these 3 segments have something special?
smth special like?
and does that help tell you something?
BCD...
mhm
and can you use that here?
Like add it here?
so it forms
AB+BD+BC+DC>AD+BD?
you got BC + CD > BD
If you add AB to both sides, you can still write the inequality as,
AB + BC + CD > AB + BD
oh yesh
Here you got both BD and BC+DC on the same side, but otherwise yea
OH
then
so uhh, you got what now?
Closed by @dry ledge
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 have you tried?
the number could be square of prime number
cuz it have only 3 factors
it is indeed a square of a prime number
but the hint will be:
How does one make "pqrs" into "pqrspqrs"?
but i am stuck at how can i find possible values now
tbh listing could probably do
there isnt much prime numbers between 37 and 100
so pqrspqrs will have prime fractorization: x² × 73 × 137
yeah frrr
only 14
but its wrong answer
yess
thats my approximation tbh so ofc its inaccurate
its fine
and buy Guy's suggestion, you can list out the possible factors
yeah ig cuz it have only 3 prime number
what?
i mean must p, q, r, s be unique
no its not
listing make sense tbh so we could go with that, i cant figure a better way for this
i think i got it
just do 10001*(the sum of every p^2)
hint: list out all possible combinations of
x²(73)(137)
i dont think factorizing 10001 could help
its probably better to leave it there since we dont utilize it anyway
and
lets 73^2 * 73 * 137 = 8
note that there will be a special case when x=73
Cheers!
im so sorry i just reread the problem
alright sorry for any damage i might have caused
@surreal bay 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.
i dont get the solution, like its relying on there already being elements in B, and whether to add it to B or not right?
but arent we starting with B not having any elements (empty set)?
tbh i feel like im misunderstanding the concept of the solution cause like almost every sentence doesent make sense to me 😭
Its perfectly ok for B to be empty at the start. We only use it to count the number of elements of B which x divides
If B is empty then the count is 0
Yes they even wrote that in the solution
Are you asking why does this proof work?
im asking why would you think of the proof, or the idea rather
The main observation is what they said about when introducing a new number to B, it doesn't change the parity for bigger numbers
Olymiad questions are built on observations like this. First you make the observations and then you glue them together for a final solutions.
Closed by @jagged flare
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.
my guess is $|z| < 1$ but i am trying to justify that
artemetra
i have barely worked with complex series tho
by root test i get
$$
\lim_{ k \to \infty } \sqrt[k]{ \left\lvert \frac{2k+i}{k+2i} \right\rvert } \lvert{ z }\rvert = \lim_{ k \to \infty } \lvert{ z }\rvert
$$
artemetra
Why is that obvious?
(2k+i)/(k+2i) converges in norm
R = 1 is the radius of convergence, as you've found, but the root test fails if the limit turns out to be 1
So we are certain the series converges if |z| < 1 and diverges if |z| > 1, but we don't know yet what happens when |z| = 1
does it?
so the terms of the series look like 2z^k for large k
so |z| still can't be 1 right
@stone wagon Has your question been resolved?
.close
Closed by @stone wagon
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 was doing a problem involving a 50 ft long rope that weighs 0.6 lb/ft hanging off a 140 ft building and the work required to pull it and portions of it up. I managed to solve it but I am still very confused.
the total work to lift the rope is 750 ft-lb
but
then it asks for the work requried to lift the top half of the rope and its a quarter of the work even though you are lifting half of the rope and wouldnt the rest of the rope weigh it down?
then it asks for the work required to lift the bottom half of the rope but the integral that is used is different
the function is 15 instead of 3/5x which is what was used to find the top half
why is this the case?
and when they add the two works together is comes out as 562.5 ft-lbs which is what im most confused about. Shouldnt the work required to lift both individual half of rope sum to the work required to lift the whole thing?
I am very confused with these problems
<@&286206848099549185>
@limber narwhal Has your question been resolved?
@limber narwhal Has your question been resolved?
@limber narwhal Has your question been resolved?
<@&286206848099549185>
thank you
lowkey this shit confusing me just a bit
for your last question
I can tell you that calculating just the work to lift the whole thing with one calculation
ok
yea
meaning the less weight you pull
15 comes from .6x25
im still figuring out why though
yea
i dont get why calculating the work to pull the entire rope comes out as 750 ft-lbs but when you calculate both halves and add them it is 562.5 ft-lbs
so
think about it like this
lets say you have a rope hanging right (new scenario)
and it takes 3N to pull it up half way
just using newtons cause why not
bare with me
3N to pull it up half way
so you have the 2nd half still hanging
which means its only going to take 1.5N to pull that up half way
and .75N for the next half
thats whats happening
take 750/2
thats 375
divide that by 2 its 187.5
375+187.5 = 562.5
idk what that is
but basically
15 is the force needed to lift the rope up at the bottom of the first half
or start of the 2nd half
.6lb/ft x 25 ft = 15
the ft cancels and you are left with 15lb
yea
what are you confused on
Closed by @limber narwhal
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.
This is meant to be basic review, but I have no clue what it is and have gotten like 10 questions in a row wrong
I understand what like sin, cos, theta, and degrees are, I am just so confused by all this
took me some time
you can draw it as a square where vector A is the hypotenuse
yes
alright Imma try the next problem, give me one sec
not sure why I got this wrong
lemme show my work, one sec
correct answer was Acos(theta) being negative
negative I get, it's negative x
but Idk why cos
in the last problem
what you were solving for was the y value
do you know the unit circle?
yes
and in this one the x
So you know that on the unit circle the y value is sine
and x is cos
its the same here
wait one sec
because one time it wanted me to use sin for the x I think
which was got me all confused
yeah you use sine for x depending on what angle is given
but in this case
the angle given is at the origin
just like the unit circle
so sine is y
and cos is x
I cooked
right right
different problem
this diff one was for the y
I figured it out now, thank you
just know that the -y changes nothing on the x problem
right I know
but this was for a y problem
oh gotcha
you can just throw things onto the unit circle
for these problems
and its easy
I am so annoyed one sec
I was on the last problem
got it wrong
and now I gotta do another 4
I thought since it's positive Y that it would be positive magnitude
and that it'd be sin because it's y
think about it with sines actually function
its sine(theta) = opposite/hypotenuse
so look at theta
and whats opposite of it
so it'd be cos
yeah
alright
opposite in that case is x
for sine
on unit circle
it would be like 90-theta
and sine would work
alr I got it all done
thanks for the help, that'll be all for today, physics done, time for english
then calc 1 HW tmmrw (kill me)
I recommend you learn the unit circle really well
I'm sure I'll be back then
I've got it memorized
I just hadn't ever dealt with questions like these before
type shit
.close
Closed by @oblique estuary
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.
@gentle zephyr Has your question been resolved?
the way i did it is to suppose || (x,y)-(-1,8) ||<=delta, this got me that |x+1|<=delta, |x-8|<= delta and some bound on |x| and |y|, then i showed |xy+8| < …insert magic… < delta times some constant using these previous ineq
if you want to play osu and chill tonight it’s also fine
no
it’s late here
I wanted to say something too
absolute value is a specific case of the pythagorean distance
correct?
if x=0 or y=0 in (x,y) you can compute it and yeah you will get |x| and |y|
you are not getting my point
||(x,y) - (-1,8)|| is always geq 0
and since delta is stricly greater than 0 we get
0 < ||(x,y) - (-1,8)|| < delta
any pointers ?
we get that f(x,y) = xy where lim xy = -8
from the precise definition of the limit
my first message are the main pointers
why are you equating ||(x,y) - (-1,8)|| <= delta is a strict inequality
🤔
can you deduce from this |x+1| < delta and |y-8| < delta?
min{1,epsilon/10}
I dont think I follow? @dense badge
the goal is to have some inequalities to play with and then try to bound |xy+8| by delta times some constant using those ineq
those two were useful
the manipulation on |xy+8| were kinda tricky
I dont understand the triangle inequality in multivariable
this is correct but the proof i got did not required it
like the ineq is true but might not be useful
how did you get to the point that you want to prove x - 1 delta y - 8 delta
@dense badge
are you using triangle inequality or not?
suppose 0 < sqrt((x+1)^2+(y-8)^2) < delta
then i shower |x+1|<delta and |y-8|<delta
for me that looks like triangle inequlity, no?
yes but later, those two ineq are not from the triangle ineq, both can be deducedfrom sqrt((x+1)^2+(y-8)^2) alone
the end goal is to have delta as a function of epsilon here delta(eps) = min{1,eps/10}
to get that function, one strategy is to get |xy+8|<delta C where C is some constant because then you will be able to choose delta = eps/C to complete the proof
it’s thinking backward a bit
but it work
wait a second my dude
suppose 0 < sqrt((x+1)^2+(y-8)^2) < delta
then i shower |x+1|<delta and |y-8|<delta
how did we get here ?
why is delta in terms of epsilon?
ok do you agree that |x+1| <= sqrt((x+1)^2) <= sqrt((x+1)^2+(y-8)^2)
the last ineq comes from the fact sqrt() is increasing and (y-8)^2 is always greater than 0
these ineq also make sense geometrically
if you think about your disk of radius delta centered at (-1,8)
i would say it’s tricky math
there are a bunch of trickshots and techniques to bound stuff
absolutely terrifying exams where you need creativity
k the first trick is to add and subtract y as in |xy +y -y +8|
then you have to bound this to be able to use |x+1| < delta and |x-8| < delta
what do i do with 8 - y
that’s -(y-8) and the absolute value can eat the -
triangle
yes but
but what?
I am still hardstuck
oh, here is where absolute value kills negative
ohhh
I see it now
yes!
unfortunately (|y|+1) is not constant 😭
we need to write |y| in terms of some constant
sum and subtract 8
and then bound by delta
haha! triangle inequality again
yours is a much stronger statement
is crazy that we both conclude almost the same thing in diferent ways?
have you taken anal?
yes of course
is it painful?
i said it before, it’s this kind of tricky stuff
you say its tricky but you are very well versed in it
at least on basic calculus, u know?
meh, at my uni we have like a sequence of 3 course on analysis then i took some extras because i kinda enjoy more this kind of pure math
like an intro to functional analysis and a intro to complex analysis, both went over my head but i mean it was cool while it lasted kek
what do you get when you use |y| < delta + 8 in the sequence of inequality from before?
that’s where the min(1, smt) came from
we have to force an extra bound on delta to simplify this last ineq further
this can be done with a min when defining it
wdym?
like wouldn’t it be nice if delta was ALWAYS less than 1
then we could say that delta(delta+9) <= 10 delta
because i can make this true when defining delta in term of epsilon
using delta(eps) = min(1,smt)
Suppose |xy + 8| < 1 , then we get that we can define a delta such that both conditions are covered
that’s not the same as using this, btw you assume there that eps=1. Once you are convinced of my delta(eps)=min{1,eps/10} you can just plug eps in it.
how do I find that delta
what i’m proposing is to defined delta so it’s always the case that delta < 1
so that delta(delta+9) <= 10 delta is always valid
no, I understand
I completely understand
we have
|xy + 8| <= delta(delta + 9) < epsilon
and we need to define a delta(delta + 9) < 1
well, we dont know that |xy + 8| < epsilon
but if we define delta correctly
we will get there
how do I find
delta(delta + 9) < epsilon?
delta = epsilon/(delta + 9)?
delta is positive so this never gets to division by 0
delta can only be defined in terms of of epsilon here or else with what you have now it’s like recursive
in a way the trickery with the min can be avoided for this particular exercices because they only want you to get deltas for eps=1 and another one
delta(delta+9) < 1 is a quadratic inequality isn’t it
you could just check when x(x+9) < 1
?
is that a typo?
i mean is tricky we dont need to use wolfram
cant we draw this inequality in the number line?
yea
oh now that i think about it i would just throw in the quad formula get the roots and deduce the interval
but yeah it’s kinda ass
the trick with the the min{} is cleaner in this regard
,w x(x+9) < 1
k when you put 1 in the min when defining delta, you can use in your proof that delta <= 1 always
welcome to the gulag
is not good enough because we need |xy + 9| <= delta(delta + 9) < 1
we are not allowed that delta = 1
true but this is the constraint imposed by the min not the euclidean distance
if you let delta = min{1, smth} and smth > 1 then we get delta = smth
and since |xy + 9| <= smth^2 + 9smth < 1
we get an absurdity, no?
maybe think of delta = min{1,eps/C} this way, i need delta to be less than 1 for this last inequality i.e. the delta(delta+9) <= 10 delta to hold AND i also need delta to be less than eps/C a certain constant to finish the eps-delta argument
if delta <=1 the delta+9 <= 10
unfortunately no, but can pick delta=eps/10 to end the proof in that case
go back to the main proof, this
just the delta+9 you want to keep the other delta to transform it into a epsilon
dude what the fk is going on, I dont get it
this is the strat i was yapping about at the start
it’s like 1:16 am here
so not the prime time for an eps-delta argument tbh
we can stop here, unfortunately the proof is like very close from complete
worst case you have the quadratic inequality from earlier
yeah I sort of understand but is tricky
in general, i think a good way to do eps-delta proofs is to do them two part, 1 the scratch work where you get to the point of defining delta in terms of of epsilon and then a cleaner rewrite of the argument with the chosen delta(eps) to see it always work
I think we can leave it at here
but you have most of it and imo you learned some tricks of the trade +smt -smt, the ineq with distance etc.
I just hope we dont getan exercise like this in my exam, I thought I liked hard problems in the exam
but anal is a beast on its own
also this is like 3rd week of uni and its eating me alive
killed my motivation that theres always someome who is stronger than u, in preuni it was different
unsolicited advice here, imo comparing with others sometime makes you think you are better than you are or conversely that you are shittier than you are.
people have different backgrounds, tbh raw ability and exposure to this stuff
I appreciate the help, I will try to put all the pieces together tomorrow morning, is almost 3am here and I wasted almost completely my friday, I was supposed to be having fun in osu dude, not doing a epsi delta proof at midnight, you know what I mean, is not like I didnt had fun, but sometimes the creativity steps made it so tiring and we spent like at least 2 hours at this point in a single exercise dude, mostly because I didnt understood your hints, is crazy what undergraduate degree does to a human
yeah mathcord is a rabbit hole on that kind of shit
i should have waited to see you tomorrow in the help channels on this one
anyways gn my dude
.close
Closed by @gentle zephyr
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.
$\lim_{x\to 0}\frac{e^{2x}-1}{x\sin(x)}$
ihave<skissue>
my intuition tells me this is undefined
i tried lhospital cause its 0/0 and i got
$\frac{2e^{2x}}{\sin(x)+x\cos(x)}$ which im not too sure how to find the limit to 0 either?
ihave<skissue>
lim x-->0 (sinx/x) = 1
use the fact on the question itself
you can lhop again
wasnt that only for indeterminant form..?
what do the num and denom now approach
$\lim_{x\to 0}\frac{e^{2x}-1}{x\sin(x)}=\lim_{x\to 0}\frac{x(e^{2x}-1)}{x^2\sin(x)}=\lim_{x\to 0}\frac{e^{2x}-1}{x^2}=\lim_{x\to 0}\frac{e^{2x}}{x}$
ah shit yeah
[michael vsauce voice] or can you?
my bad
oh wait
dont listen to my bumbling ass
ihave<skissue>
oh and e^2x converges>0 while /x diverges, so its undefined?
your doing limit!!! pog
i think so
yeah , you can also use expantion after the 3rd step , if you have covered it
thats what i was about to type because thats what i did 
taylor?
yes
snipe
e^x = 1+x+x^2/2 + \dots but as x->0 its enough to consider only the first 3 terms
yeah cause the denominator has x^2 , terms after x^2 in e^x , i.e. x^3 , x^4 , will just vanish
Closed by @jagged flare
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.
pls help, I have no clue how to get the zeroes of this function
Zeroes?
yeah, isnt that how I'd get the coordinates of the points that intersect the x-axis?
Oh i got u
See on axis the value of y=0
Do we agree on that
yeah
are you supposed to use a graphing tool for this question
well, you can always make your own graph by maybe plotting g at integers and sketching
We can consider g(x)=y
that might take too much time, I get 1.5 minutes per question
yup
So the whole expression=0
okay
Can u factorize it?
no linear factors over rationals
yeah evaluating small integers is quite helpful
g(-1)=3
g(0)=1
g(1)=-1
g(2)=-3
g(3)=19
i think basically you want to numerically get the roots
Yea doesnt
@loud steeple what does this series of calculations tell you where the zeroes are?
0 and somewhere between 2 and 3?
whoops typo, i calculated g(0) wrongly
but yeah, the main idea is you can narrow down to find the roots
somewhere between 0 and 1, and 2 and 3
it's multiple choice so that might be the intended method
alternatively, newton method out the roots, if you have access to a calculator
I do have access to a calculator
then you can very well newton method out the roots
newton method?
In numerical analysis, the Newton–Raphson method, also known simply as Newton's method, named after Isaac Newton and Joseph Raphson, is a root-finding algorithm which produces successively better approximations to the roots (or zeroes) of a real-valued function. The most basic version starts with a real-valued function f, its derivative f′,...
not sure if you were taught this before
though i feel this might be one of the more difficult questions that you can afford to spend slightly more time on
at the cost of spending slightly less time on the easier questions
yeah it took too much time when I was doing it the first time I just had to guesstimate the answer and move on
yeah so i would guess the two roots are near 0.5 and 2.5, and newton method it out, I would get...
,calc 0.5-(0.5^4-20.5^3-0.5^2+1)/(40.5^3-60.5^2-20.5)
Result:
0.78125
,calc 0.78125-(0.78125^4-20.78125^3-0.78125^2+1)/(40.78125^3-60.78125^2-20.78125)
Result:
0.72352259659614
,calc 0.72352259659614-(0.72352259659614^4-20.72352259659614^3-0.72352259659614^2+1)/(40.72352259659614^3-60.72352259659614^2-20.72352259659614)
Result:
0.72125934712853
about 0.72 for one root, and you should be able to get the other root pretty fast
yeah that makes it easier
with a calculator where you can iterate a calculation (there's certain scientific calculators that let you do that), it's much faster
yeah okay I got the answer as D, which definitely took more time than it should've but this new newton's method thing is pretty neat lmao
getting the first root definitely helped a ton
when you are done with this channel you can .close
lemme check out the other root tho
,calc 2.5-(2.5^4-22.5^3-2.5^2+1)/(42.5^3-62.5^2-22.5)
Result:
2.371875
,calc 2.371875-(2.371875^4-22.371875^3-2.371875^2+1)/(42.371875^3-62.371875^2-22.371875)
Result:
2.349262539011
,calc 2.35-0.72
Result:
1.63
yeah
Closed by @loud steeple
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.
!status
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
write
well not midway exactly
*right
js at the beginning
well you can see sin and cos in the identity
so it would make sense to write tan theta as sin theta/cos theta right?
well theres also p^2 and q^2 terms so better to multiply both sides with p/q
do you know about componendo and dividendo?
no-
hmm well
when we are given a/b = c/d'
this implies a+b/a-b = c+d/c-d
this is called componendo dividendo
can you see how we can use it here?
im not sure if im supposed to use that-
yes
so p sin theta/ q cos theta = p^2/q^2
or you can divide both sides by cos
right?
how huh
i multiplied both sides with p/q
p tan theta - q / p tan theta + q when you divide both sides by cos
so p/q=p sine theta/ q sine theta?
