Thank you so much for featuring this puzzle! It was a really nice surprise. The way you figured out and explained the logical steps concerning the regions was really excellent. Regarding the point in the solve where you used the 'longish' logical chain to deduce which the last yellow cell was, I will gladly let you know what the intended next step was: the 7 and the 9 in the green region must be the two cells in column 5 (7's and 9's in col 2-4 are used up by orange, blue and yellow). Placing that pair immediately makes the 24 clue useful to resolve row 8. I worried a little bit about this step as a sticking point (since especially the 9's in col 2-4 are not so easy to spot). But there is also an alternative way to allow progress on the 9's. By appreciating that the 9 in yellow will always be in row 3 (regardless of whether yellow goes left or up), the grey 9 in R1C1 can be deduced, and then the only remaining place for a 9 in column 5 will be in green. I did actually consider to replace one more question-mark by a number, in order to make the irregular Sudoku part towards the end of puzzle easier, but I decided against it because 1) I really like the type of arguments combining regions and columns like the one used with the 7's and the 9's, and 2) because of the possible alternate (more step-wise) method, I figured solvers would be able to finish the puzzle even if the did not spot the intended way at that crux. Interestingly, this puzzle was kind of set by accident. I solved two 7x7 puzzles by SudokuExplorer based on this ruleset, and immediately wanted to play around with the logic a bit myself. So I opened Penpa and started experimenting, got in a really amazing flow, and before I knew it, I had actually finished setting a 9x9. Sometimes you have days when everything you try ends up working out...
Dr Josefsson, that is one heck of a puzzle! Absolutely gorgeous! That intended solve path you described above, is just unbe-fricking-lievable. More of your puzzles on CtC, I say.
I also got super stuck on that spot and your explanation is really helpful! I’m not sure exactly how knowing the 7-9 situation helps you figure out the top part of the puzzle, though.
Congratulations Simon and Mark! You found an enormously successful way to connect with not just 5,000 patreon members, but also with thousands more through the podcasts, the interviews, the merch, and the channel. You have changed lives, cracked cryptic by cracked cryptic. Cheers!
20:15 An easier way to deduce that Yellow and Blue can't be part of the same region is simple addition of the two totals: 20 + 23 = 43 (jealously guarded secret). Together they make 7 squares, but you need 2 more squares to get to a 9-square region and you also need 2 to get to 45. You can't have two 1's in the same 9 cell region.
For some reason I understood this confusing puzzle more than any other one on this channel. I love every video I have seen from Simon, but I have to admit I generally feel very dumb. Usually when he says "I know some of you are screaming at me right now" I'm like nope. Actually I am not because I have no clue where to start. This is the only time I think I will ever be able to say I was lowkey screaming at Simon through my computer screen and that makes me happy :)
Don't forget you have the line tool! You can mark a × on a boundary where you know the regions on either side of the line are different even if you don't know which regions they are, which can be helpful to keep track of things and means you don't need to use dummy colours as much.
Thinks: "It's too long. It's too complicated. There's no way I'm trying this. I'll just watch the master at work." Attempts it anyway. Solves it somehow. And I'm so glad I did. That was marvellous. The logic in that and how those clues work out is brilliant. Perhaps a hunt on LM for more Jesper in the next couple of weeks would be fun for me although infodumping about it to everyone will only further confirm in their minds that I'm odd. Looking forward to the Murder. That sounds excellent too. Deep joy. Can't beat a good murder mystery.
The OCD part of my brain wanted him to mark the 2 known red in the lower right for so long. Even if you don't know the 3rd one, mark those 2. You know, I just need to listen without looking at the screen.
I really enjoyed when you got stuck at 30:10, because you usually do those deductions in your head before articulating them (well) to the audience, and this was a great insight into what’s happening in your head during those quiet bits.
"It's not bifurcation if I can see it, and I can see it" ... "Didn't expect that... ooh, didn't think about that... Ah, but it still doesn't work" ... "It's not bifurcation if I can see it, and I saw it, there, I said it"
23+20 is different from 45 (the sum from 1 to 9) so there must be a "2" left if the two regions are the same, and it is impossible to sum 2 numbers to "2". So the 20 and 23 might be distinct regions... Great puzzle! Thank you for sharing!
I get that it didn't really matter tot he solve as he kept fine track of what was supposed to be there, but anyone else mildly annoyed that he took so long to color the two squares in the bottom right area that had to be part of red regardless of which way around the 12 pair went?
If I face a disciplinary hearing for not performing at my job as I should be I'm coming after you, SImon and Mark. How am I supposed to focus on work when there is so much fun to be had at CTC??
What a lovely solve, it really showed the incredible logic here at play. Jesper, you've done a fantastic job setting this puzzle, even if it was a bit beyond my limited timeframes haha.
A huge mazel tov on reaching 5K patrons. I am proud to be among them. You have given countless people hundreds of hours of top level brain exercise - with good cheer and Douglas Adamesque sense of humour. I want a chocolate teapot as the reward for the jubilee whodunnit.
Brilliant puzzle! I thought at 17:30 you could have also said that whether the 20 clue started with 3 or 5, because the two clues sum to 43, you would need two 1's in the region to make it work.
An hour of solving can really boggle the mind, in the end row nine was filled all but the nine and you began to panic. Great solve and much respect for your endurance on these longer puzzles
What a brute of a puzzle, but an utter joy. First time I broke it about three-quarters of the way through, but it was such a clever puzzle was willing to give it another go.
59:50 yellow area must have 9 on R3, which makes 78 pair in C1 78, 47, 67 in R1-R3 kills 7 in R3C5 putting 4 in R1C2 would end up with 3 and 5 in R6 and R7 forcing to put 3 and 5 onto 24 sum in R8, so R1C2 must be 3
67:16. What an amazing puzzle! I got stuck for the longest time where I just couldn't solve the lower right quadrant (3 squares in the 28 X-sum region and 23 X-sum region combined). I spent 30 minutes getting to that point but about 20 of it was trying to figure it out. Then I paused, went away for a while and came back. I reset the puzzle, because I needed a clean slate and was able to see the logic which had eluded me earlier. And from there everything followed, but there was a lot of tricky logic involved. It never felt straightforward. But once I got to the end, it was an amazing feeling. Thank you, Jesper for the great puzzle.
46:54 "Is there something easy we could do at this point?" Yeah, I was asking you to paint two cells red (R8C7, R7C7) for about 10 minutes now and also in the top right corner paint R1C8 and R1C9 pink ("purple"). It's not much, but it's easy and useful at least as far as making progress goes.
69:30 for me!! What a great puzzle!! I always find it so incredible when puzzles like this one have a unique solution and a way to get there from a completely blank grid. My respects for the setters who are able to create such beautiful puzzles.
Could you possibly have a chat with your app devs to make the colors slightly less saturated and slightly lighter? The red really clashes with the blue pencil marks. The same is a little true of orange, blue and purple too. Even changing the hue/brightness values by 10% to make them a tiny bit more pastel would probably work - I don't mean full-on near-white or anything!
@@terracottapie Of course, but if this was the case, the point is more I'd like the channel itself to use a lighter option if there is one. It's not even AS bad on the app itself, but with TH-cam's compression, between the slight biurring and color flattening, it makes it very hard to see here.
To reiterate a theme from your video with Rob Stephenson it is encouraging watching you find your way through difficult puzzles with good humor. Giving your viewers hope to get through their challenges with good humor as well.
I really like this puzzle, BUT... I had an issue with the two question marks at the bottom of column 9, where a certain solution seemed accurate... but it wasn't apparent how it broke the puzzle until maybe a half an hour later, when I was 1:14:00 in. So now I have to restart. Restarted, and solved it at 49:29 on that attempt.
At 35:40 why is Simon so certain that the yellow regions turns upwards and not downwards? Since the grey cell could be red, I don't see a way to decide whether yellow goes up or down. What is the logical path to determine if yellow goes up?
20:00 For Blue (3 cells summing to 20) and Yellow (4 cells summing to 23) to join, you would also have needed get the remaining 2 cells to sum to 2, which would have been quite hard.
at 4:53, when trying to describe the example - "look at the 10 clue HERE", would be much easier to follow if the cursor hadn't switched to a single pixel.
@@stephenbeck7222 Yeah, possibly a conflict between the image editor and the streaming software? maybe opening the image in a different browser tab would reduce some variables.
At the 1-hour mark, where Simon comes dangerously close to resorting to bifurcation, There is still work to do on the bottom row, placing a 2 and getting an 8-9 pair. Also, in row 8, columns 5-7, (using the 24 clue outside the grid), Simon couldn't determine the combo to add to 18, BUT, he could eliminate 6 as a candidate in column 7, and narrow down the digit in column 5. I think Simon could have eliminated the 6 in r8c5, and once he did, he could determine that row 8, cols 2-4 had to be 6-5-9 (you couldn't pair a 6 with an 8-6 combo nor a 7-7 combo). In sum, I know there was lots to still be had on the bottom two rows at the 1-hour mark, and I'm guessing Simon could crack the puzzle at that point (in time), at that point (in the grid). LOL I know he felt under time pressure.
I didn't even try that one, but I decided to try out that Mystery hunt and I'm happy to say that I solved the first puzzle with no issue. Getting that killer sudoku app was a good investment.
19:41 There are two reasons for the blue (20) 3-cell region and the yellow (23) 4-cell region to be in different regions: 1) The 3-cell region has to be 3, 8 and 9. The 4-cell region has to be 4, 5, 6 and 8. They overlap in the 8. 2) The 3-cell region is a 20 X-sum, and the 4-cell one is a 23 one. You'd have 43 in two cells, and you can't have two 1s to round the region.
I dont know if he figures this out later as I paused to video to comment but at 16:00 when he tries to figure out if it can be a 5, it can’t (unless I’m wrong) because that would make both question mark clues 1s in the L shape region
Never in my wildest dreams did I think I could beat Simon at a build-you-own irregular sudoku, but I solved the puzzle in under 45 minutes. It's amazing how much sudoku logic makes sense to me when I'm operating on only 5 hours of sleep.
Not necesarisialyrily, because they would have to have a cell that was joining them ... if the cell that was in both regions was a 9️⃣ then you've only got digits adding up to 42 at that point.
At @33:50, how can you know that the 89 square is purple? Could it not be a yellow 9? I’ve been staring at it for a bit but if that’s sound logic I simply cannot see it...
He already knew exactly 3 squares in the corner area were not purple. Once he gets the 12 pair, he knows that those 3 cells must be in rows 7 & 8; one in the row with the purple 2 and two in the row with the purple 1. In fact just moments later he fills in the cells in row 6 & row 9 all as purple. Not sure why he realized it for r6c9 before the rest, though.
@@thecommexokid That's not quite right. He shows a viable solution at 30:22, where the 1,2 pair are NOT both purple and the three non-purple squares are NOT in rows 7 and 8. The reason he marks r6c9 as purple first is because there are only two places in the 3x4 corner region, that purple inhabits, that can be 8 and 9, namely the two cells in column 9 either side of the 12 pair. Therefore, these have to be the purple 8 and 9. Only once he knows r6c9 is a purple 8 or 9 can he rule out the possibility shown earlier at 30:22, which then allow him to mark the other cells in rows 6 and 9 as also purple.
I want to see Simon vs Mark Sudoku set challenge. Mark creates one that Simon solves and Simon creates a puzzle that Mark solves. I would be interested in seeing what kind of sudoku puzzle Simon would create.
20:00 I am just watching and kibitzing a bit. Simon ruled out joining yellow and the 345 strip above it by noticing that yellow needs a 5. But the two totals are 23 and 20, so can never form a 9-cell region. The same does *not* apply to marrying the 3-strip to yellow. Although a 7-cell region cannot add to 43 ever, it could take in one more cell and become a possible 8-cell missing a 2. That would leave an 11 or 12 cell region in the south-east which then must be one 9-cell region and 2 or 3 cells to be absorbed by red. The reds and yellows *must* take, between them, 3 cells from the south-eastern region and the south-east must have its own 9-cell region. I think that is right.
Everyone knows Maverick committed the crime. Note: I have not solved anything, so if this is in fact the solution, don't comment about it please, let's not spoil things for people.
Note: commenters really do discuss solutions in the comments and to expect otherwise is actually unfair to the people that wish to discuss solutions. Anyone who expects otherwise will be disappointed whenever it happens because it does happen. Rather than expect this from everybody else, accepting things the way they actually are goes a long way towards maintaining peace in one's life. it's better to avoid the comment section until you have either solved the puzzle or given up That way, a puzzle need never be spoilt for you, and also you never need to be disappointed by the realisation that there's no way to make other people conform to how you wish they would behave.
@@tadperry1817 my point was more about making sure the joke wouldn't inadvertently spoil a fun reveal for people. If this comment had been a serious discussion about the challenge, then I would not have included a warning. But here it was meant to be something people can enjoy as a joke, regardless of their intentions to complete the challenge or not. Or I may just be a bit optimistic. 🤔
@@ragnkja In all things, you will never be disappointed if you don't expect others to conform to your idea of how they should behave. Not all people are going to provide spoiler alerts or avoid discussing solutions, and being aware of that is the best idea as a reader of the comments. It is not on them that you avoid spoilers, it is on you.
@@tadperry1817 The users of the CTC Discord server are actually pretty good about marking spoilers. If you click on them to make them readable, that’s definitely on you. And while TH-cam comments may contain unmarked “spoilers” for that video, it’s usually easy enough to just watch the video before reading the comments - after all, it’s right there, and you have to actively open/scroll down to the comments to read them, so they’re not that hard to avoid if you want to avoid spoilers. So it’s really on both parties. If a TV Tropes page says that “due to the nature of the trope, spoilers will be unmarked”, it’s up to you to decide if you want to take the risk that a work you don’t want spoiled is there, or if you want to avoid reading the section for a particular medium, or even the whole page.
At 34:45, I isn't necessary for there to be only 3 purple cells in that block of 6. It could be 4 if one of them wrapped around adjacent to the 1 clue. That would let the yellow 3 into column 7, row 6.
Really amazing puzzle.. I took over 80 mins and realised i had made a massive mistake... who knew that 3 does not equal 4 :O... anyway I did it again and it took me 50 mins... really good puzzle.
I would have colored more regions earlier (some 2-color) and more simple deduction. For example in the bottom green row the 2 and the 8,9 pair could be filled in easily. With 2 already in green, blue, orange, yellow and purple, it can only be part of grey. And it has to be on either row 2 or 3. And that also solves where the blue 2 can go. That determines the blue zone and fill in 2 more purple colors.
As always, a delightful video. To answer your question, Simon, the ‘bifurcation’ moment could have been easily avoided with the following: 3 must go in the top row, second place. It couldn’t be a 4, which would connect it with the orange below, making the 4-5 pair on row 4 break orange (two 4s in the same color). That fixes the 3s and solves the final yellow square. Hope that’s useful.
A grey 4 in r1c2 could leave the possibility of a grey 5 in r4c2 and an orange 4 in r4c3, no? That doesn't give you two 4s in orange, so doesn't break (unless I'm missing something).
20:00 The two question marks on the right side of rows 7 and 8 are a 1/2 pair. Let's see how long it takes him to realize it... They can't be 4 because they would meet up with red and repeat the 4. They can't be 5 because they would (again) meet red and complete the region, isolating the bottom corner as a region that is too small. They can't be 3 because they already see a 3 in the column. The only things left are 1 and 2.
"being likely isnt the same as something that we can force to happen" I run into that problem with every sudoku that takes me too long. start going by the likely option then have to redo the last 10 deductions when the chain i was on runs into a problem
Which is what we call „bifurcation“. As sometimes seen in Mark‘s videos (his speed in solving still is incredible, even if he takes wild guesses, though).
19.09 you can also say that blue on row 4 amb yellow of row 5 cannot be the same colour because they sum 43 and still need 2 cells more, it s another way to see it
1:06:36 finish. I got off to a faster start than Simon with the logic, but slowed down near the end allowing him to catch up. That's what typically happens.
17:00 my take: if orange and yellow is one region it will brake straight away, reason: the first number from 20 must be 3 or 5, and yellow region is locked with 4568 meaning orange region cannot be 389 because double 8, nor 569 or 578 because the double 5.
10 or 15 minutes, ha! I solved the thing all the way down to some 7/9 pairs without the 24, decided it was indeterminate, came back the next day and then saw the 24 clue.
Got it in under 3h... Well, I have a lot more to learn and speed up. One thing though: when you got stuck and "bificated", you could have seen the 24 clue much earlyier. You had to have 20 in 3 cells but on the right side you already knew you needed 3 large digits (79/789), therefore 20 had to be built with only 1 of those numbers. The only way of doing so is by using 965. So you tried it the other way around at the end (and still got it right ofcours). :)
Frustrated at Simon for not coloring them but even more frustrated at myself for fixating on it when I can see perfectly well that even if he colored them it wouldn’t lead to any new deductions.
28:30 The answer to that is yes, not no. If you take and make the cell that takes yellow, r7c9 a 1, and then take a red on r8c9, this is entirely possible.
My favorite puzzles are always these ones where you need to color in the irregular sudoku as you discover the regions, but they take me so long to solve...
How did he know @12:15 the 3 needed to be there and not 4 or 5? Was it because he knew a single gap was needed and therefore it could not have been 4 and 4 since that break sudoku and then 5 or 6 left made it so 3 was the only thing leave a gap since 4 would meet 5 and overlap with 6? I think that is the logic yes?
At 41:35 he assumes green and blue are not the same to color r7c4 green. No real harm done, he shortly afterwards starts thinking about whether blue and green are different, and even though his explanation why they are not still assumes r7c4 is green, his reasoning would work just as well if it had been left uncolored. And once we know blue and green are not the same region he can color r7c4.
Silly question, I solved a sudoku solver from Cracking the cryptic. When I finished solving I received a popup that said congratulations, then it asked the following: Did you spot the Empty Rectangle on 7. I am still learning notations when it comes to locations on the puzzle and don't know what on 7 means. I am used to notations where rows are letters and columns are #s. When it just gives a number is it talking about an entire 3x3 area? As in 1,2,3 next row 4,5,6 next row 7,8,9? Thank you.
I don't know for sure, but I would expect it means empty rectangle that could be applied to the digit 7 somewhere in the grid, to either place a 7 or remove a 7 as a candidate. For cell references, I think they always use the convention of r4c7, for example, to mean row 4 column 7. And if referring to boxes, they're numbered left to right, top to bottom, so 1 2 3 4 5 6 7 8 9
31 minutes and 33 seconds in and I realize there's an 8 that can be filled in... LOL, that's the only number I got so far, and I got it a completely different way than he did. I'm so proud of myself :p
I would usually watch both your and Mark's videos once I get home from work, but I'm afraid I've been too busy bobbins-ing up the Patreon Murder Mystery puzzles.
Realising the last yellow had two possible cells (3 or 5) was enough to simple use colours rather than numbers to allocate further cells their assigned colour.
1:02:58 for me today. Even with a huge 15 minutes gap where i was second-guessing my logic of the lower right corner. This was hard, but i liked the logic.
I had a mistake at the end, so I debugged it for almost an hour to see where things went wrong. I miscolored the field in the lower left corner very early. Everything went well and then an hour later I lacked a 4
No there isn't an alternate. If you switch 3/5 then you MUST switch all 3/5s - this puts a 5 at the top of r2 which is wrong. It puts a 5 at the right end of r4 which is wrong. It puts a 3 in the bottom left corner which is wrong. It puts a 5 at the bottom of c5 which is wrong. It puts a 5 at the left end of r5 which is wrong. Swapping 3 with 5 puts SIX broken digits in the edge cells of the puzzle - broken in that those digits have to match the number of cells they can 'see'. ie the bottom left cell sees 5 green cells & thus you can't blithely stick a 3 in there instead!
It's a useful exercise if you believe you have an alternative solution to find the point in the video where Simon's solution first diverges from your solution. Since Simon only enters a digit when he believes it to be forced, you're effectively saying there's a point in his solve where he makes a mistake. I haven't seem anyone else in the comments suggesting Simon's logic was faulty at any point. "The 3/5 pairs in the bottom right can switch over" doesn't help, I'm afraid. There are no 3/5 pairs in a 'deadly pattern' that could simply be switched over without impacting other digits.
Thank you so much for featuring this puzzle! It was a really nice surprise. The way you figured out and explained the logical steps concerning the regions was really excellent. Regarding the point in the solve where you used the 'longish' logical chain to deduce which the last yellow cell was, I will gladly let you know what the intended next step was: the 7 and the 9 in the green region must be the two cells in column 5 (7's and 9's in col 2-4 are used up by orange, blue and yellow). Placing that pair immediately makes the 24 clue useful to resolve row 8. I worried a little bit about this step as a sticking point (since especially the 9's in col 2-4 are not so easy to spot). But there is also an alternative way to allow progress on the 9's. By appreciating that the 9 in yellow will always be in row 3 (regardless of whether yellow goes left or up), the grey 9 in R1C1 can be deduced, and then the only remaining place for a 9 in column 5 will be in green. I did actually consider to replace one more question-mark by a number, in order to make the irregular Sudoku part towards the end of puzzle easier, but I decided against it because 1) I really like the type of arguments combining regions and columns like the one used with the 7's and the 9's, and 2) because of the possible alternate (more step-wise) method, I figured solvers would be able to finish the puzzle even if the did not spot the intended way at that crux.
Interestingly, this puzzle was kind of set by accident. I solved two 7x7 puzzles by SudokuExplorer based on this ruleset, and immediately wanted to play around with the logic a bit myself. So I opened Penpa and started experimenting, got in a really amazing flow, and before I knew it, I had actually finished setting a 9x9. Sometimes you have days when everything you try ends up working out...
Dr Josefsson, that is one heck of a puzzle! Absolutely gorgeous! That intended solve path you described above, is just unbe-fricking-lievable. More of your puzzles on CtC, I say.
I would have never thought of that logic, but it is absolutely beatiful!
Funnily enough I spotted the 9s but not the 7s being restricted there in C5
I also got super stuck on that spot and your explanation is really helpful! I’m not sure exactly how knowing the 7-9 situation helps you figure out the top part of the puzzle, though.
Come on! That was next level sir! Great job. It's just 8AM and I'm mentally fatigued. My boss would not like that....
Congratulations Simon and Mark! You found an enormously successful way to connect with not just 5,000 patreon members, but also with thousands more through the podcasts, the interviews, the merch, and the channel. You have changed lives, cracked cryptic by cracked cryptic. Cheers!
Amen!
20:15 An easier way to deduce that Yellow and Blue can't be part of the same region is simple addition of the two totals: 20 + 23 = 43 (jealously guarded secret). Together they make 7 squares, but you need 2 more squares to get to a 9-square region and you also need 2 to get to 45. You can't have two 1's in the same 9 cell region.
Denmark are playing England and I am overly invested in whether a square is purple or not. I never thought I would turn out this way.
I love these kinds of color "fill in the boxes first" sorts of puzzles... even if they make my brain hurt even more than regular sudoku.
SAMEE!!
For some reason I understood this confusing puzzle more than any other one on this channel. I love every video I have seen from Simon, but I have to admit I generally feel very dumb. Usually when he says "I know some of you are screaming at me right now" I'm like nope. Actually I am not because I have no clue where to start.
This is the only time I think I will ever be able to say I was lowkey screaming at Simon through my computer screen and that makes me happy :)
Don't forget you have the line tool! You can mark a × on a boundary where you know the regions on either side of the line are different even if you don't know which regions they are, which can be helpful to keep track of things and means you don't need to use dummy colours as much.
Thinks: "It's too long. It's too complicated. There's no way I'm trying this. I'll just watch the master at work."
Attempts it anyway. Solves it somehow. And I'm so glad I did. That was marvellous. The logic in that and how those clues work out is brilliant. Perhaps a hunt on LM for more Jesper in the next couple of weeks would be fun for me although infodumping about it to everyone will only further confirm in their minds that I'm odd.
Looking forward to the Murder. That sounds excellent too. Deep joy. Can't beat a good murder mystery.
Here I was after getting into a snag after 40 minutes and seemingly good progress, maybe I should let it rest and try again tomorrow.
The OCD part of my brain wanted him to mark the 2 known red in the lower right for so long. Even if you don't know the 3rd one, mark those 2. You know, I just need to listen without looking at the screen.
that and the two purples in the top-right corner, was driving me up the wall
I spent half the video wondering when he'd mark the green two.
omg yes! and the purple ones had me dying inside lmao
I really enjoyed when you got stuck at 30:10, because you usually do those deductions in your head before articulating them (well) to the audience, and this was a great insight into what’s happening in your head during those quiet bits.
I love the longer videos. Grab some popcorn 🍿
"It's not bifurcation if I can see it, and I can see it" ... "Didn't expect that... ooh, didn't think about that... Ah, but it still doesn't work" ... "It's not bifurcation if I can see it, and I saw it, there, I said it"
After a lot of struggling with other puzzles, finally a decent time for me: 56min :D I just love these unknown region themed Sudokus :D
Congrats on 5000 patrons!
23+20 is different from 45 (the sum from 1 to 9) so there must be a "2" left if the two regions are the same, and it is impossible to sum 2 numbers to "2". So the 20 and 23 might be distinct regions... Great puzzle! Thank you for sharing!
I get that it didn't really matter tot he solve as he kept fine track of what was supposed to be there, but anyone else mildly annoyed that he took so long to color the two squares in the bottom right area that had to be part of red regardless of which way around the 12 pair went?
I was also annoyed about it
Yes! I was quite bothered.
Took me a while to colour in those two as well.
Yes, as well as the fact that the top right needed to have two cells in purple as blue only needed one more square. Grrrrrr.
Same
You know you've released this right in the middle of the England-Denmark football game and I choose CTC 👍
it's a no-brainer...
I got a "watching the footy?" text. "No, I'm watching someone solve a difficult Sudoku", which is a perfectly normal response in my world :)
Oh. Football thing. That’s why it’s gone quiet.
Since Phistomefel does not play for Denmark or England, and not even for Germany, there is nothing to watch.
Delayed gratification has its points too. I watched the game first and THEN went off to CTC. Yaay England; hard luck Denmark.
I really want a montage of Simon saying things like "Now we know that blue and yellow are different colors!"
If I face a disciplinary hearing for not performing at my job as I should be I'm coming after you, SImon and Mark. How am I supposed to focus on work when there is so much fun to be had at CTC??
Amazing puzzle! love the logic required to construct the different regions. i found it quite tough but quite fun :)
Too my surprise, I managed to solve it within ~90 minutes, including a restart due to an error I couldn’t find. Tricky, but extremely fun puzzle!
What a lovely solve, it really showed the incredible logic here at play. Jesper, you've done a fantastic job setting this puzzle, even if it was a bit beyond my limited timeframes haha.
A huge mazel tov on reaching 5K patrons. I am proud to be among them. You have given countless people hundreds of hours of top level brain exercise - with good cheer and Douglas Adamesque sense of humour. I want a chocolate teapot as the reward for the jubilee whodunnit.
Brilliant puzzle! I thought at 17:30 you could have also said that whether the 20 clue started with 3 or 5, because the two clues sum to 43, you would need two 1's in the region to make it work.
An hour of solving can really boggle the mind, in the end row nine was filled all but the nine and you began to panic. Great solve and much respect for your endurance on these longer puzzles
What a brute of a puzzle, but an utter joy. First time I broke it about three-quarters of the way through, but it was such a clever puzzle was willing to give it another go.
59:50
yellow area must have 9 on R3, which makes 78 pair in C1
78, 47, 67 in R1-R3 kills 7 in R3C5
putting 4 in R1C2 would end up with 3 and 5 in R6 and R7 forcing to put 3 and 5 onto 24 sum in R8, so R1C2 must be 3
67:16. What an amazing puzzle! I got stuck for the longest time where I just couldn't solve the lower right quadrant (3 squares in the 28 X-sum region and 23 X-sum region combined). I spent 30 minutes getting to that point but about 20 of it was trying to figure it out. Then I paused, went away for a while and came back. I reset the puzzle, because I needed a clean slate and was able to see the logic which had eluded me earlier. And from there everything followed, but there was a lot of tricky logic involved. It never felt straightforward.
But once I got to the end, it was an amazing feeling.
Thank you, Jesper for the great puzzle.
This video and puzzle was absolutely wonderful, also definitely had the most amount of times I've said "Cmon Simon Notice x!!!!!"
46:54 "Is there something easy we could do at this point?"
Yeah, I was asking you to paint two cells red (R8C7, R7C7) for about 10 minutes now and also in the top right corner paint R1C8 and R1C9 pink ("purple"). It's not much, but it's easy and useful at least as far as making progress goes.
69:30 for me!! What a great puzzle!! I always find it so incredible when puzzles like this one have a unique solution and a way to get there from a completely blank grid. My respects for the setters who are able to create such beautiful puzzles.
I want "Stupidly Clever" on a mug or t shirt
I hadn't realized that's an oxymoron, cool
Could you possibly have a chat with your app devs to make the colors slightly less saturated and slightly lighter? The red really clashes with the blue pencil marks. The same is a little true of orange, blue and purple too. Even changing the hue/brightness values by 10% to make them a tiny bit more pastel would probably work - I don't mean full-on near-white or anything!
One idea might be to let the user choose what color the pencil marks and full on grid entries are, as opposed to always being blue.
@@terracottapie Of course, but if this was the case, the point is more I'd like the channel itself to use a lighter option if there is one. It's not even AS bad on the app itself, but with TH-cam's compression, between the slight biurring and color flattening, it makes it very hard to see here.
Hooray! I waited about 40 mins for you to ask where does the 2 go in green?
Same.
To reiterate a theme from your video with Rob Stephenson it is encouraging watching you find your way through difficult puzzles with good humor. Giving your viewers hope to get through their challenges with good humor as well.
I really like this puzzle, BUT... I had an issue with the two question marks at the bottom of column 9, where a certain solution seemed accurate... but it wasn't apparent how it broke the puzzle until maybe a half an hour later, when I was 1:14:00 in. So now I have to restart.
Restarted, and solved it at 49:29 on that attempt.
At 35:40 why is Simon so certain that the yellow regions turns upwards and not downwards?
Since the grey cell could be red, I don't see a way to decide whether yellow goes up or down.
What is the logical path to determine if yellow goes up?
20:00 For Blue (3 cells summing to 20) and Yellow (4 cells summing to 23) to join, you would also have needed get the remaining 2 cells to sum to 2, which would have been quite hard.
at 4:53, when trying to describe the example - "look at the 10 clue HERE", would be much easier to follow if the cursor hadn't switched to a single pixel.
It always does that for Simon when he drags an image on top of his browser windows. Maybe something to do with his image editor software?
@@stephenbeck7222 Yeah, possibly a conflict between the image editor and the streaming software? maybe opening the image in a different browser tab would reduce some variables.
At the 1-hour mark, where Simon comes dangerously close to resorting to bifurcation,
There is still work to do on the bottom row, placing a 2 and getting an 8-9 pair.
Also, in row 8, columns 5-7, (using the 24 clue outside the grid), Simon couldn't determine the combo to add to 18, BUT, he could eliminate 6 as a candidate in column 7, and narrow down the digit in column 5.
I think Simon could have eliminated the 6 in r8c5, and once he did, he could determine that row 8, cols 2-4 had to be 6-5-9 (you couldn't pair a 6 with an 8-6 combo nor a 7-7 combo).
In sum, I know there was lots to still be had on the bottom two rows at the 1-hour mark, and I'm guessing Simon could crack the puzzle at that point (in time), at that point (in the grid). LOL
I know he felt under time pressure.
I didn't even try that one, but I decided to try out that Mystery hunt and I'm happy to say that I solved the first puzzle with no issue. Getting that killer sudoku app was a good investment.
12:12 I love how the secret is spoken about on this channel.
19:41 There are two reasons for the blue (20) 3-cell region and the yellow (23) 4-cell region to be in different regions:
1) The 3-cell region has to be 3, 8 and 9. The 4-cell region has to be 4, 5, 6 and 8. They overlap in the 8.
2) The 3-cell region is a 20 X-sum, and the 4-cell one is a 23 one. You'd have 43 in two cells, and you can't have two 1s to round the region.
The little annoyance when there's 9 colors in the software and Simon prefers to repeat colors for different regions, lol.
Keep up the great videos!
Light grey & Black is not a good color in my opinion
I dont know if he figures this out later as I paused to video to comment but at 16:00 when he tries to figure out if it can be a 5, it can’t (unless I’m wrong) because that would make both question mark clues 1s in the L shape region
At least Simon doesnt apologize for apologizing.
Alternative title: Simon anthony trying to disproof the 4 color theorem
Never in my wildest dreams did I think I could beat Simon at a build-you-own irregular sudoku, but I solved the puzzle in under 45 minutes. It's amazing how much sudoku logic makes sense to me when I'm operating on only 5 hours of sleep.
Great set and solve 👏👏
14:29 the 23-clue and 28-clue regions can also never be joined because they would be adding up to at least 51.
Not to mention that if you make the first cell in the 28 clue a 5 or 6 and join the 2 regions, you'll get 2 5s or 2 6s in the region.
Not necesarisialyrily, because they would have to have a cell that was joining them ... if the cell that was in both regions was a 9️⃣ then you've only got digits adding up to 42 at that point.
Congrats on the 5k Patrons! Fantastic!
At @33:50, how can you know that the 89 square is purple? Could it not be a yellow 9? I’ve been staring at it for a bit but if that’s sound logic I simply cannot see it...
I think if you make it a yellow 9 the 9 in purple can't go anywhere
He already knew exactly 3 squares in the corner area were not purple. Once he gets the 12 pair, he knows that those 3 cells must be in rows 7 & 8; one in the row with the purple 2 and two in the row with the purple 1. In fact just moments later he fills in the cells in row 6 & row 9 all as purple. Not sure why he realized it for r6c9 before the rest, though.
Thank you! Simple now that I see it. :)
@@thecommexokid That's not quite right. He shows a viable solution at 30:22, where the 1,2 pair are NOT both purple and the three non-purple squares are NOT in rows 7 and 8.
The reason he marks r6c9 as purple first is because there are only two places in the 3x4 corner region, that purple inhabits, that can be 8 and 9, namely the two cells in column 9 either side of the 12 pair. Therefore, these have to be the purple 8 and 9. Only once he knows r6c9 is a purple 8 or 9 can he rule out the possibility shown earlier at 30:22, which then allow him to mark the other cells in rows 6 and 9 as also purple.
As much as I’d love to watch this video right now, I’m busy doing the murder mystery sudoku hunt! I’ll have to watch later :)
Brilliant puzzle
Great solve by Simon
I want to see Simon vs Mark Sudoku set challenge. Mark creates one that Simon solves and Simon creates a puzzle that Mark solves. I would be interested in seeing what kind of sudoku puzzle Simon would create.
20:00 I am just watching and kibitzing a bit. Simon ruled out joining yellow and the 345 strip above it by noticing that yellow needs a 5. But the two totals are 23 and 20, so can never form a 9-cell region. The same does *not* apply to marrying the 3-strip to yellow. Although a 7-cell region cannot add to 43 ever, it could take in one more cell and become a possible 8-cell missing a 2. That would leave an 11 or 12 cell region in the south-east which then must be one 9-cell region and 2 or 3 cells to be absorbed by red.
The reds and yellows *must* take, between them, 3 cells from the south-eastern region and the south-east must have its own 9-cell region.
I think that is right.
Everyone knows Maverick committed the crime.
Note: I have not solved anything, so if this is in fact the solution, don't comment about it please, let's not spoil things for people.
Note: commenters really do discuss solutions in the comments and to expect otherwise is actually unfair to the people that wish to discuss solutions. Anyone who expects otherwise will be disappointed whenever it happens because it does happen.
Rather than expect this from everybody else,
accepting things the way they actually are goes a long way towards maintaining peace in
one's life.
it's better to avoid the comment section until you have either solved the puzzle or given up That way, a puzzle need never be spoilt for you, and also you never need to be disappointed by the realisation that there's no way to make other people conform to how you wish they would behave.
@@tadperry1817 my point was more about making sure the joke wouldn't inadvertently spoil a fun reveal for people.
If this comment had been a serious discussion about the challenge, then I would not have included a warning. But here it was meant to be something people can enjoy as a joke, regardless of their intentions to complete the challenge or not.
Or I may just be a bit optimistic. 🤔
@@tadperry1817
The best is to mark spoilers and/or have a separate thread for those who want to discuss the solution.
@@ragnkja In all things, you will never be disappointed if you don't expect others to conform to your idea of how they should behave.
Not all people are going to provide spoiler alerts or avoid discussing solutions, and being aware of that is the best idea as a reader of the comments.
It is not on them that you avoid spoilers, it is on you.
@@tadperry1817
The users of the CTC Discord server are actually pretty good about marking spoilers. If you click on them to make them readable, that’s definitely on you.
And while TH-cam comments may contain unmarked “spoilers” for that video, it’s usually easy enough to just watch the video before reading the comments - after all, it’s right there, and you have to actively open/scroll down to the comments to read them, so they’re not that hard to avoid if you want to avoid spoilers. So it’s really on both parties. If a TV Tropes page says that “due to the nature of the trope, spoilers will be unmarked”, it’s up to you to decide if you want to take the risk that a work you don’t want spoiled is there, or if you want to avoid reading the section for a particular medium, or even the whole page.
At 34:45, I isn't necessary for there to be only 3 purple cells in that block of 6. It could be 4 if one of them wrapped around adjacent to the 1 clue. That would let the yellow 3 into column 7, row 6.
Whoops, nope, that would strand a single cell next to the 1 clue. Carry on!
Really amazing puzzle.. I took over 80 mins and realised i had made a massive mistake... who knew that 3 does not equal 4 :O... anyway I did it again and it took me 50 mins... really good puzzle.
I would have colored more regions earlier (some 2-color) and more simple deduction. For example in the bottom green row the 2 and the 8,9 pair could be filled in easily. With 2 already in green, blue, orange, yellow and purple, it can only be part of grey. And it has to be on either row 2 or 3. And that also solves where the blue 2 can go. That determines the blue zone and fill in 2 more purple colors.
34:23
"Stupidly clever" has to be one of the funnier things you've said
As always, a delightful video. To answer your question, Simon, the ‘bifurcation’ moment could have been easily avoided with the following: 3 must go in the top row, second place. It couldn’t be a 4, which would connect it with the orange below, making the 4-5 pair on row 4 break orange (two 4s in the same color). That fixes the 3s and solves the final yellow square. Hope that’s useful.
A grey 4 in r1c2 could leave the possibility of a grey 5 in r4c2 and an orange 4 in r4c3, no? That doesn't give you two 4s in orange, so doesn't break (unless I'm missing something).
Took me over 90 minutes (92:08), and I may have resorted to some bifurcation late in the day ... but so help me, I solved this one!
Insane puzzle!
just finished the last simon video, this now looks quite interesting. in this storm, i can't imagine anything better than watching sudoku
20:00 The two question marks on the right side of rows 7 and 8 are a 1/2 pair. Let's see how long it takes him to realize it...
They can't be 4 because they would meet up with red and repeat the 4. They can't be 5 because they would (again) meet red and complete the region, isolating the bottom corner as a region that is too small. They can't be 3 because they already see a 3 in the column. The only things left are 1 and 2.
Less than 5 minutes. Not bad.
"being likely isnt the same as something that we can force to happen" I run into that problem with every sudoku that takes me too long. start going by the likely option then have to redo the last 10 deductions when the chain i was on runs into a problem
Which is what we call „bifurcation“. As sometimes seen in Mark‘s videos (his speed in solving still is incredible, even if he takes wild guesses, though).
19.09 you can also say that blue on row 4 amb yellow of row 5 cannot be the same colour because they sum 43 and still need 2 cells more, it s another way to see it
1:06:36 finish. I got off to a faster start than Simon with the logic, but slowed down near the end allowing him to catch up. That's what typically happens.
How do you know placing the green 2 would not have helped sooner ?
:D
That was a blast!
17:00 my take: if orange and yellow is one region it will brake straight away, reason: the first number from 20 must be 3 or 5, and yellow region is locked with 4568 meaning orange region cannot be 389 because double 8, nor 569 or 578 because the double 5.
21:33 that 11 and 20 cannot be same region, reason: if 11 is 29, the 9 sees the 389 in the 20, if 11 is 317 or 326, the 3 sees the 389 in the 20.
Brilliant thank you
needing that 24 clue at the end was evil.. i was stuck there at the end for about 10 or 15 minutes thinking i was missing some sudoku...
10 or 15 minutes, ha! I solved the thing all the way down to some 7/9 pairs without the 24, decided it was indeterminate, came back the next day and then saw the 24 clue.
45 minutes flew by
Edit: I panicked so hard trying to figure out how in the world he was going to solve it in 6 mins
Its comical that "I dont wanna know about a two if it wouldnt have done anything" and there was a two that wouldnt have done anything :P
Got it in under 3h... Well, I have a lot more to learn and speed up.
One thing though: when you got stuck and "bificated", you could have seen the 24 clue much earlyier. You had to have 20 in 3 cells but on the right side you already knew you needed 3 large digits (79/789), therefore 20 had to be built with only 1 of those numbers. The only way of doing so is by using 965.
So you tried it the other way around at the end (and still got it right ofcours). :)
Does anyone else get bothered by Simon's refusal to make the 2 cells in column 7 red?
It's the first time that I've literally yelled at the screen. We knew the region of those two cells for so long!
Frustrated at Simon for not coloring them but even more frustrated at myself for fixating on it when I can see perfectly well that even if he colored them it wouldn’t lead to any new deductions.
What in the hell happened at @12:34 that through a purple 5 into the grid?
28:30 The answer to that is yes, not no. If you take and make the cell that takes yellow, r7c9 a 1, and then take a red on r8c9, this is entirely possible.
My favorite puzzles are always these ones where you need to color in the irregular sudoku as you discover the regions, but they take me so long to solve...
How did he know @12:15 the 3 needed to be there and not 4 or 5? Was it because he knew a single gap was needed and therefore it could not have been 4 and 4 since that break sudoku and then 5 or 6 left made it so 3 was the only thing leave a gap since 4 would meet 5 and overlap with 6? I think that is the logic yes?
@46:15 the creepy Simon look came out again
23:00 Simon asks a question the answer to which is yes!
At 41:35 he assumes green and blue are not the same to color r7c4 green. No real harm done, he shortly afterwards starts thinking about whether blue and green are different, and even though his explanation why they are not still assumes r7c4 is green, his reasoning would work just as well if it had been left uncolored. And once we know blue and green are not the same region he can color r7c4.
Silly question, I solved a sudoku solver from Cracking the cryptic. When I finished solving I received a popup that said congratulations, then it asked the following: Did you spot the Empty Rectangle on 7. I am still learning notations when it comes to locations on the puzzle and don't know what on 7 means. I am used to notations where rows are letters and columns are #s. When it just gives a number is it talking about an entire 3x3 area? As in 1,2,3 next row 4,5,6 next row 7,8,9? Thank you.
I don't know for sure, but I would expect it means empty rectangle that could be applied to the digit 7 somewhere in the grid, to either place a 7 or remove a 7 as a candidate.
For cell references, I think they always use the convention of r4c7, for example, to mean row 4 column 7. And if referring to boxes, they're numbered left to right, top to bottom, so
1 2 3
4 5 6
7 8 9
31 minutes and 33 seconds in and I realize there's an 8 that can be filled in... LOL, that's the only number I got so far, and I got it a completely different way than he did. I'm so proud of myself :p
I would usually watch both your and Mark's videos once I get home from work, but I'm afraid I've been too busy bobbins-ing up the Patreon Murder Mystery puzzles.
Love Love Love 💕
Realising the last yellow had two possible cells (3 or 5) was enough to simple use colours rather than numbers to allocate further cells their assigned colour.
First rule of the secret is, we TALK about the secret
"Let us know who committed the Crime" Sure ... Mark in the conservatory with a spanner wearing purple. Let see how wrong I can be.
1:02:58 for me today. Even with a huge 15 minutes gap where i was second-guessing my logic of the lower right corner. This was hard, but i liked the logic.
Knowledge bomb for you Simon: 20+23=43.
I had a mistake at the end, so I debugged it for almost an hour to see where things went wrong. I miscolored the field in the lower left corner very early. Everything went well and then an hour later I lacked a 4
I'm never cross with you for apologizing, Simon. I just want you to be kind to yourself even though it's hard
I discovered, in solving it myself, that there is an alternate solution by which the 3/5 pairs in the bottom right can switch over
No there isn't an alternate. If you switch 3/5 then you MUST switch all 3/5s - this puts a 5 at the top of r2 which is wrong. It puts a 5 at the right end of r4 which is wrong. It puts a 3 in the bottom left corner which is wrong. It puts a 5 at the bottom of c5 which is wrong. It puts a 5 at the left end of r5 which is wrong. Swapping 3 with 5 puts SIX broken digits in the edge cells of the puzzle - broken in that those digits have to match the number of cells they can 'see'. ie the bottom left cell sees 5 green cells & thus you can't blithely stick a 3 in there instead!
It's a useful exercise if you believe you have an alternative solution to find the point in the video where Simon's solution first diverges from your solution. Since Simon only enters a digit when he believes it to be forced, you're effectively saying there's a point in his solve where he makes a mistake. I haven't seem anyone else in the comments suggesting Simon's logic was faulty at any point.
"The 3/5 pairs in the bottom right can switch over" doesn't help, I'm afraid. There are no 3/5 pairs in a 'deadly pattern' that could simply be switched over without impacting other digits.
"Stupidly clever!" (с) Simon @34:24
"oh nooo...that was so close to being interesting"
Me @ all my hobbies
BTW, the colour is pink. Not purple. Its PINK
It has blue & red in it thus it's purple. Pink does not have blue in it.
I can't even manage to get past the instructions without peeking from the video...