Mistaken logic at 13:28 the arrows add up to 8 each when the number in the circle is 8, not 7. Instead, realize that if 8 goes in the circle then the only positions that 3 and 9 are able to go in in column 7 is r5 and r8, but neither fit in r8 due to the arrow sum.
I took me a while to solve that puzzle, including quite a bit of bifurcation. When I watched the video to see how it's done without too much bifurcatino, it was quite disappointing that the video didn't provide such a way due to the wrong reasoning there. Thanks for pointing out the correct way. But the rest of Mark's solve is also such a mess that it didn't help me to see a really straightforward path.
Thank you, Mark, for doing my puzzle and for the well wishes! I wanted to build in some fairly advanced classic Sudoku logic into the lower digits by giving plenty of interesting information about high digits. I managed to get both a swordfish and a kite-string in there (hence the “fishing” part of the title). Mark was very close to seeing the swordfish, and could have done so had he placed a 2 in box 7. It’s interesting to see the variety of solving times and reactions!
Hi Scott. Great puzzle, I especially liked how the ‘fish’ interacted with the cages. I left quite a long comment a few days ago with an attempt at a solve guide as I’d noticed people were getting stuck because of Mark’s errors and the difficulty with what I believe was the last major step in the solve. So I’d be interested to see how it compared to your intended path (I don’t actually know what a kite-string is but I did find several jellyfish in there, including a finned one).
I am not happy with explanation given at 13:34 since it considers pairs of numbers that add to 7 on the arrows but this entire deducion is based on 8 placed on the bulb of the termo at r7c7. After taking out 789 from the box, remaining digits add to 21 with 2 arrows adding to 8 so r7c9 + r9c9 add to 5 and there is no conflict with the 7 cage above.
I went back from the end of the video to see what error led to incorrectly removing that one pencilmark. Going from around 31:40 to 33:40 Mark notes where the 1's could go IF there was a 1 in r4c2... then notes somewhere that a 1 cannot go UNDER THAT CIRCUMSTANCE, and treats it like that applies regardless of circumstance. Needless to say, when r4c2 turned out NOT to be a 1, that left this deduction completely baseless and, as we discover in the end, it was in fact false.
13:10 I cannot follow this logic He starts with the hypothesis that the circled cell is an 8, but then a few seconds letter his reasoning says the arms add to 7 instead of 8? If you look at box 9, placing an 8 in the top left leads to a 7 in the bottom left, a 9 in the middle and a 14 pair in the cells not on the arrows. From there, I do not see how this breaks.
Better logic is below... **Spoiler alert!! If you want to try to figure out it first, check this later!!** R7C7 indeed is an either 8 or 9. BUT, look at where else 9 can be in row 7 and column 7. If R7C7 is 8, it forces 9 into R7C5 and R5C7 by basic sudoku. Then, that forces 9 into R3C9 and R4C3 to satisfy the cage rule. NOW there nowhere left to put at 9 in the 14 cage in row 9! Therefore, you can mark 9 in R7C7 and carry on.
Mark's logic was clearly wrong here for the reason stated above. I waited for him to see his error, but I don't think he ever did. Right conclusion but for the wrong reason.
@@adamdobrowolski2510 i used completely different logic. because two ,3s are forced to be in 2 of the 3 rows in columns 3 and rows 2, 5 and 8. That meant there was only room for 1 more 3 in rows 2 5 and 8 in columns in 6 and 7. so it forces a 3 to be in the 15 cage. which makes it a 1239 cage and puts the 9 in the circle. Edit: P.S. lol your thinking i give up a lot quicker because by the time you put the 9 in r4c3 i would gave up going further, because after that many steps my brain would be fried doing that all in my head.
A puzzle that requires such a huge amount of finagling, it's unsurprising to me that some errors occured. What deserves special praise is that this video is here at all. Many lesser channels would hide their mistakes under the rug. We all do the undoing when solving away from a spotlight. Given the fearsome puzzles you've solved, you're more than allowed the odd logical oops.
Mark was so close with his logic at 37 minutes. The problem was he’d failed to highlight the cage containing 4 in columns 6 and 7. If he had, he would have discovered a finned jellyfish (4 columns, 3 rows and a cage, plus the finned digit in row 7 column 1) which forces row 8 column 2 to be a 5. It’s a beautiful step but it’s very difficult to spot, so it was a great effort.
I finished in 127 minutes. That was one of the hardest finishes I have ever done. The beginning I was soaring through. I spotted the many x wings and was cruising. My production slowed down a little, but I felt good. At about 30 minutes in, I became hard stuck and just couldn't progress. Then, I spotted a brilliant trick and noticed that the arrow in box 9 couldn't be an 8, due to a 2 already ruled out, limiting it to 1357, but putting an 8 in would also force r9c7 to be a 7, breaking it. That was discovered at about 60 minutes in. I struggled for the rest of the time. I just couldn't see anything. Then, I finally got lucky by looking at r2c8 and seeing if that was limited. Surprisingly, both 6 and 8 break the puzzle in the same way by putting two 7's in r7c2&3&4 or two 4's in r6&7c4 depending on direction. If that is the intended way, then I have no idea how to intentional spot that, other than stumbling upon the correct cell to see. The beginning was so nice, but the ending almost killed me. I enjoyed it, though. Great Puzzle!
A very enjoyable video. Mark’s ability to recover from mistakes is amazing and fun to watch. Mark’s ability to comment his thought process while solving is uncanny. Also, I thank the people who comment on oversights Mark doesn’t catch at the time, and provide alternative ways to get the solve back on track.
Mark never notices that col 3 contained pairs 1,3; 1,2; 1,3 for an easy placement of a 2 in col 3, and then a 2 in col.4. At this point it solves quite quickly.
The 2 in box 4 was obvious for a much longer time. When ruling out 2 from the yellow cells, AND a 2 pointing down in col 2, there was only one cell left in box 4.
After I zoomed at the start and felt like I had made a lot of progress in only 8 minutes, it ground to a halt and I never found where to go next (I'm 90 minutes in with no new deductions in 82 more minutes and Mark made incorrect deductions in the video around the same point so watching is no help to get me unstuck this time). Dropping this one, it has defeated me.
as a long time follower of this channel and a big fan of your videos Marks. I don't understand why you wouldn't invest in a silent pc or at least in a proper microphone setup. The videos you present are a joy to watch and the solution approach is often stunning. Bun listening to it is a way to ear-cancer :/ Please please.....the solution is such a small investment
Is the solution to this meant to be based on "This is not quite an X-wing / Swordfish /Jellyfish, because this rogue cell could still contain digit X and that would spoil the pattern. But if that _wasn't_ X, the ${scary sea creature} would definitely be true. So we can still eliminate anything that would definitely be ruled out in either case" ?
I missed to placed one digit in the arrow of box 7, and made it almost till the end. When everything got wrong. So I could understand why it take you too long to solve this one. Not an easy puzzle! Thanks for that Mark! 😊
To anyone that was confused about how to solve this puzzle, especially due to Mark’s solve having a few errors (it’s easily done), here’s an attempt at a solve guide. Start by filling in all the restricted cages e.g. the 3 cage in row 1. Then consider what digits are used up in some rows/columns and how that limits other cages e.g. the 17 cage and 29 cage take up all the 8s and 9s in rows 3 and 4. Remember to combine killer cage totals if necessary. The 2 cell cages that add to 14 and 15 can instead be considered together as 4 cells that add to 29. You should now have all the cages filled except for the 7, 12 and 15 cages in rows 6 and 7. This is where the puzzle title comes in. Very often the title is a hint to what you need to look for in the solve and this puzzle is no exception; we’re looking for fish i.e. swordfish (3 rows and columns where a digit is restricted) and jellyfish (the same but 4 rows and columns). It should become apparent that 1s, 2s and 3s are quite restricted, which forces a 2 into the 7 cage and all of them into the 15 cage, as well as placing a few others around the grid. Now comes the trickiest bit. The final fish is in fact a finned jellyfish on the digit 4 (it is almost a jellyfish but for 1 possibility that adds an extra row/column) and is difficult to spot (if you’re struggling, focus on where 4 can go in columns 1 and 9 and use that as a guide). Once you have found it, remember that either the fin must be true or the jellyfish must be true, so you are looking for them to force the same result somewhere in the grid. This should reveal the last remaining cage and from there it is just a matter of sudoku. It’s a clever puzzle but the 2nd half in particular is not easy, so hopefully this will help in some way. Feel free to ask if you need further pointers, I’ve tried to provide just enough information to solve without the risk of people getting stuck for ages.
Wow, Mark, I am so impressed! When Scott first showed me this puzzle I took about 2 hours to solve it, some deductions you made were in seconds that took me half an hour. And I (when I asked Scott for a hint of what to look at next at a certain point) got told "look at the title" which was mystifying to me. Bow fishing? (But I knew it was a bow and arrow kind of bow ... but the 'fishing' word did not make sense to me for quite a while.) Well, needless to say, I was fascinated to watch you solve this puzzle of Scott's, and thank you for wishing him a happy birthday! I was watching this video so late because I had a meeting this evening, and it is possible he has not even realized that you did his puzzle tonight! 🎊
A "Swordfish" is the 3-by-3 analogue of an X-wing, followed by a series of increasingly-fiendish ocean creatures as more rows and columns are introduced ..... Mark's careless miss of why R8C8 could not be 9 probably meant he also missed some other things that would have shown up in the course of proving it, and made or rubbed out a pencil mark that could have come in useful later on. I think the proper technique is to notice that there are some "near-miss fish" where there is an extra square in one 3*3 box that could still be the digit in question, and if so would kill the fish; but some eliminations are still possible in either case, whether the extra square is the "fishy" digit or the fish pattern exists.
@@bluerizlagirl Yes, there is an extra square in at least one place that dissuaded me from looking at it as a swordfish. I will have to give this some thought and maybe try Scott's puzzle again with both Mark's solve in mind as well as more of an eye toward finding swordfish. Thanks for the tip!
I really liked the first part of this puzzle, and then it got to the harder stuff and I wasn't a super fan of the rest of it. Totally wasn't expecting combining combining the 1/3 in the killer columns with the 1/3 in columns 1 and 7 to cause a hidden triple in column 9 to be part of the solution path. And I'm still not really certain how to disambiguate the 45 in box 7 after.
I’ve left a ‘solve guide attempt’ comment which you may be interested in. I believe you had to find 4 jellyfish, with the last one being a finned jellyfish (if there is such a thing!) that neatly resolved the 4,5 pair in box 7. I initially thought the same as yourself but mainly as my first solve was a bit stop start. The actual solve path is very smooth and took just under half an hour at speed (so probably 45 minutes to an hour at normal pace). It’s also well signposted by the puzzle title.
Mark's logic about the 8 or 9 in box 9 is incorrect. If you plug an 8 in the circle it does not break because of the 7 cage in column 9. It breaks because of the 15 cage in columns 6 & 7. Mark got away with a lucky guess.
13:30, an 8 in R8C1 would force a 1 into R5C1, and a 4 in both R4C2 and R8C2 (By this time, 2 must already be in R9C1, explained at 25:07). Therefore the 8 in box 7 must be in R7, and R7C7 must be a 9
23:23 for me, solver number 7. Made some silly mistakes that cost me some time: first I treated the arrows in Box 9 like a zipper and put digits in the wrong spots, then I had some bad pencilmarks that led me to erroneous conclusions.
Love how mark loves to pencilmark and colour But doesn't like to colour Lows(1-4) only 12 and 123....and then ignores the pencilMarking on B7arrow. His solves still great too watcj
I didn't have the timer going, took around half an hour on and off The cages are almost all forced, and then there's some clever logic with the small digits
If basic sudoku is beneath you (that is, checking row, column, box after each real digit put in), then why not use the conflict markers that your software provides?
Better idea: Get Sven to update the software to show the conflict checker markings for the viewers at home, while Simon or Mark have to struggle on without them!
Some time ago a viewer who is a psychologist, I believe, posted a helpful comment about the types of reasoning involved in solving complex sudoku (such as this puzzle). This person said that there are two sorts of reasoning going on, the surface reasoning of "since I have put a six here, I should check to see if it eliminates any pencil marks" sort of thing. What you have called "basic sudoku." The other kind of reasoning is the deeper kind, where the internal logic of the puzzle is being worked on. This is the level of reasoning that Mark (and Simon) are usually employing, because the puzzles they try to solve are quite complicated and there is minimal information supplied by the constructor (on purpose - this is the point of these puzzles). This viewer's main point was that it is quite difficult to effectively switch between those two levels of reasoning, and when Mark (or Simon) seem to be "missing" something that is obvious to us as viewers, it is because they are not switching to the surface level. We have the luxury of not having to be in the deep reasoning ourselves, we can just look around the grid as we please. So your disdain of Mark's not using basic sudoku is probably a misplaced criticism. Using the conflict checker and responding to it by updating pencil marks would have the same effect, pulling them out of the deeper reasoning into the surface, and their solves would be less enjoyable to them and therefore, by extension, to us. I wish that your comment had been less mean-spirited, by the way - it would have been possible for you to simply say, "I wonder why you don't use the conflict checker?" and leave out the "beneath you" part of your remark. There is no need to be mean about it.
What is all that nonsense around 13:30 about an 8 arrow adding to 7! It works perfectly well as an 8 arrow, at least in that part of the solve. When this happens I skip to the end to see if the poor logic got the correct solution. If not I go back and continue watching. Otherwise I go do something else. Yep. I’m off to do something else!
I give up on this one. It's fine but then everything grinds to a halt. Any progress I make is through trial and error. I spent fifty something minutes, decided to restart, and then went at the same damn pace with no improvement for another fifty something minutes. You know what? This puzzle is BAD, actually.
It’s a tricky puzzle but it’s actually incredibly cleverly put together. I would highly recommend you give it another chance but this time consider a change of solve approach, as I believe this is the reason for things grinding to a halt in the middle (I experienced the same on my first solve). I’ve left a comment with an attempt at a solve guide where I’ve tried to provide enough information to progress with the solve but still enjoy finding things yourself. You’re of course welcome to say it’s one of those sudokus where you’d prefer to forget about it but it’s definitely not a bad puzzle or solve path, it just takes understanding the hint in the title and a bit of solving experience.
Not my best solve... made decent progress in the beginning, then ended up bifurcating to find a contradiction, which did allow me to complete in 33:19 (conflict checker off). Many thanks to Scott for a cool, if a bit brutal, puzzle!
Mistaken logic at 13:28 the arrows add up to 8 each when the number in the circle is 8, not 7.
Instead, realize that if 8 goes in the circle then the only positions that 3 and 9 are able to go in in column 7 is r5 and r8, but neither fit in r8 due to the arrow sum.
yeah that threw me
Task failed successfully
"That's amazing".
Well, yes, I was certainly amazed.
I took me a while to solve that puzzle, including quite a bit of bifurcation. When I watched the video to see how it's done without too much bifurcatino, it was quite disappointing that the video didn't provide such a way due to the wrong reasoning there. Thanks for pointing out the correct way. But the rest of Mark's solve is also such a mess that it didn't help me to see a really straightforward path.
Come on, as if any of us has never made a logical error that we got lucky with.
(EDIT: That's a reply to bibliopolist)
Thank you, Mark, for doing my puzzle and for the well wishes! I wanted to build in some fairly advanced classic Sudoku logic into the lower digits by giving plenty of interesting information about high digits. I managed to get both a swordfish and a kite-string in there (hence the “fishing” part of the title). Mark was very close to seeing the swordfish, and could have done so had he placed a 2 in box 7. It’s interesting to see the variety of solving times and reactions!
Hi Scott. Great puzzle, I especially liked how the ‘fish’ interacted with the cages. I left quite a long comment a few days ago with an attempt at a solve guide as I’d noticed people were getting stuck because of Mark’s errors and the difficulty with what I believe was the last major step in the solve. So I’d be interested to see how it compared to your intended path (I don’t actually know what a kite-string is but I did find several jellyfish in there, including a finned one).
I am not happy with explanation given at 13:34 since it considers pairs of numbers that add to 7 on the arrows but this entire deducion is based on 8 placed on the bulb of the termo at r7c7. After taking out 789 from the box, remaining digits add to 21 with 2 arrows adding to 8 so r7c9 + r9c9 add to 5 and there is no conflict with the 7 cage above.
I went back from the end of the video to see what error led to incorrectly removing that one pencilmark. Going from around 31:40 to 33:40 Mark notes where the 1's could go IF there was a 1 in r4c2... then notes somewhere that a 1 cannot go UNDER THAT CIRCUMSTANCE, and treats it like that applies regardless of circumstance. Needless to say, when r4c2 turned out NOT to be a 1, that left this deduction completely baseless and, as we discover in the end, it was in fact false.
Luckily he didn't use that part not being a 1 effectively at all.
Bow fishing probably means fishing with a bow and arrow, in which case you're using arrow sudoku to find swordfish and jellyfish.
Bow fishing, as in finishing with a bow and arrows. Hence the arrows and the two lined arrow meant to look like a bow.
I'm guessing he realised by the time he came to design the thumbnail for the video?
13:10 I cannot follow this logic
He starts with the hypothesis that the circled cell is an 8,
but then a few seconds letter his reasoning says the arms add to 7 instead of 8?
If you look at box 9, placing an 8 in the top left leads to a 7 in the bottom left, a 9 in the middle and a 14 pair in the cells not on the arrows.
From there, I do not see how this breaks.
Better logic is below... **Spoiler alert!! If you want to try to figure out it first, check this later!!**
R7C7 indeed is an either 8 or 9. BUT, look at where else 9 can be in row 7 and column 7. If R7C7 is 8, it forces 9 into R7C5 and R5C7 by basic sudoku. Then, that forces 9 into R3C9 and R4C3 to satisfy the cage rule. NOW there nowhere left to put at 9 in the 14 cage in row 9!
Therefore, you can mark 9 in R7C7 and carry on.
Mark's logic was clearly wrong here for the reason stated above. I waited for him to see his error, but I don't think he ever did. Right conclusion but for the wrong reason.
@@adamdobrowolski2510 How is 9 forced into R4C3? It can still go in R4C4, which puts a 9 in R9C3 in the 14-cage.
@@adamdobrowolski2510 i used completely different logic. because two ,3s are forced to be in 2 of the 3 rows in columns 3 and rows 2, 5 and 8. That meant there was only room for 1 more 3 in rows 2 5 and 8 in columns in 6 and 7. so it forces a 3 to be in the 15 cage. which makes it a 1239 cage and puts the 9 in the circle.
Edit: P.S. lol your thinking i give up a lot quicker because by the time you put the 9 in r4c3 i would gave up going further, because after that many steps my brain would be fried doing that all in my head.
@@sgtpepper1790 Not just _can_ go in r4c4, it _must_ go in r4c4 because that's the only place it can go in box 5.
A puzzle that requires such a huge amount of finagling, it's unsurprising to me that some errors occured. What deserves special praise is that this video is here at all. Many lesser channels would hide their mistakes under the rug. We all do the undoing when solving away from a spotlight. Given the fearsome puzzles you've solved, you're more than allowed the odd logical oops.
Mark was so close with his logic at 37 minutes. The problem was he’d failed to highlight the cage containing 4 in columns 6 and 7. If he had, he would have discovered a finned jellyfish (4 columns, 3 rows and a cage, plus the finned digit in row 7 column 1) which forces row 8 column 2 to be a 5. It’s a beautiful step but it’s very difficult to spot, so it was a great effort.
I finished in 127 minutes. That was one of the hardest finishes I have ever done. The beginning I was soaring through. I spotted the many x wings and was cruising. My production slowed down a little, but I felt good. At about 30 minutes in, I became hard stuck and just couldn't progress. Then, I spotted a brilliant trick and noticed that the arrow in box 9 couldn't be an 8, due to a 2 already ruled out, limiting it to 1357, but putting an 8 in would also force r9c7 to be a 7, breaking it. That was discovered at about 60 minutes in. I struggled for the rest of the time. I just couldn't see anything. Then, I finally got lucky by looking at r2c8 and seeing if that was limited. Surprisingly, both 6 and 8 break the puzzle in the same way by putting two 7's in r7c2&3&4 or two 4's in r6&7c4 depending on direction. If that is the intended way, then I have no idea how to intentional spot that, other than stumbling upon the correct cell to see. The beginning was so nice, but the ending almost killed me. I enjoyed it, though. Great Puzzle!
A very enjoyable video. Mark’s ability to recover from mistakes is amazing and fun to watch. Mark’s ability to comment his thought process while solving is uncanny. Also, I thank the people who comment on oversights Mark doesn’t catch at the time, and provide alternative ways to get the solve back on track.
You need to sort your sound out Mark. When you're not speaking, white noise builds up, until you speak again, when it stops!
The laptop fan needs a break😮
Lol wasn't sure if it was his or mine for awhile... I need new technology.
He also needs a better mic. Maybe we can do a gofundme: Better laptop and mic for Mark
Mark never notices that col 3 contained pairs 1,3; 1,2; 1,3 for an easy placement of a 2 in col 3, and then a 2 in col.4. At this point it solves quite quickly.
Solves quickly how? After I put those digits in, I'm exactly as stuck as before.
The 2 in box 4 was obvious for a much longer time. When ruling out 2 from the yellow cells, AND a 2 pointing down in col 2, there was only one cell left in box 4.
After I zoomed at the start and felt like I had made a lot of progress in only 8 minutes, it ground to a halt and I never found where to go next (I'm 90 minutes in with no new deductions in 82 more minutes and Mark made incorrect deductions in the video around the same point so watching is no help to get me unstuck this time). Dropping this one, it has defeated me.
as a long time follower of this channel and a big fan of your videos Marks. I don't understand why you wouldn't invest in a silent pc or at least in a proper microphone setup. The videos you present are a joy to watch and the solution approach is often stunning. Bun listening to it is a way to ear-cancer :/ Please please.....the solution is such a small investment
Is the solution to this meant to be based on "This is not quite an X-wing / Swordfish /Jellyfish, because this rogue cell could still contain digit X and that would spoil the pattern. But if that _wasn't_ X, the ${scary sea creature} would definitely be true. So we can still eliminate anything that would definitely be ruled out in either case" ?
70:02
Sublime genius.
I missed to placed one digit in the arrow of box 7, and made it almost till the end. When everything got wrong. So I could understand why it take you too long to solve this one. Not an easy puzzle! Thanks for that Mark! 😊
To anyone that was confused about how to solve this puzzle, especially due to Mark’s solve having a few errors (it’s easily done), here’s an attempt at a solve guide. Start by filling in all the restricted cages e.g. the 3 cage in row 1. Then consider what digits are used up in some rows/columns and how that limits other cages e.g. the 17 cage and 29 cage take up all the 8s and 9s in rows 3 and 4. Remember to combine killer cage totals if necessary. The 2 cell cages that add to 14 and 15 can instead be considered together as 4 cells that add to 29. You should now have all the cages filled except for the 7, 12 and 15 cages in rows 6 and 7. This is where the puzzle title comes in. Very often the title is a hint to what you need to look for in the solve and this puzzle is no exception; we’re looking for fish i.e. swordfish (3 rows and columns where a digit is restricted) and jellyfish (the same but 4 rows and columns). It should become apparent that 1s, 2s and 3s are quite restricted, which forces a 2 into the 7 cage and all of them into the 15 cage, as well as placing a few others around the grid. Now comes the trickiest bit. The final fish is in fact a finned jellyfish on the digit 4 (it is almost a jellyfish but for 1 possibility that adds an extra row/column) and is difficult to spot (if you’re struggling, focus on where 4 can go in columns 1 and 9 and use that as a guide). Once you have found it, remember that either the fin must be true or the jellyfish must be true, so you are looking for them to force the same result somewhere in the grid. This should reveal the last remaining cage and from there it is just a matter of sudoku. It’s a clever puzzle but the 2nd half in particular is not easy, so hopefully this will help in some way. Feel free to ask if you need further pointers, I’ve tried to provide just enough information to solve without the risk of people getting stuck for ages.
Thanks for this - I am going to consider your approach as I solve this puzzle again!
@@emilywilliams3237 You’re very welcome Emily, I hope you find it helpful. Scott did a fantastic job with this one.
Wow, Mark, I am so impressed! When Scott first showed me this puzzle I took about 2 hours to solve it, some deductions you made were in seconds that took me half an hour. And I (when I asked Scott for a hint of what to look at next at a certain point) got told "look at the title" which was mystifying to me. Bow fishing? (But I knew it was a bow and arrow kind of bow ... but the 'fishing' word did not make sense to me for quite a while.) Well, needless to say, I was fascinated to watch you solve this puzzle of Scott's, and thank you for wishing him a happy birthday! I was watching this video so late because I had a meeting this evening, and it is possible he has not even realized that you did his puzzle tonight! 🎊
A "Swordfish" is the 3-by-3 analogue of an X-wing, followed by a series of increasingly-fiendish ocean creatures as more rows and columns are introduced .....
Mark's careless miss of why R8C8 could not be 9 probably meant he also missed some other things that would have shown up in the course of proving it, and made or rubbed out a pencil mark that could have come in useful later on.
I think the proper technique is to notice that there are some "near-miss fish" where there is an extra square in one 3*3 box that could still be the digit in question, and if so would kill the fish; but some eliminations are still possible in either case, whether the extra square is the "fishy" digit or the fish pattern exists.
@@bluerizlagirl Yes, there is an extra square in at least one place that dissuaded me from looking at it as a swordfish. I will have to give this some thought and maybe try Scott's puzzle again with both Mark's solve in mind as well as more of an eye toward finding swordfish. Thanks for the tip!
I really liked the first part of this puzzle, and then it got to the harder stuff and I wasn't a super fan of the rest of it. Totally wasn't expecting combining combining the 1/3 in the killer columns with the 1/3 in columns 1 and 7 to cause a hidden triple in column 9 to be part of the solution path. And I'm still not really certain how to disambiguate the 45 in box 7 after.
I’ve left a ‘solve guide attempt’ comment which you may be interested in. I believe you had to find 4 jellyfish, with the last one being a finned jellyfish (if there is such a thing!) that neatly resolved the 4,5 pair in box 7. I initially thought the same as yourself but mainly as my first solve was a bit stop start. The actual solve path is very smooth and took just under half an hour at speed (so probably 45 minutes to an hour at normal pace). It’s also well signposted by the puzzle title.
Really like this one, great flow, finished in 15:48.
Are you sure you didn't make a lucky error? Quite suspicious to finish in 15 minutes when Mark struggles for an hour.
Mark's logic about the 8 or 9 in box 9 is incorrect. If you plug an 8 in the circle it does not break because of the 7 cage in column 9. It breaks because of the 15 cage in columns 6 & 7. Mark got away with a lucky guess.
The correct logic was to say that if you put an 8 I'm the circle, the rest of the cage is 1-2-4, which breaks 3 in columns 2-3-5-6
Collumns 3-4-6-7, sorry
8 in the circle forces an 8 into r1c8, which makes the 12 cage 57, which breaks the 5789 cells at the top of c1.
@50:41 great "Siobhan Sharpe" mention - what a character!
13:30, an 8 in R8C1 would force a 1 into R5C1, and a 4 in both R4C2 and R8C2 (By this time, 2 must already be in R9C1, explained at 25:07). Therefore the 8 in box 7 must be in R7, and R7C7 must be a 9
23:23 for me, solver number 7. Made some silly mistakes that cost me some time: first I treated the arrows in Box 9 like a zipper and put digits in the wrong spots, then I had some bad pencilmarks that led me to erroneous conclusions.
Since 25:49, there was a 1-3 pair in C3
38:22 Fun puzzle, thanks!
The layout is screaming set theory to me! Forcing a minima on the gaps between the cages, maybe?
Applying some basic sets you end up with very close sums (columns sum up to 42 and rows to 45) so not enough to progress further without extra info
Great puzzle.
Love how mark loves to pencilmark and colour
But doesn't like to colour Lows(1-4) only 12 and 123....and then ignores the pencilMarking on B7arrow.
His solves still great too watcj
All the 4s given? You are talking about the 11 cages and knights constraint?
Yes that's the one he was referring to. A video last 7/24 (I remember because it was my wedding anniversary).
I didn't have the timer going, took around half an hour on and off
The cages are almost all forced, and then there's some clever logic with the small digits
39:56. Lovely puzzle
2:29:14 - Phew!
@44:23, I was yelling about that because of the 13 pair in column 3 for a while...
25:43 for me. Got a little bit stuck towards the end, but nice puzzle overall!
Solved in 55:42.
If basic sudoku is beneath you (that is, checking row, column, box after each real digit put in), then why not use the conflict markers that your software provides?
Better idea: Get Sven to update the software to show the conflict checker markings for the viewers at home, while Simon or Mark have to struggle on without them!
Some time ago a viewer who is a psychologist, I believe, posted a helpful comment about the types of reasoning involved in solving complex sudoku (such as this puzzle). This person said that there are two sorts of reasoning going on, the surface reasoning of "since I have put a six here, I should check to see if it eliminates any pencil marks" sort of thing. What you have called "basic sudoku." The other kind of reasoning is the deeper kind, where the internal logic of the puzzle is being worked on. This is the level of reasoning that Mark (and Simon) are usually employing, because the puzzles they try to solve are quite complicated and there is minimal information supplied by the constructor (on purpose - this is the point of these puzzles). This viewer's main point was that it is quite difficult to effectively switch between those two levels of reasoning, and when Mark (or Simon) seem to be "missing" something that is obvious to us as viewers, it is because they are not switching to the surface level. We have the luxury of not having to be in the deep reasoning ourselves, we can just look around the grid as we please. So your disdain of Mark's not using basic sudoku is probably a misplaced criticism. Using the conflict checker and responding to it by updating pencil marks would have the same effect, pulling them out of the deeper reasoning into the surface, and their solves would be less enjoyable to them and therefore, by extension, to us. I wish that your comment had been less mean-spirited, by the way - it would have been possible for you to simply say, "I wonder why you don't use the conflict checker?" and leave out the "beneath you" part of your remark. There is no need to be mean about it.
What is all that nonsense around 13:30 about an 8 arrow adding to 7! It works perfectly well as an 8 arrow, at least in that part of the solve. When this happens I skip to the end to see if the poor logic got the correct solution. If not I go back and continue watching. Otherwise I go do something else. Yep. I’m off to do something else!
37:34 for me
"Wow, that's a tough puzzle..." Er, no it isn't... You just made a complete pig's ear of it.
Completed in 22m14s.
We can't all be as clever as you. I thought it was difficult.
00:37:31
Why am I the zeroth like again? I’m starting to get a complex.
I give up on this one. It's fine but then everything grinds to a halt. Any progress I make is through trial and error. I spent fifty something minutes, decided to restart, and then went at the same damn pace with no improvement for another fifty something minutes.
You know what? This puzzle is BAD, actually.
It’s a tricky puzzle but it’s actually incredibly cleverly put together. I would highly recommend you give it another chance but this time consider a change of solve approach, as I believe this is the reason for things grinding to a halt in the middle (I experienced the same on my first solve). I’ve left a comment with an attempt at a solve guide where I’ve tried to provide enough information to progress with the solve but still enjoy finding things yourself. You’re of course welcome to say it’s one of those sudokus where you’d prefer to forget about it but it’s definitely not a bad puzzle or solve path, it just takes understanding the hint in the title and a bit of solving experience.
Not my best solve... made decent progress in the beginning, then ended up bifurcating to find a contradiction, which did allow me to complete in 33:19 (conflict checker off). Many thanks to Scott for a cool, if a bit brutal, puzzle!