I was a bit worried, when I saw this puzzle appear on the channel. I thought it was extremely difficuIt and maybe too much for a solve on camera. But of course Simon killed it!
idk if you remember me. im picoton, the guy who asked for help on lmd. honestly, seeing simon bifurcate at ur puzzle made me feel a little bit better about myself. what a tough nut this was. truly a remarkable setting.
My man is out here solving a blank sudoku with one of the most complicated rulesets to wrap ones head around and is apologizing because he needs time to think. You have my respect for just solving this sudoku but the fact that you try to appease the 0.01% of people who are able to advance faster than you is just such a nice touch.
You said at one point that if you weren't making a video you'd have whipped out pencil and paper and taken notes about the thermos - I can't speak for everyone, but personally I would love seeing stuff like that in future videos, either via a document reader or some MS Paint doodles. Doesn't much matter if it's a bit janky, it'd still be fun in my opinion!
Update: Okay I whipped out some graph paper to do some working out, and after your logic with 8s and even/odd thermo length, you can solve all thermo identities (except for 1 length) immediately. From your logic: 8: 87654321 7: 7654321 6: 876543 4: 8765 2: 87 Each even digit is in exactly 4 thermos, removing 6 as a possible digit on the remaining odd length thermos thermos. That forces the 5 length thermo. 5: 54321 4 is now on 4 thermos and this removes it as a possibility from the 3 length thermo. 3: 321 This places 7, 5, and 3 on 5 thermos each and 1 on 4 thermos. From this we can conclude that 7, 5, and 3 must be in corners, and the 1 length thermo has the same value as the remaining corner. Team paper!
I can’t tell if he’s not getting a paper or notepad out for audience reasons (which he apparently should reconsider), or to challenge himself (which I love to see and wouldn’t want to take that from him!)
@@Michael-kp4bd At 25.30 he says if he were not doing a video, he would get a notebook out, so it seems to be definitely the former reason, and I do hope he considers doing it in future.
I think it would be worth some experimentation with this idea, as there are clearly some Sudoku puzzles that require a bit of thinking away from the grid. Perhaps an extra camera trained on a section of desk with paper, pencils and Scrabble tiles, for quickly visualising possible combinations of digits?
Simon, please do NOT make me glare at you over the internet. The fact that you were able to solve this at all without using pencil and paper to sketch out possibilities... is amazing. The polar /opposite/ of "too stupid". You're dealing with a technical issue: you need to have a way to write things off the grid and have them be visible. Whether this is a notepage file that you drag into and out of frame, or a piece of paper with a second camera (possibly your phone?) pointed at it... either will work. When you have a moment, please see if there's a way you can work with the tools that work for you. (And yes, that's easier said than done, I know.) Because we also enjoy your comments about the puzzle and the solve - especially when they're kind. And that includes you being kind to yourself.
Wow, language police. Just let Simon talk through the puzzle in the way he thinks best. I don't see any problem with him being humble about it. Actually I think it's commendable.
Actually you're just "woman-splaining" by trying to force Simon to talk the way you would. Men think differently and tend to sharpen their skills best by being harder on themselves. You don't understand this and want to shut it down. It's healthy.
Hey guys, I am the 40-year-old Thomas mentioned at the beginning of the video and I am proud that I got my own "Happy birthday" section in one of your videos. Many thanks in particular to my brother Martin who apparently made this possible. I am touched. However, I am still waiting for the promised chocolate cake. ;-) Simon and Mark, please keep up the fantastic content. With your logically comforting manner, you are a real enrichment for the stressful everyday life. No matter how long your videos are. ;-) Many greetings and thanks again, Thomas
I was able to easily solve this puzzle in under 20 minutes by staring dumbfoundedly at the blank grid for 19 minutes then skipping to the end of the solve and copying Simon's solution.
And the word "Blue" has an even number of letters in it; this is true not just in English, but also in French (bleu), German (blau), Spanish (azul), and Portuguese (açul)!
Very nice solve Simon ! Next time when you’re attempting such puzzles, please use a notepad (or excel etc) for you to keep track them (and not just try to memorize them all). It’s easier for you and for us viewers to visualise all that.
I honestly love witnessing Simon’s ability to keep it all in his head, use the software as he sees best, and brilliantly pushing through. Just my preference, and it’s also fun that Mark’s play style is so different, so I get to see two geniuses handle things differently. Just my preference though, I totally admit it’s me simply not wanting him to change 😄
@@Michael-kp4bd Yeah I absolutely agree. It’s great to see such great ability to consider and memorize all of them. But sometimes it’s too tedious and for us viewers we can’t grasp them all at once. So imo with a notepad on screen we also can visualise what he’s doing and considering and follow through with the process.
1:15:00 I felt for Simon here, as this was literally one of the first deductions I made before even writing in any numbers. After realizing the four 8's on thermos are 1-8, 3-8, 5-8 and 7-8, you can also similarly deduct that he four 2's on the thermos must be 1-8, 1-7, 1-5 and 1-3. So you immediately find the distribution of all the thermos except for the single cell. The single cell is odd, and is either 1 (which we have so far used only four of on our thermos), or whatever duplicate occurs in the corners.
1:38:34 My path through this was: - The F in column 2 requires an adjacent 7 on its thermo, so R8C2 can't be a 7. - This puts a 7 into R8C6, since the corner is already ruled out. - Now you consider where F goes in C7. Since F requires an adjacent 7 on its thermo, F can now only be in R3 and connects to R2C7
I was just solving this puzzle recently and thought "this break-in is too difficult, probably won't see it on CTC" so kudos to Simon who did the impossible I guess. time to dig in!
I kept hoping Simon would notice that his 4 8s logic applies to all the even digits. So that means that his 4 odd thermos need to start with 1s so you can put 4 2s in the puzzle (except the 1 cell thermo)
Well no worries, because once he sees it, he calls himself “So stupid” many times in a row for not recognizing this sooner 🤣 I quickly missed that it could be helpful for any digit besides 8, props to seeing it!
I completely agree he should not be so hard on himself. But I will say I spent 45 minutes pleading with him to ask the question “how many 2s are in the middle”. He never seems to listen to me from the past when I talk to him from the future.
0:55 This immediately makes me think that, because 36=1+2+3+4+5+6+7+8, those are the thermos' lengths, and their differences accordingly 0, 1, 2, 3, 4, 5, 6, 7. ... And I see that Simon thought the same. But gosh it became practically unsolvable from there. Simon, I admire you for your solving ability and your patience to stick at this beast for nearly 2 hours.
@@NichtcrawlerX From the rules: "The difference between bulb and tip must be unique for each thermometer." The maximum difference is 7, and that can account for a maximum of 8 of the middle 36 cells. The next biggest difference is 6, which can account for up to 7 more cells. Then a difference of 5 takes up up to 6 more cells. And so on, until a difference of 0 takes up a maximum of 1 cell. So the maximum number of total cells covered by thermos is 8+7+6+5+4+3+2+1=36. According to the rules, each of the 36 non-edge cells must be on a thermo, so 36 is also the minimum. Therefore, the above maximums must be the case - there must be a thermo of each possible difference (0 to 7), and each thermo must take up its maximum possible number of cells (1 to 8).
I am currently watching, the next step could be where to hide the 8 and the 1. A thermo of length 7 should drop one 1 or one 8 on the perimeter, etc. for shorter thermo. But you can only hide 4 of each on the perimeter. From the 16 evens and 12 odds on the perimeter, I think it means there are four 8 hiding on the perimeter, so the length 7 thermo is 1-7. OK I was too lazy to think any further. I did not see the odd/even length. After having my breakfast, four 8 on the thermo means all the even length thermo (as explained by Simon) end with 8, they are 1-8, 3-8, 5-8 and 7-8. On the perimeter there must be three 1 and 2, two 3 and 4 and one 5 and 6. So there are 6 odd and 6 even unaccounted for on the perimeter.
at around 27 to 30 minute mark, Simon has deduced a 1-8, 1-7, 3-8, 5-8 and 7-8 thermos. If there are four 8s in the sides, there are four 8s in the middle. This is the same for all other even digits. The above given only shows two 2s and three 4s and the only way to get both 2s is adding a 1-5 and a 1-3 line, leaving only the single box thermos unknown.
for anyone who did not enjoy the 10-minute bifurcation chain, I have a gift for you! at 1:23:35, what Simon failed to appreciate is that r3c6 is now a C, and C must be connected to 3 and D, but r3c7 can be neither. by the same token, C is also not in r3c7, and must be in r3c8. you can attempt to solve from this point.
even after he failed to realize this, his long bifurcation chain proved very simply that there had to be a 7 in one of r6c2 or r7c2, which placed a 7 in the bottom row, which he also could have used w/o the entirety of the bifurcation chain
@@RichSmith77 There is some stronger logic, C cannot go in the domino because it can't connect to the cell above it (7 or 8) so it must connect to the other digit in the domino, but it can't be 3 or D.
I know it's been said several times but watching you figure out those possibilities without pen and paper was amazing! And your ability to explain it in a way that is both interesting to watch and understandable to the viewer is phenomenal. The logic of this puzzle has been one of a kind for sure and you cracked it! Well done!
Haha this is amazing! It's the puzzle I did not dare to recommend because I imagined Simon being stuck and sad.... and then he solves it anyway! I'm looking forward to the next hour and a half. :)
Thank you so much for the birthday wishes! My brother was very happy and he promised to comment on this video when he's done with partying tonight. :) Have a wonderful weekend!
Oh Simon please DON'T ever scold yourself stupid.... you are by far the most brilliant and eloquent solver that makes my day bright on this monstrously hard puzzle with you juggling all those combinations in your head ... Please take a bow yourself as u have a powerful powerful memory brain. Love yourself more Simon.... Great solve all around again. 👍👍👍👍
Bloody heck, this has got to be one of the, if not The, most impressive solves I've seen from Simon. Keeping all that info in his head, wrapping his head around it all, and then starting to fill in.
Rules: 03:10 Let's Get Cracking: 06:24 What about this video's Top Tier Simarkisms?! Three In the Corner: 6x (55:47, 56:21, 56:33, 56:57, 1:24:55, 1:43:28) Bobbins: 3x (26:24, 1:26:58, 1:42:02) Maverick: 3x (02:42, 03:04, 03:07) And how about this video's Simarkisms?! Sorry: 22x (04:44, 14:43, 25:30, 25:30, 28:05, 34:51, 34:51, 40:34, 41:11, 42:23, 45:36, 49:11, 54:14, 58:18, 1:00:34, 1:02:23, 1:03:39, 1:03:39, 1:03:39, 1:14:55, 1:15:13, 1:22:31) Hang On: 12x (25:12, 36:16, 46:53, 46:53, 1:17:25, 1:33:03, 1:33:06, 1:33:11, 1:39:39, 1:42:37, 1:43:21, 1:44:42) Ah: 11x (28:54, 56:21, 59:29, 59:29, 1:00:17, 1:09:30, 1:20:10, 1:24:35, 1:48:03, 1:48:33, 1:49:24) Wow: 9x (08:32, 23:35, 23:35, 34:41, 1:05:34, 1:06:22, 1:06:35, 1:08:59, 1:52:13) By Sudoku: 7x (1:01:45, 1:04:06, 1:06:38, 1:06:54, 1:32:03, 1:43:16, 1:44:47) In Fact: 6x (45:00, 46:36, 1:23:39, 1:43:16, 1:46:29, 1:47:46) Obviously: 5x (03:24, 33:50, 51:54, 1:00:34, 1:39:10) Unique: 5x (05:14, 05:20, 09:10, 1:12:17, 1:16:22) Nature: 4x (08:56, 33:36, 1:05:46, 1:27:24) Lovely: 3x (44:47, 1:21:28, 1:51:23) What on Earth: 2x (13:28, 34:41) The Answer is: 2x (21:59, 1:20:25) Beautiful: 2x (21:45, 1:45:33) Brilliant: 2x (01:59, 33:45) Deadly Pattern: 2x (1:48:32, 1:49:36) Come on Simon: 2x (54:41, 1:01:36) Shouting: 2x (01:35, 01:55) Surely: 2x (30:11, 30:11) Let's Take Stock: 2x (1:12:31, 1:12:34) What Does This Mean?: 2x (56:25, 1:28:55) That's Huge: 2x (1:11:31, 1:29:36) Pencil Mark/mark: 2x (48:11, 1:08:45) Useless: 1x (45:16) Bother: 1x (26:11) Nonsense: 1x (46:40) In the Spotlight: 1x (1:43:30) Break the Puzzle: 1x (1:34:58) Fascinating: 1x (16:02) Incredible: 1x (02:52) Gorgeous: 1x (1:20:10) Masterpiece: 1x (1:52:03) Whoopsie: 1x (37:37) Phone is Buzzing: 1x (1:04:43) Scrabble Tiles: 1x (37:12) The Power of Maths: 1x (1:18:54) Thingy Thing: 1x (1:26:38) Cake!: 1x (01:51) Symmetry: 1x (44:57) Most popular number(>9), digit and colour this video: Twenty Eight (8 mentions) One (190 mentions) Blue (4 mentions) Antithesis Battles: High (2) - Low (1) Odd (98) - Even (89) Higher (2) - Lower (1) White (2) - Black (0) Row (15) - Column (7) 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!
At 1:28:36, this bifurcation mess can be avoided by finding that F in column 7 can only be in r3c7 because the 7 in column 6 can only be in row 7 and 8. From there, this puzzle becomes much more elegant to solve
My juggler friends and I, somewhat imfamously, often think up a new, complex multi-person juggling pattern, then spend half an hour talking about it, trying to understand how it works and what the implications are, before any actual juggling takes place. I can't think why I'm suddenly reminded of that here. 🤔
1:23:39 It's easy to prove that R3C6 is not even. Its value is 2or4, but its corner mark is not B, but C. (by sudoku) And as we see in R2C4 C is connected to 3 and to D. But R3C7 can't be either (by sudoku, 3 is in the region, D is in the row) It means that our R3C6 cell has nothing to connect to. And that means it's not even. Thus it's odd :)
Logically, I got as far as figuring out that there couldn’t be any thermometers in the perimeter. I could follow your logic, once you pointed it out. I just couldn’t get there myself. Great solve. Watching you solve puzzles makes me a better puzzle solver, and I can’t thank you enough for that!
Simon...just absolutely more incredible solving by you!! Loved how you approached this ruleset! Thank you as always for showcasing brilliant setters, rulesets and puzzles on a daily basis
Simon, what a video - epic! Your reasoning and logic at the beginning to work out how the thermos had to work was mind-blowing. Thanks so much for this video and for doing everything that you do to make sure that we get two sudoku videos every day - not to mention cryptic crosswords! (I also loved the little side trip into Oliver!)
That was crazy, I tried it for 30 mins getting no clues, so I just reverted to the channel and really enjoyed Simon cracking it. I almost felt the pain
I think you need to add a notepad to the side of the sudoku app. It would have been quite helpful here. Knowing that the even-long thermos all end in 8, so 78, 5678, 345678, and 12345678, and there's 1234567, if you wrote it down you'd quickly see that it accounts for all four sixes. So the 5-long thermo can only be 12345, and now all fours are gone too. So the three length thermo is 123, with one final odd digit being the last 'one length thermo'. But I think you need to write it down to see that.
It was just about possible to use Mark's favourite, of using an empty area of the grid (the centre) for "aide memoire" pencil marking. I entered the first and last digits into a cell to denote the full length of the thermo, knowing it also had all the digits between. So 17 stood for the 1234567 thermo. This way I was able to reach the same conclusion, but without pencil and paper. However, I then needed to write them down on paper anyway, when it became necessary to delete my aide memoire 😂.
Simon did say it was 100% on LMD. Unfortunately, there are trolls on LMD that actually watched this channel and watch just to get the answers so they can pad their scores. That wouldn't be too bad but then this certain troll marks the puzzle down in the rating. This puzzle now has an 88% approval rating because of this certain troll. It just shame that LMD allows this person to continue manipulating the approval rating. On top of this this troll is in the top 10 of solvers. It is pretty sad that a person like this having no honor is allowed to take away the joy that setters get when they see how well their creation is doing. I'm not going to name the person but he watches this channel and others so he can get the solution code and continue the cycle of jacking up his ratings while knocking down the setters rating. A true troll that needs to get evicted from LMD.
Is there a way to report them on LMD? It seems to be pretty obvious that, at least for this puzzle, that's exactly what happened. Such a shame that someone would do that, but hopefully there's a way to stop them.
You are very honorable not to mention the name of the person. Too bad the person does not have the same sense of honor not to do this. I don't understand people sometimes. No, that's not right, I very often don't understand people!
I think Simon needs an iPad that he can project over the puzzle when he’s doing these very complicated conditions and sets to better explain how he’s getting there. I was able to follow, but it would be amazing to see it drawn out to see the beauty in the analysis ❤
I appreciate how you take the needs of your colour-blind viewers into consideration when choosing colours. With that sort of thing in mind, please consider that it is difficult and unpleasant for some viewers to hear you call yourself stupid, or to apologize for taking too long on a puzzle. It is from a place of compassion that I make this comment. Please be kinder to this person I care about (i.e. yourself)!
He started talking dumb things about himself since 2021, we have worked hard not letting that continue and it have never worked. I think he's taking comments like this as sarcasm and doesn't understand that nobody (that includes us viewers) is perfect.
Is this a record? First pencil marking at 48 minutes (minus birthdays)? But I was flabberghasted by the fact that any pencil marking was possible! Simon you are a genius!
30:00 I believe that it can be stated around this point that there must be not only 4 lots of the digit 8 in the central cells, but also 4 lots of the digits 6, 4 and 2. And this should fix most of the odd-length thermometers to be 1-7, 1-5 and 1-3. Correct? Only about the single-cell thermometer I am not sure...
Yeah, I noticed that as well. I don't think one can nail down the single cell thermo at that point other than that it needs to be odd. However, the 7 known thermos fix at least one each of 3, 5, and 7 in the corners, and then the fourth corner has to be the same digit as the single cell thermo because otherwise there would be too many of that digit in the grid.
Hats off to you Simon, I am a very competitive person, and keep challenging myself to reach certain conclusions faster than you, some puzzles I see stuff before you do and some I don't, but the one thing that I NEVER EVER win at is how fast your brain finds proper grid divisions to mark cells that are equal to each other... You are just pure genius.
I like to imagine that in his previous life in financial services, Simon also insisted on doing all calculations in his head without recourse to paper, pencil, or computer. 😀
51:37 Funny how he does all this thinking, instead of just realizing 7 is in r3,c3 because of the 1,3,5 triple he has penciled in the ling "snake" region which contains that cell, leaving 7 as its only option.
For puzzles like this I would not mind if you had a second screen where you took notes, and scrolled to it occasionally. It's incredibly impressive you can keep all of this in your head, but I certainly cannot, so I just had to trust that you knew what you were talking about, rather than looking for deductions myself. Great solve :)
Hey Simon, thank you for showing the law of leftovers again and in detail. This was a point where I was stuck. I think pen and paper or a second screen with an excel-sheet would have been a great help for you. I managed to figure out the nature of the thermos quite well after visualizing my knowledge about the thermos. Even more impressive though that you got there without any tools. Kudos!
This is only my 3rd video this year because university is stressing me out 😪 but I'm so glad I'm here to see Simon solve this wonderful Sudoku and to read your wonderful comments, you awesome people.
27:00 Dont we know all thermos at that point? There are 4 8s 4 6s 4 4s 4 2s in the perimeter which means there are 4 of 2/4/6/8 in the middle. The only thermos that can pick up the 4 8s are the even thermos which means they have to be 12345678 345678 5678 78 Which also means we have used up 4 8s 3 6s 2 4s and 1 2. Now the odds the biggest one can only be 1234567 so all 6s are gone which means the 5 length thermometer can only be 12345 which is the only way is does not pick up another 6. Then the other two have to be 123 and 1. That finishes all thermos. Did i make an error there? Sorry for my bad english by the way. On a second thought i dont think the single cell thermo is not forced to be a one because no even digit is involved but the rest should be correct. But it is a 1/3/5/7
This. Was. Awesome. I kept thinking I was to the end of the possible deductions - with parity, number combination, everything - and I just kept finding more! I don't think I could have done it without a pen and paper, but Simon's brain is remarkable. Beautiful puzzle. Brilliant.
Remarkable puzzle which I would not have solved in a decade of Sundays without some prompting from Simon. How he does it all in his head I will never know. Ditto how anyone can concoct a puzzle like this!
I've done very little of the puzzle so far, barely started and went to bed, but just from thinking about it at my work I can tell you it is amazing, the sheer amount of information you can extract essentially just from the corners is stunningly beautiful
Stubborn as I am, I had to solve this before watching your solution (that's still how I check my brain is still working), so that's where my afternoon went. And as I started watching yours, I thought "I admire Simon can do these puzzles without pencil and paper and with very little marking"... and I confirm that! Absolutely brilliant solve! But I spent almost an hour on the same place where you got stuck, and I'm kinda pleased to see there wasn't anything evident that I missed, it was just very very hard.... Kudos!!!
37:18 Not sure if it would have helped getting to a solution any quicker, but my path through this was to use the green cells in R3 (let's call Group A) to make two groups either side: R3 C1/C2 (Group B) and R3 C6/C7/C8 (Group C) Then because Group A, B and C are all in diff regions the cells in R1C1 + R2 C1/C2/C3/C4 have to be Group C + 2 of Group A (3rd is already in R4C2), and R2 C5/C6/C8 have to be Group B + 1 of Group A. Then because of the Group A digit in R4C2 - we now know that R4C8 & R5C8 have to be the same as the two Group A digits in the R1C1 + R2 C1/C2/C3/C4 block, and one of R2 C5/C6/C8 has to equal R4C2 Still some shenanigans after that, but less i think 😊
U don’t have to apologize for not figuring it out quickly. I watch because i enjoy watching the big revelations, and the longer I have to wait, the better!
I know we all deal with distractions differently, but there's not much I like more than the sound of small planes flying by. For me, it's an absolute highlight of your videos!
This one was hard ... but I got it. 231:20, solve counter 283. I started out with coloring the whole grid (which turned out to be not very fruitful, but bring some results), then did some logic (including a tiny bit of set theory) to figure out which thermos need to exist (which didn't need that coloring at all - except for the fact that two of the corner cells had the same color), then tried to combine both of them.
Crazy how fast he realized there are four 8’s in the center and how long it took him to consider that there are also four 1’s… would’ve taken me ages to get the former, but having gotten it the latter would’ve followed instantly for me😂
Very proud of myself for deductions on this one. Simon shot ahead and came up with a whole load of complex deductions that it took me forever to understand! But pretty soon after he worked out that there were four 8s in the middle, I deduced the nature of all the thermos, by considering the quota of other even digits. I haven't finished watching yet so I'm not even sure whether Simon works that out or solves it another way! I definitely wouldn't have been anywhere near a break in on my own!
Amazing, bonkers sudoku puzzle! Mark often uses the sudoku grid to log down his thoughts almost like it’s in the margins which very often proves so helpful in his explanations. Simon please feel free to write it all down on the grid rather than try keeping it all in your head, which is admittedly way more impressive and functional than mine! It really isn’t bifurcation. I know I’d be hugely grateful to see you scribble explanations down on an empty grid. Thank you for this fantastic channel. Sheer enjoyment.
Always appreciate you picking colors that work for colorblind viewers! My Dad was red-greed colorblind so I've always been aware of friendlier color combos.
Bifurcation was not necessary, but sometimes the next steps were hard to find. Also Simon, two things - first, please take notes if it will save your sanity, and second, if you are stupid, what are the rest of us?
1:03 I think we expect there to be exactly three 5s in the perimeter; based on the original logic about the makeup of the thermos, there are exactly five 5's on them. Well, unless the one-cell thermo is a 5, but I think that has to be a 1; otherwise there are only four 1s on thermos and therefore four 1s on the perimeter, which would require that none of the corners is a 1 and whatever digit is repeated in two corners would only appear twice, which is not enough.
Simon in multiple videos: "If I wasn't doing this puzzle on camera I would do X." Do it man! Do what you need to do! Maybe just upload a picture of your work and put it on screen when you're doing it. Or open up a word document, or paint to do your notes in a way we can see.
This was a classic example of "the solve would be better if you had a notepad window" ... there's just too much for you to try to keep track of in your head here, let alone for _us_ to try to keep track of while watching ... pretty much no-one is going to try to solve this without keeping an aide memoire, there's no shame in doing that, and *it would make the solve easier to follow and more enjoyable to watch.* And then we wouldn't be treated to the spectacle of you berating yourself for temporarily forgetting one of 20 or more complex deductions that you had made previously, which (and I mean this in the nicest possible way) is absolutely ridiculous!
I solved it... with a notepad to remember all of the thermos. im very impressed at how simon could do all of that in his head. without simon's insights that there are 4 8s and 4 2's, and the only way to even out the digits is to have 4 double odd ended thermos and 4 balanced ones, forcing there to be 4 thermos that end with 8s (and start with once of each odd digit), 3 that start with 1 and end with every odd digit (the 4th 2 is in the 1-8 thermo) but 1 plus one odd digit in the length 1 thermo The way i deduced everything is by noticing that there are exactly 4 of each even digit in the center, and that the 4 long thermos (length 5-8) have to contain a 4, using the all up. the length 4 thermo cant use a 4 so has to be a 5-8, using up the last 6 (the other 3 have to be in length 6-8 themos). the length 3 thermo needs at least 1 even digit, but can't use 4 nor 6, so has to be a 123 thermo and the length 5 thermo can't use 6, so has to be a 1-5. these use up all the 2s. that leaves 2 length thermo as a 78 pair and the 6 length as a 3-8, using up the last 8. this makes the length 7 thermo a 1-7 and you only have odd digits left to place in the border + the single cell thermo. the test at 1:23:38 was easy to break. like simon said, r3c6 is c, and c is surrounded by 3 and D (in r2c5), but r3c7 nor the 2 even cells can be either, so r3c6 isn't c, and so not even.
9:47, triangular number for 8 is 36. There are 36 squares in the center of the grid which must contain thermos. Every thermo in the center of the grid must contain consecutive numbers. All corner digits of the puzzle are odd. Each thermo must have either equal odd and even digits or 1 more odd digit than even, the 4 corner cells being double counted decreasing the odd count around the perimeter by 4. Also, the irregularities of grid make r3c345 equal at least two odd digits, and r2c4, r3c4, r3c5 comprise of at least two even digits, proving r3c3 and r6c3 odd and r2c4 and r4c2 even. The single digit thermo is tricky because we are left with 4 1s, 3 3s, 3 5s, and 3 7s, where if every corner is different, the 1 is the 1 cell thermo or if there is a repeat in a corner, that is the 1 cell thermo, with 1s never appearing in a corner.
I did take Simon's early thoughts about working out the possibilities on pencil and paper, which helped me greatly in starting the solve! I think it would be nice to figure out a way to use this kind of outside-the-grid work in these videos, because it's very helpful in puzzles like these. Without finding any digits, I discovered the makeup of the largest seven thermos, and also discovered that there were 357 in the corners, along with whichever of 1357 made up the lone cell thermo. All of this I figured out just because I knew there could be no more than four of every even digit in the 6x6, along with the fact Simon also discovered about the four 8s required to be in every even length thermo.
The 28 digits in the perimeter are four sets of the digits 1-8 less the corner digits (which have effectively been counted twice: once in their rows and once in their columns). The 36 digits that appear on the thermometers must therefore be four sets of 1-8 plus the digits in the corners -- which we know are odd (from the grey circles) and cannot include more than two of any digit (because each one sees two others). We also know the thermometers must increase strictly by single steps, just to swallow up all the available digits. The biggest possible difference between bulb and tip is 7 (for a 12345678 thermometer) and the smallest is 0 (for the single-cell thermometer). This allows a maximum of eight thermometers, which must account for 36 cells. The minimum length of a thermometer with a non-zero difference is 2; the maximum length of a thermometer is one more than its difference. Since 1+2+3+4+5+6+7+8=36, all the thermometers must increase strictly by single steps; otherwise, digits will be left out that cannot possibly be part of any thermometer. Now knowing that the eight different-length thermometers are formed from four sets of 1-8 plus another four odd digits, we can consider how they might fit together by pairing them off to create total lengths of nine; which must consist of four complete sets of 1-8 with an additional odd digit. Clearly we need to go to both extremes within each pair, to avoid having too many "middle" digits (2-7) and not enough "outside" digits (1 and 8). If we tried something like 12 and 2345678, we would need an extra 2, which is not available. 123 and 345678 is a set of 1-8 plus 3. 1234 and 45678 needs an extra 4, so is impossible. 12345 plus 5678 is a set of 1-8 plus 5. 123456 plus 678 is not possible. 1234567 and 78 is a set of 1-8 plus 7. 12345678 could actually go with any other odd digit, tempting as it might be to assume this ought to be 1. This fixes the thermometers as being 12345678, 1234567, 345678, 12345, 5678, 123, 78 and some other odd digit; and all except the one-cell thermometer must either start from 1 and finish on an odd number, or start from an odd number and finish on 8. The corner cells are 3, 5, 7 and the digit on the 1-cell thermometer. All this is a lot to try to explain in words, but quite easy to see if you have some props -- such as numbered tokens, or even just pencil and paper -- to try it with. But of course, introducing sequences involving props into a CTC video would require new filming and editing techniques .....
For the hard part: (i hope this is logically correct and helpful): - if there's a 6 or 8 in R6/7C2 there needs to be an accompanying 7 - if it's a 6/7 pair, it needs to be 6 in R6C2 und therefore a 5 in R6C3 - therefore D is 5 - since all thermos with an odd number as their highest digit need to have an odd number of cells the thermo in R2/3C4/5 needs to end in a 1 - the 1 can't go into R4C5, because if it were in R3C6 the even number in R3C7 wouldn't be able to exist (can't be an 8, because in this scenario G is 8, and can't be anything else, since it's thermo couldn't exit on both sides) - All of this leaves us with a 1 missing from R3, which can't connect to a 2 in it's collumn or row, therefore breaking
WOOOOW what a puzzle. took me close to 3 hours to do on my own. I also went back later just for fun and checked if it was gonna be easier, knowing where the thermos were... it ONLY disambiguates if you respect the rules about unique distances on the thermo (other than that its quite easy to solve if you know where the thermos are) either way INCREDIBLE
It took me a *lot* longer, but I was able to solve it after a couple of hours. I started with the same deductions of odds and evens inside the middle grid. Then I moved on to the thermometer compositions. Looking at the correct number of odds and evens, I believe all of them are forced. Except the 1-length. Then either there are no repeats on the corners, which pushes the 1 to be the 1-length. Or there is one repeat, which can be the 3, 5, or 7, making that the one-length. There can't be 2 repeats. I was able to show next that there couldn't be 4 different corners, pushing the 1 out of all the corners into the sides. And requiring exactly one repeated corner. I believe that would have helped Simon solve the '78' in r67c2, since the 1 would be required in r3c45. Awesome sudoku and great to see Simon make 3 hours worth of deductions in about 12 minutes.
Simon, you are definitely too hard on yourself. I envy your prowess with maths and those digety thingy things. I love sudoku but suck at math. And you are not a silly man. ❤
As of 13:30 in the vid i came to notice that, in the empty 6x6, theres eight groups of empyy cells separated by lines, containing 1,2,3,4,5,6,7,8 cells. Totaling 36.
I am not sure I could have solved this one no matter how long I stared at it. I got the parity/lengths/totals bits but trying to figure out where to go from there was beyond me at least for long enough to watch this and just enjoy the solve. You have a dark mind Nahileon. 😈
I was confused by the need to have only single digit rising thermometers until I finally just counted it out more distinctly. With the 36 squares in the middle, you need a maximum sized thermometer for each difference between bulb and tip, which means each length of thermometer from 1-8 cells must be present to give 36 squares. 8+7+6+5+4+3+2+1=36.
I am just getting started with solving such difficult Sudoku puzzles and it took me super long to wrap my head around this and i had to peek into Simons solution some times because I got stuck but in the end I managed to solve it and I absolutely loved this amazing puzzle!
This is an impressive puzzle... And seeing it done... Well, my attempt to create an invisible thermo puzzle feels... less terrible that I didn't get all that far with it. Thank you Simon, Thank you Nahileon.
Had to watch to 1:43:50 to solve. Did not really understand how to approach this one at all, just used sudoku in the end to populate the rest of the grid knowing that Simon wasn't ever skipping digits on the thermos. The difference counts on all the final thermos are all different, so I think I got it, but no way would I have been able to finish this without all that help. No shame in that though, was still great fun :D
You might have saved yourself about a half hour of time if you had written down the thermo combos on paper - took me less than 5 minutes to figure out. However I am impressed that you kept them all straight in your head.
103:24 for me. I came up with the break-in quite quickly, EXCEPT I wrongly assume the tricky 1-cell thermometer which could be any odd number is 1. I have to spend at least 20 minutes to find my mistake. Nevertheless, this is really a stunning puzzle.
This one was one of a series of very vexing puzzles for me; I eventually figured the other ones out, but I failed on this one. I actually got some parts on my own (the part about 'leftovers', the knowledge that all thermos has to alternate odd-even, the knowledge that all odd-length thermos were bookended by odd numbers), but I missed the fact that every even number besides "8" had to connect on both sides. After that, I got it in my head that the four corners had to be all different (I was right about 7 of the 8 thermos, but forgot that the single-cell one didn't have to be "1"); from there, I went a long way before breaking; at that point, I completely gave up. I kinda wish we'd get a video of a puzzle that they couldn't solve, if only to see if I could get one they didn't. Until that day, though, I'll just remain in awe that they could figure something like this out.
Before you went into parity colouring of the grid you instead could have one continuation of the pattern you already detected for the differences. You already concluded you have to have thermometers of each length 1-8 and the rules also only allow each difference of 0-7 once only. The 1-8 thermo uses up the 7 difference, the next shorter thermo you finally concluded has to be 1-7, as the 8 digit has to be reserved for the even length thermos. Well, that's a very special deduction but can be generalized when thinking about the differences you can have with different lengths. What you cannot have is a thermo of length 7 with a gap, as that would go from 1-8 ad have a difference of 7 again, so the 7-cell long thermo has to have a difference of 6 only and be consecutive. Well, the next shorter length 6 now also can't have a difference of 6 as that is taken by the 7-cell-thermo. This ripples down and you can conclude all thermos are also renban lines and have no gaps. I made a wrong conclusion from that, though, that when all even-length thermos end in 8 to have 4 8s in the 6x6 all odd-length thermos start at 1. As you can have a repeat odd in a corner, that can be the single cell thermo within the 6x6 and it doesn't need to be 1. I failed at that. I'm still proud the renban deduction turned out to be correct.
At the start Simon kept saying "If I wasn't on camera, I would be taking notes" and I keep wondering, why not just open Notepad or something? Or just use the 8x8 grid to take some notes? I just opened the puzzle and put some numbers in and very quickly realized that we need a minimum of five 7s in the middle, five 5s, five 3s and at least four 1s. What took Simon a LOT of time and mental effort to sort out (the composition of the thermos) took me about 5 minutes after taking a few notes. Therefore, I think if he had let himself do the same, this could have been an hour long video instead.
I was a bit worried, when I saw this puzzle appear on the channel. I thought it was extremely difficuIt and maybe too much for a solve on camera. But of course Simon killed it!
At the point where Simon proved a 6 was impossible in the domino at the bottom of column 2, was there a different line of thought?
idk if you remember me. im picoton, the guy who asked for help on lmd. honestly, seeing simon bifurcate at ur puzzle made me feel a little bit better about myself. what a tough nut this was. truly a remarkable setting.
Just beautiful. Good work.
Just such incredible setting from you!! Brilliance!! Would love to see a setting video, or share any insights in how you came up with this!
Well done! That was fantastic!
My man is out here solving a blank sudoku with one of the most complicated rulesets to wrap ones head around and is apologizing because he needs time to think. You have my respect for just solving this sudoku but the fact that you try to appease the 0.01% of people who are able to advance faster than you is just such a nice touch.
You said at one point that if you weren't making a video you'd have whipped out pencil and paper and taken notes about the thermos - I can't speak for everyone, but personally I would love seeing stuff like that in future videos, either via a document reader or some MS Paint doodles. Doesn't much matter if it's a bit janky, it'd still be fun in my opinion!
Update: Okay I whipped out some graph paper to do some working out, and after your logic with 8s and even/odd thermo length, you can solve all thermo identities (except for 1 length) immediately. From your logic:
8: 87654321
7: 7654321
6: 876543
4: 8765
2: 87
Each even digit is in exactly 4 thermos, removing 6 as a possible digit on the remaining odd length thermos thermos. That forces the 5 length thermo.
5: 54321
4 is now on 4 thermos and this removes it as a possibility from the 3 length thermo.
3: 321
This places 7, 5, and 3 on 5 thermos each and 1 on 4 thermos. From this we can conclude that 7, 5, and 3 must be in corners, and the 1 length thermo has the same value as the remaining corner.
Team paper!
I can’t tell if he’s not getting a paper or notepad out for audience reasons (which he apparently should reconsider), or to challenge himself (which I love to see and wouldn’t want to take that from him!)
@@Michael-kp4bd At 25.30 he says if he were not doing a video, he would get a notebook out, so it seems to be definitely the former reason, and I do hope he considers doing it in future.
I think it would be worth some experimentation with this idea, as there are clearly some Sudoku puzzles that require a bit of thinking away from the grid. Perhaps an extra camera trained on a section of desk with paper, pencils and Scrabble tiles, for quickly visualising possible combinations of digits?
I agree -- watching Simon's thought process (and Mark's) is fun, and a bit of note-taking on screen might make the expositions clearer.
Simon, please do NOT make me glare at you over the internet. The fact that you were able to solve this at all without using pencil and paper to sketch out possibilities... is amazing. The polar /opposite/ of "too stupid".
You're dealing with a technical issue: you need to have a way to write things off the grid and have them be visible. Whether this is a notepage file that you drag into and out of frame, or a piece of paper with a second camera (possibly your phone?) pointed at it... either will work.
When you have a moment, please see if there's a way you can work with the tools that work for you. (And yes, that's easier said than done, I know.)
Because we also enjoy your comments about the puzzle and the solve - especially when they're kind.
And that includes you being kind to yourself.
This!!!! Hope the kindness you radiate out to the world around you can shine back onto yourself.
Wow, language police. Just let Simon talk through the puzzle in the way he thinks best. I don't see any problem with him being humble about it. Actually I think it's commendable.
Actually you're just "woman-splaining" by trying to force Simon to talk the way you would. Men think differently and tend to sharpen their skills best by being harder on themselves. You don't understand this and want to shut it down. It's healthy.
@@michaellautermilch9185I hope that people you interact with show you the kindness that you show to others
Found the “alpha”
Hey guys,
I am the 40-year-old Thomas mentioned at the beginning of the video and I am proud that I got my own "Happy birthday" section in one of your videos. Many thanks in particular to my brother Martin who apparently made this possible. I am touched. However, I am still waiting for the promised chocolate cake. ;-)
Simon and Mark, please keep up the fantastic content. With your logically comforting manner, you are a real enrichment for the stressful everyday life. No matter how long your videos are. ;-)
Many greetings and thanks again,
Thomas
ich hoffe, dass du einen schönen geburtstag hattest😇
I watched it happen and I still couldn't really tell you what happened... a remarkable puzzle and a well-earned solve by Simon. Wow
Amen!
@@longwaytotipperary Ditto! 😀
@@davidrattner9 😄
I played it twice and still am not sure
I was able to easily solve this puzzle in under 20 minutes by staring dumbfoundedly at the blank grid for 19 minutes then skipping to the end of the solve and copying Simon's solution.
Well done, mate. 💪🏼
A one-minute copy time is pretty impressive.
He had us in the first half….
My kind of solve!😂
6:46 "...and that is because 'O' and 'orange' begin with the same letter."
One of my favorite knowledge bombs!
And the word "Blue" has an even number of letters in it; this is true not just in English, but also in French (bleu), German (blau), Spanish (azul), and Portuguese (açul)!
Very nice solve Simon ! Next time when you’re attempting such puzzles, please use a notepad (or excel etc) for you to keep track them (and not just try to memorize them all). It’s easier for you and for us viewers to visualise all that.
Yes, please!
Mark would absolutely have done that!
I honestly love witnessing Simon’s ability to keep it all in his head, use the software as he sees best, and brilliantly pushing through. Just my preference, and it’s also fun that Mark’s play style is so different, so I get to see two geniuses handle things differently.
Just my preference though, I totally admit it’s me simply not wanting him to change 😄
Yeah I agree, would be really cool to see a notepad cam
@@Michael-kp4bd Yeah I absolutely agree. It’s great to see such great ability to consider and memorize all of them. But sometimes it’s too tedious and for us viewers we can’t grasp them all at once. So imo with a notepad on screen we also can visualise what he’s doing and considering and follow through with the process.
1:15:00 I felt for Simon here, as this was literally one of the first deductions I made before even writing in any numbers.
After realizing the four 8's on thermos are 1-8, 3-8, 5-8 and 7-8, you can also similarly deduct that he four 2's on the thermos must be 1-8, 1-7, 1-5 and 1-3. So you immediately find the distribution of all the thermos except for the single cell. The single cell is odd, and is either 1 (which we have so far used only four of on our thermos), or whatever duplicate occurs in the corners.
1:38:34 My path through this was:
- The F in column 2 requires an adjacent 7 on its thermo, so R8C2 can't be a 7.
- This puts a 7 into R8C6, since the corner is already ruled out.
- Now you consider where F goes in C7. Since F requires an adjacent 7 on its thermo, F can now only be in R3 and connects to R2C7
Thank you. That was a more satisfying path through this section of the puzzle.
Yeah, I did the same logic and then the puzzle resolves once you See that r3c6 must be a 5 to connect to the 678.
I was just solving this puzzle recently and thought "this break-in is too difficult, probably won't see it on CTC" so kudos to Simon who did the impossible I guess. time to dig in!
I kept hoping Simon would notice that his 4 8s logic applies to all the even digits. So that means that his 4 odd thermos need to start with 1s so you can put 4 2s in the puzzle (except the 1 cell thermo)
Well no worries, because once he sees it, he calls himself “So stupid” many times in a row for not recognizing this sooner 🤣
I quickly missed that it could be helpful for any digit besides 8, props to seeing it!
1:15:00 I don't like when simon is so hard on himself for missing something :(
you aren't stupid simon you are brilliant
This one was particularly heartbreaking in this regard 😢
I REALLY wish he would not beat himself up and apologize so much
I completely agree he should not be so hard on himself. But I will say I spent 45 minutes pleading with him to ask the question “how many 2s are in the middle”. He never seems to listen to me from the past when I talk to him from the future.
0:55 This immediately makes me think that, because 36=1+2+3+4+5+6+7+8, those are the thermos' lengths, and their differences accordingly 0, 1, 2, 3, 4, 5, 6, 7. ... And I see that Simon thought the same. But gosh it became practically unsolvable from there. Simon, I admire you for your solving ability and your patience to stick at this beast for nearly 2 hours.
But what actually dictates the 8 Thermos of 8 unique lengths part? I feel like that part of the logic is glossed over.
You are right, at first I didn't question it but now that you ask I don't see why that has to be the case
@@NichtcrawlerX From the rules: "The difference between bulb and tip must be unique for each thermometer."
The maximum difference is 7, and that can account for a maximum of 8 of the middle 36 cells.
The next biggest difference is 6, which can account for up to 7 more cells.
Then a difference of 5 takes up up to 6 more cells.
And so on, until a difference of 0 takes up a maximum of 1 cell.
So the maximum number of total cells covered by thermos is 8+7+6+5+4+3+2+1=36. According to the rules, each of the 36 non-edge cells must be on a thermo, so 36 is also the minimum. Therefore, the above maximums must be the case - there must be a thermo of each possible difference (0 to 7), and each thermo must take up its maximum possible number of cells (1 to 8).
I am currently watching, the next step could be where to hide the 8 and the 1. A thermo of length 7 should drop one 1 or one 8 on the perimeter, etc. for shorter thermo. But you can only hide 4 of each on the perimeter.
From the 16 evens and 12 odds on the perimeter, I think it means there are four 8 hiding on the perimeter, so the length 7 thermo is 1-7.
OK I was too lazy to think any further. I did not see the odd/even length.
After having my breakfast, four 8 on the thermo means all the even length thermo (as explained by Simon) end with 8, they are 1-8, 3-8, 5-8 and 7-8. On the perimeter there must be three 1 and 2, two 3 and 4 and one 5 and 6.
So there are 6 odd and 6 even unaccounted for on the perimeter.
@@steve470 I will print your comment, mark it and hung it on the wall
at around 27 to 30 minute mark, Simon has deduced a 1-8, 1-7, 3-8, 5-8 and 7-8 thermos. If there are four 8s in the sides, there are four 8s in the middle. This is the same for all other even digits. The above given only shows two 2s and three 4s and the only way to get both 2s is adding a 1-5 and a 1-3 line, leaving only the single box thermos unknown.
for anyone who did not enjoy the 10-minute bifurcation chain, I have a gift for you! at 1:23:35, what Simon failed to appreciate is that r3c6 is now a C, and C must be connected to 3 and D, but r3c7 can be neither. by the same token, C is also not in r3c7, and must be in r3c8. you can attempt to solve from this point.
even after he failed to realize this, his long bifurcation chain proved very simply that there had to be a 7 in one of r6c2 or r7c2, which placed a 7 in the bottom row, which he also could have used w/o the entirety of the bifurcation chain
Okay, I attempted to solve it from this point - and I still needed to bifurcate. 😂
(I'm no doubt missing a neat logical step.)
How dare you assert that Simon should so quickly resort to sudoku logic? 😂
@@RichSmith77 There is some stronger logic, C cannot go in the domino because it can't connect to the cell above it (7 or 8) so it must connect to the other digit in the domino, but it can't be 3 or D.
I do hope he covers this logic tomorrow
I know it's been said several times but watching you figure out those possibilities without pen and paper was amazing! And your ability to explain it in a way that is both interesting to watch and understandable to the viewer is phenomenal. The logic of this puzzle has been one of a kind for sure and you cracked it! Well done!
That isn't a puzzle for mortal men to solve. All the props to you Simon for undergoing said tribulation just to bestow us such a video.
Love it how Simon prefers to look at the screen in his mind above the monitor rather than the monitor! He is seeing his puzzle in the air.
Haha this is amazing! It's the puzzle I did not dare to recommend because I imagined Simon being stuck and sad.... and then he solves it anyway! I'm looking forward to the next hour and a half. :)
Thank you so much for the birthday wishes! My brother was very happy and he promised to comment on this video when he's done with partying tonight. :) Have a wonderful weekend!
Oh Simon please DON'T ever scold yourself stupid.... you are by far the most brilliant and eloquent solver that makes my day bright on this monstrously hard puzzle with you juggling all those combinations in your head ...
Please take a bow yourself as u have a powerful powerful memory brain.
Love yourself more Simon....
Great solve all around again. 👍👍👍👍
Bloody heck, this has got to be one of the, if not The, most impressive solves I've seen from Simon. Keeping all that info in his head, wrapping his head around it all, and then starting to fill in.
Rules: 03:10
Let's Get Cracking: 06:24
What about this video's Top Tier Simarkisms?!
Three In the Corner: 6x (55:47, 56:21, 56:33, 56:57, 1:24:55, 1:43:28)
Bobbins: 3x (26:24, 1:26:58, 1:42:02)
Maverick: 3x (02:42, 03:04, 03:07)
And how about this video's Simarkisms?!
Sorry: 22x (04:44, 14:43, 25:30, 25:30, 28:05, 34:51, 34:51, 40:34, 41:11, 42:23, 45:36, 49:11, 54:14, 58:18, 1:00:34, 1:02:23, 1:03:39, 1:03:39, 1:03:39, 1:14:55, 1:15:13, 1:22:31)
Hang On: 12x (25:12, 36:16, 46:53, 46:53, 1:17:25, 1:33:03, 1:33:06, 1:33:11, 1:39:39, 1:42:37, 1:43:21, 1:44:42)
Ah: 11x (28:54, 56:21, 59:29, 59:29, 1:00:17, 1:09:30, 1:20:10, 1:24:35, 1:48:03, 1:48:33, 1:49:24)
Wow: 9x (08:32, 23:35, 23:35, 34:41, 1:05:34, 1:06:22, 1:06:35, 1:08:59, 1:52:13)
By Sudoku: 7x (1:01:45, 1:04:06, 1:06:38, 1:06:54, 1:32:03, 1:43:16, 1:44:47)
In Fact: 6x (45:00, 46:36, 1:23:39, 1:43:16, 1:46:29, 1:47:46)
Obviously: 5x (03:24, 33:50, 51:54, 1:00:34, 1:39:10)
Unique: 5x (05:14, 05:20, 09:10, 1:12:17, 1:16:22)
Nature: 4x (08:56, 33:36, 1:05:46, 1:27:24)
Lovely: 3x (44:47, 1:21:28, 1:51:23)
What on Earth: 2x (13:28, 34:41)
The Answer is: 2x (21:59, 1:20:25)
Beautiful: 2x (21:45, 1:45:33)
Brilliant: 2x (01:59, 33:45)
Deadly Pattern: 2x (1:48:32, 1:49:36)
Come on Simon: 2x (54:41, 1:01:36)
Shouting: 2x (01:35, 01:55)
Surely: 2x (30:11, 30:11)
Let's Take Stock: 2x (1:12:31, 1:12:34)
What Does This Mean?: 2x (56:25, 1:28:55)
That's Huge: 2x (1:11:31, 1:29:36)
Pencil Mark/mark: 2x (48:11, 1:08:45)
Useless: 1x (45:16)
Bother: 1x (26:11)
Nonsense: 1x (46:40)
In the Spotlight: 1x (1:43:30)
Break the Puzzle: 1x (1:34:58)
Fascinating: 1x (16:02)
Incredible: 1x (02:52)
Gorgeous: 1x (1:20:10)
Masterpiece: 1x (1:52:03)
Whoopsie: 1x (37:37)
Phone is Buzzing: 1x (1:04:43)
Scrabble Tiles: 1x (37:12)
The Power of Maths: 1x (1:18:54)
Thingy Thing: 1x (1:26:38)
Cake!: 1x (01:51)
Symmetry: 1x (44:57)
Most popular number(>9), digit and colour this video:
Twenty Eight (8 mentions)
One (190 mentions)
Blue (4 mentions)
Antithesis Battles:
High (2) - Low (1)
Odd (98) - Even (89)
Higher (2) - Lower (1)
White (2) - Black (0)
Row (15) - Column (7)
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!
Bah ! (1:44:57)
I want to know how many times Simon said "thermometer" in this puzzle
No chocolate teapot?
Honourable mention for "Thingy thingy" at 59:09
I think it would be nice to also have the time stamp of when Simon gets his first digit in the grid.
At 1:28:36, this bifurcation mess can be avoided by finding that F in column 7 can only be in r3c7 because the 7 in column 6 can only be in row 7 and 8. From there, this puzzle becomes much more elegant to solve
Thank you! Very elegant and I knew there must be a way :)
My juggler friends and I, somewhat imfamously, often think up a new, complex multi-person juggling pattern, then spend half an hour talking about it, trying to understand how it works and what the implications are, before any actual juggling takes place. I can't think why I'm suddenly reminded of that here. 🤔
1:23:39
It's easy to prove that R3C6 is not even.
Its value is 2or4, but its corner mark is not B, but C. (by sudoku)
And as we see in R2C4 C is connected to 3 and to D.
But R3C7 can't be either (by sudoku, 3 is in the region, D is in the row)
It means that our R3C6 cell has nothing to connect to. And that means it's not even. Thus it's odd :)
This dude really thought “one of the sudoku triangular numbers is a perfect square” and then made an amazing puzzle utilizing that fact!
Logically, I got as far as figuring out that there couldn’t be any thermometers in the perimeter.
I could follow your logic, once you pointed it out. I just couldn’t get there myself.
Great solve. Watching you solve puzzles makes me a better puzzle solver, and I can’t thank you enough for that!
Simon...just absolutely more incredible solving by you!! Loved how you approached this ruleset! Thank you as always for showcasing brilliant setters, rulesets and puzzles on a daily basis
Simon, what a video - epic! Your reasoning and logic at the beginning to work out how the thermos had to work was mind-blowing. Thanks so much for this video and for doing everything that you do to make sure that we get two sudoku videos every day - not to mention cryptic crosswords! (I also loved the little side trip into Oliver!)
Wonderfully expressed as usual from you!! 🩵💙
That was crazy, I tried it for 30 mins getting no clues, so I just reverted to the channel and really enjoyed Simon cracking it. I almost felt the pain
I think you need to add a notepad to the side of the sudoku app. It would have been quite helpful here. Knowing that the even-long thermos all end in 8, so 78, 5678, 345678, and 12345678, and there's 1234567, if you wrote it down you'd quickly see that it accounts for all four sixes. So the 5-long thermo can only be 12345, and now all fours are gone too. So the three length thermo is 123, with one final odd digit being the last 'one length thermo'.
But I think you need to write it down to see that.
Yeah, this was an easy win just from noting down the options
It was just about possible to use Mark's favourite, of using an empty area of the grid (the centre) for "aide memoire" pencil marking. I entered the first and last digits into a cell to denote the full length of the thermo, knowing it also had all the digits between. So 17 stood for the 1234567 thermo. This way I was able to reach the same conclusion, but without pencil and paper.
However, I then needed to write them down on paper anyway, when it became necessary to delete my aide memoire 😂.
Simon did say it was 100% on LMD. Unfortunately, there are trolls on LMD that actually watched this channel and watch just to get the answers so they can pad their scores. That wouldn't be too bad but then this certain troll marks the puzzle down in the rating. This puzzle now has an 88% approval rating because of this certain troll. It just shame that LMD allows this person to continue manipulating the approval rating. On top of this this troll is in the top 10 of solvers. It is pretty sad that a person like this having no honor is allowed to take away the joy that setters get when they see how well their creation is doing. I'm not going to name the person but he watches this channel and others so he can get the solution code and continue the cycle of jacking up his ratings while knocking down the setters rating. A true troll that needs to get evicted from LMD.
Is there a way to report them on LMD? It seems to be pretty obvious that, at least for this puzzle, that's exactly what happened. Such a shame that someone would do that, but hopefully there's a way to stop them.
Yikes. That is truly horrific.
You are very honorable not to mention the name of the person. Too bad the person does not have the same sense of honor not to do this. I don't understand people sometimes. No, that's not right, I very often don't understand people!
Already reported. But the admins of LMD decided to do nothing.
According to them, there was no evidence to act against this person
What are you on about?
I think Simon needs an iPad that he can project over the puzzle when he’s doing these very complicated conditions and sets to better explain how he’s getting there. I was able to follow, but it would be amazing to see it drawn out to see the beauty in the analysis ❤
I appreciate how you take the needs of your colour-blind viewers into consideration when choosing colours. With that sort of thing in mind, please consider that it is difficult and unpleasant for some viewers to hear you call yourself stupid, or to apologize for taking too long on a puzzle. It is from a place of compassion that I make this comment. Please be kinder to this person I care about (i.e. yourself)!
He started talking dumb things about himself since 2021, we have worked hard not letting that continue and it have never worked. I think he's taking comments like this as sarcasm and doesn't understand that nobody (that includes us viewers) is perfect.
Two hours of Simon's video?! What a treat!
Is this a record? First pencil marking at 48 minutes (minus birthdays)? But I was flabberghasted by the fact that any pencil marking was possible! Simon you are a genius!
30:00 I believe that it can be stated around this point that there must be not only 4 lots of the digit 8 in the central cells, but also 4 lots of the digits 6, 4 and 2. And this should fix most of the odd-length thermometers to be 1-7, 1-5 and 1-3. Correct? Only about the single-cell thermometer I am not sure...
Yeah, I noticed that as well. I don't think one can nail down the single cell thermo at that point other than that it needs to be odd. However, the 7 known thermos fix at least one each of 3, 5, and 7 in the corners, and then the fourth corner has to be the same digit as the single cell thermo because otherwise there would be too many of that digit in the grid.
52:25 Classic Simon: doing incredible logic on the thermometer's parity but not seeing yet that the 57 pair in r2c2 and r3c3 is solved by sudoku :-)
What a lad, calling out his video length +/- 10 minutes
1:28:25 Simon unintentionally being whisked into the alluring world of bifurcation
Hats off to you Simon, I am a very competitive person, and keep challenging myself to reach certain conclusions faster than you, some puzzles I see stuff before you do and some I don't, but the one thing that I NEVER EVER win at is how fast your brain finds proper grid divisions to mark cells that are equal to each other... You are just pure genius.
Thanks for your videos. It really helps while staying days in hospital to search and think with you
I like to imagine that in his previous life in financial services, Simon also insisted on doing all calculations in his head without recourse to paper, pencil, or computer. 😀
51:37 Funny how he does all this thinking, instead of just realizing 7 is in r3,c3 because of the 1,3,5 triple he has penciled in the ling "snake" region which contains that cell, leaving 7 as its only option.
Getting my popcorn for 2hrs of pure joy 😊😊😊😊🎉
For puzzles like this I would not mind if you had a second screen where you took notes, and scrolled to it occasionally. It's incredibly impressive you can keep all of this in your head, but I certainly cannot, so I just had to trust that you knew what you were talking about, rather than looking for deductions myself. Great solve :)
i’ve wanted a notepad feature on the solving site for quite some time!
Hey Simon, thank you for showing the law of leftovers again and in detail. This was a point where I was stuck.
I think pen and paper or a second screen with an excel-sheet would have been a great help for you. I managed to figure out the nature of the thermos quite well after visualizing my knowledge about the thermos.
Even more impressive though that you got there without any tools. Kudos!
This is only my 3rd video this year because university is stressing me out 😪 but I'm so glad I'm here to see Simon solve this wonderful Sudoku and to read your wonderful comments, you awesome people.
27:00 Dont we know all thermos at that point? There are 4 8s 4 6s 4 4s 4 2s in the perimeter which means there are 4 of 2/4/6/8 in the middle. The only thermos that can pick up the 4 8s are the even thermos which means they have to be 12345678 345678 5678 78 Which also means we have used up 4 8s 3 6s 2 4s and 1 2. Now the odds the biggest one can only be 1234567 so all 6s are gone which means the 5 length thermometer can only be 12345 which is the only way is does not pick up another 6. Then the other two have to be 123 and 1. That finishes all thermos. Did i make an error there? Sorry for my bad english by the way. On a second thought i dont think the single cell thermo is not forced to be a one because no even digit is involved but the rest should be correct. But it is a 1/3/5/7
This. Was. Awesome. I kept thinking I was to the end of the possible deductions - with parity, number combination, everything - and I just kept finding more! I don't think I could have done it without a pen and paper, but Simon's brain is remarkable. Beautiful puzzle. Brilliant.
Simon: "I'm so stupid, so stupid, sorry"
Remarkable puzzle which I would not have solved in a decade of Sundays without some prompting from Simon. How he does it all in his head I will never know. Ditto how anyone can concoct a puzzle like this!
I've done very little of the puzzle so far, barely started and went to bed, but just from thinking about it at my work I can tell you it is amazing, the sheer amount of information you can extract essentially just from the corners is stunningly beautiful
"was it visualizable from a bit further back so we can claim it wasnt bifurcation, which it was a bit" made me LOL
Stubborn as I am, I had to solve this before watching your solution (that's still how I check my brain is still working), so that's where my afternoon went. And as I started watching yours, I thought "I admire Simon can do these puzzles without pencil and paper and with very little marking"... and I confirm that! Absolutely brilliant solve! But I spent almost an hour on the same place where you got stuck, and I'm kinda pleased to see there wasn't anything evident that I missed, it was just very very hard.... Kudos!!!
37:18 Not sure if it would have helped getting to a solution any quicker, but my path through this was to use the green cells in R3 (let's call Group A) to make two groups either side: R3 C1/C2 (Group B) and R3 C6/C7/C8 (Group C)
Then because Group A, B and C are all in diff regions the cells in R1C1 + R2 C1/C2/C3/C4 have to be Group C + 2 of Group A (3rd is already in R4C2), and R2 C5/C6/C8 have to be Group B + 1 of Group A.
Then because of the Group A digit in R4C2 - we now know that R4C8 & R5C8 have to be the same as the two Group A digits in the R1C1 + R2 C1/C2/C3/C4 block, and one of R2 C5/C6/C8 has to equal R4C2
Still some shenanigans after that, but less i think 😊
U don’t have to apologize for not figuring it out quickly. I watch because i enjoy watching the big revelations, and the longer I have to wait, the better!
I know we all deal with distractions differently, but there's not much I like more than the sound of small planes flying by. For me, it's an absolute highlight of your videos!
I think it would be fun if he tracked down “Maverick” and had him on camera sometime! 😊
Stunning puzzle, with layer after layer of original logic. One of my favourites on the channel so far! ❤
This one was hard ... but I got it. 231:20, solve counter 283.
I started out with coloring the whole grid (which turned out to be not very fruitful, but bring some results), then did some logic (including a tiny bit of set theory) to figure out which thermos need to exist (which didn't need that coloring at all - except for the fact that two of the corner cells had the same color), then tried to combine both of them.
Crazy how fast he realized there are four 8’s in the center and how long it took him to consider that there are also four 1’s… would’ve taken me ages to get the former, but having gotten it the latter would’ve followed instantly for me😂
It is outrageous that Simon has not placed the thermo bulbs!
😂
I am proud to find out what is the list of thermos (except the single one) just at the beginning. But the rest - rock hard! Simon, you are the Master.
Very proud of myself for deductions on this one. Simon shot ahead and came up with a whole load of complex deductions that it took me forever to understand! But pretty soon after he worked out that there were four 8s in the middle, I deduced the nature of all the thermos, by considering the quota of other even digits. I haven't finished watching yet so I'm not even sure whether Simon works that out or solves it another way!
I definitely wouldn't have been anywhere near a break in on my own!
Amazing, bonkers sudoku puzzle! Mark often uses the sudoku grid to log down his thoughts almost like it’s in the margins which very often proves so helpful in his explanations. Simon please feel free to write it all down on the grid rather than try keeping it all in your head, which is admittedly way more impressive and functional than mine! It really isn’t bifurcation. I know I’d be hugely grateful to see you scribble explanations down on an empty grid. Thank you for this fantastic channel. Sheer enjoyment.
Always appreciate you picking colors that work for colorblind viewers! My Dad was red-greed colorblind so I've always been aware of friendlier color combos.
Bifurcation was not necessary, but sometimes the next steps were hard to find. Also Simon, two things - first, please take notes if it will save your sanity, and second, if you are stupid, what are the rest of us?
Love your question… his half hour solves often leave me feeling dumb… this is a whole other level
1:03 I think we expect there to be exactly three 5s in the perimeter; based on the original logic about the makeup of the thermos, there are exactly five 5's on them. Well, unless the one-cell thermo is a 5, but I think that has to be a 1; otherwise there are only four 1s on thermos and therefore four 1s on the perimeter, which would require that none of the corners is a 1 and whatever digit is repeated in two corners would only appear twice, which is not enough.
My brain got exhausted just by watching Simon solve this puzzle
Simon in multiple videos: "If I wasn't doing this puzzle on camera I would do X." Do it man! Do what you need to do! Maybe just upload a picture of your work and put it on screen when you're doing it. Or open up a word document, or paint to do your notes in a way we can see.
This was a classic example of "the solve would be better if you had a notepad window" ... there's just too much for you to try to keep track of in your head here, let alone for _us_ to try to keep track of while watching ... pretty much no-one is going to try to solve this without keeping an aide memoire, there's no shame in doing that, and *it would make the solve easier to follow and more enjoyable to watch.*
And then we wouldn't be treated to the spectacle of you berating yourself for temporarily forgetting one of 20 or more complex deductions that you had made previously, which (and I mean this in the nicest possible way) is absolutely ridiculous!
I solved it... with a notepad to remember all of the thermos. im very impressed at how simon could do all of that in his head. without simon's insights that there are 4 8s and 4 2's, and the only way to even out the digits is to have 4 double odd ended thermos and 4 balanced ones, forcing there to be 4 thermos that end with 8s (and start with once of each odd digit), 3 that start with 1 and end with every odd digit (the 4th 2 is in the 1-8 thermo) but 1 plus one odd digit in the length 1 thermo
The way i deduced everything is by noticing that there are exactly 4 of each even digit in the center, and that the 4 long thermos (length 5-8) have to contain a 4, using the all up. the length 4 thermo cant use a 4 so has to be a 5-8, using up the last 6 (the other 3 have to be in length 6-8 themos). the length 3 thermo needs at least 1 even digit, but can't use 4 nor 6, so has to be a 123 thermo and the length 5 thermo can't use 6, so has to be a 1-5. these use up all the 2s. that leaves 2 length thermo as a 78 pair and the 6 length as a 3-8, using up the last 8. this makes the length 7 thermo a 1-7 and you only have odd digits left to place in the border + the single cell thermo.
the test at 1:23:38 was easy to break. like simon said, r3c6 is c, and c is surrounded by 3 and D (in r2c5), but r3c7 nor the 2 even cells can be either, so r3c6 isn't c, and so not even.
9:47, triangular number for 8 is 36. There are 36 squares in the center of the grid which must contain thermos. Every thermo in the center of the grid must contain consecutive numbers. All corner digits of the puzzle are odd. Each thermo must have either equal odd and even digits or 1 more odd digit than even, the 4 corner cells being double counted decreasing the odd count around the perimeter by 4. Also, the irregularities of grid make r3c345 equal at least two odd digits, and r2c4, r3c4, r3c5 comprise of at least two even digits, proving r3c3 and r6c3 odd and r2c4 and r4c2 even. The single digit thermo is tricky because we are left with 4 1s, 3 3s, 3 5s, and 3 7s, where if every corner is different, the 1 is the 1 cell thermo or if there is a repeat in a corner, that is the 1 cell thermo, with 1s never appearing in a corner.
I did take Simon's early thoughts about working out the possibilities on pencil and paper, which helped me greatly in starting the solve! I think it would be nice to figure out a way to use this kind of outside-the-grid work in these videos, because it's very helpful in puzzles like these.
Without finding any digits, I discovered the makeup of the largest seven thermos, and also discovered that there were 357 in the corners, along with whichever of 1357 made up the lone cell thermo. All of this I figured out just because I knew there could be no more than four of every even digit in the 6x6, along with the fact Simon also discovered about the four 8s required to be in every even length thermo.
The 28 digits in the perimeter are four sets of the digits 1-8 less the corner digits (which have effectively been counted twice: once in their rows and once in their columns). The 36 digits that appear on the thermometers must therefore be four sets of 1-8 plus the digits in the corners -- which we know are odd (from the grey circles) and cannot include more than two of any digit (because each one sees two others). We also know the thermometers must increase strictly by single steps, just to swallow up all the available digits.
The biggest possible difference between bulb and tip is 7 (for a 12345678 thermometer) and the smallest is 0 (for the single-cell thermometer). This allows a maximum of eight thermometers, which must account for 36 cells. The minimum length of a thermometer with a non-zero difference is 2; the maximum length of a thermometer is one more than its difference. Since 1+2+3+4+5+6+7+8=36, all the thermometers must increase strictly by single steps; otherwise, digits will be left out that cannot possibly be part of any thermometer.
Now knowing that the eight different-length thermometers are formed from four sets of 1-8 plus another four odd digits, we can consider how they might fit together by pairing them off to create total lengths of nine; which must consist of four complete sets of 1-8 with an additional odd digit. Clearly we need to go to both extremes within each pair, to avoid having too many "middle" digits (2-7) and not enough "outside" digits (1 and 8). If we tried something like 12 and 2345678, we would need an extra 2, which is not available. 123 and 345678 is a set of 1-8 plus 3. 1234 and 45678 needs an extra 4, so is impossible. 12345 plus 5678 is a set of 1-8 plus 5. 123456 plus 678 is not possible. 1234567 and 78 is a set of 1-8 plus 7. 12345678 could actually go with any other odd digit, tempting as it might be to assume this ought to be 1.
This fixes the thermometers as being 12345678, 1234567, 345678, 12345, 5678, 123, 78 and some other odd digit; and all except the one-cell thermometer must either start from 1 and finish on an odd number, or start from an odd number and finish on 8. The corner cells are 3, 5, 7 and the digit on the 1-cell thermometer.
All this is a lot to try to explain in words, but quite easy to see if you have some props -- such as numbered tokens, or even just pencil and paper -- to try it with. But of course, introducing sequences involving props into a CTC video would require new filming and editing techniques .....
For the hard part: (i hope this is logically correct and helpful):
- if there's a 6 or 8 in R6/7C2 there needs to be an accompanying 7
- if it's a 6/7 pair, it needs to be 6 in R6C2 und therefore a 5 in R6C3
- therefore D is 5
- since all thermos with an odd number as their highest digit need to have an odd number of cells the thermo in R2/3C4/5 needs to end in a 1
- the 1 can't go into R4C5, because if it were in R3C6 the even number in R3C7 wouldn't be able to exist (can't be an 8, because in this scenario G is 8, and can't be anything else, since it's thermo couldn't exit on both sides)
- All of this leaves us with a 1 missing from R3, which can't connect to a 2 in it's collumn or row, therefore breaking
WOOOOW what a puzzle. took me close to 3 hours to do on my own. I also went back later just for fun and checked if it was gonna be easier, knowing where the thermos were... it ONLY disambiguates if you respect the rules about unique distances on the thermo (other than that its quite easy to solve if you know where the thermos are) either way INCREDIBLE
I did appreciate that b being either 2 or 4 was either a two-b or not-two-b.
😄
It took me a *lot* longer, but I was able to solve it after a couple of hours.
I started with the same deductions of odds and evens inside the middle grid. Then I moved on to the thermometer compositions. Looking at the correct number of odds and evens, I believe all of them are forced. Except the 1-length. Then either there are no repeats on the corners, which pushes the 1 to be the 1-length. Or there is one repeat, which can be the 3, 5, or 7, making that the one-length. There can't be 2 repeats. I was able to show next that there couldn't be 4 different corners, pushing the 1 out of all the corners into the sides. And requiring exactly one repeated corner. I believe that would have helped Simon solve the '78' in r67c2, since the 1 would be required in r3c45.
Awesome sudoku and great to see Simon make 3 hours worth of deductions in about 12 minutes.
Imbalance In Favour Of Oddness is my new bandname, thanks Simon!
Bifurcation in a CtC video, and that too by Simon! Now I’ve seen everything :)
Simon, you are definitely too hard on yourself. I envy your prowess with maths and those digety thingy things.
I love sudoku but suck at math. And you are not a silly man. ❤
As of 13:30 in the vid i came to notice that, in the empty 6x6, theres eight groups of empyy cells separated by lines, containing 1,2,3,4,5,6,7,8 cells. Totaling 36.
I'm amazed how Simon can deduce and visualize all thermo combinations in its head. I had to use notepad... but it went much faster ;)
I am not sure I could have solved this one no matter how long I stared at it. I got the parity/lengths/totals bits but trying to figure out where to go from there was beyond me at least for long enough to watch this and just enjoy the solve. You have a dark mind Nahileon. 😈
I was confused by the need to have only single digit rising thermometers until I finally just counted it out more distinctly. With the 36 squares in the middle, you need a maximum sized thermometer for each difference between bulb and tip, which means each length of thermometer from 1-8 cells must be present to give 36 squares. 8+7+6+5+4+3+2+1=36.
I really tried to wrap my mind around it but the thermometer logic just breaks my brain 🤯
I am just getting started with solving such difficult Sudoku puzzles and it took me super long to wrap my head around this and i had to peek into Simons solution some times because I got stuck but in the end I managed to solve it and I absolutely loved this amazing puzzle!
This is the first time I've ever seen Simon bifurcate! Can't say I could possibly blame you for it though, unbelievable solve
This is an impressive puzzle... And seeing it done... Well, my attempt to create an invisible thermo puzzle feels... less terrible that I didn't get all that far with it. Thank you Simon, Thank you Nahileon.
Had to watch to 1:43:50 to solve. Did not really understand how to approach this one at all, just used sudoku in the end to populate the rest of the grid knowing that Simon wasn't ever skipping digits on the thermos. The difference counts on all the final thermos are all different, so I think I got it, but no way would I have been able to finish this without all that help. No shame in that though, was still great fun :D
You might have saved yourself about a half hour of time if you had written down the thermo combos on paper - took me less than 5 minutes to figure out. However I am impressed that you kept them all straight in your head.
103:24 for me. I came up with the break-in quite quickly, EXCEPT I wrongly assume the tricky 1-cell thermometer which could be any odd number is 1. I have to spend at least 20 minutes to find my mistake. Nevertheless, this is really a stunning puzzle.
This one was one of a series of very vexing puzzles for me; I eventually figured the other ones out, but I failed on this one.
I actually got some parts on my own (the part about 'leftovers', the knowledge that all thermos has to alternate odd-even, the knowledge that all odd-length thermos were bookended by odd numbers), but I missed the fact that every even number besides "8" had to connect on both sides. After that, I got it in my head that the four corners had to be all different (I was right about 7 of the 8 thermos, but forgot that the single-cell one didn't have to be "1"); from there, I went a long way before breaking; at that point, I completely gave up.
I kinda wish we'd get a video of a puzzle that they couldn't solve, if only to see if I could get one they didn't. Until that day, though, I'll just remain in awe that they could figure something like this out.
Before you went into parity colouring of the grid you instead could have one continuation of the pattern you already detected for the differences. You already concluded you have to have thermometers of each length 1-8 and the rules also only allow each difference of 0-7 once only. The 1-8 thermo uses up the 7 difference, the next shorter thermo you finally concluded has to be 1-7, as the 8 digit has to be reserved for the even length thermos. Well, that's a very special deduction but can be generalized when thinking about the differences you can have with different lengths.
What you cannot have is a thermo of length 7 with a gap, as that would go from 1-8 ad have a difference of 7 again, so the 7-cell long thermo has to have a difference of 6 only and be consecutive. Well, the next shorter length 6 now also can't have a difference of 6 as that is taken by the 7-cell-thermo. This ripples down and you can conclude all thermos are also renban lines and have no gaps.
I made a wrong conclusion from that, though, that when all even-length thermos end in 8 to have 4 8s in the 6x6 all odd-length thermos start at 1. As you can have a repeat odd in a corner, that can be the single cell thermo within the 6x6 and it doesn't need to be 1. I failed at that.
I'm still proud the renban deduction turned out to be correct.
At the start Simon kept saying "If I wasn't on camera, I would be taking notes" and I keep wondering, why not just open Notepad or something? Or just use the 8x8 grid to take some notes? I just opened the puzzle and put some numbers in and very quickly realized that we need a minimum of five 7s in the middle, five 5s, five 3s and at least four 1s. What took Simon a LOT of time and mental effort to sort out (the composition of the thermos) took me about 5 minutes after taking a few notes. Therefore, I think if he had let himself do the same, this could have been an hour long video instead.
How you did this without having to write stuff down I can't even imagine!