23% Beyond the Riemann Hypothesis - Numberphile

Поделиться
HTML-код
  • Опубликовано: 13 янв 2025

Комментарии • 540

  • @kostasch5686
    @kostasch5686 Год назад +436

    6:15 The reason π is used for pi, is because the word circumference in greek is περίμετρος which has the first letter π. The reason π(x) is used is because the prime numbers in greek are called πρώτοι αριθμοί which litterally translates to first numbers. Yet again π is the first letter. Coincidental that the letter is the same.

    • @theemperor-wh40k18
      @theemperor-wh40k18 Год назад +97

      For those who don't read greek: the first one is "perimetros" and the second one "protoi aritmoi".

    • @ferretyluv
      @ferretyluv Год назад +8

      I thought it’s a capital pi that’s used for primes.

    • @johndickinson82
      @johndickinson82 Год назад +11

      That’s for series multiplication like how sigma is to sum

    • @soupisfornoobs4081
      @soupisfornoobs4081 Год назад +12

      ​@@theemperor-wh40k18 those who don't speak greek wouldn't know that "oi" is pronounced "i"

    • @demoman1596sh
      @demoman1596sh Год назад +21

      @@soupisfornoobs4081Not sure that’s really all that important, is it? would in any event have been pronounced /oi/ at some earlier stage of Ancient Greek, even though it has been pronounced /i/ for a long time now.

  • @adandap
    @adandap Год назад +84

    This reminded me of the Anchorman quote "60% of the time it works every time"

  • @stevensutton4677
    @stevensutton4677 Год назад +920

    So this is a sort of Parker Reimann Hypothesis?

    • @wierdalien1
      @wierdalien1 Год назад +57

      Parker meme

    • @codycast
      @codycast Год назад +7

      Huh?

    • @imnimbusy2885
      @imnimbusy2885 Год назад +67

      The Riemann-Parker Postulate?

    • @tdurran
      @tdurran Год назад +4

      Hilarious😂

    • @wbfaulk
      @wbfaulk Год назад +70

      In the same way that "Reimann" is a Parker spelling of "Riemann".

  • @TessaLucy
    @TessaLucy Год назад +36

    Brady being a questioning viewer is such a good device for information

  • @standard_limbo
    @standard_limbo Год назад +224

    So many strange things tie back to the Riemann hypothesis. It's fascinating. I'm glad he took the time to explain it so clearly.

  • @nickjohnson410
    @nickjohnson410 Год назад +143

    I think I gained a 23% increase in understanding of the Reimann Hypothesis.
    Thank You

  • @hammadusmani7950
    @hammadusmani7950 Год назад +16

    It's inspiring to see, hear, and experience the growth of this channel. I wouldn't be surprised if the people who eventually do solve the Reimann Hypothesis are huge fans! It's not just that, it's the quality has been so consistent. Thank you for this.

  • @rediculousman
    @rediculousman Год назад +264

    I can tell that this guy is IN THE ZONE at the moment. I love when I'm like that and my topic of focus is so clear. It's just the looming wipeout of depression that comes later that wrecks me.

    • @housellama
      @housellama Год назад +19

      Oof, I feel that to my bones. The grad school depression and burnout is REAL

    • @la6beats
      @la6beats Год назад +1

      Same

    • @thenoobalmighty8790
      @thenoobalmighty8790 Год назад +4

      Need to socialise 😅

    • @jamesnorrah1316
      @jamesnorrah1316 Год назад

      I get what you mean 😢

    • @frtzkng
      @frtzkng 13 дней назад

      ADHD hyperfocus be like

  • @ilovezsig
    @ilovezsig Год назад +268

    Love the enthusiasm of this guy

    • @muskyoxes
      @muskyoxes Год назад +12

      Every sentence is like "i can expand on this for an hour", and it makes me want any one of those hours

    • @NLGeebee
      @NLGeebee Год назад +17

      True, but he speaks so fast in short bursts, he lost me after 3 minutes…

    • @BernardoHenriques4
      @BernardoHenriques4 Год назад

      @@NLGeebeeyeah… hard to understand for a non-native speaker like myself…

    • @victorcossio
      @victorcossio Год назад

      @@NLGeebee Actually I watched this at 0.75x speed

    • @iboremytherapist
      @iboremytherapist Год назад +7

      Asperger’s syndrome or adderall

  • @tzombikos9718
    @tzombikos9718 Год назад +125

    i really enjoy the longer videos with jared, his explanations are great.

  • @KB_13247
    @KB_13247 Год назад +9

    it never stops amazing me how the riemann hypothesis links to so many different things and all things related to prime numbers, which means it's related to basically all of mathematics.

  • @alexstixx
    @alexstixx Год назад +31

    I spent years figuring out what this video explained in a short amount of time. Really great video.

  • @syedsamaan5053
    @syedsamaan5053 Год назад +11

    This guy sounds so excited while explaining mathematics like imagine as a kid you discover something and are eager to show that to your parents and friends and siblings....... This excitement in his voice is kinda interesting at the same time being kinda contagious too😊😊

  • @tyleringram7883
    @tyleringram7883 Год назад +254

    I loved the way he explained it and a different aspect of it. Lots of people will tell you about a graph and a critical line and a complex plain, but that visual representation is way to complex.

    • @nazgullinux6601
      @nazgullinux6601 Год назад +8

      Complex Plane* But yes.

    • @velteau
      @velteau Год назад +22

      Joke's on you mate, this guy's explanation was too complex for me as well.

    • @UnknownCleric2420
      @UnknownCleric2420 Год назад +7

      Complex, heehee

    • @Lighthouse_out_of_order
      @Lighthouse_out_of_order Год назад +2

      @@nazgullinux6601 You're all imagining stuff

    • @Alexagrigorieff
      @Alexagrigorieff Год назад +1

      @@nazgullinux6601 These spellings are related, anyway, just like "sheer" and "shire", which also mean "plain" and "plane".

  • @Lotrfan1991
    @Lotrfan1991 Год назад +140

    Super clear explanation. This guy is awesome

  • @DrTacoPHD665
    @DrTacoPHD665 Год назад +15

    Very happy to see Jared back again

  • @bens4446
    @bens4446 11 месяцев назад +19

    That last 0.000000001% is a lot more important in math than it is in household cleaners.

    • @SunShine-xc6dh
      @SunShine-xc6dh 10 месяцев назад +1

      Not really. They use i without it having any definable value.

    • @michaelsmith4904
      @michaelsmith4904 9 месяцев назад

      but 0.00000000001% of te remaining germs will reproduce exponentially and grow back to the original quantity in log time!

    • @MatthijsvanDuin
      @MatthijsvanDuin Месяц назад +1

      @@SunShine-xc6dh i has a perfectly rigorous definition. If you don't consider it to have a value, that's just a limitation of your definition of "value". If you don't _have_ a definition, then what you're saying boils down to "this is something I'm not used to, therefore I reject it". Keep in mind that many sorts of numbers we take for granted now were at some point in history not considered to exist, such as negative numbers, zero, and irrational numbers.

    • @frtzkng
      @frtzkng 13 дней назад

      @@SunShine-xc6dh i^2 = -1

    • @jurajvariny6034
      @jurajvariny6034 2 дня назад

      That depends on your personal value system 😆

  • @TheDrakmannen
    @TheDrakmannen Год назад +6

    I love watching these videos and just letting my ’tism run wild

  • @rosiefay7283
    @rosiefay7283 Год назад +19

    1:27 My favourite way to do maths! Think of a phenomenon, write a program to seek examples and write them, run it, read 'em, look for patterns, ...

  • @ΘάνατοςΧορτοφάγος

    It is so nice to have the words hypothesis and theorem used correctly on youtube...

    • @AySz88
      @AySz88 Год назад +6

      ​​@jash21222Eh, probably misuse of "theory" is being lumped in with it.

    • @columbus8myhw
      @columbus8myhw Год назад +3

      It's unfortunate that the words "theorem" and "theory" are so similar; the difference in meaning is quite large.

    • @btf_flotsam478
      @btf_flotsam478 Год назад +3

      Mathematics has a different use of it than the (other) sciences.

    • @btf_flotsam478
      @btf_flotsam478 Год назад +1

      Mathematics has a different use of it than the (other) sciences.

    • @WAMTAT
      @WAMTAT Год назад

      But that's just a hypothesis, a game hypothesis

  • @debayanbanerjee
    @debayanbanerjee Год назад +2

    11:38 Brady touches upon a very important (but seemingly innocuous) question about pure science and applied science. Engineering is also applied science in action. Any engineer will tell you the amount of assumptions, scenarios etc. that we 'limit' a case to just so that we get closer to the answer for that specific instance. Just a short example, we know that the earth is spherical, but all spheres are locally plane (flat) and that is why we make do with flat rulers to measure the length of a table for instance. Because for most ordinary measurements it is enough to assume that we are dealing with flat surfaces. So even though 99.9% leaves out a lot of numbers, but for a specific case 99.9% is as close to a perfect result one might get.

  • @ancientswordrage
    @ancientswordrage Год назад +69

    More of this guy please, he's great

  • @JosephGallagher
    @JosephGallagher Год назад +12

    I feel like I talk just like him a lot of the time. The little pauses while still decoding the concept in mine mind.

  • @graduator14
    @graduator14 Год назад +43

    Always nice to see some progress!

  • @IsoYear
    @IsoYear Год назад +1

    when he says "a question comes up" and his face lights up i know im about to have my mind blown. i never got to see this side of math in chemistry

  • @mobius32
    @mobius32 Год назад +25

    Jared is so enthusiastic and excited about this subject it practically radiates through the screen. Find someone who talks about you the same way Jared Lichtman talks about the Bombieri-Vinogradov theorem!

  • @3Max
    @3Max Год назад +22

    I felt like he kept dodging your Q about "what does it mean to even out" -- did you ever get an answer to it? or anyone here know? :) I guess if you relax "how even it needs to be", you could probably "get there faster" in a sort of trivial/definitional manner? All that to say, it was a great Q, but you should have kept pushing for an answer! :P

    • @japanada11
      @japanada11 Год назад +30

      "Leveling out" is meant in the limit as x goes to infinity - do the number of primes with each allowed digit converge to the expected mean value?
      The point is that you allow b to grow as well, not just x. To illustrate an extreme example, suppose you consider all the primes up to b/2 in base b (eg all the primes up to 500 in base 1000). No matter what b you pick, the result is very far from being uniform across all the allowed digits: the larger half of your digits are never reached at all! Taking b to infinity doesn't help: no matter what b you pick, half of your digits will get nothing.
      A more interesting example is considering all the primes up to b in base b. There still just aren't enough primes to go around; even if you take b to infinity, you'll always find many "allowed" digits with zero primes having that digit (never getting anywhere close to the expected average)
      But if you consider way more primes, say all primes up to 10^b in base b, and you take b to infinity, then the proportion of primes with each last digit DOES converge to the expected mean. The digits themselves are changing because you're changing the base b; I'm saying that as you increase b, no matter what base-b digit you pick, the ratio between the expected average number of primes with that digit and the actual number gets closer to 1 the larger b gets.
      The Riemann hypothesis would say that you don't need nearly that many primes in order to see the convergence: just taking all the primes up to b^2 in base b would be enough. The 2023 result says the same thing happens if you only consider all primes up to b^1.63 - though this only holds for "most" b. That is, as you increase b, it's possible that for some very sparse sequence of bases, the distribution jumps away from being uniform; but as long as you take a sequence of bases b that avoids these rare troublemakers, then the proportion of primes up to b^1.63 with each allowed last digit converges to the expected value as b goes to infinity.

    • @sachamm
      @sachamm Год назад +3

      @@japanada11 I think you explained this part of it better than the video.

    • @3Max
      @3Max Год назад +1

      @@japanada11 that is definitely helpful, thanks for the details! Though I might still need to just read the papers to fully understand it. Based on their dialogue I was imagining the definition to be more focused on identifying some type of bound on "the variance across the digit-buckets" when you look at all (or "almost all") bases up to x^0.5 (or x^0.61, etc). Fun stuff!

    • @brotatobrosaurus5411
      @brotatobrosaurus5411 Год назад

      I agree and I'd also like to know more about the local probabilities around each prime. it's nice to know when the probability evens out when considering the space of all primes. but are there some primes where certain last digits "disappear" for a while?

  • @FunWithBits
    @FunWithBits Год назад +5

    Finally, a video with Riemann Hypothesis in the title that I can understand.

  • @MorganHayes
    @MorganHayes Год назад +6

    I love how this guy always brings colorful metaphors

  • @iluvtacos1231
    @iluvtacos1231 Год назад +118

    This hypothesis is so far over my head that it's not even funny.
    But I'm glad people like him are able to grasp it.

    • @redryder3721
      @redryder3721 Год назад +29

      I forgot what tab I was on and I thought this was a video about a talking parrot. I got very confused by your comment.

    • @youtubepooppismo5284
      @youtubepooppismo5284 Год назад +2

      It's not hard really. it just means that whenever ζ(z)=0, then either z is a negative integer or z is on the critical line (z = it+1/2)

    • @aug3842
      @aug3842 Год назад

      @@youtubepooppismo5284i think they might be talking more about the deep connections it implies

    • @anticorncob6
      @anticorncob6 Год назад +7

      ​@@youtubepooppismo5284
      I'm guessing the OP doesn't understand what that function really is.
      It has a simple formula for Re(z) > 1, but to understand how it's defined elsewhere you need to understand analytic continuation. From my understanding there is only one function on the complex plane that is analytic everywhere except at 1 and matches the formula where Re(z) > 1.
      And if the OP is reading this, "analytic" on the complex numbers means it's differentiable everywhere it's defined.

    • @youtubepooppismo5284
      @youtubepooppismo5284 Год назад +1

      @@anticorncob6 Analytic doesn't mean it's differentiable everywhere, that would be holomorphic. Analytic is when its Taylor series locally converges. Althought analytic and holomorphic are equivalent on the complex plane so you can use them interchangably, but they have different meanings. You are correct in saying that there is a unique analytical continuation of the zeta function for Re(z)

  • @jccusell
    @jccusell Год назад +2

    This guy seems very nervous and extremely confident at the exact same time. Also very ackward and very elloquent at the same time. Very interesting.

  • @ChemicaLove
    @ChemicaLove 11 месяцев назад +20

    This guy talks like how I type

  • @danielstanev5685
    @danielstanev5685 Год назад +24

    This guy dodged brady's question "How do you measure that digits even-out" like 3 times, which was the thing I was most curious about . I feel like his explanations were just too superficial.

    • @peterjoeltube
      @peterjoeltube Год назад +10

      I'm wondering if the "real" explanation was lost in editing because it was deemed to hard for the average viewer to understand, but I'd have preferred to see something I didn't understand that the completely unsatisfying result that we got instead.

    • @jrenema
      @jrenema Год назад +8

      No he didn't, he mentioned it can be done in terms of the variance

    • @obscurity3027
      @obscurity3027 6 месяцев назад +1

      He explained that it’s all about the standard deviation. He didn’t expand on that or provide any actual data, but he did technically answer the question.

  • @AnnevanRossum
    @AnnevanRossum 7 месяцев назад

    I first had to check if RUclips wasn't set on speed 1.5 when he started to speak. Then I set it to that and it was still intelligible, amazing clear speech.

  • @lindavid1975
    @lindavid1975 Год назад +3

    As Brady says, that was a unique angle/approach to the RH, which I had not seen before - I had often wondered if working in another base would throw up something more interesting - but as usual, "I've done so. Arne Saknussemm"

  • @calvincrady
    @calvincrady Год назад +2

    10:11 "Almost all" and "over 99.9%" are different; does anyone know which it is? In other words: as x → ∞, the percentage of bases b < √x for which (*) holds goes to some constant k; does the Bombieri-Vinogradov theorem state that k = 100% or that k > 99.9%? (The big-O notation on the wikipedia page for the B-V theorem makes me think it's the former, but I understood _almost none_ of that article :P)

  • @error.418
    @error.418 Год назад +5

    So thankful he took the time to show bases other than 10, much appreciated

  • @PunmasterSTP
    @PunmasterSTP Год назад +2

    I still don't entirely understand how "closeness" or "convergence" was measured in this sense, but the video was still fascinating to watch!

  • @zubairiafaisal5340
    @zubairiafaisal5340 Год назад +1

    love this guys new take on an old topic

  • @Piggye98
    @Piggye98 Год назад +19

    I HAVEN’T SEEN NUMBERPHILE IN SO LONGGGGG
    THIS IS BRINGING BACK MEMORIES 😭😭😭

    • @ZelenoJabko
      @ZelenoJabko Год назад +1

      Maybe subscribe to the channel then? They are posting regularly, this is your fault

  • @brightsideofmaths
    @brightsideofmaths Год назад +23

    Use the Riemann hypothesis to clean the house :D

  • @MyTBrain
    @MyTBrain Год назад +1

    Why does it seem like the limit is going towards x^0.618 (1/phi)?
    Edit: Timestamp 16:07 is where I'm talking about

  • @MathsMadeSimple101
    @MathsMadeSimple101 Год назад +2

    One day this channel will reach ten million subscribers. I'm calling it.

  • @psmirage8584
    @psmirage8584 Год назад +2

    I love how Brady comes up with the example of the Fermat's Last Theorem - that a single counterexample would destroy it. Riemann Hypothesis is no different. But, from what I understand, Fermat's Last Theorem has now been proven. It's sort of a "race" between finding a single counterexample to the Riemann Hypothesis and finding a proof for it. I'm rooting for the proof, of course, but a counterexample would be fascinating.

  • @agargamer6759
    @agargamer6759 Год назад +3

    Crazy how so much is connected to the Reimann hypothesis

  • @g-nonymousgems3047
    @g-nonymousgems3047 Год назад +2

    Funny how the length of the video 20:27 (2027) is also a Prime Number.

  • @JabeRaddle
    @JabeRaddle Год назад +1

    Me at 2am not having taken a math class in a decade: hmm yes yes the Riemann Hypothesis of course

  • @Oskoreii
    @Oskoreii Год назад +1

    What are the practical applications that he is takling about? 17:21

  • @theadamabrams
    @theadamabrams Год назад

    Brady tried to ask at 7:28 and again at 14:28 and 16:08, but I never heard a clear answer about what "evening out SOON" means. I get that lim_(x→∞) π(x;a,b) / (π(x)/φ(b)) = 1, but it doesn't make any sense to me to say that a statement about limits "holds once x > b²". And the versions with "b < √x" or "b < x⁰ᐧ⁶¹⁷" make even less sense to me because if we stop at some finite x, then π(x;a,b) won't actually be π(x)/φ(b), regardless of b. I still have no idea what the actual claim about "closeness" is.

  • @johnchessant3012
    @johnchessant3012 Год назад +13

    This seems like a really neat result, but I'm afraid I'm not understanding what it's saying. Doesn't the "~" symbol in π(x; b, a) ~ π(x)/ϕ(b) imply that we're taking the limit as x -> infinity? (specifically, the limit of the ratio of both sides equals 1) So when we say b < x^(1/2), won't all x eventually satisfy that as we let x -> infinity?

    • @hypnogri5457
      @hypnogri5457 Год назад +7

      He is saying that you can set b to sqrt(x) and it will still hold. You can let the base grow at the same time as the x

    • @cjc6063
      @cjc6063 Год назад

      (from someone who doesn't understand much of what was said) it typically means 'proportional to', no?

    • @carstenmeyer7786
      @carstenmeyer7786 Год назад +2

      The property you're looking for is "asymptotically equivalence", represented by "~". The definition is
      *"an ~ bn" "an / bn --> 1 for n --> oo"*
      Roughly speaking, if *"an ~ bn",* they have roughly the same behavior for large *n.*

    • @tomalata5742
      @tomalata5742 Год назад

      From my understanding...Yes they will but that is the Riemann Hypothesis. The first is just the Prime Number Theorem, the RH simply asserts that we don't need to take limit to infinity but rather x only needs to be bigger than the square of b for the relation to hold

  • @TomLeg
    @TomLeg Год назад +22

    My first question is what are the characteristics of that 0.1%?

    • @jimmyh2137
      @jimmyh2137 Год назад +11

      The 0.1% is not proven. It MIGHT follow Riemann, or it might be different and break.

    • @jessehammer123
      @jessehammer123 Год назад +12

      My understanding is that it’s not a 99.9-0.1 split or something like that, it’s an “almost all”-“almost none” split where the probability of it working is exactly 1, but there may still be failure cases. For example, it could fail for the primes, or for the powers of 2, or for the set {1, 50000000}, or never. The point is that we know it works “every” time.

    • @tomalata5742
      @tomalata5742 Год назад +1

      The 0.1% was him trying to be a bit friendly (at least to pass his point) but when the technical word "almost" is used in Mathematics it simply means the probability of finding anything that violates what was being discussed is zero. It is just that when talking about probabilities (or any measure at that) of infinite sets a probability of zero does not mean impossible just improbable

    • @seneca983
      @seneca983 Год назад +1

      @@tomalata5742 I'm not sure assigning a probability here would make sense (without choosing some arbitrary distribution for different bases). I think what is meant here is that when you look at bases up to a limit and that limit grows then the share of the bases where the hypothesis doesn't hold tends to zero.

    • @tomalata5742
      @tomalata5742 Год назад

      @@seneca983 Hello I'm finding trouble following. Kindly elaborate. From my understanding sometimes we can make general statement about distributions without knowing the particulars of distribution for example, the chebyshev's inequality. we know it holds for a class of distributions that satisfy certain properties, the same could apply for this one, that is, the statement holds for the class of distributions that model the distribution of primes under different bases

  • @jonadabtheunsightly
    @jonadabtheunsightly Год назад +5

    The distribution of primes actually does have a connection to pi (or, if you prefer correctness over deeply-rooted tradition, to tau). 3blue1brown did an entire video on this connection.

    • @bigmouthfisheyes
      @bigmouthfisheyes Год назад

      The expressions 6n +/- 1 produce all prime numbers greater than three, and many more composite numbers. If we knew exactly where the composite numbers would appear in these sequences, we could infer the location of all of the prime numbers. Am I understanding this correctly? Of what use would this be to anyone?

    • @danielmurogonzalez1911
      @danielmurogonzalez1911 Год назад

      Can you link me to the video you are referring to? I'm very curious about this video.

  • @PplsChampion
    @PplsChampion Год назад +2

    ive rewatched this one a few times already, it's teasingly deep and addictive

  • @bentoomey15
    @bentoomey15 Год назад +1

    @3:40 --- that looks more like Dirichlet's theorem on primes in arithmetic progression in action than it does the Prime Number Theorem.

    • @bentoomey15
      @bentoomey15 Год назад

      I now see he's calling it the PNT for APs. I'm a fan of removing people's names from theorems, I like it!
      (but maybe the "for arithmetic progressions" should be used the first time it's mentioned for clarity).

  • @samanthaalejandre693
    @samanthaalejandre693 Год назад +4

    I couldn’t help but notice, the number 0.617 is very close to the golden ratio that is approximately 0.618. I wonder if there is a connection 🤔

    • @docwunder
      @docwunder Год назад

      I actually doubt that.
      The golden ratio is the solution of the equation
      x²-x-1=0.
      You may find an infinite number of either functional results which may come close to that without having anything to do with it.

    • @Derek_Bell
      @Derek_Bell 11 месяцев назад

      ​​@@docwunderYet the golden ratio shows up in other places
      Say you wanted to solve the differential equation
      f'(x) =f^-1 (x), where f^-1 is the inverse function of x
      The solution involves x^phi

  • @shingofan
    @shingofan Год назад +3

    13:58 AND THIS IS TO GO FURTHER BEYOND!

  • @benoitalain5833
    @benoitalain5833 Год назад

    He never defined "evens out", he just said it's about the standard deviation. Be a bit more formal next time maybe? Also he's waving hands a lot and not explaining how the Riemann Hypothesis based result works, or any other results. Also not being able to give examples of bases that work for b < x^0.617 makes it even more vague. Maybe we proved that it works for all b < x^0.617, pushing the upper bound for b a bit beyond what we could prove using RH? I'm left with a bit of confusion on what exactly is the result that I should be excited about.

  • @davidgillies620
    @davidgillies620 Год назад +2

    It's not generally appreciated how many theorems in number theory (which deals with integers) are statistical (and thus involving real numbers). The Prime Number Theorem in its simplest form is an example.

  • @pepega3344
    @pepega3344 Год назад +8

    I still don't get what does 'starts to even out' mean... If you look at variance, what limit of variance do you set?

    • @xenmaifirebringer552
      @xenmaifirebringer552 Год назад +3

      Yeah, he never explained what limit for "evening out" is being considered. I feel he should also have given some examples of an increasing b value as oppossed to just a constant base, since it seems hard to visualize.

    • @severnkariuki9129
      @severnkariuki9129 Год назад +2

      I suppose infinite evening out. 0 variance as x tends to infinity.

  • @ronin6158
    @ronin6158 Год назад +1

    'Im not so fast at writing these numbers down... especially if they go on forever.'
    Yogi Berra-ism if Ive ever heard one.

  • @bwderge187
    @bwderge187 Год назад +1

    I really like this dude. He could talk math to me all day

  • @josuel.9598
    @josuel.9598 Год назад +28

    When he said James Maynard was “up there “ I thought he meant he passed away😅

    • @xyz.ijk.
      @xyz.ijk. Год назад +2

      Thankfully not!
      I thought he meant James was watching over this presentation because of his recent Field's Medal award ... but I think James is humble and almost incapable of such hegemonistic thoughts.

    • @clutchmatic
      @clutchmatic 10 месяцев назад

      As an economist, I was excited about Keynesian economics until I remembered that it is John, not James

  • @SorteKanin
    @SorteKanin Год назад +9

    I still feel like I'm missing what "evenness" more precisely means here.

    • @tomalata5742
      @tomalata5742 Год назад

      I believe it meant that the height of the bars becoming the same with small fluctuations in this case

    • @SorteKanin
      @SorteKanin Год назад

      @@tomalata5742 That's not very precise. How small are these fluctuations allowed to be? It's only 0 in the limit.

  • @jamesbliss758
    @jamesbliss758 Год назад +3

    Jared's cadence reminds me of Christopher Walken. Lovely video :)

  • @Alex-yj9lw
    @Alex-yj9lw 11 месяцев назад +1

    May your passion for mathmatics rub off in my life so I may find something im so passionat about, amazing lecture 🥳

  • @YawnGod
    @YawnGod Год назад +1

    "Morally speaking..."
    I choke-coughed.

  • @samirfarsane2379
    @samirfarsane2379 Год назад

    I Have solved the Riemann hypothesis 8 years ago. I've reached out to many academics. NONE ANSWERED. Reached to this channel (multiple hosts) NONE ANSWERED. So I enjoyed the result and kept it to myself. ABSOLUTELY SERIOUS HERE. NO JOKES!!!

    • @rand0m_694
      @rand0m_694 Год назад

      Proof?

    • @samirfarsane2379
      @samirfarsane2379 Год назад

      @@rand0m_694 hint rather: the reason no one could solve it is because everyone is still using Euclid's 5000 year old primarility check formula. I found a better one and it opened a sea of new mathematics for me.

    • @MrM1729
      @MrM1729 10 месяцев назад +1

      He has the proof, but it’s too large to fit in the comment box.

    • @samirfarsane2379
      @samirfarsane2379 10 месяцев назад

      @@MrM1729 if you believe that the proof to one of Math's hardest problems should be posted in the comments, then I should believe your entire knowledge base should fit inside these comments.

    • @Giantcrabz
      @Giantcrabz 2 месяца назад

      big if true

  • @benbertrand3946
    @benbertrand3946 Год назад +2

    Nice to watch and thanks for sharing

  • @androidlogin3065
    @androidlogin3065 Год назад +4

    What distribution would have if counting not how many times last digit of a prime, instead counting how many times last digit of a prime is equal to next prime last digit?
    Example: 2, 3, 5, 7, 11, 13, 17 -> in such there are not tow consecutive primes that have the same last digit, so count is still 0 up to that point.
    How many two consecutive primes have '1' as last digit? And '3'? ... so on.
    How that counts grow?
    Hope it is clear.

    • @japanada11
      @japanada11 Год назад +1

      Numberphile did a video on this question! Look up "The Last Digit of Prime Numbers"

    • @TheEternalVortex42
      @TheEternalVortex42 Год назад +1

      You are just asking how many prime gaps of 10 there are. There is the "Polignac's conjecture" which states that there are infinite primes with gaps of 2n (for n=1 this is the twin prime conjecture). Currently it's only proven for 2n >= 246.
      If you want to know how often this occurs as a %, then it tends to 0 as n -> infinity because the average gap in primes goes like log(n).

    • @killerbee.13
      @killerbee.13 Год назад

      @@TheEternalVortex42 This isn't exactly that, because it's not a prime gap of exactly 10, it's a prime gap of 10n. e.g. 31, 51 both end in 1 but have a gap of 20. Though if there are infinitely many primes with a gap of 250 (the next multiple of 10 after 246) then there are infinitely many adjacent primes with the same last digit. Though I think that the question was more getting at "is the sequence pairwise equidistributed?" rather than "are there infinitely many pairs where the any last digit is repeated at all", which would be satisfied even if the sequence 3, 3 happened only finitely often for example.

  • @danamulter
    @danamulter Год назад +1

    This one I could follow, but it wavers back and forth about being specific. "Kind of nice even distributions" versus "we want a complete understanding". There's this handwaving feel about how meticulous you are in the distribution being equal, versus the tiny incremental testing of x.617. It sounds like digging to be exact while handwaving the parts that say you're exact.

  • @sdrc92126
    @sdrc92126 Год назад +1

    I love Mathematica for exactly this purpose

  • @AZALI00013
    @AZALI00013 Год назад +1

    very cool result !! >:0

  • @snuff248163264
    @snuff248163264 Год назад +1

    this dude could go on that for hours i can tell

  • @robertr7923
    @robertr7923 Год назад

    how are those histograms correct? For example at 4:55, we look at 1000 primes, but the charts go up to 80. And the chart for 2 and 5 was 1 at the beginning of the video

    • @SunShine-xc6dh
      @SunShine-xc6dh 10 месяцев назад

      Nah it's all the primes up to 1000 not the first 1000 primes.

  • @azrobbins01
    @azrobbins01 Год назад

    What are the round things in the floor? Speakers? Ventilation? Access panels for power outlets and data ports? Surely not water drains.

  • @DavidPanofsky
    @DavidPanofsky Год назад +17

    If the √x result links back to the zeros along the critical line of 1/2, what does this x^.617 result say about RH (if anything)?

    • @lppunto
      @lppunto Год назад +24

      So part of the reason this is tricky to answer is that the relationship is not actually to non-trivial zeros of the zeta function, but to non-trivial zeros of a more general class of functions known as L-functions (this is the "generalized" part of the "Generalized Riemann Hypothesis"). Now, the x^1/2 bound seen here is directly tied to the conjecture that non-trivial zeros of l-functions lie on the critical line Re(z) = 1/2, just as in the Riemann Hypothesis. Very loosely speaking (I don't understand the details myself), the Bombieri-Vinogradov theorem can be viewed as saying something about how the zeros of different L-functions interact, rather than saying something stronger about just one L-function. That's why it's a bit "orthogonal" to the GRH: it is in one way stronger by saying things about multiple functions at once, but weaker in terms of what it's claiming about those functions - namely, just some amount of cancellation between their zeros, rather than restraining them all to a line.

    • @_Wombat
      @_Wombat Год назад

      @@lppunto at this point I feel like mathematics approaches the realm of magic tbh

  • @noterictalbott6102
    @noterictalbott6102 Год назад +2

    I'm not sure if this guy has ever been on the channel before but i like him!

  • @Mrstealurgrill
    @Mrstealurgrill Год назад +3

    I still don’t get what is b and x.

  • @sdrc92126
    @sdrc92126 Год назад +2

    In an infinite world of numbers, only a vanishingly small percentage are useful.

  • @rif6876
    @rif6876 Год назад

    Always good to see new mathematicians, I like this guy
    Base "e" - someone must have tried it, and I cannot believe the results were uninteresting.

    • @therealax6
      @therealax6 Год назад +2

      If you follow the rule that all digits must be smaller than the base (so that the only digits allowed are 0, 1 and 2), integers above 2 don't even have a finite representation, so there's no last digit to look at.

  • @danfg7215
    @danfg7215 Год назад +4

    I still don't understand the Riemann Hypothesis, but I'll try to understand it a bit better next time it's mentioned or explained in other videos.

    • @sapwho
      @sapwho 7 месяцев назад

      The riemann hypothesis states that the all the zeroes of riemann’s zeta function exist on complex plane’s line x=1/2

    • @danfg7215
      @danfg7215 7 месяцев назад

      @@sapwho hi, I really appreciate your comment, and I'm baffled by how little it cleared things up for me regarding the riemann hypothesis or its significance.

    • @sapwho
      @sapwho 7 месяцев назад +1

      @@danfg7215 are you familiar with the zeta function 😂

    • @sapwho
      @sapwho 7 месяцев назад +1

      @@danfg7215 are you familiar with the zeta function itself? that may be the place to start ❤️

  • @midcam
    @midcam 11 месяцев назад

    16:05 Shouldn't it be x*600 in the middle at 20%? It doesn't seem logic otherwise for me. Nevertheless, great video!

  • @matthewsullivan3804
    @matthewsullivan3804 Год назад +1

    you guys wouldnt know, but this guy can ball fr

  • @konstantinosadamopoulos9918
    @konstantinosadamopoulos9918 Год назад +1

    So happy he said fi for φ and not fai

  • @mmburgess11
    @mmburgess11 Год назад +1

    "I'm not buyin' it!". This is why we love Brady.

  • @BLenz-114
    @BLenz-114 Год назад +3

    Have you ever had someone talk about prime numbers in other bases? He touched on it here, but it got me wondering. I'd be looking for a pretty simple "Intro to other base primes".

    • @davidgustavsson4000
      @davidgustavsson4000 Год назад +3

      A prime in any other base is still a prime.

    • @BLenz-114
      @BLenz-114 Год назад +1

      @@davidgustavsson4000
      Hmmm. Yeah I guess 5 is still 5, it just looks different. A bit tough to get my head around. I'm not a mathematician, can you tell? 😉

    • @davidgustavsson4000
      @davidgustavsson4000 Год назад

      @@BLenz-114 for example, 111 base 2 (= 7 base X) is prime, because it doesn't matter which base you express it in.

    • @blableu4519
      @blableu4519 Год назад

      Any prime isn't divisible by any number other than itself and one, so setting the base to any number won't change that fact.

  • @fatmn
    @fatmn Год назад

    And this is to go even further beyond (the Riemann Hypothesis)!!!

  • @bholdr----0
    @bholdr----0 Год назад

    @9:41: OOOOOHHHH!! Dang! Wow... it's like finding Pi in continuing fractions a la Ramanujan, or in disparate, unexpected physical processes and...
    Then... wait... I still can't see Riemann clearly.

  • @StrayCatInTheStreets
    @StrayCatInTheStreets Год назад

    This is such an excellent video. Bravo

  • @mrborisak
    @mrborisak Год назад

    his enthusiasm is great

  • @SobTim-eu3xu
    @SobTim-eu3xu 6 месяцев назад

    Oh, my new most loved mathematician!)

  • @Pseudify
    @Pseudify Год назад +2

    I had trouble following the connection with RH and the b < x^exp equations. For starters, if RH was proven, would that exponent become a new value? Like 1? Or are they measuring different things? Also, I don’t understand the concept of “beyond RH”.

    • @ethanbottomley-mason8447
      @ethanbottomley-mason8447 Год назад +2

      The connection with the Riemann hypothesis is that it yields better approximations of the prime counting function. The generalized Riemann hypothesis then gives better approximations for Dirichlet L functiona which is what you want for this problem in particular. If you do some contour integration and use Perron's forumula, you can show that the Riemann hypothesis is equivalent to the bound psi(x) = x + O(sqrt(x)log(x)) which with some more integration can be shown is equivalent to the bound pi(x) = Li(x) + O(sqrt(x)log^2(x)). You get something similar for the Dirichlet L functions under the GRH which is what is really being used here.

    • @billcook4768
      @billcook4768 Год назад +2

      The Riemann Hypotheses only gets you to 1/2 for this particular question. But if proven, it always works. There are other ways to get similar results, and actually stronger results (ie you get to evenness faster) but only work most of the time. These other ways have been proven.

    • @tomalata5742
      @tomalata5742 Год назад +1

      @@billcook4768From my understanding the results are not necessarily stronger than the Riemann hypothesis but rather allow to you to do away with it and achieve the required result. Case in point here we see that RH asserts the statement is true for all b < sqrt(x) but then Bombieri-Vinogradov a slightly weaker result show that the statement holds for almost all b's which isn't equal to RH which says that it holds for all b, but this result allows us to do away with RH in some of the cases. Even the subsequent statements are just but the tightening of Bombieri-Vinogradov theorem

  • @Dorgpoop
    @Dorgpoop Год назад

    Jared is a good explainer. Obviously this is a subject thats far too complicated to explain comprehebsively, but he did a great job at giving an intuitive understanding of the core principle and offered glimpses of why this problem is so important to many different areas of maths.

  • @not1not2but3
    @not1not2but3 Год назад +3

    Ok, Is it me, or does he have the EXACT same cadence as the “Autistic news reporter” from the onion. Don’t mean it in a bad way, but that was the first thing I heard.

    • @Giantcrabz
      @Giantcrabz 2 месяца назад

      😂😂😂😂

  • @KulakOfGulag
    @KulakOfGulag Год назад +2

    Dmitry Bivol very smart guy i see

  • @MikePaixao
    @MikePaixao Год назад +2

    The solution to the hypothesis ended up being a way to predict quantum randomness

  • @friggy1899
    @friggy1899 Год назад +4

    So what am I getting from this is that you can do the Riemann Hypothesis 23% faster?

    • @2424Lars
      @2424Lars Год назад

      The Riemann Hypothesis sets an upper limit, and these numbers are 23% below that limit

  • @christianellegaard7120
    @christianellegaard7120 Год назад +11

    I find that one of the fun things about primes is that there are all these hard and fast rules about them, except for the single digit numbers.
    It just seems cheeky that 2 and 5 sneak in there at the beginning.

    • @Jesin00
      @Jesin00 Год назад +12

      That's because the thing that makes primes special is that every single one of them introduces a new rule that only it breaks. No prime is divisible by 2, except 2. No prime is divisible by 5, except 5. No prime is divisible by 23, except 23. And so on for every prime. It just happens that 2 and 5 are the ones that are visible in the last digit because they're the factors of 10 which we chose as our base to write numbers in.

    • @GynxShinx
      @GynxShinx Год назад +1

      Even 3 isn't THAT prime. If you look at a chart of primes in base 10, the ones ending in 3 go prime, prime, composite, prime, prime, composite, except the very first one "3" which breaks the pattern and is prime.

    • @CutleryChips
      @CutleryChips Год назад

      Haha what kind of rules are you using? 0,1,2,3,4,5,6,7,8,9 is manmade single digits. Let’s say we had one more digit for ten, T. In this system, T is ten, 10 is eleven, …, 19 is twenty, 1T is twenty-one and so on.
      Then these are the primes in Base-11.
      2,3,5,7,10,12,16,18,21,27,29,34,38,3T,43,49,54,56,61,65,…
      For comparison, these are the primes in Base-10.
      {2,3,5,7,11,13,17,19,23,29,31,37,41,43,47,53,59,61,67,71,…}
      Notice 65 is a prime in base-11.
      That is 6*11+5 =71 in base 10.
      So, there’s nothing special about 2 or 5.

    • @Jesin00
      @Jesin00 Год назад

      @@GynxShinx That pattern gets broken up by more and more composites as the numbers get bigger; you just might not notice it in reasonably-sized tables.

    • @GynxShinx
      @GynxShinx Год назад +1

      @@Jesin00 Yeah, still funky though. I feel like in another universe where 3 was larger, whatever that means, it may have been composite, where 7 on the other hand just seems very prime, even in other bases.

  • @AdamCallison
    @AdamCallison Год назад

    When Brady said that James Maynard is "up there somewhere" and pointed the camera at the ceiling, I genuinely thought he had died

  • @muskyoxes
    @muskyoxes Год назад +1

    I'm already shocked that up to a million it's almost exactly equal for primes to be 1 mod 3 or 2 mod 3. If it was random, wouldn't one of them drift a couple hundred away from the other? There should be some kind of theorem that is the "power" that holds them more closely together.
    This video makes the point that each remainder has equal chance, but so does a coin flip have equal chance, yet heads and tails will drift apart.
    I guess this is saying that the _total_ is constrained, not the probability, which is cool, and out of my reach to know how the constraint works

    • @Temari_Virus
      @Temari_Virus Год назад +1

      Sometimes the distributions do drift apart considerably, but like with coin flips, we expect any significant drifts to be less and less likely as the numbers go up, but they can still happen at any point