Sudoku Players' Forum (clone)

Hello to all sudoku enthusiasts who frequent this forum on which I have just registered.
Being French and not fluent in English, I use an electronic translator to do my translations, so I apologize for my poor English, but I hope I will be understandable.
On this forum, I will explain the technique I have been using since 2011 and that I have perfected down to the last detail. I call it "technique des pistes" in French (technical of tracks in English) and in summary TDP.
To be followed in other articles....
Robert

Hi Robert,

Mauriès Robert wrote:On this forum, I will explain the technique I have been using since 2011 and that I have perfected down to the last detail. I call it "technique des pistes" in French (technical of tracks in English) and in summary TDP.
To be followed in other articles....

I guess this is it? My first impression is that it seems to be quite similar to what I already use via coloring. The idea is explained here, though it's somewhat difficult to follow because David explained it in terms of text-based grid markings that no one but himself would ever use. Nevertheless, in practice it's very easy and powerful. I would be very surprised (and impressed) if your technique provided any additional powers! Seems unlikely (I'd rather bet the opposite), but I guess we'll find out if you do follow up with those other articles.

Btw, some time ago someone posted about a similar technique that also follows two tracks simultaneously. Funny enough, his post quickly disappeared after I mentioned the similarity with David's GEM. So... I don't want to sound discouraging, but that idea is far from new. I don't claim to know who invented it first, though. (It's also possible that I've misunderstood something and your idea is about something else.)

Either way, you're welcome to the forum! Like I told the other guy (who disappeared), you're very welcome to participate in the daily Puzzles as well. We have another skilled Frenchman (Cenoman) there too.

Hello SpAce,
Thank you for your feedback. Indeed the TDP (https://www.top-sudoku.com/sudoku/fr/techniques-des-pistes.php) is similar to the coloring technique you use on Hodoku and probably to that of David Bird (GEM) which I have never studied in depth. It certainly differs only in the way it is formulated. So I don't claim to have invented this method of resolution, but I have been using it since 2011 as a self-taught sudoku without having tried to find out if it already existed elsewhere. In France, there is another technique called "Virtual Coloring" very similar developed by Bernard Borrelly.
Nevertheless, I would like to give my vision of this technique and hope to bring some new features. So I will do the articles announced, because the article you mention that is mine is only a popularization.
As for the article you mention (here), it is only from 2019 and totally ignores my work published since 2013.
Sincerely
Robert

TDP Part 1
See TDP Theory (part 1) which replaces the original text of this post.
Robert

Did i understand it right, that your approach is a kind of "grouped Medusa coloring" with it's own elimination rules ?
There you start with an either/or condition of single digits (e.g. bivalue/bilocation) and follow both ways. Then you can eliminate all candidates, which would be eliminated from both paths (or if a path leads to a contradiction, you can eliminate the starting digit).
You seem to start with an either this group of digits or that.

It reads as if its doing multi digit template conditional testing via cycling multiple tiles {46656 templates} together that either are true as a n set, or contain a contradiction thus false. Manually instead a program that can compare all p.o.m tiles for each cycling the (1..9) digit combination set together to form eliminations.

1 digit - template delete/ most fish.
2 sets: ie multi coloring,
2-3sets for 3d medusa,
4 sets for sk-loop.

partial template sets for combinations of digits via comparison for completed N-sets {ie subsets}.

if i am correct, this has been around since pre: 2005/2006 long before the name i listed above came out usually these ideas fall under POM/template delete.
{ which some hints to bifurcation{nisho} / traveling pairs and braids theorem which is based on studying pom. which is in the 2006-2007 time frame}

not to knock anything you are presenting,
but if it it something new to us i am sure
champagne whom is from the french Sudoku forms can preamble on, or mayhaps help with translating for you.

Hello Eleven, I'm glad you read my article and thank you for your question.
The TDP I present uses in a certain way, such as Medusa 3D, Forcing Chains, etc... the "true/false" principle. So from this point of view there is no novelty and it is normal that we achieve the same results.
What I originally wanted to do in 2011 (without knowing Medusa at the time) was to build a global theory where, in everything I read, things were presented case by case without identifying a general principle. This gave rise to multiple techniques. I happen to have published my work only in France and in French, and as a result they are totally ignored by the Anglo-Saxon world. Medusa 3D has only been on display since 2015, I believe (http://www.sudokuwiki.org/3D_Medusa). Without seeking parternity, I still want to expose the TDP as a global theory of sudoku with elements that I have never read elsewhere.
The foundation of TDP is based on a network of two suites of candidates, one of which is necessarily true and the other false, and this generates the well-known properties found in Medusa3D on the elimination or validation of candidates. The notion of anti-track seems to me not to have been used elsewhere, and it is on the basis of this notion that I build the whole TDP. I hope that with my next articles you will better understand my approach, knowing that these articles will not cover all the aspects that can be developed, the forum being not a pedagogical book.
You can also consult my publication in French through this link.
With kind regards
Robert

Hello StrmCkr, thank you for your intervention on the subject of TDP.
I know the POM method (http://www.sudokuwiki.org/Pattern_Overlay) invented by Myth Jellies. It seems to me that it is different from TDP, POM not being from my point of view a global theory.
I suggest you read my answer to Eleven above, as well as my next articles.
With kind regards.
Robert

Hi Robert,

Thanks for your reply and the initial presentation. It's a bit heavy on theory for my taste, but hopefully we'll see more practical examples later.

Mauriès Robert wrote:Medusa 3D has only been on display since 2015, I believe (http://www.sudokuwiki.org/3D_Medusa).

That's completely false. As far as I know, 3D Medusa is a very old technique, preceding the likes of GEM (which fully contains 3D Medusa, but is much more powerful). According to David, he's used GEM since 2006. Added: I just found the first reference to Medusa on this forum. It's from 2005, and I believe by the original inventor (Bob Hanson). Here's the link to his site.

StrmCkr knows the history of the various techniques better. In any case, it makes no sense to compare your technique with 3D Medusa, as the latter is obviously much much weaker (at least the very basic version presented on SudokuWiki -- but I don't think any of the more powerful extensions are called 3D Medusa these days). The only relevant comparison is between your technique and GEM, and I would be very interested in seeing an actual comparison of those two.

Here's a recent example of an SE 9.0 puzzle that I solved quite effortlessly with GEM using Hodoku coloring tools. My rough solving steps are found later in the thread. How would you approach that puzzle with TDP?

Hi Robert,

since it was a bit hard to understand for me, what you are doing, i wanted to clarify, what the similarities and differences to known techniques in the english world are.

Medusa coloring goes back to at least 2008, with variations, see e.g. this link.

But i had a quick look at your site and saw a small 2-digit path sample, which cannot be done with that (however i would have done it another way). So you must have developed a bigger theory around it.

I don't care much about originator claims, and i don't doubt, that you found that on your own (many of us have found the basic and some advanced techniques on our own, before reading about them).

eleven wrote:Medusa coloring goes back to at least 2008, with variations, see e.g. this link.

I just proved that it goes back to at least 2005. However, your link proves that GEM has also been known in 2008:

It is akin to a watered down GEM and has surely been done by others.

It ("watered down GEM" also implies that GEM was probably considered the most powerful version of the advanced coloring techniques at the time. I haven't seen anything more powerful since then either, except maybe multi-digit multi-coloring (like champagne's "full tagging" but that's not humanly applicable. Until proven otherwise, I consider GEM the gold standard for manual coloring.

SpAce wrote: I haven't seen anything more powerful since then either, except maybe multi-digit multi-coloring (like champagne's "full tagging" but that's not humanly applicable.

Hi SpAce,
your remark push me back years ago.
Undoubtly, the "full tagging" belongs to the coloring family. It worked however in a different way. The first step was to find all kinds of subchains made of bi-values (and equivalences when specific properties as "symmetry of given" were applied), then to apply a classic chaining step to the sub chains.
As you noticed, this is not the way skill players like to work, so, I turned to other ways to attack puzzles. SKFR has been written using the full tagging code as basis. Next versions are much closer to Sudoku Explainer view. My feeling here is that we are closer to the attack done in Sudoku explainer, but with a start using a set of candidates instead of a unique candidate. In Sudoku Explainer, the expansion is hidden and only chains of interest are produced.

TDP Part 2
See TDP Theory (part 2) which replaces the original text of this post.
Robert

Your image did not work. Here is an example, how to post one:

Code: Select all

[img]http://i56.tinypic.com/xlwn7b.jpg[/img]

This is the grid.

Code: Select all

+-------------------------+-------------------------+-------------------------+
| 1      A4'78*   268     |B2'78*   9      A268'4   | 27      5       3       |
| 9       78*     5       |B2'78*   1       3       | 4       6       27      |
| 2346    2347    236     | 257    B6'2    A4'25-6  | 9       1       8       |
+-------------------------+-------------------------+-------------------------+
| 7       1       2389    | 6       238     289     | 358     4       59      |
| 23      389     4       | 12389   5       1289    | 6       3789    79      |
| 5       6       389     | 389     4       7       | 38      2       1       |
+-------------------------+-------------------------+-------------------------+
| 8       349     1       | 2359    236     2569    | 27      379     245679  |
| 36      5       7       | 4       2368    2689    | 1       389     269     |
| 346     2       369     | 13589   7       15689   | 358     389     4569    |
+-------------------------+-------------------------+-------------------------+
UR 78r12c24: 4r1c2 = 2r12c4

POM not being from my point of view a global theory.

- interesting that you don't think/consider its a global theory when its a n-set satisfaction problem and pom is able to cycle all 46656 templates for each digit in full combinations and partitions of uniform sections to solve n-set satisfaction. template delete/pom is the bases for and confirmation for pretty much every technique developed to-date. the biggest problem for pom is how long it take it to find/confirm eliminations in very large sets of information.
tabling

I don't care much about originator claims, and i don't doubt, that you found that on your own (many of us have found the basic and some advanced techniques on our own, before reading about them).

agreed many of us have multiple overlapping discoveries. I will contest misinformation.

do not use this source as a date reference: to http://www.sudokuwiki.org/3D_Medusa

Andrew stewards source material is often not actuate and only updated to the dates he publicized or adapted it to his solver.

the real sudokuwiki was lost and Andrew conveniently acquired the rights to the name to draw more people to his page.
his original website can still be referenced and entered via googling scanraid

we have a recovered copy of some of the original content on this sudopedia

I happen to have published my work only in France and in French, and as a result they are totally ignored by the Anglo-Saxon world.

that's also a misconceptions, many of the contributes on this site spend hours combing multiple multi-langue sites for information to adapt/translate directly and we reference all source material back to the original webpages.
many of those that are here are not Anglo-Saxon.

so where exactly is your predated work/research not a recently publish paper.

which by the way has a publication date of: sunday, March 18, 2018, 4:18:24 PM

the biggest headache we will acknowledge is the loss of 3 of the major original source material. aka sudoku.org, sudoku programmers forum, our original hosting site
a lot of the stuff on this page has been restored by very dedicated people from data base backups from our own web-crawlers. because of this some topics dates of publication are only verifiable via the waybackmachine even then its an approximation based on multiple tracing sources.

closing remarks, as i said before do not get me construed i will digest what you post and critique and evaluate it for merit.

interesting als a & als b with strong links on 7, transport on digit 4 eliminates digit 2&6.

Code: Select all

+---------------------+----------------------+-------------------+
| 1       (478)  268  | (278)  9     2468    | 27   5     3      |
| 9       (78)   5    | (278)  1     3       | 4    6     27     |
| 236(4)  3(47)  236  | 25(7)  (26)  5-26(4) | 9    1     8      |
+---------------------+----------------------+-------------------+
| 7       1      2389 | 6      238   289     | 358  4     59     |
| 23      389    4    | 12389  5     1289    | 6    3789  79     |
| 5       6      389  | 389    4     7       | 38   2     1      |
+---------------------+----------------------+-------------------+
| 8       349    1    | 2359   236   2569    | 27   379   245679 |
| 36      5      7    | 4      2368  2689    | 1    389   269    |
| 346     2      369  | 13589  7     15689   | 358  389   4569   |
+---------------------+----------------------+-------------------+


or you know we could do a simple....
Finned X-Wing: 2 c14 r35 fr1c4 fr2c4 => r3c56<>2 ->> R3C5 = 6 ....

and both really avoid the unique rectangle arguments.

for giggles ALS - W - RING

Code: Select all

+-------------------+----------------------+-------------------+
| 1     (478)  26-8 | (278)   9     (2468) | 27   5     3      |
| 9     (78)   5    | (278)   1     3      | 4    6     27     |
| 2346  34(7)  236  | 5-2(7)  (26)  45-26  | 9    1     8      |
+-------------------+----------------------+-------------------+
| 7     1      2389 | 6       238   289    | 358  4     59     |
| 23    39-8   4    | 12389   5     1289   | 6    3789  79     |
| 5     6      389  | 389     4     7      | 38   2     1      |
+-------------------+----------------------+-------------------+
| 8     349    1    | 2359    236   2569   | 27   379   245679 |
| 36    5      7    | 4       2368  2689   | 1    389   269    |
| 346   2      369  | 13589   7     15689  | 358  389   4569   |
+-------------------+----------------------+-------------------+