Let k=√(100+√n) +√(100-√n) k²=200+2√(10000-n) Since k is an integer, k² (perfect square) is also an integer. And since 200 is also an integer, one way to make this true is to make √(10000-n) an integer, which makes 10000-n a perfect square. Hence, set m²=10000-n, mEz+. So we have: k²=200+2m (1) and n=10000-m² (2). Notice that in (1), the RHS are sum of two even numbers, that means LHS is an even perfect square. From(2), notice that minimize the value of n will maximize the value of m and k and since n>=1 => 10000-m²>=1, then 0=
Oops I was assuming things had to stay as integers throughout, so I was assuming the sqrt(n) was also an integer. If you go that way, I came up with n=9,216 which sums to 16 in the original equation.
@@blackpenredpen Yeah I guess I was just assuming the left term and the right term were both integers already. Because, what are the odds that they would be non-integers and then exactly cancel out to make an integer once you add them up? And yet that's exactly the solution you presented. How cool!
That shouldn't be weird. If you take any number of the form x = a+b√n you can rewrite it as x² = a²+nb²+2ab√n x = √(a²+nb²+2ab√n) So, any integer can be written as sum of radicals of this weird form. For example, 4 = √(15+4√11)+√(15-4√11) which I obtained just calculating (2±√11)² Such numbers and similar expressions became popular after Cardano's formula to solve cubics. Indeed, a third degree and complex version of the previous example is the solution of the classic x³=15x+4 The Cardano formula for depressed/reduced cubics x³=3px+2q is x = ³√R(+) + ³√R(-) with R(±) = q±√(q²-p³) You can easily check that such x is ONE solution for the cubic. For the cubic x³=15x+4 we have p=5, q=2 so R(±) = 2±√(2²-5³) = 2±√-121 = 2±11i Such numbers have as one of their cubic roots 2±i, since (2±i)² = 4-1±2*2i = 3±4i (2±i)³ = 6-4±(2*4+3)i = 2±11i making x = ³√R(+) + ³√R(-) = 2+i + 2-i = 4 which satisfies 4³ = 64 = 15*4+4 By the way, this is also the actual origin of complex numbers. I hope I didn't make any mistake.
Wow I have not seen your videos since I was in college and your channel was so small. Now you have hundreds of thousands of subscribers. This is amazing. Congratulations
Solved the bonus problem by doing a substitution: n = -t, which then turned, considering that i^2 = -1, the final equation into a k^2 = 200 + sqrt(100^2 + t) where I have to find the smallest non-negative t. Using the same method I found 22 as the first smallest potential k (after 20) and, respectively, t = 10164 => n = -10164
I found the solution a different way! It is possible to rewrite √(100 + √n) = √a + √b, where a and b are integers and a>b. If you square both sides you get that whatever n is, it can be written as 4a(100-a). Writing the sum with our new numbers we get that the sum of the two terms is just 2√a. For this to be an integer, since a
for the common types of problems involving sqrt[a+sqrt(b)]+sqrt[a-sqrt(b)], i usually find out first if i can get something in the form of sqrt[x+2 sqrt(y)], then figure out if i can factor it using (a+b)^2, and if i can i can write it as sqrt x_2 + sqrt y_2, and similarily, i can do the same thing with the other part of the equation such that the y_2's cancel out giving us 2 sqrt x_2
I found the same result and i want to explain my solution. Let me call the expression f(n) and substitute n with x belonging to R so that f(x) is a real CONTINUOUS function. Existence conditions implie that x must belong to the bounded interval [0, 10000]. f(0) = 20 whereas f(10000) = sqrt(200) that is similar to 14.14 Now, f(x) is monotonous decreasing ( HINT: study the positivity of the derivative function) and this means that 20 is his maximum value and 14.14 is his minimum value. Thanks to Darboux's theorem ( or the intermediate value theorem) we can claim that f(x) attains every integer value from 20 to 15. From 20 to 15 x will increase from 0 to 10000 since f(x) is monotonous decreasing. So compute the counter image of K from 19 to 15 and if x is an integer we stop the algorithm, else the result is impossible. To compute x knowing K we write the inverse function: x= 10000-((K^2-200)/2)^2. f^(-1)(19) is not an integer, we pass to 18. f^(-1)(18) is an integer and we find that x=6156. This implies that x=6156 is the smallest positive integer which makes f(x) an integer. The proof is now concluded. I hopefully you appreciate my demonstration and if I made mistakes please don't hold it against me, I'm only an italian mathematician that wanted to share his solution with the global community ^.^ All the best, D.M.
Other way to solve its calling sqrt(100+sqrt(10))= b then b+sqrt(200-b²)=N then by using the quadratic formula we get (N+sqrt(400-N²))/2 then just search for the lowest N value with b= some natural number +sqrt(natural number) as we want b to be the least start with 19 and then move on to 18, and thats it with 18 you get 9+sqrt(19)
I did the same method lol and i found that expression gives values that are less or equal to 20 and when the expression is equal to 18 then it gives the lowest N ...and N=
My answer before watching the video is 96^2 My approach is to let the expression equal to x and then make x^2 a perfect square, I got in the form 2(100 +√(100^2-n) so I concluded 50 + √(50^2 - n/4) must be a perfect square. Now since the √ is always +ve, I checked for the first perfect square after 50, which is 64 and that gave an answer for n, which was 4(50-14)(50+14), then I simplified a little and wrote in terms of power of primes (luckily it only contained powers of 2 and 3) to solve for √n, and I got it's value 96, which I verified that it makes both square roots perfect square so answer is integer without simplifying But the value of n I got is pretty large, so I suspect I may have made a mistake somewhere and missed a smaller solution, but I'm almost sure my answer is correct (Unless you remove the positive integers and change it to non-negative :P)
Wow, turns out he didn't realise he had to make that expression a perfect square and mistakenly just solved for it to be an integer. So, he didn't get it right on his first try. Yay!
Never mind, I just realised my mistake halfway through the video, the answer is actually 4(50-31)(50+31) or ((18)^2)(19) which gives after simplifying gives (9+√19)+(9-√19) = 18 Earlier, I actually mistakenly solved for the largest value of n, because I solved for smallest value of x, in fact I have to solve for biggest value of x to get smallest n
Regarding the bonus part, the biggest negative integer is the negative integer closest to zero. If there are no other possible solutions between the claimed -10164 and -1, inclusive, then -10164 would be the biggest (assuming it works).
I think I've actually solved this by little logic and trial error method. Since [ 100 - sqrt (n) ] must be positive perfect square, sqrt (n) < 100 Trial error values of 100 +- sqrt (n) fails for squares of 11, 12 and 13, more specifically 100 - 21, 100 - 44 and 100 - 69 (as these differences are not perfect squares). But it succeeds for sqrt (n) = 96, as both 100 + 96 and 100 - 96 are perfect squares. Hence n = 96^2 = 9216 EDIT 1 : I tried this orally without using any calculator as I thankfully remember perfect squares till 111. EDIT 2 : I checked the solution and was sad to know my answer is not correct here. No issues. It would have been correct for a rational square root result. Anyway. I enjoyed trying this one.
@@mojtabasaleh8842 It's wrong because square root of 100 +- sqrt (n) can be irrational, which on addition, result in an integer. That's something which we didn't think of.
The problem with the tournaments is that their purpose is to trip people up on mental calculations (for the most part) rather than any rigorous type of math involved. Seeing or scratch writing 10000, might as well be shorthand to 10k or 1kk, or any other moniker.
lets consider n=4k than (root/100-,+//n)=(/100-,+2//k) we know that if a+b=100 and a*b=k we can write that as /a -,+/b a is bigger than b so we can get positive number(for the negative one) if we do all these we get (/a+/b)+ (/a-/b)=2/a so a must be x^2 and a=81 and b=19 perfectly fits into this so 2/81 =18 n=4*81*19=6156 ''/'' = square root
Other solution: If we have sqrt{a+\sqrt{b}}=sqrt{x}+sqrt{y} then sqrt{a-sqrt{b}}=sqrt{x}-sqrt{y} With x=frac{a+\sqrt{a^2-b}}{2} and y=frac{a-sqrt{a^2-b}}{2} [the proof is easy, try it] So, in this problem we have that the sum is sqrt{100+sqrt{n}}+sqrt{100-\sqrt{n}}=2(sqrt{x})=sqrt{4x} in Z With x=frac{100+\sqrt{10000-n}}{2} and 4x=200+2\sqrt{10000-n} and how do we know that $4x=k^2\inZ$, we continue as bprp.
@@blackpenredpen typically computer scientists. and i have no idea why they ever fumbled into that obnoxious definition. of course the naturals should start with 1. 0 is hardly "natural". didn't it take life something like 4.5 billion years to even conceptualize it? whereas the positive integers took merely 4.49999 billion.
@@sharpnova2 When I was a computer science major, I never heard anyone there talk about 0 as a natural number (I don't think they even used the term). From what I've heard, it's more of a geographical thing: 0 is often a natural number in Europe, but not in the U.S.
This was how I approached it by looking at cases where n was a square number. Turns out that isnt the best solution, but I kinda just assumed it would be the case. n=9,216 is nice and neat in that things are integers at every step.
@@Harlequin_3141 Yeah when I was in the gym I wasn’t able to square it all in my head and work it out that way so at first I was just looking for numbers that fit and found that. But later when I got home I managed to reduce the equation and get an inequality that showed the only possible solutions were 6156 and 9216. Also n=9216 makes the equation it’s smallest possible integer solutions so I prefer it.
A quick python script to verify the answer : from math import sqrt j = 0 while True: j += 1 f = sqrt(j) s = sqrt(100+f)+sqrt(100-f) if s == int(s): break
Solution to the question at the end: With the same approach as in the video we get k² = 200 + 2sqrt(100² - n), and for the sake of (my own) better understanding, let -n = m; in which case we're looking for the smallest positive integer m. so k² = 200 + 2sqrt(100² + m), if m was equal to zero, k² would be 400, so we're looking for the perfect squares bigger than 400 but as small as possible 21² = 441 doesn't do the job, as k² is also even. take 22² = 484 then 484 = 200 + 2sqrt(100² + m) 284 = 2sqrt(100² + m) 142 = sqrt(100² + m) 142² = 100² + m m = 142² - 100² = (142-100)(142+100) = 42*242 = 484*21 = 9680 + 484 = 10164; therefore n = -10164 plug this into any programm able to work with complex numbers, you get an integer in the original expression.
The answer to your challenge question:- (-6156) because let n = -m, where m is positive int. when we find for m, for smallest, it is similar to solving for -n for biggest. If I am right, pls give a heart!!
There is a related approach as well - consider 100 + \sqrt(n) = ( a + \sqrt(b) )^2 for some a,b. Then, 100 - \sqrt(n) = ( a - \sqrt(b) )^2. It follows that 100 = a^2 + b while n = 4 a^2 b = 4 a^2 (100 - a^2). However, there are two conditions for this to hold: 1) a > \sqrt(b) else the \sqrt(100 - \sqrt(n)) = \sqrt(b) - a and the result is no longer an integer. This also means a^2 > b = 100 - a^2 => a^2 > 50. 2) b > 0 (This is to ensue n>0, and for the bonus can we changed to b < 0). This gives a^2 < 100. Given n is a decreasing function in 50 < a^2 < 100, choose a^2 = 81 and correspondingly n = 4 x 81 x 19 = 6516. The solution for bonus is similar with b < 0. which corresponds to a^2 > 100. Choose a = 11 to get n = - 4 x 121 x 21 = 10164. (Assuming the square root with positive real part is considered) In fact, the set of all admissible n corresponds to all a^2 >= 50 or a >= 8.
First examine all singular digits. Square every one. 0^2 = 0 1^2 = 1 2^2 = 4 3^2 = 9 4^2 = 16 5^2 = 25 6^2 = 36 7^2 = 49 8^2 = 64 9^2 = 81 We define an even number to be a number divisible by two with no remainders left and we define odd to be a number when divided by two has a remainder. Notice that every odd number produces an odd number when squared. This is why if we work backwards that we can conclude that ‘k’ is even. I would test bigger numbers to prove this to you further but it satisfies for every number because the only way you will get an even number is if your last digit is defined as even. i.e: 100^2 = 10000 and has no impact in the ones place. 102^2 = 10404. 102 has impact in the ones place because there exists and number in the ones place for 102.
Both expressions k(n) and k²(n) are strictly decreasing and admit no local maximum (the maximum is n = 0). Which is more or less what BPRP did when he said k=20 and which is wrong. Then your method would find the closest n that gives k an integer, which would be again equivalent to what he said.
@@PackSciences yes i think you are right...the graph does not feature a local minima and maxima is at 20..so by my method it is impossible to find an integer. I guess hit and trial is the only way to solve it
My solution was different, but incorrect, because of an incorrect assumption I made. Moreover, I did this algorithmically rather than mathematically. I made a couple of assumptions based on the fact that our result, k is an integer and n is a positive integer: 1. n squared must be less than 100 2. each term of k individually must be an integer (an incorrect assumption based on the given facts) Then, based on this, I introduced a new term, p = sqrt(n) Now my objective was to find an integer, p, at an equal distance from 100 in both directions, landing on a perfect square. This gave me p = 96, or n = 9216, which is a solution, but not the answer to the question, as it is not the smallest value of n. This as all based on the incorrect assumption that p would be an integer, stemming from the assumption that each term of k would be an integer. This method gives me k = 16, however.
It's weird how all the videos in english say that zero is not positive nor negative, while in french we learned it was actually both. In a question like this it would say n is strictly positive, if it only said positive then zero would be an acceptable answer.
The idea of the solution is quite easy but not making a numerical mistake along the way was impossible for me. I came up with a slightly different tactic from yours but I had to calculate the thing like three times until I eventually uncovered all the mistakes (things like 2x199=399 or whatever). I suppose simple 4-operation calculators aren`t allowed on the exam, right?
*Related Problem* Prove that sum of [ C(4k-2, 2k) / (2k-1) x (n/20^4)^k ] from k = 1 to k = Infinity equals 1/20 where C(n,r) is binomial co-efficient function. You will find that n = 6156
There is quiet a few natural numbers that makes the expression a positive integer. 6156 give integer because (9±sqrt(19))^2 = 100±18sqrt(19)=100±sqrt(6156)
I missed that it's about finding a positive integer for n, so I stopped once I found the trivial n=0 case. But the method I used easily spat out 6126 (which interestingly enough when subtracted from 10,000 gives exactly 62 squared).
More generally, given a and any k such that a/2 ≤ k² ≤ a, then n = 4k²(a-k²) makes √(a+√n) + √(a-√n) an integer, namely, 2k. In the case above, a = 100; so 100/2 ≤ k² ≤ 100 implies k = 8, 9 or 10, making n = 9216, 6156 or 0 and √(a+√n) + √(a-√n) = 16, 18 or 20. n = 0 is explicitly excluded, making n = 6156 the minimum. (9216 = 96², in which case √n is an integer, 96, whereas 6156 = 4·9²·19 makes √n = 18√19.)
I hope someone makes a general case for the cube root video. I’m quite interested that that thing just simplifies to 2, so I’m hoping to discover more of such.
@@MathSolvingChannel so here are two specific examples of the problem: ua-cam.com/video/_6Pl_MNphf4/v-deo.html ua-cam.com/video/5pa1AryylpM/v-deo.html These are specific examples. How do I generalize/generate other examples with other numbers inside the nested roots? Any leads would be highly appreciated. Thank you!
@@kimveranga The general way to solve this type of problems is to use substitution. Let a=1st cubic root term, b=2nd cubic root term. so you will get a^3+b^3 = some integer and a*b=another integer, usually. then you use the complete cube formula to convert them into a cubic equation respect to a+b, then you solve this cubic equation. I also did two videos on how to apply this method to solving this type of problems. and here are the links: ua-cam.com/video/0P6L581BxuI/v-deo.html ua-cam.com/video/UbLWWNgp-sE/v-deo.html
@@MathSolvingChannel I’m quite familiar on how to solve those things. My question is how do you even come up with the square roots inside cube roots problem like this. How do you generate these kinds of problems? Thank you!
@@kimveranga Let me confirm if I understand your question correctly, do you mean: why they put this form of square root expression inside this cubic root? why not they put others such as 11+sqrt(97), or in general speaking, for what kind of square root expression this problem is solvable?
I think before concluding "k is even" you also have to prove that you can't choose an n so that the square root comes out to a half number (like 80.5) which when multiplied by two comes out as an odd integer
Good catch, but it turns out this is always impossible. The square root of a positive integer is either an integer or irrational (look up the proof of the irrationality of 2, it's a classic!)
I'm only in 9th grade, and I do these equations so I can get ahead in math, I want to go to MIT, thankfully you have helped me expand my math knowledge. These problems are difficult but honestly I understand them a little, I was off to a good start but then got lost about halfway through trying it. Watching you do it made it make more sense though.
I decided to take this a step further and created a small Python script to find all possible values. The value n must be greater than 0, but less than or equal to 10,000. There are only two values, the one he found, and 9216.
What's wrong in: let k=√(100-√n)+√(100+√n) Squaring both sides(and skiping a step or two) K²=100-√n+2√(100²-n)+100+√n K²=2(100+√(10,000-n)) Rearranging to get: (K²/2)-100=√(10,000-n) Squaring both sides to get (K⁴/4)-100k²+10,000=10,000-n Rearranging to get 100k²-(k⁴/4)=n You can try finding integer solutions from here but it doesn't work out. I don't know where the mistake is.
Ok I got the right answer after seeing your working out.(also pretty surprised seeing that everyone's first step is the same. I thought I did something horribly wrong in the beginning but turns out it's the same as everyone.)
I tried to find any value of n that will yield the integer in the expression and I found 396. sqrt(100+sqrt(n)), putting n=396 sqrt(100+sqrt(396)) = sqrt(100+2sqrt(99)) = sqrt(1+2.1.sqrt(99)+(sqrt(99))^2) = sqrt((1+sqrt(99))^2) = 1+sqrt(99) In similar fashion, sqrt(100-sqrt(396)) becomes 1-sqrt(99) now , the given expression becomes 1+sqrt(99)+1-sqrt(99) =2 , which is an integer. I do know that second term can be sqrt(99)-1, in that case the solution won't work but I don't see any reason why cant we put 1-sqrt(99). Is it because the term becomes negative? I'm really confused.
Why isnt 361 a vakid answer? Root of 361 can be +19 or -19. If you choose appropriate roots, you can get 18 as the final answer. Can somwone explain it to me please??
Hello blackpenredpen, I would like to ask you if you accept user submissions with regards to question and papers and if you do, how we can contact you. I love the questions you put on your channel and allows me to practice my craft, thank you so much.
Hi Justin. I am quite busy this semester. You can send the problem to my email blackpenredpen@gmail.com but I am sorry that cannot promise if I can make a video on it soon.
@@blackpenredpen Stay safe and keep on bringing the videos! Old timers like me who haven't studied rigorous theoretical math or even basic algebra in the past 7 years still like to keep up with things. I ended up going the biochemistry route, but I still have a longing for that math itch from my 2nd degree in pure math.
What I'd probably do is at first, deduce that n must be a square number and 100-sqrt(n) AND 100+sqrt(n) must be a square number. then plug in sqrt(n)=1,2,3... Since sqrt(n) cannot be equal to 10, it will be one of the values from 1-9
Thanks for this problem! I worked the dual on my channel, i.e. found largest natural number, n, such that sqrt(100+sqrt(n)) + sqrt(100-sqrt(n)) is positive integer. Got n=9216 which is in the domain=[0,10000]
I did this and got 9,216, which is wrong, but I’m still happy because it is the smallest value that also remains an integer when taken the square root. I did consider that maybe the answer would not necessarily abide by this rule but I didn’t really know how to solve the problem if that was the case.
2:19 you are talking about he just tried a little troll or a showed a little little mistake but the question is why he said that so let me explain:- integer + integer is always integer but the square root of various integers are irrational (which is not integer) so if the square root part comes out to be integer we wont care about rest it will be integer which will be answer but the he said that there is square on answer which means we are not getting integer hence not the answer most importantly he knew it was closest perfect square (n should be integer) he set them both equal and square rooted both sides and equated(rest explaination is mentioned in this reply by me)
Because we're looking for the smallest possible n>0, so we're looking for the largest perfect square less than 100^2 (to get an integer out of the radical), which would be 99^2.
I'm a bit confused, granted I am well out of my range of knowledge. Doesn't √(100+√n)+√(100-√n) rearrange to 10+ √√n + 10 - √√n and then the √√n cancel out to just leave twenty?
So were you allowed a calculator for this problem, and if not then how would you know thats the correct solution when you plug 6156 into the initial equation?
Well it's not zero. I guess n is a square since the root of a root is irrational. For root of n, I use x. The value of x must be between 1 and 100 or else we obtain complex solutions from 100-x being negative under the root. The larger root will be 101 to 200. There are four roots in this range: 121, 144, 179, 196. Of these we search from smallest to find the first which is also a square at 200-x. It can't be 121, because 200-121 = 79 and that isn't a square of an integer. Likewise, we eliminate all but 196, because 200-196 = 4 .... an integer square as 2². So we now see that x is 96, because √(100+96) and √(100-96) are integers. Since x is the root of n, I think that n = 96²
how is it clear that 20 (n=0) is the biggest value of k? i don't see how k is strictly decreasing as n increases. i plotted the graph on desmos and i can see that it is decreasing, but how could one tell algebraically/analytically? EDIT: ah, i can see that k^2 is strictly decreasing as n increases. is that enough to determine that k is also strictly decreasing though? 2nd Edit: yes, because k is always positive, so a decreasing k^2 implies a decreasing k!
can you explain how those two irrationals make a rational. because it is not like adding φ and 1-φ. they are two different irrationals. what part of those numbers is the one canceling out? is that a stupid question?
@@ipcheng8022 is there anywhere you recomend to look up for this one. I am an engineer so I am not familiar with where you go to discuss math. I haven't done math for the degree for 3 years now so yeah that didn't help either
@@hellNo116 i am an engineer too, so i tried it on Python. when the two terms in the original equation are seperated, both of them have a lot of decimal places. First term: 13.358898943540673 Second: 4.641101056459326 Here is the thing that I cant figure out why: the decimal places were aligned too perfectly, so that all of the sum at the decimal places are nine. I believe this is not a mathematical issue but a technical issur, where computers automatically round up the 17.99999999999 to 18. On the other hand, I cant figure out why the result is so close to 18. There is no terms to should be cancelled due to the additional square roots applied to the two terms. It could just be coincidence, or this is some math theories that we dont know about. This is why I strongly support that 9216 should be the correct answer because it can be proven for sure that results of all terms are integers
@@ipcheng8022 rounding errors are thing. they don't always round that way. it has to do with the printed versus the saved value of the number. also it is really depending on how the computer is programmed to deal with them. Also that is an approximation. My question isn't if the approximation add up, but how the irrational parts cancel out
@@hellNo116 i cant find any other similar cases or related theories that can explain this phenomenon, which is why i can only explain it from a technical view.
Don't you assume that the function is descending at all times? Otherwise the largest output wouldn't necessarily mean you get the lowest input, and you also can't know that the range of the output value is less than 20. Shouldn't you prove this?
My try: let's say that the given thing equals k. k^2=200+2sqrt(10000-n). Since k is an integer, k^2 should be a square number. n is a positive integer, and k^2 must be smaller than 400. The largest square number smaller than 400 is 19^2=361, however this must be even and it is 18^2=324. This means, sqrt(10000-n)=62, 10000-n=3844, therefore n=6156. Uh, I didn't copy the video but I solved it in an identical way, I guess.
0:31 if you did pause
👇
Let k=√(100+√n) +√(100-√n)
k²=200+2√(10000-n)
Since k is an integer, k² (perfect square) is also an integer. And since 200 is also an integer, one way to make this true is to make √(10000-n) an integer, which makes 10000-n a perfect square.
Hence, set m²=10000-n, mEz+. So we have:
k²=200+2m (1) and n=10000-m² (2).
Notice that in (1), the RHS are sum of two even numbers, that means LHS is an even perfect square.
From(2), notice that minimize the value of n will maximize the value of m and k and since n>=1 => 10000-m²>=1, then 0=
@@mathmathician8250 You copied my solution :P
Bonus Q answer: n=-10164 solution steps: ua-cam.com/video/rYT_GEO0UOw/v-deo.html
@@anshumanagrawal346 Nah I didn't it took me 30 minutes to type in on my phone :)
Don't see in 2x speed at 0:46
The solution is simple: *define 0 to be a positive integer*
😆
🤨🤨🤭😜😅🤣🤣🤣🤦🏻♂️clever
This person is going places
Lmao
@@benedictchoong Which places?
Oops I was assuming things had to stay as integers throughout, so I was assuming the sqrt(n) was also an integer. If you go that way, I came up with n=9,216 which sums to 16 in the original equation.
That is cool! It would have been more fun that way 😆
Did the same thing!
I had the same error, thought that sqrt(n) was an integer and had n=9216, glad to see that i wasnt the only one 😅
@@blackpenredpen Yeah I guess I was just assuming the left term and the right term were both integers already. Because, what are the odds that they would be non-integers and then exactly cancel out to make an integer once you add them up? And yet that's exactly the solution you presented. How cool!
I did the same
It is very weird that the squre root of 6156 is not even an integer, but when it was plugged into this equation, the result is an integer.
The magic of square root
Also I'm confused
It's a perfect square less than 10,000. Namely, 62 squared = 3844, and 3844 + 6156 = 10000. Now why it's 62 squared of all values, I couldn't say.
That shouldn't be weird. If you take any number of the form
x = a+b√n
you can rewrite it as
x² = a²+nb²+2ab√n
x = √(a²+nb²+2ab√n)
So, any integer can be written as sum of radicals of this weird form. For example,
4 = √(15+4√11)+√(15-4√11)
which I obtained just calculating
(2±√11)²
Such numbers and similar expressions became popular after Cardano's formula to solve cubics. Indeed, a third degree and complex version of the previous example is the solution of the classic
x³=15x+4
The Cardano formula for depressed/reduced cubics
x³=3px+2q
is
x = ³√R(+) + ³√R(-)
with
R(±) = q±√(q²-p³)
You can easily check that such x is ONE solution for the cubic. For the cubic
x³=15x+4
we have
p=5, q=2
so
R(±) = 2±√(2²-5³)
= 2±√-121
= 2±11i
Such numbers have as one of their cubic roots 2±i, since
(2±i)²
= 4-1±2*2i
= 3±4i
(2±i)³
= 6-4±(2*4+3)i
= 2±11i
making
x = ³√R(+) + ³√R(-)
= 2+i + 2-i
= 4
which satisfies
4³ = 64 = 15*4+4
By the way, this is also the actual origin of complex numbers.
I hope I didn't make any mistake.
Wow I have not seen your videos since I was in college and your channel was so small. Now you have hundreds of thousands of subscribers. This is amazing. Congratulations
Thanks. I am happy and grateful about it.
Maybe correct the grammar?
Solved the bonus problem by doing a substitution: n = -t, which then turned, considering that i^2 = -1, the final equation into a k^2 = 200 + sqrt(100^2 + t) where I have to find the smallest non-negative t. Using the same method I found 22 as the first smallest potential k (after 20) and, respectively, t = 10164 => n = -10164
I found the solution a different way!
It is possible to rewrite √(100 + √n) = √a + √b, where a and b are integers and a>b. If you square both sides you get that whatever n is, it can be written as 4a(100-a). Writing the sum with our new numbers we get that the sum of the two terms is just 2√a. For this to be an integer, since a
for the common types of problems involving sqrt[a+sqrt(b)]+sqrt[a-sqrt(b)], i usually find out first if i can get something in the form of sqrt[x+2 sqrt(y)], then figure out if i can factor it using (a+b)^2, and if i can i can write it as sqrt x_2 + sqrt y_2, and similarily, i can do the same thing with the other part of the equation such that the y_2's cancel out giving us 2 sqrt x_2
I found the same result and i want to explain my solution.
Let me call the expression f(n) and substitute n with x belonging to R so that f(x) is a real CONTINUOUS function.
Existence conditions implie that x must belong to the bounded interval [0, 10000].
f(0) = 20 whereas f(10000) = sqrt(200) that is similar to 14.14
Now, f(x) is monotonous decreasing ( HINT: study the positivity of the derivative function) and this means that 20 is his maximum value and 14.14 is his minimum value.
Thanks to Darboux's theorem ( or the intermediate value theorem) we can claim that f(x) attains every integer value from 20 to 15.
From 20 to 15 x will increase from 0 to 10000 since f(x) is monotonous decreasing.
So compute the counter image of K from 19 to 15 and if x is an integer we stop the algorithm, else the result is impossible.
To compute x knowing K we write the inverse function: x= 10000-((K^2-200)/2)^2.
f^(-1)(19) is not an integer, we pass to 18.
f^(-1)(18) is an integer and we find that x=6156.
This implies that x=6156 is the smallest positive integer which makes f(x) an integer.
The proof is now concluded.
I hopefully you appreciate my demonstration and if I made mistakes please don't hold it against me, I'm only an italian mathematician that wanted to share his solution with the global community ^.^
All the best, D.M.
👌
That’s beautiful
Are you a mathematician
@@gscreationss yep!!
Please give me some tips for a future mathematician (😜it's me) what I want to learn well
I have covered the exact same problem a couple of months ago in my channel. Glad you covered the same problem too!
Their explanation is basically almost identical. good thing
Quite a stander math Olympiad number theory question. Great video as always
Thanks.
I'd say this is far too easy for an Olympiad problem
Junior Olympiad
srsly??? olympiad???😂, go check out olympiad kiddo
2:38
my confidence,
30 seconds before the exam ends
Other way to solve its calling sqrt(100+sqrt(10))= b then b+sqrt(200-b²)=N then by using the quadratic formula we get (N+sqrt(400-N²))/2 then just search for the lowest N value with b= some natural number +sqrt(natural number) as we want b to be the least start with 19 and then move on to 18, and thats it with 18 you get 9+sqrt(19)
Ah! Nice one.
The fact that I can't understand it, it's the proof that I can't get into Harvard
I did the same method lol and i found that expression gives values that are less or equal to 20 and when the expression is equal to 18 then it gives the lowest N ...and N=
My answer before watching the video is 96^2
My approach is to let the expression equal to x and then make x^2 a perfect square, I got in the form 2(100 +√(100^2-n) so I concluded 50 + √(50^2 - n/4) must be a perfect square. Now since the √ is always +ve, I checked for the first perfect square after 50, which is 64 and that gave an answer for n, which was 4(50-14)(50+14), then I simplified a little and wrote in terms of power of primes (luckily it only contained powers of 2 and 3) to solve for √n, and I got it's value 96, which I verified that it makes both square roots perfect square so answer is integer without simplifying
But the value of n I got is pretty large, so I suspect I may have made a mistake somewhere and missed a smaller solution, but I'm almost sure my answer is correct (Unless you remove the positive integers and change it to non-negative :P)
Looks like he's taking a similar approach, I thought I was clever to find this easy way :P
Wow, turns out he didn't realise he had to make that expression a perfect square and mistakenly just solved for it to be an integer. So, he didn't get it right on his first try. Yay!
Never mind, I just realised my mistake halfway through the video, the answer is actually 4(50-31)(50+31) or ((18)^2)(19) which gives after simplifying gives (9+√19)+(9-√19) = 18
Earlier, I actually mistakenly solved for the largest value of n, because I solved for smallest value of x, in fact I have to solve for biggest value of x to get smallest n
well done
Regarding the bonus part, the biggest negative integer is the negative integer closest to zero. If there are no other possible solutions between the claimed -10164 and -1, inclusive, then -10164 would be the biggest (assuming it works).
Yayy did I get it right with n=-10164 then? Not used to getting BPRP problems right lol
I think I've actually solved this by little logic and trial error method.
Since [ 100 - sqrt (n) ] must be positive perfect square,
sqrt (n) < 100
Trial error values of 100 +- sqrt (n) fails for squares of 11, 12 and 13, more specifically 100 - 21, 100 - 44 and 100 - 69 (as these differences are not perfect squares).
But it succeeds for
sqrt (n) = 96, as both
100 + 96 and 100 - 96 are perfect squares.
Hence n = 96^2 = 9216
EDIT 1 : I tried this orally without using any calculator as I thankfully remember perfect squares till 111.
EDIT 2 : I checked the solution and was sad to know my answer is not correct here. No issues. It would have been correct for a rational square root result. Anyway. I enjoyed trying this one.
This is the same method and result I used/got
I also got your answer initially and assumed each of the roots are positive integers
Yeah I used the same method
I did the same,but why this answer is wrong?
@@mojtabasaleh8842
It's wrong because square root of 100 +- sqrt (n) can be irrational, which on addition, result in an integer. That's something which we didn't think of.
I was the problem czar for this tournament, during my freshman fall! I hope you enjoyed the problems!
I thought bprp went mad when setting 10,000-n to 9801 😂
Wow, I tried solving this and ended up making the exact same mistake as you did! It must have stumped many more students during the tournament
The problem with the tournaments is that their purpose is to trip people up on mental calculations (for the most part) rather than any rigorous type of math involved. Seeing or scratch writing 10000, might as well be shorthand to 10k or 1kk, or any other moniker.
lets consider n=4k than (root/100-,+//n)=(/100-,+2//k) we know that if a+b=100 and a*b=k we can write that as /a -,+/b
a is bigger than b so we can get positive number(for the negative one)
if we do all these we get (/a+/b)+ (/a-/b)=2/a so a must be x^2 and a=81 and b=19 perfectly fits into this so 2/81 =18
n=4*81*19=6156
''/'' = square root
His cursor is bigger than my self-esteem
ua-cam.com/video/HMHLL4YAXdc/v-deo.html
Other solution:
If we have sqrt{a+\sqrt{b}}=sqrt{x}+sqrt{y} then sqrt{a-sqrt{b}}=sqrt{x}-sqrt{y}
With x=frac{a+\sqrt{a^2-b}}{2} and y=frac{a-sqrt{a^2-b}}{2}
[the proof is easy, try it]
So, in this problem we have that the sum is
sqrt{100+sqrt{n}}+sqrt{100-\sqrt{n}}=2(sqrt{x})=sqrt{4x} in Z
With x=frac{100+\sqrt{10000-n}}{2}
and 4x=200+2\sqrt{10000-n}
and how do we know that $4x=k^2\inZ$, we continue as bprp.
Can You Solve This Harvard MIT Math Tournament Problem?
*no*
Happy teachers' day sir😇
Thanks. I don’t think we actually have that here but I appreciate it! 😃
@@blackpenredpen I think you have teachers day at 5 October we are from India 🇮🇳
Really nice initiative
Thanks.
you know things getting complicated when he pull out the blue pen
I don't pause the video and try it when you say, because I've usually already had a go from the thumbnail and now I want to know the answer.
ua-cam.com/video/HMHLL4YAXdc/v-deo.html
I have a general question: is it ok to say naturals instead of positive integers?
Some people count 0 as a natural number. So it’s better to just say pos integers.
@@blackpenredpen typically computer scientists. and i have no idea why they ever fumbled into that obnoxious definition. of course the naturals should start with 1. 0 is hardly "natural". didn't it take life something like 4.5 billion years to even conceptualize it? whereas the positive integers took merely 4.49999 billion.
@@sharpnova2 When I was a computer science major, I never heard anyone there talk about 0 as a natural number (I don't think they even used the term). From what I've heard, it's more of a geographical thing: 0 is often a natural number in Europe, but not in the U.S.
There are only two integer solutions to this problem. Obviously one for n=6156 but the other is n=9216.
This was how I approached it by looking at cases where n was a square number. Turns out that isnt the best solution, but I kinda just assumed it would be the case. n=9,216 is nice and neat in that things are integers at every step.
@@Harlequin_3141 Yeah when I was in the gym I wasn’t able to square it all in my head and work it out that way so at first I was just looking for numbers that fit and found that. But later when I got home I managed to reduce the equation and get an inequality that showed the only possible solutions were 6156 and 9216. Also n=9216 makes the equation it’s smallest possible integer solutions so I prefer it.
Elegantly explained Boss!!
Awww.... 🦒❤😯😍😍
It's really adorable 😅😅
bruh
bruh
yes i aggree
🤩🤩🦒
@@jimmykitty giraffe love grass
A quick python script to verify the answer :
from math import sqrt
j = 0
while True:
j += 1
f = sqrt(j)
s = sqrt(100+f)+sqrt(100-f)
if s == int(s):
break
print(j)
Solution to the question at the end:
With the same approach as in the video we get k² = 200 + 2sqrt(100² - n), and for the sake of (my own) better understanding, let -n = m; in which case we're looking for the smallest positive integer m.
so k² = 200 + 2sqrt(100² + m), if m was equal to zero, k² would be 400, so we're looking for the perfect squares bigger than 400 but as small as possible
21² = 441 doesn't do the job, as k² is also even. take 22² = 484
then 484 = 200 + 2sqrt(100² + m)
284 = 2sqrt(100² + m)
142 = sqrt(100² + m)
142² = 100² + m
m = 142² - 100² = (142-100)(142+100) = 42*242 = 484*21 = 9680 + 484 = 10164; therefore n = -10164
plug this into any programm able to work with complex numbers, you get an integer in the original expression.
The answer to your challenge question:-
(-6156)
because let n = -m, where m is positive int.
when we find for m, for smallest, it is similar to solving for -n for biggest.
If I am right, pls give a heart!!
ua-cam.com/video/HMHLL4YAXdc/v-deo.html
There is a related approach as well - consider 100 + \sqrt(n) = ( a + \sqrt(b) )^2 for some a,b.
Then, 100 - \sqrt(n) = ( a - \sqrt(b) )^2. It follows that 100 = a^2 + b while n = 4 a^2 b = 4 a^2 (100 - a^2).
However, there are two conditions for this to hold:
1) a > \sqrt(b) else the \sqrt(100 - \sqrt(n)) = \sqrt(b) - a and the result is no longer an integer. This also means a^2 > b = 100 - a^2 => a^2 > 50.
2) b > 0 (This is to ensue n>0, and for the bonus can we changed to b < 0). This gives a^2 < 100.
Given n is a decreasing function in 50 < a^2 < 100, choose a^2 = 81 and correspondingly n = 4 x 81 x 19 = 6516.
The solution for bonus is similar with b < 0. which corresponds to a^2 > 100. Choose a = 11 to get n = - 4 x 121 x 21 = 10164. (Assuming the square root with positive real part is considered)
In fact, the set of all admissible n corresponds to all a^2 >= 50 or a >= 8.
3:40, I feel like I kinda understand why but I can't really wrap my head around why that would make k even
First examine all singular digits. Square every one.
0^2 = 0
1^2 = 1
2^2 = 4
3^2 = 9
4^2 = 16
5^2 = 25
6^2 = 36
7^2 = 49
8^2 = 64
9^2 = 81
We define an even number to be a number divisible by two with no remainders left and we define odd to be a number when divided by two has a remainder.
Notice that every odd number produces an odd number when squared. This is why if we work backwards that we can conclude that ‘k’ is even.
I would test bigger numbers to prove this to you further but it satisfies for every number because the only way you will get an even number is if your last digit is defined as even.
i.e:
100^2 = 10000 and has no impact in the ones place. 102^2 = 10404. 102 has impact in the ones place because there exists and number in the ones place for 102.
successfully solved this! and i furthermore challenged myself to do it all in my mind! great video as always
What if we found the maxima of the given expression at the first place and then put that in place of k to get n(of course less than 20).
Both expressions k(n) and k²(n) are strictly decreasing and admit no local maximum (the maximum is n = 0). Which is more or less what BPRP did when he said k=20 and which is wrong. Then your method would find the closest n that gives k an integer, which would be again equivalent to what he said.
@@PackSciences yes i think you are right...the graph does not feature a local minima and maxima is at 20..so by my method it is impossible to find an integer. I guess hit and trial is the only way to solve it
ua-cam.com/video/HMHLL4YAXdc/v-deo.html
omg your 1 hand marker transitions are so cold
0:28 "that expression gave us an integer" ("= - 2" pops up)
The minus sign is wrong (just nitpicking).
My solution was different, but incorrect, because of an incorrect assumption I made. Moreover, I did this algorithmically rather than mathematically.
I made a couple of assumptions based on the fact that our result, k is an integer and n is a positive integer:
1. n squared must be less than 100
2. each term of k individually must be an integer (an incorrect assumption based on the given facts)
Then, based on this, I introduced a new term, p = sqrt(n)
Now my objective was to find an integer, p, at an equal distance from 100 in both directions, landing on a perfect square. This gave me p = 96, or n = 9216, which is a solution, but not the answer to the question, as it is not the smallest value of n. This as all based on the incorrect assumption that p would be an integer, stemming from the assumption that each term of k would be an integer. This method gives me k = 16, however.
I made the same mistake and got the same answer as you. It seems like the usual technique so it’s the first time I encountered such a question!
rohitchaoji, i did the same mistake
We did same mistake.I found it 96 too
4:49 let k=2p, square both sides and you will get n=f(p)
It's weird how all the videos in english say that zero is not positive nor negative, while in french we learned it was actually both.
In a question like this it would say n is strictly positive, if it only said positive then zero would be an acceptable answer.
The idea of the solution is quite easy but not making a numerical mistake along the way was impossible for me. I came up with a slightly different tactic from yours but I had to calculate the thing like three times until I eventually uncovered all the mistakes (things like 2x199=399 or whatever). I suppose simple 4-operation calculators aren`t allowed on the exam, right?
*Related Problem*
Prove that sum of [ C(4k-2, 2k) / (2k-1) x (n/20^4)^k ] from k = 1 to k = Infinity equals 1/20 where C(n,r) is binomial co-efficient function.
You will find that n = 6156
Yeah I'm just gonna admit Alexa just halfass solved it for me.
There is quiet a few natural numbers that makes the expression a positive integer.
6156 give integer because
(9±sqrt(19))^2 = 100±18sqrt(19)=100±sqrt(6156)
I missed that it's about finding a positive integer for n, so I stopped once I found the trivial n=0 case. But the method I used easily spat out 6126 (which interestingly enough when subtracted from 10,000 gives exactly 62 squared).
More generally, given a and any k such that a/2 ≤ k² ≤ a, then n = 4k²(a-k²) makes √(a+√n) + √(a-√n) an integer, namely, 2k. In the case above, a = 100; so
100/2 ≤ k² ≤ 100 implies
k = 8, 9 or 10, making n = 9216, 6156 or 0 and √(a+√n) + √(a-√n) = 16, 18 or 20. n = 0 is explicitly excluded, making n = 6156 the
minimum. (9216 = 96², in which case √n is an integer, 96, whereas 6156 = 4·9²·19 makes √n = 18√19.)
I hope someone makes a general case for the cube root video. I’m quite interested that that thing just simplifies to 2, so I’m hoping to discover more of such.
can you be more specific? what cubic root problem?
@@MathSolvingChannel so here are two specific examples of the problem:
ua-cam.com/video/_6Pl_MNphf4/v-deo.html
ua-cam.com/video/5pa1AryylpM/v-deo.html
These are specific examples. How do I generalize/generate other examples with other numbers inside the nested roots? Any leads would be highly appreciated. Thank you!
@@kimveranga The general way to solve this type of problems is to use substitution. Let a=1st cubic root term, b=2nd cubic root term. so you will get a^3+b^3 = some integer and a*b=another integer, usually. then you use the complete cube formula to convert them into a cubic equation respect to a+b, then you solve this cubic equation. I also did two videos on how to apply this method to solving this type of problems. and here are the links:
ua-cam.com/video/0P6L581BxuI/v-deo.html
ua-cam.com/video/UbLWWNgp-sE/v-deo.html
@@MathSolvingChannel I’m quite familiar on how to solve those things. My question is how do you even come up with the square roots inside cube roots problem like this. How do you generate these kinds of problems? Thank you!
@@kimveranga Let me confirm if I understand your question correctly, do you mean: why they put this form of square root expression inside this cubic root? why not they put others such as 11+sqrt(97), or in general speaking, for what kind of square root expression this problem is solvable?
I think before concluding "k is even" you also have to prove that you can't choose an n so that the square root comes out to a half number (like 80.5) which when multiplied by two comes out as an odd integer
Good catch, but it turns out this is always impossible. The square root of a positive integer is either an integer or irrational (look up the proof of the irrationality of 2, it's a classic!)
@@alexmcdonough4973 Thanks! That's good to know
So fun though very intuitive 💪
ua-cam.com/video/HMHLL4YAXdc/v-deo.html
I'm only in 9th grade, and I do these equations so I can get ahead in math, I want to go to MIT, thankfully you have helped me expand my math knowledge. These problems are difficult but honestly I understand them a little, I was off to a good start but then got lost about halfway through trying it. Watching you do it made it make more sense though.
ua-cam.com/video/HMHLL4YAXdc/v-deo.html
I decided to take this a step further and created a small Python script to find all possible values. The value n must be greater than 0, but less than or equal to 10,000. There are only two values, the one he found, and 9216.
This is because k^2 is at least 200, so k is at least 15. Since k is even and less than 20, that only leaves solutions for k=16 and k=18.
ua-cam.com/video/HMHLL4YAXdc/v-deo.html
What's wrong in: let k=√(100-√n)+√(100+√n)
Squaring both sides(and skiping a step or two)
K²=100-√n+2√(100²-n)+100+√n
K²=2(100+√(10,000-n))
Rearranging to get:
(K²/2)-100=√(10,000-n)
Squaring both sides to get
(K⁴/4)-100k²+10,000=10,000-n
Rearranging to get
100k²-(k⁴/4)=n
You can try finding integer solutions from here but it doesn't work out. I don't know where the mistake is.
Ok I got the right answer after seeing your working out.(also pretty surprised seeing that everyone's first step is the same. I thought I did something horribly wrong in the beginning but turns out it's the same as everyone.)
I tried to find any value of n that will yield the integer in the expression and I found 396.
sqrt(100+sqrt(n)), putting n=396
sqrt(100+sqrt(396)) = sqrt(100+2sqrt(99)) = sqrt(1+2.1.sqrt(99)+(sqrt(99))^2) = sqrt((1+sqrt(99))^2) = 1+sqrt(99)
In similar fashion,
sqrt(100-sqrt(396)) becomes 1-sqrt(99)
now ,
the given expression becomes
1+sqrt(99)+1-sqrt(99)
=2 , which is an integer.
I do know that second term can be sqrt(99)-1, in that case the solution won't work but I don't see any reason why cant we put 1-sqrt(99). Is it because the term becomes negative?
I'm really confused.
Why isnt 361 a vakid answer? Root of 361 can be +19 or -19. If you choose appropriate roots, you can get 18 as the final answer. Can somwone explain it to me please??
n=361 looks good to me. I was working on the same trick but for some reason went upwards to 100+(21) and 100-(-21) making n=441
The last time I watched was the 100 integrals video. Now he is a wizard 😳. Math God
Hello blackpenredpen, I would like to ask you if you accept user submissions with regards to question and papers and if you do, how we can contact you. I love the questions you put on your channel and allows me to practice my craft, thank you so much.
Hi Justin. I am quite busy this semester. You can send the problem to my email blackpenredpen@gmail.com but I am sorry that cannot promise if I can make a video on it soon.
@@blackpenredpen no problem
@@blackpenredpen Stay safe and keep on bringing the videos! Old timers like me who haven't studied rigorous theoretical math or even basic algebra in the past 7 years still like to keep up with things. I ended up going the biochemistry route, but I still have a longing for that math itch from my 2nd degree in pure math.
It’s great to watch a 5 min video and confirm I had no business going to MIT. 😂
What I'd probably do is at first, deduce that n must be a square number and 100-sqrt(n) AND 100+sqrt(n) must be a square number.
then plug in sqrt(n)=1,2,3... Since sqrt(n) cannot be equal to 10, it will be one of the values from 1-9
Idk about y'all but I don't know how to do this math but I'm still watching 😭
Me to
It’s pretty easy if you notice that it is simply 9 + sqt(19) + 9 - sqt(19), then it goes 81*19*4=6156
where did you get the numbers from?
ua-cam.com/video/HMHLL4YAXdc/v-deo.html
My lucky day , i just opened the phone to search this type of sqrt exercise
2:39 🤣🤣🤣 I like this part..
😆
Me too.😂
Thanks for this problem!
I worked the dual on my channel, i.e. found largest natural number, n, such that sqrt(100+sqrt(n)) + sqrt(100-sqrt(n)) is positive integer.
Got n=9216 which is in the domain=[0,10000]
I was able to solve this question on my own ... and yea it made my day :P
100+n/100-n->ⁿ defined as an postive integer ->10000n->200+2×->->18
This goat is the smartest
I've watched two math videos and now youtube thinks i love math
😆
I did this and got 9,216, which is wrong, but I’m still happy because it is the smallest value that also remains an integer when taken the square root. I did consider that maybe the answer would not necessarily abide by this rule but I didn’t really know how to solve the problem if that was the case.
Please try with k=16.Then n is a square number=36×256.
What is the value of n that gives the smallest integer solution to sqrt(100 + sqrt(n)) - sqrt(100 - sqrt(n))?
smallest integer: 0
smallest positive integer: 396
other positive solutions:
1536
3276
5376
7500
9216
9996
Yay captions!
Sorry, can someone explain to me how does sqrt(100^2-n) become sqrt(99^2)?
2:19 you are talking about
he just tried a little troll or a showed a little little mistake
but the question is why he said that
so let me explain:-
integer + integer is always integer but the square root of various integers are irrational (which is not integer)
so if the square root part comes out to be integer we wont care about rest it will be integer which will be answer but the he said that there is square on answer which means we are not getting integer hence not the answer
most importantly he knew it was closest perfect square (n should be integer) he set them both equal and square rooted both sides and equated(rest explaination is mentioned in this reply by me)
Because we're looking for the smallest possible n>0, so we're looking for the largest perfect square less than 100^2 (to get an integer out of the radical), which would be 99^2.
Ohh ok, I understand now. Thanks guys 👍
can you put all the other questions' answers please I tried to solve them but unsure of it's correct
Sir how you write maths symbol in this video thumbnail, plz guide me
I get it. You are channeling that Pokemon ball to give you the answers. Good thinking.
I'm a bit confused, granted I am well out of my range of knowledge. Doesn't √(100+√n)+√(100-√n) rearrange to 10+ √√n + 10 - √√n and then the √√n cancel out to just leave twenty?
Is there some imaginary reason this doesn't work maybe?
Biggest negative integer n = -10164 ?
yes, correct, same result here :)
How can sq root contain a negative integer?
@@teslaaf5830 it becomes complex numbers :)
@@MathSolvingChannel so u solved the problem in complex numbers? and how come a negative integer came as a result.. i tried but didnt get the ans
@@MathSolvingChannel can you post that solution in ur youtube channel
Sorry for my dumbness, but i didnt undestand why sqrt(10000 - n) can be set to sqrt(99^2)
10000 is 100^2, and n cannot be zero but we want to maximize sqrt(10000 - n). That way, the maximum possible answer is sqrt((100-1=99)^2).
So were you allowed a calculator for this problem, and if not then how would you know thats the correct solution when you plug 6156 into the initial equation?
Well it's not zero. I guess n is a square since the root of a root is irrational. For root of n, I use x. The value of x must be between 1 and 100 or else we obtain complex solutions from 100-x being negative under the root. The larger root will be 101 to 200. There are four roots in this range: 121, 144, 179, 196. Of these we search from smallest to find the first which is also a square at 200-x. It can't be 121, because 200-121 = 79 and that isn't a square of an integer. Likewise, we eliminate all but 196, because 200-196 = 4 .... an integer square as 2². So we now see that x is 96, because √(100+96) and √(100-96) are integers. Since x is the root of n, I think that n = 96²
how is it clear that 20 (n=0) is the biggest value of k? i don't see how k is strictly decreasing as n increases. i plotted the graph on desmos and i can see that it is decreasing, but how could one tell algebraically/analytically?
EDIT: ah, i can see that k^2 is strictly decreasing as n increases. is that enough to determine that k is also strictly decreasing though?
2nd Edit: yes, because k is always positive, so a decreasing k^2 implies a decreasing k!
because you cant have n>100^2, otherwise Sqrt(100-Sqrt(n)) is a complex number and this problem restricts us to the positive (real) integers.
@@colereynolds2080 I figured it out in my edits lol thanks tho
Can you explain the Beimar formula for calculating prime numbers?
can you explain how those two irrationals make a rational. because it is not like adding φ and 1-φ. they are two different irrationals. what part of those numbers is the one canceling out? is that a stupid question?
not at all. in fact, i believe this is the core to the debate whether 6156 is the correct answer or not.
@@ipcheng8022 is there anywhere you recomend to look up for this one. I am an engineer so I am not familiar with where you go to discuss math. I haven't done math for the degree for 3 years now so yeah that didn't help either
@@hellNo116 i am an engineer too, so i tried it on Python. when the two terms in the original equation are seperated, both of them have a lot of decimal places.
First term: 13.358898943540673
Second: 4.641101056459326
Here is the thing that I cant figure out why: the decimal places were aligned too perfectly, so that all of the sum at the decimal places are nine. I believe this is not a mathematical issue but a technical issur, where computers automatically round up the 17.99999999999 to 18. On the other hand, I cant figure out why the result is so close to 18. There is no terms to should be cancelled due to the additional square roots applied to the two terms. It could just be coincidence, or this is some math theories that we dont know about. This is why I strongly support that 9216 should be the correct answer because it can be proven for sure that results of all terms are integers
@@ipcheng8022 rounding errors are thing. they don't always round that way. it has to do with the printed versus the saved value of the number. also it is really depending on how the computer is programmed to deal with them.
Also that is an approximation. My question isn't if the approximation add up, but how the irrational parts cancel out
@@hellNo116 i cant find any other similar cases or related theories that can explain this phenomenon, which is why i can only explain it from a technical view.
I tought that square roots were separate part of the formula.
WolframAlpha gets it wrong sometimes…I got 361…nvm I got it wrong by using - in the first radical instead of +…..
n=2944 is what my brain told me, because i didnt expect 19 not working
As always nice video, My favourite math channels
1) blackpenredpen
2)mathematics mi
3)flammable maths
4)dr.peyam
5)Michael pen
Don't you assume that the function is descending at all times? Otherwise the largest output wouldn't necessarily mean you get the lowest input, and you also can't know that the range of the output value is less than 20.
Shouldn't you prove this?
@blackpenredpen how can I send my q to you to solve..?
On the bonus question, does biggest mean greatest or |greatest| ?
Sir please video on roots of 4 degree polynomial and sir I like your teaching thanks for open this channel
The ball in his hand turned to Pokémon and he has grown a beard all in a matter of time.
Probably couldn't do it when I was in highschool tbh but I can now
Much much n much interesting❤❤❤❤
+1 again 😆🙂
we can just take n = 0
it will be inferior to 6156
and k will equal to 20
what are you saying about this please answer me ?
Hehe, didnt expect the sum of two irrational roots to become integer.
Me neither. Any explanation to that, anyone?
ok never mind. The irrational parts just cancel out.
Plot twist
There is arceus in that pokeball
ua-cam.com/video/HMHLL4YAXdc/v-deo.html
My try: let's say that the given thing equals k. k^2=200+2sqrt(10000-n). Since k is an integer, k^2 should be a square number. n is a positive integer, and k^2 must be smaller than 400. The largest square number smaller than 400 is 19^2=361, however this must be even and it is 18^2=324. This means, sqrt(10000-n)=62, 10000-n=3844, therefore n=6156. Uh, I didn't copy the video but I solved it in an identical way, I guess.
7:48 -10164, I think