Help to bring an SE 9.3 [Correction 9.5] to around SE 7.5

Post the puzzle or solving technique that's causing you trouble and someone will help

Re: Help to bring an SE 9.3 to around SE 7.5

Postby denis_berthier » Thu Nov 12, 2020 3:25 pm

DEFISE wrote:here is only a summary of my resolution at depth 2, because my program does not output the exact syntax of whips (only right links):
1) I prove that (puzzle + hint 7L3C5) has no solution, using several whips[n], n <= 10.
2) I delete the candidate 7L3C5
3) I solve the puzzle using several whips[n], n <= 10.

OK, I can reproduce this with SudoRules, with the same max lengths.
As of now, it seems to be the simplest solution - n8r3c5 is a whip[10]-backdoor. But one may ask, why should you try precisely this candidate?
Last edited by denis_berthier on Fri Nov 13, 2020 5:51 am, edited 1 time in total.
denis_berthier
2010 Supporter
 
Posts: 1983
Joined: 19 June 2007
Location: Paris

Re: Help to bring an SE 9.3 to around SE 7.5

Postby DEFISE » Thu Nov 12, 2020 6:14 pm

denis_berthier wrote:As of now, it seems to be the simplest solution - n7r3c5 is a whip[10]-backdoor. But one may ask, why should you try precisely this candidate?


I chose the 78r3c5 pair because it was the best start found by my DFS optimization program (with criterion = minimum number of tree ends). See Robert's post above.
This is not a very rational justification, especially because the branches of my DFS tree are developed using “basic techniques” i.e. alignements (whip [1]) and subsets.
Let's say I'm lucky !
N.B: I suppose you wanted to write "n8r3c5 is a whip[10]-backdoor" (not n7r3c5).
DEFISE
 
Posts: 68
Joined: 16 April 2020
Location: France

Re: Help to bring an SE 9.3 to around SE 7.5

Postby denis_berthier » Fri Nov 13, 2020 5:51 am

DEFISE wrote: I suppose you wanted to write "n8r3c5 is a whip[10]-backdoor" (not n7r3c5).

OK, thanks. I corrected.
denis_berthier
2010 Supporter
 
Posts: 1983
Joined: 19 June 2007
Location: Paris

Previous

Return to Help with puzzles and solving techniques