dailysudoku.com

[Robert Ferre]

George, I would he interested in your notation system.
.
Keith, I don't understand unique rectangles. I see the ones that you pointed out, although the cells have other values, too, and similar values nearby. So, I don't understand the logic that leads to the exclusions that you included with each example. I have learned that X-boxes and Swordfish have direction. Is the same true for unique rectangles?

[Marty R.]

Keith, I just asked a similar question in the May 2 thread.

[keith]

Robert and Marty,

Give me a day or two, I will post an introduction to unique rectangles.

Marty,

I was thinking of you the other day. I drove from Detroit to Ithaca, NY, passing close to Rochester. (And with Al Franken on XM radio, broadcasting that day from Rochester!)

Best wishes,
Keith

[George Woods]

I label the boxes alphabetically so Box 1 is a and box 9 is i
i.e.

a b c
d e f
g h i

so if the first move is in box e and is the insetion of a 9 I write it down as e9. If I need to refer explicitly to a cell I write it - example Fa i.e. the first cell in box F

Generallysomething like c9 is adequate, and it is hopefully ovious where to put the 9 and why! Otherwise I use additional symbols e.g. underline means "this is the only number that can go in this cell" - i.e see that all other numbers 1 to 9 can be "seen" from this cell. A bar above I use to show that it is deduced by looking along the row, and two bars the column. I use an exclamation mark to show a deduction based on an implied arrangement (e.g.that number must lie in 2 or 3 cells - allowing a deduction elesewhere). No symbol generally means obvious and at the start of a puzzle is usually"only cell that can hold this number", but near the end of a puzzle is likely to be "the only number that is allowed in this cell". I try to use commas to indicate a change of idea, so e673 means 6 then 7 and 3 are entered into e in that order, but done as a "triple" e6,7 implies the 6 and 7 derived from different threads of argument.

I sometimes write f(67) or f66 impling that I have written 67 into two cells, or e66 that 6 lies in only two of the cells, and I write very small 6 in each to remind me.

Clearly you can expand this nomenclature. What I have found is that when a puzzle "explodes", I can backtrack, and usually the mistake turns out to be only one or two moves ago!

I find the alphabetic description easier to use and is more compact e.g Fe is more compact than r5c8

[Robert Ferre]

That was helpful, George. Thanks. It is based not only on numerical possibilities, it also indicates some of the logic involved. I think I can use a little more work on my logic, and my "seeing." I'm working on it.

Robert

[George Woods]

For what it is worth here is an example of my solution to today's puzzle (may4)

a1,e1,h1.i1,b7,e4||,col2(28)d5,f7|,d7,i7|,g7c7a7,a5g6,h6,h9,c2,g3|,d8g2,d62,a48,c4|b4c5,h52,c2,e89,b3,9,c3,e63,f62,c296,i8,g458,f89,i9,i9534,a69

I have used | and || instead of bars over to indicate look in rowand look in column

The col2(2 is put in because it is a key move i.e. scan col2 for 2and 8 and you will find a "hidden pair"