No, you don't have to be a genius. Just train your mind into spotting candidates and use good techniques, then practice a lot. You will be there before you know it.
True, the title is misleading. He just solves the puzzle explaining strictly what he needs (and not explaining everything). From the title you would think the video discusses techniques, by name, you can use when facing different situations.
I love these “how to” videos. At my advanced age I’m just discovering Sudoku, and have finally graduated to the hard puzzles. For me it’s so easy to follow the logic while you do it, but still tricky to apply it on my own. I suppose that’s what makes it fun.
I've also just started to use the more 'advanced' techniques, and while their logic is very clear, they are very difficult to spot at the beginning! These videos help a lot though.
The logic is simple to understand and apply once shown but spotting it on your own without blindly trying 100s of options is really tricky. Is it intuition perhaps?
Th3e hardest thing is where to start. I simply rotate through the numbers. When completed, much is understood and all the obvious eliminated. I thin one could simply rotate through alll the numbers again, mechanically until it was completed, but that is boring. It is how I would program a computer.
Thanks for the comment Glen. I'm in the elderly group too and have just started the advanced. I usually end up with a lot of numbers in each box which makes my head spin. Hopefully it'll get easier to spot the not so obvious solutions.
I discovered it pretty much at the point the UA-cam algorithm made it explode (about a week ago) and have been mostly playing Sudoku and disregarding sleep. Case in point, it's almost 3AM and I'm so ready to start another puzzle before I go to sleep.
I’ve loved sudoku for a long time, since watching them solve puzzles and explaining their thinking and ways to solve I’ve found myself enjoy even more.... however I still find I get stuck on some where there are no more possible logic..... still lots to learn 😀📸😀
Great to see/hear Mark actually explain what he's doing in a way that makes clear sense for a change, rather than purely brute forcing a puzzle open leaving us to figure out what and how he's done it
@Fighter-jy1xo The bottom row would need a 3. If you placed the 3 in the 7th row, bottom row wouldn't get a 3 anymore. So 3 can only be places in the bottom row cell.
It's the key to solving this particular puzzle without having to resort to some much harder reasoning over chains of connected squares. Also, I sort of agree with the objection raised to using the "unique rectangle" rule but in the end if a sudoku rule is that the solution must be unique then it seems as legitimate to use it as the rule of no repeated digits. HOWEVER, I have come across puzzles where applying it leads to an error (because the pairs of digits gets fixed by some other means so there is actually no ambiguity in the answer, just the initial appearance of it).
I still don't quite get it, why wasn't it possible to put the 3 at the top corner cell instead of the bottom corner, if we'd put a 3 on the left centred cell?
@@yiutung4427 Yeah 3 can definitely be on the top of the cell. This means this guy already knows the solution and is just "solving" it for internet clout.
9:25 If you're confused here, he didn't do the best job explaining the deduction behind the 3-7 solve. So to clarify, if row 8 column 2 (r8c2) is a 3, that would mean that in box nine, 3 would be constricted to row 7. For now it doesn't matter we don't know the exact cell for box nine where the 3 would go, we can use that row 7 to eliminate the candidate 3 in box eight r7c6. If the 3 candidate in r7c6 is eliminated, that means 3 would go in r9c6. Now if r8c2 is a 7, then in box eight the 7 would go in r7c6, leaving the 3 to go in r9c6. Either way, r9c6 will always be a 3.
Thank you for this clarification! I literally paused at this section to try and figure out what this meant, and when I didn't get it, I checked comments. Appreciate the breakdown!
Your explanation is great, thankyou, very much appreciated. I was stuck at this very point. And even repeating his explanation many times, I still was'nt seeing the logic, and thinking it was really just his best intelligent guess. Then I read your comment, and realised what was going on. Yes, he should have explained it the way you have; I would'nt have got it otherwise. Thankyou! I notice you posted this 2yrs ago, a long time, for all I know you may have died since then, but if you hav'nt I say 'thankyou' to you. Best wishes from Alan, in England.
I love the meta strategy of uniqueness. I've never thought of that. I love situations where you can take outside, seemingly irrelevant information, and make informed decisions for strategy and tactics. I also really appreciate your attitude of pointing that out, and then showing how to solve the puzzle with its self contained logic. Beautiful video my friend.
I've been a sudoku fiend for two days now. Got stuck on an expert puzzle, came here. I don't think I want to do sudoku anymore. Seeing someone who actually knows what they're doing solve in an infinitely more elegant way has crushed me. Maybe another day if I can learn to enjoy it again. TLDR: I thought I was doing sudoku, but I was not.
As someone who used to work on Sudoku puzzles by filling out all possibilities in all cells and then eliminating them, I can attest to the fact that it takes ages to solve it that way. Also, the Snyder notation that Simon and Mark use have made some of the very hard sudoku puzzles very much accessible, along with speeding up my solves. As a side note, I can solve almost all of NYT hard sudoku puzzles using the elimination idea very quickly. NYT puzzles come with the auto completion check mark, that I use very often and I find that almost all their puzzles can be solved efficiently by simply eliminating candidates and finding hidden doubles and triples.
I have been doing sudoku for last 4-5 years, I have started loving it , my day is incomplete without daily dose of sudoku in morning newspaper, but I used to get stuck with very hard sudoku sometimes , today I have learned the technique , all thanks to you sir.
So happy to see Mark solve a puzzle like this. Great for newer members of the channel, and super great for me who struggles with higher logic issues with tougher puzzles. I aspire to get to the point where I scan a grid, and finds a cell jumping out to me that needs attention.
The handicap of not using uniqueness just makes your solves more interesting to watch. I do appreciate you pointing out when it is available as it is a useful technique that I would not hesitate to use myself. I also find the topic of uniqueness interesting in relation to the star battles and other puzzles that almost require you to use it.
A couple of weeks ago I ran into one of their puzzles that had 2 solutions just like the 7-9 issue raised here. Except that puzzle had a knights move rule, which ruled out of the 2 possible solutions. I didn't notice the knights move, and was ready to triumphantly post how there was a 2nd solution. I was actually in process of typing my comment, when I noticed the knights rule. Damn...I should know better than to ever question these guys in sudoku!
Bill Young They would not be solving a puzzle that has two distinct solutions, because the people who make them test the puzzles before publishing them, since many of them are world-class puzzle crafters aided by software, they are not your average Joe from the street, most of the time. And if the puzzle is completely machine-generated, then it only validates further the uniqueness. I think it is silly to argue against the use of uniqueness.
If the goal were just to present the finished solution, I'm sure they'd be all for using uniqueness, as it's just another tool to get to the answer as fast as possible. However, the point of the channel is to not only get the answer right, but to show how they arrive at the answer. I guess it feels like they're not using the intended solve path if they skip ahead using uniqueness, though to most of us it's as much a rule as saying each box has to contain all of the digits 1-9. It's ok to use uniqueness to help get unstuck, but you'll never see Mark or Simon resort to that on purpose, and honestly, at their skill level, they shouldn't have to.
@@federicocavagnero5226 I can picture myself returning to this puzzle every now and then until I'm finally capable of spotting this type of trick. Good luck with your upcoming puzzle attempts! :)
I put this puzzle to www.sudoku-solutions.com/ and the logic solving by the machine got stuck before figuring out the 3 in the 7 pair in box 8. I believe this proves that this puzzle is not solvable by using only the popular techniques (including the X-Wing, XYZWing stuff.)
@@chongchonghe3748 Thanks for the link! I'll use it with some sudoku puzzles that are giving me headaches. I used another website (don't remember the link) to try one puzzle and the logic chain broke after several and switched strategies to endlessly removing candidates from every cell until something fell apart.
But I don't think the advice is entirely applicable. The problem is that this channel is an educational channel to an extent, at least they have proclaimed themselves to be. But if one of them is not being careful or attentive about how they go about solving puzzles, then not only is it not entertaining, but it actually takes away from whatever lesson they are trying to cover. As such, criticizing the channel due to the fact that they often miss very simply stuff that ends up leading to cluttered notation and less intuitive solution paths is completely legitimate. It's not hate, and it's not being triggered, it's a legitimate issue the channel could address and make the experience better for everyone. I still love the channel. But it's perfectly fine to state the flaws.
@@angelmendez-rivera351 I agree with what you say for this particular video, because it mentions "techniques" in its title. Therefore I understand that some viewers may expect the puzzle to be solved the best way possible. When the video title doesn't explicitly says that the video is a tutorial, then I think viewers should not expect it to be a tutorial. Most of the time, the videos are just about 2 guys showing themselves doing what they like, which is solving puzzles. It happens that in those videos, they also give so adivce if they spot a particular solving path, but once again, if the video title doesn't suggest that it's a tutorial, then I would not expect it to be. I guess some people would prefer them to solve the puzzles beforehand, and just make a walkthrough of the solving in the video. But that would remove those magic moments when one of them spots a beautiful trick in the puzzle, and is amazed by it.
Using uniqueness is a perfectly valid way of _solving_ a puzzle, but (like bifurcation), it solves at the expense of appreciating the beauty of the logic involved. As you guys highlight the beauty of these puzzles rather than merely demonstrate the quickest way to solve it, I really appreciate that you refrain from using uniqueness.
But uniqueness still doesnt rely on using only logical deduction. The “logic” works like this: “Oh, this puzzle needs to have only 1 solution because all the other puzzles do”, which is not deduction but induction. It’s going to be right every single time... right until its not because there is no acutal logical basis for its validity! I just cant help but feel that sure, its a strategy in speedsolving but just doesn’t have a home in casual solving.
It depends on whether you regard "there is only one solution" as just a working assumption, or as one of the rules that defines a valid puzzle. If you see it as a rule, then it's a legitimate basis for logic just like any other rule. Personally I'd say that it IS a rule - a puzzle with more than one solution isn't just non-standard, it's defective - but I agree that it's usually a less satisfying route to the solution.
@@MTTR01 disagree but my motivation to comment is more that what does and doesn't have a home in casual solving shouldn't be regulated? Being entertained? You're doing it right.
I, obviously, can't speak for everyone, but when I catch something you or Simon miss I am most certainly not triggered. I am elated, because it is such a rare occurrence.
Triggered is such a bad description though. More like you just want to put that digit in so badly that it makes your fingers twitch. "Triggered" implies we have a negative emotional response to it. Like panic, flashbacks, fear or panic. Which is far from the truth. It's just that we can't convey the message which makes us so eager to comment about it. Sure most of their viewers are probably happy they spot something they didn't spot. But when it's so obvious that even a first time solver should be able to see it, it does get a bit too much to bare just watching it being missed over and over again. Still not triggering... If that was the case then people seeing someone walking straight out onto a busy road in front of traffic would be triggered rather than concerned for their safety. It's that human response of wanting to be helpful that can't be satisfied that's so uncomfortable. I feel elated if i spot something advanced and not immediately obvious before Mark and Simon though. That doesn't instill the same unease.
I'm 44 seconds in to the video. Yes! Exactly! Thank you! I found you somewhere early in the shut down. No idea how. Became addicted. Went full Patreon. The next day tried your methods. Rethought patronage. Let's do this!!
This was my first and only hard solve and I did it in about 1hr and 20ish minutes. Before this I had only attempted hard and had only solved a couple of easy ones. By watching this channel for about a week or two, now I felt like I was able to do some logic which I was not able to do before no matter how hard I tried. It was not obviously at the level shown in the vid. But I am still happy to be able to solve this on my own.
Several months ago, you presenters as well as some high-profile setters in the comments turned me around to the idea of using uniqueness as a logical method. We've recently encountered a sudoku where that's not the case, but that was a very special case, I'm sure you know what I'm talking about. So I was actually perfectly prepared for you to use uniqueness and accept it. But THEN, you turn it around again! And for some reason I'm super happy about it!
I just started getting reinterested in sudoku after remembering how my math teacher would make us do a puzzle to start off class lol! It's super helpful to hear you think out loud, I didn't know about the Snyder notation, but will definitely start using it!! Great video, much love!!!
Super helpful! I appreciate more of an explanation of what Snyder notation is. It's been referenced but I've had a hard time nailing down what specifically it is, so that was really helpful. Also appreciated the tip about using pairs to deduce other parts of the grid. Always learn something new on this channel. =)
2:40 I use a dot or point system where I poke a dot at the position the number would have in a 9 digit grid per box. So if a cell has a potential of being a 6 or an 8, I would place a dot on the horizontal center axis to the right to represent a possible 6 and another on the vertical center axis at the bottom for the 8 position. In many cases I get to the point where deduction cannot eliminate options to a single option and therefore guessing between possible option 'A' or 'B' (eg 6 or 8) is a wild guess that can give progress until the knot occurs and then there's no going back.
Fascinating. Thanks illustrating your either/or technique for bivalue cells Mark. I find them to be quite powerful. I can always get out of a jam by testing them. I tend to use them at endgame when all else fails, but now I realize I can use them mid-game as well.
Agree. But note he pencilled in 7s to box 8 but that's only supported if the 78 in box 3 is an 7. Having watched these for years I know it won't matter in the end but I love the (super rare) times I get to say, "Aha!".
Brilliant! Ignore the critics, your method is superior. @12:45 Your logic is valid logic, can't have 2 solutions, so 9 doesn't belong in those 2 squares, has to go in r1c3. Again, you're educating all of us, ignore the critics.
I agree that it's harder to spot things when you fill every cell with all possible values it could have. So I'm limiting my pencil marks to 2 (max 3). But that chain inference placing 3 at R8C6 that started at R7C2 with 3and7 as possible values is mind-blowing. I will never spot that ever.
Hi! Do you have any tips for training yourself to become more aware of those chain reaction patterns that limited the three and the seven in the bottom middle square? Thanks! I love your videos!
Isabel Erwin practice by doing tons of sudokus. You’ll start to recognize patterns like that by yourself very quickly. You just have to do a few sudokus
Notating pairs as much as possible helps a lot, and, when you get stuck, checking what happens with each option in some pair. Pairs that are not "pairs of pairs within the same box" tend to be more likely to be involved in interesting chains, per my experience (but that might just be some bias that I have because of the way I solve sudokus, the things I spot more easily or the things I miss more often, etc., so "you mileage might vary"). Sometimes you can spot an interesting chain when a cell "is seen" by two different pairs, or is surrounded by other cells that are very restricted. Somehow, over time you start "feeling" that some cells are "in the middle of something" and some cells are just "quiet", uninteresting by now... I'm not sure there are exact criteria for that, but for sure that helps a lot with the solving speed (note how Mark always says "where should we look now..." when he is stuck: he is not looking for something specific, he is looking for "interesting" cells or patterns ;-) )
@@wickedsamurai3323 I found a sudoku app called HoDoKu. It might not be the best application, but you can ask it to make a puzzle that specifically has the strategy you want to practice. It is great, and it is free.
Took me 18:43 to solve; definitely one of the harder classics on the channel. Really enjoyed solving it. As a side note, I've never really agreed with people who think uniqueness rectangles are a non-satisfying, or even, invalid way of solving classic sudokus. Mainly for two reasons, 1. As far as I know (I'll admit that I haven't researched this extensively myself, but I've read it on multiple sites/forums discussing sudoku strategies), while the strategy itself is based on the assumption that 'there is only one solution', even if you're solving a 'defective' sudoku with multiple solutions, using uniqueness will NOT lead to a wrong solution at the end of the road, just one of the multiple correct solutions. In other words, as long as the puzzle you are solving is a classic sudoku that has valid solution(s), there is actually no instance when applying uniqueness will lead to a contradiction, so it's ok to use it. The only case where you should actively refrain from using uniquness strategies is when you're solving variant sudokus; in this case, since classic rules aren't the only rules enforcing a unique solution, applying uniqueness can, and usually will, lead you to the wrong solution. 2. While Unique Rectangles are introduced as fairly advanced techniques on a lot of sudoku tutorials/videos, I actually find that 'classic' Unique Rectangles (such as the example in this particular video) are fairly common and very easy to spot. They can often help out in tough spots, helping to avoid long chains. Again, I'll admit that the same can't be said for 'hidden' or 'extended' Unique Rectangles, but if you're solving a classic that requires these strategies to progress, it's probably an extremely difficult one anyway. Obviously, I understand that everyone has different reasons and styles/preferences when it comes to puzzles, and if using uniqueness to solve sudokus just doesn't 'feel right' to someone, that's perfectly ok. However I personally love URs and use them to solve classic sudokus at every opportunity, and encourage others to do the same.
I'm quite happy to use uniqueness if I can't spot any other way but I get the fact that to do so is not using the internal logic of the puzzle, instead using the external logic and assumption of it being a good sudoku. It's a bit like one of those "if you have $10, what can you buy to satisfy all" puzzles and turning round and assuming a good answer is "Feck it, I'll use my Credit Card"
Charles Fox Yes, it is ultimately external logic, but I do not see why this is an issue. It's such a purist, naive, and immature mentality to not consider external logic as part of the logic. Besides, some puzzles are explicitly designed such that the only way to arrive at the solution is precisely to assume that it has a unique solution. And if you don't ever use this rule, you'd have to simply literally guess the solution, or more formally use bifurcation, which many people also treat as not an act of using logic - though by definition, bifurcation is internal logic via the principle of reductio ad absurdum.
Charles Fox That's a false equivalence fallacy. Not only are there are no Rubik's cube puzzles designed with the intent that external logic is required, but uniqueness is not even present. Sudoku and Rubik's cube is fundamentally present, in that a restricted binary deductive logic is required in sudoku, not in a Rubik's cube, which is purely combinatorial. So, your argument is invalid. I know you're trying to demonstrate the absurdity of my statement by making a comparison with a statement you think is equally as absurd that logically follows from my statement, but unfortunately, it doesn't work: they're not logically similar, and you can't prove my statement isn't absurd because it genuinely isn't. You can't prove a false statement if you're using the correct set of deductive rules. You can choose to prohibit the use of uniqueness as a valid deduction in your puzzle solving habits. Sure. But as far as the actual theory goes, there is legitimately no good reason to argue against it. Comparing it to a Rubik's cube isn't going to help the "purist's" case, if it can even be called that.
Agree with everything you've written so far, and I just want to add that uniqueness opens up some beautiful logical strategies that can get just as complex as the "normal" logic we use in these puzzles! There are so many variations of uniqueness patterns, going way beyond simple URs. I feel that the classic sudoku experience would be less "rich" without it (and even some variants too, though one has to be very careful using them in variants, it must be "sealed off" and not overlap with the variant logic to work ;) ).
Because the two cells are the only two cells in that column where 3 and 5 can exist. If you put an 8 there you wouldn't have anywhere to place both 3 and 5.
to follow up, it also is why they use the center fill-in when they are putting the only possible candidates for a box. when there still are unknown possibilties they place the numbers in the corners. it is a nice a elegant way to keep track of what you have figured out....or not figured out. ...so if you see 2 boxes with a 35 pair they redo those as center then remove the corners. it then eliminates a possibility for the other numbers they had in the corner which then often leads to knowing where that other digit goes. finding pairs is key to solving these puzzles. on very difficult possible you often will have to find a very hidden triple! hehe
10:12, completely missed that, ended up making a chain completing the puzzle of that cell was a three. Worked out every candidate trying to find an error and it turned out to be the final solution. Took me 40 minutes.
At 9:22, assume it's a 3 for the 37 pair, why then 3 goes to the bottom on the 9th row when 3 could also go to the 7th row? I think this might be a flawed logic? Please comment.
No it is not a mistake. If row 8, column 2 cross is assumed to be 3, that means for row 9, where can the 3 go in row 9, it can only go into row 9, column 6, bottom. So look at row 9, not at box 8.
@@wynandstockel9592you are assuming 3 in box 7 at the place of 37 because 3 is not marked anywhere elese in box 7, if you see 3 can also come in box 7 bottom 3rd column next to 2 , with this the logic of 3 can only come on 9th row failed
9:46 inference is only a guess at this point since 7 can also be top right cell of the bottom right box. Therefore guessing is part of the ?techniques'?
Thanks, I'm looking for more clues. The chief take home from this is NOT to put all possibilities in each box, but to work with the ones that can be only 2 choices, or possibly 3.
I could not understand how the 3 came in the 8 block... down side of “2”..logic I missed .. 3 can be on above of “2” also.can please explain in detail..??
Do you mean that 3-8 in the upper right corner of box 9? He did explain it. There are only two possibilities for that square because box 9 already has 179 and the column has 246 and the row has 5. The only numbers left are 3 and 8.
I first watched... then I did the sudoku and watch again... makes so much sense. Thanks .. I'm sure I have become a better player . Yes, u r sure not bad for the country 😂😂😂❤
9:19... someone please explain. Is that some kind of rule? Shouldn't all the 3's an 7's have been pencil marked first in the bottom left box before concluding where the 3 goes in bottom middle box?
At about 11:16 you made an assumption that a 3 can only go in row 8 column 2. It seems to me that t is also possible at this point for it to go into row 7 column 3. Ultimately, it turned out that you were correct but I cannot see how that assumption is provable with the information available at the time. Am I missing something or were you lucky?
I think you're on to something -- I can't see how 3 at this point has been eliminated from r7c8 and r8c9 that an 8 on r3c9 would guarantee the 3 would appear in r7c9, and thus in r8c2. I mean, being him, I can very well imagine he worked it out in his head some other way, but he has not explained it XD
The 3/7 deduction was because that cell had to be a 3 or 7 due to the rest of the column and row, so once the 7 was eliminated as an option because of it's position in box 8 that cell had to be the 3 regardless of where it could have possibly gone in that box.
Can someone explain how he got a possible 3 in box 7 (column 2, row 8). As far as I can tell that 3 can also be possible in all the other empty cells except column 1, row 7. He puts it in around 8:17.
He did actually explain it. It can only be 3 or 7 because all the other numbers are taken by the column and row it is joined to. The column it is in already has 124568 and it can't be 9 because it's already in the row so here are only two possibilities left.
Uniqueness is a genuine strategy. Instead of looking for crazy patterns that only a genius could see, one can simply use uniqueness and simply solve it. Insisting on the idea that using uniqueness is not pure is BS, because the other harder logic that he finds would simply not work if sudoku was not unique.
I'd agree on that. The uniqueness technique is using one simple assumption: 'This puzzle has a unique solution'. The uniqueness trait is the _default,_ and is _expected_ by the vast majority of people who solve sudokus. Sudokus without a unique solution are considered by many as invalid, and I'd reckon that most sudoku setters are taking pride in the fact that their sudokus are unique. Also, many other more complicated techniques are relying on uniqueness, like every 'unique rectangle' technique, for example. In my opinion, this 'uniqueness is not pure' idea is completely arbitrary nonsense.
Good puzzle, nicely solved. I can understand the objections to using uniqueness rather than logic to place a number but I say don't hit on the solver for using uniqueness, hit on the setter for allowing that situation to occur in the first place.
Unless the puzzle has a very narrow solve path, I’m not sure how well a setter can avoid all possible scenarios where the solver might notice uniqueness. See the Kurt Schneider recent video where he sees it like half a dozen times in 30 seconds in a random star battle puzzle.
As long as the puzzle can be logically solved without relying on the uniqueness logic, it doesn't much matter how many times you spot the option to use it. It's just that, if a setter creates a puzzle that NEEDS a Unique Rectangle trick to solve it, then it's not a valid Sudoku, as it has three solutions. At LEAST three :P A fun joke puzzle might be one where uniqueness problems are everywhere, and while it has like 20+ valid solutions, if you apply the Unique Rectangle logic, you can force it down to just one. Might be interesting for the channel, as a way to make use of uniqueness for a day without having to forego the beauty of a puzzle. The beauty is in how ridiculous a concept it is :P
I have found that if you have pencilled in say a 1 in the top left square in column 1 and 2 then it follows that in the two sqares below the 1 will be either in Rowe's 1and 3 or 2 and 3. This has helped me many times.
Thanks for this very informative video! Managed to solve this one by myself first thanks to watching other videos on your channel. So it was interesting to then watch your solve and compare my results.
I very much enjoyed this video! Im happy that you bring the explanations down to my level and take your time explaining some of the concepts (particularly the uniqueness bit)
I enjoy your solving. question though at 6:30, could you restrict the center cell to 8 and 9 ? Does not make a difference later on for this sudoku, but I found that is sometimes usefull on other sudokus to restrict 2 out of 3 triplets digits in only 2 cells. this way triplets in line 5 and the center box would be perfect (set of 2 numbers rotating in all 3 cells)
YAY! I actually solved this by myself. Been ages since I last did a sudoku puzzle though the alternative one's really confuse me with all their different rules which I don't even understand most of them but still watching them being solved 'till something hopefully clicks in my brain so I can understand it lol Also loved your comments from 5:10 onwards. I've only just found your channel which is great for me trying to solve them before watching your video to see any difference.xx
9:26 See? This is what it all comes down to sometimes. Making assumptions. In this case it's the beautiful simple logic. You have two choices A & B if they both lead to the same result then they're either both true or both false. I love this!
It's a FT (forcing test) with the 37 in RC82. If RC82 is a 3, the 3 in the bottom right hand block goes into to R7, which punts the 3 in C6 to R9. If RC82 is a 7, a 7 would go into RC76, again punting the 3 in C6 to R9. Either way the 3 in C6 is in R9.
I wondered a couple of times why you guys avoid the unique solution, so thank you very much for explaining! :) For myself I just figured out that it is harder to spot, when additional rules apply. For example a knights move could solve the double 79 pair and it would have been a mistake to say it is a 4 because of uniqueness. Thank you again for the explanation :)
@@nacho__063 He can't put a 7 in R7C6 because once he puts a 3 in R8C2 the only place for a 7 in Block 7 is the same row you want to put a 7 in Block 8.
Alternate inference chain... On 9:30?? Some tricky techniques i need to learn. Xy chain on r3c9?. Solved it in 22:33. I need to practice more on those diabolical techniques
I have just discovered sudoku. I'm hoping with fingers crossed that a fair portion of seeing where the numbers can go comes with repetition & practice because watching this fella makes me want to chuck it in before I start! How tf can he see these patterns so quickly - i.e where he saw to put the 3's straight away. Jings
I got severely interrupted twice while doing this, so I don't have an exact time,, but I think I spent half an hour actively solving it. Didn't use the inference chain that Mark showed, or the naked single in box six, I managed to find a different logic path that still led to the same conclusion. Which means it's a good puzzle, since it can be solved for the same unique solution using different methods of attack.
Possibly a dumb question, but at 8mins 46secs you say the 9 blocks out C2 R8 (understood) and that 9s must therefor be in C2 R1 or R2. But why can't it be in C1 R2 or C3 (R1 or R3)? Thanks. Nicely done!
That was a masterful lesson in how to make the difficult look easy, thank you. Is the technique you used two times equivalent to the XY wing technique? Either way it still needs much practice to spot these possibilities.
I love CTC's solving style so much I made a video on my channel where I analyze each step of this solve. In addition to explaining the strategies used, I add a couple of pause the video moments and some alternate solving paths, similar to how someone analyzes a chess game.
Ok duh! It is because once the 3 is placed in box 7 it means that the 7s in box 7 have to be in the top row! Which rules out the 7 in the top row of box 8!
I think it is a pattern (e.g. 25, 25 in two boxes middle) you have to select in your brain and then test the pattern selected if it conflicts you change to another then you get the answer quickly.
Today I learned that the best technique for hard sudokus is being a genius.
@Brock Savage 😃😃😃😂
Now I feel hopeless
No, you don't have to be a genius. Just train your mind into spotting candidates and use good techniques, then practice a lot. You will be there before you know it.
True, the title is misleading. He just solves the puzzle explaining strictly what he needs (and not explaining everything). From the title you would think the video discusses techniques, by name, you can use when facing different situations.
🤣
I love these “how to” videos. At my advanced age I’m just discovering Sudoku, and have finally graduated to the hard puzzles. For me it’s so easy to follow the logic while you do it, but still tricky to apply it on my own. I suppose that’s what makes it fun.
I've also just started to use the more 'advanced' techniques, and while their logic is very clear, they are very difficult to spot at the beginning! These videos help a lot though.
The logic is simple to understand and apply once shown but spotting it on your own without blindly trying 100s of options is really tricky. Is it intuition perhaps?
Th3e hardest thing is where to start. I simply rotate through the numbers. When completed, much is understood and all the obvious eliminated. I thin one could simply rotate through alll the numbers again, mechanically until it was completed, but that is boring. It is how I would program a computer.
Makes sense sudoku is np complete. Hard to solve but easy to verify a solution.
Thanks for the comment Glen. I'm in the elderly group too and have just started the advanced. I usually end up with a lot of numbers in each box which makes my head spin. Hopefully it'll get easier to spot the not so obvious solutions.
Anyone else bingeing this channel? I’m not the best at sudoku but watching this channel is just relaxing
Watching daily!
I discovered it pretty much at the point the UA-cam algorithm made it explode (about a week ago) and have been mostly playing Sudoku and disregarding sleep. Case in point, it's almost 3AM and I'm so ready to start another puzzle before I go to sleep.
@@alf10087 We all are!
I’ve loved sudoku for a long time, since watching them solve puzzles and explaining their thinking and ways to solve I’ve found myself enjoy even more.... however I still find I get stuck on some where there are no more possible logic..... still lots to learn 😀📸😀
Yes! Wouldn't have believed it until I watched one myself and now I am deep down the rabbit hole.
Great to see/hear Mark actually explain what he's doing in a way that makes clear sense for a change, rather than purely brute forcing a puzzle open leaving us to figure out what and how he's done it
9:30 analysis for placement of the three in the bottom row. Brilliant. I need to add this to my very small bag of tricks.
This was the defining moment of this puzzle for sure
Anybody explain 9:30 min step how that cell 3 is possible I couldn't understood.. explain pls
@Fighter-jy1xo The bottom row would need a 3. If you placed the 3 in the 7th row, bottom row wouldn't get a 3 anymore. So 3 can only be places in the bottom row cell.
@@snigdhasharma8666 but that 3 for bottom row can also add in 3rd bottom column next to 2
Look up y-wing technique
The logic starting from 09:20 to 10:00 has been what I have been missing for quicker very hard sudoku times. Thanks a lot!
same! i never used that technique before but when he explained it seems so obvious lol. will implement it in my next puzzle for sure
It's the key to solving this particular puzzle without having to resort to some much harder reasoning over chains of connected squares. Also, I sort of agree with the objection raised to using the "unique rectangle" rule but in the end if a sudoku rule is that the solution must be unique then it seems as legitimate to use it as the rule of no repeated digits. HOWEVER, I have come across puzzles where applying it leads to an error (because the pairs of digits gets fixed by some other means so there is actually no ambiguity in the answer, just the initial appearance of it).
I still don't quite get it, why wasn't it possible to put the 3 at the top corner cell instead of the bottom corner, if we'd put a 3 on the left centred cell?
@@yiutung4427 Exactly what I was thinking. The answer is further down the comments by tiotito31
@@yiutung4427 Yeah 3 can definitely be on the top of the cell. This means this guy already knows the solution and is just "solving" it for internet clout.
what.... I got the exact same grid like him at 9:19 and I got super hard stuck, then he did this with 3 and 7 and I'm mindblown...
9:25 If you're confused here, he didn't do the best job explaining the deduction behind the 3-7 solve. So to clarify, if row 8 column 2 (r8c2) is a 3, that would mean that in box nine, 3 would be constricted to row 7. For now it doesn't matter we don't know the exact cell for box nine where the 3 would go, we can use that row 7 to eliminate the candidate 3 in box eight r7c6. If the 3 candidate in r7c6 is eliminated, that means 3 would go in r9c6. Now if r8c2 is a 7, then in box eight the 7 would go in r7c6, leaving the 3 to go in r9c6. Either way, r9c6 will always be a 3.
Thanks! That helped a lot
Thank you for this clarification! I literally paused at this section to try and figure out what this meant, and when I didn't get it, I checked comments. Appreciate the breakdown!
Good work
That’s exactly where I stopped, because it didn’t make sense.
Your explanation is great, thankyou, very much appreciated. I was stuck at this very point. And even repeating his explanation many times, I still was'nt seeing the logic, and thinking it was really just his best intelligent guess. Then I read your comment, and realised what was going on. Yes, he should have explained it the way you have; I would'nt have got it otherwise. Thankyou! I notice you posted this 2yrs ago, a long time, for all I know you may have died since then, but if you hav'nt I say 'thankyou' to you. Best wishes from Alan, in England.
I love the meta strategy of uniqueness. I've never thought of that. I love situations where you can take outside, seemingly irrelevant information, and make informed decisions for strategy and tactics. I also really appreciate your attitude of pointing that out, and then showing how to solve the puzzle with its self contained logic. Beautiful video my friend.
I've been a sudoku fiend for two days now. Got stuck on an expert puzzle, came here. I don't think I want to do sudoku anymore. Seeing someone who actually knows what they're doing solve in an infinitely more elegant way has crushed me. Maybe another day if I can learn to enjoy it again. TLDR: I thought I was doing sudoku, but I was not.
As someone who used to work on Sudoku puzzles by filling out all possibilities in all cells and then eliminating them, I can attest to the fact that it takes ages to solve it that way. Also, the Snyder notation that Simon and Mark use have made some of the very hard sudoku puzzles very much accessible, along with speeding up my solves.
As a side note, I can solve almost all of NYT hard sudoku puzzles using the elimination idea very quickly. NYT puzzles come with the auto completion check mark, that I use very often and I find that almost all their puzzles can be solved efficiently by simply eliminating candidates and finding hidden doubles and triples.
Very useful techniques showed in this video! See 8:11, 9:18, and 12:31!
I have been doing sudoku for last 4-5 years, I have started loving it , my day is incomplete without daily dose of sudoku in morning newspaper, but I used to get stuck with very hard sudoku sometimes , today I have learned the technique , all thanks to you sir.
So happy to see Mark solve a puzzle like this. Great for newer members of the channel, and super great for me who struggles with higher logic issues with tougher puzzles. I aspire to get to the point where I scan a grid, and finds a cell jumping out to me that needs attention.
The handicap of not using uniqueness just makes your solves more interesting to watch. I do appreciate you pointing out when it is available as it is a useful technique that I would not hesitate to use myself. I also find the topic of uniqueness interesting in relation to the star battles and other puzzles that almost require you to use it.
A couple of weeks ago I ran into one of their puzzles that had 2 solutions just like the 7-9 issue raised here. Except that puzzle had a knights move rule, which ruled out of the 2 possible solutions. I didn't notice the knights move, and was ready to triumphantly post how there was a 2nd solution. I was actually in process of typing my comment, when I noticed the knights rule. Damn...I should know better than to ever question these guys in sudoku!
Bill Young They would not be solving a puzzle that has two distinct solutions, because the people who make them test the puzzles before publishing them, since many of them are world-class puzzle crafters aided by software, they are not your average Joe from the street, most of the time. And if the puzzle is completely machine-generated, then it only validates further the uniqueness. I think it is silly to argue against the use of uniqueness.
Thanks, I learn something
If the goal were just to present the finished solution, I'm sure they'd be all for using uniqueness, as it's just another tool to get to the answer as fast as possible. However, the point of the channel is to not only get the answer right, but to show how they arrive at the answer. I guess it feels like they're not using the intended solve path if they skip ahead using uniqueness, though to most of us it's as much a rule as saying each box has to contain all of the digits 1-9. It's ok to use uniqueness to help get unstuck, but you'll never see Mark or Simon resort to that on purpose, and honestly, at their skill level, they shouldn't have to.
I hate showing my work :( but at least these folks don’t mind
This might be one of the more useful videos this channel has ever produced. Well done Mark!
The trick with the 7 in box 8 was my weakness. I couldn't spot that.
same for me
@@federicocavagnero5226 I can picture myself returning to this puzzle every now and then until I'm finally capable of spotting this type of trick. Good luck with your upcoming puzzle attempts! :)
I put this puzzle to www.sudoku-solutions.com/ and the logic solving by the machine got stuck before figuring out the 3 in the 7 pair in box 8. I believe this proves that this puzzle is not solvable by using only the popular techniques (including the X-Wing, XYZWing stuff.)
@@chongchonghe3748 Thanks for the link! I'll use it with some sudoku puzzles that are giving me headaches. I used another website (don't remember the link) to try one puzzle and the logic chain broke after several and switched strategies to endlessly removing candidates from every cell until something fell apart.
I've been doing sudoku for 15 years and have never even considered that as a strategy. To have seen it instantly is masterful.
Fantastic advice: "don't get triggered if you can possibly avoid it"
I need to work on this a lot... always getting frustrated with Simon :x
And this is why we need trigger warnings on the internet so people can avoid the things that do that :)
But I don't think the advice is entirely applicable. The problem is that this channel is an educational channel to an extent, at least they have proclaimed themselves to be. But if one of them is not being careful or attentive about how they go about solving puzzles, then not only is it not entertaining, but it actually takes away from whatever lesson they are trying to cover. As such, criticizing the channel due to the fact that they often miss very simply stuff that ends up leading to cluttered notation and less intuitive solution paths is completely legitimate. It's not hate, and it's not being triggered, it's a legitimate issue the channel could address and make the experience better for everyone. I still love the channel. But it's perfectly fine to state the flaws.
"Don't get triggered if you can possibly avoid it." --Seneca
@@angelmendez-rivera351 I agree with what you say for this particular video, because it mentions "techniques" in its title. Therefore I understand that some viewers may expect the puzzle to be solved the best way possible.
When the video title doesn't explicitly says that the video is a tutorial, then I think viewers should not expect it to be a tutorial.
Most of the time, the videos are just about 2 guys showing themselves doing what they like, which is solving puzzles. It happens that in those videos, they also give so adivce if they spot a particular solving path, but once again, if the video title doesn't suggest that it's a tutorial, then I would not expect it to be.
I guess some people would prefer them to solve the puzzles beforehand, and just make a walkthrough of the solving in the video. But that would remove those magic moments when one of them spots a beautiful trick in the puzzle, and is amazed by it.
Using uniqueness is a perfectly valid way of _solving_ a puzzle, but (like bifurcation), it solves at the expense of appreciating the beauty of the logic involved. As you guys highlight the beauty of these puzzles rather than merely demonstrate the quickest way to solve it, I really appreciate that you refrain from using uniqueness.
But uniqueness still doesnt rely on using only logical deduction. The “logic” works like this: “Oh, this puzzle needs to have only 1 solution because all the other puzzles do”, which is not deduction but induction. It’s going to be right every single time... right until its not because there is no acutal logical basis for its validity! I just cant help but feel that sure, its a strategy in speedsolving but just doesn’t have a home in casual solving.
It depends on whether you regard "there is only one solution" as just a working assumption, or as one of the rules that defines a valid puzzle. If you see it as a rule, then it's a legitimate basis for logic just like any other rule. Personally I'd say that it IS a rule - a puzzle with more than one solution isn't just non-standard, it's defective - but I agree that it's usually a less satisfying route to the solution.
Paul O'Brien Yeah, didn’t think of it that way... I guess that i would just not assume it as an axiom. That is a good argument
@@MTTR01 disagree but my motivation to comment is more that what does and doesn't have a home in casual solving shouldn't be regulated? Being entertained? You're doing it right.
Am I the only one who finds the uniqueness logic fascinating?
Thank you for taking the time to explain how the cells tell off on each other.
I, obviously, can't speak for everyone, but when I catch something you or Simon miss I am most certainly not triggered. I am elated, because it is such a rare occurrence.
Triggered is such a bad description though. More like you just want to put that digit in so badly that it makes your fingers twitch.
"Triggered" implies we have a negative emotional response to it. Like panic, flashbacks, fear or panic. Which is far from the truth. It's just that we can't convey the message which makes us so eager to comment about it.
Sure most of their viewers are probably happy they spot something they didn't spot. But when it's so obvious that even a first time solver should be able to see it, it does get a bit too much to bare just watching it being missed over and over again.
Still not triggering... If that was the case then people seeing someone walking straight out onto a busy road in front of traffic would be triggered rather than concerned for their safety.
It's that human response of wanting to be helpful that can't be satisfied that's so uncomfortable.
I feel elated if i spot something advanced and not immediately obvious before Mark and Simon though. That doesn't instill the same unease.
me: try clicking & typing on the screen
It's usually because they are busy solving something more important
I'm 44 seconds in to the video. Yes! Exactly! Thank you! I found you somewhere early in the shut down. No idea how. Became addicted. Went full Patreon. The next day tried your methods. Rethought patronage. Let's do this!!
Could someone explain at 9:20 why box 8 square 3 couldnt be a 3? That part is confusing me. Thanks!
This was my first and only hard solve and I did it in about 1hr and 20ish minutes. Before this I had only attempted hard and had only solved a couple of easy ones. By watching this channel for about a week or two, now I felt like I was able to do some logic which I was not able to do before no matter how hard I tried. It was not obviously at the level shown in the vid. But I am still happy to be able to solve this on my own.
Several months ago, you presenters as well as some high-profile setters in the comments turned me around to the idea of using uniqueness as a logical method. We've recently encountered a sudoku where that's not the case, but that was a very special case, I'm sure you know what I'm talking about. So I was actually perfectly prepared for you to use uniqueness and accept it. But THEN, you turn it around again! And for some reason I'm super happy about it!
I just started getting reinterested in sudoku after remembering how my math teacher would make us do a puzzle to start off class lol! It's super helpful to hear you think out loud, I didn't know about the Snyder notation, but will definitely start using it!! Great video, much love!!!
Super helpful! I appreciate more of an explanation of what Snyder notation is. It's been referenced but I've had a hard time nailing down what specifically it is, so that was really helpful. Also appreciated the tip about using pairs to deduce other parts of the grid. Always learn something new on this channel. =)
2:40 I use a dot or point system where I poke a dot at the position the number would have in a 9 digit grid per box. So if a cell has a potential of being a 6 or an 8, I would place a dot on the horizontal center axis to the right to represent a possible 6 and another on the vertical center axis at the bottom for the 8 position.
In many cases I get to the point where deduction cannot eliminate options to a single option and therefore guessing between possible option 'A' or 'B' (eg 6 or 8) is a wild guess that can give progress until the knot occurs and then there's no going back.
Fascinating. Thanks illustrating your either/or technique for bivalue cells Mark. I find them to be quite powerful. I can always get out of a jam by testing them. I tend to use them at endgame when all else fails, but now I realize I can use them mid-game as well.
Thank you, your cracking of the 3 and 7, and then the 7 8 and 9 was excellent. Enjoy your even pace reasoning and clear illustration.
Agree. But note he pencilled in 7s to box 8 but that's only supported if the 78 in box 3 is an 7. Having watched these for years I know it won't matter in the end but I love the (super rare) times I get to say, "Aha!".
Brilliant! Ignore the critics, your method is superior. @12:45 Your logic is valid logic, can't have 2 solutions, so 9 doesn't belong in those 2 squares, has to go in r1c3. Again, you're educating all of us, ignore the critics.
I agree that it's harder to spot things when you fill every cell with all possible values it could have. So I'm limiting my pencil marks to 2 (max 3). But that chain inference placing 3 at R8C6 that started at R7C2 with 3and7 as possible values is mind-blowing. I will never spot that ever.
I love watching this channel, it makes me feel like a genius as the puzzle is solved before my eyes hahaha
Hi! Do you have any tips for training yourself to become more aware of those chain reaction patterns that limited the three and the seven in the bottom middle square? Thanks! I love your videos!
Isabel Erwin practice by doing tons of sudokus. You’ll start to recognize patterns like that by yourself very quickly. You just have to do a few sudokus
Notating pairs as much as possible helps a lot, and, when you get stuck, checking what happens with each option in some pair. Pairs that are not "pairs of pairs within the same box" tend to be more likely to be involved in interesting chains, per my experience (but that might just be some bias that I have because of the way I solve sudokus, the things I spot more easily or the things I miss more often, etc., so "you mileage might vary"). Sometimes you can spot an interesting chain when a cell "is seen" by two different pairs, or is surrounded by other cells that are very restricted. Somehow, over time you start "feeling" that some cells are "in the middle of something" and some cells are just "quiet", uninteresting by now... I'm not sure there are exact criteria for that, but for sure that helps a lot with the solving speed (note how Mark always says "where should we look now..." when he is stuck: he is not looking for something specific, he is looking for "interesting" cells or patterns ;-) )
@@jacobbassam6616 Man I've been doing them every day for months and I still can't spot those chains
Wicked Samurai interesting. Maybe watch some videos. Try puzzles that use that specific strategy.
@@wickedsamurai3323 I found a sudoku app called HoDoKu. It might not be the best application, but you can ask it to make a puzzle that specifically has the strategy you want to practice. It is great, and it is free.
Took me 18:43 to solve; definitely one of the harder classics on the channel. Really enjoyed solving it.
As a side note, I've never really agreed with people who think uniqueness rectangles are a non-satisfying, or even, invalid way of solving classic sudokus. Mainly for two reasons,
1. As far as I know (I'll admit that I haven't researched this extensively myself, but I've read it on multiple sites/forums discussing sudoku strategies), while the strategy itself is based on the assumption that 'there is only one solution', even if you're solving a 'defective' sudoku with multiple solutions, using uniqueness will NOT lead to a wrong solution at the end of the road, just one of the multiple correct solutions. In other words, as long as the puzzle you are solving is a classic sudoku that has valid solution(s), there is actually no instance when applying uniqueness will lead to a contradiction, so it's ok to use it.
The only case where you should actively refrain from using uniquness strategies is when you're solving variant sudokus; in this case, since classic rules aren't the only rules enforcing a unique solution, applying uniqueness can, and usually will, lead you to the wrong solution.
2. While Unique Rectangles are introduced as fairly advanced techniques on a lot of sudoku tutorials/videos, I actually find that 'classic' Unique Rectangles (such as the example in this particular video) are fairly common and very easy to spot. They can often help out in tough spots, helping to avoid long chains. Again, I'll admit that the same can't be said for 'hidden' or 'extended' Unique Rectangles, but if you're solving a classic that requires these strategies to progress, it's probably an extremely difficult one anyway.
Obviously, I understand that everyone has different reasons and styles/preferences when it comes to puzzles, and if using uniqueness to solve sudokus just doesn't 'feel right' to someone, that's perfectly ok.
However I personally love URs and use them to solve classic sudokus at every opportunity, and encourage others to do the same.
I'm quite happy to use uniqueness if I can't spot any other way but I get the fact that to do so is not using the internal logic of the puzzle, instead using the external logic and assumption of it being a good sudoku. It's a bit like one of those "if you have $10, what can you buy to satisfy all" puzzles and turning round and assuming a good answer is "Feck it, I'll use my Credit Card"
Charles Fox Yes, it is ultimately external logic, but I do not see why this is an issue. It's such a purist, naive, and immature mentality to not consider external logic as part of the logic. Besides, some puzzles are explicitly designed such that the only way to arrive at the solution is precisely to assume that it has a unique solution. And if you don't ever use this rule, you'd have to simply literally guess the solution, or more formally use bifurcation, which many people also treat as not an act of using logic - though by definition, bifurcation is internal logic via the principle of reductio ad absurdum.
@@angelmendez-rivera351 that's good because I solve the rubiks cube by taking it apart and putting it back together again... cause it's how it's made.
Charles Fox That's a false equivalence fallacy. Not only are there are no Rubik's cube puzzles designed with the intent that external logic is required, but uniqueness is not even present. Sudoku and Rubik's cube is fundamentally present, in that a restricted binary deductive logic is required in sudoku, not in a Rubik's cube, which is purely combinatorial. So, your argument is invalid. I know you're trying to demonstrate the absurdity of my statement by making a comparison with a statement you think is equally as absurd that logically follows from my statement, but unfortunately, it doesn't work: they're not logically similar, and you can't prove my statement isn't absurd because it genuinely isn't. You can't prove a false statement if you're using the correct set of deductive rules.
You can choose to prohibit the use of uniqueness as a valid deduction in your puzzle solving habits. Sure. But as far as the actual theory goes, there is legitimately no good reason to argue against it. Comparing it to a Rubik's cube isn't going to help the "purist's" case, if it can even be called that.
Agree with everything you've written so far, and I just want to add that uniqueness opens up some beautiful logical strategies that can get just as complex as the "normal" logic we use in these puzzles! There are so many variations of uniqueness patterns, going way beyond simple URs. I feel that the classic sudoku experience would be less "rich" without it (and even some variants too, though one has to be very careful using them in variants, it must be "sealed off" and not overlap with the variant logic to work ;) ).
14:44 why 8 must be in that box? It can be put also in the two boxes above right?
Because 2 columns under eliminates the 4 squares and the 2 squares with the notes are reserved by the digits 3 and 5 so thats why the 8 is left 🤗
14:40 Why can 8 not go into the top two cells of column 3 (with the 3 and 5)?
Because the two cells are the only two cells in that column where 3 and 5 can exist. If you put an 8 there you wouldn't have anywhere to place both 3 and 5.
to follow up, it also is why they use the center fill-in when they are putting the only possible candidates for a box. when there still are unknown possibilties they place the numbers in the corners. it is a nice a elegant way to keep track of what you have figured out....or not figured out. ...so if you see 2 boxes with a 35 pair they redo those as center then remove the corners. it then eliminates a possibility for the other numbers they had in the corner which then often leads to knowing where that other digit goes. finding pairs is key to solving these puzzles. on very difficult possible you often will have to find a very hidden triple! hehe
10:12, completely missed that, ended up making a chain completing the puzzle of that cell was a three. Worked out every candidate trying to find an error and it turned out to be the final solution. Took me 40 minutes.
At 9:22, assume it's a 3 for the 37 pair, why then 3 goes to the bottom on the 9th row when 3 could also go to the 7th row? I think this might be a flawed logic? Please comment.
No it is not a mistake. If row 8, column 2 cross is assumed to be 3, that means for row 9, where can the 3 go in row 9, it can only go into row 9, column 6, bottom. So look at row 9, not at box 8.
@@wynandstockel9592you are assuming 3 in box 7 at the place of 37 because 3 is not marked anywhere elese in box 7, if you see 3 can also come in box 7 bottom 3rd column next to 2 , with this the logic of 3 can only come on 9th row failed
Thanks for this video learned something new with that chaining link. Very helpful.
9:46 inference is only a guess at this point since 7 can also be top right cell of the bottom right box. Therefore guessing is part of the ?techniques'?
Thank you. You've given me a new way of solving the puzzles.
Thank you for this in depth look at your process. I’m new to the channel and had been wondering what you meant by “naked single”. Very insightful!
Thanks, I'm looking for more clues. The chief take home from this is NOT to put all possibilities in each box, but to work with the ones that can be only 2 choices, or possibly 3.
I tried to solve one puzzle from past few days. Today I solved it after watching this video. I learnt the way of using the notes. thank u
I could not understand how the 3 came in the 8 block... down side of “2”..logic I missed .. 3 can be on above of “2” also.can please explain in detail..??
Do you mean that 3-8 in the upper right corner of box 9?
He did explain it. There are only two possibilities for that square because box 9 already has 179 and the column has 246 and the row has 5. The only numbers left are 3 and 8.
Great channel ! I discovered you by chance a few days ago and you made me realize that Sudoku can really be thrilling !
I first watched... then I did the sudoku and watch again... makes so much sense. Thanks ..
I'm sure I have become a better player .
Yes, u r sure not bad for the country 😂😂😂❤
Whyd you fill 6 before 4 in the middle left box, it could've been the other way around also couldn't it? At 3:52
8:11 For me that cell was the key to solving the entire puzzle.
This master solver saying "don't get triggered" has me cracking up! Nice one, man.
9:19... someone please explain. Is that some kind of rule? Shouldn't all the 3's an 7's have been pencil marked first in the bottom left box before concluding where the 3 goes in bottom middle box?
At about 11:16 you made an assumption that a 3 can only go in row 8 column 2. It seems to me that t is also possible at this point for it to go into row 7 column 3. Ultimately, it turned out that you were correct but I cannot see how that assumption is provable with the information available at the time. Am I missing something or were you lucky?
I think you're on to something -- I can't see how 3 at this point has been eliminated from r7c8 and r8c9 that an 8 on r3c9 would guarantee the 3 would appear in r7c9, and thus in r8c2. I mean, being him, I can very well imagine he worked it out in his head some other way, but he has not explained it XD
The 3/7 deduction was because that cell had to be a 3 or 7 due to the rest of the column and row, so once the 7 was eliminated as an option because of it's position in box 8 that cell had to be the 3 regardless of where it could have possibly gone in that box.
@@bmac5077 Thanks, I see it now.
The tip about uniqueness was awesome. Thank you.
For those new to Sudoku or are stuck at an intermediate level, these sort of videos are absolutely essential. Thanks!
Can someone explain how he got a possible 3 in box 7 (column 2, row 8). As far as I can tell that 3 can also be possible in all the other empty cells except column 1, row 7. He puts it in around 8:17.
He did actually explain it. It can only be 3 or 7 because all the other numbers are taken by the column and row it is joined to. The column it is in already has 124568 and it can't be 9 because it's already in the row so here are only two possibilities left.
Uniqueness is a genuine strategy. Instead of looking for crazy patterns that only a genius could see, one can simply use uniqueness and simply solve it. Insisting on the idea that using uniqueness is not pure is BS, because the other harder logic that he finds would simply not work if sudoku was not unique.
I'd agree on that. The uniqueness technique is using one simple assumption: 'This puzzle has a unique solution'. The uniqueness trait is the _default,_ and is _expected_ by the vast majority of people who solve sudokus. Sudokus without a unique solution are considered by many as invalid, and I'd reckon that most sudoku setters are taking pride in the fact that their sudokus are unique. Also, many other more complicated techniques are relying on uniqueness, like every 'unique rectangle' technique, for example.
In my opinion, this 'uniqueness is not pure' idea is completely arbitrary nonsense.
Good puzzle, nicely solved. I can understand the objections to using uniqueness rather than logic to place a number but I say don't hit on the solver for using uniqueness, hit on the setter for allowing that situation to occur in the first place.
Unless the puzzle has a very narrow solve path, I’m not sure how well a setter can avoid all possible scenarios where the solver might notice uniqueness. See the Kurt Schneider recent video where he sees it like half a dozen times in 30 seconds in a random star battle puzzle.
As long as the puzzle can be logically solved without relying on the uniqueness logic, it doesn't much matter how many times you spot the option to use it. It's just that, if a setter creates a puzzle that NEEDS a Unique Rectangle trick to solve it, then it's not a valid Sudoku, as it has three solutions. At LEAST three :P
A fun joke puzzle might be one where uniqueness problems are everywhere, and while it has like 20+ valid solutions, if you apply the Unique Rectangle logic, you can force it down to just one. Might be interesting for the channel, as a way to make use of uniqueness for a day without having to forego the beauty of a puzzle. The beauty is in how ridiculous a concept it is :P
I have found that if you have pencilled in say a 1 in the top left square in column 1 and 2 then it follows that in the two sqares below the 1 will be either in Rowe's 1and 3 or 2 and 3. This has helped me many times.
I'm new to this and I do mean new and this is really helpful in getting an understanding of the process involved, many thanks.
9:40 - why can’t the bottom row, third box across be a 3 ? And same for the box 2 above that?
this is the alternate universe where Hank found a sudoku book on Walter's toilet
Thanks for this very informative video! Managed to solve this one by myself first thanks to watching other videos on your channel. So it was interesting to then watch your solve and compare my results.
I very much enjoyed this video! Im happy that you bring the explanations down to my level and take your time explaining some of the concepts (particularly the uniqueness bit)
U send hard sudoku puzzles for me
I enjoy your solving. question though at 6:30, could you restrict the center cell to 8 and 9 ? Does not make a difference later on for this sudoku, but I found that is sometimes usefull on other sudokus to restrict 2 out of 3 triplets digits in only 2 cells. this way triplets in line 5 and the center box would be perfect (set of 2 numbers rotating in all 3 cells)
5:59 my first miss. I had 5 being possible in box 4's 1, 7, and 9 squares. But since box 5 requires a 5 in the bottom row, that constrains box 4.
YAY! I actually solved this by myself. Been ages since I last did a sudoku puzzle though the alternative one's really confuse me with all their different rules which I don't even understand most of them but still watching them being solved 'till something hopefully clicks in my brain so I can understand it lol Also loved your comments from 5:10 onwards. I've only just found your channel which is great for me trying to solve them before watching your video to see any difference.xx
At 11.20 a possible 7 at r7c6 is ignored. Why? Please.
At around 10:00 you explain why a three is place in R9-C6. Is there a name for the technique because it totally eludes me. Thanks.
Can someone explain the 3 and 7 trick he did? I tried to read the comments but i still dont get it
I really enjoyed the deductions, thank you Mr. Holmes.
9:26 See? This is what it all comes down to sometimes. Making assumptions. In this case it's the beautiful simple logic. You have two choices A & B if they both lead to the same result then they're either both true or both false. I love this!
At 9:37, he rules out the 3 in r7c6 but I don’t understand how/why?
It's a FT (forcing test) with the 37 in RC82.
If RC82 is a 3, the 3 in the bottom right hand block goes into to R7, which punts the 3 in C6 to R9.
If RC82 is a 7, a 7 would go into RC76, again punting the 3 in C6 to R9.
Either way the 3 in C6 is in R9.
I wondered a couple of times why you guys avoid the unique solution, so thank you very much for explaining! :)
For myself I just figured out that it is harder to spot, when additional rules apply. For example a knights move could solve the double 79 pair and it would have been a mistake to say it is a 4 because of uniqueness.
Thank you again for the explanation :)
The precious one! Two powerful methods I did not know. Thank you very much!
Learned 2 things; Naked singles and why you fill some possible numbers in the middle while others go in the corners. Very useful. Thanks!
Hi.. got lost at 11:21. Why can't a 7 also go in R7C6? What did i miss?
I have the same question
@@nacho__063 He can't put a 7 in R7C6 because once he puts a 3 in R8C2 the only place for a 7 in Block 7 is the same row you want to put a 7 in Block 8.
@@oldcar8580 ohh I see, It was a matter of the threes, couldn’t see he bigger picture. Thank you!
Amazing video and puzzle, great tips. Thanks !
मराठी
छ
Brilliantly solved with good teaching
Alternate inference chain... On 9:30?? Some tricky techniques i need to learn. Xy chain on r3c9?. Solved it in 22:33. I need to practice more on those diabolical techniques
I have just discovered sudoku. I'm hoping with fingers crossed that a fair portion of seeing where the numbers can go comes with repetition & practice because watching this fella makes me want to chuck it in before I start! How tf can he see these patterns so quickly - i.e where he saw to put the 3's straight away. Jings
I got severely interrupted twice while doing this, so I don't have an exact time,, but I think I spent half an hour actively solving it. Didn't use the inference chain that Mark showed, or the naked single in box six, I managed to find a different logic path that still led to the same conclusion. Which means it's a good puzzle, since it can be solved for the same unique solution using different methods of attack.
Genius! I enjoyed every single second.
Could someone explain 8:17, why is the bottom left corners 3*3's centre cube a 3 or 7?
Because all other numbers are already in its same column or line.
Wow, that happened fast! I had to pause and go back to watch a few things over, but THANK YOU!
You are absolutely fantastic. I love the way you solve
Possibly a dumb question, but at 8mins 46secs you say the 9 blocks out C2 R8 (understood) and that 9s must therefor be in C2 R1 or R2. But why can't it be in C1 R2 or C3 (R1 or R3)? Thanks. Nicely done!
How do you know at 14:38 that the 3 now makes a naked 35 pair? Why can't a 4 go there?
That was a masterful lesson in how to make the difficult look easy, thank you. Is the technique you used two times equivalent to the XY wing technique? Either way it still needs much practice to spot these possibilities.
This pandemic has brought me to this channel out of nowhere and i can't stop watching. Absolutely brilliant!
I have solved this puzzle in 45 minutes after I heard how you do it. Then it is easy to follow and solve
yourself
I love CTC's solving style so much I made a video on my channel where I
analyze each step of this solve. In addition to explaining the
strategies used, I add a couple of pause the video moments and some
alternate solving paths, similar to how someone analyzes a chess game.
I couldn't follow the step at 11:14. How did you rule out a 7 in row 7 column 6?
Ok duh! It is because once the 3 is placed in box 7 it means that the 7s in box 7 have to be in the top row! Which rules out the 7 in the top row of box 8!
@11:15 can anybody explain to a novice here why can 3 only go in the middle cell? Why not the top right of that block?
Great Job and very informative! I learnt a lot! Thank you so much!
I understand the solution but how you spotted it was incredible :O
At 8:06 in the middle box lower left why is it either 3or 7
Because in column 2, you only have 3, 7 and 9 missing. And the 9 is ruled out by the 9 in row 8 (in the last box)
Your comment might be slightly confusing since a box refers to the bigger ensemble containing digits 1,2,3,4,5,6,7,8,9
having watched your videos again I can now feel confident in my skills
I think it is a pattern (e.g. 25, 25 in two boxes middle) you have to select in your brain and then test the pattern selected if it conflicts you change to another then you get the answer quickly.
if not "uniqueness" then what other technique could be used to solve 3/7? appreciate help!