Thank you so much for this feature I really enjoyed from start to finish and it was hit every logic I did intended. Glad you enjoyed it as well, hopefully to see more in future, thanks again and take care.
@@kennetsdad Great ruleset. I had never noticed that there must be a *minimum* of *15* friendly cells in any 9x9 sudoku. And this is the first time evert that I used *placeholder lines‼* More exactly, eight *region sum* placeholder lines. Each of them was a diagonal spanning three cells in three different boxes, and represented either itself (negative slope) or its mirror image with respect to *column 5* (positive slope). It was very fun to use them and they made it easy enough for me to find the intended trick to rule out the option with negative slope. I love using *placeholder digits* to simplify my notation (see my playlist for more information) so it was natural for me to extend the concept to lines. *Placeholding* is quite an elegant notation technique, and in some cases it can be extremely powerful. In other cases it is still powerful enough to be useful to solvers like me, that are relatively numpty compared to Simon and Mark. It always simplifies notation. Sometimes massively, sometimes just enough to make your solve a little bit smoothier. Sometimes I need simplified notation to unveil the logic path that would be otherwise hidden by *cumbersome pencilmarking.* Some other times I just enjoy solving puzzles compatible with this technique as smoothly as possible, even when they are easy enough to be solved without it.
Magnificent construction. I had never noticed before that the minimum number of friendly cells in a 9x9 sudoku is 15❗ I enjoied solving by means of *placeholder lines,* which is the same technique used by Simon @30:10, although I only used eight lines and I colored them in blue, as I considered them to be *region sum* lines. Each of them was a 3-cell long diagonal with *negative slope* (while Simon used positive slope) and represented either itself or its mirror image with respect to *column 5.* The trick to rule out the negative slope, based on the black dot in *row 2,* was phenomenal❗ I love using *placeholder digits* whenever they are applyable, and it was natural for me to extend that concept to *lines.* Actually, I call them *placeholder lines* because they represent sets of *placeholder digits.* See my separate comment for more details.
as someone who falls asleep watching these videos sometimes, its not you guys being boring, its that you have nice and calming voices and i struggle to sleep when its silent so these videos create an environment that i feel relaxed enough to sleep in
For me, the videos are the perfect mix between mentally engaging enough to keep the Bad Thoughts away and calming enough to not prevent me from falling asleep. I hold this channel very dear. Please keep it up!
I usually fall asleep to your videos Simon, it is a lovely soothing voice, that helps with my tinnitus. Please do not feel bad about it, it is a lovely quality.
I like it when Simon always says when it comes to the secret of sudoku that he only shares this with his very favourite people, and "if you are watching this video, you are definitely one of my favourite people". Well, I can tell that we wouldn't watch the video if Simon wasn't one of our favourite people. ;) Nice solve, btw. This one got me stuck twice to be honest :(
I sleep to the videos because i need to be interested and engaged in something to turn off my brain, i watch awake too! But you've got a very nice calm voice that feels safe while keeping my interest in something so i can let myself relax enough to sleep. I have rough insomnia.
"That 1 is a 1 you would like to meet at a party. That is the antithesis of a Simon." Cracked me up!! Ahahaha! PS: I would like to say that Simon is in fact the ONLY kind of person I would like to meet at a party.
To sleep to a CtC video does not make it a bad video, it simply means the speaker behind it has a fantastic and soothing voicebox to help put us at ease :)
You can deduce the number of friendly digits with a slightly simpler argument: For each digit there are 3 ways it can be friendly (row, column, box). So for each digit we have to identify a friendly cell three different ways (it can be the same cell). We saw that 1, 5, 9 can be triple friendly, so you can have just one friendly cell with the digit which counts for the three ways it can be friendly; a minimum of 3 cells. 2,3,4,6,7,8 can be at most double friendly, so we must have at least two different friendly cells of each; a minimum of 12 distinct cells. 12+3=15, the absolute minimum number of friendly cells in a sudoku puzzle..
@inspiringsand123 I am so glad you still do these analysis comments. I see the list of Simarkisms has grown since I last checked. "If I trust my pencil marks" is a surprising addition, but it makes absolute sense to include it, because he really does say that pretty much at least once a video.
I was hoodwinked into thinking this was a Phistomefel puzzle. Glad I checked the comments and was able to disillude that misapprehension! Interesting puzzle, I have no idea how Simon solved it as I type this, but I saw r4c5 had to be the same as r3c4 or r3c6 and chose one, remembering that I hadn't actually disambiguated which. This led to being able to colour all the box borders, and as luck would have it the one I chose showed an impossibility. From there I had to redo the colouring reflecting and it solved nicely in just over an hour for me.
fantastic puzzle, and classic Dorlir. I think I would have guessed that was a Dorlir puzzle if it was a 'Guess The Setter' challenge. Stunning minimalism and clever combining of two rulesets
I really enjoyed this solve! I got it in less than an hour, which is amazing for me! I liked that there is quite a small amount of very hard logic in the puzzle, that once you get to, it basically solves itself. Thank you for bringing it to my attention!
With forty four minutes of meta analysis and only the remaining ten for harvesting almost all of the digits, you know you’re in for some mind blowing logic…
Funny how at 46:36 Simon decides to remove the 8 pencilmark (instead of restoring the cross-box pencilmark for 8s) but would have needed it for its very next deduction on the black dot ;)
Gotta say, this is the limit of the clickbaitiness (of the video title) for me. If I see the devil graphic and the name of the great man, I want to see the great man. This is unfair to Dorlir, who is an awesome setter too - now I'm disappointed right from the start even though this puzzle is probably great. Other setters don't need to live in the shadow of Phistomephel. Maybe I'm over-interpreting, but I'm going to watch it anyway so please don't go quite this far with the clickbait.
"Phistomefel-ian" on the title says clearly from the start that the puzzle is not from Phistomefel, but that it only applies his principle. So I'd point out poor reading and impulsive clicking as the only cause of disappointment, not the title.
@@williammorris7279 Yeah, after watching the whole video it did feel like clickbait a little, unnecessarily so. But aren't viewers to blame for what we choose to watch?
@dbarrustele We are! It just isn't what we expect from this channel. They succeed by ignoring the normal youtube ploys. Well, mostly: Simon is especially adept at elongating videos and getting comments. "How do I pronounce this name? Chocolate cake. Which colour do I choose? I somehow cannot scan. Do you like long videos? I am uninteresting at parties."
For placing the odds and evens in the magic square, I use this argument: 1) A line cannot be even + odd + odd, because that sum is even but must be 15 which is odd. 2) Wherever the first even digit is placed, another even digit is opposite it along the line crossing the five, otherwise that line is even-odd-odd. 3) The same is true of the remaining two even digits. They are opposite each other along the five. 4) These two pairs can't be aligned in a + shape because the top row would be odd-even-odd. 5) Similarly they can't be one-orthogonal, one-diagonal, because it creates odd-odd-even in some position. 6) After reflection/rotation the only formation for two pairs remaining is the X formation. 7) Therefore the evens are positioned in that X shape, i.e. in the corners.
UA-cam's new "feature" of making anything monetizable into an automatic search link has informed me that wrogn is now a fashion brand. Or possibly has been for a while. That's a shame, I hope they don't get litigious over the use of the word in sudoku puzzles.
Great puzzle. My break in differed significantly from Simon's. I used the constraints for 9s in row 8 to identify two possible length 3 diagonals of 9s. I then used their interaction with the magic square to determine which direction diagonal was correct.
Extraordinarily brilliant solve 👏😏Particularly the *placeholder notation* used @30:10, which is similar to what I used, except that I drew only the 3-cell long *diagonals,* and I colored them in blue, as I considered them to be *region sum* lines. I did not need to draw the 2-cell lines. Each of them was a 3-cell long diagonal with *negative slope* (while Simon used positive slope) and represented either itself or its mirror image with respect to *column 5.* They can be called *placeholder lines* because they represent sets of three identical *placeholder digits.* *Placeholding* is quite an elegant notation technique, and in some cases it can be extremely powerful. In other cases it is still powerful enough to be useful to solvers like me, that are not as brilliant as Simon and Mark. It always simplifies notation. Sometimes massively, sometimes just enough to make your solve a little bit smoothier. Sometimes I need simplified notation to unveil the logic path that would be otherwise hidden by *cumbersome pencilmarking.* Some other times I just enjoy solving puzzles compatible with this technique as smoothly as possible, even when they are easy enough to be solved without it.
I finished in 164 minutes. This was such a tough puzzle for me to see, especially regarding the friendly cells. I found the zipper pattern pretty quickly and it created one of the coolest patterns I have seen. It was so pretty and kind of looked like paracord. It took me a really long time to appreciate the friendly cells limitation. I was only focusing on one aspect of the columns without realizing that it would force the rows to be displaced as well. Once I understood that, I saw that maximizing rows and columns used for each friendly cell led to 1-9 being placed on the long negative diagonal. It looked like it worked and gave me 15 friendly digits. However, it broke and I couldn't see why. It took me another long time to see that the digits don't just have to share rows and columns, but that they could share row/column and box. Doing that finally forced a 1 and 9. The puzzle finally finished itself. That was rough for a while. I am glad I was able to find it as it is an amazing minimalist piece of art. Great Puzzle!
@@CarlintVeld Because friendly cells are limited, we can imagine the most efficient way of populating them by having them share as many rows/columns/boxes as they can. We can put 1 in r1c1, 2 in r2cc2, 3 in r3c3, etc. We can see that this gives us 9 friendly cells already. If we look at the boxes, we can see that 1, 5, and 9 already have their friendly box digits. This leaves 6 boxes that have to contain their friendly box number. This now gives 9 plus 6, which is perfect as it is 15 and the number we were looking for. However, this is not the only efficient way to pack this structure. Let's imagine 2 in r2c2 was shifted to r2c3. The 2 is still friendly with its row, but column 2 no longer sees that 2. We know a 2 must exist in the column 2, so that gives a plus 1 to our friendly count and breaks our minimum of 15. However, if we shift the 2 in r2c2 to r2c4, the extra 2 we created in column 2 gets cancelled out by the 2 in r2c4 sharing its friendliness with the box, so our friendly count remains at 15. You can think of these numbers on the diagonal as having to share at least two of rows/columns/boxes, so the numbers can shift along the row to match their box instead of staying in the initial spot on the negative diagonal. The exception to that is 1, 5, and 9. If we tried to move 1 like we did with 2, we can see that no matter how we move it, we generate an additional friendly 1 in the row or column. Therefore, 1, 5, and 9 are forced as they have no play. Hope this helps!
This 4-star-out-of-5-for-difficulty looked like a 3-star, at most, when I watched you solve this, Simon. You explain things so well, and you lead us in your thought process, it makes the puzzle look easy! Thanks for how wonderfully you have led me into being a better solver (and I hear yours and Mark's voices in my head when I solve, so often), through these videos.
I feel there are many more positions for the friendy cells and that Simon got lucky 😅 Like the 1 can go in any position on the row, column or box providing 3 instances of the 1
I noticed that, too, but in fact as he'd already pencil-marked the possible locations for friendly digits, it turned out that it wasn't necessary to verify them in the end.
I got lucky assuming that the double friendlies were in their own box. I missed the possibility that the double friendlies could be on the diagonal. Luckily, I guessed right.
If I understand what you are asking for, it can be done already. If you explore the color pad you will see that there is a way to customize every color and have three palettes, in fact. But perhaps I don't understand your suggestion.
@@malvoliosf Ah. No, I don't think it can be one operation. But you can double click all of the 5s and make them all blue. But that is after they are already in the grid.
I couldn't figure this one out. After penciling in a lot of digits, and of course getting the 5 in the middle, I was getting a feel for where it wanted me to place the digits and guessed most of 1, 9, and 3s, which I guessed correctly, first time. Then the rest finally cascaded, and I completed it. It was unsatisfying that I couldn't logic those first few digits. Hopefully Simon can show me what I missed.
That's no three in the corner, no three in the spot... light... Simon's losing patience. (The setters tried to keep up with him but they couldn't do it.)
Simon got incredibly lucky with this one, didn't he? He didn't even consider the possibility the the "doubly" friendly cell can share a column + box, plus another one elsewhere.
I still don't get it why can't the 1's be put three times? 😬 edit: I think I get it because of the exactly N friendly digits constraint which is also the minimum! Still Simon got lucky as he did not include the columns as potential placements. Edit: that would yield a third placement for a digit and would break the N friendly digits constraint. I get it I guess 😅
The only thing you did not catch was that the diagonals propagate pacman style across the outside border... That could have gotten you the one in row 6 and 7 a bit earlier
I think the rules for this puzzle might need cultural revision. "normal reading order" doesn't mean the same thing everywhere in the world. Rules should specify left to right,top to bottom. It would be fascinating if this puzzle were to work for multiple reading orders, and positively mind blowing if it worked for all reading orders.
@@RichSmith77 I don't dispute that, but with the plethora of automatic AI translators and the potential for universal appeal for sudoku, a puzzle might not always get to someone with the rules still in English. While the puzzle itself does not originate in Japan, the name Sudoku does, and Japanese is traditionally a vertical language, top to bottom, right to left. And since the fun of these puzzles is the technicality of the rules I feel ambiguity should be avoided. And yes, this is a bit pedantic. 🙂
Hi Simon (please excuse my bad english), this is a very nice riddle, but i have one question: the second rule is telling, that "a border square is a 2x2 square that... and so on...." When i read the rules, i can´t see, HOW MANY border sqaures in this sudoku are placed. You draw 12 short yellow lines with length "two" and 8 longer lines with length "three"... this is the maximum of possible lines. OK. But where in the rules is standing, that tis maximun has to be placed in the riddle ? When i tried to solve it without seeing the video i would know, that there ARE border squares with a doubled digit, but how much or where border squares have to been placed is not known... Am i wrong, or is the description of the rules "incolplete" ? Greetings, Thomas
I think the thumbnail for this should be changed. Saying "Phistomefel-ian" in the title is fine to have as a description, but having the red demon in the thumbnail implies that this a sudoku by Phistomefel. I know that that was my thought upon seeing it. I clicked this video thinking it was a miracle sudoku set by Phistomefel.
It may be misleading, but if you pay attention to the title, you can at least suspect it is not by Phistomefel. Then you can read the description or watch the introduction and understand that the *red demon* is there because one of his puzzles inspired the constructior. Some other times the *red demon* appeared in the thumbnail just because he recommended that puzzle to Simon.
78:00 with a bit of luck. I "concluded" that the cells that were counted had to be in the row/box pair (eg 3 in row 3 AND box 3). I broke it at the very end and the video showed me that I neglected thinking that the counted cells could be in a row/column pair (eg 8 in row 8 AND column 8).
Got confused about the ruleset wording for the border squares its implied there are at least two. But if im reading it right then technically there dont habe to be any at all
Any 2x2 square that crosses a box border is a border square. There are four border lines, each with 8 border squares, but the border squares at the four intersections will be double counted, so (4 x 8) - 4 = 28 border squares.
@@RichSmith77 the word 'any' is key here. I was looking for spme kind of identification on which ones were border squares. Of course the video clarifies. But i usually tey to solve it at least a bit before watching
@@christopherbowers7236 Isn't the restriction that it crosses a border a sufficient enough condition to define "any" 2x2 square that meets this specified criteria as being a border square? There are 8x8=64 possible 2x2 squares in the grid. Any of these that cross a border are being called "border squares". That seems like a perfectly acceptable use of the word "any" to me.
@@christopherbowers7236 Sorry. When I saw 'any' in your previous post, I thought you were quoting it from the rules and saying its presence in the rules was the issue. My mistake. I should have re-read the rules more closely. I'm not sure if using "a" instead of "any" makes [any/a] difference though. The rules are still stating exactly what 2x2 border square is. It's one of the 64 2x2 squares in the grid that crosses at least one box border. That's the complete definition of what a 2x2 border square is.
I deducted the logic of a magic square in your app ealier today. The 5 in the middle I agree with. But how I did another way. If every line needs to be 15 then you need either 3 odd numbers, or 2 even and 1 odd number otherwise the sum wouldnt be odd. Hence the even numbers need to be in the conrers.
39:18 It's funny to me which rows and columns Simon looked at to see that that various members of that triple repeated digit sees 7 colors, because actually just the single cell r3c3 sees all 7 of them on its own...
47:30 Always strange when Simon seams to miss obvious things. Here he only highlights r3c1,2,9 and r4c1,8 but doesn't highlight r4c9 Doesn't he realize that r3c1 is the same as r4c9?
What he proved was, that the MINIMUM amount of friendly cells is 15. And since as you Said we know that there has to be exactly 15 friendly cells, we can force 1 and 9 in the corners.
The point is that Simon loves to show that he is smart. We all knew N was 15 as soon as we read the rules. He likes to do Sudoku ignoring rules and proving things and then saying, oh I didn't need to do that, as a soft brag. And yes he is smarter at Sudoku than I am. And many times his approach to ignoring rules and proving things teaches me new ways to approach these puzzles. But it is indeed a low-key flex. ALSO - he does multiple puzzles per day - so remembering each puzzles' rules can be cumbersome.
@@RedBarchetta2019 This is a bit of an unkind approach to what Simon does. Simon's goal is that no one, even if they have never seen a particular rule set or logical step before, will leave the puzzle and throw up their hands saying that they can never understand variant sudoku. Simon explains so that the newcomer to the channel will understand and become a fan of variant sudoku.
I could see the coloring pattern so quickly, but for the life of me, I could not see how to prove or disambiguate anything without just plugging in a number and rolling with it aka "guess and check." So I gave up after 2 hours to just enjoy the video. 😅
On the webpage version, all the rules are in one big paragraph. Would be very much improved if the rules got separated with a few empty lines, extra 's.
hmmm I haven't started yet this isn't clear to me, but I'm making an assumption that the "N" in the friendly cells refers to the same "N" in the magic square?
with all due respect, the phistomefel puzzles aren't very fun but it's just complicated figuring out the rules. Puzzles with new or complicated rule sets are complicated and cumbersome because of this, not because the logic is necessarily difficult
I am pretty sure that you are in the minority thinking the Phistomefel's puzzles are not very fun. Most people would describe them more as "the very best sort of puzzle using a particular rule set" and full of beautiful (not necessarily difficult) logic. There are puzzles by Phistomefel and also by Dorlir in some of the apps - they are not necessarily difficult at all, but they do show the unique strengths that each has as a constructor.
Don't say it's Phistomephel when it's not. You're better than that. You don't need to lie to people to get them to watch. You're like Field of Dreams. If you solve it, they will come.
Thank you so much for this feature I really enjoyed from start to finish and it was hit every logic I did intended. Glad you enjoyed it as well, hopefully to see more in future, thanks again and take care.
A very cool solve path. It's like you hid the maximum palindrome lines behind a bizarre rule set.
@@kennetsdad Great ruleset. I had never noticed that there must be a *minimum* of *15* friendly cells in any 9x9 sudoku.
And this is the first time evert that I used *placeholder lines‼*
More exactly, eight *region sum* placeholder lines. Each of them was a diagonal spanning three cells in three different boxes, and represented either itself (negative slope) or its mirror image with respect to *column 5* (positive slope).
It was very fun to use them and they made it easy enough for me to find the intended trick to rule out the option with negative slope.
I love using *placeholder digits* to simplify my notation (see my playlist for more information) so it was natural for me to extend the concept to lines.
*Placeholding* is quite an elegant notation technique, and in some cases it can be extremely powerful. In other cases it is still powerful enough to be useful to solvers like me, that are relatively numpty compared to Simon and Mark. It always simplifies notation. Sometimes massively, sometimes just enough to make your solve a little bit smoothier.
Sometimes I need simplified notation to unveil the logic path that would be otherwise hidden by *cumbersome pencilmarking.* Some other times I just enjoy solving puzzles compatible with this technique as smoothly as possible, even when they are easy enough to be solved without it.
Magnificent construction. I had never noticed before that the minimum number of friendly cells in a 9x9 sudoku is 15❗
I enjoied solving by means of *placeholder lines,* which is the same technique used by Simon @30:10, although I only used eight lines and I colored them in blue, as I considered them to be *region sum* lines.
Each of them was a 3-cell long diagonal with *negative slope* (while Simon used positive slope) and represented either itself or its mirror image with respect to *column 5.*
The trick to rule out the negative slope, based on the black dot in *row 2,* was phenomenal❗
I love using *placeholder digits* whenever they are applyable, and it was natural for me to extend that concept to *lines.*
Actually, I call them *placeholder lines* because they represent sets of *placeholder digits.*
See my separate comment for more details.
I am always so impressed with your setting Dorlir! Another gem from you!!
What a marvelous puzzle, Dorlir. Bravo!
as someone who falls asleep watching these videos sometimes, its not you guys being boring, its that you have nice and calming voices and i struggle to sleep when its silent so these videos create an environment that i feel relaxed enough to sleep in
A safe space is always nice to have.
For me, the videos are the perfect mix between mentally engaging enough to keep the Bad Thoughts away and calming enough to not prevent me from falling asleep. I hold this channel very dear. Please keep it up!
They do. They combine the best parts of Bob Ross and Adam and Jaime friom Mythbusters,
@@sirenier same for me lol
I do this too! For the same reason. I sometimes feel bad for using them to sleep, so I will revisit a video to see it through.
I usually fall asleep to your videos Simon, it is a lovely soothing voice, that helps with my tinnitus. Please do not feel bad about it, it is a lovely quality.
I like it when Simon always says when it comes to the secret of sudoku that he only shares this with his very favourite people, and "if you are watching this video, you are definitely one of my favourite people".
Well, I can tell that we wouldn't watch the video if Simon wasn't one of our favourite people. ;)
Nice solve, btw. This one got me stuck twice to be honest :(
I absolutely love when a puzzle results in a 100% colored grid. 🎉
I sleep to the videos because i need to be interested and engaged in something to turn off my brain, i watch awake too! But you've got a very nice calm voice that feels safe while keeping my interest in something so i can let myself relax enough to sleep. I have rough insomnia.
"That 1 is a 1 you would like to meet at a party. That is the antithesis of a Simon." Cracked me up!! Ahahaha!
PS: I would like to say that Simon is in fact the ONLY kind of person I would like to meet at a party.
To sleep to a CtC video does not make it a bad video, it simply means the speaker behind it has a fantastic and soothing voicebox to help put us at ease :)
I normally shout at the screen expressing where some numbers go. Not this one... just sat back dumfounded and in awe on how you solved it!!!
Yeah, Simon's a bit of a genius.
It really feels like a perfect solve.
You can deduce the number of friendly digits with a slightly simpler argument: For each digit there are 3 ways it can be friendly (row, column, box). So for each digit we have to identify a friendly cell three different ways (it can be the same cell). We saw that 1, 5, 9 can be triple friendly, so you can have just one friendly cell with the digit which counts for the three ways it can be friendly; a minimum of 3 cells. 2,3,4,6,7,8 can be at most double friendly, so we must have at least two different friendly cells of each; a minimum of 12 distinct cells. 12+3=15, the absolute minimum number of friendly cells in a sudoku puzzle..
One of the most unbelievably beautiful sudokus I've ever played
These videos are the best parts of my day. I love you guys!
Same here.
Slightly renaming your colours, Simon:
'All the threes are brown, and the eights are grey...' 🎵
Man now I might make 3s brown and 5s gray as a convention just to sing that song.
@@Sully800 'Fives' does rhyme better, I agree. 👍
Rules: 06:58
Let's Get Cracking: 10:34
Simon's time: 41m56s
Puzzle Solved: 52:30
What about this video's Top Tier Simarkisms?!
The Secret: 6x (02:08, 08:32, 08:32, 10:46, 10:56, 12:41)
Bobbins: 2x (47:01, 47:46)
Three In the Corner: 1x (49:38)
And how about this video's Simarkisms?!
Ah: 12x (07:24, 16:49, 23:56, 27:10, 34:36, 35:17, 37:40, 39:41, 39:54, 41:01, 46:55, 51:19)
Hang On: 9x (16:29, 16:47, 27:35, 41:01, 43:58, 43:58, 43:58, 45:45, 46:09)
Pencil Mark/mark: 8x (27:05, 27:51, 33:33, 37:57, 43:06, 46:24, 46:42, 47:31)
Obviously: 7x (02:18, 04:00, 12:56, 15:18, 28:10, 32:11, 34:52)
Brilliant: 4x (03:09, 04:41, 29:14, 35:20)
By Sudoku: 4x (43:11, 46:16, 48:51, 49:26)
Cake!: 4x (04:39, 05:34, 05:34, 06:42)
Sorry: 3x (24:13, 30:03, 39:54)
I've Got It!: 3x (23:56, 23:56, 35:08)
Wow: 3x (52:24, 52:39, 52:39)
Goodness: 2x (05:30, 52:28)
Bonkers: 2x (01:05, 53:15)
Bizarre: 2x (29:21, 29:29)
Proof: 2x (13:52, 16:24)
Fabulous: 2x (53:24, 53:24)
Nature: 2x (44:12, 48:34)
Symmetry: 2x (41:21, 41:46)
Weird: 2x (16:06, 28:07)
Nonsense: 1x (46:49)
Clever: 1x (35:17)
Beautiful: 1x (35:17)
Fascinating: 1x (29:01)
Elegant: 1x (13:36)
Hypothecate: 1x (15:59)
Take a Bow: 1x (53:29)
Shouting: 1x (05:48)
Of All Things: 1x (49:26)
If I Trust my Pencil Marks: 1x (46:42)
Shenanigans: 1x (00:20)
Surely: 1x (23:09)
In Fact: 1x (48:27)
I Digress: 1x (00:52)
Intriguing: 1x (31:48)
That is Sick: 1x (27:21)
Triangular Number: 1x (11:08)
Most popular number(>9), digit and colour this video:
Fifteen (19 mentions)
One (90 mentions)
Black (25 mentions)
Antithesis Battles:
Even (7) - Odd (1)
Black (25) - White (0)
Row (42) - Column (35)
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!
End of magic square proof: 16:30 :)
@inspiringsand123 I am so glad you still do these analysis comments. I see the list of Simarkisms has grown since I last checked. "If I trust my pencil marks" is a surprising addition, but it makes absolute sense to include it, because he really does say that pretty much at least once a video.
Maybe we should give Dorlir his own graphic?🧐
Dorlir surely a combination of math professor and wizardry!
I was hoodwinked into thinking this was a Phistomefel puzzle. Glad I checked the comments and was able to disillude that misapprehension! Interesting puzzle, I have no idea how Simon solved it as I type this, but I saw r4c5 had to be the same as r3c4 or r3c6 and chose one, remembering that I hadn't actually disambiguated which. This led to being able to colour all the box borders, and as luck would have it the one I chose showed an impossibility. From there I had to redo the colouring reflecting and it solved nicely in just over an hour for me.
you made it look easy because you showed the magic square parities so fast, superb solve
fantastic puzzle, and classic Dorlir. I think I would have guessed that was a Dorlir puzzle if it was a 'Guess The Setter' challenge. Stunning minimalism and clever combining of two rulesets
I really enjoyed this solve! I got it in less than an hour, which is amazing for me! I liked that there is quite a small amount of very hard logic in the puzzle, that once you get to, it basically solves itself. Thank you for bringing it to my attention!
With forty four minutes of meta analysis and only the remaining ten for harvesting almost all of the digits, you know you’re in for some mind blowing logic…
This just looks wonderful to dig into! Can't wait til I'm home to try it out.
This puzzle is a mathematician's dream...all about patterns and permutations
46:40 - My new favorite phrase: "the jiggery-pokery with the nonsenses"
Funny how at 46:36 Simon decides to remove the 8 pencilmark (instead of restoring the cross-box pencilmark for 8s) but would have needed it for its very next deduction on the black dot ;)
It occurs that with a slightly different placement of the black dots, you could have the relevant diagonal pairings go exactly the opposite way.
Gotta say, this is the limit of the clickbaitiness (of the video title) for me. If I see the devil graphic and the name of the great man, I want to see the great man. This is unfair to Dorlir, who is an awesome setter too - now I'm disappointed right from the start even though this puzzle is probably great. Other setters don't need to live in the shadow of Phistomephel. Maybe I'm over-interpreting, but I'm going to watch it anyway so please don't go quite this far with the clickbait.
"Phistomefel-ian" on the title says clearly from the start that the puzzle is not from Phistomefel, but that it only applies his principle. So I'd point out poor reading and impulsive clicking as the only cause of disappointment, not the title.
Agreed!!!
@@dbarrusteleWhat you state is exactly the definition of "clickbait"! And there is the misleading graphic as well as the title.
@@williammorris7279 Yeah, after watching the whole video it did feel like clickbait a little, unnecessarily so. But aren't viewers to blame for what we choose to watch?
@dbarrustele We are! It just isn't what we expect from this channel. They succeed by ignoring the normal youtube ploys. Well, mostly: Simon is especially adept at elongating videos and getting comments. "How do I pronounce this name? Chocolate cake. Which colour do I choose? I somehow cannot scan. Do you like long videos? I am uninteresting at parties."
For placing the odds and evens in the magic square, I use this argument:
1) A line cannot be even + odd + odd, because that sum is even but must be 15 which is odd.
2) Wherever the first even digit is placed, another even digit is opposite it along the line crossing the five, otherwise that line is even-odd-odd.
3) The same is true of the remaining two even digits. They are opposite each other along the five.
4) These two pairs can't be aligned in a + shape because the top row would be odd-even-odd.
5) Similarly they can't be one-orthogonal, one-diagonal, because it creates odd-odd-even in some position.
6) After reflection/rotation the only formation for two pairs remaining is the X formation.
7) Therefore the evens are positioned in that X shape, i.e. in the corners.
Don't we get the N immediately, since it's always the same for magic squares?
@FrequencySpectre-y1b we do)
Yes -- but isn't it more fun to make us spend the necessary few seconds to realize that?
that's also how I interpreted the rules
You knew that, and Simon knew that, but as well as a solver he is also a teacher, so he wanted to explain why.
I miss an overload Wrogn puzzle, with so many clues that reading them takes effort.
UA-cam's new "feature" of making anything monetizable into an automatic search link has informed me that wrogn is now a fashion brand. Or possibly has been for a while.
That's a shame, I hope they don't get litigious over the use of the word in sudoku puzzles.
Absolutely wonderful puzzle.
Just amazing, it's amazing how that pattern exists
Yes a Dorlir puzzle, amazing mind setting and an amazing mind solving.
Now then. Haven't got past the intro yet but the 'state of enmity / n-mity' joke is worth the admission fee alone!
Great puzzle. My break in differed significantly from Simon's. I used the constraints for 9s in row 8 to identify two possible length 3 diagonals of 9s. I then used their interaction with the magic square to determine which direction diagonal was correct.
Interestingly I got the break in completely differently with where does nine go in box 8 with it only being able to go into yellow
Extraordinarily brilliant solve 👏😏Particularly the *placeholder notation* used @30:10, which is similar to what I used, except that I drew only the 3-cell long *diagonals,* and I colored them in blue, as I considered them to be *region sum* lines. I did not need to draw the 2-cell lines.
Each of them was a 3-cell long diagonal with *negative slope* (while Simon used positive slope) and represented either itself or its mirror image with respect to *column 5.*
They can be called *placeholder lines* because they represent sets of three identical *placeholder digits.*
*Placeholding* is quite an elegant notation technique, and in some cases it can be extremely powerful. In other cases it is still powerful enough to be useful to solvers like me, that are not as brilliant as Simon and Mark. It always simplifies notation. Sometimes massively, sometimes just enough to make your solve a little bit smoothier.
Sometimes I need simplified notation to unveil the logic path that would be otherwise hidden by *cumbersome pencilmarking.* Some other times I just enjoy solving puzzles compatible with this technique as smoothly as possible, even when they are easy enough to be solved without it.
See also *Solving sudoku with placeholder digits.*
I noticed that the matched diagonals also go over the edges, and I was hoping that would come up, made it a little easier for me (:
I finished in 164 minutes. This was such a tough puzzle for me to see, especially regarding the friendly cells. I found the zipper pattern pretty quickly and it created one of the coolest patterns I have seen. It was so pretty and kind of looked like paracord. It took me a really long time to appreciate the friendly cells limitation. I was only focusing on one aspect of the columns without realizing that it would force the rows to be displaced as well. Once I understood that, I saw that maximizing rows and columns used for each friendly cell led to 1-9 being placed on the long negative diagonal. It looked like it worked and gave me 15 friendly digits. However, it broke and I couldn't see why. It took me another long time to see that the digits don't just have to share rows and columns, but that they could share row/column and box. Doing that finally forced a 1 and 9. The puzzle finally finished itself. That was rough for a while. I am glad I was able to find it as it is an amazing minimalist piece of art. Great Puzzle!
Why are the 1, 5 and 9 forced? I can't seem to grasp that aspect
@@CarlintVeld Because friendly cells are limited, we can imagine the most efficient way of populating them by having them share as many rows/columns/boxes as they can. We can put 1 in r1c1, 2 in r2cc2, 3 in r3c3, etc. We can see that this gives us 9 friendly cells already. If we look at the boxes, we can see that 1, 5, and 9 already have their friendly box digits. This leaves 6 boxes that have to contain their friendly box number. This now gives 9 plus 6, which is perfect as it is 15 and the number we were looking for. However, this is not the only efficient way to pack this structure. Let's imagine 2 in r2c2 was shifted to r2c3. The 2 is still friendly with its row, but column 2 no longer sees that 2. We know a 2 must exist in the column 2, so that gives a plus 1 to our friendly count and breaks our minimum of 15. However, if we shift the 2 in r2c2 to r2c4, the extra 2 we created in column 2 gets cancelled out by the 2 in r2c4 sharing its friendliness with the box, so our friendly count remains at 15. You can think of these numbers on the diagonal as having to share at least two of rows/columns/boxes, so the numbers can shift along the row to match their box instead of staying in the initial spot on the negative diagonal. The exception to that is 1, 5, and 9. If we tried to move 1 like we did with 2, we can see that no matter how we move it, we generate an additional friendly 1 in the row or column. Therefore, 1, 5, and 9 are forced as they have no play. Hope this helps!
@42:20: at this point I solved the magic square by looking at where 9 goes into box 8. :-) Great puzzle!
I will never ever get tired of hearing simon call me one of his favorite people ❤
This 4-star-out-of-5-for-difficulty looked like a 3-star, at most, when I watched you solve this, Simon. You explain things so well, and you lead us in your thought process, it makes the puzzle look easy! Thanks for how wonderfully you have led me into being a better solver (and I hear yours and Mark's voices in my head when I solve, so often), through these videos.
I still do not understand why friendly numbers couln'd be placed in box"column logic.
I feel there are many more positions for the friendy cells and that Simon got lucky 😅 Like the 1 can go in any position on the row, column or box providing 3 instances of the 1
Loved this puzzle. With all the colouring Simon did, I would just say he forgot to check the "Friendly" cells.
I noticed that, too, but in fact as he'd already pencil-marked the possible locations for friendly digits, it turned out that it wasn't necessary to verify them in the end.
I would have expected a deduction about how the unmatched digits at the ends of the borders would match each other.
Solved in 54:42!
I got lucky assuming that the double friendlies were in their own box. I missed the possibility that the double friendlies could be on the diagonal. Luckily, I guessed right.
Simon, you should ask Sven to change to problem so you can assign a color to any “big” number. It’d save a lot of time.
If I understand what you are asking for, it can be done already. If you explore the color pad you will see that there is a way to customize every color and have three palettes, in fact. But perhaps I don't understand your suggestion.
@@emilywilliams3237 I mean, can you set it up so that if you set a cell to 5, the cell’s color is automatically changed to blue?
@@malvoliosf Ah. No, I don't think it can be one operation. But you can double click all of the 5s and make them all blue. But that is after they are already in the grid.
@@emilywilliams3237 Exactly. I’d like to make that automatic.
Hi mark!!! Watching after a while today 😅😊
Finished in 25:36 by following along with the video.
I couldn't figure this one out. After penciling in a lot of digits, and of course getting the 5 in the middle, I was getting a feel for where it wanted me to place the digits and guessed most of 1, 9, and 3s, which I guessed correctly, first time. Then the rest finally cascaded, and I completed it. It was unsatisfying that I couldn't logic those first few digits. Hopefully Simon can show me what I missed.
Why assume that all the identical digits in the squares have to be on the positive diagonal?
Well Simon proved that the negative diagonals weren't possible so concluded it could only be positive diagonals
Testers. Please find a puzzle for Simon with 3 in the corner. It would make him so happy.
I didn't realize the exactly N friendly cells is an important constraint until 1.5 hrs past xD
Finally solved it in sub-2 hours.
just in time for bed
Opposite for me.
Wake up and see a new sudoku to solve haha
That's no three in the corner, no three in the spot... light... Simon's losing patience. (The setters tried to keep up with him but they couldn't do it.)
Simon got incredibly lucky with this one, didn't he? He didn't even consider the possibility the the "doubly" friendly cell can share a column + box, plus another one elsewhere.
There wasn't any flaw in the logic on this puzzle anywhere by Simon, actually he explained in a very clear way.
No such "doubly" friendly cell exist when 1,5 and 9 are gone.
I still don't get it why can't the 1's be put three times? 😬 edit: I think I get it because of the exactly N friendly digits constraint which is also the minimum! Still Simon got lucky as he did not include the columns as potential placements. Edit: that would yield a third placement for a digit and would break the N friendly digits constraint. I get it I guess 😅
@istvanmagi473 I think you are right. Need to check my solve again.
Ah no: The only values for which column and box intersect are 1, 5 and 9, and they are needed for the triply friendly cells.
The only thing you did not catch was that the diagonals propagate pacman style across the outside border... That could have gotten you the one in row 6 and 7 a bit earlier
I think the rules for this puzzle might need cultural revision. "normal reading order" doesn't mean the same thing everywhere in the world. Rules should specify left to right,top to bottom. It would be fascinating if this puzzle were to work for multiple reading orders, and positively mind blowing if it worked for all reading orders.
Yes, but "normal reading order" being written in English is kind of a clue to what it most likely means.
@@RichSmith77 I don't dispute that, but with the plethora of automatic AI translators and the potential for universal appeal for sudoku, a puzzle might not always get to someone with the rules still in English.
While the puzzle itself does not originate in Japan, the name Sudoku does, and Japanese is traditionally a vertical language, top to bottom, right to left.
And since the fun of these puzzles is the technicality of the rules I feel ambiguity should be avoided.
And yes, this is a bit pedantic. 🙂
@@fristlsat4663 That's fair. 👍
@@fristlsat4663 Good point. I would say interesting, rather than pedantic.
Just my imagination, or has this one been featured before? I seem to remember a lot of the logic?
Antithesis of a Simon?? You are so friendly, Simon, would love to meet you at a party!
Hi Simon (please excuse my bad english),
this is a very nice riddle, but i have one question: the second rule is telling, that "a border square is a 2x2 square that... and so on...."
When i read the rules, i can´t see, HOW MANY border sqaures in this sudoku are placed.
You draw 12 short yellow lines with length "two" and 8 longer lines with length "three"... this is the maximum of possible lines. OK.
But where in the rules is standing, that tis maximun has to be placed in the riddle ? When i tried to solve it without seeing the video i would know, that there ARE border squares with a doubled digit, but how much or where border squares have to been placed is not known...
Am i wrong, or is the description of the rules "incolplete" ?
Greetings, Thomas
I think the thumbnail for this should be changed. Saying "Phistomefel-ian" in the title is fine to have as a description, but having the red demon in the thumbnail implies that this a sudoku by Phistomefel. I know that that was my thought upon seeing it. I clicked this video thinking it was a miracle sudoku set by Phistomefel.
It may be misleading, but if you pay attention to the title, you can at least suspect it is not by Phistomefel.
Then you can read the description or watch the introduction and understand that the *red demon* is there because one of his puzzles inspired the constructior.
Some other times the *red demon* appeared in the thumbnail just because he recommended that puzzle to Simon.
78:00 with a bit of luck. I "concluded" that the cells that were counted had to be in the row/box pair (eg 3 in row 3 AND box 3). I broke it at the very end and the video showed me that I neglected thinking that the counted cells could be in a row/column pair (eg 8 in row 8 AND column 8).
Got confused about the ruleset wording for the border squares its implied there are at least two. But if im reading it right then technically there dont habe to be any at all
Any 2x2 square that crosses a box border is a border square.
There are four border lines, each with 8 border squares, but the border squares at the four intersections will be double counted, so (4 x 8) - 4 = 28 border squares.
@@RichSmith77 the word 'any' is key here. I was looking for spme kind of identification on which ones were border squares. Of course the video clarifies. But i usually tey to solve it at least a bit before watching
@@christopherbowers7236 Isn't the restriction that it crosses a border a sufficient enough condition to define "any" 2x2 square that meets this specified criteria as being a border square?
There are 8x8=64 possible 2x2 squares in the grid. Any of these that cross a border are being called "border squares". That seems like a perfectly acceptable use of the word "any" to me.
@@RichSmith77 the clue did not use the word 'any' it just says 'a'
@@christopherbowers7236 Sorry. When I saw 'any' in your previous post, I thought you were quoting it from the rules and saying its presence in the rules was the issue. My mistake. I should have re-read the rules more closely.
I'm not sure if using "a" instead of "any" makes [any/a] difference though. The rules are still stating exactly what 2x2 border square is. It's one of the 64 2x2 squares in the grid that crosses at least one box border. That's the complete definition of what a 2x2 border square is.
I deducted the logic of a magic square in your app ealier today. The 5 in the middle I agree with. But how I did another way.
If every line needs to be 15 then you need either 3 odd numbers, or 2 even and 1 odd number otherwise the sum wouldnt be odd. Hence the even numbers need to be in the conrers.
59:03 here. I never cease to be amazed with Dorlir's puzzles. Sheer genius and an absolute pleasure to solve.
Imagine having a 5 cell 379 triple
42:20 for me. Probably not that hard, but I made a bunch of mistakes.
Super early for a super empty grid!
39:18 It's funny to me which rows and columns Simon looked at to see that that various members of that triple repeated digit sees 7 colors, because actually just the single cell r3c3 sees all 7 of them on its own...
47:30 Always strange when Simon seams to miss obvious things. Here he only highlights r3c1,2,9 and r4c1,8 but doesn't highlight r4c9 Doesn't he realize that r3c1 is the same as r4c9?
46:58 It actually can't be 48, because that would rule 8 out of all of its potential "doubly friendly" cells identified earlier
I was thinking, didn't he have 8s pencil marked in that row at one stage? Trust Simon to have removed them just before they would have been useful.
Dont magic squares always have to add to 15? So once it said magic square, you know N is 15.
I think they pencil marking supported this but not formalised verbally
I got confused why it was identified really quickly that N was 15, then there was a long amount of time wasted proving that?
What he proved was, that the MINIMUM amount of friendly cells is 15. And since as you Said we know that there has to be exactly 15 friendly cells, we can force 1 and 9 in the corners.
The point is that Simon loves to show that he is smart. We all knew N was 15 as soon as we read the rules. He likes to do Sudoku ignoring rules and proving things and then saying, oh I didn't need to do that, as a soft brag. And yes he is smarter at Sudoku than I am. And many times his approach to ignoring rules and proving things teaches me new ways to approach these puzzles. But it is indeed a low-key flex. ALSO - he does multiple puzzles per day - so remembering each puzzles' rules can be cumbersome.
@@RedBarchetta2019 This is a bit of an unkind approach to what Simon does. Simon's goal is that no one, even if they have never seen a particular rule set or logical step before, will leave the puzzle and throw up their hands saying that they can never understand variant sudoku. Simon explains so that the newcomer to the channel will understand and become a fan of variant sudoku.
Rules seem ambiguous. Are we meant to assume N is the same for both rules? If not, then the N number of friendly cells is just trivial, no?
Same letter repeated in one set of rules is usually representing the same number, yes.
I could see the coloring pattern so quickly, but for the life of me, I could not see how to prove or disambiguate anything without just plugging in a number and rolling with it aka "guess and check." So I gave up after 2 hours to just enjoy the video. 😅
On the webpage version, all the rules are in one big paragraph. Would be very much improved if the rules got separated with a few empty lines, extra 's.
hmmm I haven't started yet this isn't clear to me, but I'm making an assumption that the "N" in the friendly cells refers to the same "N" in the magic square?
yes, generally variables of the same symbol the same variable. otherwise they easily could've change one of them to a different letter
Also, what else could N refer to in the statement "there are exactly N friendly cells" other than the N previously mentioned?
with all due respect, the phistomefel puzzles aren't very fun but it's just complicated figuring out the rules. Puzzles with new or complicated rule sets are complicated and cumbersome because of this, not because the logic is necessarily difficult
I am pretty sure that you are in the minority thinking the Phistomefel's puzzles are not very fun. Most people would describe them more as "the very best sort of puzzle using a particular rule set" and full of beautiful (not necessarily difficult) logic. There are puzzles by Phistomefel and also by Dorlir in some of the apps - they are not necessarily difficult at all, but they do show the unique strengths that each has as a constructor.
red is rong (noted)
Very nice puzzle; sloppy solve from me. 100:22 after so, so many false starts.
Another phistomofel classic!
Phistomefel had nothing to do with it.
Don't say it's Phistomephel when it's not. You're better than that. You don't need to lie to people to get them to watch. You're like Field of Dreams. If you solve it, they will come.
47:31 for my time. Fun little puzzle. The opening has a not too difficult but rather unique logic.