The New Sudoku Players' Forum

[eleven]

Hurray! We finally got there! We are agreed your implication 'stream' is not a stream but is disjointed and has gaps in it - and that's from my efforts to put your first example into words.

I never said, that this notation is a stream in your sense (i don't even know, what a stream in your sense is), and it does not have "gaps", given that you are able to find out by yourself, that if you have a number a in a unit and a bivalue cell ab in the same unit, that this must be b then. Too hard ?

Because you were clever enough to 'ski off-piste' in your first week, you wrongly believe all beginners should be able to. But they need nursery slopes first which is what Eureka AICs gives them.

They do not. It is easier, if they leave Eureka AIC off-piste for a long time. That is my opinion. Yeah.

Just keep to the framework of alternating links and plug-in the Booleans as you find them. You are free to express the Booleans as you like provided the reader is able to make sense of them – after all, like you, they are all intelligent. So why is it that you need any further explanation?

The "framework of alternating links" is not needed at all, it is just a special way to see the things. And you just have to look at the pains of all the AIC newcomers to learn the notation over 10 years in different forums to be very sure, that further explanation is needed.
Maybe people were too shy or too polite, but in all the time i have posted in my notation i cannot remember, that anyone had a problem to understand it. But often i was offended by narrow-minded AIC purists.

All this time you have been a (lazy) mischief maker haven't you? So can we call a stop to it now?

Had a bad day ? I only answered your criticism of my notation.

Finally let me admit I'm not as smart as you. I need the discipline of AICs and rigorous checking they allow to help me analyse patterns such as Junior Exocets to find the eliminations they provide. That's is what I meant when I said they would take users on a longer journey.

Yes, this was a fine work. But i did not think, that you could really believe, that it was this notation, which let you find all this. It was the picture.

[DonM]

[
[ For DonM's benifit: I most emhaptically did find this manually.
However, I don't consider myself to be a "good" manual solver, or even a "typical" manual solver. ]

I don't doubt that it is a manual solution. There are some very obvious giveaways that a solution is computer solver-derived. What I respect about manual solvers more than anything else including whether they are 'good', 'great', 'typical' or whatever is both their having the courage to expose their solutions to the microscope of others (knowing that there might be errors) and having the tenacity to find their own solution even though they know they could likely run the puzzle thru a computer solver and get a quick (albeit generic/unoriginal) fix.

[blue]

[
[ For DonM's benifit: I most emhaptically did find this manually.
However, I don't consider myself to be a "good" manual solver, or even a "typical" manual solver. ]

I don't doubt that it is a manual solution. There are some very obvious giveaways that a solution is computer solver-derived. What I respect about manual solvers more than anything else including whether they are 'good', 'great', 'typical' or whatever is both their having the courage to expose their solutions to the microscope of others (knowing that there might be errors) and having the tenacity to find their own solution even though they know they could likely run the puzzle thru a computer solver and get a quick (albeit generic/unoriginal) fix.

Thanks Don. I see I was so emphatic about it, that I spelled it wrong

[sultan vinegar]

[ BTW: Thank you sultan_vinegar, for offering to discuss it. I wish I had had the time to respond that day. ]

...

If it's of any consequence, my opinion on the "legal strong link or not" issue, was that yes it was perfectly legal -- perhaps even new and inovative (although I had my reservations on how new it was) -- and that given that, what I had presented, was a perfectly legal AIC.

While I'm away, I wonder if you have (or anyone else has) any thoughts on that specific issue.

Hi blue, no problem with the delay. I actually misread your initial post about the strong link between the two ALS; I thought you wanted to directly link the ALS in c3 with the AALS in c7, hence my ridiculous AMSLS! Now I see that is not the case, so please ignore the AMSLS.

If we're keeping a scorecard of legal vs non-legal for your proposed strong link, then mark me down in the legal camp (with the (1|7) correction that you mentioned included in the chain of course). For what it's worth, I would write the chain with no passenger candidates (see my previous comments explaining why) as:

(45)r27c3 = (17)r2c239 - (17#1 = 3)r1247c7 - (3 = 9)r79c38 => r9c3 <> 4,5.

Given that a = b means at least one of a and b are true, why can't the doubters provide a counter-example in which both (45)r27c3 and (17)r2c239 are false?

The only spoiler to (45)r27c3 is if r2c3 = 1. The only spoiler to (17)r2c239 is if r2c3 = 4. Both spoilers can't be true given that they are different candidates that share a cell. Hence at least one of (45)r27c3 and (17)r2c239 is true.

Well done creating this innovative grouped strong link

[David P Bird]

Eleven, I'm sorry but the language barrier between us is bigger than I thought. I don't know how much that has affected our conversation and how much has been caused by you playing games with me.

SteveG48, there's no harm in showing candidates in an ANS in an order that suited you, but I wouldn't do it myself.

------------------------------

All, I'm sorry, it was foolish of me to expect anyone to express an opinion on what's best for the forum when they are mainly concerned with their own points of detail or in in-fighting.

Nevertheless I hope I have given everyone food for thought as they try to achieve the changes or status they seek.

David P Bird

[daj95376]

Code: Select all

+---------------------+----------+----------------------+
| 4568  9     2458    | 1  36  7 | (456)   2456    235  |
| 46    (37)  (14)    | 5  9   2 | (1467)  8       (13) |
| 56    37    125     | 8  36  4 | 9       2567    1235 |
+---------------------+----------+----------------------+
| 3     2     89      | 4  7   1 | (56)    569     58   |
| 19    5     7       | 6  2   8 | 13      39      4    |
| 18    4     6       | 3  5   9 | 17      27      128  |
+---------------------+----------+----------------------+
| 7     1     (45)*   | 2  8   6 | (345)   (3)45*  9    |
| 2     6     3       | 9  4   5 | 8       1       7    |
| 459   8     (9)-45* | 7  1   3 | 2       45*     6    |
+---------------------+----------+----------------------+

45r27c3 = 137r2c239 - (17=3456)r1247c7 - 3r7c8 =UR= 9r9c3 => -45r9c3


Why stop at merging the first two ALS terms? You could merge the two ALS terms and the AALS term into:

Code: Select all

45r27c3 =r2c29= 3r1247c7 - 3r7c8 =UR= 9r9c3 => -45 r9c3


However, I'll stick with:

Code: Select all

(45=1*)r27c3 - (1=7)r2c29 - (*17=3)1247c7 - 3r7c8 =UR= 9r9c3 => -45 r9c3


I know it should be (*1|7=3)1247c7, but I've decided that the reader has some responsibility towards understanding fundamental, multi-value interactions.

_

[ronk]

Getting back to the original topic of this thread, ArchieTech's Aug 5th puzzle:

Code: Select all

+---------------------+----------+----------------------+
| 4568  9     2458    | 1  36  7 | (456)   2456    235  |
| 46    (37)  (14)    | 5  9   2 | (1467)  8       (13) |
| 56    37    125     | 8  36  4 | 9       2567    1235 |
+---------------------+----------+----------------------+
| 3     2     89      | 4  7   1 | (56)    569     58   |
| 19    5     7       | 6  2   8 | 13      39      4    |
| 18    4     6       | 3  5   9 | 17      27      128  |
+---------------------+----------+----------------------+
| 7     1     (45)*   | 2  8   6 | (345)   (3)45*  9    |
| 2     6     3       | 9  4   5 | 8       1       7    |
| 459   8     (9)-45* | 7  1   3 | 2       45*     6    |
+---------------------+----------+----------------------+

using UR <45>r79c38:

45r27c3 = 137r2c239 - (17=3456)r1247c7 - 3r7c8 =UR= 9r9c3 => -45r9c3; stte

I find the AHS version of this easier to understand.

(9)r9c3 =AUR= (3)r79c8 - (3)r7c7 = (3-1)r5c7 = (hp17)r26c7 - (hp46)r2c17 = (45)r27c3 ==> r9c3<>45

[sultan vinegar]

Why stop at merging the first two ALS terms? You could merge the two ALS terms and the AALS term into:

Code: Select all

45r27c3 =r2c29= 3r1247c7 - 3r7c8 =UR= 9r9c3 => -45 r9c3


That's my AMSLS from page 2. I'm not sure about the =r2c29= qualifier though? If you are listing the houses of the A*LS then it should be =r2c37=. Perhaps it means something else?

I find the AHS version of this easier to understand.

(9)r9c3 =AUR= (3)r79c8 - (3)r7c7 = (3-1)r5c7 = (hp17)r26c7 - (hp46)r2c17 = (45)r27c3 ==> r9c3<>45

I like that too. Your final link between the hp46 and np45 captures the vibe of the "new" type of link that blue was advocating, but written a different way. Gee, if we didn't have this rich diversity in thinking and notation then we wouldn't be getting all these wonderful solutions!

[daj95376]

That's my AMSLS from page 2. I'm not sure about the =r2c29= qualifier though? If you are listing the houses of the A*LS then it should be =r2c37=. Perhaps it means something else?

Sorry, but I haven't been following all of the details in earlier posts. So, I missed your AMSLS.

When I merged my A*LS terms, I only included the cells from the second ALS.

Code: Select all

  1st ALS       2nd ALS         AALS
============   ==========   =============


(45=1*)r27c3 - (1=7)r2c29 - (*17=3)1247c7 - 3r7c8 =UR= 9r9c3 => -45 r9c3


45r27c3            =r2c29=       3r1247c7 - 3r7c8 =UR= 9r9c3 => -45 r9c3


This "condensing of terms" reminds me of when I use to write XY-Chains as:

Code: Select all

(n=m)cell_a   list_of_cells   (k=n)cell_z  =>  -n peer_cells


This idea went over like a lead balloon. Maybe blue will have better luck.

_

[blue]

I'll post this: it's the first part of something that I've been writing, but that I won't be able to get back to for several hours
Dinner and usual Friday night card game are calling, and I can't escape either

It has been noticeable that the advocates for change were remarkably silent over the first few days of this discussion, but now two Blue, and DAJ, are contributing. What I'd like to know from them (and the others) is do they accept that a compromise standard is sensible or do they want to proceed with one customised to their specific needs?

<snip>
I have much to say now, and (again) no time to say it.
I'll be back later (today) with time on hand.

<snip>

This was my first post in this thread ... a week ago now ! The colored highlights are new.
<snip>
While I'm away, I wonder if you have (or anyone else has) any thoughts on that specific issue.

Last things first:

Why stop at merging the first two ALS terms? You could merge the two ALS terms and the AALS term into:

Code: Select all

45r27c3 =r2c29= 3r1247c7 - 3r7c8 =UR= 9r9c3 => -45 r9c3


However, I'll stick with:

Code: Select all

(45=1*)r27c3 - (1=7)r2c29 - (*17=3)1247c7 - 3r7c8 =UR= 9r9c3 => -45 r9c3


I know it should be (*1|7=3)1247c7, but I've decided that the reader has some responsibility towards understanding fundamental, multi-value interactions.

I find the AHS version of this easier to understand.

(9)r9c3 =AUR= (3)r79c8 - (3)r7c7 = (3-1)r5c7 = (hp17)r26c7 - (hp46)r2c17 = (45)r27c3 ==> r9c3<>45

Danny asks why I didn't I go farther, and introduce some (I think) new notation, that David would think was some "new notation, contrived merly to suit my needs", but instead, would stick with what David would call a "memory chain", and Denis would say "no, it doesn't use memory, it uses left to right AND splitting".

Ron prefers a chain that ends with a quantum naked pair.

"No thank you please", to both ... with all due respect.

If I had to change something, I'ld go with this:

(9=45)r79c3 - (4=137)r2c239 - (17=3456)r1247c7 - 3r7c8 =UR= 9r9c3 => r9c3=3

If someone wanted to write that like this, I wouldn't have a problem with it:

(9=4)r79c3 - (4=17)r2c239 - (17=3)r1247c7 - 3r7c8 =UR= 9r9c3 => r9c3=3

This is what I first "saw" in the PM, and not that it's relevant here, but I liked the idea (Don's idea ?) that a posted solution should convey what the solver was thinking at the time.

Parenthetical remark: That's what I saw, but since it was a chain like "9r9c3 = ... = 9r9c3", and since I knew that chains like that, can usually be modified to produce something allows either eliminating candidates other than 9 in r9c3, or eliminating the other 9(s) in a house containing r9c3, and since doing that almost always reduces the number of '=' symbols by one, I decided to try that. That lead me to changing my thinking about the logic at start of the chain, from "(9=45)r79c3 - 4r2c3 = 1r2c3" (in the short term), to "45r27c9 = 1r2c3". However, since in the line of thought, the 1 in r2c3, coupled with the 137 in r2c27 ... it meant I would need to write "45r27c9 = 137r2c237". I thought that was interesting, and I didn't recall seeing anything like it before, and so I thought I'ld share it -- share it, with the idea in mind, that it might open up new lines of thought for manual solvers.

Now back to the fist thing ...

All, I'm sorry, it was foolish of me to expect anyone to express an opinion on what's best for the forum when they are mainly concerned with their own points of detail or in in-fighting.

Nevertheless I hope I have given everyone food for thought as they try to achieve the changes or status they seek.

I'll express some opinions and (maybe) some ideas. But as I'm sure you're aware ... most prople people will tuck them away (at best), and continue doing what they've always done. Others may seek the enjoyment of an argument on some of the details. I'm an optimist though, and I'll hope for the best. I enjoy a polite argument too. On the flip side, I'm prepared to bend with the wind, and hope that I can always make sense of what other people write.

--

I've been working on this last part, and David, I'm sorry it has taken so long.
Much to say, again/still ... and in the moment ... once again ... no time to say it all

[sultan vinegar]

When I merged my A*LS terms, I only included the cells from the second ALS.

Ok, that makes more sense, but what would do if there was a fourth A*LS to merge?

Ron prefers a chain that ends with a quantum naked pair.

"No thank you please", to both ... with all due respect.

I incorrectly called that a naked pair above. It doesn't fit my definition of quantum pair either because it consists of two candidates in two cells. Perhaps composite pair (cp) is a better term? In the end:

(9)r9c3 =AUR= (3)r79c8 - (3)r7c7 = (3-1)r5c7 = (hp17)r26c7 - (hp46)r2c17 = (cp45)r27c3 ==> r9c3<>45

is just shorthand for

(9)r9c3 =AUR= (3)r79c8 - (3)r7c7 = (3-1)r5c7 = (hp17)r26c7 - (hp46)r2c17 = (4-1)r2c3 = (np45)r27c3 ==> r9c3<>45

and what I (and probably Ron) spot as a single step.

[ronk]

Perhaps composite pair (cp) is a better term? In the end:

(9)r9c3 =AUR= (3)r79c8 - (3)r7c7 = (3-1)r5c7 = (hp17)r26c7 - (hp46)r2c17 = (cp45)r27c3 ==> r9c3<>45

is just shorthand for

(9)r9c3 =AUR= (3)r79c8 - (3)r7c7 = (3-1)r5c7 = (hp17)r26c7 - (hp46)r2c17 = (4-1)r2c3 = (np45)r27c3 ==> r9c3<>45

and what I (and probably Ron) spot as a single step.

Not quite, more like ...

(9)r9c3 =AUR= (3)r79c8 - (3)r7c7 = (3-1)r5c7 = (hp17)r26c7 - (hp46)r2c17 = (4)*r2c3 - (4=5)*r7c3 ==> *r9c3<>45

... where r2c3 and r7c3 obviously both see r9c3. Seems unnecessarily detailed, just because 4r2c3 is part of an AHS and 5r7c3 is part of an ALS, albeit a single bivalued cell. As to your composite pair suggestion, at this point in time I say perhaps.

[daj95376]

If I had to change something, I'ld go with this:

(9=45)r79c3 - (4=137)r2c239 - (17=3456)r1247c7 - 3r7c8 =UR= 9r9c3 => r9c3=3

If someone wanted to write that like this, I wouldn't have a problem with it:

(9=4)r79c3 - (4=17)r2c239 - (17=3)r1247c7 - 3r7c8 =UR= 9r9c3 => r9c3=3

This is what I first "saw" in the PM, and not that it's relevant here, but I liked the idea (Don's idea ?) that a posted solution should convey what the solver was thinking at the time.

Parenthetical remark: That's what I saw, but since it was a chain like "9r9c3 = ... = 9r9c3", and since I knew that chains like that, can usually be modified to produce something allows either eliminating candidates other than 9 in r9c3, or eliminating the other 9(s) in a house containing r9c3, and since doing that almost always reduces the number of '=' symbols by one, I decided to try that.

I like either discontinuous loop. FWIW, the common practice of shortening discontinuous loops doesn't seem to work well when a UR is the first or last SL. At least, not for me. It's an exception to the rule, so to speak.

Ok, that makes more sense, but what would do if there was a fourth A*LS to merge?

My response was more tongue-in-cheek than a suggestion for anyone to implement. However, there's no real limit on connecting A*LS terms. You list the first and last A*LS ... and then you list =cells= for intermediary A*LS terms. Thus, my reference to writing an XY-Chain -- which can be viewed as numerous bivalue ALS cells being linked in a compact form. Example:

Code: Select all

 (5=6)r3c1 - (6=3)r3c5 - (3=7)r3c2 - (7=3)r2c2 - (3=1)r2c9 - (1=4)r2c3 - (4=5)r7c3  =>  -5 r9c1,r13c3

 -or-

 (5=6)r3c1 r3c52,r2c293 - (4=5)r7c3  =>  -5 r9c1,r13c3


Here, the equal signs (=) are superflous since each ALS term occupies a single cell.

_

[denis_berthier]

what David would call a "memory chain", and Denis would say "no, it doesn't use memory, it uses left to right AND splitting".

I consider it important to clarify any misunderstanding here. I wouldn't say "no, ..."
After I introduced what I called nrct-chains at that time (before I simplified my vocabulary, as the nrc appearing everywhere was somewhat redundant and in any case not appropriate for other games than Sudoku), various intuitive descriptions were suggested: memory, accumulating ... (by David, DonM, and maybe others - apologies if I'm forgetting anyone or any other such description).

I have nothing against such descriptions:
- at local "time" t within the chain, one uses his memory from previous times to justify the left to right local positive inference;
- one can also consider he is carrying a bag of knowledge for the previous steps and the bag gets heavier and heavier as he goes forward; I like this image because complexity grows fast with length and you soon feel it; as you know, the feeling of heaviness is not linear wrt real weight.

However, there are lots of different chains "with memory":
- whips, braids, g-whips, g-braids
- all the generalized whips and braids, i.e. those including more complex sub-patterns than g-candidates
- the 2D projections of all the above
- all the versions of the above without z-candidates

One of the rules of the nrc-notation is that any pattern must be prefixed by its name and length, so that the reader knows in advance what is being spoken of. "memory" or "accumulating" is not enough for such descriptions.
A side advantage (still for the reader) of this rule is, whenever you feel the need of introducing a new pattern, either as a standalone one or as a sub-pattern in a larger one, you have to define it first before using it. It may be an additional burden for the writer but, considering the pages of subsequent explanations required otherwise, I think it's a good methodological constraint (BTW, independent of any specific notation, unless it is too cumbersome).
It has also the advantage of putting into light what's really new in a proposed solution.

[blue]

I'll post this: it's the first part of something that I've been writing, but that I won't be able to get back to for several hours

<snip>

Now back to the fist thing ...

All, I'm sorry, it was foolish of me to expect anyone to express an opinion on what's best for the forum when they are mainly concerned with their own points of detail or in in-fighting.

Nevertheless I hope I have given everyone food for thought as they try to achieve the changes or status they seek.

I'll express some opinions and (maybe) some ideas. But as I'm sure you're aware ... most prople people will tuck them away (at best), and continue doing what they've always done. Others may seek the enjoyment of an argument on some of the details. I'm an optimist though, and I'll hope for the best. I enjoy a polite argument too. On the flip side, I'm prepared to bend with the wind, and hope that I can always make sense of what other people write.

--

I've been working on this last part, and David, I'm sorry it has taken so long.
Much to say, again/still ... and in the moment ... once again ... no time to say it all

Call me a liar if you like, but after thinking about what I originally wanted to say, and how to say it best ... I'm afraid I'm going to take a pass.
Some of it would have involved expressing pipe dream hopes & wishes based on a view of things through rose colored glasses.
The rest would have been pointing out positive and negative aspects of various competing alternatives, and still not drawing any conclusions, or making any specific reccomendations.