dailysudoku.com

[AZ Matt]

So I am working on comprhending swordfish and other, trickier fish.

Is it always the case, in the end analysis, that you are trying to find a set of cells for one number in 1) rows, or 2) columns, where the number has to fit in each line, regardless how the numbers fall, so that if you find the set in rows, for example, you can elminate the number as a candidate in extraneous cells in related columns (or vice versa)?

I ask this because it seems I have come across the term "turbo fish," and even had one of my solving techniques described as a "sashimi with a fin," but it seems that no matter how complicated it gets, that is the basic theory.

So in the end, is it true that an x-wing is the simplest form of this solving technique, a swordfish is two interdependent x-wings (the next level?), and other fish are more complicated variations on this theme (e.g., boxes become implicated, or other solving techniques become involved)?

[keith]

AZ,

You need to read the Advanced Solution Techniques forum on sudoku.com.

Follow th link I posted the other day to Mike Barker's list of links to solution techniques.

One progression:

The most basic fish is a naked single.

The next is an X-wing, in which two candidates in each of two rows line up in two columns.

The next is a swordfish, in which three candidates in three rows line up in three columns.

Jellyfish for four, squirmbag for five. You will never need a squirmbag.

Each of these patterns is a collection of overlapping rectangles; when you sketch them out, they have an even number of sides or "links".

A "finned" fish is one where the pattern is true, if not for one extra candidate cell. A "sashimi" fish is where the pattern is true, except one cell is missing the candidate.

A Turbot Fish is of a different family and is a pattern that has five sides. The best way to understand them is to first read the thread on "Strong links for Beginners", then the one on "Turbot Fish".

Good luck!

Keith

[AZ Matt]

Thanks for confirming my understanding. The problem with the explantions of fish is that they try to state the concept as a succinct rule -- they don't explain the underlying "philosophy" of the broader, general concept. Now that I know what I am doing and what I am looking for, I see the trick -- you are simply taking the candidates of one number and seeing how they can "fit" (or, more often, can't fit) into the puzzle. The names seem to attach to the number of cells (generally) and the patterns involved.