View previous topic :: View next topic |
Author |
Message |
| angusj Site Admin
| Joined: 18 Jun 2005 | Posts: 406 | : | | Items |
|
Posted: Wed Jun 14, 2006 11:49 pm Post subject: Re: possible bug in simple sudoku 4.1w?? |
|
|
frobozzz wrote: | **BUT** if I have the pair filter on the "Exclude several..." choice is dimmed out!! |
Yes, that's by design - if you have "Show all candidates while Filtering" unselected, as you evidently do. It strikes me a generally counter-intuitive to want to exclude multiple candidates while filtering. |
|
Back to top |
|
|
| rkral
| Joined: 21 Oct 2005 | Posts: 233 | : | | Items |
|
Posted: Fri Jun 30, 2006 1:52 pm Post subject: |
|
|
angusj,
I was surprised to see SS ver 4.1z hint a multi-coloring exclusion (r5c8<>9) before a hidden triple (378 in r5c246).
Code: |
*-----------*
|...|1.2|3.4|
|...|...|...|
|...|4.5|6.7|
|---+---+---|
|1.5|...|7.2|
|...|...|...|
|7.6|...|8.3|
|---+---+---|
|3.7|5.8|...|
|...|...|...|
|5.8|6.1|...|
*-----------*
*--------------------------------------------------------------------*
| 68 57 9 | 1 678 2 | 3 58 4 |
| 2468 57 234 | 3789 36789 3679 | 129 12589 15 |
| 28 1 23 | 4 389 5 | 6 289 7 |
|----------------------+----------------------+----------------------|
| 1 38 5 | 389 34689 3469 | 7 469 2 |
| 249 *38 24 |*23789 15 *34679 | 149 14569 156 |
| 7 249 6 | 29 15 49 | 8 15 3 |
|----------------------+----------------------+----------------------|
| 3 249 7 | 5 249 8 | 124 1246 16 |
| 249 6 1 | 379 23479 3479 | 5 37 8 |
| 5 24 8 | 6 37 1 | 24 37 9 |
*--------------------------------------------------------------------* |
|
|
Back to top |
|
|
| angusj Site Admin
| Joined: 18 Jun 2005 | Posts: 406 | : | | Items |
|
Posted: Fri Jun 30, 2006 11:01 pm Post subject: |
|
|
rkral wrote: | I was surprised to see SS ver 4.1z hint a multi-coloring exclusion (r5c8<>9) before a hidden triple (378 in r5c246). |
Yes, although multi-coloring is a more advanced technique (because its logic is more complicated to understand), I believe once the technique is mastered it's easier to apply. (I find hidden triples and hidden quads almost impossible to spot without help.) |
|
Back to top |
|
|
| rkral
| Joined: 21 Oct 2005 | Posts: 233 | : | | Items |
|
Posted: Sat Jul 01, 2006 3:33 am Post subject: |
|
|
angusj wrote: | rkral wrote: | I was surprised to see SS ver 4.1z hint a multi-coloring exclusion (r5c8<>9) before a hidden triple (378 in r5c246). |
Yes, although multi-coloring is a more advanced technique (because its logic is more complicated to understand), I believe once the technique is mastered it's easier to apply. |
If I had found your order of techniques posted 3 months ago, I would not have been surprised.
angusj wrote: | Naked Singles
Hidden Singles
Naked Pairs
Locked Candidates
Naked Triples
Naked Quads
Hidden Pairs
XWing
Swordfish
Colors
MultiColors
Hidden Triples
XYWing
Hidden Quads |
|
|
Back to top |
|
|
| ltuser
| Joined: 04 Jul 2006 | Posts: 1 | : | | Items |
|
Posted: Tue Jul 04, 2006 4:04 pm Post subject: |
|
|
I realize that SS hasn't been updated in over a year, so it may be that this suggestion is pointless, but ...
Why are Unique Rectangles -- at least the easier types -- among the tactics? They're *easier* to spot than naked pairs since you only need look for a pair that is in a box and in line -- much quicker to search the entire grid for URs than naked pairs -- I always look for both tactics at the same time -- and they're *very* common. In fact, by not including them, they through a wrench in the ratings SS assigns -- puzzles that "require" an xy-wing or swordfish or coloring -- much more difficult tactics to spot -- can often be solved by UR instead.
For example, at random, I pulled up the following Extreme from 4.1z:
Code: | *-----------*
|.9.|...|1.7|
|...|61.|.23|
|...|...|.6.|
|---+---+---|
|..3|1.9|6..|
|.6.|.7.|.3.|
|..8|5.6|7..|
|---+---+---|
|.5.|...|...|
|17.|.68|...|
|3.2|...|.1.|
*-----------* |
The hint system requires these tactics:
Naked Pairs, Triples, Quads, Locked Candidates 1+2, Coloring 4s, X-wing, Multiple Coloring 8s.
However, any reasonably adept solver would require only naked pairs and locked candidates to get to this point:
Code: |
*-----------*
|.9.|..3|1.7|
|..7|61.|923|
|231|...|56.|
|---+---+---|
|723|1.9|6..|
|.6.|.72|.31|
|418|536|792|
|---+---+---|
|.5.|..1|.7.|
|17.|.68|...|
|3.2|...|.16|
*-----------*
*-----------------------------------------------------------*
| 56 9 456 | 248 2458 3 | 1 48 7 |
| 58 48 7 | 6 1 45 | 9 2 3 |
| 2 3 1 | 4789 489 47 | 5 6 48 |
|-------------------+-------------------+-------------------|
| 7 2 3 | 1 48 9 | 6 458 458 |
| 59 6 59 | 48 7 2 | 48 3 1 |
| 4 1 8 | 5 3 6 | 7 9 2 |
|-------------------+-------------------+-------------------|
| 689 5 469 | 234 24 1 | 23 7 489 |
| 1 7 49 | 234 6 8 | 23 45 459 |
| 3 48 2 | 79 59 57 | 48 1 6 |
*-----------------------------------------------------------* |
There is now an ultra-obvious UR at r78c47 that eliminates the 4s from the rest of column 4, making an immediate placement of an 8 at r5c4. Only singles are required to finish the puzzle -- no coloring, no x-wing, no multiple coloring.
I'm not complaining that I need help spotting the UR -- but that puzzles that can be solved with UR as the toughest tactic cannot be reasonably rated as Extreme or even Expert. One loads up an Extreme in order to try to tackle a Swordfish, Coloring or XY-wing -- but cannot ignore the much simpler solution. This happens *very* often.
I know there are plenty of other advanced tactics that can be used to eliminate the need for Swordfish, X-wing, XY-wing, etc -- but they are all generaly just that -- advanced tactics -- that are most likely *more* difficult to apply -- so it isn't really the same thing. The relatively late discovery and rise to popularity of the UR is an anomoly and they cannot be considered an advanced or difficult tactic. The solver that finds a 5-cell forcing chain that skips past a swordfish and multi-coloring will probably *not* feel cheated, but instead feel clever for finding the chain.
Here's another one at random from SS:
Code: | *-----------*
|.9.|..2|7..|
|...|98.|..2|
|7..|...|.8.|
|---+---+---|
|.73|5..|..6|
|.8.|...|.1.|
|4..|..3|97.|
|---+---+---|
|.5.|...|..7|
|8..|.61|...|
|..6|8..|.4.|
*-----------* |
Simple Sudoku needs Locked Candidates, Naked Pairs, Coloring and Swordfish. All I needed was ONE UR and the rest was singles:
Code: | *-----------*
|.98|..2|7.4|
|..4|987|..2|
|7.2|...|.89|
|---+---+---|
|173|598|426|
|.89|7..|.1.|
|4.5|..3|978|
|---+---+---|
|.51|..9|867|
|847|.61|.9.|
|9.6|875|.41|
*-----------*
*-----------------------------------------------------------*
| 356 9 8 | 136 135 2 | 7 35 4 |
| 356 136 4 | 9 8 7 | 1356 35 2 |
| 7 136 2 | 1346 1345 46 | 1356 8 9 |
|-------------------+-------------------+-------------------|
| 1 7 3 | 5 9 8 | 4 2 6 |
| 26 8 9 | 7 24 46 | 35 1 35 |
| 4 26 5 | 126 12 3 | 9 7 8 |
|-------------------+-------------------+-------------------|
| 23 5 1 | 234 234 9 | 8 6 7 |
| 8 4 7 | 23 6 1 | 235 9 35 |
| 9 23 6 | 8 7 5 | 23 4 1 |
*-----------------------------------------------------------* |
r58c79 is a simple UR Type I allowing an immediate placement of 2 at r8c7. Only singles remain.
Oh, and isn't it time for an option to include the diagonals? |
|
Back to top |
|
|
| angusj Site Admin
| Joined: 18 Jun 2005 | Posts: 406 | : | | Items |
|
Posted: Tue Jul 04, 2006 11:03 pm Post subject: |
|
|
ltuser wrote: | Why are Unique Rectangles -- at least the easier types -- among the tactics? |
Because I believe Unique Rectangles techniques are based on an unreasonable premise - ie that you know there is only one solution. I believe that part of pleasure in solving Sudoku is proving that there is only one solution. To me, it's a bit like solving a maze by working from the finish to the start. Yes, I understand that others feel differently. |
|
Back to top |
|
|
| rkral
| Joined: 21 Oct 2005 | Posts: 233 | : | | Items |
|
Posted: Tue Jul 04, 2006 11:42 pm Post subject: |
|
|
angusj wrote: | I believe Unique Rectangles techniques are based on an unreasonable premise - ie that you know there is only one solution. I believe that part of pleasure in solving Sudoku is proving that there is only one solution. |
Since SS tells us there is only one solution (by not telling us there are multiple solutions), that argument seems weak. |
|
Back to top |
|
|
| Sartak
| Joined: 06 Apr 2006 | Posts: 3 | : | | Items |
|
Posted: Sat Jul 15, 2006 7:52 pm Post subject: |
|
|
I recently switched to Linux and I'm very happy to report that Simple Sudoku runs fairly well under WINE. I've encountered a strange bug -- when you minimize and restore the window, there's a glaring display error that makes the game unplayable. Here's a screenshot:
Oh well, otherwise it runs like a peach.
Edit: I'm not expecting you to fix this at all; this is certainly a WINE bug (considering this behavior has never happened that I know of under Windows). |
|
Back to top |
|
|
| Joey_Alva
| Joined: 03 Sep 2006 | Posts: 6 | : | Location: Sweden | Items |
|
Posted: Sun Sep 03, 2006 7:38 am Post subject: |
|
|
Hi there, just downloaded your program and started playing around with it, I'm kinda new to Sudoku so I have a question about an error message I get when entering a puzzle from a local newspaper.
I get the, puzzle is asymetrical do you want to continue? yes no error when I've completed filling in the puzzle and presses start.
What does this mean? Does it mean that the puzzle has more than one solution or that it gives too little information so that it is just a guessing game?
The puzzle I try to insert:
Code: |
*-----------*
|5.4|1.9|.3.|
|7.3|86.|4..|
|91.|.42|...|
|---+---+---|
|...|...|8..|
|.92|4..|.5.|
|.36|..8|71.|
|---+---+---|
|.87|.91|..2|
|.51|..4|96.|
|.4.|.53|1.7|
*-----------*
|
Is taken from http://www.dn.se a Swedish newspaper. |
|
Back to top |
|
|
| angusj Site Admin
| Joined: 18 Jun 2005 | Posts: 406 | : | | Items |
|
Posted: Sun Sep 03, 2006 10:23 am Post subject: |
|
|
Joey_Alva wrote: | I get the, puzzle is asymetrical do you want to continue? yes no error when I've completed filling in the puzzle and presses start. |
It's a warning, not an error. Most puzzles (especially from newspapers) are symmetrical. Therefore, if a puzzle is entered manually that is asymmetrical, then you will be given a warning in case you've made a simple keying in error. |
|
Back to top |
|
|
| falcons75scp
| Joined: 09 May 2006 | Posts: 4 | : | Location: Oklahoma | Items |
|
Posted: Mon Sep 18, 2006 3:07 pm Post subject: |
|
|
Angus,
Thanks again for creating such a useful and helpful program. I use it virtually every day.
I visited your web site this morning and was somewhat surprised to see an updated version of Simple Sudoku (4.2e - the previous that I had downloaded was 4.1z). I saw no mention of the update here. However, I've noted that the program behaves differently now upon startup. Previously, it started in the design mode. Now each time it starts with a new puzzle - is this intentional? Is there a startup option to prevent this?
I know that the program is freeware, but ........... do you take contributions? The program has real value to me (and to many others, I suspect). I would like to make a small donation if that is possible. This is not unheard of for freeware - perhaps others feel as I do.
Steve |
|
Back to top |
|
|
| angusj Site Admin
| Joined: 18 Jun 2005 | Posts: 406 | : | | Items |
|
Posted: Mon Sep 18, 2006 10:37 pm Post subject: |
|
|
falcons75scp wrote: | Previously, it started in the design mode. Now each time it starts with a new puzzle - is this intentional? Is there a startup option to prevent this? |
Hi Steve. Yes, this is intentional - I thought it would be less confusing for newbies. No, there's currently no option to startup in design mode.
Quote: | I know that the program is freeware, but ........... do you take contributions? The program has real value to me (and to many others, I suspect). I would like to make a small donation if that is possible. This is not unheard of for freeware - perhaps others feel as I do. |
Thanks for the thought but I'm happy just to receive the occasional (or not so occasional) thanks such as yours. |
|
Back to top |
|
|
| falcons75scp
| Joined: 09 May 2006 | Posts: 4 | : | Location: Oklahoma | Items |
|
Posted: Tue Sep 19, 2006 1:42 pm Post subject: |
|
|
Angus,
One of the great features of the program is being able to enter puzzles from outside sources and use the program as a solving aide. As you know, this REQUIRES design mode unless the user knows how paste a puzzle or how to create a file of the puzzle to load into the program.
Therefore, what might be confusing for newbies to Simple Sudoku is how to get OUT of the design mode once in it. Those of us who have been using Simple Sudoku for a while know that the F3 key does this. But (unless I missed it somewhere) there is nothing to clue newbies to the program of this feature.
Your thoughts?
Steve |
|
Back to top |
|
|
| angusj Site Admin
| Joined: 18 Jun 2005 | Posts: 406 | : | | Items |
|
Posted: Thu Sep 21, 2006 10:38 pm Post subject: |
|
|
falcons75scp wrote: | Your thoughts? |
Hi again Steve.
No thoughts really. Do you have suggestions?
Angus |
|
Back to top |
|
|
| falcons75scp
| Joined: 09 May 2006 | Posts: 4 | : | Location: Oklahoma | Items |
|
Posted: Tue Oct 03, 2006 4:54 am Post subject: |
|
|
Angus,
OK, I now see that F3 (start) shows up on the menu (apparently after 10 givens are entered). I frequently use Simple Sudoku for Killer Sudoku when that many givens is not initially available. Since Simple Sudoku is not actually designed for Killer Sudoku and there are (apparently) no valid regular Sudoku puzzles with fewer than 17 givens - the "problem" that I pointed out is a non-issue for newbies. In other words.................
Never mind.
Steve
|
|
Back to top |
|
|
|