This post concludes my series of posts on basic solving, the actions taken to identify all candidates. As the October posts have shown, sysudoku basic solving also identifies and marks strong links and locked sets in boxes and lines. It is designed to find these elements exhaustively, with minimum effort.
The following posts describe advanced solving techniques, which combine these elements to make candidate removals and find new clues. Advanced techniques exploit slinks and winks directly, expanding their reach by stringing them into chains and other formations. Much later, the blog explores extreme techniques, so called because they exploit indirect and collective consequences of Sudoku rules.
I have several thoughts to share with you about Sysudoku basic solving now, before turning my attention to advanced topics.
For one thing, it applies to every Sudoku puzzle, from the Monday morning one star, to the celebrated monster puzzle designed to be unsolvable. The goal of finding all candidates is not reached for easier ones, because candidates all melt into clues. But the process serves that quick result as well.
As you read through this blog, we will share basic solving of puzzles that illustrate advanced and extreme techniques of all kinds. You will have a complete record of many of my solving efforts, thanks to the very compact traces and colorful diagrams. I hope that you will be intolerant of any mistakes you find, and will send in a comment. I’ll fix it, no charge.
These early posts have illustrated the process of marking which follows up on any new clues or candidate removals. Throughout the blog, posts provide traces of basic and advanced marking, along with the grid diagrams. The traces follow a marking procedure that exploits new discoveries immediately, while retaining every bit of new information for possible use later. We are hunter gatherers all. All effects of a new cause are listed together, with effects on the cause number listed first. Then each effect is processed in turn as a cause. When prior recorded but unexploited information is needed, the trace reader and writer know where to go for it. When all consequences of that new event are exhausted, the solver knows it.
A systematic and reproducible solving process is a means of judging the level of difficulty of puzzles. Many experts rate puzzles on how far their computer codes go into the stockpile of algorithms to solve them. That is systematic and reproducible. But for my goal of human engineered Sudoku solving, the system has to be a mental one. Basic solving turned out to be naturally systematic enough to be reproducible, given the tracing rules for marking. My attempt to extend this into advanced and extreme solving, termed the Order of Battle, remains a work in progress.
Generally speaking, on my experience, Sysudoku basic solving covers five star puzzles in newspapers and magazines. Three stars generally may have one or two “3f” lines to fight you with, after box marking. But that doesn’t mean they aren’t fun.
Disconcertingly enough, I know some “natural” solvers, who don’t bother pencil marking on basic level puzzles. Yes, they just write down the clues. Maybe you are, or will be, among them. But if not, it doesn’t matter. If you are coming here from the current posts, you know I’m solving celebrated monsters and reviewing the most difficult collections published, and I can’t do it without pencil marks. But with the bypass stage of box marking introduced in October 2014, I’m getting closer.
I have another Dave Green puzzle with which to end the basic solving posts. I saw it just in time for a revision of this one. It’s this “teapot” *** published 2/05/14 by King Features in the Akron Beacon Journal. Your mission is to read through the trace below and find the times that a special box formation from the beginner’s page is used to leapfrog the numerical order of box marking. You know to get out a grid and mark it up as you read, right? The puzzle illustrates how far such box marking departures can go.
Then after following the link above, try to see how far you can go with the bypass. The box marking trace is your checkpoint.
By the way, at the original posting of 11/01/11, there was no beginner‘s page, and no bypass. The blog evolves.
I am following the tracing, and try as I might, I cannot see how you get the NE5m after 2: Nm, NE2 – I see NE5t for NEr1 – what am I missing?
Thanks for the correction, Rich. It is NE5t, along with N1m, as the effects of NE2.
It seems that the notation NW4 is missing after N4m. I see now that after that NEr1c2 must be 4.
Not quite, Rich. N4 is an effect of the naked pair, not the 4m. But thanks for commenting.