The New Sudoku Players' Forum

[ArkieTech]

Code: Select all

 *-----------*
 |..4|.8.|...|
 |..2|1..|83.|
 |89.|..4|1..|
 |---+---+---|
 |.4.|5..|.73|
 |7..|.1.|4..|
 |..5|..7|6..|
 |---+---+---|
 |.17|.43|9..|
 |.8.|6..|.4.|
 |...|9..|...|
 *-----------*


Play/Print this puzzle online

[Leren]

Code: Select all

*----------------------------------------------------*
| 1     356f 4  | 237g 8    256 |a5-7a  2569  25679  |
|e56    7    2  | 1    56   9   | 8     3     4      |
| 8     9    36 | 237  2356 4   | 1     256   2567   |
|---------------+---------------+--------------------|
|f69    4    1  | 5    69   8   | 2     7     3      |
| 7    g236  8  | 23h  1    26  | 4     59    59     |
| 239   23   5  | 4    239  7   | 6     18    18     |
|---------------+---------------+--------------------|
|d256d  1    7  | 8    4    3   | 9    c256c c256c   |
| 235   8    9  | 6    257  125 |b357b  4     1257   |
| 4     2356 36 | 9    257  125 |b357b  18    125678 |
*----------------------------------------------------*

Kraken Row 5 Digit 3: (Read from right to left)

(7-5) r1c7 = 5r89c7 - 5r7c89 = 5r7c1 - (5=6) r2c1 - 6r4c1 = 6r5c2 - 3 r5c2;

(7-5) r1c7 = 5r89c7 - 5r7c89 = 5r7c1 - 5r2c1 = (5-3) r1c2 = 3r1c4 - 3 r5c4; => - 7 r1rc7, stte

Too late at night to turn it around into a linear AIC

Leren

[SpAce]

I have no idea if this is correct despite the recent education on the subject (*), but I have to try...

Code: Select all

.---------------.----------------.-----------------------.
| 1    356   4  | 237  8     256 | 5-7  *2569  *2569(+7) |
| 56   7     2  | 1    56    9   | 8     3      4        |
| 8    9     36 | 237  2356  4   | 1    *256   *256(+7)  |
:---------------+----------------+-----------------------:
| 69   4     1  | 5    69    8   | 2     7      3        |
| 7    236   8  | 23   1     26  | 4    *59    *59       |
| 239  23    5  | 4    239   7   | 6     18     18       |
:---------------+----------------+-----------------------:
| 256  1     7  | 8    4     3   | 9    *256   *256      |
| 235  8     9  | 6    257   125 | 357   4      125-7    |
| 4    2356  36 | 9    257   125 | 357   18     12568-7  |
'---------------'----------------'-----------------------'

MUG(2569)r1357c89+2 using internals

(7)r13c9 => -7 r1c7,r89c9; stte

(*) Edit. I think it might actually be correct (unless someone tells me it's not). Without the internals we'd have a square quad in b3 and two identical and parallel quads in c89, which sure looks like a MUG to me. I can't see any way to disrupt it, as any external placements would always remove both instances of the same digit from the pattern rows, thus retaining the paired quads and possibly introducing URs, BUG-Lites, or zero-solution states.

...in case that doesn't work, here's a backup:

Code: Select all

.-------------------.-----------------.-------------------.
|   1     a356   4  | b237  8     256 | 57   2569  25679  |
| a[5]-6   7     2  |  1    56    9   | 8    3     4      |
|   8      9     36 |  237  2356  4   | 1    256   2567   |
:-------------------+-----------------+-------------------:
| d(6)9    4     1  |  5    69    8   | 2    7     3      |
|   7     d236   8  | c23   1     26  | 4    59    59     |
|   239    23    5  |  4    239   7   | 6    18    18     |
:-------------------+-----------------+-------------------:
|   256    1     7  |  8    4     3   | 9    256   256    |
|   235    8     9  |  6    257   125 | 357  4     1257   |
|   4      2356  36 |  9    257   125 | 357  18    125678 |
'-------------------'-----------------'-------------------'

(53)b1p42 = r1c4 - r5c4 = (36)b4p51 => -6 r2c1; lcte

(That can be easily extended into stte using b9 (as in Leren's solution), but it just gets longer for no good reason.)

[Ngisa]

Code: Select all

+--------------------+---------------------+------------------------+
| 1      c356     4  |b237     8       25-6| 57      2569    25679  |
|d56      7       2  | 1      e56      9   | 8       3       4      |
| 8       9       36 | 237     2356    4   | 1       256     2567   |
+--------------------+---------------------+------------------------+
| 69      4       1  | 5       9-6     8   | 2       7       3      |
| 7       236     8  | a23     1      a26  | 4       59      59     |
| 239     23      5  | 4       239     7   | 6       18      18     |
+--------------------+---------------------+------------------------+
| 256     1       7  | 8       4       3   | 9       256     256    |
| 235     8       9  | 6       257     125 | 357     4       1257   |
| 4       2356    36 | 9       257     125 | 357     18      125678 |
+--------------------+---------------------+------------------------+


(6=2)r5c46 - r1c4 = (3-5)r1c1 = r2c1 - (5=6)r2c5 => - 6r1c6,r4c5; stte

Clement

[SpAce]

(6=23)r5c46 - r1c4 = (3-5)r1c1 = r2c1 - (5=6)r2c5 => - 6r1c6,r4c5; stte lcte

Hi Clement! Just a couple of typos (colored corrections above). I guess "lclste" is the standard term but since it's just one pointing pair, I think "lcte" is more descriptive.

[SteveG48]

Actually, I've wondered why we don't use btte (basics to the end) instead of lclste.

[SteveG48]

Code: Select all

 *------------------------------------------------------------------------------*
 |  1      d356     4       |e237     8       256     | 57      2569    25679   |
 |cg56      7       2       | 1       56      9       | 8       3       4       |
 |  8       9      c36      | 237     2356    4       | 1       256     2567    |
 *--------------------------+-------------------------+-------------------------|
 | g69      4       1       | 5      f69      8       | 2       7       3       |
 |  7       236     8       |f23      1      f26      | 4       59      59      |
 |  239     23      5       | 4       239     7       | 6       18      18      |
 *--------------------------+-------------------------+-------------------------|
 |bh26-5    1       7       | 8       4       3       | 9     ai256   ai256     |
 |  235     8       9       | 6       257     125     | 37-5    4       127-5   |
 |  4       2356    36      | 9       257     125     | 37-5    18      12678-5 |
 *------------------------------------------------------------------------------*


5r7c89 = r7c1 - (5=36)b1p49 - 3r1c2 = r1c4 - (3=269)b5p246 - (9=56)r24c1 - 5r7c1 = 5r7c89 => -5 r7c1,b9p4679 ; stte

[Wecoc]

Actually, I've wondered why we don't use btte (basics to the end) instead of lclste.

Well, "basics to the end" sounds a bit vague to me.
Some consider all of the simple sudoku set as basic. Others consider quads as advanced. It can vary depending on context, too.
stte and lclste are more static definitions.

[eleven]

MUG(2569)r1357c89+2 using internals

Nice. (T8)

[SpAce]

MUG(2569)r1357c89+2 using internals

Nice. (T8)

Thanks! A lucky catch. I'm pretty sure it was the first non-trivial MUG I've spotted, and it wouldn't have happened without our discussion. It was nice of Dan to provide a live specimen right after we talked about them, and also nice of everyone else to leave it for me to catch No better way to reinforce learning than getting to put theory into practice right away. I still can't promise I'll be spotting them with much certainty, but at least this gives some hope.

[SpAce]

Actually, I've wondered why we don't use btte (basics to the end) instead of lclste.

Well, "basics to the end" sounds a bit vague to me.
Some consider all of the simple sudoku set as basic. Others consider quads as advanced. It can vary depending on context, too.
stte and lclste are more static definitions.

I used to wonder about the same, but I tend to agree with Wecoc. It has been seen even here that not everyone has the same idea about "basics", but "lclste" is unambiguous as long as one knows what it means: locked candidates and/or locked sets (and singles, of course). Also, the latter allows dropping the "ls" if locked sets aren't needed, which gives a more accurate picture about what remains to be solved, though it isn't quite standard.

To me the non-standard "lcte" is somewhere between stte and lclste, but actually closer to stte since locked candidates are almost as easy to spot as singles. On the other hand, some real lclste finishes can be quite tedious if many locked sets remain, especially if they're larger than pairs. Technically we can also use "lste" if only locked sets remain, but that distinction doesn't usually make as much difference.

PS. Besides aiming for stte finishes, there's also a possibility for an additional challenge I once suggested when someone posted puzzles that were solvable with just basics. What if singles were the only basic move allowed both before and after the main move(s)? That would make many puzzles considerably harder, or outright impossible to solve as one-steppers (even if they're otherwise solvable with basics alone). Perhaps not a very interesting challenge as no one would really solve that way, but might be fun to try sometime.

[SteveG48]

Actually, I've wondered why we don't use btte (basics to the end) instead of lclste.

Well, "basics to the end" sounds a bit vague to me.
Some consider all of the simple sudoku set as basic. Others consider quads as advanced. It can vary depending on context, too.
stte and lclste are more static definitions.

I used to wonder about the same, but I tend to agree with Wecoc. It has been seen even here that not everyone has the same idea about "basics", but "lclste" is unambiguous as long as one knows what it means: locked candidates and/or locked sets (and singles, of course). Also, the latter allows dropping the "ls" if locked sets aren't needed, which gives a more accurate picture about what remains to be solved, though it isn't quite standard.

Yes, basics is ambiguous in general, but my point is that in this forum we've chosen to define basics in a particular way to get everyone at the same starting point. In this context, btte would mean the same set of moves.

[SpAce]

Yes, basics is ambiguous in general, but my point is that in this forum we've chosen to define basics in a particular way to get everyone at the same starting point. In this context, btte would mean the same set of moves.

That would work for me. It would look cleaner and its meaning would probably be easier to guess too (if one knows what stte means)— lclste makes zero sense for someone who hasn’t looked up the definition, and it’s easy to misspell too. Then again, btte only differs from stte by one letter so the distinction might be easier to miss (not that it hugely matters).

If you’re officially suggesting this change or option, I vote Yes (unless someone provides good counter-arguments). One counter-argument might be that it conflicts with historical solutions, and not everyone would probably start using it with new ones either, so passive readers would have to learn two terms for the same concept and understand they’re synonyms. That doesn’t bother me much, though.

[SteveG48]

Yes, basics is ambiguous in general, but my point is that in this forum we've chosen to define basics in a particular way to get everyone at the same starting point. In this context, btte would mean the same set of moves.

That would work for me. It would look cleaner and its meaning would probably be easier to guess too (if one knows what stte means)

The easier to guess part is what I was thinking. In any case, it really doesn't matter much. We get along pretty well.

When I first came here I enjoyed a few days trying to guess what lclste meant before I finally asked. I had managed stte earlier, but I would probably never have guessed lclste.

[SpAce]

The easier to guess part is what I was thinking. In any case, it really doesn't matter much. We get along pretty well.

Yes, it's definitely not a high priority. Then again, I don't mind evolution when better ideas come around. Insisting on backwards compatibility is a nuisance in many cases.

When I first came here I enjoyed a few days trying to guess what lclste meant before I finally asked. I had managed stte earlier, but I would probably never have guessed lclste.

The same, though I don't think I guessed even stte. I probably found some reference pretty early on that mentioned both, so I'm not sure if I had to ask, but even so I know I've had to look up the standard spelling of "lclste" several times when I've actually used it myself ("btte" would have been much easier to remember and type correctly). Without the reference or asking, at least lclste would have remained a complete mystery -- especially since I didn't even know what "locked candidates" meant when I joined the party (I still think it's a very unintuitive name -- why not something like "locked group" which would go well with "locked set" and allow the plural form too?).