Hey Tulrak here, thanks for another feature and I'm glad you enjoyed it, the previous one of mine you solved was the pi snake one (still my own favorite). Good criticisms on the rules, thanks, I'm still quite new to setting, so this is very valuable to me and I see how my thinking was flawed on it. I thought that because fog is not a clue then it's less important and should be at the end, but it actually makes sense to put global rules first. Good solve too, you missed a few basic Sudoku steps (the 4 in box 6 really stood out to me for a long time) that would've made the solve easier, but using Sudoku in a Sudoku puzzle can be easily overlooked, so no complaints there, you still got through it very well. Now for a fan theory: I wonder if spotting the fact that the 1 in the 13 sandwich isn't proven was partly the reason why you let your guard down with the 8 and 27 sandwiches. You thought you had seen through my trick and avoided putting an unproven 1 in R3C3 but what you couldn't have known was that the 13 sandwich was a double bluff and it's actually the thought of 8 and 27 sandwiches overlapping that was too good to be true. Anyway I quite enjoyed that you made the mistake, because it makes me feel better about making similar mistakes myself and having to undo steps (even if some of the digits were shown to be true by fog). I hope that solvers didn't just place the 1 in R3C3 because the 5 and 8 next to the 9 seemed convenient for a 13 sandwich, I remember setting it with the thought that the 1 can not be placed until later.
Well you'll be pleased to know that I waited til I'd placed the 1 in box 7 before I completed the 13 clue. I did also worry about the possibility of a 162 = 9 arrow. The 9 would have had to be in R7C1 in that case and I'm not sure if I might have since the glimplse of a line at that time. I got the 6 and 7 in row 7 much later than bremseter though so he didn't have to worry about that one!
3 out of the 10 Sudokus I've made so far are fog of war, this one was my first try at a fog puzzle. So there's a chance he might cover the others, although since he said that he has been doing a lot of fog puzzles recently, then perhaps he's looking for something different.
administrative trick: with some of these fog puzzles, i have found it so hard to see candidates, and that may be critical, that i start by colouring the whole grid yellow.
14:10 With the 27 and 28 clues, I discovered that it was impossible to put either 1 or 9 in block 5. Either would require three cells adding to 7 or 8 without using a 1. That places 1 and 9 in block 5. 37:20 Yes, I placed that 1. It was a mistake without ruling out the other possibility.
Haha, it's always easier when you mistakenly place correct digits ;) You weren't meant to, I kept it in mind during setting that the 1 is not placeable, but forcing it to not be there seemed like too obvious of a trick, so I let it be there in the end.
do bear in mind that many setters may not have english as their primary language, and even among those that do, you are unlikely to find communications experts, or teachers. as to the order in which rules are presented, i agree with your position. as to the need to explicitly state negative constraints, the wording "cells separated by V sum to 5" implies no negative constraint, whereas the wording "cells summing to 5 are indicated with a V" does. (*) however, most people do not parse words that precisely, so it does help to explicitly state whether the negative constraint is active. (*) there is an interesting quirk here. these phrases break the symmetry of the IF THEN causality. thus, in the first case, IF [no V] THEN [indeterminate] makes sense, but in the second case, IF [sum .ne. 5] THEN [existence of V indeterminate] could lead to nonsense, with orphaned Vs. in other words, this must refer to an IFF.
I started to solve this, got terribly stuck pretty early in, so decided to watch BremSter's solve. He got to the point I did when I suddenly realized something I was missing that was staring me in the face... finished the puzzle in 13:39 after that, but I don't feel like that time is fair, given I probably spent about 5 minutes missing the obvious. 😅 Good puzzle, props to Tulrak, and good solve from BremSter!
There's also a route through without assuming it's a V in column 6, which arrived at a very cool deduction. I didn't trust my eyes 100% that it was a V.
I know you are looking at some standardisation of words for rules. One part of this is to get xml, json, or equivalent for the puzzles. Then they can be emailed, loaded into apps, loaded into solvers etc....
Hey Tulrak here, thanks for another feature and I'm glad you enjoyed it, the previous one of mine you solved was the pi snake one (still my own favorite).
Good criticisms on the rules, thanks, I'm still quite new to setting, so this is very valuable to me and I see how my thinking was flawed on it. I thought that because fog is not a clue then it's less important and should be at the end, but it actually makes sense to put global rules first.
Good solve too, you missed a few basic Sudoku steps (the 4 in box 6 really stood out to me for a long time) that would've made the solve easier, but using Sudoku in a Sudoku puzzle can be easily overlooked, so no complaints there, you still got through it very well.
Now for a fan theory: I wonder if spotting the fact that the 1 in the 13 sandwich isn't proven was partly the reason why you let your guard down with the 8 and 27 sandwiches.
You thought you had seen through my trick and avoided putting an unproven 1 in R3C3 but what you couldn't have known was that the 13 sandwich was a double bluff and it's actually the thought of 8 and 27 sandwiches overlapping that was too good to be true.
Anyway I quite enjoyed that you made the mistake, because it makes me feel better about making similar mistakes myself and having to undo steps (even if some of the digits were shown to be true by fog). I hope that solvers didn't just place the 1 in R3C3 because the 5 and 8 next to the 9 seemed convenient for a 13 sandwich, I remember setting it with the thought that the 1 can not be placed until later.
Well you'll be pleased to know that I waited til I'd placed the 1 in box 7 before I completed the 13 clue. I did also worry about the possibility of a 162 = 9 arrow. The 9 would have had to be in R7C1 in that case and I'm not sure if I might have since the glimplse of a line at that time. I got the 6 and 7 in row 7 much later than bremseter though so he didn't have to worry about that one!
It's a beautifully designed puzzle. I hope we'll see more fog of war sudokus from you.
3 out of the 10 Sudokus I've made so far are fog of war, this one was my first try at a fog puzzle. So there's a chance he might cover the others, although since he said that he has been doing a lot of fog puzzles recently, then perhaps he's looking for something different.
I originally put the 1 in R3C3 but then realised it wasn’t proved so I took it back out again until it resolved quite beautifully actually.
Interesting puzzle.
Good solve.
Nice puzzle and solve, thanks
A very nice flow in this solve.
administrative trick:
with some of these fog puzzles, i have found it so hard to see candidates, and that may be critical, that i start by colouring the whole grid yellow.
35:28 for me - I do like a nice FOW puzzle.
This one wasn't too bad! Definitely a fun solve.
I was not expecting to sit down to this long of a puzzle. I'm glad I did though. Very lovely puzzle!
What a great fog puzzle. I didn't got everything so I watch you to get help. It was quite hard for me, but the solve was pretty great! 🐋
44:51 for me, some very transitive deductions…
24:40, thank you!
14:10 With the 27 and 28 clues, I discovered that it was impossible to put either 1 or 9 in block 5. Either would require three cells adding to 7 or 8 without using a 1. That places 1 and 9 in block 5.
37:20 Yes, I placed that 1. It was a mistake without ruling out the other possibility.
Haha, it's always easier when you mistakenly place correct digits ;)
You weren't meant to, I kept it in mind during setting that the 1 is not placeable, but forcing it to not be there seemed like too obvious of a trick, so I let it be there in the end.
@@Tul-rak One well-placed mistake is worth -- how much?
do bear in mind that many setters may not have english as their primary language, and even among those that do, you are unlikely to find communications experts, or teachers.
as to the order in which rules are presented, i agree with your position.
as to the need to explicitly state negative constraints,
the wording "cells separated by V sum to 5" implies no negative constraint, whereas
the wording "cells summing to 5 are indicated with a V" does. (*)
however, most people do not parse words that precisely, so it does help to explicitly state whether the negative constraint is active.
(*) there is an interesting quirk here. these phrases break the symmetry of the IF THEN causality.
thus, in the first case, IF [no V] THEN [indeterminate] makes sense, but
in the second case, IF [sum .ne. 5] THEN [existence of V indeterminate] could lead to nonsense, with orphaned Vs. in other words, this must refer to an IFF.
I started to solve this, got terribly stuck pretty early in, so decided to watch BremSter's solve. He got to the point I did when I suddenly realized something I was missing that was staring me in the face... finished the puzzle in 13:39 after that, but I don't feel like that time is fair, given I probably spent about 5 minutes missing the obvious. 😅 Good puzzle, props to Tulrak, and good solve from BremSter!
There's also a route through without assuming it's a V in column 6, which arrived at a very cool deduction. I didn't trust my eyes 100% that it was a V.
I ruled out an X so it had to be a V.
Very nice Fog puzzle. Finished in 13:35
Great puzzle... done in 25 mins...
I know you are looking at some standardisation of words for rules. One part of this is to get xml, json, or equivalent for the puzzles. Then they can be emailed, loaded into apps, loaded into solvers etc....
30:45 ... wow! Can't believe I cracked this one.
18:08 :)
31:57 for me
nice puzzle