SudokuSolver Forum

Congratulations Richard, ando more
I just enjoyed doing the Regular Clueless Explosion 1A, like the good old times.
I appreciate many of the new features, and keep on discovering new things
There are some details on the way the navigation is done wich is not so natural for me, but I will have to use it more before proposing any change.
You are doing a very good programming

_________________
Nothing can both be and not be

Glyn, slight correction. 1B stalls with standard options.

If you use the following options file: then 1B solves - it needs a rather complex ALS move to unlock it.

1D is, I'm sorry to say unsolvable in SudokuSolver

Rgds
Richard

Hi Richard I'll a look at that. Perhaps it's the ALS that I used when my rod was stolen.

One thing that might be useful is the retention of colouring when changing grids. I have been informed that an Explosion has been found that does not surrender to JSudoku without Brute Force, sounds another Ronin.

_________________
I have 81 brain cells left, I think.

I have posted it here as CE-57226, so that everyone can have a shot at it.

_________________
Quis custodiet ipsos custodes?

Normal: [D  Y-m-d,  G:i]     PM->email: [D, d M Y H:i:s]

Richard, just wanted to say that whatever improvements you've made to SudokuSolver also seems to make it run a lot more smoothly. Good work!

Just to let you all know - I've made a few quick updates to the program.

http://www.rcbroughton.co.uk/sudoku/SudokuSolver.msi

Main items was to add in Centre Dot elimination technique and fix a couple of bugs in the Save/Load routines.

I've also added in a small Gattai selection function when pasting a new puzzle definition to avoid any confusion when SudokuSolver can't determine what type of puzzle it should be. As an example - if you paste in a string containing 9 x 81 grid definitions, there are three Gattai types that use this number of grids. If the program is currently using a compatible puzzle defininition, then that puzzle type will be assumed, otherwise a small window will pop up to allow you to select from a short list of compatible Gattai puzzles.

Rgds
Richard

You should put the line of the central grid as the 5th line instead of the 3rd.

This code works:
Code:
090400870100000009048070200020016000601900084007000600000600000003090100000000000
000090020000002308080130400008010006030009700000007000010500807000040002000070010
040070000100090000209003010000800000004900030900030400003029040705300000020050000
000000000001040200000003000006000300310004905000930010004090150100000009085001060
600000008090100030000000000000706010000000000010304000000000000030001040400000003


WHEN PASTED THIS CODE IT CHANGED THE POSITION OF THE NUMBERS. THE FULL PUZZLE CHANGED.

Richard, EDIT on 31 may 08 based on new informationI still need more time using the program to really be more complete with my comments, but for now I have done enough regular Clueless/Explosions/Samurays as to comment on two specific problems I have found:
1, The cursor keys control both the movement on the grid and the advance-retrocess on the filter pad. That impeedes to do fast alternate moves from both of them. Keep cursor keys for the grid. CluelessHelper did the navigation on the filter pad via < and > key but it required an uppercase and was not in the same position in all keyboards. Since anyway the navigation is only "next digit", "previous digit" I propose:
a coulpe of leters.
adjacent and in the left part of the keyboard of preference.
which are not changing from english to british to french to spanish keyboard.
like "s""d" or better "x""c".
EDIT Using the CTRL+cursor keys it is possible to change the digit selected without loosing the position in the grid! Allow for fast actions! Left hand on CTRL key, while right one takes care of cursor-keys
2, The above would impact the navigating that now goes arround "Hightlght Type", "Cell Types", filter pad 1-9, "Clear" and "Setup". First I propose to take out of the navigation the "Clear" and "Setup", they are not frequent enough to require fast navigation among them and nit linked to filters. Then, for "Hightlght Type", "Cell Types", I would leave the navigation with the same keys as with the filter pad, but with a different loop for each one, meaning that when you are in "Hightlght Type" the couple of keys take you from "All Digits", to "Any Digits" to "Only Digits" to "All Digits" and so on, same for "Cell Types" and of course for filter pad 1-9 digits that I mentionned above in point 1, close loop from 1 to 9.
EDIT Change is being considered for future releases, stay tuned!
Well now I continue to use it and once I get any other remarks I wil coma back to you

_________________
Nothing can both be and not be

That sounds perfect! Did I miss something in the changelog, or is this not in the new 3.1.0 version?

Two requests (in addition to my previous one for candidate marking):

1. Print each grid of a multi-grid puzzle, one grid per sheet of paper, full-size (8"x11" in the U.S., A4 or whatever is standard elsewhere). Currently, SS prints grids too small to be useful (to me, anyway).

2. Upon exporting candidates, enable selection (via CTRL-A or similar) of entire block (for copying to Windows clipboard)

And a bug:

Export of solution seems to not be working (at least for Samurai). Instead, it exports the unsolved puzzle code for the upper-left grid, 5 identical rows.