The New Sudoku Players' Forum

[Mauriès Robert]

Hi all,
This puzzle is a bit more difficult than the previous one, but your resolutions will be interesting for sure.

...82..54.....62..........7.974...18.2..8..7.85...746.1..........41.....57..98...

puzzle: Show

my resolution: Show

After simplifying the grid by the basic techniques (6 placements and 6 eliminations), we use "step by step" the tracks technique (https://www.assistant-sudoku.com/Pdf/TDP-anglais.pdf ), studying only anti-tracks P'(E) (green marking) without developing their conjugated tracks P(E), this for different successive generating sets E.

1) P'(2r3c1) : (-2r3c1) => (2r3c3->5r2c3)->8r7c3->9r8c1 ... => -2r8c1 => r3c1=2, r8c8=2.



2) P'(3r56c3) : (-3r56c3) => 3r4c1->9r2c1->9r7c3->(2r9c3 et 58r23c3) ... => -3r2379c3



3) P'(3r23c8) : (-3r23c8) => 3r7c8->1r9c7->1r2c9 ... => -3r2c9.
4)P'(3r23c8) : (-3r23c8) => 3r7c8->3r9c4->(5r3c5 and 5r8c6)->5r4c5->3r4c7 ... => -3r13c7.
3r23c8 => -3 r7c8.



5) P'(9r2c1) : (-9r2c1) => 3r2c1->3r3c8->5r3c4->5r2c3 ... => -9r2c3.



6) P'(9r2c1) : (-9r2c1) => 3r2c1->3r1c6->(5r3c4 et 5r8c6)->5r4c5->3r4c7->1r9c7->1r2c9 ... => -9r2c9.



7) P'(9r7c3) : (-9r7c3) => 9r8c1->9r2c8 ... => -9r7c8 => r7c8=8 and end of the puzzle by induction (singles).

[denis_berthier]

.

Code: Select all

Resolution state after Singles and whips[1]:
   +----------------------+----------------------+----------------------+
   ! 7      136    369    ! 8      2      139    ! 1369   5      4      !
   ! 39     1348   3589   ! 357    13457  6      ! 2      389    139    !
   ! 2369   13468  235689 ! 35     1345   13459  ! 13689  389    7      !
   +----------------------+----------------------+----------------------+
   ! 36     9      7      ! 4      356    2      ! 35     1      8      !
   ! 4      2      136    ! 356    8      135    ! 359    7      359    !
   ! 8      5      13     ! 9      13     7      ! 4      6      2      !
   +----------------------+----------------------+----------------------+
   ! 1      368    23689  ! 23567  34567  345    ! 35789  2389   3569   !
   ! 2369   368    4      ! 1      3567   35     ! 35789  2389   3569   !
   ! 5      7      236    ! 236    9      8      ! 13     4      136    !
   +----------------------+----------------------+----------------------+
168 candidates.

I found something odd in the lengths of your "anti-tracks" (shorter than usual).
So, applying successively my function "try-to-eliminate" to each of your eliminated candidates (with some multiple eliminations in your path corresponding to several lines in mine), I get the following (same numbering).

1) whip[4]: c3n2{r9 r3} - c3n5{r3 r2} - c3n8{r2 r7} - b7n9{r7c3 .} ==> r8c1≠2
hidden-single-in-a-column ==> r3c1=2
hidden-single-in-a-row ==> r8c8=2

2) z-chain[4]: c3n5{r2 r3} - c3n8{r3 r7} - b7n9{r7c3 r8c1} - r2c1{n9 .} ==> r2c3≠3
z-chain[4]: c3n5{r3 r2} - c3n8{r2 r7} - b7n9{r7c3 r8c1} - r2c1{n9 .} ==> r3c3≠3
z-chain[3]: b4n3{r6c3 r4c1} - r2c1{n3 n9} - b7n9{r8c1 .} ==> r7c3≠3
z-chain[4]: c3n2{r9 r7} - b7n9{r7c3 r8c1} - r2c1{n9 n3} - b4n3{r4c1 .} ==> r9c3≠3

3) z-chain[3]: c8n3{r3 r7} - r9c7{n3 n1} - c9n1{r9 .} ==> r2c9≠3

4) whip[6]: b6n3{r5c7 r5c9} - r9n3{c9 c4} - r8c6{n3 n5} - r5c6{n5 n1} - c3n1{r5 r6} - c3n3{r6 .} ==> r1c7≠3
braid[6]: c8n3{r3 r7} - r9n3{c7 c4} - r8c6{n3 n5} - r3c4{n3 n5} - c5n5{r8 r4} - r4c7{n5 .} ==> r3c7≠3
whip[1]: b3n3{r3c8 .} ==> r7c8≠3

5) biv-chain[4]: c3n5{r2 r3} - r3c4{n5 n3} - c8n3{r3 r2} - r2c1{n3 n9} ==> r2c3≠9

6) whip[8]: r2c1{n9 n3} - r1n3{c2 c6} - c6n9{r1 r3} - c6n1{r3 r5} - r6c5{n1 n3} - r4n3{c5 c7} - r9c7{n3 n1} - b3n1{r1c7 .} ==> r2c9≠9
naked-single ==> r2c9=1
hidden-single-in-a-column ==> r9c7=1

7) finned-x-wing-in-rows: n9{r2 r8}{c1 c8} ==> r7c8≠9
stte


Now, referring to your words below, the million dollar question. Of the following 3 statements, which is true?
1) you have finally adopted my definition of length without saying it (which is plagiarism in the usual sense)
2) you have an explanation of how you find exactly the same lengths as me instead of the longer lengths you used to find before your long absence from this website
3) you have relaxed your definition of plagiarism

I could with antitracks (and tracks as well as DEFISE does) design a fixed length antitrack pattern and use this "simplest first" principle, except that manually this is very tedious or impossible and intellectually it would be like plagiarizing your concepts.

[Mauriès Robert]

.
Now, referring to your words below, the million dollar question. Of the following 3 statements, which is true?
1) you have finally adopted my definition of length without saying it (which is plagiarism in the usual sense)
2) you have an explanation of how you find exactly the same lengths as me instead of the longer lengths you used to find before your long absence from this website
3) you have relaxed your definition of plagiarism

I could with antitracks (and tracks as well as DEFISE does) design a fixed length antitrack pattern and use this "simplest first" principle, except that manually this is very tedious or impossible and intellectually it would be like plagiarizing your concepts.

Hi Denis,
Neither plagiarism nor change of method, and if my resolution corresponds here to the "simpler first" that you practice it is only by chance, especially since I am unable to produce a resolution with your methods correctly. Besides, as you know, for a manual resolution it is very difficult to apply the "simpler first" principle and my resolutions are manual.
But I don't see the similarity that you invoke, my second step for example doing in one time the 4 eliminations that you do in 4 times ! If I had done "simpler first" I would have done these eliminations one by one and the lengths of the sequences would have been shorter.
Robert

[denis_berthier]

.
I'm not talking of the simplest-first strategy but of the lengths of the individual chains. And I'm not talking of those aggregated to make multiple eliminations but of all the other ones.

[Mauriès Robert]

.
I'm not talking of the simplest-first strategy but of the lengths of the individual chains. And I'm not talking of those aggregated to make multiple eliminations but of all the other ones.

I refer you to my previous resolutions, for example Robert's puzzle 2020-11-02 or Robert's puzzle 2020-09-22 etc..., to see that it is not new that I use anti-tracks made of quite short sequences contrary to what you write. I've been doing this since I introduced my tracks technique on this forum, but not always of course, because I also sometimes use long sequences, use sets of tracks by crossing the tracks, forks, all the things you usually criticize. But this is my technique of resolution, I don't need to plagiarize. Having said that, I see no reason not to use a short sequence when I see it in the course of my resolution. I also try to be as pedagogical as possible in the presentation of my resolutions to those who read me with fairly short sequences when I can, which is not to be confused with your resolution strategy.
One thing is sure, I build my resolutions before seeing yours, so I don't risk copying them either in the order of the steps or in the length of the sequences.
Robert

[denis_berthier]

One thing is sure, I build my resolutions before seeing yours, so I don't risk copying them either in the order of the steps or in the length of the sequences.

Totally beside the point. I never suggested you are copying my resolution paths one by one.

I already established long ago that anti-tracks are a downgraded (and much posterior) version of braids (or S-braids if you allow inner Subsets): lack of the left-linking candidates and of the CSP-Variables which allow to follow the logic without requiring any reconstruction.
For you, an (anti-) track was a set. It seems you now define it as a sequence (also copied from whips/braids).
No notion of length was ever defined in your "book" or referred to in your resolutions. You even claimed you didn't need any. And you are now obviously adopting mine (without saying it).

[Mauriès Robert]

One thing is sure, I build my resolutions before seeing yours, so I don't risk copying them either in the order of the steps or in the length of the sequences.

Totally beside the point. I never suggested you are copying my resolution paths one by one.

I already established long ago that anti-tracks are a downgraded (and much posterior) version of braids (or S-braids if you allow inner Subsets): lack of the left-linking candidates and of the CSP-Variables which allow to follow the logic without requiring any reconstruction.
For you, an (anti-) track was a set. It seems you now define it as a sequence (also copied from whips/braids).
No notion of length was ever defined in your "book" or referred to in your resolutions. You even claimed you didn't need any. And you are now obviously adopting mine (without saying it).

You are going astray Denis and you have wrong ideas, you have already made all these criticisms to me since I came on this forum to present and use the technique of the tracks, and we have largely explained ourselves. So we are not going to start debating about it again, just reread our exchanges where everything is already said !

[denis_berthier]

just reread our exchanges where everything is already said !

Absolutely. My conclusions there are beyond any discussion.

[DEFISE]

I found W-rating = 6 and my "Few-steps" program give a 5 steps solution in W7:

Single(s): 7r1c1, 2r4c6, 4r5c1, 2r6c9, 9r6c4, 4r9c8
Box/Line: 1c2b1 => -1r1c3 -1r2c3 -1r3c3
Box/Line: 6c9b9 => -6r7c7 -6r8c7 -6r9c7
whip[4]: r3n2{c1 c3}- c3n5{r3 r2}- c3n8{r2 r7}- b7n9{r7c3 .} => -2r8c1
Single(s): 2r8c8, 2r3c1
whip[7]: r6c5{n3 n1}- r5c6{n1 n5}- r8c6{n5 n3}- c1n3{r8 r2}- r1n3{c2 c7}- c8n3{r2 r7}- c2n3{r7 .} => -3r4c5
whip[5]: b6n3{r4c7 r5c9}- r4n3{c7 c1}- r2c1{n3 n9}- r2c9{n9 n1}- c7n1{r1 .} => -3r9c7
Single(s): 1r9c7, 1r2c9
whip[6]: b7n9{r7c3 r8c1}- r2c1{n9 n3}- c8n3{r2 r3}- r3c4{n3 n5}- r2n5{c4 c3}- r2n9{c3 .} => -9r7c8
Box/Line: 9c8b3 => -9r1c7 -9r3c7
whip[7]: r4c7{n3 n5}- r4c5{n5 n6}- c1n6{r4 r8}- b7n9{r8c1 r7c3}- r7n2{c3 c4}- r7n6{c4 c9}- r9c9{n6 .} => -3r5c9
Box/Line: 3c9b9 => -3r7c7 -3r7c8 -3r8c7
STTE