The Return Of A Sudoku Legend

** TODAY'S PUZZLE **
The sudoku world has missed one of its giants but we're delighted to see a new puzzle from the great Jay Dyer. It's called Cave Of Killers and it's just wonderful.
Play the puzzle at the link below:
sudokupad.app/aem1ckhlvc
Rules:
Normal sudoku rules apply. Shade some cells so that the unshaded cells form a single orthogonally connected area, and each connected shaded region touches the grid boundary. A circle is always unshaded, and the digit in the circle equals the number of unshaded cells visible in a straight line from that cell, including the cell itself. Shaded cells obstruct the view. Each shaded region contains a corner clue which equals the sum of the digits in that region. Digits may not repeat in a shaded region, and the digit in the clued cell equals the size of the region.
** NEW SANDRA & NALA FOG PACK **
A bonus sudoku pack was released on Patreon this week of 5 fog of war sudokus to celebrate the arrival of spring (in the Northern hemisphere) from Sandra & Nala. It's completely free - just click our Patreon link below.
Other treats on Patreon include:
- the new James Bond sudoku hunt competition;
- Simon's latest forays into the world of Islands Of Insight;
- Mark's video looking at the new OneUp puzzle from Rodolfo Kurchan;
- his solve of Region Geometry by Emre Kolotoğlu (3hr 36min long...!);
- and Mark's latest solve of The Times Club Monthly cryptic crossword
/ crackingthecryptic
** GET OUR FOG KICKSTARTER DELUXE & OUR BOOKS **
Check out this link for the kickstarter books and Fog Novella we've created over the years:
coffeebean.games/product-cate...
▶ SUDOKU PAD - Use Our Software For Your Puzzles ◀
You can input classic sudoku puzzles into our software and help support Sven, the programmer responsible for the wonderful user interface we all use to play these puzzles everyday. The app also comes with 12 handmade puzzles from us:
iOS:
apps.apple.com/us/app/svens-s...
Steam: store.steampowered.com/app/17...
Android:
play.google.com/store/apps/de...
ALSO on Amazon: Search for “SudokuPad”
▶ Contents Of This Video ◀
0:00 Theme music & "I've missed you" message
0:57 Puzzle intro
1:55 600k Subscribers Pack
3:42 Happy Birthday bonanza
7:58 Rules
12:29 Start of Solve: Let's Get Cracking
▶ Contact Us ◀
Twitter: @Cracking The Cryptic
email: [e-mail]
Our PO Box address:
Simon Anthony & Mark Goodliffe
Box 102
56 Gloucester Road
London
SW7 4UB
(Please note to use our real names rather than 'Cracking The Cryptic'.)
**************************************************************
▶ CTC FAN DISCORD SERVER◀
/ discord
▶ OUR BACK CATALOGUE - ALL CATEGORISED WITH LINKS!◀
tinyurl.com/CTCCatalogue
▶ NEW CRACKING THE CRYPTIC MERCHANDISE◀
teespring.com/en-GB/stores/cr...
▶SEND US PUZZLES TO SOLVE/CONTACT US◀
[e-mail]
▶FOLLOW US◀
Twitter: @crypticcracking
Instagram (for how to solve daily clues from The Times): crackingthe...
We also post the Wordle In A Minute videos on TikTok.

[@jonathancard4466]

The most frustrating thing about shouting an answer at the screen is knowing that Simon doesn't see it because he's being more clever somewhere else in the puzzle.

[@iambicpentakill]

To me, I feel like someone slumped on the couch and eating potato chips, while yelling at the professional athletes on their tv

[@katiekawaii]

​@@iambicpentakillHaha that's exactly it! 😅

[@koenth2359]

Generally, when Simon doesn't see something, it's because he has an intuition for where the breakthrough in the puzzle is to be found, and he'a actively looking for that. A number that can be filled in (but is a dead end) does not draw his attention. But then again, he also has those moments where he goes to great lengths to prove something overlooking simple logic that proves the same.

[@joethornton5321]

Also a recorded video does not allow for two-way communication, neither through space nor time.

[@jaydyer3999]

Thanks for the solve! I'm glad you enjoyed it as much as you did. It's good to be back doing puzzle stuff, so hopefully there will be more puzzles on the way in the near future. I can see that a few commenters have spotted your deliberate mistake 😉 - While the 21-cage doesn't go North as you said, it could still have gone into r4c5 and then headed left. Luckily you didn't end up bypassing any of the interesting steps in the solve, as you probably would have resolved it pretty soon using the 37-cage anyway. You just got to do some stuff in the bottom left a bit sooner than expected.

[@JohnDBlue]

If I may ask, is it intentionally not mentioned that the shaded regions can't touch each other? I noticed a few other commenters were also confused by this. Simon seems to be basing his assumption on them not sharing edges on the logic of previous puzzles, in my opinion, as the rules don't say anything about the possible connections between shaded regions.

[@spryteXO]

​@@JohnDBlueif they touch each other orthogonally they become the same region. If they touch diagonally the green can't connect to itself.

[@chironchangnoi]

Jay, your puzzles are always such a joy to solve, and to watch Simon solve. I had a lot of head scratching bafflement watching this solve, both because I couldn't see the logic but I could see Simon missing the sudoku. Especially on the easy early question about where the 7 went in the 37 cage. I really loved the way the 37 and 24 cages interacted. Thank you so much for the wonderful, mend bending, puzzling you bring to this community!

[@stevesebzda570]

I was wondering about that r4c5 cell. Thanks, Jay. Continued now . (It's a shortcut, but I've gotta follow Simon) Thanks again .

[@slurpcurry4253]

I came down to comment this as well. He could’ve gone “north” from that central cell, and then all the way to the left hand perimeter of the grid

[@Sujisan4]

Can't think of a more deserving youtube channel to hit 600k subscribers. You are only the best channel on KZbin.

[@sjm6280]

His "I've missed you" was so sincere, and he also returned from his break calling pink... pink! Simon is great

[@fordprefect9027]

One of the reasons I love this channel is that the logical elegance is explained with enjoyable eloquence.

[@longwaytotipperary]

@fordprefect9027 so very well said!!!! Perfectly stated!!!

[@davidrattner9]

Indeed ford...brilliantly written from you!!

[@chrishawkins963]

at 25:00 onwards, isnt simon missing that the central cell can move to the adjacent cells even if it doesnt move in that direction? i.e the final region he draws could have r4c5 instead of r5c4 as far as i can tell. you can still get to 30:30 either way and prove it doesnt by considering how the r2c3 cage, r1c5, and the r1c8 cage interact and its a shame that it was missed

[@Manigo1743]

Yes, his logic is flawed.

[@CyberTom1965]

@@Manigo1743 Simon always gets away with such oversights. If i do it, the puzzle breaks - usually not immediately, but half an hour later.

[@stephenmarseille5425]

I agree. He missed that and got lucky!

[@MarkBennet10001]

I know this took me some thought to resolve - Simon was lucky. But though it required thought, it was not hugely deep - non-trivial though.

[@sxmchn]

Was going to say the same thing.

[@longwaytotipperary]

So glad you’re back and that you missed us while having fun! Missed hearing birthdays- I really like that part!! Reminds me that we are a community that cares about you and Mark, sudoku and each other!! Also good to see a Jay Dyer puzzle!!!

[@Pulsar77]

3:00 Did you just say "a pack of 600000 themed puzzles"? Well, that's going to keep me busy for a few days 🙂

[@kennetsdad]

Kennet thanks you very much for the birthday shoutout, Simon!

[@kathyjohnson2043]

happy birthday!

[@longwaytotipperary]

Happy birthday Kennet! Lovely family!

[@kennetsdad]

@@longwaytotipperary thank you; that’s kind.

[@kennetsdad]

@@kathyjohnson2043 we thank you

[@joshzaid8589]

i love this channel haha, but sometimes I'm just sitting here shouting in my head "think where does the 7 go in the 37 cage" for half an hour

[@davidrattner9]

Great to have you back Simon from holiday and that you had a wonderful time. We missed you as much you missed us. Your passion for what you do is exemplary!!

[@zealot2147]

46:26 for me, the logic flows very well when you realize only shaded cells can spring from the clues and have limited sizes. Using the necessity of them reaching the edge to narrow where they can move combined with them needing to block some circles made a solve path that’s challenging but never feels impossible. It looks very daunting on open but has a lot of useful info as long as you can parse it. I’m happy to solve it in just over half the video length time

[@benimanah]

Truly love your videos - perfect for when you're stoned off your gourd and want to make yourself feel smart

[@IOSARBX]

Cracking The Cryptic, This video is fantastic! I liked it a lot!

[@meiter]

What a lovely surprise to see another Jay Dyer puzzle. Oh what have I missed this setter's work

[@RecreationalCynic]

Finished it in 92:49. I wasn't familiar with "cave rules", but saw early on that cave rules are just a special application of yin-yang: yin is the single unshaded region in the center and yang is the group of shaded regions connected to an imaginary ring of shaded cells outside the grid (or the other way around, whichever boats your float). I was able to apply the checkerboard secret thanks to learning it on this channel. The sudoku at the end was the toughest part for me, but placing the rightmost digit of the 24 cage unravelled everything afterward. Also, I thoroughly enjoy listening to the birthdays and missed them for the past week or so.

[@iambicpentakill]

I didn't make the connection of the imaginary ring outside the grid, thank you

[@emilywilliams3237]

It's interesting, I was thinking just the other day that it had been a while since Jay Dyer had appeared on the channel, and voila, here we are! It is always fun to see you do a cave or yin-yang puzzle, Simon, because you enjoy them so much. Thanks for this video - and welcome home.

[@soapagel]

27:08 Is there a reason the cage cant take r4c5? I cant see one.

[@73dsotm]

That is just what I'm thinking also!

[@paulv3lv]

Yea I'm completely lost why the 21 cage couldn't have gone to r4c5 as well.

[@MaierFlorian]

if you do, what is your next step? you can't visit column 4 or you bump into another shaded region, so you need to take r3c5, than either r3c6 or r2c5. In Both ways, you need to finish with r2c6 and r1c6. In doing so, the circle in r1c5 sees the whole of this shaded cage in terms of sudoku, and the digits in the shaded area are 123456, leading to at least 7 in the circle (this is the point simon absolutely misses to explain properly, took me like 5 minutes to understand what the heck is going on). But the circle only can see up to 5 unshaded cells now, so it can't be at least 7, leading you to a repeated digit either in column 5 or box 2.

[@stephenmarseille5425]

Simon was wrong to green it for the reason he gave... he got lucky! I wish I knew the logic though.

[@MaierFlorian]

@@stephenmarseille5425 I literally explained every single step, why he did not get lucky but actually knew what he was doing. He just did fail to explain his thoughts properly.

[@stevieinselby]

I needed a bit of help here, although I really shouldn't have. I figured out that the wall in r5c5 couldn't go _up_ because you'd have a stacking problem in the circle in r1c5, but failed to notice that exactly the same trick could be pulled with all-but-one of the other routes to the edge 🙄. With that hint, it then took me about an hour. 👀 Simon, looking at three cells that he has pencil-marked 5s into and explaining in depth why they can't be from 1-2-3-4 ... never change, Simon!

[@LednacekZ]

27:07 how does he know that r4c5 is unshaded? How did he rule out the possibility of the center shaded area extend along the 4th row to r4c1?

[@MaierFlorian]

you can't go to r4c6, obviously, and if you go to r4c4, you need to go the whole 4th column straight up, bumping into another shaded region, which renders the corner clues incorrect.

[@LednacekZ]

@@MaierFlorian which one. you can make a 2 square region for 9, and the region of 37 can be made too without touching.

[@stephenmarseille5425]

He didn't know, he got lucky :/

[@10prozenthimmel]

Why would you need to go up column 4? Just go from r5c5 to r4c5 then turn left and continue to r4c1.

[@MaierFlorian]

@@LednacekZ the 37 region, obviously? If you go up column 4, you visit r2c4 which is right next to the 37 clue?

[@ericpraline1302]

Lovely puzzle. Jay's puzzles are usually quite tricky, but I always feel there is some clean logic to be found so that one never gives up hope. That central cell was delightful.

[@anaayoung9142]

I tried this puzzle, I couldn't finish at first so I watched your video to help me. At 27min he didn't consider that R4C5 can be part of the 21 cage. As I didn't want to follow this without consider that cell, so I decided to try again the puzzle. If you are lost like me, just try again. Think about the 37 cage (it can be on the 3row?) and the R1C5 circle (how he will connect to the other unshaded cells? With the 23 pair in row 1, he can only be 4 or 5). The logic between this 2 and the cage 21 helped me through this part. And I was able to finish the puzzle after that. 🎉 If I could solve you can solve it too!! 😊

[@Millenassang]

Very nice one from Jay Dyer! Can t wait for a zetamath return as well

[@CauchyIntegralFormula]

71m29s. This was beautiful! Lots of fun logic, even right from the get-go figuring out how that 21 in the middle was connected to the ourer edge

[@frankjiang1857]

Finished in 52:47. Lovely use of the positioning of the killer clues and circles to make the break-in and the rest of the puzzle in general. This was a delight to solve! Fun puzzle!

[@MattYDdraig]

45:38 Elegant as ever with some really neat logic as expected from Jay Dyer. I found the sudoku elements quite tricky after finishing the clues.😂

[@angec9908]

That 5 in box 5 may not have pushed the solution forward but it was bugging me so I’m glad you saw it sooner rather than later 😂 (and I always thank you when you finally put on your sudoku hat)

[@OUdarling]

27:50 when Simon starting talking about boundaries and 6s, I thought I was watching Cracking the Cricket 😂 Fun puzzle, and it’s always good to see a Jay Dyer construction - even if I can’t wrap my head around them most of the time!

[@damienneighors]

I can’t believe I’ve been watching you for years and just never subbed. Can’t wait for 600k

[@chocolateboy300]

I finished in 135 minutes. This was really fun to logically work through. Although, I had my colors reversed and that did mess me up. Great Puzzle!

[@skiesofarcadia4885]

Welcome back, Simon! ❤️

[@marssang]

Simon assumed the first step of the center cage needs to be taken towards the boundary it connects to, and didnt even account for the possiblity that the first step could be the sidestep it eventually needs to take anyways. He doesnt usually miss logic like that, as thats not sudoku.

[@Gonzalo_Garcia_]

25:52 for me. What a fantastic puzzle, loved it!!

[@compiling]

Ah, heck. I should have thought about the rules a bit more, because I didn't realise that the shaded regions can't connect. It seems obvious now. I managed to finish after you told me how the puzzle actually works.

[@shawncarter7188]

I enjoy hearing the birthdays and news

[@user-gj5uc4yx3i]

I think a cleaner way to notate which cardinal direction the 21 cage could not go may have been to use the pen tool (perhaps red) to draw a 3 cell long edge line between box 5 and the box in the ruled out direction. Simon might not be 100% back from holiday just yet.😊😊

[@SamAHill]

Just started, but it's not clear to me from the rules that the shaded regions can't touch each other?

[@JohnDBlue]

I also can't see any reason they can't touch.

[@steve470]

The intended meaning (the only meaning that makes the puzzle work) is that a "shaded region" is defined as all of the shaded cells that touch each other. If two "shaded regions" touched each other, they'd be the same one shaded region. The wording of the rules may not have made that perfectly clear, but that is how the puzzle is intended to be solved.

[@SamAHill]

I confused the shaded regions with "cages", which can be next to each other of course. Not a big deal, but I know Simon likes to avoid taking his viewers' knowledge of terminology for granted. I would add a line to the rule that shaded regions cannot touch orthogonally.

[@SiIentSong]

​@@steve470 Sorry if I miss anything, but how do you know that the puzzle is intended to be solved that way?

[@RichSmith77]

Isn't it simply how regions are defined in the rules? The word "region" first appears in the phrase "each connected shaded region" so all the shaded cells that are connected define a region.

[@stangerrits6712]

Around 55:00 Simon seems to be completely forgetting the fact that he needs a gray somewhere above the two circles in box 8, to prevent them having the same value. Interesting, because a moment before he explains quite thoroughly why that has to be the case :) Based on this, the 24 region could never be 4, and because of the pencil marked 5s in box 6, it can't be 5 either. Nice part of the solve, to see how much work that 24 cages does!

[@xerodeus2337]

27:05 - why couldn't the wall go up 1 square and then straight left to the perimeter? Did Simon get lucky that this was the right way for the wall to go?

[@ozzycommander]

This is going to be a very good reason....... Brain think of the reason. That is going to be a 2024 Sudoku quote contender

[@Jocio13]

Hello, thanks a lot for all your very entertaining vidéos ! I have a question on the deduction done at 27:07 => the cell r4c5 is set to green I have understood that this grey zone r5c5 cannot go to the north side but for me, to go to the west side, this grey zone could have used r4c5 before going full left, no ?

[@RoderickEtheria]

Did Simon jump to a conclusion about r4c5? Could he not have taken it and then raced to the left?

[@brentusfirmus]

Yeah seems like he didn't realise and got lucky. The other comments say that you can still deduce the top three cells of box 4 will be shaded anyhow, and later in the puzzle the 37 cage dictates that r4c5 can't be shaded.

[@piarittersporn]

That was hard work but a very brilliant puzzle.

[@toms7114]

Once Simon got the purple pencil marks in box 8 and 9 it had to go in box 7 and the only thing purple could be was the 1. Drives me nuts when Simon stops his logical train of thought before it gets an answer because the next step is Sudoku.

[@homewort-gf2ni]

Welcome back, Simon! Could someone explain to me why, at 37:00 or so in the video, the circled digit in r1c5 couldn't have been a 1, and pincered by the 37 region?

[@MrSuperclove]

The unshaded cells must form a contiguous region 😉

[@MarkBennet10001]

This one is a beauty

[@tchpowdog]

How did he rule out r4c5 from being shaded?

[@stevesebzda570]

@28:28 it's unclear to me why that cell above the "6" can't be in the gray? (It's still 6 away ultimately to the "East") And all the "123456" wouldn't be seeing that green circle. Why can't it go north one cell then to the east?

[@stevesebzda570]

I'll have to continue with it being green -- Simon's gone too far. Could r4c5 have been gray though? Thanks ?

[@Oznej]

Lovely puzzle! Btw, it now has 1000 more solves within 24 h of this video. That's CTC for you! 😊

[@bait6652]

Saw.the first Obs...but that 2nd obs needed was doozy. Never would have thought to occlude what he did...but after that the coloring and puzzle went quick... Reminded me of SC creepcrawl or gameof life spread.

[@RichSmith77]

Oops. The chagrin you feel when you come to the comments and see everyone pointing out a mistake Simon made, and you realise you made exactly the same mistake. Of course it could still have gone to r4c5 and the left, even though it "doesn't go north [to the edge]"! Doh! At least I could restart the puzzle and do it correctly the second time. 🙂

[@RichardDamon]

24 / 4 / 24 IS "Palindromic" if you think of it as NUMBERS and not DIGITS, and as you said 4 / 24 / 24 is Palindromic as digits, so it works both ways using different meanings.

[@crazypantaloons]

I have no idea how I got that. I was throwing down colors, lines, and all of the sudden I used up all the clues!

[@Ardalambdion]

Oh no, Simon is greeted with chainsaws when he returns home. Someone obviously mad about sudoku, other logical puzzles and weird humour.

[@andrewhearne9199]

At 27:08 why is the row 4 column 5 cell green? I understand that grey has to go west, but it could go up first then directly across.

[@ellengustafsson574]

At around 30:00 couldn't you make that checker board work? To me it seems pretty easy to connect the green even if r1c7 is shaded

[@ellengustafsson574]

Oh nevermind I see it now, the six cell region can't touch the wall if that is a checkerboard

[@Lombax999]

I'm a bit confused about the rules, and would appreciate some clarification if anyone can help me out. The way I read the rules as written, nothing seems to prohibit shaded regions from touching each other orthogonally, yet at 18:00 Simon makes a deduction as if this were the case. I can quite easily see ways to make shaded regions touch each other without breaking the rules (they could touch and still both reach an edge, and the unshaded region can still all be orthogonally connected around them). Is this an oversight in how the rules are written? Is it meant to specify that the shaded regions can't touch each other orthogonally? Or am I missing something? *edited for clearer wording

[@abrodeur]

It may be little unclear but for this type of puzzle connect shaded region always includes all shard cells orthogonal connected. Since all corn values are different, none of them can connect together without causing a contradiction. With in the puzzle community they try to make sure rule set is all that needs to be follow but there are many assumed rules based on type of puzzle.

[@Lombax999]

@@abrodeur Thanks for responding. I've seen some other comments with similar concerns and caught up a bit on the situation regarding this ruleset. Just in case anyone reads this who has the power to act on this feedback, I'd like to say that I think the extra clarity is needed in the rules. For context, I've been following CTC and doing their puzzles for a while now, and I think it's important to not include assumptions like this if they can avoid it. I mean, who's to say someone might not make a ruleset with a small variation such as this in the future, right? It's also just helpful for all the rules lawyers like myself out there to be as clear as possible, and it doesn't really have any downside to add a couple more words to the rules. I get where they're coming from, but that's just my 2 cents.

[@steve470]

65:19 for me. That one was a bit mind-twisting!

[@user-hz3my6yv8w]

Where is the rule that prohibits two shaded regions from sharing a boundary?

[@zealot2147]

It’s implied. If 2 regions were to touch the clues don’t work. They would be one region, because touching, but one of the sums of cells in region clues would be wrong. To put it more simply if they touched they’d be one region but they are two. Just doesn’t work

[@JohnDBlue]

​@@zealot2147 what implies they would somehow merge if they were touching?

[@user-hz3my6yv8w]

@@zealot2147 With respect, that's an assumption, not an implication. Unless there is a definition somewhere that defines a "region" that way. In which case, that's a rule (and one that's needed to solve this puzzle) that is not documented. The 9 boxes of standard sudoku are often called regions, and they certainly share edges.

[@moosesox]

In variant sudoku, I’ve never heard of the 3x3 boxes being called regions. Only in irregular sudoku, or for crafted region sum boxes like this. In cave sudoku, a shaded region is one block of shaded cells. An unshaded region is one block of unshaded cells. Once two wall sections touched, it would become one bigger region and both clues would be wrong.

[@JohnDBlue]

​@@moosesox I know where the assumption comes from, however the rules should never be left open to misinterpretation. Everything else is explained no matter how basic it is, no?

[@gibbbon]

placed a green cell too fast on 26:59, could have been part of the grey 6

[@TurquoizeGoldscraper]

I'm just going to watch.

[@DarrenNakamura]

I broke it twice but I finished with a total time of 101:45. Good puzzle, I just applied bad logic to it.

[@deathpigeon2]

Whenever you say something is interesting, then make a mean comment about how you aren't fun at parties because that's something you find interesting, it's always something I also find interesting.

[@tremkl]

Is it weird that in my head canon, Jay Dyer is related to Danny and Dani Dyer? 😂

[@dysphoricpeach]

ooo a long one!

[@srwapo]

59:37 with three peaks at the video. 1) didn't see how to start, missed the rule about the digit in the cell with a clue equals the size of the region. 2) had to see how the 6 in the center of the grid had to go to the left. 3) didn't see how to deal with the 24 region, looked at enough to see that I had to do some trial and error to work that one out.

[@andytaylor445]

At 29:29, why couldn't the 6 cell wall, go directly north, then west?

[@tamergaber00]

"Once wiggled"

[@WimmekeVL]

Alas just clicking subscribe is not that easy if you're watching on a TV set. It kind of expects you to have some kind of account. TV's don't have much of a keyboard other than their remote control, so I haven't gotten around to even trying to enter account credentials there.

[@Zardoz37]

Does 'corner clue' mean that the clue has to appear in the topmost then leftmost cell of the shaded region? Simon's explanation in the Rules section (regarding the 37 region) broke this constraint, but if that's not what 'corner clue' means, then I'm not sure what it means. In fact, 'corner clue' cannot mean what I was expecting it to mean: it breaks the 13 region, which would then have to be 2 cells in the bottom row containing a 2, which is impossible. But I still don't know what 'corner clue' actually means. Just that it's in the corner of a cell? Ie no restriction on where in the region the clue appears?

[@RichSmith77]

Yes, it simply means the clue is in a corner of a cell. You're overthinking it.

[@vandelay33]

At 26:30 how does he eliminate r6c9 as the destination of the grey area? It could go via r6c6 which isnt seen by r5c9

[@linkzmax]

R6C6 cannot be shaded or it would create a checkerboard by connecting to either the 21 or 24 clue

[@rankinepiper]

I’m having a hard time understanding why the central cage couldn’t drop down and head east. I get it can’t go up the way, by when he blocks r6c5 it could still go to r6c6 and see 7 cells . What am I missing

[@11122276]

It can't see 7 cells as it totals to 21. But I also fail to see why it can't go east.. [edit] maybe I mistunderstand your comment.

[@RichSmith77]

If it dropped down and then headed east, it would connect with the 24 clue in box 9.

[@11122276]

25:00ish I am flummoxed. Why does he conclude that r5c9 sees the entirety of any gray region that heads east? There aren't knight/king constraints... what am I missing...?

[@musicmilo]

The central cage can't touch the 24 corner clue, and the circle in box 5 prevents the cage from going north until it reaches the 6th box. That means that regardless of when the cage cuts north and moves east, the circle on the east border can see the entirety of the central cage. That would eliminate the digits 1-6 as a value for the circle, and would need to be at least a 7. However, that implies the circle would be able to see at least 7 unshaded cells, and there's not enough unshaded cells. At most it would be able to see 6. Hope that clears it up

[@11122276]

I feel like a prize doofus. It's a sudoku. Completely failed to see the 3x3 boxes.

[@CatzHoek]

Isn't the deduction that column 5, row 4 is green at 26:53 false? Could have been that you go one north and then all the way to the west. or?

[@Raven-Creations]

At that point in Simon's solve, you're right. If he'd done more with the 37 region first, he'd have found it had to go into R3C6, removing that option. The 23 pair in R1 meant that R1C5 had to be at least 4, which meant the 37 region could only take at most one cell in R1. It also had to avoid the 11 region (which is now 3 cells). This meant it had to extend into R3C6.

[@LednacekZ]

49:33 for me. quite tricky start.

[@Raven-Creations]

You said the 21 region can't go north and made R4C5 green. That was bad logic. It can't hit the top edge, but it could have gone north and then gone west. You should have thought about the 37 region first. You could grey the cells in R4 in box 4, which makes the 9 region two cells, and the 11 region three cells. R1C5 has to be at least 4, but can't be 5 or the 37 cannot reach the side. The 37 can only take one cell in R1. This means the 37 must extend to the east, and avoid the 11, so it must take R3C6. This forces the 21 to take its final position. Once you worked out that R5C1=5, you could have made R4C4=5

[@therealedvaize]

(Edit: Missed one bit, check replies)For people wondering how the logic works with the r4c5 at 25:00. Yes simon is just missing that but it gets disambiguated quite quickly by the logic with r1c5 later. You can determine that r4c1, r4c2, r4c3 and r4c4 all have to be part of the 21 region because you have to take them if you take r4c5 and if you take r5c4 you also have to take them to make sure r5c1 works. Now you know most of how the region goes except for the last cell (r5c4 or r4c5), you can green around those gray cells and r5c1 still has to be 3, 4 or 5. Now if you make r5c4 green you can't make it work even with 5 because r7c1 couldn't reach a region.

[@keithtemple8418]

Why can't r5c1 be an 8?

[@therealedvaize]

@@keithtemple8418 Wow I missed that apparently, but I got around it a different way. If we claim r4c1-5, r3c5 is still green and r3c6 also gets green to avoid a checkerboard. Same logic applies that simon did that the 37 clue has to extend to r2c6 and that r1c7 has to be green. 11 clue extends and r2c7 also gets greened and now the 37 clue is trapped in. It can only get to five cells to not break r1c5. Feels a bit like bifurcating but it's not insane.

[@AG_247]

Of course, it gets disambiguated, that's not the point. Simon unfortunately missed it (around min 27:00 to 29:10). The logic was correct, when he said "we have to go west", but at this point, r4c5 should still remain unknown instead of making it green. Then the solver should figure out, which 6 cells could be possible and be able to discard r5c2, r5c3 leaving two possible cages (5 cells in row 4 plus the circle OR 4 cells in row 4 plus r5c4 and the circle).

[@therealedvaize]

@@AG_247 Yeah it just seemed people were wondering how it got disambiguated. My point was though that it's not an error that would end up breaking the puzzle at the last minute, it actually broke quite quickly, but it's just unfortunate that simon missed it.

[@eporama]

Yesterday was also 4 4! 4! (Or 4! 4 4!)

[@stevenlacrosse4876]

1:00:27

[@theredstoneengineer6934]

36:43 for me

[@zachnichols8642]

1:15:50 nice

[@nezarnassif2967]

I FINALLY FOUND A MISTAKE 😅😅😅😅😈😈😈😈

[@clairerothero724]

If I met Simon at a party, I would be so excited I wouldn't know how to behave. "Interesting" just depends on finding your people, and you've certainly found them, Simon!

[@JohnDBlue]

Am I misreading the rules or just not thinking correctly - I can't see why two shaded regions can't touch each other based on them. Hmm.

[@steve470]

The intended meaning (the only meaning that makes the puzzle work) is that a "shaded region" is defined as all of the shaded cells that touch each other. If two "shaded regions" touched each other, they'd be the same one shaded region. The wording of the rules may not have made that perfectly clear, but that is how the puzzle is intended to be solved.

[@smokybros1267]

At 18:10, when he says that r2c2 must be unshaded, I don't see that logic. The rules do not state that the shaded regions cannot touch, but he is talking like that is a rule.

[@andrewnanninga1966]

“Each shaded region contains a corner clue which equals the sum of the digits in that region.” If two shaded regions touch and have different values that would be impossible. In the case of r2c2, if the shaded regions touched then the region would have to sum to 9 and 37 simultaneously.

[@MaierFlorian]

he explains it, there is no explicit rule but you break the puzzle if you do let them touch diagonally in every Cage- or Yin-Yan-Puzzle per definition.

[@Esperi74]

​@@andrewnanninga1966 I don't see anything in the wording that precludes 2 regions sharing an edge and still being regarded as separate regions. Just draw a border between them? E.g. if the 9 region in the top right was R1C1-3 - draw a box around those, have the 37 region touch the top or left side, as long as you don't isolate an unshaded cell, you can still draw a contiguous unshaded region throughout the rest of the grid. In other words, nothing says that two shaded regions sharing an edge are then considered "merged" into a single region.

[@JohnDBlue]

​​@@MaierFlorian this is neither a cage nor a yin yang puzzle - and cages are absolutely allowed to touch each other in typical killer cage sudokus Edit: ah I just realized you must be thinking of checkerboard avoidance which is completely different from sharing edges

[@MaierFlorian]

@@JohnDBlue ah, now I get your point. Shaded regions can't share an edge because that would make that region bigger, blowing up the numbers in the corners (and would make both numbers the same). Maybe the rules don't mention that really explicitly but that's standard wording of this kind of shading puzzles so I guess it's just considered common knowledge that you can't have invisible boundaries between cages but they union once they share an edge.

[@matthewjohnson6360]

I always just skip to the rules.

[@AG_247]

I sometimes did this as well, but no longer do. Thankfully so, because otherwise I would have missed the awesome story about that one British man who creates riddles and solves them for decades, and corrects them manually and tracks records..

[@iceberg54321]

The ruleset is broken, the circle cells only count if they see themselves, and unshaded cells directly horizontal or vertical from the circle in the cell. The definition of a straight line does not stop you from drawing a straight line at an angle that is not blocked by a shaded cell. You could draw a straight line horizontally or vertically from a corner or extreme edge of a circle cell to pick up more cells, or draw it at an angle to pick up more unshaded cells that are not blocked by a shaded cell.

[@michaellautermilch9185]

Lol points for creativity

[@mipsuperk]

8:45 what does orthogonally connected mean? It doesn't mean anything, the expression is total gibberish. Orthogonal means perpendicular. Two adjacent cells are not perpendicularly connected, that doesn't even mean anything. Please setters, stop it.

[@RichSmith77]

I'm afraid that that boat sailed a long time ago. It's been used that often in the variant sudoku world, it has now come to mean something, in this context, simply by repeated usage. There must be thousands of puzzles that have rules with the word "orthogonal" meaning "connected along an edge" by now.

[@abrodeur]

This were simplify of definition in high school catchs up with you. orthogonality is the generalization of perpendiculity. Perpendicular is special case when lines are orthogonal. Kind of like how square is special case of rectangle when all sides equal length. So your correct say two cells are perpendicular is gibberish but your not correct to then say two cells orthogonal is also being gibberish. There is reason why we have two different terms. In high school yoir told they mean the same thing since you only every see concepted applied to lines. orthogonal can refer to other object not just lines and there relationship. When extending idea of orthogonal to other concepts often you define two lines in this context and test to see of there perpendicular or you use the vector definition of perpendicular aka vector are perpendicular if there dot product of the two vector is zero. From my understanding cells consider orthogonal if they share the edge since the direction of connection is orthogonal to edge they connect at. E.g if two cells are connect horizontal aka (r1c1, r1,c2) then there edges where they meet at is vertical. More mathematically cells are orthogonal if you draw a line from the center point of two cells and that line is perpendicular to edge where the two cells meet.

[@RichSmith77]

@@abrodeur Orthogonal simply means two directions at 90⁰ (a right-angle) to each other. The two diagonals of the grid are also orthogonal lines. If all the cells only connected at their corners, then those would also be orthogonal connections, all at ±45⁰ to the vertical (if the connecting lines continue through the cell centres). It's just, in the current sudoku world, the orthogonal directions are always taken to be vertical and horizontal, to match the grid lines, and "orthogonally connected" means pairs of cells are connected, either vertically or horizontally, along an edge.

[@mirtexxan]

First! :)

[@emilnilsson1935]

third

[@_-_-Sipita-_-_]

37:59 for me