Thanks for the feature! I thought of the title because it seemed like a miracle that it solved and I had just watched Encanto. So fun to see Simon working through one of my own puzzles for once!
So that’s what something that’s been spit straight from the depths of hell looks like… i hope for the people that have crossed you that you’re not a petty person haha congrats on setting this
Simon getting 85% of the way through the puzzle via magnificent logic and failing to complete the last 15% in 5 seconds with basic scanning is the most Simon thing imaginable 😆
Thank God. I thought i was the only one having that feeling. I even wrote a comment cause it rly triggered me somehow... min 15: Oh that was fast... WHAT THE VIDEO IS 1:04:24!?!? :D
48:50 for all the people that want to enjoy this epic sentence again. In that moment. this sentence seemed completely fine, but reading it on it's own is simply fantastic.
Yeah, i was disappointed it wasn't framed this way. "Cells joined by black dot are in a 1:2 ratio, circle are consecutive, by a V sum to 5, by an X sum to 10. All black dots, circles, V's and X's are given." Simon would have freaked out for 5 minutes worrying that the puzzle was set incorrectly.
I miss the days when "finned swordfish" was a strange phrase. under any other circumstance, it would be safe to assume a swordfish does in fact have fins haha.
when he was first trying to tigger bounce the 2's that tripped him up really quickly. He was so close to saving like 30 minutes of trying to prove the placements of the 2'ss. if he just put the 2's one box over that would have solved it.
Simon's refusal to do Sudoku in a puzzle always equally amuses and amazes me, because he's literally so damn clever, that he'll find a complicated workaround. Great Puzzle, and a great solve.
Suggestion for something to add to the Antithesis Battles section: "Has to be" vs "Can't be" Might be problematic if you want to add in variations such as "Must be" and "Could never be"
I love that Simon has gone from, "Mark is having a joke at my expense" in The Miracle, to "this has got to be the most remarkable discovery in the history of sudoku!" upon seeing this puzzle! The setters have shown that almost anything is possible!
That was pretty much the response to the original miracle too. He started wby saying that it can't be possible and was going to be angry at Mark for setting him up, and ended by enthusing over the puzzle.
@@richardfarrer5616 What the OG commenter meant (i assume) is, that Simon completely skipped the "Mark set me up to have a laugh"-phase and immediately assumed, that this puzzle is possible, thereby highlighting, that setters have gotten so incredibly good, that almost anything seems possible.
This was the most brutal sudoku I ever solved without peeking at the solution in the video. Took more than 3 hours. But now I feel like a genius. Congrats to fafrd for setting, and to Simon for solving this!
I'm proud of myself of actually solving this in way less time than Simon did (while I usually need way more) 😄. I needed just under 30 minutes. He wasted SO much time finding out the pattern of the 2's, but his mistake in the beginning was putting one in the corner of the 9x9 grid when trying to find the pattern. There are only two possible patterns (one and its mirror), and only one fits when you take the cage into account. The rest naturally flows from there.
Quick question, did it show you that you solved correctly? I just finished the puzzle and it told me I made a mistake but when I look at the video my solution is 100% the same. Was looking for someone that solved it as well and not sure if they read 800+ comments per video
@@oisyn- same for me, took me longer to figure out how a single box should look compared to Simon. But after that I had it done in 10 minutes. Visualising how the grids have to intersect came to me pretty fast as soon as I realised they'd all need to be offset by one and how that perfectly fits in the 11x11 grid and then it was just a very easy sudoku that quickly became clear it was a few copy pastes between grids.
@@simonb.6281 yeah it's bugged, if you hit f12 you can see the 'errors' in the console. And for some reason the website thinks that there should be 1 + 1 = 3 inside the cage and not 7 + 4 = 11
Hi, it’s about the efficient use of the 11x11 space, only 40 unused cells allowed - “simplifies” into an efficient stacking problem. The basic pattern rises to the left, so shift it as far left as possible to minimise the waste on that side and build towards the right. Any other approach uses up the 40 too quickly, looks like the 9 boxes don’t fit but it really is because you have no useable spaces left.
Yes, I remember a puzzle from Pjotr V. using the same constraints, which was featured on the channel and I think there also was a setter's video from Peter, if I'm not mistaken.
It took me until almost the end of the video to unpack what you getting at...Now that I get it that would have been genius...Watching Simon's head explode over that would have been awesome...
I've been following this channel for a long time and when I saw this Sudoku I thought "I have to try it!" and I think I've just solved this magnificent puzzle, I say "I think" because if I'm not mistaken, these Sudoku puzzles that do not have the standard 9x9 cell format are not recognized correctly by the program when checking the result but going through all the boxes I have not detected no error (Now when I see the video I will check it). It has taken most of my quiet night shift at work but it has been one of the most rewarding experiences of my life. I don't consider myself particularly good at solving Sudoku puzzles but this channel has taught me so many techniques and tricks that I've finally been daring with the more "affordable" problems and I think this is my third or fourth Sudoku puzzle solved. I want to thank this channel for the content it makes and the way they teach us. You help people not only to think but to value themselves a little more and not give up in the face of obstacles.
Mine also told me I was wrong! Heartbreaking after spending so many hours on it… sounds like you had more integrity that I did though, I just turned on conflict detection and skipped to the end of the video
I struggle to make myself watch movies anymore because my brain sees an hour and a half to two hours as being too much of a time investment, but I can somehow easily watch Simon do nothing but make a pretty pattern using logic for 50 minutes only getting the 2's in. And I will also likely go on to watch several more Cracking the Cryptic videos afterwards. Splendid.
This would be a funny joke for a future fog of war puzzle. The puzzle is initially all fogged out, but as the fog clears there are no further clues in the entire grid. Extra points for a long complicated ruleset.
@@nonyobisniss7928 Seventeen setters are setting to work on this one right now, and Mark is preparing to send one to Simon with the usual "open this live" instrux....
I started laughing as soon as Simon opened the puzzle as I knew exactly what how this would play out. I played around quite a bit with anti-XV and non-consecutive minimalistic puzzle ideas a couple years ago, and trying to combine different sets of restrictive rules to make a functional sudoku that ideally devolves to just a set of 8 solutions (before any clues) that could be resolved into a unique solution with a single clue acting as a orientation resolver by breaking symmetry. My experience was just a lot of frustration where solutions either had too many solutions or 0 solutions. I never considered the idea of adding a deconstructed grid as a way of making this work. I'd imagine Fafrd arrived at this puzzle in a similar way, recognizing how constrained a box can be made with the combined restrictions, where anti-Kropki brings down a boxes 3x3 options to just one solution (plus rotations and reflections), and found the brilliant solution of using a deconstructed Sudoku to create a puzzle that could have a valid solution.
once he had a complete box and the 2s in place, my brain just started rotating (open and filled in the puzzle) and I sat and watched the last 6 minutes of his solve fascinated that he didn't see it :P
Everytime you feel you've seen everything possible in a sudoku a brilliant constructor comes up with something totally mind blowing.. Simon's total ecstasy at the end of this was just priceless
I was gashing my teeth late in, when you had 3 3x3s left to solve BUT had one "wing" of the "propeller" entirely filled in and STILL kept trying to LOGIC the rest out... you built the propeller because it TELLS YOU what the rest of the 3x3 MUST be! You had this effectively entirely solved more than a minute before you "solved" the puzzle! This was a magical puzzle and a fantastic experience to watch.
Ngl, I don't watch much puzzle content. But, the way this madlad solves puzzles with joy, wonder, and ruthless logic is compelling. Thank you for the dynamite videos!
I absolutely love how, after an incredible, hour-long session of deduction and pattern recognition, Simon was completely stunned to find sudoku disambiguating the end! Absolutely fantastic solve, and for a rarity, perhaps an even better puzzle! Immaculate!
For anyone that tries to solve it themselves, the site appears to be bugged at the moment and won't register that this is the right answer. You can copy it number-by-number and it will think something is wrong. Extra notes below on solving it faster. Solving can actually go a lot faster when you consider the 11 cage. Since it must contain two numbers, that means it must be in one of the boxes; the cage touches the bottom of the grid, so we know that its box runs along the bottom. And because we know that 2 is at the centre of each box, that means that the 2 in that box can only be in one of three spots: the top of the 11 cage and the two cells beside it. By narrowing down the two locations at 36:00 and 40:00, it defaults to the space on the left without needing to check the rest. As for the contents of the 11 cage, there are only two possibilities. 5 and 6 can't be adjacent and 3 and 8 are separated in the box pattern, leaving only 4 and 7 or 2 and 9 as the combo. Since we've already established that the 2 can't go in the 11 cage, that means it must be 7 on top and 4 on the bottom (flipping them would push 1, 6, and 8 off the grid). From there, the 2 spots solve themselves and the box placements rotates 180 degrees every time it moves clockwise to the next box.
48:58 “These twos are all far enough apart that we don’t run into prehistoric wheel trouble.” That’s a sentence that has never been said before. 😂 I love Simon. I could watch these videos all day.
Loved it! Pencil and paper attack shows 2 must be in middle etc.. Then it becomes an efficient stacking problem, 121 cells available, 81 get used, so only 40 spaces, don’t use them up too quickly. Three box pattern rises to the left, so shift it as far left as possible to minimise the waste on that side and expand to the right. (Anything else uses up the spaces too quickly, looks like boxes don’t fit but cause is you running out of useable spaces.)
Simon, you are so funny. When you said (1:03:20) that "it's perfectly doable, once you start to think about it..." I laughed out loud. You forgot to mention the part about needing to be a genius to begin with, and that years and years of experience wouldn't hurt when doing a puzzle like this. Thanks for the chuckle. And very well done, Fafrd!
The 2 is always at the box's center, and is impervious to rotation and reflection. Therefore, a 2 needs to be in every column and row, barring the edges. We must also maintain the separation needed for the boxes to non-overlap. 27:30 I think that the box has eight possible permutations, which means that one has to be repeated. There should be something unique about the two positions. The 4 can go into any of the four corners, and the 79 pair can resolve on either side of the 4. 36:50 The 3x3 box is correct. The 2 just turns out to be in the least intuitively obvious cell. 53:00 Fafrd has made Simon work his tail off to just begin solving the puzzle. The rest of the puzzle will take ten or eleven minutes.
This is the craziest puzzle you've featured that I've managed to solve! Super fun. Pretty darn please with myself and a very satisfying piece of sudoku theory crafting
I haven't played Sudoku in maybe years, and I absolutely loved this video. Seeing you chug through what initially seems to be impossible, ending up at such a beautifully elegant solution. Also I hope Fafrd sees this video, I can't imagine any higher praise for a puzzle maker than this video.
Solved it in just 9:28. Very nice puzzle. Few minutes to figure out unique arrangement of digits due to orthogonality constraints. Few minutes to work out unique arrangement of boxes due to centers. The cage positioned one of the boxes. Then, from the box arrangement the 4 corners + center box must have have the same orientation and the 4 edge boxes must have the same orientation. Then, just a matter of filling in boxes.
When I was solving this I totally forgot to check for all the negative constraints after deducing the pattern of each 3x3 box, and it actually led to an interesting bit of symmetry. You can show that if each of the 3x3s boxes inserted into an 11x11 grid are rotations/reflections of the same digit pattern, then: 1. All the 3x3 boxes are laid out in the same rotationally symmetric pattern as here (although it can be reflected as well). 2. All four 3x3 boxes touching the edge of the grid and the center 3x3 must be identical. 3. All four of the remaining 3x3 boxes must be identical as well. So there really aren't many degrees of freedom here! You could make a very similar puzzle where the rules are that each 3x3 box must be a magic square (because again that locks you into a single digit pattern in each 3x3 which can be rotated/reflected), with a small number of simple clues scattered about to deal with the degrees of freedom mentioned in 1/2/3 above.
The most amazing thing about this to me is that of the 9 sudoku squares, 5 of them share the same pattern for corners and center, and the orthogonal positions from center are the same pattern, but rotated 180 degrees. There is no mirroring at all which is amazing!
Finally got a faster solve than Simon, by about 4 minutes! The 'propeller' lines that Simon drew for the final stage actually provide a way to visually solve the position of the 9 squares. If you imagine you have a box that is 11x11 and you have 9 (unnumbered) 3x3 squares in it each with a propeller drawn on them, then the puzzle is to arrange them so that none of the propellers line up. If you start with all nine squares together (in a normal sudoku grid) in the bottom left of your box then it very quickly becomes clear that you can remove all the vertical alignment by shifting the middle row of squares to the right by one, and the top row to the right by two. If you now repeat this with the columns, the only moves you can make are left column up 2, middle column up 1, and you get the pattern above. The only other possible sequences of shifting rows and columns (that don't get 'stuck') either produce the same pattern, or its mirror image, which is invalidated by the 11-cage.
Fafrd, this is simply amazing. This is the most awe inspiring puzzles I've seen on the channel. Well done my friend, well done. Again... it's simply amazing!
I finished in 121:08 minutes. This was absolutely an incredible sudoku. The fact that this is even possible is a testament to the brilliance of the constructor Fafrd. This truly was an incredible discovery. It took me a while to spot that 4 was the most limited and being that limited, forces it to always create the same pattern with 2 in the middle. It took me even longer to try to fit all the 2's in the grid. I don't know why it took me so long to see that the most efficient wasn't a knight's move away, but rather a super knight's move, which is three moves and a turn. That solidified it for me and the rest was a cake walk. I very much enjoyed this puzzle and I still can't believe this is possible. Congratulations, Fafrd. This was a thing of beauty. Great Puzzle!
As always, your enthusiasm, Simon, is entertaining and inspiring. Thanks so much for opening Mark's email and attempting this puzzle, as bonkers as it seemed at first. I loved this video.
32:35 finish. Whenever there is an 11x11 puzzle with the "build your own non-overlapping 3x3s" constraint, I always start them in the same way. The nine cells at the intersections of rows and columns 3/6/9 each have to be in a 3x3, and moreover each one must be in a 3x3. Given that both cells of the 11 cage have to be in a 3x3, you know that the bottom right cage is in rows 9-11, with the center 2 in row 10. A quick test of r10c8-10 shows that r10c9 and r10c10 will break, giving you your first 2 in r10c8. The distance restrictions will allow you to place the rest of the 2s, and work from there. Excellent puzzle! And look, Simon! EVERY box has a 3 in the corner!
I was so lucky today! Once I figured out that there was essentially only one possible way to fill a 3x3 box observing the constraints, modulo rotations and reflections, the puzzle reduced to finding the placements of the central 2s. That screamed *modulo 3* to me, since the column/row rule of sudoku applies, the rows *and* columns of the 2s must be staggered. Efficient placement of the 2s almost by definition makes symmetry inevitable. Hence the 2s form a square but a skewed square to ensure the mod 3 condition. Rows 1,2,3 go with one residu class mod 3, rows 4.5.6 with a second residu class mod 3, and rows 7,8,9 with the remaining residu class mod 3. The 11 box disambiguates the situation. The rest is standard sudoku with simple constraints. I was able to finish in 33:02 but I used pen and paper to do the maths. It helped hugely that I recently wrote up a proof of a theorem in combinatorics that involved mod 5 classes and their arrangement which had the same flavour. 99.9% of the time I am totally baffled by the high-end puzzles on CtC but today I was very lucky..
That was such a joy to watch!! I've never seen Simon get so excited when he solved a puzzle! And what a puzzle! The finished grid is just stunning! By far the best solve/reaction I've seen Simon do!
Yeah, that‘s what I noticed too. Mind you I failed miserably at deducing the pattern of the cage. However if you check the pattern you notice that the cage can only be 47 or 29 and with that you can basically bruteforced the approximate position of the 2 in the bottom right and continue on simons solving path
@@RichSmith77 yeah, so you can rule out the rest of that row and then basically move on with what simon did by eliminating the possibilities one by one.
@@RichSmith77 Exactly (not that I would have seen this on my own, mind you). So there were in fact fewer possibilities to check: Simon could have saved himself, oh I dunno, maybe 2 whole minutes on that.
@Richard Smith only the two positions listed above. Because the 11 cage is on the edge you have to take Simon's original grid and flip it to find a pair that will fit...either the 47 or the 29 pair. And with that we know the order. If the 2 is in the cage so is the 11, same for the 4 and 7. So either the 2 is in the cage or the 7 is...2 positions.
This puzzle is mad! There is so much effort upfront just to figure out what is going on. But as soon as that's done, the filling of the puzzle itself becomes incredibly simple! Very fun.
20 minutes for me! This is a great puzzle and I really enjoyed it. I complete almost all the puzzles on this channel (eventually) and habitually take 2x, 3x or 4x what Simon or Mark take. This is a very very rare exception where I was much faster. I started by drawing a graph of the digits with pencil and paper (OK Simon couldn't reasonably do that on video). Then join all the allowable pairs. It's a very sparse graph and by referring to it, it's easy to quickly (a) make a valid 3x3 box and (b) rule out any other possible 3x3 boxes apart from rotations/reflections. Now you have to put 9 3x3 boxes in an 11x11 grid with none sharing a row or column. That sort of "staggered" pattern comes immediately to mind. Perhaps I've seen a similar puzzle before or can just see intuitively that it works and is the only way (apart from its reflection). Now it's clear that the centre of one of the boxes must go just left of the top half of the cage. The rest is just slamming all the digits in as quick as possible. I don't often watch a whole video here (spending too long solving them!). I usually just watch the intro and explanation and sometimes the end to check. But I had to watch this one to see what took so long! Feeling smug now! If only I could consistently solve them anywhere near as quick as Mark or Simon!
He lost a bunch of time not realizing that the first 2 only had three possible positions to rule out. He wasn’t simply concerned he was getting trolled he thought he messed up something with his initial assessment that the two is in the center. Sometimes being super clever is a disadvantage.
56:30 There is an overlooked conclusion there. Because of the 9 in the bottom centre square of the top right box, the top centre square of the middle box must be a 6. This expedites the puzzle by 5-6 minutes, all other things being equal.
This felt like a pencil puzzle that Simon created a sudoku grid for in order to solve the pencil puzzle which then gave the real sudoku grid for Simon to finally solve. Very pleasant evolution of the puzzle
Wow this is a pure gem ! For the first time I managed to solve the puzzle without seeing any of the video, and I still watched it for the pleasure of seeing the deduction process ! Thank you Fafrd and thank you Simon for this wonderful moment !!
I've never watched this channel before, but watched this one straight through, and so pleased I did. This was a lovely puzzle, and I really appreciate the energy of the video.
I solved this one together with my sister and it took us roughly 40 minutes. This didn't feel very sudokuey, but was very fun as a kind of more general logic puzzle. I can't even imagine how you come up with this.
I'm convinced that he intentionally skips over basic scanning to figure out the hardest way possible of doing a puzzle so that the videos are longer for monetization... He could have used sudoku basic from the very first box once he gets the windmills. That being said. This is a truly beautiful puzzle standing ovation for Fafrd.
Simon, at 0:58:25, you had actually solved the puzzle by saying it was identical, but you didn’t realize the scope of that statement. You diligently still did it square by square! After you had just said it was identical, the bottom left corner became obvious and then the entire left side. Great job!
What's absolutely beautiful to see is also the blocks repeat themselves. It's a giant X with the center block that is identical, and the four side blocks are identical. Genius of the HIGHEST order. I literally rewatched the whole video just to follow the logic again. What sublime beauty.
I wasn't following this channel for a longer time (I had to concentrate on my life, for short). But this got me hooked again. It's absolutely astonishing how you got to the solution. The last steps were easier than you made them, really just pure sudoku, but to get there at all, I would have given up at several milestones ahead already. But Simon (the sorcerer), of course, found them all in the right order in which you needed them to step forward. Wow, and 1 hour for it while also explaining this all to us, that's marvelous.
Honestly watching an expert solve puzzles like this - that are far beyond my abilities - is the same rush as watching sport at its best. The hour I spent watching this video absolutely flew by. What an elegant puzzle.
I came at this slightly differently as regards the 2s, when I really thought about it. If the 9x9 square is constrained for positioning of 2s, this immediately suggests some sort of symmetrical pattern. It being a 9x9 square, you can repeat the symmetry for all but 1 square, and there will be a symmetry of some form around a value of a multiple of 2. Then, the 9th 2 must be placed in a place where it is the symmetrical counterpart to itself, and so, one 2 should be right in the middle of the grid. This results in 4 identical shapes being made, and in no shape can you actually place 3 2s. This means the pattern on 2s must be such that all 2s form a rotationally symmetric pattern. Given the 11 cage, there are only a few places to form this symmetrical pattern, given the only 11 totals in the sudoku boxes at the end are 29 and 47. The only way to make the pattern work for rotational symmetry are to use the extremes of either two central rows or two central columns in each of the shapes left from the central 2 and the 2 in the bottom right sits on the very bottom of the 9x9 square.
Yes. This sounds similar to my approach. Although I did waste some time, like Simon, trying to put the bottom right box (with the cafe) as far into the corner as it could go. (I think there was a uniqueness argument that could be used to rule out 2-9 in the 11 cage, since I couldn't see how you could then prevent two solutions by flipping each box around its vertical axis. But I could be wrong about that.) I eventually realised the boxes would most likely be arranged symmetrically, although I'm not sure I had proof for it, or it was just intuition.
Amazing puzzle, amazing solve with such an amazing community day in and day out!!! Will Shortz legendary indeed. Fabulous interview with him. Used to also see his puzzles in Games Magazines in mid/late 80s to early 90s. Genuis he is.
I actually made a mistake while solving this (I assumed that 2-9 was the only pair in the solved square that added to 11) but it ended up being worth it. The puzzle ends up breaking with the impossibility of arranging the regions, but by that time I had worked out the required region pattern and it ended up being easy once I placed the 11 correctly
I decided to have a go with this before watching Simon solve it. It took me about 90 minutes; the first 80 minutes or so was figuring out the only valid 3x3 and then the grid layout; the rest came together rather quickly. I was afraid I might have made a mistake somewhere because the web site doesn't validate it, but skipping ahead I see I thankfully reached the same conclusion as Simon. A beautiful puzzle and incredibly satisfying to solve. Now to watch Simon and see if his logic was the same as mine. :)
26:18 I immediately thought of the greatest video in this channel that probably got most of us addicted to it, the fabulous OG Miracle Sudoku. Lots of tigger bounces in that one!
Anybody else finish the puzzle (which I was SHOCKED with!) and get a “Bobbins! That doesn’t look quite right” from the checker?! So anticlimactic to go through that whole puzzle only to be told I was wrong and then come back here and see Simon get the exact same grid but have it check cleanly
Did he have it check cleanly? I can't find it in the video. [edit: ah it's at 1:01:51] And YES I encountered that same issue. Although it says "puzzle doesn't contain solution". Also weird how it doesn't check for conflicts, I went over the entire grid to see maybe I made a mistake somewhere 🙄
This was amazing. I don't understand why youtube only recommended this channel now. Finally something really good. I know I'll be binging this channel now...
I really wish I could like a video multiple times. This has truly been something else. Purely based on logic, simple rules and a miracle solution. This might just be my favourite puzzle Simon has ever solved!
Less than a minute in, and I'm clicking "Like". I know I am in for a treat éspecial! This solve is going to be epic - I laughed out loud at seeing the puzzle reveal. What a construction. Stupendous. I'm settling in with my cup of tea and biscuit.
Absolutely brilliant! Took me 1:17hr to solve. Amazing that it's actually got a unique solution. Hat's off to Fafrd for the mind-bendingly constraining ruleset!
It's always nice to see how you solve these puzzles. I went about this with a much more brute force oriented process of elimination way and eventually came to the same result but checking your solution and seeing how well you're able to create visuals to systematically break something like this down was really nice. Thanks for introducing me to this beautiful puzzle.
Hi Simon, I don't know if you've ever heard of point groups (and their crystollagraphic extrapolation, space groups) but give the role of reflections and rotations in puzzles like this one, you'd qutie possibly find the subject interesting.
30:04 what im thinking is since this is a repeating pattern, no matter how you slice it, you will end up with the same configuration of 3x3's with at least one of them in a corner. so the real goal I think of the 11 cage, is to give you one box with a starting orientation which you can then use to propagate the rest of the boxes.
That's what I did. First thing I saw was that the centre of each box had to be a two. Second thing I saw was that the two in box 9 had to be in row 10, and that each row in each box had to contain the same digits. From there I decided that the only way that was possible was if the middle digit was a two. Solve path was entirely different from what Simon did but the end result was the same.
If the box with the 2 in the middle is the only possible 3x3 placement, does that not immediately dictate the 2 goes either in the top box of the 11 killer, or 1 to the left or right? Of course that 2 could be not the only possibility, but I cannot think of any reason why it would allow the 2 to be anywhere else but the bottom row of the pseudo 9th box the 11 cage is in, if that is the only 3x3 pattern allowed. (Currently 42 minutes into the video, just breaking my mind over why one would need to test any of the other cells of the pseudo box 9 but the bottom line)
Awesome! I really liked the logic for this puzzle. I've been working on something similar, using negative constraints, and the way this one breaks down is cool
Loved this puzzle! I think there's a bug in the web app though; I got the same solution as Simon (only after watching the first half of the video, I hasten to add), but when I validated I got the dreaded "bobbins" error message. Looking at the browser console, the validator found three errors, all of which seem false. This one is the most obviously wrong: "Error in cage sum in cage r10c9-r11c9. Expected "3" and found "11"" 11 in r10c9-r11-c9 is literally the only instruction we start with, even if all my solving was wrong that one has to be correct.
Same- loved the puzzle, but got frustrated when my solution got "Bobbins!" Checked that it matched with this video solution and got even more frustrated....
1:03:00 Not only that but the rules aren't newly-invented rules carefully crafted to ensure uniqueness. The non-consecutive, no 1:2, no 5 and no 10 rules are already familiar to variant sudoku solvers.
That is an incredible discovery - Fafrd, take not just a bow but a sheaf of arrows to go with it 😆 I feel like I lucked into the disposition of the boxes, but if you've figured out that the 2s have to be in separate rows and columns then it is relatively easy to visualise that they must be in some kind of stepped arrangement, where box 5 is centred in the grid and the ones above, below, left and right are offset by 1, and the corner boxes are offset by another 1 in each direction.
How fun that you're friends with Will Shortz! I don't know him, but I feel as though I do because I listened to him on the radio for years and years. Nice to know that you know each other, my two friends I don't know.
![This Sudoku Is Astonishing [MUST WATCH]](http://i.ytimg.com/vi/W3h5_lkqnTU/mqdefault.jpg)
Thanks for the feature! I thought of the title because it seemed like a miracle that it solved and I had just watched Encanto. So fun to see Simon working through one of my own puzzles for once!
This is a remarkable puzzle. Thanks, and congratulations too!
Extremely elegant!
Genius and sublime from you! Encanto was great movie also. 🙂
So that’s what something that’s been spit straight from the depths of hell looks like… i hope for the people that have crossed you that you’re not a petty person haha congrats on setting this
@@davidrattner9 ditto on both! 😁
Simon getting 85% of the way through the puzzle via magnificent logic and failing to complete the last 15% in 5 seconds with basic scanning is the most Simon thing imaginable 😆
I swear, the moment I saw 3 corners and one side of the puzzle I was hopping he would see he could just copy and paste everything xD
Thank God. I thought i was the only one having that feeling. I even wrote a comment cause it rly triggered me somehow... min 15: Oh that was fast... WHAT THE VIDEO IS 1:04:24!?!? :D
I was thinking the same thing "man! Look at the over 5-2-7, 7-2-5 grids! They're all the same just copy paste!" Heheh
this perfectly reflects my feelings throughout the video
I think a lot of people in this thread are forgetting that reflections were allowed.
"These 2s are all far enough apart that we don't run into prehistoric-wheel-trouble" must be the craziest sentence ever uttered.
48:50 for all the people that want to enjoy this epic sentence again. In that moment. this sentence seemed completely fine, but reading it on it's own is simply fantastic.
We need shorts with random strange sentences from Simon
I've run into prehistoric wheel trouble in the past. Trust me, it's not something you want to do.
The moment Simon fully transformed into an Alice in wonderland character
I laughed out loud at that sentence.
The only missing thing for this puzzle was missed opportunity to define rules as "all possible kropki dots, X's and V's are shown"
oh god, it should have been called XV.II.MMXXIII, today's date, "XV." from the rules and "II" for the 2 that brakes the puzzle
🤣🤣
Yeah, i was disappointed it wasn't framed this way. "Cells joined by black dot are in a 1:2 ratio, circle are consecutive, by a V sum to 5, by an X sum to 10. All black dots, circles, V's and X's are given."
Simon would have freaked out for 5 minutes worrying that the puzzle was set incorrectly.
@@BenRHarshhahahah he'd have definitely threatened to call Mark to give him a piece of his mind
The only missing thing was a wrogn in the title lmao
Flying geese, square prehistoric michelin tyres, tigger bounces. The language of sudoku gets stranger with every passing day!
I miss the days when "finned swordfish" was a strange phrase. under any other circumstance, it would be safe to assume a swordfish does in fact have fins haha.
@@vidiot5533true hahaha
The best puzzles are the ones where Simon is absolutely CONVINCED the puzzle is broken. His joy at finding the way through after that is infectious.
when he was first trying to tigger bounce the 2's that tripped him up really quickly. He was so close to saving like 30 minutes of trying to prove the placements of the 2'ss. if he just put the 2's one box over that would have solved it.
Simon's refusal to do Sudoku in a puzzle always equally amuses and amazes me, because he's literally so damn clever, that he'll find a complicated workaround. Great Puzzle, and a great solve.
i was YELLING
@@acapellascience It's not a CTC video when you're not yelling at your screen.
@@acapellascience I’m so glad you’re here, I have several of your OG songs memorized from when I was back in middle school
THAT'S A 9 THAT'S A 9 THAT'S A 9
@@bingolingo6555 The 7 in the central propeller was locked in very early because of the conflicting 5 on the vertical... I was foaming at the mouth.
Rules: 01:09
Let's Get Cracking: 07:47
Simon's time: 54m06s
Puzzle Solved: 1:01:53
What about this video's Top Tier Simarkisms?!
Bobbins: 4x (05:03, 32:12, 52:22, 57:38)
Maverick: 2x (01:25, 01:27)
You Rotten Thing: 1x (58:47)
And how about this video's Simarkisms?!
Ah: 15x (07:50, 13:37, 18:06, 18:06, 18:06, 19:02, 19:04, 20:58, 35:26, 35:32, 39:29, 46:58, 50:45, 56:54, 58:35)
Hang On: 13x (00:59, 00:59, 15:15, 16:03, 24:02, 31:41, 33:48, 35:20, 43:20, 43:39, 43:39, 43:39, 1:01:04)
Obviously: 12x (11:41, 12:22, 14:53, 18:21, 24:09, 24:32, 24:32, 24:34, 37:10, 54:19, 57:31, 59:58)
Sorry: 4x (02:43, 20:58, 45:40, 54:01)
Wow: 4x (14:35, 25:15, 48:49, 48:52)
Lovely: 3x (06:21, 17:36, 59:16)
Beautiful: 3x (31:49, 1:03:07, 1:03:09)
Shouting: 3x (05:11, 06:37, 06:58)
In Fact: 3x (00:20, 28:55, 33:38)
Cake!: 3x (05:28, 06:06, 06:19)
What on Earth: 2x (20:01, 1:02:18)
Nonsense: 2x (02:09, 37:07)
Fascinating: 2x (41:04, 41:07)
Incredible: 2x (1:00:14, 1:00:16)
Ridiculous: 2x (01:48, 01:51)
Bonkers: 2x (1:02:27, 1:02:30)
Surely: 2x (34:07, 1:02:23)
Goodness: 1x (43:20)
Naughty: 1x (58:45)
Horrible Feeling: 1x (21:07)
Brilliant: 1x (05:23)
Break the Puzzle: 1x (45:33)
Going Mad: 1x (43:39)
Magnificent: 1x (49:15)
Unbelievable: 1x (1:02:15)
Whoopsie: 1x (50:45)
We Can Do Better Than That: 1x (47:36)
*****ing: 1x (12:40)
Symmetry: 1x (47:51)
Most popular number(>9), digit and colour this video:
Eleven (16 mentions)
Two (180 mentions)
Green (4 mentions)
Antithesis Battles:
Even (8) - Odd (0)
Black (2) - White (1)
Column (8) - Row (5)
FAQ:
Q1: You missed something!
A1: That could very well be the case! Human speech can be hard to understand for computers like me! Point out the ones that I missed and maybe I'll learn!
Q2: Can you do this for another channel?
A2: I've been thinking about that and wrote some code to make that possible. Let me know which channel you think would be a good fit!
Tigger bounce
12:40 "*****ing the grid" 😂
Michelin Tire (Most popular phrase) - though it morphs into "prehistoric tire"
Suggestion for something to add to the Antithesis Battles section: "Has to be" vs "Can't be"
Might be problematic if you want to add in variations such as "Must be" and "Could never be"
This is a good thing that exists
He actually forgot about the concept of horizontality for the last 10 minutes of the solve, which made it so much more fun.
@@oNtuobAwoH I was constantly pointing them out to him…”but there’s a 5 there!! Has to be a 3 because of the 8! !
I love that Simon has gone from, "Mark is having a joke at my expense" in The Miracle, to "this has got to be the most remarkable discovery in the history of sudoku!" upon seeing this puzzle! The setters have shown that almost anything is possible!
That was pretty much the response to the original miracle too. He started wby saying that it can't be possible and was going to be angry at Mark for setting him up, and ended by enthusing over the puzzle.
@@richardfarrer5616 What the OG commenter meant (i assume) is, that Simon completely skipped the "Mark set me up to have a laugh"-phase and immediately assumed, that this puzzle is possible, thereby highlighting, that setters have gotten so incredibly good, that almost anything seems possible.
The phrase "by prehistoric tire logic" is my new favourite thing in the whole entire world.
34:25 for those who wants to relive the moment of the creation of the name.
@@williamnathanael412Thanks, go ahead!
This is the happiest I've seen Simon after completing a puzzle. Well deserved, Simon. Amazing setting by Fafrd.
This was the most brutal sudoku I ever solved without peeking at the solution in the video. Took more than 3 hours. But now I feel like a genius. Congrats to fafrd for setting, and to Simon for solving this!
I'm proud of myself of actually solving this in way less time than Simon did (while I usually need way more) 😄. I needed just under 30 minutes. He wasted SO much time finding out the pattern of the 2's, but his mistake in the beginning was putting one in the corner of the 9x9 grid when trying to find the pattern. There are only two possible patterns (one and its mirror), and only one fits when you take the cage into account. The rest naturally flows from there.
Quick question, did it show you that you solved correctly? I just finished the puzzle and it told me I made a mistake but when I look at the video my solution is 100% the same. Was looking for someone that solved it as well and not sure if they read 800+ comments per video
@@simonb.6281 I just solved it myself and have the same problem. I triple checked it and then checked the solution and it’s perfectly right.
@@oisyn- same for me, took me longer to figure out how a single box should look compared to Simon. But after that I had it done in 10 minutes. Visualising how the grids have to intersect came to me pretty fast as soon as I realised they'd all need to be offset by one and how that perfectly fits in the 11x11 grid and then it was just a very easy sudoku that quickly became clear it was a few copy pastes between grids.
@@simonb.6281 yeah it's bugged, if you hit f12 you can see the 'errors' in the console. And for some reason the website thinks that there should be 1 + 1 = 3 inside the cage and not 7 + 4 = 11
This formation of digits was discovered before, but that only one cage was necessary for this 11×11 grid is fascinating
Hi, it’s about the efficient use of the 11x11 space, only 40 unused cells allowed - “simplifies” into an efficient stacking problem. The basic pattern rises to the left, so shift it as far left as possible to minimise the waste on that side and build towards the right. Any other approach uses up the 40 too quickly, looks like the 9 boxes don’t fit but it really is because you have no useable spaces left.
Yes, I remember that ruleset being featured in a single box of a sudoku once.
Yes, I remember a puzzle from Pjotr V. using the same constraints, which was featured on the channel and I think there also was a setter's video from Peter, if I'm not mistaken.
Simon always just casually forgetting the most basic rules of Sudoku. Love his methods but it cracks me up every time.
Would've been even funnier if the rules fully stated Kropki and XV rules and then said "all possible dots, Xs and Vs are given"
That's positively evil. I love it.
Missed opportunity
It took me until almost the end of the video to unpack what you getting at...Now that I get it that would have been genius...Watching Simon's head explode over that would have been awesome...
I've been following this channel for a long time and when I saw this Sudoku I thought "I have to try it!" and I think I've just solved this magnificent puzzle, I say "I think" because if I'm not mistaken, these Sudoku puzzles that do not have the standard 9x9 cell format are not recognized correctly by the program when checking the result but going through all the boxes I have not detected no error (Now when I see the video I will check it).
It has taken most of my quiet night shift at work but it has been one of the most rewarding experiences of my life. I don't consider myself particularly good at solving Sudoku puzzles but this channel has taught me so many techniques and tricks that I've finally been daring with the more "affordable" problems and I think this is my third or fourth Sudoku puzzle solved.
I want to thank this channel for the content it makes and the way they teach us.
You help people not only to think but to value themselves a little more and not give up in the face of obstacles.
This would be one of those kind comments that Simon particularly enjoys. Congratulations on the solve. 🙂
Mine also told me I was wrong! Heartbreaking after spending so many hours on it… sounds like you had more integrity that I did though, I just turned on conflict detection and skipped to the end of the video
Not only is this grid a miraculous discovery, the use of that domino cage to disambiguate is utterly masterful.
And on top of that, the clue is 11 in the 11x11 grid. Inconceivable.
Every other day Simon does a puzzle which is "The most amazing discovery in Sudoku"! What a great time to be alive and puzzling.
This is the 'Miracle Square' as first discovered by PjotrV - but I have never seen it used quite like this! Lovely.
I have also discovered it in the past. In another context than sudoku per se.
I remembered I'd seen this square before. Here's PjotrV's puzzle being solved on CtC,
ua-cam.com/video/jd8vUosXg0k/v-deo.html
Thank you for naming it! I had remembered that it had come up in a puzzle but I couldn't remember much more.
I struggle to make myself watch movies anymore because my brain sees an hour and a half to two hours as being too much of a time investment, but I can somehow easily watch Simon do nothing but make a pretty pattern using logic for 50 minutes only getting the 2's in. And I will also likely go on to watch several more Cracking the Cryptic videos afterwards. Splendid.
Of course it's sudoku after 1 hour that stops Simon from finishing the puzzle in an instant :'D what a great idea that puzzle is :)
Nice Fog of War puzzle. Wait. Wait, THERE'S NO FOG
This would be a funny joke for a future fog of war puzzle. The puzzle is initially all fogged out, but as the fog clears there are no further clues in the entire grid. Extra points for a long complicated ruleset.
And there's almost no point starting with the one visible clue.
Hilarious!
😆
@@nonyobisniss7928 Seventeen setters are setting to work on this one right now, and Mark is preparing to send one to Simon with the usual "open this live" instrux....
Simon's excitement at finishing this puzzle fuels me.
I never thought i would be such excited about an other person solving a sudoku and me watching him. Wow, love this channel.
I started laughing as soon as Simon opened the puzzle as I knew exactly what how this would play out. I played around quite a bit with anti-XV and non-consecutive minimalistic puzzle ideas a couple years ago, and trying to combine different sets of restrictive rules to make a functional sudoku that ideally devolves to just a set of 8 solutions (before any clues) that could be resolved into a unique solution with a single clue acting as a orientation resolver by breaking symmetry. My experience was just a lot of frustration where solutions either had too many solutions or 0 solutions. I never considered the idea of adding a deconstructed grid as a way of making this work. I'd imagine Fafrd arrived at this puzzle in a similar way, recognizing how constrained a box can be made with the combined restrictions, where anti-Kropki brings down a boxes 3x3 options to just one solution (plus rotations and reflections), and found the brilliant solution of using a deconstructed Sudoku to create a puzzle that could have a valid solution.
As soon as Simon got the 2's my brain went, 'There it is! It's the pattern! It's the slightly weirdly rotated square!'
once he had a complete box and the 2s in place, my brain just started rotating (open and filled in the puzzle) and I sat and watched the last 6 minutes of his solve fascinated that he didn't see it :P
Everytime you feel you've seen everything possible in a sudoku a brilliant constructor comes up with something totally mind blowing.. Simon's total ecstasy at the end of this was just priceless
I was gashing my teeth late in, when you had 3 3x3s left to solve BUT had one "wing" of the "propeller" entirely filled in and STILL kept trying to LOGIC the rest out... you built the propeller because it TELLS YOU what the rest of the 3x3 MUST be! You had this effectively entirely solved more than a minute before you "solved" the puzzle!
This was a magical puzzle and a fantastic experience to watch.
Ngl, I don't watch much puzzle content. But, the way this madlad solves puzzles with joy, wonder, and ruthless logic is compelling.
Thank you for the dynamite videos!
I always enjoy when a setter takes iconic variant sudoku roles and make a puzzle the relates surprisingly little to sudoku.
I absolutely love how, after an incredible, hour-long session of deduction and pattern recognition, Simon was completely stunned to find sudoku disambiguating the end! Absolutely fantastic solve, and for a rarity, perhaps an even better puzzle! Immaculate!
"These twos are all far enough apart that we don't run into prehistoric wheel trouble" may be the best sentence yet uttered on this channel.
For anyone that tries to solve it themselves, the site appears to be bugged at the moment and won't register that this is the right answer. You can copy it number-by-number and it will think something is wrong. Extra notes below on solving it faster.
Solving can actually go a lot faster when you consider the 11 cage. Since it must contain two numbers, that means it must be in one of the boxes; the cage touches the bottom of the grid, so we know that its box runs along the bottom. And because we know that 2 is at the centre of each box, that means that the 2 in that box can only be in one of three spots: the top of the 11 cage and the two cells beside it. By narrowing down the two locations at 36:00 and 40:00, it defaults to the space on the left without needing to check the rest.
As for the contents of the 11 cage, there are only two possibilities. 5 and 6 can't be adjacent and 3 and 8 are separated in the box pattern, leaving only 4 and 7 or 2 and 9 as the combo. Since we've already established that the 2 can't go in the 11 cage, that means it must be 7 on top and 4 on the bottom (flipping them would push 1, 6, and 8 off the grid). From there, the 2 spots solve themselves and the box placements rotates 180 degrees every time it moves clockwise to the next box.
48:58 “These twos are all far enough apart that we don’t run into prehistoric wheel trouble.”
That’s a sentence that has never been said before. 😂
I love Simon. I could watch these videos all day.
I love the symmetry in this puzzle, figure out 1 single box, rotate it once for box 2 and then just spread them around apart.
Loved it! Pencil and paper attack shows 2 must be in middle etc.. Then it becomes an efficient stacking problem, 121 cells available, 81 get used, so only 40 spaces, don’t use them up too quickly. Three box pattern rises to the left, so shift it as far left as possible to minimise the waste on that side and expand to the right. (Anything else uses up the spaces too quickly, looks like boxes don’t fit but cause is you running out of useable spaces.)
That “there’s not much in this grid” comment is golden
I proper belly laughed at that 😆
Simon, you are so funny. When you said (1:03:20) that "it's perfectly doable, once you start to think about it..." I laughed out loud. You forgot to mention the part about needing to be a genius to begin with, and that years and years of experience wouldn't hurt when doing a puzzle like this.
Thanks for the chuckle. And very well done, Fafrd!
The 2 is always at the box's center, and is impervious to rotation and reflection. Therefore, a 2 needs to be in every column and row, barring the edges. We must also maintain the separation needed for the boxes to non-overlap.
27:30 I think that the box has eight possible permutations, which means that one has to be repeated. There should be something unique about the two positions. The 4 can go into any of the four corners, and the 79 pair can resolve on either side of the 4.
36:50 The 3x3 box is correct. The 2 just turns out to be in the least intuitively obvious cell.
53:00 Fafrd has made Simon work his tail off to just begin solving the puzzle. The rest of the puzzle will take ten or eleven minutes.
This is the craziest puzzle you've featured that I've managed to solve! Super fun. Pretty darn please with myself and a very satisfying piece of sudoku theory crafting
My solution was way more involved than Simon's somehow lol
I can't believe I was glued to the screen for a full hour! Wonderful. Loved it. Thank you Fafrd and everyone.
I haven't played Sudoku in maybe years, and I absolutely loved this video. Seeing you chug through what initially seems to be impossible, ending up at such a beautifully elegant solution.
Also I hope Fafrd sees this video, I can't imagine any higher praise for a puzzle maker than this video.
Solved it in just 9:28. Very nice puzzle.
Few minutes to figure out unique arrangement of digits due to orthogonality constraints.
Few minutes to work out unique arrangement of boxes due to centers. The cage positioned one of the boxes.
Then, from the box arrangement the 4 corners + center box must have have the same orientation and the 4 edge boxes must have the same orientation.
Then, just a matter of filling in boxes.
What an incredible puzzle. Love Simon's stalwart refusal to do sudoku whenever a more interesting logical path is available.
This was incredibly satisfying to watch. Your joy and enthusiasm for sudoku is what got me into the hobby in the first place!
When I was solving this I totally forgot to check for all the negative constraints after deducing the pattern of each 3x3 box, and it actually led to an interesting bit of symmetry. You can show that if each of the 3x3s boxes inserted into an 11x11 grid are rotations/reflections of the same digit pattern, then:
1. All the 3x3 boxes are laid out in the same rotationally symmetric pattern as here (although it can be reflected as well).
2. All four 3x3 boxes touching the edge of the grid and the center 3x3 must be identical.
3. All four of the remaining 3x3 boxes must be identical as well.
So there really aren't many degrees of freedom here! You could make a very similar puzzle where the rules are that each 3x3 box must be a magic square (because again that locks you into a single digit pattern in each 3x3 which can be rotated/reflected), with a small number of simple clues scattered about to deal with the degrees of freedom mentioned in 1/2/3 above.
The most amazing thing about this to me is that of the 9 sudoku squares, 5 of them share the same pattern for corners and center, and the orthogonal positions from center are the same pattern, but rotated 180 degrees. There is no mirroring at all which is amazing!
Finally got a faster solve than Simon, by about 4 minutes! The 'propeller' lines that Simon drew for the final stage actually provide a way to visually solve the position of the 9 squares. If you imagine you have a box that is 11x11 and you have 9 (unnumbered) 3x3 squares in it each with a propeller drawn on them, then the puzzle is to arrange them so that none of the propellers line up. If you start with all nine squares together (in a normal sudoku grid) in the bottom left of your box then it very quickly becomes clear that you can remove all the vertical alignment by shifting the middle row of squares to the right by one, and the top row to the right by two. If you now repeat this with the columns, the only moves you can make are left column up 2, middle column up 1, and you get the pattern above. The only other possible sequences of shifting rows and columns (that don't get 'stuck') either produce the same pattern, or its mirror image, which is invalidated by the 11-cage.
Now I finally understand why we don't see any clue-less XV+kropki Sudokus with non-consecutive constraint.
Fafrd, this is simply amazing. This is the most awe inspiring puzzles I've seen on the channel. Well done my friend, well done. Again... it's simply amazing!
I finished in 121:08 minutes. This was absolutely an incredible sudoku. The fact that this is even possible is a testament to the brilliance of the constructor Fafrd. This truly was an incredible discovery. It took me a while to spot that 4 was the most limited and being that limited, forces it to always create the same pattern with 2 in the middle. It took me even longer to try to fit all the 2's in the grid. I don't know why it took me so long to see that the most efficient wasn't a knight's move away, but rather a super knight's move, which is three moves and a turn. That solidified it for me and the rest was a cake walk. I very much enjoyed this puzzle and I still can't believe this is possible. Congratulations, Fafrd. This was a thing of beauty. Great Puzzle!
As always, your enthusiasm, Simon, is entertaining and inspiring. Thanks so much for opening Mark's email and attempting this puzzle, as bonkers as it seemed at first. I loved this video.
32:35 finish. Whenever there is an 11x11 puzzle with the "build your own non-overlapping 3x3s" constraint, I always start them in the same way. The nine cells at the intersections of rows and columns 3/6/9 each have to be in a 3x3, and moreover each one must be in a 3x3. Given that both cells of the 11 cage have to be in a 3x3, you know that the bottom right cage is in rows 9-11, with the center 2 in row 10. A quick test of r10c8-10 shows that r10c9 and r10c10 will break, giving you your first 2 in r10c8. The distance restrictions will allow you to place the rest of the 2s, and work from there. Excellent puzzle!
And look, Simon! EVERY box has a 3 in the corner!
I was so lucky today! Once I figured out that there was essentially only one possible way to fill a 3x3 box observing the constraints, modulo rotations and reflections, the puzzle reduced to finding the placements of the central 2s. That screamed *modulo 3* to me, since the column/row rule of sudoku applies, the rows *and* columns of the 2s must be staggered. Efficient placement of the 2s almost by definition makes symmetry inevitable. Hence the 2s form a square but a skewed square to ensure the mod 3 condition. Rows 1,2,3 go with one residu class mod 3, rows 4.5.6 with a second residu class mod 3, and rows 7,8,9 with the remaining residu class mod 3. The 11 box disambiguates the situation. The rest is standard sudoku with simple constraints.
I was able to finish in 33:02 but I used pen and paper to do the maths. It helped hugely that I recently wrote up a proof of a theorem in combinatorics that involved mod 5 classes and their arrangement which had the same flavour.
99.9% of the time I am totally baffled by the high-end puzzles on CtC but today I was very lucky..
That was such a joy to watch!! I've never seen Simon get so excited when he solved a puzzle! And what a puzzle! The finished grid is just stunning! By far the best solve/reaction I've seen Simon do!
I think once you establish the 9x9 you need to go back to the 11 cage and decide there’s only a couple places 2 can go.
Yeah, that‘s what I noticed too. Mind you I failed miserably at deducing the pattern of the cage. However if you check the pattern you notice that the cage can only be 47 or 29 and with that you can basically bruteforced the approximate position of the 2 in the bottom right and continue on simons solving path
Aren't there three places where the 2 can go at that point? Either at the top of the 11 cage, or either side of it?
@@RichSmith77 yeah, so you can rule out the rest of that row and then basically move on with what simon did by eliminating the possibilities one by one.
@@RichSmith77 Exactly (not that I would have seen this on my own, mind you). So there were in fact fewer possibilities to check: Simon could have saved himself, oh I dunno, maybe 2 whole minutes on that.
@Richard Smith only the two positions listed above. Because the 11 cage is on the edge you have to take Simon's original grid and flip it to find a pair that will fit...either the 47 or the 29 pair. And with that we know the order. If the 2 is in the cage so is the 11, same for the 4 and 7. So either the 2 is in the cage or the 7 is...2 positions.
This puzzle is mad! There is so much effort upfront just to figure out what is going on. But as soon as that's done, the filling of the puzzle itself becomes incredibly simple!
Very fun.
20 minutes for me! This is a great puzzle and I really enjoyed it. I complete almost all the puzzles on this channel (eventually) and habitually take 2x, 3x or 4x what Simon or Mark take. This is a very very rare exception where I was much faster. I started by drawing a graph of the digits with pencil and paper (OK Simon couldn't reasonably do that on video). Then join all the allowable pairs. It's a very sparse graph and by referring to it, it's easy to quickly (a) make a valid 3x3 box and (b) rule out any other possible 3x3 boxes apart from rotations/reflections. Now you have to put 9 3x3 boxes in an 11x11 grid with none sharing a row or column. That sort of "staggered" pattern comes immediately to mind. Perhaps I've seen a similar puzzle before or can just see intuitively that it works and is the only way (apart from its reflection). Now it's clear that the centre of one of the boxes must go just left of the top half of the cage. The rest is just slamming all the digits in as quick as possible. I don't often watch a whole video here (spending too long solving them!). I usually just watch the intro and explanation and sometimes the end to check. But I had to watch this one to see what took so long! Feeling smug now! If only I could consistently solve them anywhere near as quick as Mark or Simon!
Perhaps I have a slight advantage that I'm not worrying if I'm being trolled by an unsolvable puzzle? I just assume it's valid and get on with it.
He lost a bunch of time not realizing that the first 2 only had three possible positions to rule out. He wasn’t simply concerned he was getting trolled he thought he messed
up something with his initial assessment that the two is in the center.
Sometimes being super clever is a disadvantage.
I tried this puzzle too, I typically am not able to solve the puzzles Simon can but I could today. Idk how but I got a diffrent solution
@@alexsadler3496 solution should be unique. Check yours for errors.
56:30
There is an overlooked conclusion there.
Because of the 9 in the bottom centre square of the top right box, the top centre square of the middle box must be a 6. This expedites the puzzle by 5-6 minutes, all other things being equal.
This felt like a pencil puzzle that Simon created a sudoku grid for in order to solve the pencil puzzle which then gave the real sudoku grid for Simon to finally solve. Very pleasant evolution of the puzzle
Wow this is a pure gem ! For the first time I managed to solve the puzzle without seeing any of the video, and I still watched it for the pleasure of seeing the deduction process ! Thank you Fafrd and thank you Simon for this wonderful moment !!
Absolutely fantastic!!! Simon’s excitement at solving this is sunshine on this gloomy day! 👏🏻👏🏻👏🏻👏🏻👏🏻👏🏻👏🏻👏🏻👏🏻
The sun will come out tomorrow.....😄
@@davidrattner9 I appreciate the thought, but weather reports say more rain. 😕
I've never watched this channel before, but watched this one straight through, and so pleased I did. This was a lovely puzzle, and I really appreciate the energy of the video.
I solved this one together with my sister and it took us roughly 40 minutes. This didn't feel very sudokuey, but was very fun as a kind of more general logic puzzle. I can't even imagine how you come up with this.
I'm convinced that he intentionally skips over basic scanning to figure out the hardest way possible of doing a puzzle so that the videos are longer for monetization... He could have used sudoku basic from the very first box once he gets the windmills. That being said. This is a truly beautiful puzzle standing ovation for Fafrd.
I dont think i have seen Simon so happy at a solve than i have tonight
Julius Caesar - veni, vidi, vici
Simon Anthony - vidi, vici, veni 😁
Simon, at 0:58:25, you had actually solved the puzzle by saying it was identical, but you didn’t realize the scope of that statement. You diligently still did it square by square! After you had just said it was identical, the bottom left corner became obvious and then the entire left side. Great job!
Fafrd is a warrior hero in Fritz Lieber's Fafrd and the Gray Mouser series of Sword and Sorcery.
What's absolutely beautiful to see is also the blocks repeat themselves. It's a giant X with the center block that is identical, and the four side blocks are identical. Genius of the HIGHEST order. I literally rewatched the whole video just to follow the logic again. What sublime beauty.
I wasn't following this channel for a longer time (I had to concentrate on my life, for short). But this got me hooked again.
It's absolutely astonishing how you got to the solution. The last steps were easier than you made them, really just pure sudoku, but to get there at all, I would have given up at several milestones ahead already. But Simon (the sorcerer), of course, found them all in the right order in which you needed them to step forward. Wow, and 1 hour for it while also explaining this all to us, that's marvelous.
oh hey olaf! fancy seeing you here
Honestly watching an expert solve puzzles like this - that are far beyond my abilities - is the same rush as watching sport at its best. The hour I spent watching this video absolutely flew by. What an elegant puzzle.
I loved watching this unfold, I was yelling at my screen when you forgot the cage, but other than that, it was amazing
I came at this slightly differently as regards the 2s, when I really thought about it. If the 9x9 square is constrained for positioning of 2s, this immediately suggests some sort of symmetrical pattern. It being a 9x9 square, you can repeat the symmetry for all but 1 square, and there will be a symmetry of some form around a value of a multiple of 2. Then, the 9th 2 must be placed in a place where it is the symmetrical counterpart to itself, and so, one 2 should be right in the middle of the grid. This results in 4 identical shapes being made, and in no shape can you actually place 3 2s. This means the pattern on 2s must be such that all 2s form a rotationally symmetric pattern. Given the 11 cage, there are only a few places to form this symmetrical pattern, given the only 11 totals in the sudoku boxes at the end are 29 and 47. The only way to make the pattern work for rotational symmetry are to use the extremes of either two central rows or two central columns in each of the shapes left from the central 2 and the 2 in the bottom right sits on the very bottom of the 9x9 square.
Yes. This sounds similar to my approach. Although I did waste some time, like Simon, trying to put the bottom right box (with the cafe) as far into the corner as it could go.
(I think there was a uniqueness argument that could be used to rule out 2-9 in the 11 cage, since I couldn't see how you could then prevent two solutions by flipping each box around its vertical axis. But I could be wrong about that.)
I eventually realised the boxes would most likely be arranged symmetrically, although I'm not sure I had proof for it, or it was just intuition.
Baffling! Amazing! Stupefying! Thank you for the genius solve Simon and thanks Fafrd for a genuine piece of art!!! ❤🎉
Amazing puzzle, amazing solve with such an amazing community day in and day out!!! Will Shortz legendary indeed. Fabulous interview with him. Used to also see his puzzles in Games Magazines in mid/late 80s to early 90s. Genuis he is.
I actually made a mistake while solving this (I assumed that 2-9 was the only pair in the solved square that added to 11) but it ended up being worth it. The puzzle ends up breaking with the impossibility of arranging the regions, but by that time I had worked out the required region pattern and it ended up being easy once I placed the 11 correctly
I decided to have a go with this before watching Simon solve it. It took me about 90 minutes; the first 80 minutes or so was figuring out the only valid 3x3 and then the grid layout; the rest came together rather quickly. I was afraid I might have made a mistake somewhere because the web site doesn't validate it, but skipping ahead I see I thankfully reached the same conclusion as Simon.
A beautiful puzzle and incredibly satisfying to solve. Now to watch Simon and see if his logic was the same as mine. :)
The sheer joy Simon has at the end when he gets it right is a balm for my soul.
Thanks for the video.
26:18 I immediately thought of the greatest video in this channel that probably got most of us addicted to it, the fabulous OG Miracle Sudoku. Lots of tigger bounces in that one!
Anybody else finish the puzzle (which I was SHOCKED with!) and get a “Bobbins! That doesn’t look quite right” from the checker?! So anticlimactic to go through that whole puzzle only to be told I was wrong and then come back here and see Simon get the exact same grid but have it check cleanly
Did he have it check cleanly? I can't find it in the video. [edit: ah it's at 1:01:51] And YES I encountered that same issue. Although it says "puzzle doesn't contain solution". Also weird how it doesn't check for conflicts, I went over the entire grid to see maybe I made a mistake somewhere 🙄
Same for me. Something is broken I guess
Same! I had to watch to figure out if I'd gotten it right. Funny how missing that little bit of software hooray is a little bit of a let down!
Same!
Yep - came looking for this comment!
52:15 "Let's re-invent the wheel, so to speak" that was nearly as beautiful as the puzzle itself.
Clickbait title: Solving an impossible sudoku with propellers and prehistoric Michelin tyres.
This was amazing. I don't understand why youtube only recommended this channel now. Finally something really good. I know I'll be binging this channel now...
These miracle puzzles are getting more and more amazing. 😂
I really wish I could like a video multiple times. This has truly been something else. Purely based on logic, simple rules and a miracle solution. This might just be my favourite puzzle Simon has ever solved!
Less than a minute in, and I'm clicking "Like". I know I am in for a treat éspecial! This solve is going to be epic - I laughed out loud at seeing the puzzle reveal. What a construction. Stupendous. I'm settling in with my cup of tea and biscuit.
Absolutely brilliant! Took me 1:17hr to solve. Amazing that it's actually got a unique solution. Hat's off to Fafrd for the mind-bendingly constraining ruleset!
I love how simon took a very strange sudoku and managed to make it look like a star-battle for a section of his solve.
What does star battle mean?
It's always nice to see how you solve these puzzles. I went about this with a much more brute force oriented process of elimination way and eventually came to the same result but checking your solution and seeing how well you're able to create visuals to systematically break something like this down was really nice. Thanks for introducing me to this beautiful puzzle.
Hi Simon, I don't know if you've ever heard of point groups (and their crystollagraphic extrapolation, space groups) but give the role of reflections and rotations in puzzles like this one, you'd qutie possibly find the subject interesting.
30:04 what im thinking is since this is a repeating pattern, no matter how you slice it, you will end up with the same configuration of 3x3's with at least one of them in a corner. so the real goal I think of the 11 cage, is to give you one box with a starting orientation which you can then use to propagate the rest of the boxes.
I think the key to figuring out where the two goes might be to start in the center box.
That's what I did. First thing I saw was that the centre of each box had to be a two. Second thing I saw was that the two in box 9 had to be in row 10, and that each row in each box had to contain the same digits. From there I decided that the only way that was possible was if the middle digit was a two.
Solve path was entirely different from what Simon did but the end result was the same.
it’s great hearing your train of thought. encouraging, mate! i love the minimal puzzles too
If the box with the 2 in the middle is the only possible 3x3 placement, does that not immediately dictate the 2 goes either in the top box of the 11 killer, or 1 to the left or right? Of course that 2 could be not the only possibility, but I cannot think of any reason why it would allow the 2 to be anywhere else but the bottom row of the pseudo 9th box the 11 cage is in, if that is the only 3x3 pattern allowed. (Currently 42 minutes into the video, just breaking my mind over why one would need to test any of the other cells of the pseudo box 9 but the bottom line)
Awesome! I really liked the logic for this puzzle. I've been working on something similar, using negative constraints, and the way this one breaks down is cool
Loved this puzzle! I think there's a bug in the web app though; I got the same solution as Simon (only after watching the first half of the video, I hasten to add), but when I validated I got the dreaded "bobbins" error message. Looking at the browser console, the validator found three errors, all of which seem false. This one is the most obviously wrong:
"Error in cage sum in cage r10c9-r11c9. Expected "3" and found "11""
11 in r10c9-r11-c9 is literally the only instruction we start with, even if all my solving was wrong that one has to be correct.
Same- loved the puzzle, but got frustrated when my solution got "Bobbins!" Checked that it matched with this video solution and got even more frustrated....
1:03:00 Not only that but the rules aren't newly-invented rules carefully crafted to ensure uniqueness. The non-consecutive, no 1:2, no 5 and no 10 rules are already familiar to variant sudoku solvers.
That is an incredible discovery - Fafrd, take not just a bow but a sheaf of arrows to go with it 😆
I feel like I lucked into the disposition of the boxes, but if you've figured out that the 2s have to be in separate rows and columns then it is relatively easy to visualise that they must be in some kind of stepped arrangement, where box 5 is centred in the grid and the ones above, below, left and right are offset by 1, and the corner boxes are offset by another 1 in each direction.
No BOW jokes (is my stern reply!)
How fun that you're friends with Will Shortz! I don't know him, but I feel as though I do because I listened to him on the radio for years and years. Nice to know that you know each other, my two friends I don't know.