10:37. I've gotten incredibly good and fast(?) at hexagonal shaped slitherlink but I rarely practice normal square shaped so it's interesting how much knowledge is transferred between them haha Edit: yeah, watching you and as others have said, using colors to keep track of which cells are 'inside' and 'outside' helps a TON. Especially if you're relatively inexperienced and don't know a lot of patterns already, it definitely helps keep track of everything. Two squares that are the same color can never have a line between them, two that are different must have a line, 2's must always have 2 of each color surrounding them, etc.
Wonderful to see all these sub 5 minute solves in the comment section after just spending 51 mins on it. Well, first time I ever tried to solve one of these, So I was just satisfied with success dialog popping up in the end.
This type of puzzle is all about pattern recognition, so the people that did it in less than 5 minutes (including me) don't have to think about it from scratch.
I do about ten to a dozen Slitherlink puzzles every day and my number one tip to anyone out there new to Slitherlink is *learn the patterns* . There are many, many patterns in Slitherlink, a lot of them very basic and quick to learn, and there are a number of advanced techniques to learn once you have the basic ones mastered. Do this and you will find you can usually complete a good two thirds of a puzzle without any real problem. Oh, and I really do recommend sticking to the easy ones until you are well used to solving them because some of the harder ones can be very, very difficult and almost impossible to do without a good knowledge of advanced techniques.
I find it can be very helpful to colour in cells which are "inside" and "outside" the loop. To get started, if there's a line on the edge of the grid, the cell touching it must be inside. Parity (inside/outside) stays the same between cells with an x between them and changes when there's a line between them. Really helps sometimes. The website you're using even supports that by clicking on the inside of a cell.
Slitherlink has been a favourite of mine, particularly as a Sudoku break since Slitherlinks appeared in the Puzzler Sudoku Monthly. Always try doing it with logic but the harder ones need a little bifurcation or guessing a fork until it breaks. Colouring the cells is useful for hard ones. The colour will always be the same for cells with a cross between them. Obvious but it often gives you visual insight beyond just using lines as well as the ability to see impossibilities quicker. Aways click the 'Enable to Input BGColor when the center of the cell is clicked' option. I learned some more little logic nuggets from you Mark.
One thing that helps tremendously is to colour the cells which we know are inside the loop and the cells which are outside the loop (there's even an "Enable to Input BGColor when the center of the cell is clicked" option in the Slitherlink software). It's particularly relevant when there are opposite components of the loop coming together. Using that, I completed the puzzle within 10 minutes (granted, I've had lots of prior experience with Slitherlink so I easily recognize some common patterns without any explanation needed). By the way, background colouring would help in the castle wall puzzle, too, except it's aligned to the lines going through the middles of cells rather than along their borders so this won't work with the software used earlier. Now, if someone made software that had cells aligned to the potential castle borders - with the givens occupying the corners between cells - that could make the area colouring so much easier to execute, and possibly streamline solving of castle wall at least a litte.
Re: castle wall coloring, I wonder if there's a convenient way to color cell corners, since that should be basically equivalent to shifting the line and coloring the cells. Not a perfect solution, but it could be a workaround.
it would be helpful too to color the "speculative" lines a different color from "finished" lines, when working multiple possibilities so that you remember which lines to undo. this could be especially helpful in cases like the diagonal 3s where some lines are ambiguous until you solve more of the puzzle.
One tip i would give to you is to look more at numbers 1 and 3 because there are only 4 ways you draw lines around them compared to 6 (i think) around number 2.
I'd never heard of a slitherlink puzzle and am really pleased to have been able to solve it after watching the introduction. Even if it took 45m-1h... These are fun!
For your hypotheses about a 0 and 2 diagonally adjacent: If there are no additional clues horizontally or vertically adjacent to the 2, then the only remaining way to disambiguate which side of the 2 the pair of lines goes on is for the dot on the 2 farthest away from the 0 to be occupied. In that case, the lines along the 2 can't go through that dot, and must go through the dot next to the 0. And so, if you have a 0 and 2 like that, the fact that the puzzle has a unique solution means that this dot must be occupied, not only giving you the two lines that occupy the 2, but two additional lines that occupy that dot! (On the other hand, this deduction can only ever be a shortcut, not something necessary for a solve. If the puzzle has a unique solution, then something outside of this analysis must be forcing that dot to be occupied, which means that that something is also sufficient to disambiguate the lines along the 2.)
Solved in 26:58 for me, but I am still new to Slitherlink and also learned a few patterns along the way (like the 1-3 combo along an edge, or two 3s diagonally apart from each other).
I'm sure you'll get the hang of it quickly when you do more of these, but currently you're still not seeing a lot of the very basic small scale patterns. For instance with the 3 in the bottom right. Whichever edge the 2 diagonally above it was gonna have of the remaining two options, it was always going to enter the 3 at the shared corner. And whenever you have an edge entering a 3 at a corner you'll have two edges opposite that corner. You would have had similar pattern if that 3 had been a 2 by the way. Because whenever you have an edge entering a 2 at a corner (i.e. an edge that cannot turn away from the 2) there must be an edge exiting that 2 at the opposite corner. And often one of the two directions in which it can exit is already excluded for some reason so then it must be the other one. Another fairly basic pattern you missed was with the 2 diagonally below the middle 0. You correctly spotted that the 0 leaves only two options for the edges around the 2: a corner pointing towards or away from the 0. What you failed to notice is that in both cases the loop exits the 2 at the two corners on the other diagonal. And that means that any edges already going into those corners from outside the 2 cannot turn away from the 2. Spotting that would have allowed you to see much more easily that the edge coming in from above wasn't going along the top of the 1 and connect with the 3's left edge.
I've seen a suggestion when paper solving Slitherlink before to draw that line diagonally through a cell like the 2 in the last paragraph (r7c6). This "virtual connection" combined with the work in the top right of the puzzle means that the short inner loop can't close via the line above the 2 in r8c7, so must go down the sides of it.
A more general rule is that any square with a 2 has the property that adding the number of lines coming in to two opposite corners gives an even result, and any square with a 1 or 3 it gives an odd result. This can be surprisingly useful.
man this is a cool puzzle, never even knew it existed before. it took me 30 mins to finish it because i made a mistake that broke it but i used the mistake to get a clue i shouldnt really have at that point in the game so that was kinda nice.
I got around 6 minutes to solve this. Over the years I've developed a completely different method to solve. I start similarly with the 3's. But then I color the squares based on if the square is inside or outside the loop. Then each number in the puzzle interacts with the adjacent colors in different ways which are used to color more squares. The inside squares must all connect to each other. The outside squares must connect to the outside.
3:38 for me, felt a bit slower because I'm used to just klick to get a line, not drag it between the dots. I've been doing slitherlinks for years, they are quit addictive and fairly simple. They just take longer the bigger they are. Simple pattern recognition with occasional counting of loop ends.
yeah it took me longer than i care to admit just to figure out how to make the first line. I clicked so many times and could only get an x it was driving me crazy.
I do enjoy slitherlink, so very happy to see them on the channel. Look forward to doing this one, but will need to wait a little before I can get started sadly.
This one was actually surprisingly doable, the only piece of missing logic I had was counting the line ends that meant the two on the left had to come up :o
5:23 for me, but I play slither link quite a bit and know several shortcuts. I have a really hard time with the sudoku's they post and they typically takes me 45 minutes.
For your speculative line to illustrate why it can never be present, the trial mode functionality of the software might be very useful since it turns undoing the speculation into a single button press.
You wouldn't be the other person who's like 500 'huge' puzzles deep on a slitherlink app, would you? Assuming there's only one other, it's a funny thought to me so I've decided that's how it is
One of the more important techniques (in my probably uninformed opinion) is the idea of "one of." To explain, look at 10:40 with the 2 he just completed at the left. The left segment has two options, either left or up. Therefore, 'one of' those two segments is in the loop. Not both, because then it would fork, and not neither, because then the segment couldn't go anywhere. Now with the two immediately up-left of it, we know one of the bottom or right sides are in the loop. So where is the second one? It has to be one of the left or top sides. Furthermore, we can get another segment from that. Because the top left corner of that two is along an edge of the puzzle, where does that next 'one of' segement go from there. It HAS to go up, meaning the left side of that next two is also part of the loop. It takes a bit to get this (feel free to try x-ing that left side and convincing yourself it must lead to a contradiction) but once you can see it fast it really makes these easier. It can even apply where there isn't a segment to start with. For example, look at the two up-right of a three at the bottom right of the puzzle. At 14:19, he gets it so the top side is x-ed and the right is filled in. Now, one of the two bottom-left sides must be filled in to complete the two. This propogates along the diagonal the exact same way. Now, one of the two top-right sides of the three must be filled in. Therefore, both of the two at the other corner (the bottom left) can be filled in immediately (as he does later). Since this also x-es the bottom side of the two to the left of the three, one of the top-left sides of that two must be filled, meaning one of the bottom-right sides of the two up-left from there must be filled, and therefore one of the top-left cells of that two must be filled (This is the two three cells left of the original one). This can be useful later when the top cell is x-ed, and therefore the left side must be filled. Anyway I think it's important to get this down, especially to get really fast times. If you start looking for it at corners, you'll find it everywhere.
1 diagonally next to 0 always gives you two x because going into the dot between them would be a dead end. Also every dot that gets its third x immediately gets its fourth x.
That was slitherlink bifurcation. With the right tools (making sure that you x off impossible paths, and colouring the inside and the outside, knowing the patterna around 2s and 3s) it is much easier.
@@tomcollyer641 I don't try the puzzles usually, but I probably could have done it in a few minutes. Mostly because I have poor coordination, there was nowhere really I felt stuck on that puzzle.
This video was a bit painful to watch as you missed a lot of easier patterns. Would you be interested in a video response on how to do it quickly going into more into some micro-patterns that help to solve those?
I'm the same, but I love slitherlinks because in most cases a what if you encounter just means there is a logic rule you haven't noticed yet. Unless you are using an app that creates a randomised puzzle, in which case you can actually end up with one that will require trial and error.
Please do more Slitherlink puzzles! This was very fun and I'd love to see more.
10:37. I've gotten incredibly good and fast(?) at hexagonal shaped slitherlink but I rarely practice normal square shaped so it's interesting how much knowledge is transferred between them haha
Edit: yeah, watching you and as others have said, using colors to keep track of which cells are 'inside' and 'outside' helps a TON. Especially if you're relatively inexperienced and don't know a lot of patterns already, it definitely helps keep track of everything. Two squares that are the same color can never have a line between them, two that are different must have a line, 2's must always have 2 of each color surrounding them, etc.
Wonderful to see all these sub 5 minute solves in the comment section after just spending 51 mins on it.
Well, first time I ever tried to solve one of these, So I was just satisfied with success dialog popping up in the end.
I'm right there with you! This was my first time with one of these and finished in a similar time frame. Sub 5 minutes, we've got a ways to go :)
congrats on your first solve !!
At least I'm not the only one who did about 50 mins :)
It was my first one, too.
This type of puzzle is all about pattern recognition, so the people that did it in less than 5 minutes (including me) don't have to think about it from scratch.
I do about ten to a dozen Slitherlink puzzles every day and my number one tip to anyone out there new to Slitherlink is *learn the patterns* . There are many, many patterns in Slitherlink, a lot of them very basic and quick to learn, and there are a number of advanced techniques to learn once you have the basic ones mastered. Do this and you will find you can usually complete a good two thirds of a puzzle without any real problem. Oh, and I really do recommend sticking to the easy ones until you are well used to solving them because some of the harder ones can be very, very difficult and almost impossible to do without a good knowledge of advanced techniques.
I find it can be very helpful to colour in cells which are "inside" and "outside" the loop. To get started, if there's a line on the edge of the grid, the cell touching it must be inside. Parity (inside/outside) stays the same between cells with an x between them and changes when there's a line between them. Really helps sometimes. The website you're using even supports that by clicking on the inside of a cell.
Slitherlink has been a favourite of mine, particularly as a Sudoku break since Slitherlinks appeared in the Puzzler Sudoku Monthly. Always try doing it with logic but the harder ones need a little bifurcation or guessing a fork until it breaks. Colouring the cells is useful for hard ones. The colour will always be the same for cells with a cross between them. Obvious but it often gives you visual insight beyond just using lines as well as the ability to see impossibilities quicker. Aways click the 'Enable to Input BGColor when the center of the cell is clicked' option. I learned some more little logic nuggets from you Mark.
One thing that helps tremendously is to colour the cells which we know are inside the loop and the cells which are outside the loop (there's even an "Enable to Input BGColor when the center of the cell is clicked" option in the Slitherlink software). It's particularly relevant when there are opposite components of the loop coming together. Using that, I completed the puzzle within 10 minutes (granted, I've had lots of prior experience with Slitherlink so I easily recognize some common patterns without any explanation needed).
By the way, background colouring would help in the castle wall puzzle, too, except it's aligned to the lines going through the middles of cells rather than along their borders so this won't work with the software used earlier. Now, if someone made software that had cells aligned to the potential castle borders - with the givens occupying the corners between cells - that could make the area colouring so much easier to execute, and possibly streamline solving of castle wall at least a litte.
Re: castle wall coloring, I wonder if there's a convenient way to color cell corners, since that should be basically equivalent to shifting the line and coloring the cells.
Not a perfect solution, but it could be a workaround.
Thank you for that tip! It helped my brain tremendously, and made it more enjoyable than frustrating.
it would be helpful too to color the "speculative" lines a different color from "finished" lines, when working multiple possibilities so that you remember which lines to undo. this could be especially helpful in cases like the diagonal 3s where some lines are ambiguous until you solve more of the puzzle.
@@better.better Well, you have the "Trial mode" button down below, maybe that satisfies your needs to some degree?
One tip i would give to you is to look more at numbers 1 and 3 because there are only 4 ways you draw lines around them compared to 6 (i think) around number 2.
More of these please. I absolutely loved your explanations and verbalized thought process
This is like mine sweeper with lines. Super addictive.
I'd never heard of a slitherlink puzzle and am really pleased to have been able to solve it after watching the introduction. Even if it took 45m-1h... These are fun!
For your hypotheses about a 0 and 2 diagonally adjacent: If there are no additional clues horizontally or vertically adjacent to the 2, then the only remaining way to disambiguate which side of the 2 the pair of lines goes on is for the dot on the 2 farthest away from the 0 to be occupied. In that case, the lines along the 2 can't go through that dot, and must go through the dot next to the 0. And so, if you have a 0 and 2 like that, the fact that the puzzle has a unique solution means that this dot must be occupied, not only giving you the two lines that occupy the 2, but two additional lines that occupy that dot!
(On the other hand, this deduction can only ever be a shortcut, not something necessary for a solve. If the puzzle has a unique solution, then something outside of this analysis must be forcing that dot to be occupied, which means that that something is also sufficient to disambiguate the lines along the 2.)
Solved in 26:58 for me, but I am still new to Slitherlink and also learned a few patterns along the way (like the 1-3 combo along an edge, or two 3s diagonally apart from each other).
I love slitherlink, it would be great to see a bigger scheme on the channel!
Each dot can ONLY be connected to exactly 2 lines. Those part of the loop
Exactly right. Once you identify 2 lines leading into a dot, you can immediately X out the other 2 potential lines from that dot.
I'm sure you'll get the hang of it quickly when you do more of these, but currently you're still not seeing a lot of the very basic small scale patterns.
For instance with the 3 in the bottom right. Whichever edge the 2 diagonally above it was gonna have of the remaining two options, it was always going to enter the 3 at the shared corner. And whenever you have an edge entering a 3 at a corner you'll have two edges opposite that corner.
You would have had similar pattern if that 3 had been a 2 by the way. Because whenever you have an edge entering a 2 at a corner (i.e. an edge that cannot turn away from the 2) there must be an edge exiting that 2 at the opposite corner. And often one of the two directions in which it can exit is already excluded for some reason so then it must be the other one.
Another fairly basic pattern you missed was with the 2 diagonally below the middle 0. You correctly spotted that the 0 leaves only two options for the edges around the 2: a corner pointing towards or away from the 0. What you failed to notice is that in both cases the loop exits the 2 at the two corners on the other diagonal. And that means that any edges already going into those corners from outside the 2 cannot turn away from the 2. Spotting that would have allowed you to see much more easily that the edge coming in from above wasn't going along the top of the 1 and connect with the 3's left edge.
I've seen a suggestion when paper solving Slitherlink before to draw that line diagonally through a cell like the 2 in the last paragraph (r7c6). This "virtual connection" combined with the work in the top right of the puzzle means that the short inner loop can't close via the line above the 2 in r8c7, so must go down the sides of it.
did it in 15:58, using the background colors to shade the inside & the outside of the loop really helped
A more general rule is that any square with a 2 has the property that adding the number of lines coming in to two opposite corners gives an even result, and any square with a 1 or 3 it gives an odd result. This can be surprisingly useful.
man this is a cool puzzle, never even knew it existed before. it took me 30 mins to finish it because i made a mistake that broke it but i used the mistake to get a clue i shouldnt really have at that point in the game so that was kinda nice.
22:55 really like these kinds of puzzles!
never seen this puzzle before but quite happy with my solve time of 22:49 look forward to seeing more on this channel
I got around 6 minutes to solve this. Over the years I've developed a completely different method to solve. I start similarly with the 3's. But then I color the squares based on if the square is inside or outside the loop. Then each number in the puzzle interacts with the adjacent colors in different ways which are used to color more squares. The inside squares must all connect to each other. The outside squares must connect to the outside.
3:38 for me, felt a bit slower because I'm used to just klick to get a line, not drag it between the dots. I've been doing slitherlinks for years, they are quit addictive and fairly simple. They just take longer the bigger they are. Simple pattern recognition with occasional counting of loop ends.
yeah it took me longer than i care to admit just to figure out how to make the first line. I clicked so many times and could only get an x it was driving me crazy.
Pretty sure there were slitherlink puzzles in one of the Professor Layton games. Seems very familiar.
I do enjoy slitherlink, so very happy to see them on the channel. Look forward to doing this one, but will need to wait a little before I can get started sadly.
This one was actually surprisingly doable, the only piece of missing logic I had was counting the line ends that meant the two on the left had to come up :o
5:23 for me, but I play slither link quite a bit and know several shortcuts. I have a really hard time with the sudoku's they post and they typically takes me 45 minutes.
For your speculative line to illustrate why it can never be present, the trial mode functionality of the software might be very useful since it turns undoing the speculation into a single button press.
11.06 for me. Not bad considering how out of practice I am with them. Wouldn't mind seeing more of these on the channel.
Hexagonal slitherlink is fun. You might want to try one sometime.
You wouldn't be the other person who's like 500 'huge' puzzles deep on a slitherlink app, would you?
Assuming there's only one other, it's a funny thought to me so I've decided that's how it is
One of the more important techniques (in my probably uninformed opinion) is the idea of "one of." To explain, look at 10:40 with the 2 he just completed at the left. The left segment has two options, either left or up. Therefore, 'one of' those two segments is in the loop. Not both, because then it would fork, and not neither, because then the segment couldn't go anywhere. Now with the two immediately up-left of it, we know one of the bottom or right sides are in the loop. So where is the second one? It has to be one of the left or top sides. Furthermore, we can get another segment from that. Because the top left corner of that two is along an edge of the puzzle, where does that next 'one of' segement go from there. It HAS to go up, meaning the left side of that next two is also part of the loop.
It takes a bit to get this (feel free to try x-ing that left side and convincing yourself it must lead to a contradiction) but once you can see it fast it really makes these easier. It can even apply where there isn't a segment to start with. For example, look at the two up-right of a three at the bottom right of the puzzle. At 14:19, he gets it so the top side is x-ed and the right is filled in. Now, one of the two bottom-left sides must be filled in to complete the two. This propogates along the diagonal the exact same way. Now, one of the two top-right sides of the three must be filled in. Therefore, both of the two at the other corner (the bottom left) can be filled in immediately (as he does later). Since this also x-es the bottom side of the two to the left of the three, one of the top-left sides of that two must be filled, meaning one of the bottom-right sides of the two up-left from there must be filled, and therefore one of the top-left cells of that two must be filled (This is the two three cells left of the original one). This can be useful later when the top cell is x-ed, and therefore the left side must be filled.
Anyway I think it's important to get this down, especially to get really fast times. If you start looking for it at corners, you'll find it everywhere.
About a hour and had to restart about 3 times. Bit a brute force but it told me what was impossible.
Good solve
14:31 I do Slitherlink all the time on this app for Android that has 30 puzzles. It calls it "Loopy" for some reason.
1 diagonally next to 0 always gives you two x because going into the dot between them would be a dead end. Also every dot that gets its third x immediately gets its fourth x.
The 3 on the top left could have been figured out by coloring the inny and outie of the loop. The left 3 is same color as the line coming from above
That was slitherlink bifurcation. With the right tools (making sure that you x off impossible paths, and colouring the inside and the outside, knowing the patterna around 2s and 3s) it is much easier.
Thank you
Complete: 07:27. I definitely felt slow with mouse.
I am amazed that you could do that without using polarity.
Done in 13:38
I love slitherlink! I've been playing it for years!
There's a slitherlink app which has a lot of different playfields with differently shaped cells.
About 7 minutes. I struggled with the interface. I could not put a single line for nearly 1 minute, which does not help for slitherlink.
Best
Took 4:53 which I don't think is too shabby, considering I haven't done one in about a year
11:37. Definitely out of practice and/or tired.
7:06 I get my fix of these from Conceptispuzzles, but I enjoy doing them no matter the platform.
Sorry, another pass for me. Maybe I'll try some easier ones in the future if I have more time, but can't seem to enjoy loop puzzles, just quite yet.
2:35 solve time here (felt pretty slow). Some nice logic!
16m 🙄😓
What took you so long? (I'm kidding!) 1 minute 22 here
@@tomcollyer641 I don't try the puzzles usually, but I probably could have done it in a few minutes. Mostly because I have poor coordination, there was nowhere really I felt stuck on that puzzle.
Tom Collyer sleepy and on mobile haha, didn’t want to discourage others by saying that though
11:50
2:57 ; a little bit slow, but I've missed doing these particular puzzles! Was a lot of fun to get to do one again.
anyone know the puzzle with the clues spelling HARD and EASY in different boxes.. could you please share or help me find? much appreciated
Took me an hour, takes a while to get used to for me
This video was a bit painful to watch as you missed a lot of easier patterns. Would you be interested in a video response on how to do it quickly going into more into some micro-patterns that help to solve those?
Florent Castelli I would!
Florent Castelli I know a lot of the patterns but I still failed to solve this one as quickly as I would have liked!
Ah, using over 1hour for me, nearly tried all possible routes.
The Witness!
There was a game of this on coolmathgames
8:53 for me, was a bit rusty on these, but this one was very doable :)
At 22:38 you have ruined the two you were discussing before (r7c6).
I have not got the brain for this higher level stuff I will stick to simple.
22:05 my solve I definatly need some practice with these. Very slow compared to the other comments
44:30 Solve time for me.
Sorry, I don't like these puzzles, a bit too many what-ifs for my taste.
I'm the same, but I love slitherlinks because in most cases a what if you encounter just means there is a logic rule you haven't noticed yet. Unless you are using an app that creates a randomised puzzle, in which case you can actually end up with one that will require trial and error.