The New Sudoku Players' Forum

[enxio27]

Following up from here: post263544.html#p263544

it seems that I will not be able to continue any more due recently changes in rules that I unable to follow/understand and miss a lot of learning material too

What rules are you referring to? I don't know of any rules that would not permit new users to participate in the forums.

[Leren]

Even before enxio27's post I was tempted to respond to rjamil's post because he seems confused about the forum rules for responding to puzzle postings, in particular the One Trick Pony puzzles posted by Dan over many years and more recently by 200e200w.

So, rjamil and others, here are the rules for responding to One Trick Pony and other puzzle postings :

There are no rules !!!! (other than the general forum rules that you should reply in English and not use offensive language, or otherwise disparage anyone).

However, there is a convention that has been used for many years and has proven popular. So here, yet again, is a potted description of the convention.

The challenge for a responder is to solve the puzzle with one non-basic move, if possible.

Basic moves, in this context, include naked and hidden singles, pointing and claiming intersections and naked and hidden pairs, triples and quads.

You are supposed to apply as many basic moves as possible to the puzzle until they are exhausted.

At this point, you look for a single move that will provide further eliminations/assignments that will bring the puzzle solved status to a point where it can be fully solved only by further applications of basics.

If the trailing basics include only naked and hidden singles, all you have to say after explaining your move is stte (Singles To the End).

If the trailing basics include anything else in the list you say lclste (I think this means Locked Candidates/Locked Sets To The End).

stte and lclste finishes are supposed to be of equal merit, but nobody really believes this

I don't think this convention has changed just because the puzzle poster has changed. The regular responders have been at this for so long, that following this convention is like a reflex action, you just do it without even thinking about it.

You are also quite free to respond with a multiple non basic move solution if you want to, because, as I said, there are no "rules" to stop you doing this. The whole idea of this convention is just to have fun solving puzzles.

For the Nightmare series, multiple non-basic solutions will be quite common, as these puzzles are supposed to be harder than the One Trick Ponies.

That's about it. Hope this clears up any misunderstandings about "changed rules".

Leren

PS I was wondering if this could be tidied up and made into a "sticky". This has come up up several times over the years and it will not doubt come up again. I must have gone through all this 5 or 6 times. It would be easier to refer new/confused users to the sticky.

Leren

[SteveG48]

I was wondering if this could be tidied up and made into a "sticky". This has come up up several times over the years and it will not doubt come up again. I must have gone through all this 5 or 6 times. It would be easier to refer new/confused users to the sticky.

Leren

Yes, indeed. Excellent idea.

[enxio27]

Out of curiosity, what is the range of "non-basic" moves used in these One-Trick Pony puzzles? I've pretty much stayed away because I figured that one non-basic move would be too advanced for me (anything more than swordfish is beyond my current skill level). Maybe I was wrong.

[200e200w]

Out of curiosity, what is the range of "non-basic" moves used in these One-Trick Pony puzzles? I've pretty much stayed away because I figured that one non-basic move would be too advanced for me (anything more than swordfish is beyond my current skill level). Maybe I was wrong.

When I am creating a One-Trick Pony, I am ensuring that it can be solvable with one non-basic technique, and this technique is at the AIC level (7.5-7.9 in my rating system). I remove puzzles that have a one-step solution consisting of, for example, complex fish (Franken and Mutant Fish, 9.0-9.5 in my rating system), or Krakens (10.0-12.0 in my rating system). Usually, they are rated in my rating system between 7.2 and 7.7, and because your skill level in my rating system could be rated 3.4, they would be too much challenge for you, enxio27, even to solve without guessing. Here there are solving techniques used by Hodoku for OTP #7 (except from singles):
2 x Pointing
1 x Claiming
1 x Naked Quad
1 x Skyscraper
2 x 2-String Kite
1 x Empty Rectangle
1 x Hidden Unique Rectangle
1 x Alternating Inference Chain
Rating: 7.7/1500 (My rating system/Hodoku)

200e200w

[Leren]

enxio27 wrote : Out of curiosity, what is the range of "non-basic" moves used in these One-Trick Pony puzzles?

Technically, it's everything else except basics, with one curious exception - basic fish. One quirk of the system is that basic fish never get used. All other varieties, finned, mutant and kraken are OK.

Also, Dan usually left out some of the easier moves like Swordfish, Kites, ER's and XY wings. We'd generally feel a bit let down if a puzzle solved with just one of these, but over the years a few snuck through the net.

I see that 200e200w has removed exotic fish in his One trick Ponies. Well, he can do that if he wants, because there ... are ... no ... rules !!!

Also, as you are the starter of this thread, could you consider changing the title to something like One Trick Pony Convention or similar. As I said, there are forum rules about using English, not swearing, not demeaning anybody etc

Leren

[200e200w]

I see that 200e200w has removed exotic fish in his One trick Ponies.

I removed them, because I think a One-Trick Pony should be easy to solve with a single non-basic technique not only by skilled players, but also by average and slightly-above-average players like me. So, in seeking for One-Trick Ponies, I always seek for puzzles that can be solved using a single non-basic technique that is not rated above 8.5 in my rating system, for example, AIC or Finned Fish. In the next post there will be an attachment with a list of single-step solutions for each of 7 currently published OTP's, sorted by technique used.

200e200w

[Leren]

Hi 200e200w,

We're all grateful that you have taken over the One trick Pony provider role from Dan, so you are free to filter your puzzles any way you want and you don't have to justify yourself.

I think I've said this before, there are no rules in this business.

Leren

[enxio27]

Also, as you are the starter of this thread, could you consider changing the title to something like One Trick Pony Convention or similar.

Done. I was thinking the same thing.

[200e200w]

And so,
This is not the list of all one-step solutions. This is the OTP statistic, in which you will see, which techniques are used by one-step solutions, and how many times. Statistics generated by Hodoku.

200e200w

[SteveG48]

Hi 200e200w,

We're all grateful that you have taken over the One trick Pony provider role from Dan, so you are free to filter your puzzles any way you want and you don't have to justify yourself.

I think I've said this before, there are no rules in this business.

Leren

Agreed. With Dan's puzzles, most could be solved with a fairly simple "trick", but there were always the occasional tough ones. I liked having those pop up from time to time.

[Leren]

I'm writing up a sticky on this issue and will have it published by Jason in due course. If you have comments about the wording. Post them on this thread.

Leren

[rjamil]

Hi everyone,

Subject: January 25, 2018

Hi rjamil,
First, when you post a solution that use solving techniques based on chains (like 2-String Kite, Empty Rectangle, etc.), you need to post the chain in the notation like this (my solution):

(8=9)r1c5 - r3c5 = (9-7)r3c9 = r4c9 - (7=6)r4c3 - r3c3 = (6-9)r2c1 = (9-2|7)r1c1 = (27)r1c38 => -8 r1c3; stte

Some people here will not understand the notation like this (your solution):

2-String Kite Strong Link Row 6r4c39 Column 6r25c1 => -6r2c9;
Grouped Empty Rectangle Intersection 7r4c1 Strong Link ERIs 7r48c7 => -7r8c1; stte.

It is really important to use notation that is readable to newcomers, like the notation that I use (which is used by many people on this forum). I didn't understand your notation of the Grouped Empty Rectangle, especially because you didn't present it on the board.
Second, it is also really important to present the chain on the board in the correct form. The form which you used is incorrect. The correct form for your chain presented on the board could be:

Code: Select all

.------------------.------------------.------------------.
| 279   3     78   | 4     89    6    | 5     27    1    |
| d269  24    1    | 7     3589  35   | 238   246   239-6|
| 5     47    678  | 2     389   1    | 378   467   3679 |
:------------------+------------------+------------------:
| 1     5     b67  | 9     2     8    | 4     3     a67  |
| c36   8     4    | 135   7     35   | 9     126   26   |
| 37    9     2    | 13    6     4    | 17    5     8    |
:------------------+------------------+------------------:
| 8     17    35   | 6     135   2    | 137   9     4    |
| 27    127   35   | 8     4     9    | 6     17    35   |
| 4     6     9    | 35    135   7    | 123   8     235  |
'------------------'------------------'------------------'


200e200w

Saw your advise. Allow me to clear my position.

Basically, I am not a Sudoku player. I am a programmer.

I am developing a solver and now it is able to solve few otp puzzles. I started participating only such puzzles that are solvable by my solver program. However, I also cross checked with Hodoku, SodoCue, Xsudo, etc., freely available software too.

I am neither expert in eureka notation nor interested to learn too. My intention is to solve with speed and avoid backtracking.

If any junior and/or senior member find myself unfit for puzzle topic, I apologize and stay away in future.

R. Jamil

The above correspondence is basically my response to 200e200w in pm recently.

What's in my mind is that, being a programmer, testing my solver with such OTP puzzles and try to share output in any non-specific (non-eureka) format.
Sometime, I skip some strategies in my program in order to produce one (next) move to solve/crack the puzzle. and test if any strategy check before or after other strategies will effect the solver to become faster for other puzzles too. But its totally my own idea and yet to decide any conclusion.

R. Jamil

[rjamil]

Hi again,

Subject: January 7, 2018

Code: Select all

..87..53....1..82..1...5..4.7...1..9..4.7.6..1..5...8.9..4...6..41..2....23..79..

After 22 singletons:

Code: Select all

 +------------+---------------+--------+
 | 4    69 8  | 7    2    69  | 5 3 1  |
 | 7    39 5  | 1    349  349 | 8 2 6  |
 | 236  1  26 | 368  368  5   | 7 9 4  |
 +------------+---------------+--------+
 | 2368 7  26 | 2368 368  1   | 4 5 9  |
 | 258  58 4  | 289  7    89  | 6 1 3  |
 | 1    36 9  | 5    346  346 | 2 8 7  |
 +------------+---------------+--------+
 | 9    58 7  | 4    358  38  | 1 6 2  |
 | 568  4  1  | 689  5689 2   | 3 7 58 |
 | 568  2  3  | 68   1    7   | 9 4 58 |
 +------------+---------------+--------+

1. Pointing and claiming intersection removal b7c1 Clue 6;
2. Row wise Sword Fish Cells r267c256 Clue 3;
3. Naked pair r34c5 Clues 68;
4. Pointing and claiming intersection removal b8c4 Clue 6;
5. Row wise Jelly Fish Cells r3489c1459 Clue 8;
6. XY-Wings r5c6, r4c5, r1c6 Clue 6; stte.

Too many easy steps.

R. Jamil

Hello rjamil,
Using your listed PM above, the forum's normal solving rules are to find a single "advanced" move that provides a stte finish. The accepted "basics" are described in leren's post here: http://forum.enjoysudoku.com/june-5-2016-t33143.html#p251330. I think your first five steps listed above would fall under leren's definition, so there could be only one additional (advanced) move available. Your PM can also be derived using Andrew Stuart's online solver and its first six basic steps. Those steps are generally understood to be available and are not normally listed in one's solution. Hope this helps!

SteveC

Hi SteveC,

Many thanks for pointing me the right direction.

Since I have recently started participating Puzzles thread, haven't know the rules.

This means that only one advanced (XY-Wings) move is sufficient to crack the said puzzle.

FYI, I use my solver, as it is now search Naked/Hidden single, pair, triplet, quad, Box-Line/Pointing and Claiming intersection removals, X-Wings, Sword Fish, Jelly Fish, Skyscraper/Grouped Skyscraper (under finalizing), 2-String Kite/Grouped 2-String Kite, Empty Rectangle, Dual Linked Empty Rectangle, Grouped Empty Rectangle, Dual Linked Grouped Empty Rectangle, XY-Wings Type 1 and 2, XY-Wings Type 1 Transport (exemplar 1 - 32), XYZ-Wings, XYZ-Wings Hybrid Type 1 and 2, W-Wings Type 1 and 2, WXYZ-Wings Type 1, 2a, 2b, 3, 4a and 4b, with Trial and Error (guess).

I will follow the directions as per leren's post in future.

R. Jamil

Hope, that I conveyed my position/doubt.

R. Jamil