AnotherLife wrote:In my previous post, Cenoman found a multi-sector locked set, which was equivalent to doubly linked ALS's. As far as I understand, in this example, the set r2c6789, r3c78 is also an MSLS, but now it is equivalent to triply linked ALS (r3c78) and AALS (r2c6789).
Marek's first step can be presented as two different ALS-AALS chains:
(3=6|9)r2c6 - (69245=3)b3p45678 loop: AALS (369)r2c6, ALS (234569)b3p45678, RCs (3,6,9)
or (4=25)r3c78 - (2|5=3694)r2c6789 loop: AALS (234569)r2c6789, ALS (245)r3c78, RCs (2,4,5)
Both chains are equivalent to the same MSLS: 6 cells r2c6789, r3c78; 6 Links: 369r2, 245b3
PM 6x6 (symmetric)
- Code: Select all
3r2c6 6r2c6 9r2c6
6r2c7 9r2c7 2r2c7 5r2c7
6r2c8 9r2c8 2r2c8 4r2c8
3r2c9 9r2c9 2r2c9 5r2c9
2r3c7 5r3c7
2r3c8 4r2c8
------------------------------------------
-3r2c25 -9r2c5 -2r3c9 -5b3p139
For my own first step, presented as an ALS XY-loop: (9=1)r1c1345 - (1=3)r2c14578 - (3=9)r12c6 an equivalent MSLS exists: 11 cells r1c1345, r3c14578, r12c6; 11 Links 1b1, 345r1, 2458r3, 369b2
I preferred the derived AIC (3=69)r12c6 - r1c45 = (967-1)r1c679 = (1-3)r3c9 = (3)r3c45 loop for two reasons:
- first, because the eliminations are much easily derived,
- second, because the equivalent Truth-Link balance is lighter than above:
7 Truths r12c6, 679r1, 1b3, 3r3; 7 Links 369b2, r1c679, r3c9
To me, this not a MSLS (since the Truths are not all cell-truths). But I'm not sure there is a clear agreement upon MSLS definition.
PM 7x7 (symmetric)
- Code: Select all
3r1c6 6r1c6 9r1c6
3r2c6 6r2c6 9r2c6
9r1c45 9r1c6 9r1c7 9r1c9
6r1c6 6r1c7
7r1c7 7r1c9
1r1c9 1r3c9
3r3c45 3r3c9
----------------------------------------------------
-3b2p125 -9r2c5 -3r1c6 -5r1c7 -35r1c9 -25r3c9