Today I want to share with you a neat way
to solve the towers of Hanoi puzzle just by counting in a different number system, and surprisingly this stuff relates to
finding a curve that fills Sierpinski triangle. I learned about this from a former CS lecturer of mine, his name is Keith Schwarz. And I’ve got to say, this man is one of
the best educators that I’ve ever met. I actually recorded a bit of the conversation
where he showed me this stuff, so you guys can hear some of
what he described directly. It’s weird, I’m not normally the sort of person
who likes little puzzles and games, but I just love looking at the analysis
of puzzles and games, and I love just looking at mathematical patterns and (ask): where does that come from? In case you aren’t unfamiliar, let’s just lay down what the towers
of Hanoi puzzle actually is. So you have a collection of three pegs,
and you have these discs of descending size. You think of these disks as having a hole in the middle, so that you can fit them onto a peg. The set I pictured here has five discs,
which I’ll label 0, 1, 2, 3, 4, but in principle you could have as
many discs as you want. So they all start up stacked up from
biggest to smallest on one spindle, and the goal is to move the entire
tower from one spindle to another. The rule is you can only move
one disc at a time, and you can’t move a bigger disc
on top of a smaller disk. For example, your first movemust involve moving disk 0, since any other disk has stuff on top of it that needs to get out of the way before it can move. After that, you can move disc 1, but it has to go on whatever peg doesn’t currently have disk 0, since otherwise you’ll be putting a bigger disk on a smaller one, which isn’t allowed. If you’ve never seen this before, I highly encourage you to pause and pull out some books of varying sizes and try it out for yourself, just kind of get a feel for what the puzzle is: if it’s hard, why it’s hard, if it’s not, why it’s not, that kind of stuff. Now, Keith showed me something truly surprising about this puzzle, which is that you can solve it just by counting up in binary and associating the rhythm of that counting with a certain rhythm of disc movements. For anyone unfamiliar with binary, I’m going to take a moment to do a quick overview here first. Actually, even if you are familiar with binary, I want to explain it with a focus on the rhythm of counting, which you may or may not have thought about before. Any description of binary typically starts off with an introspection about our usual way to represent numbers – what we call base-10 – since we use ten separate digits, 0123456789. The rhythm of counting begins by walking through all ten of these digits, Then, having run out of new digits, you express the next number, ten, with two digits: 10. You say that one is in the tens place, since it’s meant to encapsulate the group of 10 that you’ve already counted up to so far, while freeing the ones place to reset to zero. The rhythm of counting repeats like this: counting up nine, rolling over to the tens place, counting up nine more, rolling over to the tens place, etc. Until after repeating that process nine times, you roll over to a hundreds place, a digital that keeps track of how many groups of 100 you’ve hit, freeing up the other two digits to reset to zero. In this way, the rhythm of counting is kind of self-similar: even if you zoom out to a larger scale, the process looks like doing something, rolling over, doing that same thing, rolling over, and repeat nine times before an even larger roll over. In binary, also known as base-2, you limit yourself to two digits, 0 and 1, commonly called ‘bits’, which is short for binary digits. The result is that when you’re counting, you have to roll over all the time. After counting 01 you’ve already run out of bits, so you need to roll over to a two’s place, writing ’10’ and resisting every urge in your base-10-trained brain to read this as ten, and instead understand it to mean one group of 2 plus 0. Then, increment up to 11, which represents three, and already you have to roll over again, and since there’s a one in that two’s place, that has to roll over as well, giving you 100, which represents one group of four plus 0 groups of two plus 0, in the same way that digits in base-10 represent powers of 10, bits in base-two represent different powers of 2. So instead of talking about a ten’s place, a hundred’s place, a thousand’s place, things like that, you talk about a two’s place, a four’s place and an eight’s place. The rhythm of counting is now a lot faster, but that almost makes it more noticeable: Flip the last, roll over once. Flip the last, roll over twice. Flip the last, roll over once. Flip the last, roll over three times. Again, there’s a certain self-similarity to this pattern: at every scale the process is to do something, roll over, then do that same thing again. At the small-scale, say counting up to three, which is 11 in binary, this means flip the last bit, roll over to the two’s, then flip the last bit. At a larger scale, like counting up to fifteen, which is 1111 in binary, the process is to let the last three count up to seven, roll over to the eight’s place, then let the last three bits count up again. Counting up to 255, which is eight successive ones, this looks like letting the last seven bits count up till they’re full, rolling over to the 128’s place, then letting the last seven bits count up again. Alright, so with that mini introduction, the surprising fact that Keith showed me is that we can use this rhythm to solve the towers of Hanoi. You start by counting from zero. Whenever you’re only flipping that last bit from a 0 to a 1, move disc 0 one peg to the right. If it was already on the rightmost peg, you just loop it back to the first peg. If in your binary counting, you roll over once to the two’s place, meaning you flip the last two bits, you move disc number 1. “Where do you move it?” you might ask. Well, you have no choice. You can’t put it on top of disk 0 and there’s only one other peg, so you move it where you’re forced to move it. So after this, counting up to 11, that involves just flipping the last bit, so you move disk 0 again. Then, when your binary counting rolls over twice to the four’s place, move disc number 2, and the pattern continues like this: flip the last, move disk 0, flip the last 2, move disc 1, flip the last, move disk 0. And here we’re gonna have to roll over three times to the eight’s place, and that corresponds to moving disc number 3. There’s something magical about it, like when I first saw this, like this can’t work. I don’t know how this works, I don’t know why this works. Now I know, but it’s just magical when you see it and I remember putting together animation for this when I was teaching this, and just like… you know, I know how this works, I know all the things in it, it’s still fun to just sit and just like you know… -Watch it play out? -Oh yeah. I mean, it’s not even clear at first that this is always going to give legal moves. For example, how do you know that every time you’re rolling over to the eight’s place, the disc 3 is necessarily going to be freed up to move? At the same time the solution just immediately raise these questions like: where does this come from, why does this work, and is there a better way of doing this then having to do 2^(n-1) steps? It turns out not only does this solve towers of Hanoi, but it does it in the most efficient way possible. Understanding why this works and how it works and what the heck is going on comes down to a certain perspective on the puzzle – what CS folk might call a recursive perspective. Disc 3 is thinking, okay 2 1 and 0, you have to get off of me, I can’t really function under this much weight and pressure. And so just from disc 3’s perspective, if you want to figure out how is disc 3 going to get over here, Somehow, I don’t care how, disc 2 1 0 have to get to spindle B. That’s the only way they can move, If any of these are on top of 3, I can’t move it, any of these are at spindle C, it can’t move there. So somehow we have to get 2, 1 and 0 off. Having done that then we can move disc 3 over there. And then disc 3 says, I’m set, you never need to move me again, everyone else just figure out how to get here. And in a sense you now have a smaller version of the same problem: now you’ve got disc 0, 1 and 2 sitting on spindle B, we gotta get them to C. So the idea is that if I just focus on one disc and I think about what I’m going to have to do to get this disc to work, I can turn my bigger problem into something slightly smaller. And then how do I solve that? Well it’s exactly the same thing, disc 2 is going to say, disc 1 and disc 0, you need to, you know, it’s not you, it’s me, I just need some space, get off. They need to move somewhere, then disc 2 can move to where it needs to go, then disc 1 and 0 can do this. But the interesting point is that every single disc pretty much has the exact same strategy: They all say, everybody above me, get off, then i’m going to move, ok everyone come back on. When you know that insight you can code up something that will solve towers of Hanoi in I think five or six lines of code, which probably has the highest ratio of intellectual investment to lines of code ever. And if you think about it for a bit, it becomes clear that this has to be the most efficient solution. At every step you’re only doing what is forced upon you. You have to get discs 0 through 2 off before you can move disc 3, and you have to move disc three, and then you have to move disk 0 through 2 back on to it. There’s just not any room for inefficiency from this perspective. So why does counting in binary capture this algorithm? Well what’s going on here is that this pattern of solving a subproblem, moving a big disk, then solving a subproblem again, is perfectly paralleled by the pattern of binary counting: count up some amount, roll over, count up to that same amount again. And this towers of Hanoi algorithm and binary counting are both self similar processes, in the sense that if you zoom out and count up to a larger power of 2, or solve towers of Hanoi with more discs, they both still have that same structure: Subproblem, do a thing, subproblem. For example, at a pretty small scale, solving towers of Hanoi for two discs, move disc 0, move disc 1, move disc 0, is reflected by counting up to three in binary: flip the last bit, roll over once, flip the last bit. At a slightly larger scale, solving towers of Hanoi for three discs looks like: doing whatever it takes to solve two discs, move disc number 2, then do whatever it takes to solve two discs again. Analogously counting up to 111 in binary involves counting up to three, rolling over all three bits, then counting up three more. At all scales, both processes have this same breakdown. So in a sense, the reason that this binary solution works, at least an explanation, I feel like there’s no one explanation, but, I think the most natural one is that the pattern you would use to generate these binary numbers has exactly the same structure as the pattern you would use for towers of Hanoi, which is why if you look at the bits flipping, you’re effectively reversing this process. You’re saying what process generated these, like if I were trying to understand how these bits were flipped to give me this thing, you’re effectively reverse engineering the recursive algorithm for tower of Hanoi, which is why it works out. That’s pretty cool, right? But it actually gets cooler, I haven’t even gotten to how this relates to Sierpinski triangle. and that’s exactly what I’m going to do in the following video, part 2. Many thanks to everybody who is supporting these videos on Patreon. I just finished the first chapter of Essence of Calculus, and I’m working on the second one right now, and Patreon supporters are getting early access to these videos before I publish the full series in a few months. this video and the next one are also supported by Desmos.


100 Comments

1201 minchankim · November 13, 2018 at 8:07 am

한국어 자막좀요 ㅠㅠ

Lex Nellis · November 16, 2018 at 8:23 am

This is the first video where I knew everything prior to watching it. Defiantly a different feeling when watching.

Aperson101 · November 20, 2018 at 5:33 am

Why is this puzzle has the same names as my country’s capital city?

Guy3nder · November 20, 2018 at 5:27 pm

this recursion reminds a lot of how in physics certain principles repeat in different models. how you can see laws in physics repeat in chemistry, then in biology, and then even down to sociology and psychology. the most basic ideas in science apply to all scientific fields in some way and remain amazingly consistent. this is why i love math and science and what really makes religion look like horsecrap next to it.

My Username is Pointlessly Long and Intimidating · November 23, 2018 at 9:05 pm

This only works if the towers are allowed to be completed on the 2nd tower instead of the 3rd rip. It works in similar ways either way.

hy nguyen · November 25, 2018 at 2:32 am

i would like to know similar channel to this but in physics subject. Does any one have a list of name? thank you so much.

Laurin Neff · November 26, 2018 at 8:46 pm

There are 111 types of people: those who understand binary, those who don't understand binary, and those who didn't expect a unary joke here

Gabriel-Teofil Mititelu · November 28, 2018 at 10:45 am

@3Blue1Brown, any idea if this also works on a modified ToH problem like, N disks, but with M pegs?

Joe Smith · November 29, 2018 at 1:54 pm

Could write this in assembly the moving of bits to complete the disk tower

Nicodemeus · December 9, 2018 at 1:59 am

"Its not you its me" im dying

Rizky Maulana Nugraha · December 11, 2018 at 1:06 pm

Binary works because the algorithm reduce to binary complexity for each actions.
This is useful when you measure it's complexity.
For each depth of recursion, we essentially do 1 + 2*(steps for that depths).
To ilustrate, let f(N) be the number of steps needed to move the pegs efficiently.
It is a recursive function.
For f(N) the steps involves:
1. Moving the rest of the pegs (N-1) means f(N-1) steps
2. Moving the bottom peg count as a step
3. Moving the rest of the pegs again, means f(N-1) steps.

algo 1 and 3 is the same, so it is the same steps.

So:
f(N)=2*f(N-1) + 1

The recursion ends when f(1) = 1 (one step to move one peg).
One now can easily reduce the counts by finitely recurse until f(1), which is, if we write it as sum of power of two, is the same as writing binary in this form:
f(N) = 2^N – 1 –> which is basically all 1 in binary until N digits.
So, for 2 pegs, it is like: 11 in binary (2 digit).
3 pegs becomes: 111 in binary (3 digits).
and so on.
It's just because: 2^3+2^2+2^1+1 is equivalent to 1111 in binary for 4 pegs.
Amazing.

Badhbhchadh · December 14, 2018 at 4:31 pm

Sier-π-nski?

Ryan · December 24, 2018 at 4:42 pm

Did You Know That Hanoi Is A Real Place : https://drive.google.com/open?id=10m0h945b31w-xIP5U9TGN7iLYXFQrP_s&usp=sharing

Andres Zeballos · December 27, 2018 at 9:47 pm

I'm going to try it with my child, love the way to correlate to binary with hanoi towers, and the recursive approach

Sunny K · January 5, 2019 at 3:27 pm

5:46
this is when i got what 3b1b was saying: i made a cpde that prints out the stepbystep solving of hanoi that prints out
A C
A B
C B

where A C means move the top one in A to C. i used recalling?(재귀) functions for it: hanoi repeats itself. for blocks in A, move all but the last one to B, move the last one to C, move everything in B to C. how do i move everything except the alst from A to B and B to C? recall that function. all that the function does is move blocks from one place to another. why did i get hanoi=bi in 5:46 ?
<hanoi:moving blocks from A to C>
1.move all-1 from A to B
2.move one to C
3.move all to C(which is the same as step 1)
<binary counting>
1.increment all-1 so all digits are 1
2.add 1
3.increment again(which is the same as step 1)
a programmer's explanation on why this work. 3b1b you and your friend is a genius!

True River · January 7, 2019 at 11:04 am

This is overly complicated if the aim is to solve Hanoi. The shortest algorithm is as follows

1. First move: only legal moves involve moving the smallest. If there's an odd number of disks move the smallest to the destination; if even move smallest to the other peg. Note which direction it moved.

2. On even numbered moves make the only move available move that leaves the smallest where it was

3. On odd moves move the smallest in the same direction as step 1.

That's it!

You do not need the rest of the binary number, just the last digit. In effect you only need odd-even so a simple toggle is enough, adding in modulo 2.

And when adding mod 2, the numbers look the same in any base: 1 0 1 0 …

Binary is a distraction: it's mod 2 that matters

youra ramuser · January 11, 2019 at 7:23 pm

EXTREME SIMPLIFICATION!!

Fort-newton · January 15, 2019 at 9:36 pm

Ur wrong moving disk 3 to the C peg is redundant I think if I’m wrong please let me know

Calo Zheng · January 16, 2019 at 11:56 am

Love the smooth animation.

Kelvyn Brito · January 17, 2019 at 8:02 pm

Very cool! Great explanation!

lapidations · January 19, 2019 at 3:20 am

Starting at 8:18, when we're seeing what disk 3 is thinking, the "3" looks like its kissing lips, like "°3°", you know?

Nicod3m0 Otimsis · January 21, 2019 at 11:52 pm

I know no one cares about a comment in a years-old video, but I think that there is a (for the sake of video editing length, I understand, it is okay) left out topic. In the recursion you propose, you establish that the disk 0 should always flip to the right, and go modular once it reaches the edge, I mean, do module 3 so you return to the tower 0. So, why? And I think this is not a naive question, well, it is, until you consider the next. As it is obvious, you show a lot of examples, animated, so I could pick out a simple pattern. A) you don't care where the tower is newly created B) following your "algorithm", the towers en up solved in different towers C) the tower they finish on is related to the parity, no big deal. But, casually, the parity (I mean if it is an odd or pair number of disks) is very closely related to the last bit in binary, precisely the bit you decided to have an arbitrary behaviour. So my best guess is that because of how you established the recursion, and how you treat the last bit, or in other words, because how you decided to move the ring 0, your towers get solved in a different place. Now the obvious question, can anyone please shut up my pretentious intuition by explaining the rational math behind this so that I can stop thinking about a stupid game and go back to doing hi and elevated maths. (take the last part with some irony)

Rachel Davies · January 29, 2019 at 2:56 am

Sierpinksi’s triangle? More like complicated triforce.

Rachel Davies · January 29, 2019 at 3:00 am

Is it really the most efficient solution it seems like you would never have to move the largest disc until the one where you move it to the center.

Carter Tangeman · February 1, 2019 at 6:51 am

"The most efficient way possible"
7:58
Moves largest disc unnecessarily, changing nothing.

Nelyon Quenya · February 5, 2019 at 3:11 pm

What if you add another stick does it go to base 4?

Trevor MacIntosh · February 17, 2019 at 5:01 am

Honestly, these two videos are my favourites of yours. I’ve rewatched them so many times. SO COOL.

Alexander Skusnov · February 17, 2019 at 6:10 am

1) very fast
2) what about Rubik's cube?

Piotr Masłowski · February 17, 2019 at 12:02 pm

I've just realised how funny is the way the whole world pronounces Sierpiński's surname.

Matthew H · March 3, 2019 at 10:34 pm

1:45. There are also Tower of Hanoi apps that let you try out the puzzle by hand.

Cheeseburger Monkey · March 21, 2019 at 12:09 pm

bits
tets
QUIT s
fifs
sexes
sets
octs
nots
dets

Cheeseburger Monkey · March 21, 2019 at 12:12 pm

6:09 I DIDNT KNOW YOU OVED EXACTLY 1 DISK PER SECOND!?

Bernat Ibáñez · March 23, 2019 at 10:20 pm

When I started doing python I made a script to solve the Hanoi towers, but I didn't know what recursion was so I analyzed different cases and found a really weird relationship between numbers and letters. I thought about recursivity as you mention it here, but I solved it completely different for n disks with the most efficient moves. I could show you the code, but it is very messy and I, to this day, still not understand how the math I did really works xDD

Stefan Toljic · March 25, 2019 at 7:47 pm

Heck yeah!

GNM · April 3, 2019 at 8:18 pm

5:47
salutes in pannenkoek

Lupine Dream · April 4, 2019 at 4:44 pm

Could you count down with an inverse of the rules to say, reverse a set of data in the most efficient way ? Lets say you wanted to stack them upside down on the third stack, or by the median ?

I have so many questions :v

hoang anh vuong · April 7, 2019 at 5:54 pm

I just played with that since I was a kid(I love math). i've found the pattern, calculated the number of moves. But I didn't know why that's works. Until I saw this video.
And I was like: "wow, this is truly fascinating" (btw, I'm really proud that the puzzle's name Hanoi is our capital)

Salim Huerta · April 10, 2019 at 4:12 pm

If I wasn’t busy with anything I’d watch all of your videos non stop, well I do that now anyway but I get a headache because I’m trying to juggle it with things I’m doing already.

Salim Huerta · April 10, 2019 at 4:13 pm

My favorite thing for drives home from work

Jackie Lin · April 12, 2019 at 7:57 pm

https://hanoi-animation–p3artschool.repl.co/

[Tower of Hanoi animation]

tools: repl.it web python 3.6 + svgwrite 1.2

file format: HTML5 svg

Patriot · April 13, 2019 at 7:12 am

This works and i have solved this some years ago on a similar way; by odd and even numbers which when you turn in binary is numbers ending in 1(odd) and numbers ending in 0 (even)

Jeff Carroll · April 17, 2019 at 5:20 am

Also, if you count the starting position as 1, then the least number of moves to solve the puzzle with n disks will always be 2 to the nth power. 1 disk = 2 moves, 2 disks = 4, 3 disks = 8, 16, 32, 64, 128, 256, etc… Solution is always a power of 2.

Gameeustrupter · April 17, 2019 at 9:20 pm

Every once in a while I have to watch one of these twice, mabey because I'm in middle school though

Venom Supreme · April 21, 2019 at 2:18 pm

Isn’t it also the only way to solve it without reversing logic? Therefore any other solution I believe would contain redundancies.

Daffa_FM · April 22, 2019 at 6:27 pm

bin(3) =11
bin(7) = 111
numbin(101) = 5

Sean · April 24, 2019 at 1:55 am

I somehow did come here without watching the other video

The Good Kid Boy · April 29, 2019 at 8:59 pm

8:18
(0 3 0)

Rajat Prakash · May 7, 2019 at 6:10 am

I loved it. Thanks.

OlPa · May 24, 2019 at 5:01 pm

I think you could have highlighted the fact that you are moving the piece one right, and if you can't move it one right, two right. Moving straight left in the animation, in the case of skipping over from the middle position or the rightmost position, had me confused for a while. This mechanic also neatly explains why the piece moving can never move back onto the piece that was "trying to get rid of it", which was what I didn't originally understand.

Great video though! Thanks!

D&Darkrai · May 29, 2019 at 12:00 am

Love towers of hanoi, and there many patterns. Youll move a peice on spot over twice then two spots once then 1 spot twice etc. you also move discs in order: 1 2 1 3 1 2 1 4 1 2 1 3 1 2 1 5 1 2 1 3 1 2 1 4 1 2 1 etc. lastly if u notate your moves as ring moved in first place then left or right then how many spots (5R1 would be fifth ring right one slot) your full notation for an n solve will always be palindromic

f96 · June 2, 2019 at 11:47 am

call them tets lmao

Dobeye memes and death · June 10, 2019 at 7:45 pm

abacabadabacaba?

david ortega · June 23, 2019 at 6:45 pm

this guy is the same from majorprep?

Cheydinal · July 2, 2019 at 1:01 pm

In binary, "base 2" is written as "base 10"

Random Dude · July 16, 2019 at 3:03 pm

I had to slow down that procedure part at 6:18 just to understand how it works. And it's mind blowing!

Gregory Fenn · July 28, 2019 at 9:58 am

The audio from your guest speaker is really poor quality, it actually bored me because it reminded me of those terrible VHS science videos your teacher made you watch on rainy days (the content of what he says is fascinating, but the poor audio triggers some kind of Pavlovian reflex that makes me want to think about anything else).

Adranir Doradrië · July 29, 2019 at 8:37 pm

Now this game isn't funny anymore because I don't have to think to solve it, thanks, I hate it x)

Eric Jin · August 2, 2019 at 12:52 pm

Ohhhhhhhhhhh python XD
BTW, your function at 10:00 will break down at 1000, unless you “sys.setrecursionlimit(2147483647)''

Syaoran Code · August 3, 2019 at 6:13 am

that's so cool, now I can solve it even many disks

Mgtow Values · August 4, 2019 at 8:54 pm

The problem with all teleological explanations is one must arrive and look back and then say hey "we were always headed towards here". Occasionally, people do this in their minds.

John Brocksman · August 9, 2019 at 4:51 pm

Fuck I spent the last 30min using 6 blocks and I just realized I only needed 5, I solved 5 in like 10 minutes fml

Nisha George · August 9, 2019 at 6:53 pm

Link to the music, please?

Wa55abi · August 10, 2019 at 11:37 am

The colours of the binary digits are in opposite direction to the disc colours 🙂 Never the less, great video!

Kaczankuku · August 11, 2019 at 2:39 pm

You deal with Sierpiński triangle a long time so you finally should learn how to correct pronounce his name. I hear still the same your anglosaxon "r" and you miss the initial sound so called in Polish soft "s".
https://en.wikipedia.org/wiki/Wac%C5%82aw_Sierpi%C5%84ski – check this out.

Proxy · August 14, 2019 at 11:33 am

4:10 but why reist? it is 10… just in base2.
the defintion of what 10 is is based on the base used, but it can still be called ten and be correct. so a 10 in octal is completely different than a 10 in hexadecimal for example. but they are still called the same.

Rask R · August 15, 2019 at 6:51 pm

Is this even considered a puzzle? You just need to do first legal move available without repeating positions and that's it. No thinking at all

Tyom Sahakyan · August 16, 2019 at 12:36 pm

do the same with 3 discs 🙅🏻‍♂️

TinyFoxTom · August 19, 2019 at 3:57 pm

I came across this pattern frequently (no pun intended) while creating a computer in Minecraft. When laying down wire for 2^(n+1) addresses, the 1's place in the address looks like a sine wave of frequency f, the 2's place has a frequency of f/2, and the 2^n's place has a frequency of 2^-n*f. I never thought to apply it to the Towers of Hanoi, though. It also shows up in the rules of cellular automata. Minecraft is technically a cellular automata, but the address pattern wasn't dependent on that fact.

Anthony Nguyen · August 20, 2019 at 1:48 pm

realizes I could solve tower of Hanoi when I was younger and realizes I might not be a genius

Aayush Singh · August 27, 2019 at 12:55 pm

Towers of Hanoi!

bauagan · August 31, 2019 at 12:28 am

If I didn't grow up with so called western world numbers, I'd find this video severely informative ;D I love this planet ;D

Nguyen Chester · August 31, 2019 at 5:50 am

what does this have to do with Hanoi?

Antonio Sraffa · September 2, 2019 at 7:52 pm

Visual example of backwards induction, its ties to mathematical induction, and showing off the use for binary to manipulate two-state models…and none of that was even center-stage in this (or the next) video.

We are not worthy!!!!

E030E03 · September 6, 2019 at 9:19 am

τ⚔π

Ben D. Manevitz · September 6, 2019 at 4:48 pm

Minor quibble – you should be more clear that the rule for flipping the 2^0 bit is different than all the other bits. The 2^0 bit moves — naming the posts 0-1-2 left-to-right — in a (+1 mod 3) pattern, while the other bits go to "whichever one they can without breaking the other rules of the game."

The rule itself is necessary because unlike the other bits the 2^0 bit has choices. For instance, when you're going from 011 to 100, the 2^0 bit could go on top of either post 0 or 2. Putting it on post 2 is correct move, but putting it on post 0 doesn't violate any rule of the game.

The clarification may not be necessary, per se. But I missed it when I was watching the video and I'm feeling pedantic.

Ben D. Manevitz · September 6, 2019 at 4:50 pm

Back (many years ago) when I was learning early coding, the Towers-of-Hanoi puzzle was the way I was taught recursion. You solve for N by solving for (N-1) and then adding one move, and then solving (N-1) again. And so on…

Sumit Aggarwal · September 7, 2019 at 7:51 am

is it only applicable for 3 pegs?

Alex Loomis · September 10, 2019 at 4:43 am

Woooaaahhh fine structure constant easter egg

Jaden Max · September 19, 2019 at 12:01 pm

informative

Don't Sub Me · September 22, 2019 at 5:40 pm

This puzzle is extremely easy equalent to finding the guarenteed winning/draw pattern in tic tac toe

Victor Barroso · October 11, 2019 at 7:10 pm

Once you see disk 3 sending you a kiss, you cannot unsee it

TimmyLP · October 15, 2019 at 6:47 pm

Wow, just solved a programming task for the Towers of Hanoi, and remembered that this video existed. Thanks for saving the day and blowing my mind.

TheSpaceMan88 p · October 15, 2019 at 6:57 pm

you've done a little error at 5:59 11111111 in binary equals to 256 not 255 (which is 11111110 in binary) cause 2**8 = 256 otherwise realy nice video keep going like this (sorry for grammar error )

nj foever niso · October 22, 2019 at 5:26 am

lo

nj foever niso · October 22, 2019 at 5:32 am

Pop😛😚🤣😭🙋💆🤸🤹🤼🏄🏂⛷️🏇🎅🤶🧚🦸🧟👰👸🤴🤶

Jowat · October 22, 2019 at 11:26 am

OK… I'm sorry for seeing this video this late but…

the first thing I asked myself is "where do you move it" and actually
if it's a even disk, move it to the right
if it's a odd disk, move it to the left

I had to comment my observation…

The MinecraftR 9598 · October 31, 2019 at 5:14 am

7:14 Are you certain it’s always the most efficient? Moving the 3 seems like an extra step.

ALEXANDER BRONAUGH · November 13, 2019 at 12:33 am

24

Nick van Schoonhoven · November 23, 2019 at 12:51 am

Is there an algorithm that would tell us how many fewer moves would be required if we were to add additional pegs?

Justin Moore · December 2, 2019 at 1:50 pm

Time to go back to doing what I was programmed to do. That's to bust out of constructs.

torix · December 13, 2019 at 2:32 am

why does keith sound identical to zillion ross

Tom Kanazaki · December 21, 2019 at 2:33 am

Vietnamese Hanoi puzzle?

Griffin Shue · January 1, 2020 at 1:46 pm

dank as always 😉

J M Roblox · January 4, 2020 at 4:23 am

There are 10 Types Of People In The World:
People That Understand Binary,
People who don't,
People that thought this joke is Binary

And

People that thought this joke is ternary.

김재원 · January 12, 2020 at 12:32 pm

And you just became my favorite youtuber

Zachary Allen · January 18, 2020 at 4:02 pm

Your videos captured my attention so well in your calculus lectures, that I finally dispelled the notion that I was not a math person, and simply needed to try harder. Finally going back to school, and no longer avoiding math intensive fields! Seriously though, this video is so cool, my girlfriend generally isn’t interested in this kind of stuff, but she loved this as well. Thank you, for working hard to put this stuff out.

GG Entertainment · January 22, 2020 at 11:57 am

does it also work with more than 3 pegs?

los zdog · February 17, 2020 at 5:24 pm

There are 10 types of people: those who understand binary, those who don’t, and those who weren’t expecting a ternary joke

Brian Clarke · February 22, 2020 at 3:11 am

Rare natural occurances. Makes me think about the creation of the universe

Jared Oz · February 25, 2020 at 1:33 pm

This is EXACTLY why I'm not teaching anymore. I couldn't explain it better. Thank you.

Leave a Reply

Your email address will not be published. Required fields are marked *