22:28 I didn’t understand why there must be a 6 there. And why 1 must have a value of 1. Couldn’t 5 exist off the pink lines in box 3 making 1-2-3-4 pair valid??
Coming in after solving the puzzle myself is always interesting, because I get to see Simon leaping almost immediately into logic that didn't occur to me at all to skip entire sections of my solve, only to get stuck trying to figure out the logic I found first.
It is a *cosmic-class* construction❗I hope it will be included in the next *CTC Cosmic Hits* book. It might seem too intricate to be cosmic-class, but Simon's logic path can be simplified and hence made more enjoiable by using the *three* logic shortcuts I described in my separate comment.
Rules: 02:13 Let's Get Cracking: 04:38 What about this video's Top Tier Simarkisms?! Bobbins: 1x (1:20:35) Maverick: 1x (06:13) And how about this video's Simarkisms?! Ah: 16x (16:30, 17:02, 17:02, 27:59, 29:55, 49:53, 52:03, 54:18, 54:18, 1:02:20, 1:04:07, 1:04:07, 1:04:07, 1:11:59, 1:16:11, 1:21:49) Sorry: 10x (11:15, 13:51, 16:53, 17:02, 17:08, 21:42, 40:54, 59:38, 1:06:11, 1:13:12) Pencil Mark/mark: 9x (09:38, 21:46, 26:18, 58:13, 58:15, 1:00:29, 1:04:44, 1:05:21, 1:16:00) Weird: 8x (27:38, 27:49, 49:17, 49:58, 49:58, 50:59, 1:16:11, 1:24:03) Hang On: 7x (21:35, 37:31, 43:45, 54:18, 56:15, 1:09:08, 1:19:52) What Does This Mean?: 7x (07:50, 13:26, 19:31, 22:18, 53:37, 1:00:57) Clever: 6x (29:55, 30:02, 1:16:53, 1:22:43, 1:22:46, 1:24:37) The Answer is: 5x (20:09, 20:51, 24:25, 55:56, 1:08:17) Nonsense: 3x (23:50, 28:51, 37:35) Whoopsie: 3x (56:37, 56:37, 56:40) Fabulous: 3x (1:23:46, 1:24:03, 1:24:06) Nature: 3x (30:34, 51:07, 1:08:34) Good Grief: 2x (1:23:12, 1:23:54) Goodness: 2x (1:03:49, 1:23:46) What a Puzzle: 2x (1:23:49, 1:23:54) Out of Nowhere: 2x (1:14:53, 1:16:14) Naughty: 2x (40:16, 1:15:05) Beautiful: 2x (39:46, 55:32) Epiphany: 2x (54:51, 54:54) In Fact: 2x (06:23, 1:08:10) Obviously: 2x (03:52, 10:36) Wow: 2x (54:24, 1:15:02) Have a Think: 2x (29:12, 29:17) Useless: 1x (58:10) Bother: 1x (1:06:05) Lovely: 1x (1:18:44) Break the Puzzle: 1x (04:54) Fascinating: 1x (23:21) Incredible: 1x (01:45) Extraordinary: 1x (1:23:59) Take a Bow: 1x (1:24:54) Shouting: 1x (1:13:13) Think Harder: 1x (51:30) Propitious: 1x (00:50) Phone is Buzzing: 1x (53:33) That's Huge: 1x (19:57) Most popular number(>9), digit and colour this video: Ten (35 mentions) One (232 mentions) Green (7 mentions) Antithesis Battles: Low (6) - High (5) Even (8) - Odd (2) Row (15) - Column (11) 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!
Fans of the CtC streams may like to know that Islands of Insight (recently streamed by Simon and Mark) is currently free on Steam. Possibly only briefly, so get it while you can.
Well silly me i didn't read the rule about the adjacent 1 and 9s.... Needless to say, i stared at a deadly pattern towards the end of the puzzle for longer than I'd like to admit. However, even without using the rule the puzzle allowed for some very interesting and intricate logic all the way till the deadly pattern. Really enjoyed it
This is the sort of puzzle that my first impulse, looking at the grid before looking at the rules, I think that I can try that. Then I look at the video length and think, no matter how many explanations Simon gives which is taking up the time, I will not be able to solve this. Then I see that 1-or-10 value for the 1 and I say, nope, this is one to watch - and it was very interesting. I certainly did not follow every nuance, but I enjoy your videos, Simon. Hard puzzle, though, and yes, confusing!
I had the same thought lol. Usually Simon is good about not having "clickbate" titles. Usually if the title is "this puzzle is the most amazing puzzle ever". It actually winds up being that. But in this case, his title was 100% false. I feel like there is no way that filthy casuals like me could get anywhere with this puzzle. Granted, I'm only an hour through video but I highly doubt it's going to turn into something where it was easy in hindsight.
Okay, maybe I'm missing something, but around 44mins doesn't Simon make a logical error? I mean yes, it was correct that R8C2/R8C5 had to be an 8/9 pair, but that's only because of an X-wing on 1s in R8/9. Otherwise, the renban could have been 1, 9, 8, 7.
No, because it is already known that the U-shape in the box that holds 6 on row 7 must contain digits 1..5. Thus the X-wing is a given even if it can't be placed yet.
Even though he is right, he does not invoke the x-wing - I do think he missed the option of it being 18 (but had he remembered he would have noticed the x-wing I'm sure, so not important for the solve, just slightly unsatisfying)
Simon's solve was brilliant, but he missed *three* logic tricks that in my opinion would have made his path considerably shorter. *SHORTCUT ZERO* Immediately before my first shortcut, you need to rule out *5* from the vertical renbans in columns *2, 5,* and *8.* The most elegant way was explained below by *Richard Smith:* _"Of the six renbans of _*_length 5_*_ or greater in the bottom six rows, five of those need a _*_5._* (as shown by Simon @5:09, only one of the *U-shaped* renbans can be a _"funky"_ 6-7-8-9-1; the other two must contain a 5). _"Along with the given _*_5_*_ in _*_r4c5,_*_ those account for all six _*_5s_*_ in the bottom six rows, so you cannot put an additional _*_5_*_ on the vertical renbans."_ *FIRST SHORTCUT* 👉 The central cells of boxes *1, 2* and *3* must contain a *2-5-6* triple. For instance, in *box 1* there are two 4-cell renbans, and since they use values selected from the *1-10 set,* one of them must contain *3* and *4,* and the other must contain *7* and *8.* Hence *3, 4, 7,* and *8* are ruled out of the central cell. Moreover, *1* and *9* are ruled out of the central cell of *box 1* because they would make the vertical renban in *column 2* identical to (and hence incompatible with) one of the renbans in *box1.* For instance, if Central cell of *box 1 = 1* then the only possible configurations for the renbans would be 🔹2-3-4-5 and 6-7-8-9 in *box 1* 🔹6-7-8-9 in *column 2* and they would create an impossible 6-7-8-9 *quintuple* in *column 2* (5 cells containing only 4 candidates). For similar reasons, the same digits (1, 3, 4, 7, 8, 9) can be ruled out of the central cells of *boxes 2* and *3.*
*SECOND SHORTCUT* (available immediately after the first one) It allows you to place the first digit: 👉 Central cell of *box 2 = 2.* We have shown above that this cell must be either *2, 5,* or *6.* However, by sudoku, it cannot be *5.* Moreover, if it were *6,* only two possible configurations (2-3-4-5 and 7-8-9-1) would be available for the renbans in *box 2,* and it would become impossible to fill the renban in *column 5,* which 🔹by sudoku, cannot contain a *5* 🔹by sudoku, must be different from the renbans in *box 2* 🔹by sudoku, cannot be *1-2-3-4* (it would create an impossible 1-2-3-4-5 *sextuple* in column 5)
*THIRD SHORTCUT* Shortcuts *0* and *1* make it easier for you to figure out that the _"funky"_ U-shaped renban (containing 6-7-8-9-1) must be in *box 9,* since the vertical renban above it must contain *1-2-3-4.* In turn, this deduction allows you to disambiguate the *5-6 pair* in *row 2:* 👉 Central cell of *box 3 = 6* (as shown by Simon @21:53) 👉 Central cell of *box 1 = 5* Also, and more importantly, it makes fairly easy for you to deduce that *boxes 1, 2* and *3* must contain either: 👉 *r2c2 = r1c4 = r3c8 = 5,* 👉 *r2c1 = r3c4 = r2c8 = 6,* and 👉 *r2c3 = r3c5 = r1c7 = 1...* or the opposite configuration, with *r1c8 = 5...* However, the opposite configuration would have a *2-3-4 triple* in the first column of *box 3,* which would be incompatible with the 6-cell renban in columns *6* and *7,* as it would rule *6* out of it, by forcing *r4c7 = r1c6 = 6.* That's *cosmic class* magic, isn't it? 👽👍👍👍👍👍👍👍👍👍👍
Hi Paolo. I don't have too much time to look in detail today, but I'll come back and study it later. (If I load up my solve of this one, it tells me I spent 660 minutes on it - I don't think I was giving it my full attention for all that time, but the video title seems like it was apt - "don't worry about time ..." 😂). On your first point, I think this relies on you ruling 5 off of the four cells vertical renbans first (c2,5,8)? You can get that deduction fairly early, but not trivial, I would say. Then you know those renbans are one of 1234, 6789 or 789(10). I know I was slow to eliminate 1 and 9 from those three cells in row 2.
[Edit: Sorry, I was 100% wrong here 💩] @@RichSmith77 -The first shortcut does not require you to rule -*-5-*- out of the vertical renbans. In my solve, the configuration of those renbans was just a consequence of the first shortcut.-
_"The twist on normal renban rules is so minor that you can hardly believe how profoundly it changes all the logic. This is a masterpiece."_ (from the video description) You definitely aroused my curiosity‼ I am looking forward to solving this, and I bet you are right.
I normally skip puzzles which take Simon over an hour but the title convinced me to try. It took me 469 minutes but it's finally done. Well worth it, great puzzle.
Fun fact: this modification to renban rules breaks the concatenation property of regular renbans, which states that if two sets of digits each form a valid renban, and they share a digit in common, then their union also forms a valid renban. Here, however, we are told that both 9 and 2 are adjacent to 1, but 129 is not a valid renban, even as it is the non-disjoint union of 12 and 19, both valid renbans.
39:20 Can't put 2 on the line because it would force 6 into the middle which clashes. 41:00 ...that's for explaining -it wasn't clear to me but I also didn't expect the explanation to come later I don't mind Simon over-explaining at the start because others might well need/want it and I can always skip a bit but sometimes later in the video there are crucial gaps in the explanation.
I started with the (after the 5 from the renban in box 5) with the non-renban cells in row 8 (c2, c5, c8). These must be populated by numbers that do not overlap among the renbans along the bottom of the grid. 2 and 9 are immediately candidates (with 2 going into c8 due to the 67891 renban in that box), with 3 and 8 joining 9 as candidates in the other boxes.
I admit i did get stuck at some point but with a little help from Simon I think there is a much more straight-forward solve: 1. show that the 3 vertical lines of length 4 don't contain 5s. 2. show that each combination of the vertical line forces a different middle digit in boxes 1-3 (the digists 2,6,5). 3. determine the correct boxes-column for each combination (which would be simple enough by this point).
1:19:01 Because of the 12 in r7c1 and r7c3, either the 1 or the 2 on the 1234 renban would have to go in r1c2 or r3c2, so therefore r1c2 and r3c2 are a 26 pair.
The second secret of Sudoku would be a game changer for Simon if only he knew it: When you put a big number in a cell, you are actually allowed to remove the corresponding pencil marks 😁
I finished in 299 minutes. This was a rough one. I felt like I was spotting some good things, just not enough to crack it open. I kept trying to work on the bottom three boxes and just kept hitting a wall. I finally got lucky when I spotted what I think was the intended solution of using the main line in box 6. The 5 that is forced on that line restricts the digits of the straight line in column 8. That was enough for a break-through and I finished the puzzle. My brain feels like mush. Great Puzzle!
We can forgive Simon for backing into which line is what in box 1 at the end...figuring out that the "other" smile had to have a 3 on it was pure genius and gave me my chance to solve this awesome puzzle!
I love that Simon’s reason there had to be 345 in box nine row 1 was not because there’s a 1,2 pair in row 7. He always has a different reason than me why some logic is true.
That's great ! I did need lots of help in the early part. But, once I knew where the 6 went in box 5, I could place the 6 on the hook shaped renban in box 8. Then it wasn't too bad.
What a brilliant setting! That was quite difficult too, every deduction was hard-earned and felt so rewarding. This was not an "easy walk after break-in" at all. Totally enjoyed it. Somehow I managed to figure out the vertical lines quite early, but took longer afterwards, while Simon was quicker on row 7 and 8 (well, not so quick but quicker than me, took me a while there...)
I fell like, even though Simon and I figured out many of the same logical steps, his were in a very peculiar order that made the puzzle much harder (at least for me to follow). It may also help that I write notes in a notepad on the side to help keep track of some things. We started in the same place with the 5s in boxes 4-6. Next step was the "smiles", and realizing none of them could contain 3 repeated digits. So they must contain the 12345 and 67891 runs, otherwise 6 or 5 would repeat too many times. This meant that the 3 cells in row 8 that are not on smiles are the 3 digits that do not repeat among the 3 smile sets. This is where what I think one of the keys to the puzzle comes in: none of the 3 "struts" of length 4 in boxes 4-9 can contain a 5. For the middle strut, there's a given 5 pointing at it. For the right strut, putting a 5 in the top half would prevent the two 6 length renbans from completing (both 5s would be in row 7), putting it in the 3rd cell would also prevent it completing (both 5s would be in box 6), and putting 5 in the bottom cell would result in two many 5s in the bottom two rows, as we already have set 12345, 67891, and a third set that must contain a 5. The left strut followed the same logic of breaking the 8 renban in the top half, breaking the 6 renbans in row 7, or having two many 5s in the bottom two rows. This means the struts contain exactly three possible 4 length renbans that do not contain a 5: 1234, 6789, and 7891. From here its a combination of seeing how these struts interact with the center cells in boxes 1-3, and with the smiles (smile 67891 must match with strut 1234, both smiles in boxes 8 and 9 can't contain 5s otherwise the 6 renbans break, etc) to slowly pick it apart. Overall a very hard puzzle, with some interesting logic.
Finished in 59:45. Interesting use of an Ace card in Blackjack idea with renbans. That plus the 1 and 9 not being next to each other were the real keys of the puzzle. Fun puzzle!
I got more traction on the pinwheels at the top than Simon seems to have. On the other hand, I had less insight into the smiles at the bottom and resorted to a good deal of "brute force" elimination.
@1:20:22 Simon you knew that the 1234 can't be in the upper left most renban, because you have the orange 23 pair in r4c1 and r4c2 which precludes any 23 pair from being in column 1.
At an hour in there is a swordfish on 7s in columns 2, 4, and 5 and rows 5, 6, and 7 eliminating 7 from the Renban in columns 6 and 7 in boxes 5, 6, 9 and 8 making that 6 long Renban a 1-6 Renban which forces a 1 onto the cell r4c6. I wish Simon did more Sudoku in Sudoku puzzles.
34:13 This was an absolute gem. Had a moment at the end where I forgot about 1s and 9s not being permitted in a domino where I couldn't quite resolve the high digits, but as happens all too often it was just my forgetfulness rather than any issue with the puzzle. 😂
Somehow I've already solved this. Is this a repeat or did I perhaps stumble upon it elsewhere? Great puzzle that apparently took me 46 minutes. I remember I enjoyed it 😊
For all that Simon likes to make a big point about blue and orange being the most colourblind friendly combination of colours, he doesn't half like to use dark green and digit blue that both make it practically impossible to read pencil marks and even digits 😥. It would really help if the pink used for renban lines wasn't such a dominant colour as well.
(Not to mention that the blue and green were relevant for about a minute and should have been deleted after that, but hey, it wouldn't be a Simon solve if he didn't leave artefacts from every completed deduction littering the grid and getting in the way 😢)
@@stevieinselbyyes, he occasionally tidies up pencil marks (although sporadically, not methodically, as a general observation); but he almost never tidies up redundant colouring!
First glance at the thumbnail: it's never possible to distinguish the difference between the pairs of digits with the sum of 10, e.g. 1 and 9, and it's also impossible to place three 5's in the bottom two rows...
Tough gig this one! Puzzles me that Simon uses logic on something, but then doesn't look at the symmetrical counterpart. For example @1:14:49, he pencilmarks the 6's from one renban, but not the 2's from the other. Would've given a 26 pair. Useful? Maybe not. lol. At least, not immediately. Although the "23 pair that came out of nowhere" shortly thereafter would've been "just a 3".
Did Simon just solve that the middle vertical line can't be 1234 and therefore the rightmost line must be, only to then go on a trip around the Sudoku to solve this again 40min later? Like didn't we know from the first discovery with the 5s that it's impossible?😅
Did I miss something? 😅 At 41:20 Yes the renban in column 5 can't be a 1234, so it can't have a 12 pair in box 8. So yes it could be an 89 pair in box 8. BUT could it not also be a 7891 renban, and therefore have an 81 pair in box 8, and a 79 pair in box 5?
Fairly sure Simon overlooked it, but as others have pointed out, 1 can be easily ruled out from r8c5 by the virtual x-wing on 1's on the smiley faces (one of them is 12345 and one is 67891, between them using the two 1's in rows 8 and 9).
I need help. At 41:30, why it must be an 8-9 pair? Can't it be an 8-1 pair, and then have 7-9 in box 5? (Edit: partially explained later, discussing 2=5 in the middle of boxes 1 and 2)
No, that sounds reasonable. This puzzle can be pretty straightforward depending on how fully you grock the start. Simon did the correct logic after identifying a number of possible cases but missed that there was a more generalizable argument. Done that way it's significantly harder
How on earth did Simon conclude that R8C5 has to be an 8 or 9? Isn't there a possibility for it to be a 1, counting as 10, pairing up with an 8 above it and getting the missing 9 from Box 5?
2 bottom smiley renbans can't share 4 digits because there isn't enough space on the 3rd box to fit these 4 digits. So, he needs to avoid either 12 or 89 in box 8. Since the bottom middle strut can't be 1234, it has to be high, so it's either 6789 or 7891. In either case, he needs to avoid 8 and 9 on the smiley in box 8
@@PsychoSoldierPrometheus which still won't answer my question - you can easily avoid the 8 and 9 in Box 8's smile with a 18 on the middle and a 29 on the top but he just got rid of that possibility without explanation
@@MaierFloriannone of the bottom renbans can share 2 digits. Since Simon established that one line has 12345 and the other has 67891, he needs to populate the 3rd line with 5 digits from the "middle", so he has to use 34567, leaving 89 on the strut.
Because of 1s and 9s not being able to be consecutive. There could be a 1 on that line but it couldn't be in that box. In that box, the line could either have a 1 or 9 and either a 2 or 8, he had eliminated 1, 2, 3, 4 line, so it had to be a 8 with either a 1 or a 9. The logic failure was that it had to be a 9, it could have been a 1 but either way it defined the two lines being considered as 1, 7, 8, 9 or 6, 7, 8, 9 and the given 5 gave the order.
@@chipsounder4633 The example says 189 would be valid and 123 would be valid. It clearly doesn't says 189123 would be valid as - again - that would repeat the 1 on the line.
Not a fan of this solve / puzzle. Too long staring at nothing happening. it does seems that puzzles are being designed to obscure the solve path as much as possible and I am generally not keen on this..
This was definitely not designed to obscure the solve path. On the contrary, the repetaed geometry in boxes 1, 2, and 3, columns 2, 5, 8, and boxes 7, 8, and 9 telegraphed very well the logic needed to break-in. It was not a simple break-in, but it was not hidden. Logic shortcuts were also available. See my separate comment for details.