Page 1 of 1

Figure sum sudoku [WPF sudoku GP 2017]

Posted: Wed 13 Feb, 2019 11:57 am
by Fred76
19 - figure sum sudoku.png
19 - figure sum sudoku.png (111.46 KiB) Viewed 6494 times
Source: sudoku GP website: http://gp.worldpuzzle.org/

Potential issue: hybridation with placing pentominoes.

Re: Figure sum sudoku [WPF sudoku GP 2017]

Posted: Thu 14 Feb, 2019 9:00 pm
by detuned
They look more like tetrominoes to me, but I take the point :)

Re: Figure sum sudoku [WPF sudoku GP 2017]

Posted: Fri 15 Feb, 2019 9:24 am
by Fred76
detuned wrote:
Thu 14 Feb, 2019 9:00 pm
They look more like tetrominoes to me, but I take the point :)
Oops :roll:

Re: Figure sum sudoku [WPF sudoku GP 2017]

Posted: Sat 16 Feb, 2019 11:32 am
by DavidC
For me, I think that it is the numbers (or symbols) that should need to be placed in a sudoku, not the structural components. I am happy when regions are marked out in the beginning. I am not happy with having to mark out regions myself.

Thinking about it further, I think that I would be happy to place x-ominoes that already contained their numbers in the correct cells (let's temporarily ignore orientation and 6/9 rotational confusion), because then I would merely be using the structural shape as a container for the numbers/symbols. I might even be OK with a x-omino that didn't have all cells populated, provided at least one was. A non-touching rule for the structures would be a bit too far for me as it would be a structural building constraint rather than a number/symbol placement constraint.

Re: Figure sum sudoku [WPF sudoku GP 2017]

Posted: Sat 16 Feb, 2019 11:37 am
by DavidC
More succinctly, a sudoku is primarily about symbol placement not structure placement.

Re: Figure sum sudoku [WPF sudoku GP 2017]

Posted: Sat 16 Feb, 2019 12:32 pm
by Nilz
I don't like that definition due primarily to the use of the world 'primarily'. Similar to the way Fred has demonstrated on other types, it would be possible to create one of these puzzles where the structure placement constraint is used trivially, making that particular puzzle example a 'primarily' number placement puzzle. I hope we're not going to go as far as saying that, even within the exact same variant, some are sudokus and some aren't.

Re: Figure sum sudoku [WPF sudoku GP 2017]

Posted: Sat 16 Feb, 2019 12:41 pm
by DavidC
Even more succinctly, a sudoku is about symbol placement not structure placement :-)

Re: Figure sum sudoku [WPF sudoku GP 2017]

Posted: Sun 17 Feb, 2019 11:35 am
by Fred76
Nilz wrote:
Sat 16 Feb, 2019 12:32 pm
Similar to the way Fred has demonstrated on other types, it would be possible to create one of these puzzles where the structure placement constraint is used trivially, making that particular puzzle example a 'primarily' number placement puzzle. I hope we're not going to go as far as saying that, even within the exact same variant, some are sudokus and some aren't.
I think that if the structure placement constraint is used trivially on a particular puzzle, making it a "primarily" number placement, it should be possible to write the rules so that it is clear that it is not a structure placement constraint puzzle.
Clearly, if I read the instructions "place the given set of figures into the grid" in the instructions booklet of a tournament, I have to expect that it is not trivial (and I have to become a master on structure placement constraint puzzle in few days, to be competitive against players who are used to solve structure placement constraint puzzles very regularly).
My opinion is that not only this puzzle is not a sudoku variation (it is a puzzle hybrid featuring sudoku rules as describe by motris), but another big issue is that this puzzle is very unfair (if unfairness of not being sudoku was not sufficiant).

Fred

Re: Figure sum sudoku [WPF sudoku GP 2017]

Posted: Sun 17 Feb, 2019 11:39 am
by Fred76
I think this example is very similar to the battleship sudoku in this thread: viewtopic.php?f=33&t=1961
It has same property of being hybrid with structure placement puzzle.

Fred