The New Sudoku Players' Forum

[shye]

Code: Select all

+-------+-------+-------+
| . . . | 5 6 4 | . . . |
| 5 . . | 3 7 8 | . . 6 |
| . 6 . | . . . | . 7 . |
+-------+-------+-------+
| 1 . . | . 2 . | . . 4 |
| . 8 . | 6 . 3 | . 2 . |
| 3 . . | . 4 . | . . 9 |
+-------+-------+-------+
| . 9 . | . . . | . 4 . |
| 7 . . | 4 5 6 | . . 8 |
| . . . | 7 8 9 | . . . |
+-------+-------+-------+
...564...5..378..6.6.....7.1...2...4.8.6.3.2.3...4...9.9.....4.7..456..8...789...

estimated rating: 4.5
something i like to do is generate puzzles on YZF's solver, find ones with nice logic/setups, and then modify them to feel more in line with my own creations. this is one such puzzle!
i might make this a series, one with typically easier/standard deductions, if people like it and wish for more

stages of development: Show

this section is just a bit of insight toward the setting process. i recommend reading only after trying the puzzle first, for the best solving experience

this was the puzzle i got when generating, which had a nice one-step solution, and as far as i could tell, nothing else reasonable to finish it in few steps
..6.7.1..7...2...5.4.1.8.9.8...4...7.3.8.9.6..92...83....291......435............

first thing i did was morph and relabel, to make it look nicer (purely by personal preferences)
..3.7.2...8.3.5.1.45.....831...2...6.7.6.3.5...2.4.9..............456......789...

then i spent some time working on a different puzzle that keeps the same step and most of the same layout

i noticed r1c46 were symmetric free singles in the morphed puzzle, and that giving r2c5 didnt collapse it down, and i liked how that made the middle stack symmetric
i decided on aiming for this pattern (below), keeping the middle stack layout but making the givens in the remaining columns more evenly dispersed per row/box
...xxx...x..xxx..x.x.....x.x...x...x.x.x.x.x.x...x...x.x.....x.x..xxx..x...xxx...

since the step i wanted resolves most of the middle stack (and particularly, box 5) my focus was on directing that impact to b46 and c19
something else i do is try to keep freebies to a minimum, i like the starting position of the puzzle to have all the information the solver needs already presented (if possible). since r7c5 was a single by nature of the setup and i wasn't a fan of how giving r37c5 looks, i wanted this to be the only freeb

eventually after trying a bunch of grids, i got the final puzzle
...564...5..378..6.6.....7.1...2...4.8.6.3.2.3...4...9.9.....4.7..456..8...789...

(just an extra note, i dont care about minimality or backdoors in my own creations. if i make a minimal or backdoor-proof puzzle thats only just a nice bonus to me, my real focus is on presentation and care for solvers experience)

[Leren]

UR Type 1 (12) r37c46 => - 12 r4c3; stte Leren

[jco]

After basics

Code: Select all

.-----------------------------------------------------------------------------.
|b289     1237    123789  | 5       6       4       | 12389   1389    123     |
| 5      c124    c1249    | 3       7       8       |d1249    19      6       |
|f48      6       348     | 129     19      12      |e3458    7       35      |
|-------------------------+-------------------------+-------------------------|
| 1       57      69      | 89      2       57      | 368     368     4       |
| 49      8       4579    | 6       19      3       | 157     2       157     |
| 3       257     2567    | 18      4       57      | 15678   1568    9       |
|-------------------------+-------------------------+-------------------------|
|g68      9       58      | 12      3       12      | 567     4       57      |
| 7       123     123     | 4       5       6       | 1239    139     8       |
|a24-6    12345   12345   | 7       8       9       | 12356   1356    1235    |
'-----------------------------------------------------------------------------'

(2)r9c1 = (2)r1c1 - (2)r2c23 = (2-4)r2c7 = (4)r3c7 - (4=8)r3c1 - (8=6)r7c1 => -6 r9c1 [NP and ste]