Can We Solve Our Own (Evil) Miracle Sudoku?!

[@tombenizrilevi4803]

I love how Simon figures out the most complex ideas at the beginning of every special puzzle like that, and then misses out on 400 sudoku clues just sitting there in the grid

[@japoorva08]

The guitar piece is absolutely delightful! and so is the puzzle. Such creative gentlemen!

[@njt8185]

instruction to Robert to put out the Rubbish ?, "Bob, bins!" lol

[@sonalita_]

We need merch with "Robert - The Rubbish!" on it

[@petemagnuson7357]

I'd like to see some increasingly esoteric cryptic crossword hints towards Bobbins, personally.

[@hangugeohaksaeng]

Wow, never would have gotten that one! Thanks for solving it.

[@Zhellybelly]

Thanks, came into the comments just to figure what the hell he was on about lol.

[@jessicatait1867]

Took me a minute to get it, but what a laugh!! Brilliant.

[@msolec2000]

Nice solve, minor hatches are forgiven... BUT! Not finishing the coloring is evil. Full point and a half penalty here! So, just an 8.5/10.

[@-42-47]

10:16 "So, he's obviously tried to remove things that he thought would be helpful to me, which is quite typical of him." Man that "approachable" sudoku really made you salty huh?

[@Trias805]

It's been a vendetta from there on.

[@956675]

@@Trias805 but a great and friendly vendetta to watch.

[@remoteviewer8524]

Is it true Simon is releasing a cover of Never Mind The Bobbins by The Sex Pistols?

[@norbertrumelin8781]

I would love a video where Mark reacts on this and gives his comments if Simon follows the ideas he had at creating it

[@tobbgoblin]

Good idea. I’d love that.

[@peterhackett3815]

Second that!

[@jfkz2000]

thirded?

[@jazspencer9190]

YES THAT

[@verstrickt.und.ausgerechnet]

Yesss! 😍

[@ragnkja]

At 24:32, there’s an easier way to disambiguate the 1/9 in rows 7 and 8: now that the 13 sum in row 8 cannot be 5+8, it cannot be two digits because 6+7 is ruled out by the non-consecutive rule.

[@bradfooks4485]

Your saying the 13 clue must be 3 cells long getting the 1 & 9 in column 1

[@ragnkja]

brad fooks Correct.

[@michaelholdowsky6386]

At the end R7 had 346 tripple which sums to 13. 6 must be therefore in the middle. Must get back to my bobbins....

[@Vishnu_00]

Yup, thats how i figured out..

[@glum_hippo]

Play guitar as often as possible, please. The canned Mozart is nice, but this is such a personal touch

[@joeloedeman5160]

U2 very nice

[@undergroundmonorail]

Whew! 4:28:40, though about 40 minutes of that I wasn't working on it because I had another obligation to take care of quickly. Definitely the hardest puzzle I've ever actually solved! Not sure what possessed me to give it a shot when the *cracking the cryptic* people said it was hard, but I did it!

[@47reload]

"It's so fiendish, it should come with a health warning " ... That is the best thing I've heard in a long time! Btw, had the miracle sudoku app added to my wishlist and got it the second steam made it available earlier. Absolutely loving it!

[@Stephen-Fox]

I see we've reached the stage of Cracking the Cryptic where Simon is starting to use cryptic clues for Bobbins instead of Bobbins.

[@russellbarnett7606]

He’s really serious about not helping anyone achieve CTC Bingo

[@stellasdoesstuff]

Me, reading the title: lol I'm sure the title is just exaggerating. It can't really be 'evil' Me, 1 minute into trying the puzzle: oh my god this puzzle is evil

[@geschmacklos3032]

"This green four is not a one or a nine" got me laughing so hard I probably woke up my neighbors xD

[@cubesquared7808]

I find it amazing how one brilliant puzzle can change the entire puzzling community so much.

[@kitsterling8665]

Oh no... Simon, how could you take such a terrible leaf from Mark's book, not completing your colouring?

[@ODIRGO]

He's embarrassed for not having a single number in the grid at 26 mins of the video, I just understood the rules at 30

[@blxckbear9]

24:33 It is absolutely mesmerizing how far you can think ahead without noting it down, it reminded me of the show Sherlock, how he deducts everything just from observing, and then he explains it beautifully. Great play Mr. Simon.

[@logydavyjones5931]

I love how much you continue to explain the logic in these videos! It seems clear that you could probably do these puzzles more quickly without stopping to explain to us why we can know something for certain, so I appreciate that you put so much care and effort into explaining the why’s and how’s for us. It’s so fun for me when I try to get ahead of you, get hopelessly stumped, and then learn new ways to look at the puzzle. And kudos to Mark for making such an excellent puzzle!

[@erickehr4475]

22:41 for me. Unusually for a sandwich sudoku I didn’t finally complete the 1s and 9s until the very end. Indeed my penultimate digit was a 9. I’ll now watch the video to see if the same happens to Simon.

[@coryoubre1526]

Best video so far of not losing focus on the consecutive rule for disambiguation. Excellent job.

[@jaeusa160]

I looked at that Mitchell Lee puzzle for a bit, definitely far above my almost complete inexperience. I can see why you described it as you did. It was interesting to note the relationship between boxes 2 and 4 of the grid due to the 45 little killer contained entirely in those boxes, where the 3 digits of each half of the killer clue had the same sum as the 6 non-killer digits in the other box. But I am far too inexperienced to even begin to puzzle out how to utilize info like that, nor do I have the spare time to learn unfortunately. It just felt to me like that was something important to starting the puzzle, which was neat. :)

[@antalbojtos4413]

"It's so fiendish, frankly, it should come with a health warning" You keep cracking me up. That's not what I'd expect from a sudoku-themed video.

[@EHobbit]

I enjoyed watching the solve and I very much enjoyed hearing 38:10 till the end.

[@callumm092]

I broke in a completely different way to Simon. If you look at the first three columns, there's no way that the 1s and 9s can cross the boundary of a box - so each box must have a pair that is in the same column. Also, just wanted to say I've been thoroughly enjoying the channel. It's kept me sane over the past few months, so thank you for being there. This puzzle was the first I solved myself with no help from the video - been working up to that for a while now!

[@docghines]

I find it useful on your software to highlight in red and blue, and then purple their intersections.

[@kutjegodverdomme]

Well, we’ve heard “it’s approachable” before, haven’t we?

[@ragnkja]

There’s approachable, and then there’s “quite approachable.”

[@stpe9848]

To be honest, he does not say: Approachable by everyone. :D

[@ashwinnaidoo796]

Ho, your approaching me?

[@amaarquadri]

Loved the guitar outro. Keep them coming please! Also, any idea what the song is?

[@harleykaufman1155]

The sets of numbers in each boxes rows and columns brings me joy.

[@ruihankenguan5209]

Finished this in about 50 min. I know I'm not remotely close to Simon's level but you could actually instantly put down a 1-9 pair at r2c3-4 by reasoning about the possible patterns of the 0 sums in the first three columns. Great video!

[@jcbohn3981]

The agony when he doesn't use coloring in the las numbers

[@Necroziz666_]

I kept waiting for Simon to notice the 3 consecutive zeros at the top meant that each box had to have a 1/9 pair and with the 0 in r2 it pretty much told you exactly where the top pair went. Actually got a decent amount of a solve just based on that when normally I look at these and think I'd have an easier time making pixel art with the colored squares. LUL

[@erikkubonhalvorsen4347]

Loved the symmetry in this one. I saw that pretty early and managed to derive some short cuts in the solve. Really simplified it once you see that symmetry. Great puzzle design by Mark.

[@SWebster10]

In this puzzle, as with Mitchel Lee’s, each row in a box consists of the same triplet. Therefore if you know that two of the cells are, say 4 and 8, the other must be a 1. Is this an artefact of the non-consecutive constraint? Is it provable and therefore usable logic? Does it spoil the solve?

[@HughResnick]

Right! I noticed that after a while, and was thumbing through the comments to see if I was being an idiot savant.

[@kennystorgel686]

It it not the same on columns, so might be it is just one or the other, or not at all

[@leftysheppey]

The same is almost true for columns too. There's slight variation though, so I don't think you can use this as a hard and fast rule, if it is ultimately down to logic. E.g. there are 6 boxes with 179 columns, but 3 without in boxes 3, 6, and 9. You could rotate this puzzle 90°, and while the puzzle remains fundamentally the same, using this "rule" on the new rows might mess you up

[@SWebster10]

Kenny Štorgel Columns are just rows rotated by 90 degrees, so if a puzzle has this property for columns it’s the same thing, really

[@BlueCyann]

No, it is not a necessary consequence of the non-consecutive contraint. See for instance spatr's puzzle from August 8.

[@danielrhymer1762]

38:33. Well done Mark for that one! Liked the logic and not overly horrendous either

[@mightymac1723]

Nice U2 entrance and exit - learning so much from these videos - thanks

[@nimbletim]

marvellous outro!

[@bristolrovers27]

Fab puzzle and brilliant guitar intro and outro Tough solve

[@Now_Or_Nova]

It's a really interesting construction, when you notice the composition of the rows in each box. With hindsight (seeing the finished puzzle), it's possible to just write the numbers in according to that pattern and some simple logic.

[@Xeridanus]

The columns too, except the last three one digit of two of the triples is offset.

[@BryanLeeWilliams]

@17:00 you can go further. The 19 can't be in r4c3 because then you'd have a 19 in r3c3 and one in the domino r3c1c2. There'd be nowhere for 21 sum. plus you'd have 3 19's in row 3 because there's 1 marked on the right of the row.

[@n_mckean]

Interesting place to start. I felt the 19 pair in R2C3&4 was the obvious starting point.

[@JudeforLess]

I love how Simon gets flustered by the time it takes to place a digit every time... a) I'd be staring for hours until I'd pulled all my hair out on these diabolical ones, and b) he could start his timer about 4:30 into every puzzle he introduces!

[@Pyromonkey83]

Wow this is a stellar puzzle by Mark. If this is anything to go by for the quality of the puzzles in the app (which I'm sure it will be, given the quality of the previous apps), its going to be a jolly good time :D

[@Trias805]

Finally. I was wondering if you ever create your own puzzles and could publish one of them on YT.

[@firmbase]

When in doubt, use the given digit(s)

[@dragonborne3277]

brilliant setting by Mark and brilliant solve by Simon

[@michaelhoffman2011]

Wow. Really happy I solved it in an hour and a half. I was way to focussed trying to get all the 1-9 that I didn't see the 58 groupings... after finding them, the puzzle seemed to start to unravel itself. Great work guys, can't wait for the android app (I am the only one behind the times)

[@ThaStam]

It is interesting how every box has 3 rows, each being a 1-4-8, a 2-5-7 or a 3-6-9 triple.

[@DubioserKerl]

I also realized this. I wonder: Was this how Mark constructed the puzzle in the first place, or does it always HAVE to be that way with either Row- or Column-Triples (obviously not both at the same time)?

[@ThaStam]

@@DubioserKerl I suppose that it is easier to find a finished grid of that form and then "simply" try and find the absolutely necessary clues to make the solve unique. I would expect it to be a classic thing sudoku solvers do, when trying to set a puzzle themselves, having another one of Mark's sudokus in mind.

[@JudeforLess]

Is this not a trend for all the 'miracle' sudokus? some form of non-symmetrical pattern that carries through? I wouldn't trust it as a form of logic, though.

[@MateoJH]

And each column of 3 boxes have the same vertical triples as well. (358, 246, 179), (358, 246, 179) again, and finally (137, 246, 589)

[@johnpauladamovsky86]

I knew deep down that Simon was a fan of Doctor Evil. How could he not be? His childhood was typical, summers in Rangoon, luge lessons. Sometimes he would make outrageous claims like he invented the question mark. He would often accuse chestnuts of being lazy.

[@ManateeGag]

I never thought about using the math of a row/column to solve for the outies before. great tip.

[@filipdraber52]

At twenty minutes and twenty seconds, this was the fastest puzzle to solve for me. Brilliant, though, looking forward to downloading the app. The nonconsecutive clue and sandwich restriction interact... fascinatingly...

[@filipdraber52]

@@Dakotaidk that was my way of camouflaging the fact that I've never finished any other puzzle without help

[@sjwimmel]

Love the nylon string guitar, Simon! I've had classical guitar lessons for years in my youth so the sound is really soothing to me. I still remember the tune Romanza. It sounds like you're at the level to be able to play it, maybe look it up, it's nice.

[@truscottlee4888]

Ok that was fun!! A nice easy one to start ... woohoo ... 30 minutes ... thank you Mark and Simon for making me feel good !!

[@aere481]

Columns 4 & 6. When you see it, you realize how cruel Mark is, haha! I love it!

[@LordHypnos4]

Finish the coloring!

[@charlottestang9462]

I know you guys have done one before, but I'd love to see another video on setting

[@mwward]

I wonder if Mark meant for the solver to gain more info from the three zero sandwich totals in columns 1, 2, and 3? I'm pretty sure that the 21 and 26 totals on rows 3 and 7 forced the 1/9 pairs into their own boxes (one pair in box 1, one pair in box 4, and the final pair in box 7). Deducing that would have forced a 1/9 pair in r2c3 fairly early in Simon's solve but I'm not sure that it would have particularly sped things up.

[@Twood599]

It's interesting that you started by explaining that sums between the 1s and 9s in rows or columns not given are often large, but in columns 4 and 6 the sum was as small as good be. Devious indeed.

[@peterhackett3815]

Well that took me 1hr 25mins! I just really wanted to solve a puzzle set by Mark and I did it! Loved the logic in it was a beautiful solve.

[@nickm5977]

In individual columns and rows, triples were repeated throughout each row and column of boxes. Very satisfying grid.

[@mostman]

Cleaner path at 26:00 is to realize that row 8 can’t be a two cell sandwich because in order you make 13 without 9 you need 6 7 which is consecutive or 8 5 which would break column 7. You end up in exactly the same place Simon did with easier logic.

[@Kurgosh1]

Just under 40 minutes. I'm pretty happy about that when I look and see a 40 minute video.

[@nijsleendertz803]

Te first thing I saw when i looked at this puzzle was that the 3 zero-clues in collum 1, 2 and 3 had to be in 1 box each (they couldn’t cross a border between 2 different boxes, because if it did, it would have had to happen again (otherwise you woud end up with only one 19 square in a box), but the 21 and 26 clues wouldn’t make that possible). Because of the zero clue in row 2 you knew that a 19 square would have to be in r2c3 otherwise you would have broke the puzzle. I was quitte proud of myself spotting that in the first minute of the video.

[@mitchjemjmj]

Yeah me too! :D I started with that and when I got stuck watched the video and was surprized Simon mist it. (and used, in my idea harder, logic too find the R2-C3 relation at min 16:30).

[@MyNarsa]

arround 26min mark, much simpler logic: 13 in 2 cells -> cant be 49, cant be 58(because 58 pair on last box), cant be 67 because consecutive, so its a 3 cell 13 sum

[@RescueMichigan]

Logic at 25:00 is not easy to see as Simon says -- but unneeded because the 58 pair in r8c7 makes a 2-digit sandwich in row 8 impossible, which is easier to see.

[@simoncoolhand]

Straight away I thought Simon would notice that the combinations 19 in columns 1, 2 and 3 all had to be in single boxes, ie the 19 to make up the zero couldn't span the joint between boxes because of the 21 and 26 clues on rows 3 and 7. So very early using the 0 clue in row 2 you can get that the 19 in that row does have to span the joint between box 1 and 2 and place those in.

[@Xeridanus]

Could you explain why the 19's in the columns couldn't span the boundary between boxes?

[@Xeridanus]

@@DeletedUser410 Ah thank you! I get it now.

[@matthewdodd1262]

The puzzle has a beautiful symmetry. Each row of 3 in the 3x3 blocks contain the exact same 3 digits. Not sure if that was intentional or a product of the restrictions placed on the puzzle.

[@zoicon5]

Nice setting by Dr. Evil. I noticed that Simon's intuition that the absent sandwich clues probably meant that the values would have been too helpful if given was proven correct.

[@danshep79]

I actually started with the 15 sandwich clue - the only ways to form it with the adjacency constraint and the 4 are 3642 and 384, so that puts the 3 into the grid at r5c3 straight away and a 1/9 next to it.

[@RichSmith77]

What about 3624? I also spent sometime looking at the possibilities for the 15 clue, and realised I couldn't place the 3 because of the 3624 option.

[@carljohanr]

All the Cracking the Cryptic apps have been instabuys for me, and this one too :) Biggest fan of Thermo Sudoku so far! would love if purchases translated from the iOS version to Steam, or at least if there was a way to 'add-on' to the purchase so I don't have to pay the full price twice. No idea how the team is able to set this many puzzles...

[@ReiPreguica]

I was a bit mad in the end that the last digits weren't coloured in green when placed, but the guitar piece in the end is so soothing that now I'm glad instead

[@dooby78]

Took me 4h21m. My God, that was a war of attrition! Amazing job by Mark.

[@fuji2086]

30:09 to solve this. I really liked this puzzle. I used the all of the 0s on the first 3 columns and row 2 to immediately place the 1&9 pair in row 2.

[@stgo_g]

If anything, Mark made you look like a genius

[@nadeemhajiiqbal9214]

Simon's logic at 17:35 was sound, but there's a much simpler way of deducing this. Consider at that point, whether R4C3 can be a 1 or a 9. Well, if it is, then R3C3 ALSO has to be a 9 because of the 0 clue in C3. Which makes the rest of the cells in C3 green. Since R3 has a 21 clue, it has to be AT LEAST a four cell sandwich, which makes R3C1-2 ALSO green, And we're left with 8 green cells in Box 1!

[@Kimakae]

It's funny to me that your Android apps get released last, because I work for an app developer, and we always get held up by Apple while we can release to Android right away. Anyway, I love the ad on Steam and can't wait to start playing.

[@TheNewAccount2008]

What a great puzzle. Took me forever to get through... Wow

[@KyleBaran90]

"Can we?" Well yes, the video is posted

[@grotlek]

[Spoiler alert - don't read this before you have a go yourself] .. Great video as always. There is one thing that really gets to me when looking at these puzzles (and I'm picking on this one, even if it's not the first example of this I've seen). It was very clear to me long before the point that a lot of boxes were filled that every single box had three rows, a 148 row, a 257 row and a 369 row. I don't know whether the mathematics made this necessary but it made trying to solve the puzzle a head case of, on the one hand, knowing what the other two cells in a row were once I'd found one digit (or what the entire row would be once I'd found one digit from each of the other two rows), and on the other hand trying not to assume after picking up on the pattern and having every train of thought distracted by that pattern, hehe.

[@bryandedon7459]

He's so good at these puzzles that he constantly looks over the easy stuff and he always gets caught on one part of the rule that he forgets the other rule and the normal sudoku rule. I can't figure out half the stuff he does but it seems like I always notice an easy number he can put down but he overlooks it then gets to it and finds it out the harder way.

[@arnet95]

I had some different logic to actually start placing digits in the grid. Once you've found that the 15 clue in row 5 consists of 3 or 4 digits, you can show that it cannot be 4. Namely, if it has 4 digits, it has to be 2,3,4,6. But this then, with the 5-8 pair in box 6, creates a 5-7-8 quadruple in box 6, which cannot happen. And from that you can start placing digits like it was done in the video.

[@Trias805]

Great puzzle, Mark. But it took Simon only 35 minutes. You need to make it more "approachable" ;)

[@Coyotek4]

52:23 (approx.) ... I worked on this for about 30 minutes and then started over; I only figured out my error on the second pass-through, which took me a total of 22 minutes and 23 seconds. Nice puzzle!

[@negar1368]

this was brutal, but so satisfying to finish.

[@graduator14]

"How long did it take you to solve"? "One MILLION seconds"! **Dramatic music**

[@FireDragons42]

The end goal of the logic that Simon goes through at 24:34 can much more easily be spotted by looking at the 13 clue below it for row 8. If the 13 is in 2 cells then in must be either a 58 or a 67. Well the 67 would break the nonconsecutive rule and we now know that we can't use both 5 and 8 because of r8c7 which has either a 5 or an 8. As such we know that the 13 clue must be at least 3 digits (and since we have no more room exactly 3 digits)

[@ericjones3692]

Should have had the 1/9 in R2C3-4 within a minute. It took you until around the 18 minute mark. The 3 0s in C1-3 means each box contains adjacent 1/9s within a column. The R2 0 means there must be a 1/9 in R2, box 1, and it must overlap to box 2.

[@bencheesecake]

At 26:00, you had some very evil logic to place a 1/9 in r7c1, but looking at r8 there, the 13 cannot be in 2 cells, because of the non consecutive constraint, and it would eliminate both 5 and 8 from r8c7. This combined with the 0 sandwich total for r1 gives the 1/9 in r7c1. Other than that great solve!

[@WhoStoleMyAlias]

18:27 Could have eliminated those two cells from the start, because if you let one 1-9 pair cross the border of boxes 1 and 4 then to complete box 1 and obey the 0 sandwiches in columns 1 to 3 there has to be a second 1-9 pair crossing that same border, which breaks the 21 sandwich on row 3.

[@prateekagrawal5270]

Which app you use for solving these sudokus

[@petemagnuson7357]

There's an interesting pattern in the grid, every 3x1 tile is a scramble of either 369, 148, and 257. Similarly every 1x3 tile is composed of 358, 246, or 179. It really speeds up the last leg if you're willing to rely on the pattern holding. Once you see a 5 you know the horizontal cells in the box are a 27 and the vertical is a 38 pair.

[@davidkay6655]

What about box 3 then

[@glennmelven3414]

Those numbers don't work for columns 789.

[@nicolaidepue3970]

Needed help for most of the sudoku because I kept on making the mistake that he explains in 25:30, and I couldn't figure out what went wrong each time. But I was able to depart from the video at 28:16 once actual digits were put down. Hours were wasted over the smallest mistakes. Why was this created, and why did I find this fun?

[@julieenslow5915]

Simon! You forgot the most obvious logic of all! Who knows your Sudoku abilities as well or better than anyone else? (Mark). Just as you know his as well or better than anyone else. This puzzle was designed for/against you. We all knew that before you started! 26 minutes in and no digit in the puzzle = totally predictable. OK back to the video. Awesome! Tell Mark - Kudos! and the same to you Simon. Loved every minute of it, including the serenade!

[@julianp.2163]

16:16 Simon: „i dont know“ My brain: none at all, they are forced to move in one 3 by 3 case each in column 1,2 and 3. otherwise there must be two crossing the cases borders in that columns otherwise numbers would repeat in a case. But the sandwich sums in row 3 (21) and row 7 (26) dont allow that in a distance that short.

[@jaeusa160]

@9:00 Something interesting I noticed is that you have a... I don't know what to call it. Not an X-Wing. But in columns 1-3 of the grid, the 1s and 9s in the columns must all be adjacent. In effect, all the 1+9 pairs in boxes 1 4 and 7 of the grid are accounted for "like" an x-wing in that those 3 boxes must contain exactly three 1+9 pairs adjacent to each other in columns 1 2 and 3. If you could rule 1s and 9s out of r3c1 and r3c2, you could at least narrow column 3's 19 pair down to only being in box 1. But the 21 clue in row 3 does that. You must place a 1+9 in column 3 of the grid and it must be contained within boxes 1 4 or 7 of the grid. If you place a 1 or 9 in either of r3c1 or r3c2, it rules a second 1 or 9 out of the entirety of box 1, which breaks the box. You can't make 21 in less than 3 digits, which would rule it out of r3c2 and r3c3, and r1c3 and r2c3 would be invalid for a 1 or 9 without an adjacent partner in column 3. But you can't put three 1+9s in box 1. So all that is a very long way of saying column 3's 0 clue sandwich must be entirely in box 1 of the grid, and that r3c1 and r3c2 are not sandwich crusts. I feel like this was probably an intended early inroad into the puzzle, but as is often the case, there can be many paths to the same destination seen differently by many solvers.

[@glennmelven3414]

@@jaeusa160 If you look at columns 1-6 you will see that they are different from 7-9. 1-6 has a 358 triple whereas 7-9 has 589. So if you saw those triples existed in 1-6 and tried to use the in 7-9 you would break the puzzle.

[@jaeusa160]

@@glennmelven3414 If I made a mistaken then away it goes.......... oh wait, I shouldn't have deleted. I never said anything about column triples. The triples are only in rows.

[@Alex_Meadows]

Downloading as I type this comment! Very excited, I've been looking forward to this. If the other apps are anything to go by, I'll get up to the 9/10 puzzles and then grind to a permanent halt. I've always suspected that Mark was the more evil of the two, so it's nice to have that prejudice confirmed.

[@Alex_Meadows]

To clarify: I love watching both of them and I don't really think that Mark is evil. It's just that if, for instance, I were to place two 7s in a column, Mark would be annoyed with me while Simon would only be disappointed. Actually, now that I think about it, disappointing Simon would be much worse...

[@btestware]

I suspected you couldn't make 21 in 3 on account of the non-consecutive constraint, and worked pretty hard to convince myself it was correct - mostly by trial and error. And then Simon glanced at it, and in a moment he exclaimed: "where do you hide the seven?" Perfect logic!

[@RaghavendraBommaraju]

What's the song Simon plays at the end on his guitar? It's so familiar but I'm not able to recollect it.

[@stevieinselby]

Mothers of the Disappeared, by U2 (from The Joshua Tree)

[@RaghavendraBommaraju]

@@stevieinselby I don't know why it seemed familiar. I hadn't heard it before. It is beautiful. Thank you!

[@MichaelJPartyka]

I honestly did not think that I was going to get this one. At the 9-minute mark I thought, "Nope, that's it, there's something in here that's just beyond me." But something inside told me to keep plugging at it. It took me over an hour, but I solved it all by myself, no hints needed.

[@100beep5]

5:35 it would, in fact, be impossible to fill the grid, as whatever digit is in the centre of a box would rule out it's consecutive numbers from the whole box...

[@zackaryburns3266]

The rule states that any digit that shares an edge can't be a consecutive number, numbers that share a corner do not follow this rule.

[@moerkx1304]

@@zackaryburns3266 Well the thing you aren't getting is that Simon says it would be quite hard to fill in the grid if the consecutive restraint were to be applied to corners and Cory Mc just commented on that. And if you had read their comment thoroughly you would have seen they wrote "it would", indicating that they know that this isn't the case.