Hearing him at 7:18 changing the color of the cells when he is so intently focused just because it'll be better for people who suffer from color blindness is so genuine and one of the reasons why i think simon is absolutely amazing person!!
Not to take anything away from Simon, but he was not so thoughtful in the past and got a lot of comments on this subject. Natural evolution, if you want the channel to succeed, is to take such comments into account.
I actually think Simon missed a clue in this one: whenever there is a chain of 3 W's , the centre box MUST have a low number, since that is the only number that can chain to both numbers in the ends. At least that's what helped me solve it :)
Fun fact: in Japanese, posting a comment that ends in "w" means you're laughing (similar to our "LOL"). So all the w's in the grid mean the puzzle is laughing right along with us.
Just to add, it's because the word for laugh/smile is "warau". However, you're much more likely to see the kanji character (笑) used than the alphabetical "w".
@@davidh.4944 Are you sure about seeing the kanji more? I'm not an expert in Japanese by any means, but I do post many videos on NicoNicoDouga. I've never seen that kanji in the comments, though I have gotten plenty of "WWWW" and "WwWwW" comments. It's the same when I message my penpal (internet/keyboard pal?) Though I can chalk that up to her understanding my kanji is still limited to basic grammar and the most commonly used ones. Edit: I'm not claiming your wrong by the way, simply that what you're saying is contrary to what I've personally seen. Is it possible that this is due to say, internet culture vs texting? Or perhaps full grown adults tend to use 笑 while teens and young adults use wwww?
@@VampireBabysitter Well, it may be just a personal observation, but in my quarter-century here I have occasionally seen (笑) used, but I can't recall ever seeing (w). Perhaps it is a relatively new, and mostly online, phenomenon, since I don't spend my time hanging out in Japanese forums. It wouldn't surprise me, though. It is absolutely astounding how much the language has changed in just a generation, particularly in the number of imported English words used in everyday speech. Even in just the last decade, I've seen common terms like "business" (ビジネス), "family" (ファミリー), "trouble" (トラブル) all but replace their traditional equivalents. This could just be another instance of that.
@@davidh.4944 it also may have something to with phones. Like how shorthands like lol and lmao became commonplace in English because flip phones made typing tedious, I wouldn't be surprised at all if that exact same thing caused w to become a normal shorthand. I have no idea the Japanese layout of flip phones so maybe that's completely wrong, but considering how popular they still are in Japan, if it's as difficult to form kanji as it was to make English words that could be it.
After seeing the grid, I immediately thought the title was a reference to the fact that "w" is often used to represent laughter by Japanese, because laughter in Japanese is 笑い (warai) which starts with "w".
Simon could've used the negative constraint to remove the 3 from R8C3 The middle of two W's in a box can't have high digits, so remove the 7 and 8 from R7C3. Then R7C2 can also be reduced to pair.
First of all, I would like to thank Simon and Berni for this amazing puzzle. When I started watching videos of this channel, I found difficult to solve any puzzle. When I came across with this puzzle, I started to laugh and left it. Today, many months later, I managed to solve it and it was quite approachable. I really appreciate the work of every sudoku creator and of course you, Mark and Simon, for explaining so clearly in every video the solution of a sudoku.
Interesting puzzle. Instead of colouring known high/known low cells, I coloured according to the "polarity" of a chain: if one cell in the chain is a 1,4,6,9 then all cells are 1,4,6,9, and the same is true of 2,3,7,8. The same rule that each square, row and column must have four of each applies, but I think it's easier to see the extent to which the restrictions placed by the two known cells (one for 1469 chains and one for 2378 chains) propagate.
Same way I approached it seemed more elegant when solving and it was very satisfying. I just marked the lows and highs with pencil marks as necessary, at the end I also realized that row and column needed four of each color so it helped me figure out from which number pool I was missing.
Between knowing that each area (row/column/box) must contain 4 "high" and 4 "low" digits, and identifying the "polarity" of each chain (1469 or 2378), it is possible to color the entire grid without even using the given clues (r9c4 and r6c8) to determine what goes where.
I colored both ways: I had 2 colors for high/low, 2 colors for different chain options (2378,237 / 1469,146) - i took me longer than it took Simon but I had double fun coloring because I had almost everything colored before putting numbers in.
I found a different coloring method that helped me solve this in 15:45, which is a very short time for me. I realized that there are two groups of digits, (1-4-6-9) and (2-3-7-8). The X/V clues must always join digits from the same group (for example, the 1 joins to a 4 or 9). So, I started by coloring 4-long chain in box 3 as green. I found the 5s the same way Simon did. That means the remaining 4 cells in box 3 were not green, so I colored them purple. Those cells are joined by X/V clues to box 2, so I colored those purple, leaving 4 cells in box 2 which must be green. That chain must be all green cells, extending into box 5, and so on. Once one of these colors reaches one of the two givens, you can assign which digit group belongs to which color. When the whole grid is filled in, solve one color first, then the other, since they are essentially independent, and this becomes a divide and conquer strategy. Also within each digit group, there are two pairs that sum to 10, and one that sums to 5, so a chain of 4 within one box must be a 10-sum, 5-sum, and 10-sum in that order, which yields a lot of pairs.
Well done at 15:45. I took 17:00 I realized the two groups as well. But more than that, the ordering helps. If you write them as [6-4-1-9] and [7-3-2-8], it helps to see that the inner low numbers have two X-V relationships with their neighbors, and the outer high numbers have only one X-V relationship. That made it easy to write in the long sequences. A 4-cell sequence can immediately be written as [69] [14] [14] [69] or as [78] [23] [23] [78], depending on the color. Loads of pairs emerged. At some point the coloring didn't even matter.
This is essentially how I went about it - color in 1-4-6-9 vs. 2-3-7-8. Got all the colors except four cells that needed a break by Sudoku, then started on numbers Went to look at his solution after and was blown away that we both built around 2 colors but for *completely* different purposes.
There is a very simple rule in this puzzle that will make the solving much faster. Every time there are 3 or 4 connected cells in the same box/row/column, the middle cell(s) must contain low digits.
Yeah I spotted that logic too, what I did though was to look at all the chains as blocks of three. So with a chain of four cells I would look at the first three, then the last three. Five cells would be first/middle/last etc. So I just thought of it as "In any chain of three cells constrained by normal sudoku rules, the middle cell is low."
I always wonder if someone gave Simon a puzzle with some of the fancier constraints, killer cages, thermos, looked like it needs ring\set etc but the only way to solve it was std sudoku, how long do you think it would be before he noticed?
This is the kind of puzzle I want an entire book of! Fun, approachable, but unique. I really appreciate the puzzles that require some basic maths, they're impressive, but I really enjoy the anti-chess sudokus and ones like this where it's all more basic/visual logic
3:07 I'm completely lost from the start; why do the zero and nine square immediately become a 1 and an 8, respectively? I'm not familiar with the small-starting-number notation. What just happened?
I know this is quite late but I had the same question and figured out the answer so I figured I would put it here in case anyone else wondered in the future. It's basically the every clue is one before or after rule, since those cells have a zero and nine hint in them then they must be one at eight, respectively.
Is it proven that the negative constraint was necessary for uniqueness? I didn't see him use it until very late when the puzzle was so far done that it might have been finishable without it.
@@jaunusender6166 i solved it completely forgetting about the negative constraint, and got a different solution. Of course I may have made an error due to that... But I triple checked and didn't see anything wrong
If any digit is between two Ws that share the same box, row, or column, it has to be a low digit. If it were a high digit with any of those properties, then you would have a duplicate low digit in the same box, row, or column. I am often slow on these puzzles, but I managed to complete in 22:35 by realizing this restriction.
I noticed this right after I filled in the 5's. The first time ever I got off to a faster start than Simon. Instead of colors, I just filled the boxes with 1234 or 6789. then calcelled out numbers as I found them elsewhere.
In German, starting k is never silent, so its pronounced like the c in "cup": "cnup". You got the vowels right, though: "e" is either like the e in liver, or it sounds something like the "e" in "sire" or "higher", but not like "a".
And in case anyone was wondering: The translation is "near miss" or "just off the mark", but is frequently used in germany in similar circumstance as "close, but no cigar" (although the full saying would be "Knapp daneben ist auch vorbei" which means "a near miss is still a miss")
Yeah, and the type of pairs need to alternate if they are constrained by sudoku, so a HL pair has to connect to a LL pair, meaning you could also never have a 5 chain in a box (alternating means we get HLLHH or LHHLL, can never have two H next to eachother, and never 3 L's)
Same here. You can immediately see this pattern: _ W _ W _ W _ all over the grid. Now, only the V clues can touch. The X clues need to be on the outside, so that pattern is ALWAYS _ X _ V _ X _ . This means you get repeating patterns of " High X Low V Low X High". The interesting bit comes when you have a run of three (as in Box 7 and Box 9). Since you can always have V clues connected, it means the run is either High X Low V Low or Low V Low X High. Once you recognise this pattern, you are able to colour most of the grid, and then the V clue between R5C4 and R6C4 must be a 14 pair, as neither high digit on either side can be an 8. And then you are cooking with gas, as half the puzzle disambiguates when you apply the given 1 from box 6.
Got this one done in 13:08, this is the first time I've solved something almost 7 mins faster than Simon (yes, I know he's live-solving with commentary, but still). What worked out for me is that a chain of 3 connected cells in the same box/row/column means the middle one needs to be small (1, 2, 3, or 4), and a chain of 4 in the same box/row/column should thus have both middle cells be small.
One bit of interesting logic that I've not seen anyone mention in the comments is that in column 1, once you've put your 5 in there, you're left with 4 'W' dominoes, which means that each one of them must be a high-low pair adding up to 10. It isn't massively useful but it did give me a couple of digits. (Sorry if this has been mentioned and I missed it)
Yup, I noticed that too and used it in my solve. At the point where I was using it, I had enough other cells filled that it gave me most of the column.
13:04 "that's quite good" - ok, he sees it. 13:14 "if this can't be 2, this can't be 3 and we have a 1/4 pair" #FlipTable #ScreamAtScreen #PointAt1InTheSameRow
27 minutes! I feel pretty good about that time considering I don't usually find a break-in for 95% of the puzzles featured on this channel lol. This one was fun!
The key logic to this puzzle is that "for any chain of length 3 in a row, column or box, the middle cell is low". Simon saw it right away for one cell at 6:00, but didn't follow through with it over the whole grid. Boxes 3 and 7 could have been solved much faster with that.
I'm not sure it's "key" . I almost never finish a puzzle faster than Mark or Simon. I did finish this one quicker though, without doing anything clever other than quickly cancelling candidates that don't add to 5 or 10. Your fact about chains is certainly true but I never once thought of it.
@5:47 does he already ignore the fact that the W could indicate a sum of 5 or 10? He immediately says if r2c5 is a 1 then r2d4 "has to" be a 9, when it obviously could be a 4.
Yeah, he instantly lost me. Too confused by the workings of this one to wrap my head around it, because I spent that time going "but why? It can equal 5 too".
28:17 One of the most fun puzzles Ive ever done. Started off color coding the whole middle square, and then slowly expanding the coloring until it got cancelled out by the 8 and 1 given. Would have been much faster but for some reason I decided to keep playing the XV game for a lot longer than I should have, when I could have just moved onto regular sudoku. But I guess I had more fun my way.
Wow, I managed to finish this one in 20:33, which is the first time I've even come close to finishing one in less time than the length of the video, much less actually done so. Quite enjoyable indeed! And I see from another comment that I actually beat Simon's solve time by 11 seconds, not just the video time. I didn't do any coloring; the only markings I used were center numbers -- which ended up being lots of 2378 cells, or subsets of those. About halfway through, I had everything except 2, 3, 7, and 8 completely solved, but had only one 7 on the whole grid, and none of the others!
Hi Simon. Something I don't think anyone noticed: there's an extra bit of logic with this high-low system if you ever needed it. We know there are 81 cells total in the grid. If you know the locations of all the 5's like in this puzzle, and if you had a LOT of one color type cell, counted them and found the total to be 36, you could immediately color all remaining non-5 cells the other color.
This was probably the easiest solve i had on this channel so far, but also so much fun. I started coloring the chains because you can always only combine 4 numbers with each other in a chain of w's, and then gave secondary colors to the high and low numbers, my sudoku was a whole coloring book. colored sudokus are my favourites
15:21! Late to the party but still commenting because I'm proud of myself. Not the fastest, but the first time ever I've been faster than Simon at a solve. Same reason others have stated - realizing straight away that if a cell has 2 W's connected to it in a situation where both ends have to be different digits (so all in a line or all in the same box), they need to be one X clue and one V clue to avoid repeating a digit. Therefore, those cells must always be low digits because high digits can't be part of V clues.
After solving it I watched Simon's solve, and I didn't use the high-low idea; there was a plethora of 2-3-7-8 cells and a few 1-4-6-9 ones, but they all sorted themselves out with the help of the negative constraint.
the given 8 in R9C4 means that the 13 cell chain must be a 1-4-6-9 one. which forces the seven cell chain to be 2-3-7-8. Also liked that C1 had to have all the pairs adding to 10.
@@not.rebecca1579 Well, actually I couldn't think of a word beginning with k. It's logical, that Simon thought of a silent k, as it is silent in so many words like knight, knot, know,, kneel, whatever. You have to come up with names like Klein in Klein bottle, so it's also pronounced as a German pronounces k. English also has ck has a hard k, harder than c, but no words begin with ck, not even ckracking the cknyptic.
1:13:20! I actually finished one. I got to something like 52 minutes and saw that I had like 3 6s in some column and didn't want to unwind. Reset and remembered the tricks for the groups of 4 in a row (2378 or 1469), and then went zoomy fast!
Fun fact: '"w" is the Japanese equivalent of '"lol". The word for "laugh" is 笑う, or _warau,_ which gets abbreviated to just "w" in Japanese "text speak"
This puzzle made me realize that your app does not expect one and only proper solution, but rather checks out standard Sudoku rules. When you are left with all the 23s and 78s to finish the puzzle, you can actually enter them both ways and end up with 372 in the bottom row. Breaks the X constraint, but you still get 'Looks good to me!'
Alternative start after 5's: column 4 has 3 W in it and the 8 is already placed. It is impossible to make a 8-2 pair in this column. Either the 2 of this column has to be in row 8, or it has to be linked up with a 3, in which case there is no 3-7 pair and the 7 is in row 8. Because of the negative constrait, a 2 can't be in row 8 because then there would have to be an additional W. Since there is not, there is in fact a 2-3 pair in the column, and only a 7 can be placed in row 8. Which means that in row 7, the four linked cells that need 2 high numbers can only be made with 1469. I then proceeded to colour cells on whether they were part of the 1469 pairing, or the 2378 pairings, using central notation to find whether they're small or high numbers. Solved it in 40 minutes this way. Amazing to see this puzzle has such a different solve method available as well :O
Wow I got that in 22min, that really was approachable. The run of 4 digits above the 8 appears to only have a single solution constrained by the givens so I started there with quite a few digits which was very helpful.
Dodgy logic at 6:00 ... there's nothing to say that *that* cell can't be a 1, because it could be paired with a 4 - from the logic so far. Obviously that does break when you continue the chain. What you can say is that any time a cell is connected (by Ws) to two cells that are within the same domain (box, row, column) then that 'middle' cell _must_ be a small number. Here we go parity-colouring! (Completed in 28 minutes 👍🏻. And yes, I did laugh 😆. Love it!)
Got this done in 20:06, I easily filled in the 5s and then highlighted all the cells that couldn't have high digits due to repetition in column, row or 3x3 (Essentially part of the highlighting at 7:11 but for the whole grid) and that completely cracked it, other than the 2378s at the end, at which point the negative constrain finally came into play. Fun stuff
loved your solve, mine was along very similar lines. just a small difference in the way i started: i noticed that if there was a chain of 3 cells (all seeing each other) connected by W's then the middle cell always had to be a low digit. this made me get a few low-low pairs in a W-domino, which meant that W had to in fact be a V-domino. i then confirmed which cells had to be high digits by noticing where low digits couldn't repeat in the relevant R/C/B.
Holy crap, I've never done this well in a puzzle vs Simon's time before. It took me only 9:39 because I immediately spotted where to break into the Ws which gave a ton of data off the bat.
35:33 for me. I missed the little 9 and 0 so I had the whole grid color coded, each number had its own color. Once I looked back at the video and saw the 9 and 0 I was able to solve it, but I spent a good 5 minutes wondering what I'd done wrong. Fun video
Yep same here but caught on to the 9 0 clues when they corresponded to the high/low designation in their cells. I found it quite interesting doing it this way anyway.
This is an amazing puzzle - I absolutely loved it! It's very handy to draw a graph (in the sense of discrete mathematics) with vertex set {1,...,9}, and with all the W-relationships marked as edges. There are six edges in the graph. The graph has three connected components: 5 (an isolated vertex), and two chains, 9-1-4-6 and 8-2-3-7. The latter both go H-L-L-H (H=high, L=low), so if you ever find a chain of length 4 within a box or row/column, you know the pattern is H-L-L-H. Similarly, if you find such a chain of length 3, it must be H-L-L or L-L-H, so in particular the middle cell is L. This helped to quickly colour most of the grid.
58:00 exactly. I colored the grid blue and orange to figure out where the low digits were and where the high digits were. And I couldn't get past the 2/3 pairs in boxes 4, 7, and 9. No matter what I tried they broke. I was about to give up and I decided to delete all the digits but leave the coloring, then I started filling it in and questioned each color. I finally found 6 cells that could be the opposite of what I had. I changed those and finished the puzzle. The first time I hit check was at 33 minutes, so I would have been happy with that time, had I not messed up.
Oh, silly Simon. You got the colors backward. Orange O is for over, and Blue B is for below. :) I say that in jest but now that I'm watching the video longer, it's really messing with my head. I keep thinking you messed up in the logic. @13:30 in the video. I'm screaming at my screen. "But you have a 1 in row 6 to fix the 14 and thus all the cells in that chain." @14:20, I see that my deduction that 3 in a row or in a box, the middle one always has to be low would have helped you there.
16:35 first time I've ever been quicker (of the limited number of times I actually was able to get started lol) I found that every three w's had to be high-low-low-high very quickly and started eliminating options using the 1 and 8 instead of coloring. This gave a lot of numbers in the grid quickly!
This was a very fun puzzle. It did make me smile when I got near the end and realised the 'key' to cracking it lay with the very first number (1) that I put in the gird. Nice construction.
Delightful puzzle. Ended up proliferating the whole grid with 23s and 78s until the negative constraint down around boxes 8 and 9 sorted it all out in pretty much a single fell swoop.
27:19. Happy with my time. Agree that it helped to look at the groups with four W's (H-L-L-H) or three (L-L-H or H-L-L). This expands to the big chains of 7 and 13 cells which cross box boundaries and so are kept in chunks of four or less in each box. Also you can place all the 5's immediately. Haven't watched the video yet so I'm now going to see how Simon solves it.
This was a fun one to solve. I looked at the chains and figured them out before even touching the fives. And because I went straight to pencil marking, I didn't really use any colors.
Yeehaw! With a 20:00 solve this was my first ever being faster than Simon or Mark, even staying below Simon's actual solve time (from "Let's get cracking" 3:05 to "Yeah!" 23:49), not just the length of the video. And how nice it came out to 20:00 flat. Largely due to what others have also mentioned: realising that cells in the middle of a chain of cells seeing each other and connected by W's have to be 1234. What also helps a lot is to always think of the two groups of interconnected numbers 1469 and 2378.
Took me an hour and a half but I ACTUALLY managed to solve this on my own BEFORE watching your video (apart from the rules explanation). But now it's 3:30am and I need to sleep so I will watch this tomorrow :")
15:58 my solve. I'm normally not quick at solving sudoku's but I think when I realized that only small numbers could go where 3 cells connected with W which see each other, the puzzle became quite easy
Interesting to see the break in with the colours working out the high/low options. I started by spotting the cells with 3 connections r6c4, r7c5 - to have 3 connections, they must be a 2/3/4. From this started filling in options along the chain until I hit a restriction then following the eliminations. My first digit (after the 1, 8 and 5s) was the 6 in r7c4
Took me 34 minutes 05 seconds, but I'm proud to say that after your explanation of the rules and placing the first 8 and 1 I was able to do it all by myself :)
I didn't see the 0 and 9 clue before solving it, they don't show up in the app so I was quite surprised to see them when watching the video afterwards. I ended up using 8 different colors identifying each number and "solving" it without knowing what the actual numbers where. I just knew that "these 9 purple squares has to be the same number, these 9 blue squares has to be the same number and so on". Only the 5's were identified. First time I've ever identified all the digits by color but not knowing which color represented which digit, all I knew was which for colors were higher than 5 and which four colors were lower. Give it a try by ignoring the 0 and 9 clue, it's fun and end up looking pretty.
I focused immediately on the intersection of the "given" 1 and "given" 8, so coloring hi-lo was unnecessary since I knew from the start it was 14 or 69. As that chain goes into boxes with other chains, it tells you what they are too (23-78 for the most part). Simon dances all around the grid coloring hi-lo before honing in on the one spot it's disambiguated. One of the few times I feel I followed the more productive solution path. I'm usually amazed at his ability to anticipate the next easiest step. Of course that wasn't really essential on this puzzle.
I did a different approach first. I used 2 different colours for 1469 and 2378. Then I looked carefully about 1 and 8. Last, I focused on high / low digits. I had to count 4 green and 4 red in boxes, columns and rows.
I normally will hear the intro - hear the rules - and then dive in to a puzzle on my own to see how far I can get before watching and feeling silly for missing some obvious bit of clue. After the '5's, and realizing that there must be a 'high low low high' pattern, I actually solved this in a respectable time for myself (sub 30 minutes) - what a great puzzle that falls into place once you realize the constraints and patterns. (ie: 9 / 1 / 4 / 6 for example)
I found it easier to color high/low but to assign the center box cells a letter. I was able to nap all the letters around the grid and then numbers as they came along
And that the surrounding cells must be opposites, i.e. one high and one low, never both high or both low. Just watch out for boundary crossing when looking at ones in a 3x3. Also be careful in a few places to note the negative constraint as if you go fast, you could slip up. Cheers!
I totally used colours too! And felt the need to colour it all in before clicking check. Though I also coloured the 5s grey. And took 5 minutes longer than you even if I count your introducing-the-puzzle time...
I found an easy break in at the box with 3 w's (r6c4). It can only be 3 or 4 because off the 1 and 8. If you work the possibilities out for the adjecent cells you find that it can only be 4. From there you can fill in all the numbers.
29:40 for me. I didn't do any colouring, just started looking at the consecutive W relationships in the middle box which are heavily constrained by the initial digits. It really quickly snowballs from there (well, quickly if you're cleverer than I am. I went slow and methodical to make sure everything works as it should because I'm not great at this).
I solved this but had the 7's and 8's swapped. Took a while of checking the grid to realize there are two hints (the 9 and the 0) shown in the video that aren't on the app! Without those two hints, this puzzle has more than one solve.
The best start is box 5. We already know where the 5 is, and the sum of the three cells on the right hand side must add up to a multiple of 5 given the placings of the Ws in the rest of the box.But They cannot add up to 5, 10 or 15 because there would need to be a W between the two cells below the 5. So, the three cells must add up to 20 and the two cells below the 5 must therefore add up to 15. So, 69 or 78 and, given the position of the 1 in box 5, it is easy to work out. The rest of the puzzle then links all the way through.
21:20, this was a really approachable. I found the logic that middle digit in a sequence of 3 or 4 must be low. That high digits always have low digits as neighbours. That colouring sure helped seeing where I could fill in the rest. Really like this one. But I started with filling in all the fives. Let's see what Simon made of this.
@CrackingTheCryptic Thank you again for another fun video! As a someone who is colourblind seeing as you mentioned it in this video I'd like to suggest something: while the colours used in this video are easily enough to tell apart would it be possible to include stronger colours in the app? I have a stronger version of the "red/blue colourblindness" which and most people I know who shares this type of sight tend to like strong colours over the softer ones currently used in your app as they blend together very easily. It's just a suggestion of course and this opinion may not carter to the majority but I'd like you to know that there is a demand for stronger colours :)
This was really fun to do. I always like it when the rules make me use colours. 🎨🖌️ (Although, I have to admit that Simon messed me up quite a bit when I was watching after solving, because he coincidentally managed to exactly inverse my colours for high and low...)
This is actually approachable. Took me 1h 45min, but that may be just me… The 5s are all placeable pretty soon. Then column 1 and row 3 must be full of Xs. Then I just placed all the “high” and “low” digit pencil marks wherever an X or a V must be, always watching for negative constraints, counting the number of high and low digits per row, column, and box, and taking into account consecutive W clues. But it took me a while to also consider how many X and V dominoes there can be per box, e.g. in box 7, where the two bottom left Ws both have to be X, and exactly one of the top right ones has to be a V. The 1 and the 8 that you placed at the start constrain lots of dominoes, and standard Sudoku rules do so, too. At some point you start getting actual digits in the grid; then it’s all pretty much solved. Watching the video, I don’t know how I missed the quite obvious breakthrough around 12:00.
Hearing him at 7:18 changing the color of the cells when he is so intently focused just because it'll be better for people who suffer from color blindness is so genuine and one of the reasons why i think simon is absolutely amazing person!!
Not to take anything away from Simon, but he was not so thoughtful in the past and got a lot of comments on this subject. Natural evolution, if you want the channel to succeed, is to take such comments into account.
@@sasopalma4993 I'd say that's even better because it means that he listens to those who watch him and care about what they say much more than others
But i see your point, it's a valid one
I enjoy the fact that he continues using colors long after he needs to.
I've seen complaints in past videos where Simon didn't finish a coloring scheme. Perhaps he's trying to avoid the wrath of his ocd viewership.
He enjoys it as well, especially when he is already enjoying the puzzle.
I actually think Simon missed a clue in this one: whenever there is a chain of 3 W's , the centre box MUST have a low number, since that is the only number that can chain to both numbers in the ends. At least that's what helped me solve it :)
That's only true if the two endpoints are connected as well (in the same box / column / row), but yes that's a very useful starting point.
@@nothayley exactly.
Yep, that's what got me started after placing the 5s.
That relationship is what helped me solve it as well
Yeah if you have a cell where the w’s round a corner into another box it doesn’t apply. But that’s the rule I used to solve too!
Fun fact: in Japanese, posting a comment that ends in "w" means you're laughing (similar to our "LOL"). So all the w's in the grid mean the puzzle is laughing right along with us.
lol
Just to add, it's because the word for laugh/smile is "warau".
However, you're much more likely to see the kanji character (笑) used than the alphabetical "w".
@@davidh.4944 Are you sure about seeing the kanji more? I'm not an expert in Japanese by any means, but I do post many videos on NicoNicoDouga. I've never seen that kanji in the comments, though I have gotten plenty of "WWWW" and "WwWwW" comments.
It's the same when I message my penpal (internet/keyboard pal?) Though I can chalk that up to her understanding my kanji is still limited to basic grammar and the most commonly used ones.
Edit: I'm not claiming your wrong by the way, simply that what you're saying is contrary to what I've personally seen. Is it possible that this is due to say, internet culture vs texting? Or perhaps full grown adults tend to use 笑 while teens and young adults use wwww?
@@VampireBabysitter Well, it may be just a personal observation, but in my quarter-century here I have occasionally seen (笑) used, but I can't recall ever seeing (w). Perhaps it is a relatively new, and mostly online, phenomenon, since I don't spend my time hanging out in Japanese forums.
It wouldn't surprise me, though. It is absolutely astounding how much the language has changed in just a generation, particularly in the number of imported English words used in everyday speech. Even in just the last decade, I've seen common terms like "business" (ビジネス), "family" (ファミリー), "trouble" (トラブル) all but replace their traditional equivalents. This could just be another instance of that.
@@davidh.4944 it also may have something to with phones. Like how shorthands like lol and lmao became commonplace in English because flip phones made typing tedious, I wouldn't be surprised at all if that exact same thing caused w to become a normal shorthand. I have no idea the Japanese layout of flip phones so maybe that's completely wrong, but considering how popular they still are in Japan, if it's as difficult to form kanji as it was to make English words that could be it.
After seeing the grid, I immediately thought the title was a reference to the fact that "w" is often used to represent laughter by Japanese, because laughter in Japanese is 笑い (warai) which starts with "w".
i was thinking the same thing 草
Things you learn on CTC... :)
草
I had same reaction, actually you just spoilered to me this is not the case.
Same here
"That doesn't look like it's going to... butter any parsnips."
I’m adding this to my list of dumb shit I say 👍
Simon could've used the negative constraint to remove the 3 from R8C3
The middle of two W's in a box can't have high digits, so remove the 7 and 8 from R7C3.
Then R7C2 can also be reduced to pair.
But I like buttered parsnips.
I heard that and was like ".....I'm sorry what"
It comes from an idiom "Fine words butter no parsnips" meaning nothing is achieved through empty words and flattery.
First of all, I would like to thank Simon and Berni for this amazing puzzle. When I started watching videos of this channel, I found difficult to solve any puzzle. When I came across with this puzzle, I started to laugh and left it. Today, many months later, I managed to solve it and it was quite approachable. I really appreciate the work of every sudoku creator and of course you, Mark and Simon, for explaining so clearly in every video the solution of a sudoku.
Interesting puzzle. Instead of colouring known high/known low cells, I coloured according to the "polarity" of a chain: if one cell in the chain is a 1,4,6,9 then all cells are 1,4,6,9, and the same is true of 2,3,7,8. The same rule that each square, row and column must have four of each applies, but I think it's easier to see the extent to which the restrictions placed by the two known cells (one for 1469 chains and one for 2378 chains) propagate.
Same way I approached it seemed more elegant when solving and it was very satisfying. I just marked the lows and highs with pencil marks as necessary, at the end I also realized that row and column needed four of each color so it helped me figure out from which number pool I was missing.
Between knowing that each area (row/column/box) must contain 4 "high" and 4 "low" digits, and identifying the "polarity" of each chain (1469 or 2378), it is possible to color the entire grid without even using the given clues (r9c4 and r6c8) to determine what goes where.
I colored both ways: I had 2 colors for high/low, 2 colors for different chain options (2378,237 / 1469,146) - i took me longer than it took Simon but I had double fun coloring because I had almost everything colored before putting numbers in.
I found a different coloring method that helped me solve this in 15:45, which is a very short time for me.
I realized that there are two groups of digits, (1-4-6-9) and (2-3-7-8). The X/V clues must always join digits from the same group (for example, the 1 joins to a 4 or 9). So, I started by coloring 4-long chain in box 3 as green. I found the 5s the same way Simon did. That means the remaining 4 cells in box 3 were not green, so I colored them purple. Those cells are joined by X/V clues to box 2, so I colored those purple, leaving 4 cells in box 2 which must be green. That chain must be all green cells, extending into box 5, and so on.
Once one of these colors reaches one of the two givens, you can assign which digit group belongs to which color. When the whole grid is filled in, solve one color first, then the other, since they are essentially independent, and this becomes a divide and conquer strategy. Also within each digit group, there are two pairs that sum to 10, and one that sums to 5, so a chain of 4 within one box must be a 10-sum, 5-sum, and 10-sum in that order, which yields a lot of pairs.
Well done at 15:45. I took 17:00 I realized the two groups as well. But more than that, the ordering helps. If you write them as [6-4-1-9] and [7-3-2-8], it helps to see that the inner low numbers have two X-V relationships with their neighbors, and the outer high numbers have only one X-V relationship. That made it easy to write in the long sequences. A 4-cell sequence can immediately be written as [69] [14] [14] [69] or as [78] [23] [23] [78], depending on the color. Loads of pairs emerged. At some point the coloring didn't even matter.
This is essentially how I went about it - color in 1-4-6-9 vs. 2-3-7-8. Got all the colors except four cells that needed a break by Sudoku, then started on numbers
Went to look at his solution after and was blown away that we both built around 2 colors but for *completely* different purposes.
There is a very simple rule in this puzzle that will make the solving much faster. Every time there are 3 or 4 connected cells in the same box/row/column, the middle cell(s) must contain low digits.
Yeah I spotted that logic too, what I did though was to look at all the chains as blocks of three. So with a chain of four cells I would look at the first three, then the last three. Five cells would be first/middle/last etc. So I just thought of it as "In any chain of three cells constrained by normal sudoku rules, the middle cell is low."
yea and filling those digits insteed of playing with colors makes solve much faster
Surprising that Simon didn't see it abstractly, since he used it when given concrete choice of high or low.
Someone should make a puzzle that hides all the clues in columns and I don't think Simon would be able to solve it.
What do you mean?
I always wonder if someone gave Simon a puzzle with some of the fancier constraints, killer cages, thermos, looked like it needs ring\set etc but the only way to solve it was std sudoku, how long do you think it would be before he noticed?
Simon’s thought process is both amazing and highly comical at the same time.
This is the kind of puzzle I want an entire book of! Fun, approachable, but unique. I really appreciate the puzzles that require some basic maths, they're impressive, but I really enjoy the anti-chess sudokus and ones like this where it's all more basic/visual logic
3:07 I'm completely lost from the start; why do the zero and nine square immediately become a 1 and an 8, respectively? I'm not familiar with the small-starting-number notation. What just happened?
I know this is quite late but I had the same question and figured out the answer so I figured I would put it here in case anyone else wondered in the future.
It's basically the every clue is one before or after rule, since those cells have a zero and nine hint in them then they must be one at eight, respectively.
@@knightry ohhh damn thank you very much, I didn't understand it neither
I found myself yelling negative constraint throughout the whole video. Lol
Hahaha me too, once I notice one I just can't pay attention to anything until he resolved it
You mean the one at the top of box 6? I don't see another. (not watched the solve yet but can only be 27 or 38)
Is it proven that the negative constraint was necessary for uniqueness? I didn't see him use it until very late when the puzzle was so far done that it might have been finishable without it.
@@muskyoxes i wondered this as well. Maybe if he (or another solver) gets a chance they'll tell us.
@@jaunusender6166 i solved it completely forgetting about the negative constraint, and got a different solution. Of course I may have made an error due to that... But I triple checked and didn't see anything wrong
Tried it myself and it really was a blast! And that makes it even more enjoyable to see you solve it :) Keep going! love it :)
If any digit is between two Ws that share the same box, row, or column, it has to be a low digit. If it were a high digit with any of those properties, then you would have a duplicate low digit in the same box, row, or column. I am often slow on these puzzles, but I managed to complete in 22:35 by realizing this restriction.
I noticed this too! First time I've been able to solve a ctc puzzle without getting stuck
I noticed this right after I filled in the 5's. The first time ever I got off to a faster start than Simon. Instead of colors, I just filled the boxes with 1234 or 6789. then calcelled out numbers as I found them elsewhere.
In German, starting k is never silent, so its pronounced like the c in "cup": "cnup". You got the vowels right, though: "e" is either like the e in liver, or it sounds something like the "e" in "sire" or "higher", but not like "a".
And in case anyone was wondering: The translation is "near miss" or "just off the mark", but is frequently used in germany in similar circumstance as "close, but no cigar" (although the full saying would be "Knapp daneben ist auch vorbei" which means "a near miss is still a miss")
His excitement brushes logic aside sometimes and we get wonderfully crafted Simon-isms like "That doesn't butter any parsnips."
For me the key idea for solving a puzzle was that only low digit can have two 'w', which works a lot faster than the way Simon used
Same for me. First time I beat Simon timewise. Without explaining, for sure. And only possible in an easy puzzle.
You have to remember that only holds for cells that are also constrained by normal sudoku rules. That's how I approached it as well though.
Yeah, and the type of pairs need to alternate if they are constrained by sudoku, so a HL pair has to connect to a LL pair, meaning you could also never have a 5 chain in a box (alternating means we get HLLHH or LHHLL, can never have two H next to eachother, and never 3 L's)
Same here. You can immediately see this pattern: _ W _ W _ W _ all over the grid. Now, only the V clues can touch. The X clues need to be on the outside, so that pattern is ALWAYS _ X _ V _ X _ . This means you get repeating patterns of " High X Low V Low X High". The interesting bit comes when you have a run of three (as in Box 7 and Box 9). Since you can always have V clues connected, it means the run is either High X Low V Low or Low V Low X High.
Once you recognise this pattern, you are able to colour most of the grid, and then the V clue between R5C4 and R6C4 must be a 14 pair, as neither high digit on either side can be an 8. And then you are cooking with gas, as half the puzzle disambiguates when you apply the given 1 from box 6.
Got this one done in 13:08, this is the first time I've solved something almost 7 mins faster than Simon (yes, I know he's live-solving with commentary, but still). What worked out for me is that a chain of 3 connected cells in the same box/row/column means the middle one needs to be small (1, 2, 3, or 4), and a chain of 4 in the same box/row/column should thus have both middle cells be small.
One bit of interesting logic that I've not seen anyone mention in the comments is that in column 1, once you've put your 5 in there, you're left with 4 'W' dominoes, which means that each one of them must be a high-low pair adding up to 10. It isn't massively useful but it did give me a couple of digits. (Sorry if this has been mentioned and I missed it)
I made use of this too. Coloured dominoes purple and red if knew they contained one high and one low (as I've seen Simon do in the past).
Yup, I noticed that too and used it in my solve. At the point where I was using it, I had enough other cells filled that it gave me most of the column.
Same thing applies to row 3, yes.
13:04 "that's quite good" - ok, he sees it. 13:14 "if this can't be 2, this can't be 3 and we have a 1/4 pair" #FlipTable #ScreamAtScreen #PointAt1InTheSameRow
It was literally one of two non five's in the grid. How did it miss that so long? Then again he solved it faster than me!
27 minutes! I feel pretty good about that time considering I don't usually find a break-in for 95% of the puzzles featured on this channel lol. This one was fun!
The key logic to this puzzle is that "for any chain of length 3 in a row, column or box, the middle cell is low". Simon saw it right away for one cell at 6:00, but didn't follow through with it over the whole grid. Boxes 3 and 7 could have been solved much faster with that.
Yup, and a small corollary is that a chain of length 4 that can all see one another has to be high low low high. Very useful stuff.
I'm not sure it's "key" . I almost never finish a puzzle faster than Mark or Simon. I did finish this one quicker though, without doing anything clever other than quickly cancelling candidates that don't add to 5 or 10. Your fact about chains is certainly true but I never once thought of it.
Yep. I was screaming: "it is between two w" at the screen all the time. The other two in the triplet are then low and high.
19:06 - I've seen ambiguous XV before, but never this amusingly presented. A definite W to berni on this construction.
@5:47 does he already ignore the fact that the W could indicate a sum of 5 or 10? He immediately says if r2c5 is a 1 then r2d4 "has to" be a 9, when it obviously could be a 4.
That confused me as well.
OK, he then worked it out a few minutes later.
Yeah, he instantly lost me. Too confused by the workings of this one to wrap my head around it, because I spent that time going "but why? It can equal 5 too".
The chains in the puzzle is crazy. I was actually laughing while solving this XD It was such a good puzzle
28:17 One of the most fun puzzles Ive ever done. Started off color coding the whole middle square, and then slowly expanding the coloring until it got cancelled out by the 8 and 1 given. Would have been much faster but for some reason I decided to keep playing the XV game for a lot longer than I should have, when I could have just moved onto regular sudoku. But I guess I had more fun my way.
34 minutes without the video, finally a puzzle I can solve!
Me too, 34 minutes 05 seconds, hooray!
If I solve puzzles from these videos in less than twice the time it takes Simon I consider it a win. This one took me 47 minutes. Win achieved. :)
This one was super fun!!!
love these easier but still creative variant puzzles, the harder ones still mostly take me hours or even days to solve
Lovely puzzle, and managed to finish it even without realising the negative constraint.
Wow, I managed to finish this one in 20:33, which is the first time I've even come close to finishing one in less time than the length of the video, much less actually done so. Quite enjoyable indeed!
And I see from another comment that I actually beat Simon's solve time by 11 seconds, not just the video time. I didn't do any coloring; the only markings I used were center numbers -- which ended up being lots of 2378 cells, or subsets of those. About halfway through, I had everything except 2, 3, 7, and 8 completely solved, but had only one 7 on the whole grid, and none of the others!
Thanks to you I'm now an XV addict. I can solve the hard ones on sudoku central (about 5 digits given) in 15 minutes.
Wow, 9:38, wasn’t expecting this to go this smoothly, all the videos and solves are paying off
Hi Simon. Something I don't think anyone noticed: there's an extra bit of logic with this high-low system if you ever needed it. We know there are 81 cells total in the grid. If you know the locations of all the 5's like in this puzzle, and if you had a LOT of one color type cell, counted them and found the total to be 36, you could immediately color all remaining non-5 cells the other color.
That's a great tip about colouring the squares. Thanks for showing us.
Very fun puzzle and surprisingly simple for its ruleset. I even forgot about the negative constraint and completed it smoothly in 30 minutes.
This was probably the easiest solve i had on this channel so far, but also so much fun. I started coloring the chains because you can always only combine 4 numbers with each other in a chain of w's, and then gave secondary colors to the high and low numbers, my sudoku was a whole coloring book. colored sudokus are my favourites
15:21! Late to the party but still commenting because I'm proud of myself. Not the fastest, but the first time ever I've been faster than Simon at a solve. Same reason others have stated - realizing straight away that if a cell has 2 W's connected to it in a situation where both ends have to be different digits (so all in a line or all in the same box), they need to be one X clue and one V clue to avoid repeating a digit. Therefore, those cells must always be low digits because high digits can't be part of V clues.
After solving it I watched Simon's solve, and I didn't use the high-low idea; there was a plethora of 2-3-7-8 cells and a few 1-4-6-9 ones, but they all sorted themselves out with the help of the negative constraint.
the given 8 in R9C4 means that the 13 cell chain must be a 1-4-6-9 one. which forces the seven cell chain to be 2-3-7-8. Also liked that C1 had to have all the pairs adding to 10.
the k is not silent in German, so it's pronounced like cup with n cnup. It should be obvious by the rule "knapp daneben" means close miss.
Mir wäre niemals eine so gute Beschreibung für die Aussprache eingefallen- good job :)
@@not.rebecca1579 Well, actually I couldn't think of a word beginning with k. It's logical, that Simon thought of a silent k, as it is silent in so many words like knight, knot, know,, kneel, whatever. You have to come up with names like Klein in Klein bottle, so it's also pronounced as a German pronounces k. English also has ck has a hard k, harder than c, but no words begin with ck, not even ckracking the cknyptic.
I love the German phrase, Knapp ist auch daneben. Close(almost) is also a miss.
@@LednacekZ Yes, I leaned close, but no cigar, but it isn't the same.
@@OlafDoschke "K" is not silent in "whatever" - mainly because it does not contain a "K". Things you learn on CTC comments.
That made me smile. It's fun to do a puzzle, check the solution, and see that I did the exact same technique for coloring high/low numbers.
1:13:20! I actually finished one. I got to something like 52 minutes and saw that I had like 3 6s in some column and didn't want to unwind. Reset and remembered the tricks for the groups of 4 in a row (2378 or 1469), and then went zoomy fast!
Fun fact: '"w" is the Japanese equivalent of '"lol". The word for "laugh" is 笑う, or _warau,_ which gets abbreviated to just "w" in Japanese "text speak"
This puzzle made me realize that your app does not expect one and only proper solution, but rather checks out standard Sudoku rules.
When you are left with all the 23s and 78s to finish the puzzle, you can actually enter them both ways and end up with 372 in the bottom row. Breaks the X constraint, but you still get 'Looks good to me!'
Alternative start after 5's: column 4 has 3 W in it and the 8 is already placed. It is impossible to make a 8-2 pair in this column. Either the 2 of this column has to be in row 8, or it has to be linked up with a 3, in which case there is no 3-7 pair and the 7 is in row 8. Because of the negative constrait, a 2 can't be in row 8 because then there would have to be an additional W. Since there is not, there is in fact a 2-3 pair in the column, and only a 7 can be placed in row 8. Which means that in row 7, the four linked cells that need 2 high numbers can only be made with 1469.
I then proceeded to colour cells on whether they were part of the 1469 pairing, or the 2378 pairings, using central notation to find whether they're small or high numbers. Solved it in 40 minutes this way. Amazing to see this puzzle has such a different solve method available as well :O
Wow I got that in 22min, that really was approachable. The run of 4 digits above the 8 appears to only have a single solution constrained by the givens so I started there with quite a few digits which was very helpful.
I didn't believe the title, but after reading the rules, I opened up the puzzle and let out a chuckle when I realized what just happened.
Dodgy logic at 6:00 ... there's nothing to say that *that* cell can't be a 1, because it could be paired with a 4 - from the logic so far. Obviously that does break when you continue the chain.
What you can say is that any time a cell is connected (by Ws) to two cells that are within the same domain (box, row, column) then that 'middle' cell _must_ be a small number. Here we go parity-colouring!
(Completed in 28 minutes 👍🏻. And yes, I did laugh 😆. Love it!)
Got this done in 20:06, I easily filled in the 5s and then highlighted all the cells that couldn't have high digits due to repetition in column, row or 3x3 (Essentially part of the highlighting at 7:11 but for the whole grid) and that completely cracked it, other than the 2378s at the end, at which point the negative constrain finally came into play. Fun stuff
I did all the puzzle without the negative constraint (cause I forgot it existed :P) and it was still possible :)
@17:16 "that's not going to butter any parsnips." 😂 My new favorite British phrase
loved your solve, mine was along very similar lines.
just a small difference in the way i started: i noticed that if there was a chain of 3 cells (all seeing each other) connected by W's then the middle cell always had to be a low digit. this made me get a few low-low pairs in a W-domino, which meant that W had to in fact be a V-domino. i then confirmed which cells had to be high digits by noticing where low digits couldn't repeat in the relevant R/C/B.
Holy crap, I've never done this well in a puzzle vs Simon's time before. It took me only 9:39 because I immediately spotted where to break into the Ws which gave a ton of data off the bat.
Me too! First time ever in
35:33 for me. I missed the little 9 and 0 so I had the whole grid color coded, each number had its own color. Once I looked back at the video and saw the 9 and 0 I was able to solve it, but I spent a good 5 minutes wondering what I'd done wrong. Fun video
Yep same here but caught on to the 9 0 clues when they corresponded to the high/low designation in their cells. I found it quite interesting doing it this way anyway.
This is an amazing puzzle - I absolutely loved it!
It's very handy to draw a graph (in the sense of discrete mathematics) with vertex set {1,...,9}, and with all the W-relationships marked as edges. There are six edges in the graph. The graph has three connected components: 5 (an isolated vertex), and two chains, 9-1-4-6 and 8-2-3-7. The latter both go H-L-L-H (H=high, L=low), so if you ever find a chain of length 4 within a box or row/column, you know the pattern is H-L-L-H. Similarly, if you find such a chain of length 3, it must be H-L-L or L-L-H, so in particular the middle cell is L. This helped to quickly colour most of the grid.
12:35 when you understand the column 4; that help a lot. Nice one. Just have to understand how beginning. Cool puzzle.
58:00 exactly.
I colored the grid blue and orange to figure out where the low digits were and where the high digits were. And I couldn't get past the 2/3 pairs in boxes 4, 7, and 9. No matter what I tried they broke. I was about to give up and I decided to delete all the digits but leave the coloring, then I started filling it in and questioned each color. I finally found 6 cells that could be the opposite of what I had. I changed those and finished the puzzle. The first time I hit check was at 33 minutes, so I would have been happy with that time, had I not messed up.
Oh, silly Simon. You got the colors backward. Orange O is for over, and Blue B is for below. :)
I say that in jest but now that I'm watching the video longer, it's really messing with my head. I keep thinking you messed up in the logic.
@13:30 in the video. I'm screaming at my screen. "But you have a 1 in row 6 to fix the 14 and thus all the cells in that chain."
@14:20, I see that my deduction that 3 in a row or in a box, the middle one always has to be low would have helped you there.
16:35 first time I've ever been quicker (of the limited number of times I actually was able to get started lol)
I found that every three w's had to be high-low-low-high very quickly and started eliminating options using the 1 and 8 instead of coloring. This gave a lot of numbers in the grid quickly!
This was a very fun puzzle. It did make me smile when I got near the end and realised the 'key' to cracking it lay with the very first number (1) that I put in the gird.
Nice construction.
I actually finished this one before watching the video! :) Now I get to see the easier method I missed.
Many thanks for choosing a colorblind friendly color pair!
Delightful puzzle. Ended up proliferating the whole grid with 23s and 78s until the negative constraint down around boxes 8 and 9 sorted it all out in pretty much a single fell swoop.
27:19. Happy with my time. Agree that it helped to look at the groups with four W's (H-L-L-H) or three (L-L-H or H-L-L). This expands to the big chains of 7 and 13 cells which cross box boundaries and so are kept in chunks of four or less in each box. Also you can place all the 5's immediately. Haven't watched the video yet so I'm now going to see how Simon solves it.
This was a fun one to solve. I looked at the chains and figured them out before even touching the fives. And because I went straight to pencil marking, I didn't really use any colors.
What a nice puzzle! I somehow managed to solve this one in 18:21 without any help from the video, so I'm quite proud. :)
Yeehaw! With a 20:00 solve this was my first ever being faster than Simon or Mark, even staying below Simon's actual solve time (from "Let's get cracking" 3:05 to "Yeah!" 23:49), not just the length of the video. And how nice it came out to 20:00 flat. Largely due to what others have also mentioned: realising that cells in the middle of a chain of cells seeing each other and connected by W's have to be 1234. What also helps a lot is to always think of the two groups of interconnected numbers 1469 and 2378.
Took me an hour and a half but I ACTUALLY managed to solve this on my own BEFORE watching your video (apart from the rules explanation). But now it's 3:30am and I need to sleep so I will watch this tomorrow :")
15:58 my solve. I'm normally not quick at solving sudoku's but I think when I realized that only small numbers could go where 3 cells connected with W which see each other, the puzzle became quite easy
A really approachable puzzle... 10 minutes for me.. fun...
It did make me laugh, and then crush it thoroughly in 13:10. The chains of four cells where the middle ones need to be small makes it pretty easy.
Awesome puzzle! Got it in 37:42. Makes up for yesterday's debacle of a solve. :D
very easy solve once you figure out to check for high low parity. very clever and very approachable.
Interesting to see the break in with the colours working out the high/low options.
I started by spotting the cells with 3 connections r6c4, r7c5 - to have 3 connections, they must be a 2/3/4. From this started filling in options along the chain until I hit a restriction then following the eliminations. My first digit (after the 1, 8 and 5s) was the 6 in r7c4
Took me 34 minutes 05 seconds, but I'm proud to say that after your explanation of the rules and placing the first 8 and 1 I was able to do it all by myself :)
All chains of 4 that are in the same row or the same box must be either 9-1-4-6 or 8-2-3-7. Every time.
I'm here for that "yeahhhh" at the end
as a german, hearing you pronounce "knapp daneben" really made my day :D
I didn't see the 0 and 9 clue before solving it, they don't show up in the app so I was quite surprised to see them when watching the video afterwards. I ended up using 8 different colors identifying each number and "solving" it without knowing what the actual numbers where. I just knew that "these 9 purple squares has to be the same number, these 9 blue squares has to be the same number and so on". Only the 5's were identified. First time I've ever identified all the digits by color but not knowing which color represented which digit, all I knew was which for colors were higher than 5 and which four colors were lower. Give it a try by ignoring the 0 and 9 clue, it's fun and end up looking pretty.
I also couldn't see the 0 and 9 clue and there were multiple solutions in the end.
The 0 and 9 clue are only shown, when you open the "legacy app".
I focused immediately on the intersection of the "given" 1 and "given" 8, so coloring hi-lo was unnecessary since I knew from the start it was 14 or 69. As that chain goes into boxes with other chains, it tells you what they are too (23-78 for the most part). Simon dances all around the grid coloring hi-lo before honing in on the one spot it's disambiguated. One of the few times I feel I followed the more productive solution path. I'm usually amazed at his ability to anticipate the next easiest step. Of course that wasn't really essential on this puzzle.
I did a different approach first. I used 2 different colours for 1469 and 2378. Then I looked carefully about 1 and 8. Last, I focused on high / low digits. I had to count 4 green and 4 red in boxes, columns and rows.
I normally will hear the intro - hear the rules - and then dive in to a puzzle on my own to see how far I can get before watching and feeling silly for missing some obvious bit of clue.
After the '5's, and realizing that there must be a 'high low low high' pattern, I actually solved this in a respectable time for myself (sub 30 minutes) - what a great puzzle that falls into place once you realize the constraints and patterns. (ie: 9 / 1 / 4 / 6 for example)
I found it easier to color high/low but to assign the center box cells a letter. I was able to nap all the letters around the grid and then numbers as they came along
Once you see a cell surrounded by Ws in the same box, row, or column can never be a high digit, this puzzle is simple! Nice!
And that the surrounding cells must be opposites, i.e. one high and one low, never both high or both low. Just watch out for boundary crossing when looking at ones in a 3x3. Also be careful in a few places to note the negative constraint as if you go fast, you could slip up. Cheers!
I totally used colours too! And felt the need to colour it all in before clicking check. Though I also coloured the 5s grey. And took 5 minutes longer than you even if I count your introducing-the-puzzle time...
I found an easy break in at the box with 3 w's (r6c4).
It can only be 3 or 4 because off the 1 and 8.
If you work the possibilities out for the adjecent cells you find that it can only be 4.
From there you can fill in all the numbers.
29:40 for me. I didn't do any colouring, just started looking at the consecutive W relationships in the middle box which are heavily constrained by the initial digits. It really quickly snowballs from there (well, quickly if you're cleverer than I am. I went slow and methodical to make sure everything works as it should because I'm not great at this).
I solved this but had the 7's and 8's swapped. Took a while of checking the grid to realize there are two hints (the 9 and the 0) shown in the video that aren't on the app! Without those two hints, this puzzle has more than one solve.
The best start is box 5.
We already know where the 5 is, and the sum of the three cells on the right hand side must add up to a multiple of 5 given the placings of the Ws in the rest of the box.But They cannot add up to 5, 10 or 15 because there would need to be a W between the two cells below the 5. So, the three cells must add up to 20 and the two cells below the 5 must therefore add up to 15. So, 69 or 78 and, given the position of the 1 in box 5, it is easy to work out. The rest of the puzzle then links all the way through.
Really loved this one. And after trying it I can actually for once agree with you on calling something "approachable".
Loved it!
21:20, this was a really approachable. I found the logic that middle digit in a sequence of 3 or 4 must be low. That high digits always have low digits as neighbours. That colouring sure helped seeing where I could fill in the rest. Really like this one. But I started with filling in all the fives. Let's see what Simon made of this.
I don't know if the negative constraint is needed. This was a fun one, once I figured out high-low-low-high.
@CrackingTheCryptic Thank you again for another fun video! As a someone who is colourblind seeing as you mentioned it in this video I'd like to suggest something: while the colours used in this video are easily enough to tell apart would it be possible to include stronger colours in the app? I have a stronger version of the "red/blue colourblindness" which and most people I know who shares this type of sight tend to like strong colours over the softer ones currently used in your app as they blend together very easily.
It's just a suggestion of course and this opinion may not carter to the majority but I'd like you to know that there is a demand for stronger colours :)
once you realize how a 4-connected box works with >5 and
This was really fun to do. I always like it when the rules make me use colours. 🎨🖌️
(Although, I have to admit that Simon messed me up quite a bit when I was watching after solving, because he coincidentally managed to exactly inverse my colours for high and low...)
This is actually approachable. Took me 1h 45min, but that may be just me…
The 5s are all placeable pretty soon. Then column 1 and row 3 must be full of Xs. Then I just placed all the “high” and “low” digit pencil marks wherever an X or a V must be, always watching for negative constraints, counting the number of high and low digits per row, column, and box, and taking into account consecutive W clues.
But it took me a while to also consider how many X and V dominoes there can be per box, e.g. in box 7, where the two bottom left Ws both have to be X, and exactly one of the top right ones has to be a V. The 1 and the 8 that you placed at the start constrain lots of dominoes, and standard Sudoku rules do so, too.
At some point you start getting actual digits in the grid; then it’s all pretty much solved.
Watching the video, I don’t know how I missed the quite obvious breakthrough around 12:00.