@@studgerbil9081 Yah it was frustrating to watch, must have been like 20 of them. Also, yes he has done a similar puzzle, and yes it had that kind of entropic roping.
Yes, as soon as he applies the rule, he starts doing sudoku all over the place. At a certain point each box would have largely fallen out independently. Hilarious. Arch Simony. Bless.
Truly Simon. "Can't see the trees for the forest." He misses stuff like this regularly, but also regularly sees stuff "instantly" that takes me minutes to understand, let alone see. Even when he does finally spot the 6 in the middle.....cleans up the box....and leaves the pencil mark 9. @35:00....comments on James brilliant progression with the 2....and 3....then ignores the 4 pencil marks (which could've been done via the 56 pair, or the 2 & 3)....but notices stuff on the far side of the grid. Even when he returns to box 9 -> 8 logic....still doesn't see it. But again, continues spotting much more complicated stuff without hesitating much. Such as at 37:27.....uses the "chaperone" trick (not that complicated) to place the 9.....but the 6 in row 3 does the same thing in (imo) a slightly easier to see way. At 37:56, "I can place 4 in box 4". Yes, with excellent logic....or just place it in box 6, then box 4. And, finally...after doing a few more 4's....points out the x-wing on 4s and STILL overlooks the 3/4 conflict in box 8. All that said.....it is a great way to learn! Loving it!
40:22 “we could have done that a different way.” Meanwhile, my wife is wonder what sort of video I’m watching that is driving me so apoplectic. That 4 was trapped from 35:08 onward.
Have been away from the channel for a while and forgot the joy that is Simon making complicated logical deductions in a second... and then missing write-in digits for what feels like a lifetime
Still amazed by how fast Simon manages to forget rules literally right after using them. Cant put 3 next to 1, places 3 next to 456 and doesnt immediately say so that is a 6
These guys are incredible. I absolutely love the way he never tired of explaining the logic fully, every single time. These puzzles seem impossible, yet he makes them make sense! I even thought I might give it a go, but I can't even solve the first one on the app. Down to six cells with three digits between them, and both ways to arrange them seem equally valid with no conflicts 😢 Completely baffling to me. I guess I'll just stick to videos 😂 Keep up the great work, guys!
Definitely outstanding construction. Innovative, stratospherically cool and drastically minimalistic. Thank you *James Kopp* for designing it and thank you *Simon* for featuring so many *cosmic class* artworks on this channel. I hope this will be included in the next *CTC Cosmic Hits book.* 👏👏👏👏👏👏👏👏👏👏👏👏
I just love your delightful, jovial, quirky nature Simon. Every now and then I'll try to watch someone else solve something and as clever as they are they do tend to be a little.... flat in comparison. You're a gem
If Simon focused on completing a box rather than corner-marking the low digits, he probably would have noticed the roping in both directions throughout the grid. That really sped up my solve.
WOW. With basically 2 rules and 1 given digit. You might as well call it a Miracle Sudoku. This puzzle is more of a discovery than a constructed puzzle.
33:07 First I didn't realize what straight meant though the different of 3 is quite forcing Then I realized the difference of 3 only applies within boxes And then I peeked at the video and found out what straight meant And the I still thought it was impossible, but somehow it solved! Rollercoaster of a puzzle
I finished in 43:02 minutes. This was such a fun puzzle and a surprisingly smooth solve, considering the sparseness of the grid. I think my favorite part was the break-in utilizing the 3 in the corner to prove which side of the board the arrow was on. That was so fun to see. The rest felt very smooth as well. Great Puzzle!
Slowly savored this sudoku in 04:03:46. That included a subtle error discovered about 2 hours in that forced an almost complete reset. Extremely enjoyable puzzle though! I got a lot of work done thinking about box middles. Except for 1/2/8/9, each digit requires am exact set of four surrounding numbers when placed in the middle of a box, so if one of those four is already placed in a corner, it immediately eliminates that option from the middle.
I believe you are one of the most brilliant people I have ever interacted with (with a very loose definition of interaction being I have watched your videos), having said that you have also provided me the greater number of moments thinking or even shouting : "Come on Simon this is f***ng obvious, what is going on in your brain? "
Simon quite handily proved why he isn't normally allowed to solve James' puzzles, because clearly his solve style is disastrously incompatible with it. I'm sure Mark would have solved it in half the time, and at the very least I would have spent a lot less time shouting at the screen haha. The fact that the 5 and 6 in box 7 were placeable for over half the solve and he NEVER noticed it hurt me on an emotional level, as did the 6 in box 1
I don't know if the last few sudokus are more approachable, but I'm being able to complete almost all of them without looking at Simon's solutions :D As always, amazing video!
Two weeks ago I posted a comment hoping someone would make a “That’s 3 in the corner” puzzle and someone commented spoiler alter. And here is it! I can’t wait to watch!!!
There is an easy trick that can be applied to all boxes after you have the 1 in box 7, and that is to consider which digit can be in the middle of each row pair in a box. From there on it is straightforward, and it all solves in a beautiful way.
No, that's how I attacked this. There's one pattern (of course possible to rotate or mirror it) for 5,4/6 ,3/7 or 2/8 in the middle, and 3 for 1/9 in the middle. Only one pattern gave the 9 and 1 position that the diagonal needed.
@@henk-ottolimburg7947 I don't think I explained what I meant very clearly. I mean that adjacent digits cannot be of the same entropy (1-3, 4-6 or 7-9) so if you ignore the values and just map the entropies, each box has very few possibilities (in fact in the finished grid there are only 2, disregarding rotation etc). For example: - Box 1 has 4-6 in a downward chevron shape, 1-3 below that in another downward chevron, and 7-9 pointing up in a double-height chevron. - Box 4 has 7-9 in diagonal positions, 1-3 in a north-west pointing double chevron, and 4-6 in a south-east pointing double chevron. Those are the only two box layouts (I think they might be the only two that are physically possible) which isn't a huge help initially but once you start to get _some_ digits in the grid, you can very quickly start to see where the entropies must lie.
@@bobblebardsley Yes, I came here to say the same thing. There are only 2 configurations of low/middly/high digits within a box, and after some time that becomes a powerful tool for quickly filling boxes with digits!
@@penningmeestercgkdelft9159 I remember seeing something similar in a sudoku Simon did awhile back where each 3x3 box was a mini sudoku with its own ruleset. I don't remember exactly why it was the case in that grid (or parts of it) but it just triggered something in my memory when I saw what the non-adjacent rule was doing here. Thanks for confirming!
I found this lovely way of constructing all boxes that obey the adjacency rule. The consecutive digits are patterned in order diagonally, but skipping a diagonal row as they travel. Sort of "wrapping around the box". This is course works with reflecting across a box diagonal or cyclic permutations of the digits. You just need to figure out which direction the wrapping goes. What would seem to work but not obey this, but still be a valid box (e.g. putting 7 or 8 the wrong way around in box 1) will get ruled out by the interaction with the other boxes (effect of the roping) I painstakingly pencil marked using this, got stuck for veery long since I didn't use sudoku to rule out the bottom arrow... After that though, the rest of the grid fills itself with the aforementioned construction tactic!
Spoiler alert: Not only is there roping in the columns of the final grid, as Simon acknowledges, but there's also roping in the rows, even though it's not as dramatic.
At 33:15 another beautiful way of getting the 5 by row 3: 4 cannot go into r3c6 - r3c8 because there's only one digit left to go next with it. Following that the 6 can also not be in r3c6 - r3c8 because it can only go next to a 2 and a 9 and if it goes with a 9 there it would force the 4 into the wrong position. So r3c5 and r3c9 are a 4/6 pair, r3c6 and r3c8 are a 2/9 pair and r3c7 is a 5.
What a fun puzzle! Also, the first time I wound up with a completely different break-in from Simon! I worked out the two possible lines, and then starting playing around with possible patterns for the digits in box 7 (with the 3 in the corner!). I realized it was very hard to make an arrangement where the 1 and the 9 wouldn't break both lines. There was only one arrangement that only broke one of the lines, and I could find none that only broke the other. That quickly gave me all of box 7, which translated pretty quickly to boxes 1 and 9 because of their placed digits along the line. From there it was mostly sudoku with a few excursions into the orthogonal difference constraint.
At 8:50 the only thing I can think of is: Is that all the rules!? Let's see how this gets cracked! Edit after the break in, the way that the three in the corner is breaking symmetry is quite unexpected 🤯
Heh, I spent 4 minutes frustratingly begging you to see the 6 in the top left box :P. Meanwhile you were finding plenty of other things I would have missed :).
Very enjoyable puzzle, amazing how you can fill the grid with so little to start with. Thank you James Kopp for sharing your cleverness again with us mere mortals ;) Also, spoiler below... ... ... ... ... ... It's interesting (though not surprising with that 3+ constraint) that we have roping going on both horizontally and vertically
I enjoy this kind of solve. Where your 30 minutes in, have 20 digits and a hundred pencil marks. Then you hit a tipping point and the whole thing solves in three minutes.
The way you didn't see that the three couldn't go next to the one on the bottom is exactly how I see all digits at all times in all rulesets. Haha it's funny like that.
6 minutes 15 seconds between placing the 3 in box 1 and realizing r2c2 is a 6. Longest 6 minutes and 15 seconds of my life. Felt like the video should have almost been over.
59:18 ... I had a few false starts (I forgot that the 3-difference rule didn't cross box borders) but finally found my way to the end of this miracle-esque sudoku Nice puzzle!
Once the ground work is in place, you just go through one digit at a time. "Where are all the threes? They must be here. Okay, where are all the fours?" Every digit could be finished or nearly finished before moving to the next one
My method is a little different from Simon's. My deduction is based on below conclusion: The middle digit of the box is important. Its neighbor and neighbor ^ 2 digits should be in the corners of the box. So for middle digit 34567, its corner digits are fixed. For 28, it has 1 digit free and for 19 it has 2 digits free. After I get the box 1 done, I know the middle digit of box 2 can only be 2389 to put 1 in the corner. 2 doesn't work for sudoku. 3 doesn't work because the 45 in box 1 push both 4 and 5 to r3c6. 8 doesn't work because 9 in box 1 push 9 to r3c6. Then the candidates for row3 of box3 is {2456}. We can not find a valid solution from this set. So the middle digit of box 2 is 9! The same method can be applied to box 6.
That was unexpected, this design w/o the given has only 2 solutions+2 mirrors w.r.t the D- (4 soln total). 7 boxes are completable leaving pairwise soln in the last 2 boxes diff by the given 3. Not having the constraint across box orders fooled me at first tho
I really struggled with this, I had suspected that the rows or columns were going to have one of those hidden rules where each number had to be from 123, 456, and 789; but I couldn't find a good way to prove it so let it go... took me almost 2 hours to solve this, I almost wish I had just gone for, and played with that rule, would have solved it so quickly.
at 30:45, if it's a 5, the neiboring digit is than an 8 and the 6 is placed above the 3. You can apply the rules to set the 4 away from the six and the 2 finally has no place in the box proving it imposible.
I would love to watch you guys play The Last Express...murder mystery type game on the orient express, thought to be one of the best games in its genre. It uses stop motion graphics so you won't get the nauseous effect like some games. The special thing about it is that it uses a real time game mechanic. I played it multiple times years ago when it first came out and I just bought it again on Steam since it is on sale to play again!
43:31 for me - I got stuck at the start trying to figure out which line was the arrow; I was caught up in checking how the 3 would affect the arrow that I didn't check how the arrow affected box 7.
U can try stripping the antisymmetric components(the given 3) Orient yourself with the line...and then DD where the given 3 goes and reorient if needed
In the bottom row there is a 789 triple and the 9 in that triple is in R9C7 or R9C9. So it cannot be in R8C8 Another way to look is that if R8C8 is 9 then in the bottom row - the three cells R9C2 R9C7 R9C9 are 78
Holy missed pencil marks Batman… I’m not real good at these, but this one was a little tough to watch lol. Sorry Simon still watched and gave a thumbs up. Still waaay better than I!
Now everybody's going to start using "Within a 3x3 box, orthogonally connected cells must have a difference of at least 3" as the thing which finishes the entire puzzle after the break-in
I missed the break-in because I connected cell rule applied across boxes and got hyper-fixated on that for 25min sigh. 47:55 in the end after finally getting the break-in.
56:10, had to undo after being confused, looking at the video, being even more confused, then rereading the rules to see that the "difference of 3" rule ONLY applies within boxes.
This is just absolute peak “I’m going to ignore the clear implications of the digit I just got and decide to look across the grid instead” Simon
Yes holy moly. This was a PAIN to watch lol. Siiiiiiiiiiiiiiiiiiiiimon!!!!!!!!!!!!!!
"My phone's buzzing" at 22:09…probably Mark telling you to look at Box 1
Yeah, that six in the middle square was driving me nuts for five whole minutes.
Yeah… ever since he found the line I had been losing my mind staring at the 6
Simon seamlessly switching between using the 3-different-rule one second and totally ignoring it the next is such a Simon thing to do...
and he does it throughout the solve from start to finish. some rules are kryptonite to Simon and Mark.
@@studgerbil9081 Yah it was frustrating to watch, must have been like 20 of them. Also, yes he has done a similar puzzle, and yes it had that kind of entropic roping.
Yes, as soon as he applies the rule, he starts doing sudoku all over the place. At a certain point each box would have largely fallen out independently. Hilarious. Arch Simony. Bless.
Yes so fun to watch he see it one row but not the row under the number and so on, like 4 in box 8 :) (good to know he is human sometimes) =)
19:36 Simon being Simon…”they’re all at least 6”. Middle cell of box 1: “ 🤔🧐”
jeps. but that is not next to a 3, it is above it 😁😁😁😁 , he also placed the option 9 in the box.. with a 9 in the box
@@michaelandersen-kk4fca 9 which was his very first digit!
Truly Simon.
"Can't see the trees for the forest." He misses stuff like this regularly, but also regularly sees stuff "instantly" that takes me minutes to understand, let alone see.
Even when he does finally spot the 6 in the middle.....cleans up the box....and leaves the pencil mark 9.
@35:00....comments on James brilliant progression with the 2....and 3....then ignores the 4 pencil marks (which could've been done via the 56 pair, or the 2 & 3)....but notices stuff on the far side of the grid. Even when he returns to box 9 -> 8 logic....still doesn't see it.
But again, continues spotting much more complicated stuff without hesitating much. Such as at 37:27.....uses the "chaperone" trick (not that complicated) to place the 9.....but the 6 in row 3 does the same thing in (imo) a slightly easier to see way.
At 37:56, "I can place 4 in box 4". Yes, with excellent logic....or just place it in box 6, then box 4.
And, finally...after doing a few more 4's....points out the x-wing on 4s and STILL overlooks the 3/4 conflict in box 8.
All that said.....it is a great way to learn! Loving it!
Lol screaming look at box 1.
I about lost it when at 19:47 he just ignored the square above the 3in Box 1
Thank the Lord you were screaming. It made all the difference.
I WAS SCREAMING TOO. I FELT SO PROUD BUT ALSO ANXIOUS I SAW IT BEFORE HE DID. Literally was screaming
row 2 column 2 cant be 4 or 5 because the number 3 too close to them so it must be 6
It felt like I was watching Simon ignore 4 in box 8 for an hour
40:22 “we could have done that a different way.”
Meanwhile, my wife is wonder what sort of video I’m watching that is driving me so apoplectic. That 4 was trapped from 35:08 onward.
Have been away from the channel for a while and forgot the joy that is Simon making complicated logical deductions in a second... and then missing write-in digits for what feels like a lifetime
Still amazed by how fast Simon manages to forget rules literally right after using them. Cant put 3 next to 1, places 3 next to 456 and doesnt immediately say so that is a 6
When he did remove the 5 in center box 9 just to tease us more
Rules: 05:35
Let's Get Cracking: 08:42
Simon's time: 33m17s
Puzzle Solved: 41:59
What about this video's Top Tier Simarkisms?!
Three In the Corner: 6x (00:28, 00:30, 05:31, 15:24, 15:26, 42:14)
Bobbins: 1x (38:57)
Cooking with Gas: 1x (24:19)
Maverick: 1x (11:20)
Phistomefel: 1x (02:19)
And how about this video's Simarkisms?!
Pencil Mark/mark: 12x (11:26, 11:44, 11:46, 14:11, 16:53, 21:05, 23:06, 27:51, 28:35, 31:20, 33:43, 37:50)
By Sudoku: 7x (15:16, 22:42, 34:24, 37:08, 38:53, 40:05)
In Fact: 7x (06:35, 08:42, 20:31, 26:00, 27:22, 32:12, 37:13)
Cake!: 7x (04:08, 04:11, 04:40, 05:04, 05:13, 05:17, 05:21)
Sorry: 5x (04:38, 08:00, 17:19, 27:09, 28:14)
Beautiful: 5x (33:33, 34:31, 35:01, 37:44, 42:23)
Hang On: 5x (17:19, 18:10, 24:23, 26:46, 27:05)
Ah: 4x (17:19, 20:45, 38:05, 41:23, 41:23)
Good Grief: 3x (18:24, 27:24, 37:44)
Clever: 3x (14:50, 14:50, 34:58)
Naughty: 3x (30:16, 39:06, 39:09)
Obviously: 3x (15:59, 20:57, 28:31)
Nonsense: 2x (22:32, 28:53)
I Have no Clue: 2x (23:03, 31:35)
Lovely: 2x (29:01, 37:19)
Brilliant: 2x (04:43, 42:08)
Gorgeous: 2x (32:36, 33:50)
Wow: 2x (18:10, 29:01)
Symmetry: 2x (10:20, 10:46)
What on Earth: 1x (18:24)
Bother: 1x (35:24)
Naked Single: 1x (28:51)
In the Spotlight: 1x (00:32)
Take a Bow: 1x (42:05)
Shouting: 1x (04:46)
Bizarre: 1x (16:32)
Full stop: 1x (18:47)
Whoopsie: 1x (10:26)
We Can Do Better Than That: 1x (25:03)
Phone is Buzzing: 1x (22:09)
Next Trick: 1x (29:34)
What Does This Mean?: 1x (05:52)
Unique: 1x (00:50)
Weird: 1x (41:34)
Most popular number(>9), digit and colour this video:
Seventy Eight (3 mentions)
One (91 mentions)
Red (3 mentions)
Antithesis Battles:
Even (2) - Odd (0)
Row (8) - 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!
This is a version of a miracle sudoku, I would say, and just as fun to solve. Thanks, Simon, for the video!
3 in R3C2 is staring as hard as it can at R2C2!
43:13. Cool idea; very frustrating to actually solve because my mind kept wanting to apply the adjacency rule across boxes
If you’re wondering, he doesn’t spot it until 23:34.
I had to fast forward. I could never do what Simon does but it hurts my soul to see him struggle with something I could do!
This was amazing, brilliant work mate. The patterns in the grid at the end were fascinating
Love your content! I am now trying the daily NYT sudoku every day! :)
Those unplaced 6 in box 1 and 4 in box 8 were driving me crazy...
These guys are incredible. I absolutely love the way he never tired of explaining the logic fully, every single time. These puzzles seem impossible, yet he makes them make sense!
I even thought I might give it a go, but I can't even solve the first one on the app. Down to six cells with three digits between them, and both ways to arrange them seem equally valid with no conflicts 😢
Completely baffling to me. I guess I'll just stick to videos 😂
Keep up the great work, guys!
Columns are the same modularity and the rows are different modularity. Very clever James 🎉
I noticed the pattern in the corners almost instantly and it made this puzzle super simple after words, absolutely genius puzzle
I love Simon, but this one does hurt a bit to watch ;)
23:25 Yay, finally saw it ! 🤣🤣
Definitely outstanding construction. Innovative, stratospherically cool and drastically minimalistic.
Thank you *James Kopp* for designing it and thank you *Simon* for featuring so many *cosmic class* artworks on this channel.
I hope this will be included in the next *CTC Cosmic Hits book.*
👏👏👏👏👏👏👏👏👏👏👏👏
I just love your delightful, jovial, quirky nature Simon. Every now and then I'll try to watch someone else solve something and as clever as they are they do tend to be a little.... flat in comparison. You're a gem
Delightful! I did this as a coloring puzzle... rather surprised when you didn't pull out the crayons!
If Simon focused on completing a box rather than corner-marking the low digits, he probably would have noticed the roping in both directions throughout the grid. That really sped up my solve.
I've wondered for a while if you could make a sudoku with banding in both directions, and I was happy this puzzle showed me you can.
WOW. With basically 2 rules and 1 given digit. You might as well call it a Miracle Sudoku. This puzzle is more of a discovery than a constructed puzzle.
And one of the rules was even weakened - the three-apart condition is confined to boxes
33:07
First I didn't realize what straight meant though the different of 3 is quite forcing
Then I realized the difference of 3 only applies within boxes
And then I peeked at the video and found out what straight meant
And the I still thought it was impossible, but somehow it solved!
Rollercoaster of a puzzle
Very approachable. I used colours to start with and then noticed the roping and it then just wrote itself in
Darn, I was rooting for my favorite Simonism: "Playing Gooseberry"
40:00 "we're finding some patterns now". Classic Simon.
Solved it in 31:32 :-) It's really a kind of miracle sudoku!
Brilliant video, just what was needed today
as soon as I got 963 in one column in box 1 I figured the whole puzzel would have a modular structure
❤ this one!!! Brilliant!!
I finished in 43:02 minutes. This was such a fun puzzle and a surprisingly smooth solve, considering the sparseness of the grid. I think my favorite part was the break-in utilizing the 3 in the corner to prove which side of the board the arrow was on. That was so fun to see. The rest felt very smooth as well. Great Puzzle!
The center of each box is also unique. Nice!
Slowly savored this sudoku in 04:03:46. That included a subtle error discovered about 2 hours in that forced an almost complete reset. Extremely enjoyable puzzle though! I got a lot of work done thinking about box middles. Except for 1/2/8/9, each digit requires am exact set of four surrounding numbers when placed in the middle of a box, so if one of those four is already placed in a corner, it immediately eliminates that option from the middle.
I believe you are one of the most brilliant people I have ever interacted with (with a very loose definition of interaction being I have watched your videos), having said that you have also provided me the greater number of moments thinking or even shouting : "Come on Simon this is f***ng obvious, what is going on in your brain? "
I find these much easier when I don't start by forgetting a rule (in this case that the arrow was straight)
Simon quite handily proved why he isn't normally allowed to solve James' puzzles, because clearly his solve style is disastrously incompatible with it. I'm sure Mark would have solved it in half the time, and at the very least I would have spent a lot less time shouting at the screen haha. The fact that the 5 and 6 in box 7 were placeable for over half the solve and he NEVER noticed it hurt me on an emotional level, as did the 6 in box 1
I don't know if the last few sudokus are more approachable, but I'm being able to complete almost all of them without looking at Simon's solutions :D
As always, amazing video!
Roping in all directions
yeah, surprised Simon didn't see that as well as that damned 6 in Box 1
And they are all mod 3 equals
Miracle sudokus nearly always have roping. It's a product of the patterned rulesets they tend to have to make them actually solvable.
@@13vatra Not always, even not as a rule, but in that case all 6 box tripples in both directions are roped
Two weeks ago I posted a comment hoping someone would make a “That’s 3 in the corner” puzzle and someone commented spoiler alter. And here is it! I can’t wait to watch!!!
Brilliant but had me totally stumped!
"The 4 and 5 canoodle far too easily..." lmao
There is an easy trick that can be applied to all boxes after you have the 1 in box 7, and that is to consider which digit can be in the middle of each row pair in a box. From there on it is straightforward, and it all solves in a beautiful way.
I think we can skip 4 or 5 being next to 3, right? We can solve this with handicap
Is this one of those where there's only about 3 different patterns for each 3x3 box if you were to colour the entropy of the digits?
No, that's how I attacked this. There's one pattern (of course possible to rotate or mirror it) for 5,4/6 ,3/7 or 2/8 in the middle, and 3 for 1/9 in the middle.
Only one pattern gave the 9 and 1 position that the diagonal needed.
@@henk-ottolimburg7947 I don't think I explained what I meant very clearly. I mean that adjacent digits cannot be of the same entropy (1-3, 4-6 or 7-9) so if you ignore the values and just map the entropies, each box has very few possibilities (in fact in the finished grid there are only 2, disregarding rotation etc).
For example:
- Box 1 has 4-6 in a downward chevron shape, 1-3 below that in another downward chevron, and 7-9 pointing up in a double-height chevron.
- Box 4 has 7-9 in diagonal positions, 1-3 in a north-west pointing double chevron, and 4-6 in a south-east pointing double chevron.
Those are the only two box layouts (I think they might be the only two that are physically possible) which isn't a huge help initially but once you start to get _some_ digits in the grid, you can very quickly start to see where the entropies must lie.
@@bobblebardsley Yes, I came here to say the same thing. There are only 2 configurations of low/middly/high digits within a box, and after some time that becomes a powerful tool for quickly filling boxes with digits!
@@penningmeestercgkdelft9159 I remember seeing something similar in a sudoku Simon did awhile back where each 3x3 box was a mini sudoku with its own ruleset. I don't remember exactly why it was the case in that grid (or parts of it) but it just triggered something in my memory when I saw what the non-adjacent rule was doing here. Thanks for confirming!
I think there are only two possibilities. What is the third one?
Can it be proved that every sudoku with the 3-difference box constraint has modulo 3 roping? Knowing this does tend to trivialize these puzzles.
Was thinking the same about the roping. Looking at the solution mod 3 is not particularly constraining though as it doesn't apply horizontally
@@easternbrown That was my thoughts on reading this comment there is no horizontal roping
I found this lovely way of constructing all boxes that obey the adjacency rule. The consecutive digits are patterned in order diagonally, but skipping a diagonal row as they travel. Sort of "wrapping around the box".
This is course works with reflecting across a box diagonal or cyclic permutations of the digits. You just need to figure out which direction the wrapping goes.
What would seem to work but not obey this, but still be a valid box (e.g. putting 7 or 8 the wrong way around in box 1) will get ruled out by the interaction with the other boxes (effect of the roping)
I painstakingly pencil marked using this, got stuck for veery long since I didn't use sudoku to rule out the bottom arrow...
After that though, the rest of the grid fills itself with the aforementioned construction tactic!
It quickly becomes apparent just how powerful the restriction is. Heavy pencil marking gets the candidates for a lot of squares down in a hurry.
Beautiful puzzle, beautiful solve.
Spoiler alert: Not only is there roping in the columns of the final grid, as Simon acknowledges, but there's also roping in the rows, even though it's not as dramatic.
At 33:15 another beautiful way of getting the 5 by row 3: 4 cannot go into r3c6 - r3c8 because there's only one digit left to go next with it. Following that the 6 can also not be in r3c6 - r3c8 because it can only go next to a 2 and a 9 and if it goes with a 9 there it would force the 4 into the wrong position. So r3c5 and r3c9 are a 4/6 pair, r3c6 and r3c8 are a 2/9 pair and r3c7 is a 5.
just the title alone and i’m obligated to like the video
24:05 finish. A very nice puzzle, good logic!
What a fun puzzle! Also, the first time I wound up with a completely different break-in from Simon!
I worked out the two possible lines, and then starting playing around with possible patterns for the digits in box 7 (with the 3 in the corner!). I realized it was very hard to make an arrangement where the 1 and the 9 wouldn't break both lines. There was only one arrangement that only broke one of the lines, and I could find none that only broke the other. That quickly gave me all of box 7, which translated pretty quickly to boxes 1 and 9 because of their placed digits along the line. From there it was mostly sudoku with a few excursions into the orthogonal difference constraint.
I think this might be the first puzzle I could've beaten him lol
Very nice puzzle, thank you !!!!
At 8:50 the only thing I can think of is: Is that all the rules!? Let's see how this gets cracked!
Edit after the break in, the way that the three in the corner is breaking symmetry is quite unexpected 🤯
Crazy that he didn't notice the roping until the near end of the puzzle
Heh, I spent 4 minutes frustratingly begging you to see the 6 in the top left box :P. Meanwhile you were finding plenty of other things I would have missed :).
Very enjoyable puzzle, amazing how you can fill the grid with so little to start with. Thank you James Kopp for sharing your cleverness again with us mere mortals ;) Also, spoiler below...
...
...
...
...
...
It's interesting (though not surprising with that 3+ constraint) that we have roping going on both horizontally and vertically
excellent setting
I enjoy this kind of solve. Where your 30 minutes in, have 20 digits and a hundred pencil marks. Then you hit a tipping point and the whole thing solves in three minutes.
I wish the 600K puzzles were on the CTC app like the 500K puzzles were.
Wonderful puzzle.
13:05 for me. Fairly approachable one today.
The way you didn't see that the three couldn't go next to the one on the bottom is exactly how I see all digits at all times in all rulesets. Haha it's funny like that.
6 minutes 15 seconds between placing the 3 in box 1 and realizing r2c2 is a 6. Longest 6 minutes and 15 seconds of my life. Felt like the video should have almost been over.
59:18 ... I had a few false starts (I forgot that the 3-difference rule didn't cross box borders) but finally found my way to the end of this miracle-esque sudoku
Nice puzzle!
Once the ground work is in place, you just go through one digit at a time. "Where are all the threes? They must be here. Okay, where are all the fours?" Every digit could be finished or nearly finished before moving to the next one
My favourite part of Ludwig is every episode when they stop solving the murder to read out the birthdays.
My method is a little different from Simon's. My deduction is based on below conclusion:
The middle digit of the box is important. Its neighbor and neighbor ^ 2 digits should be in the corners of the box. So for middle digit 34567, its corner digits are fixed. For 28, it has 1 digit free and for 19 it has 2 digits free.
After I get the box 1 done, I know the middle digit of box 2 can only be 2389 to put 1 in the corner. 2 doesn't work for sudoku. 3 doesn't work because the 45 in box 1 push both 4 and 5 to r3c6. 8 doesn't work because 9 in box 1 push 9 to r3c6. Then the candidates for row3 of box3 is {2456}. We can not find a valid solution from this set. So the middle digit of box 2 is 9! The same method can be applied to box 6.
That was unexpected, this design w/o the given has only 2 solutions+2 mirrors w.r.t the D- (4 soln total). 7 boxes are completable leaving pairwise soln in the last 2 boxes diff by the given 3.
Not having the constraint across box orders fooled me at first tho
I really struggled with this, I had suspected that the rows or columns were going to have one of those hidden rules where each number had to be from 123, 456, and 789; but I couldn't find a good way to prove it so let it go... took me almost 2 hours to solve this, I almost wish I had just gone for, and played with that rule, would have solved it so quickly.
Fun, all boxes have the same tripples in the columns.
at 30:45, if it's a 5, the neiboring digit is than an 8 and the 6 is placed above the 3. You can apply the rules to set the 4 away from the six and the 2 finally has no place in the box proving it imposible.
I would love to watch you guys play The Last Express...murder mystery type game on the orient express, thought to be one of the best games in its genre. It uses stop motion graphics so you won't get the nauseous effect like some games. The special thing about it is that it uses a real time game mechanic. I played it multiple times years ago when it first came out and I just bought it again on Steam since it is on sale to play again!
43:31 for me - I got stuck at the start trying to figure out which line was the arrow; I was caught up in checking how the 3 would affect the arrow that I didn't check how the arrow affected box 7.
I "reasoned" out where it was, hit a contradiction a minute later, so knew right where to go on restart
U can try stripping the antisymmetric components(the given 3) Orient yourself with the line...and then DD where the given 3 goes and reorient if needed
I almost corner marked three cells in box 5 as well.
I don't understand why 9 pencil mark was taken out of the center square of box nine at 26:00
In the bottom row there is a 789 triple and the 9 in that triple is in R9C7 or R9C9. So it cannot be in R8C8
Another way to look is that if R8C8 is 9 then in the bottom row - the three cells R9C2 R9C7 R9C9 are 78
@@jasonbourne7047 ohhhhhhh, thank you I was going nuts trying to figure it out
@@jasonbourne7047 thank you i was wondering this as well
Belgian Whispers in boxes 🙂KFC should introduce it into their offer.
And finally on 23:34 Simon sees the centre square in box one has to be a six😊
Cracking!!
Holy missed pencil marks Batman… I’m not real good at these, but this one was a little tough to watch lol. Sorry Simon still watched and gave a thumbs up. Still waaay better than I!
This solve would have gone a lot faster if Simon didn't keep forgetting the variant rule for this puzzle.
Screen shout R2C2🎉😂❤
Now everybody's going to start using "Within a 3x3 box, orthogonally connected cells must have a difference of at least 3" as the thing which finishes the entire puzzle after the break-in
Finished in 22:36 with help from the video.
Roping all over grid - maybe as a result of the difference rule.
Do the central digits have to form another 1 to 9 set with this variant?
Its all about the mod 3.
21:33. I feel like either Simon or I did the entire middle part backwards, with him seeing the top-right and me starting from box 4
But, no confetti for that 3 in the corner.
I keep wondering whether Michael Stipe is at all aware that his song has turned into a Sudoku meme.
I missed the break-in because I connected cell rule applied across boxes and got hyper-fixated on that for 25min sigh. 47:55 in the end after finally getting the break-in.
54:30 for me. Nice puzzle!
56:10, had to undo after being confused, looking at the video, being even more confused, then rereading the rules to see that the "difference of 3" rule ONLY applies within boxes.