Solving An Expert Sudoku From India

A very good solver from India asked us to take a look at this puzzle. We're happy to oblige and it was no easy test!
▶FOLLOW US◀
Twitter: #crypticcracking
▶SUPPORT US + ACCESS CRACKING THE CRYPTIC PUZZLES/VIDEOS◀
Patreon: / crackingthecryptic
▶SEND US PUZZLES TO SOLVE/CONTACT US◀
[e-mail]
▶SOFTWARE◀
We mainly use Duncan's Sudoku Solver, which is available at www.littlegogs.com/index.shtml
▶ABOUT US◀
Hi! We're Simon Anthony and Mark Goodliffe, two of the UK's most enthusiastic puzzle solvers. We have both represented the UK at the World Sudoku Championships and the World Puzzle Championships. We're also "cryptic crossword" aficionados. Mark is the eleven-time winner of The Times championship and Simon is the former record holder for most consecutive correct solutions to The Listener crossword. We hope we can help your puzzle solving while also introducing you to some of the world's best puzzles.
Thank you for watching!
Simon and Mark

[@bhanujaysingh7012]

I never knew there was a that technique for solving the 2. Thanks a lot for solving the puzzle. You even bypassed the XYZ-wing.

[@sper1585]

I also got stuck at the 2 6 pairs in the middle. Once I saw the reasoning of those four squares couldn't be independent I was able to solve. Thank you for the video.

[@duaneraeburn8037]

I do like watching your mind work so I don't worry about the gaps I can hear the wheels spinning in that head of yours. Well done.

[@JustJohnny666]

Same. I rarely find something before him. So i'm not stressed about it

[@kasonday358]

Watching the mind of a master at their craft is always amazing. Your videos have helped me immensely to better understand the algorithms associated with Sudoku, and while I'm still somewhere between a novice and intermediate puzzle solver, I am steadily developing the tools to be a master craftsman thanks to you!

[@jamesgraczyk]

Great video! Just started really solving harder puzzles and you breaking down what you are thinking really helps me out!

[@abubakardouglas8268]

Another excellent solve to a very good puzzle. This was the first time I got to the "big pause" before you, which I regard as an achievement. It's interesting to see the way you "tidy up " at the end, in a completely different way to myself. Since watching your chanel my solving has improved greatly. Thanks.

[@florianassmuth5257]

I watched two of your videos before, I installed sudoku on my phone, and have been having a blast. I have been getting stuck on the hardest levels and this video cleared a lot up. Thanks!

[@eralipe]

Enjoyed watching you struggle a bit. Good insight into the advanced methods, loved it!

[@lorabecker586]

Magnificent solution! I learn more each time. Only been watching over a month and learn each time. My logic skills have increased exponentially.

[@Mizziri]

This was a pretty brutal one. Got to the first uniqueness trick and switched off of snyder notation. Found both uniquenesses, and cracked the rest after staring at chains for about 10 minutes before concluding that the cell in row 4, column 3 can't ever be a 2.

[@ugyennorbu9443]

Perhaps one of the hardest one that I have watched from your channel. Great work..

[@Carbunkal]

I don't mind the gaps at all. It gives me time to process what you have done and think about what I would do next. Nice work on the puzzle!

[@helenscott8202]

Struggles help me immensely. I am glad to watch someone else struggle besides me. Thank you!

[@oswaldchangerton5328]

Thank you, this is excellent to watch and learn from!

[@igotmesomeabs]

I loved soduku when I was younger but always solved it vertically or horizontally with no regard for the actual 3x3 square. I love this method and I'm going to have to start solving them again!

[@pavanka1]

Great one, i solved this by brute forcing set of assumptions, erasing and backtracking next set of assumptions, but your video really helped to spot 4square pattern, and xyz was gold. Best video explanation

[@thephysicistcuber175]

At the 20:00 mark (the point where you had to use the empty rectangle) there is an XYZ wing on 1 7 and 3 with pivot on I8 and other cells of the wing on G8 and I4 that allow elimination of a 3 from cell I9. That leads to a hidden single in row 9 (cell D9 forced to a 3) which might be another way of cracking the puzzle (haven't really checked that one).

[@Kurokuma10]

Great lesson! Thank you for the videos!

[@HULKEN1UP]

Great puzzle and solve! At the point of the empty rectangle there is also the 2s in row 2 and 6 putting that same 6 in r4c4.

[@mtmimulus7767]

I enjoyed watching that process and learning.

[@leporid257]

this is so old it doesn't even have a link to solve on your own :o

[@kaigreen5641]

Took me just under 40 minutes to complete, very fun puzzle. Lots of triples. Watching the vid now, can see you doing the same things I do "Oh that gives me that and... nothing, damn"

[@felegrin]

I really enjoyed this one. Very helpful

[@DonovanCYoung]

Nicely done.

[@danielleanderson6371]

The crazy thing is you can also use single chains at that exact spot to rule out one of the options on a 2 6 square (I forget which) to get the same effect. So even though that step does require one of the more advanced moves, you actually have a few options to choose from, depending on which you're better at spotting. I, of course, spotted neither and had to ask Duncan to give me a clue.

[@Matthias-wm8zi]

A finned Swordfish of 2s also works (I even spotted it myself :))

[@mirandaqomoyi2693]

I can follow all your logic when watching you solve but although I can solve hard puzzles I don't yet spot the patterns for the cleverer logic on my own. I'm hoping to get there gradually. I was quite pleased with myself in the beginning because I thought of Snyder notation for myself - as a philosophy graduate I think the course in logic helped.

[@55ATA3]

Thank you I enjoyed that video...

[@billjames8036]

great solve.

[@jonnybolton4659]

My way of solving the "sting in the tail" is to follow the 26 pairs around the grid (colouring), you quickly see that R2C4 and R6C3 are a 26 pair and R2C3 sees both so must be a 7

[@manuelvega1290]

I didn't know you could use uniqueness that way. that's gonna help a lot in my future Sudoku solves. Thanks for the solve!

[@NjniaVanDerWald]

though i have encountered numerous sudokus, in which that uniquess rules doesn't apply. i guess it always depends on where you get your sudoku from.

[@danpowell806]

@@NjniaVanDerWald If the uniqueness rule doesn't apply, there are at least two different solutions that match the starting condition.

[@NjniaVanDerWald]

@@danpowell806 not always. I've encountered at least two sudokus where I was left with four pairs at the end. Obviously that's not the rule but an exception.

[@leonardosouzabrum7825]

@@NjniaVanDerWald that would certainly not have only one solution, unless they are in four different boxes.

[@deucedeuce1572]

19:55 ... At the top right box (box 3) is a perfect 6-way match, so if you get one number, you automatically get the rest of them. I call it a "circle", but it can also be a triangle (with 3), a square (with 4) and can be other shapes with more, or even like an "8" or infinity symbol when the (penciled in) numbers cross (like in this situation here).

[@jurgenkloppshead8373]

Finding it hard to spot empty rectangles but always seem to find a skyscraper or finned x-wing that gives the same result. Is this always the case or do I need to work on this?

[@CreeperGoesBoom59]

I'd love to see more of the dailykillersudoku puzzles!

[@paulreader1777]

At 14:40 in the middle block of the top row with the 59 pairs and the 267 triple is it not correct to surmise that row onw column six must therefore be a 6 to leave the 59 and 27 pairs in that block as the only unresolved cells. I have, for quite a long time, used this technique intuitively but I don't know if it is correct or simply guessing.

[@varishnakov]

The amount of difficulty and the time it took for you to solve this killer puzzle is the same as when I try to solve New York Times Hard puzzles.

[@Frie_Jemi]

QUITE NICE PUZZLE! thanks to India

[@deucedeuce1572]

23:48 ... I call that a "Zipper", when solving one number solves a whole bunch of other pairs that were all unsolvable... and/or a "domino" when it doesn't just solve pairs, but the whole puzzle. (Usually it's a "zipper" if all you really have left is a bunch of pairs.)

[@Matthias-wm8zi]

13:37 There is sort of a "finned" Swordfish of 2s in columns 1, 5 and 6 which removes the 2 from C3R6. That is Resolving this mess of 26 Pairs

[@hhaavvvvii]

If you ignore the empty rectangle, is it possible to get the puzzle into a BUG? It was looking like it was real close with there being only two spots that could take three numbers.

[@fubaralakbar6800]

I solved the empty rectangle around the 22-minute mark by using alternate inference. If r6c3=2 then r2c3=7, if r6c3=6 then r4c4=6, r2c4=2, r3c2=7, therefore r3c2=7 This one took me a couple hours, which is normal for me because I do them pencil-and-paper style in a spreadsheet. It was quite a satisfying solve!

[@markseare9274]

Thanks for the classic sudoku! They help me a ton. I know they must be boring for you.

[@sagesteppe9090]

Yes

[@alexanderstelmakh8906]

15:35 there is a easy 26 chain (not 2, 6, not 6 - 2 and so on) showing that r2c3 cannot be 2, it is 7.

[@tusharagrawal7560]

Can someone plz explain me what he did at 8:45 ???

[@eternalblasphemy6526]

Took ~35 minutes but I did it on paper. Was a tricky one, liked the uniqueness trick and 267 force to break the puzzle.

[@HeroDarkStorn]

I always feel like cheating when I use "if it's here it cannot be unique" Is there always some other way to break non-unique constellations? I guess my question is, is it possible to make a puzzle with 3 possible solutions, 2 of which would be eliminated by uniqueness argument.

[@martinepstein9826]

"is it possible to make a puzzle with 3 possible solutions, 2 of which would be eliminated by uniqueness argument" Nope. If the rules of sudoku allow for more than one solution then there's more than one solution. Uniqueness arguments aren't logically valid without uniqueness. What you're describing sounds more like adding an extra rule against certain "deadly patterns".

[@Czeckie]

13:56 resolving the 8? I agree it's there but the reasoning in video seem kinda shady, what am I missing?

[@MrDrywell]

The missing numbers from the row are 1,2,3,5,6,8. There is already a couple of 2,6 pairs locking out the 2 and 6 from the remaining 4 squares. Thus, the four remaining squares must be restricted to (at most) 1,3,5,8. The 8 can be deduced as a 1, 3, and 5 already exist within the box, restricting the square from 1,3,5,8 to 8.

[@brunomendes548]

2/6 pair already locked, 1/3/5/8 for the last 4 spots of the line, and 1/3/5 already placed on the middle block just above makes 8 only go in that spot

[@Czeckie]

oh, thanks. I haven't noticed at this stage that 2/6 is locked. Interestingly enough I've logiced around it somehow.

[@peterjongsma2779]

Enjoy watching you think.

[@whocareswho]

I wonder about the 2-6 pair and uniqueness. If the 4th cell can hold 2-6-7, couldn't a valid pair in that cell be either 2-7 or 6-7? I don't see how that would create ambiguity or impossibilities in and by itself.

[@lsmart]

The point is that if r6c6 contained a 2, it would solve all 4 cells one way, and if it was a 6 it would solve all four with different numbers, yet both sets would not affect any of the remaining cells in the puzzle; hence, there would be two solutions that work. The assumption in all these sudokus is that each cell can only have one number that will lead to a viable solution, and that would clearly not be the case with a 2 or a 6.

[@niroracoon2981]

This is all lovely and great, but there is one thing that's bothering me. I couldn't solve extreme sudoku's cause the website i play on, doesnt have the sketcth option (or whatever is called). And is extremely hard to sometimes remember which numbers can go where. Thus i thought this can show me how can i organise myself, or was some trick to it. Soo i guess this is of no use for me.

[@yatagarasu1495]

the website is shite then. maybe you can import your sodukus into another sudoku site or software which lets u solve it there with marks :)

[@julieannmyers8714]

When you remove all your pencil marks, I could just scream.

[@sunriselg]

Took me 51 minutes. I kinda wanted to give up after 40 but then I did ... something ... on 2 and then the rest fell into my lap.

[@pauldensley5459]

At 10:49 R2C1 you put a 5, why not a 5 in R2C2 or R3C2?

[@knickerlover1974]

Lost me on the y wing at 15:42 .

[@kpfwt7918]

Does anyone else hear these guys voice in their head when solving puzzles at home?

[@deucedeuce1572]

9:30 ... Why doesn't the puzzle just start like this instead of making you fill in all the easy stuff on a hard puzzle?

[@TheThomasites]

8s decision @6:00. Reasoning?

[@schancey55]

The 1 3 pair in the 4th column mean that the 8 in that 3x3 cannot go there, along with the 8 in the 6th column together force the 8 to be in either of those positions in the 5th column. Since the you know where the 8s are in the 5th column, you can use that with the 8 in the 6th column to determine where the 8s in the 4th column must be.

[@Samsaptaka]

The block had two squares that could be a 2 or a 6, allowing him to eliminate the 8 from one of those squares.

[@TheThomasites]

@@schancey55 yes. Just before i read this i saw it. Silly me. Should just look more.

[@shane8037]

I bumbled through it a lot quicker than that somehow, certainly without any bent triples

[@sasikala5949]

I am an Indian

[@stuartmcconnachie]

Who’s Simon and who’s Mark?

[@darkZarchon]

Simon is the one in this video. Mark is the other regular video creator.

[@jetset9561]

It's remarkable the way our brains work - the first thing I saw in this puzzle were the two 3's in columns 5 and 6 meaning you could pencil mark then into column 4 in the bottom block - yet it took you 5 mins to get there.

[@korayb5295]

8:45 you cant put the 7 like that... there is no any reason why it cant be 6, 2 or 7... all of these 3 number can be in that box...

[@henryginn7490]

205:1 like to dislike ratio. Don't see ratios that high very often

[@CrackingTheCryptic]

Thanks Henry... I just wish the KZbin algorithm would appreciate this fact and plug us a bit more :)

[@Legender-qq5xj]

You're a God (uppercase G because he earned it)

[@Mz-ym1ho]

Not correct logic, at 11:38. Three pairs (5-9) CAN produce 5 or 9 on wing. Actually, both numbers. You just had luck with 8.

[@atis1976]

rofl

[@rogerwilco2]

9:11 Finally.

[@thephysicistcuber175]

At the 20:00 mark (the point where you had to use the empty rectangle) there is an XYZ wing on 1 7 and 3 with pivot on I8 and other cells of the wing on G8 and I4 that allow elimination of a 3 from cell I9. That leads to a hidden single in row 9 (cell D9 forced to a 3) which might be another way of cracking the puzzle (haven't really checked that one).