The New Sudoku Players' Forum

[denis_berthier]

As I like very much the double lozenge tiling with inner patterns, here's a last puzzle with this pattern.
This is an easy one.

Code: Select all

   +-------+-------+-------+
   ! . 5 . ! . 2 . ! . 1 . !
   ! 6 . . ! 3 . 8 ! . . 9 !
   ! . . 4 ! . . . ! 2 . . !
   +-------+-------+-------+
   ! . 7 . ! . 1 . ! . 9 . !
   ! 1 . . ! 8 . 4 ! . . 2 !
   ! . 8 . ! . 7 . ! . 6 . !
   +-------+-------+-------+
   ! . . 9 ! . . . ! 5 . . !
   ! 3 . . ! 1 . 5 ! . . 7 !
   ! . 6 . ! . 8 . ! . 4 . !
   +-------+-------+-------+

.5..2..1.6..3.8..9..4...2...7..1..9.1..8.4..2.8..7..6...9...5..3..1.5..7.6..8..4. #  1632 FNBTHW C28.m/S8.f
SER = 3.8


[Leren]

I cheated and used Hodoku, which gives an OTP solution with Just one Swordfish. Leren

[denis_berthier]

I cheated and used Hodoku, which gives an OTP solution with Just one Swordfish. Leren

Difficult to beat you on that.

I use Hodoku occasionally and I like is clean interface. I also like its clear definition of the rules used (though its formulation of some Uniqueness rules could be improved) and the accompanying examples. Being written in Java, it's highly portable. Like Sudoku Explainer, I have no problem running it on my Mac, even with the recent versions of MacOS that no longer support 32-bit code.

However, I can't say I'm very familiar with Hodoku (or SE). SudoRules finds a Naked-Pair and a Hidden-Pair before a Swordfish:

Code: Select all

***********************************************************************************************
***  SudoRules 20.1.s based on CSP-Rules 2.1.s, config = S
***  Using CLIPS 6.32-r773
***********************************************************************************************
hidden-single-in-a-column ==> r3c6 = 1
164 candidates, 778 csp-links and 778 links. Density = 5.82%
whip[1]: c2n4{r8 .} ==> r7c1 ≠ 4
naked-pairs-in-a-row: r8{c3 c8}{n2 n8} ==> r8c7 ≠ 8, r8c2 ≠ 2
naked-single ==> r8c2 = 4
hidden-pairs-in-a-row: r2{n1 n2}{c2 c3} ==> r2c3 ≠ 7
whip[1]: r2n7{c8 .} ==> r1c7 ≠ 7, r3c8 ≠ 7
swordfish-in-columns: n3{c2 c5 c8}{r3 r5 r7} ==> r7c9 ≠ 3, r7c6 ≠ 3, r5c7 ≠ 3, r5c3 ≠ 3, r3c9 ≠ 3
stte

Is this due to some Hodoku default priority settings (e.g. Swordfish > Pairs) or did you choose some specific settings?

[Leren]

No, just compressed use of jargon by me. The OTP move follows one or more basic moves, which for Hodoku on this puzzle were :

1. Hidden Single r3c6 = 1; 2. A pointing pair of 4's r46c1 ; 3. NP (28) r8c38; 4; Naked single r8c2 = 4; 5. NT (457) r2c578; 6. A claiming pair of 7's r8c78;

OTP move: SF (3) c258 / r357; stte

Same as you I think.

Leren

[denis_berthier]

No, just compressed use of jargon by me. The OTP move follows one or more basic moves, which for Hodoku on this puzzle were :
1. Hidden Single r3c6 = 1; 2. A pointing pair of 4's r46c1 ; 3. NP (28) r8c38; 4; Naked single r8c2 = 4; 5. NT (457) r2c578; 6. A claiming pair of 7's r8c78;
OTP move: SF (3) c258 / r357; stte

I see. Yes, same Swordfish, but I find no NT before it; instead I have a HP (but that must be the result of different priorities in Hodoku.

BTW, what does "OTP" stand for?

[Leren]

OTP stands for One Trick Pony ie a puzzle that needs only one non basic move to solve.

The daily puzzles on this forum (when they resume) are supposed to be OTP grade. Due to a historical quirk, the one trick is not supposed to be a basic fish, so they are never actually used.

Leren