Logic Masters Deutschland e.V.

am 25. Dezember 2023, 01:02 Uhr von BobAndWeave
Apologies, the updated grid is now fixed.
-----------------------------------------
Thank you for solving this question and commenting, um... although it seems a bit strange?

am 20. Dezember 2023, 14:49 Uhr von Piatato
Clever interactions during the first half. Then there was some very tough interactions toward the end. In any case it solves uniquely now. :)
---------------
Thank you for solving this puzzle and providing guidance during the process!

am 18. Dezember 2023, 14:46 Uhr von by81996672
@Piatato
Indeed, although there are some differences. But in some of the puzzles I have solved, some numbers have never appeared from beginning to end. Only by using that number in a certain clue can the puzzle have a unique answer.
So these puzzles actually conceal some complex logic to ensure their uniqueness? Okay, I've always believed that this kind of puzzle is about uniqueness.

am 18. Dezember 2023, 14:04 Uhr von Piatato
Correct, there also must be some other logic. What you are thinking about is perhaps the fact that some puzzles' main idea is a uniqueness trick? Then there is also some other logic somewhere, but often the setter will have made sure that this is something so complicated that most solvers will rather use the uniqueness trick.

am 18. Dezember 2023, 04:38 Uhr von by81996672
Hmm... Originally, CTC comes with the function of generating short connections, but I haven't noticed it yet... This is good, it can be opened directly.
I still don't quite understand how uniqueness is confirmed. In the puzzles I have solved, some used a killer cage and two ordinary digits. It's probably like the killer cage has been confirmed to be 56, with one number being 56 and the other being 569. In this case, don't we still need to use the techniques mentioned earlier to get the answer? Does uniqueness not mean that even without using this technique, the numbers here can be confirmed through other means? My understanding is that if only this technique can be used to confirm numbers, then there is no uniqueness.
Mainly, I didn't see anyone pointing out in the comments section of those puzzle pieces that may also have issues. Most puzzles either have no comments or are just praises, with very few pointing out problems. Sometimes, because the main part of the puzzle is very exciting, people choose to ignore this question. Okay, this is my last little opinion.

am 18. Dezember 2023, 01:38 Uhr von Piatato
The uniqueness trick that you mention is not actual logic that you can use to prove uniqueness. It's only a trick that a solver can use to speed up their solve, and it's based on an assumption that the puzzle is in fact unique.
Otherwise I can say that I started the puzzle last week, and thought that there were some fun and clever interactions. I hope that you will finish the puzzle. Until that happens, you may want to deactivate it.

am 17. Dezember 2023, 04:59 Uhr von by81996672
@HalfBakedLunatic
Oh! That should be because I used it in the puzzle, um... I don't know how to say it in English. For example, a regular Sudoku. R1C12 is 56, R5C1 is 56, and R5C2 is 569. To ensure the uniqueness of the puzzle, R5C2 can only fill in 9. When I was solving the puzzle myself, I also saw that a very small number of puzzles ended up using this one. Perhaps this will lead to the judgment of multiple solutions?
Then, the new version you provided is very fun, it's just that Renban has been extended a bit and combined with some things, it can actually lead to a different result. This new version is indeed simpler and smoother. Although it is a very similar puzzle, the specific process of solving it varies greatly.
I admit that the final part of my original puzzle is not good, there are other questions besides the one above. Like a clue with only a few possibilities left, assuming one of them, see if other nearby clues will contradict within a very short few steps. I often use this technique in many other puzzles, and I don't know if it's a legitimate method for those puzzles. Um, is this a bifurcation? I always thought that as long as it's extremely short within two or three steps, there should be no problem, while long ones are absolutely not. If it's long and even requires taking a screenshot, I would not hesitate to give it a low score.
I wonder if it is possible to better improve these issues. Although I have solved many puzzles, I am only using the skills I know. I hope to learn more skills so that one day I can solve some difficult puzzles that I am currently at a loss for. I also hope that more people can join the ranks of solving difficult puzzles.
Of course, with my ability to design puzzles, I cannot achieve these goals. I still have a long way to go, and I welcome guidance from others.

am 16. Dezember 2023, 15:26 Uhr von HalfBakedLunatic
You still have multiple solutions. I took the liberty of making some tweaks - this is now a valid puzzle (two of the killer cages aren't strictly necessary, but help with a smoother solution path): https://tinyurl.com/yrf55rqo

am 13. Dezember 2023, 13:16 Uhr von HalfBakedLunatic
Bifurcation is when there is no 'logical' next step that can be deduced, and the only way forward is to guess on one digit and continue on to see if it breaks - and then back up and try a different guess. The online solvers may tell you that you have a valid puzzle (with a unique solution) but only rigorous testing will ensure a logical solution path.

<edit> The various online websites (SudokuMaker, SudokuLab, F-Puzzles) all have both 'logical solvers' and 'recursive solvers'. None of them are perfect, and only work if you stick to the specific variant rules that they have programmed in. Even the best logical solvers, however, won't catch certain steps - for example if you have two palindromes that reach into a box and you know the total value of the digits on the palindromes, they won't deduce the total of the remaining three digits in the box (using this example since I came up with this trick for my puzzle 'Moriarty Leaves A Clue')

am 13. Dezember 2023, 07:15 Uhr von by81996672
@HalfBakedLunatic
@mbrandtwls
Thank you very much for pointing out the errors in the puzzle. This was my negligence, and I have now corrected the errors. I'm sorry for taking up the time of all the people who worked on this puzzle.
I should have checked it more carefully. I admit that there may have been an incorrect connection in my thinking during the process of designing the puzzle. Thank you again to all the people who made this puzzle.
Finally, I have one more request. This may be a bit thick skinnedI wonder, if you could tell me what the bifurcation means. Although I often see this word in puzzles and comments, its literal meaning is not easy to understand. If you could tell me, I would greatly appreciate it.

am 13. Dezember 2023, 00:53 Uhr von mbrandtwls
HBL asked me to try this. You have multiple solutions so it's an invalid puzzle. Of the 47 digits that are uniquely placed, it requires bifurcation to get there. Massive Fail!

am 12. Dezember 2023, 23:21 Uhr von HalfBakedLunatic
Both Sudokulab.net and Sudokumaker.app say that this puzzle has 7 solutions.

am 3. Dezember 2023, 10:23 Uhr von by81996672
Modified the difficulty level. Well... as of now, only two people have solved this puzzle. I don't know why, maybe my difficulty estimate was incorrect? I think it's three stars, Maybe it's 3.5 stars.

am 2. Dezember 2023, 11:49 Uhr von by81996672
I searched for tinyurl on my own and roughly guessed that the link might be too long. I tried to shorten it, but I'm not sure if it worked this time?

am 2. Dezember 2023, 11:27 Uhr von by81996672
Hmm... why is this link invalid? My format seems to be consistent with the video... Can anyone tell me what an effective format should be?

am 2. Dezember 2023, 11:20 Uhr von by81996672
I found a video on YouTube that taught me how to create instructions with line breaks and images, and before that, all the content was piled up together.