Page 1 of 1
DominoCue release 1.0
Posted: Sat Dec 01, 2007 7:24 pm
by Ruud
Check the
DominoCue support page for information about this new program.
Feedback appreciated as always.
Ruud
Posted: Sun Dec 02, 2007 2:58 am
by nj3h
Hi Ruud,
Ok, I'll go first.
I have never done these kind of puzzles before, so my comments may need to be discounted because of that fact.
1. I would like to see a "Are you sure?" dialog box when hitting F6.
2. I would like to see an explanation of the hint. I see that it puts a line or lines in the grid. However, In some puzzles I tried I have absolutely no reason why. What is the significance of the green boxes when a hint is asked for? Obviously they relate to the hint, but how?
3. I would like to see an option that does not automatically place other lines when the player places a domino. For example, if I place a 23, the program currently place a solid line between any pther 23s in the grid.
4. I was playing on the easiest level and when I asked for a hint, none were offered. Does this mean that there is not a unique solution or that some particular technique hasn't been implemented.
5. As far as techniques go, it would be nice to know exactly what techniques are implemented. I guess it would be nice if each technique had a name that, at least for this website, has a standard definition so we all know what is being described. At some point in the future, the ability to choose what techniques are included in generated puzzles would be nice. The user should be able to generate easy, medium, hard, etc. puzzles. Using the message thread in the forum as a starting point, a technique guide would be nice to have developed.
6. Obviously someone will want a print option, but a screen capture and running through Paint Shop Pro will work just fine. I don't think I would want to do this puzzle on paper. But that's just me.
7. I would say that as always, Ruud provides very professional looking and operating software. We all owe Ruud a round of applause for continuing to keep everyone's interest running high.
Thanks Ruud!
Regards,
George
Posted: Sun Dec 02, 2007 9:00 pm
by nj3h
All,
Maybe a simple puzzle, Level 0-3, but there are no hints available from this point. I am sure that there must be a solution, but perhaps the techniques needed haven't been incorporated into the hint system as of yet. Just curious though, shouldn't the program only generate puzzles that can be solved with the techniques covered in the hint system? A listing of which techniques that the program incorporates to date would be helpful at some point in the future.
Here is the puzzle:
0220211111322333k3uk
Also when the initial puzzle was loaded, the status line indicated Initial Singles 1, but when the progress string is pasted in the program now indicates Initial Singles 10. That confuses me.
I must say the look and feel of the program is excellent. Ruud always goes the extra yard (3.37 inches beyond a meter) in producing slick software.
Regards,
George
Posted: Mon Dec 03, 2007 3:09 am
by nj3h
Two more thoughts.
When a hint gives a number of red lines, when one clicks on one of the lines, the other red lines disappear. Is there a way to keep all hint suggestions lit up until the lines are clicked on?
Why are there sometimes two shades of green when a hint is shown? I notice a hint that highlights just two cells sometimes has two shades and sometimes the cells are the same shade.
Posted: Mon Dec 03, 2007 12:40 pm
by Pete
i concur with nj3h on
hint explination
solving guide how to
the generate easy medium hard and (interesting to Ruud) would be nice
i really feel humbled if the difficulty 3 is interesting
to nj3h "no more hint" is probably the same as the others... in that other thread there are like 3 known named techniques. I can see how to make the bfbi solver or T&E solver but im far from understanding these domino puzzles.
my suggestion:
an undo ^Z or whatever . a wrong border anywhere leaves relics all over the board.. never sure which of the lines were correct elsewhere in the puzzle after i take the last line out.
Posted: Mon Dec 03, 2007 1:11 pm
by rep'nA
nj3h wrote:All,
Maybe a simple puzzle, Level 0-3, but there are no hints available from this point. I am sure that there must be a solution, but perhaps the techniques needed haven't been incorporated into the hint system as of yet. Just curious though, shouldn't the program only generate puzzles that can be solved with the techniques covered in the hint system? A listing of which techniques that the program incorporates to date would be helpful at some point in the future.
Here is the puzzle:
0220211111322333k3uk
Also when the initial puzzle was loaded, the status line indicated Initial Singles 1, but when the progress string is pasted in the program now indicates Initial Singles 10. That confuses me.
When I load the puzzle, it is given a 6.8 rating with 5 initial singles. In any case, here is a possible solution to your puzzle:
1. Highlighting [23], we see that one of the 2's in r3c34 must be coupled with a 3, thus r3c34<>[22]. This reveals a few singles and takes us to:
Code: Select all
.-------------------.
| 0 | 2 2 | 0 2 |
| +---+---+---+---|
| 1 | 1 1 1 1 |
|---+ + + + |
| 3 2 | 2 3 3 |
| +---+ +---+---|
| 3 0 3 | 0 0 |
'-------------------'
2. Observe r23c45 which is bounded from beneath and from the right
It is easy to see that if r23c4=[13], then so does r23c5. Hence r23c4<>[13].
The puzzle now resolves with singles.
nj3h wrote:
I must say the look and feel of the program is excellent. Ruud always goes the extra yard (3.37 inches beyond a meter) in producing slick software.
Amen.
Posted: Mon Dec 03, 2007 1:59 pm
by rep'nA
Ruud,
Here is an idea. I just learned this game when I downloaded it this weekend, but I've already learned that spotting locked sets can be tedious. I've taken to writing down a table which tells me which digit combinations a given number sees. For example:
Code: Select all
.-------------------------------.
| 0 0 1 1 4 1 | 2 3 |
| + + + + + +---+---|
| 2 2 5 6 6 5 | 4 4 |
| +---+ + + + +---+---|
| 0 3 3 3 0 6 1 6 |
| +---+ + + + +---+ |
| 1 1 5 0 5 0 3 5 |
| + + +---+ + + + |
| 0 4 6 | 1 | 4 0 3 4 |
| + + + + + +---+---|
| 2 | 3 6 | 2 | 5 2 | 5 5 |
| + + +---+ + +---+---|
| 2 4 6 2 6 4 | 1 3 |
'-------------------------------'
This partial grid corresponds to the table:
Code: Select all
0: [02], [012], [123], [356], [35], [0356], [124], [0234]
1: [015], [146], [45], [6], [01]
2: [02], [025], [02], [45], [24], [6]
3: [03], [35], [036], [035], [034], [46]
4: [16], [0136], [05], [35], [236], [26]
5: [1236], [16], [0136], [04], [346], [246]
6: [356], [0456], [015], [15], [456], [36], [246], [245]
The way one constructs this table is by looking at a digit on the grid and writing down the adjacent numbers (putting up a wall kills adjacency). I construct my tables digit by digit and left to right, top to bottom. Thus,
0: [02], [012],... tells me that the first place I see a 0 in the grid (r1c1), it will be adjacent to a 0 and a 2 and the second place I see a 0 in the grid (r1c2), 0 will be adjacent to a 0, 1 and 2.
Now, roughly speaking, locked sets will appear as they do regular sudoku, although some care must be taken in the eliminations. For instance, in the row for 1 we see a singleton [6]. Naively, we can then 'eliminate' the 6 from [146] (corresponding to the 1 in r1c4). There is also a locked quint for 0 on the digits {0,1,2,3,4} which would seem to allow us to kill those numbers from [356], [35] and [0356]. All will work except for killing the 0 in [0356]. Why? Because that [00] was used as part of the locked set, namely as part of [0234].
As of now, DominoCue will run out of hints to solve the above puzzle, but using the above table, you should be able to solve it.
It would be awfully handy to have this task automated, especially for the larger puzzles.
Posted: Mon Dec 03, 2007 6:58 pm
by Ruud
Thanks for all the suggestions, George, Pete & Rep'nA (I knew there was a way to revive you, Adam
)
The rating problem is noted and will be fixed quickly. Actually, the rating you see is that of the last puzzle generated by the program. It should be updated when you paste/open an existing puzzle.
Undo is in the making, and so is the option not to automatically place additional borders. For puzzles with singles only, this feature pretty much spoils the fun.
As for solving techniques, a solving guide will be added to the DominoCue page, but this is a relatively new puzzle format for me, and I haven't found any in-depth research on solving techniques. I've found a few techniques myself and Para has also pointed out a few techniques, but there is a lot of undiscovered ground here. For me, that makes this format so interesting. What are the odds to discover a new Sudoku solving technique nowadays, compared to the chance of discovering a new Domino technique? Any Mathematicians in the audience?
Adam, I will implement your N-N view into the program. It makes life much easier looking for combinations. One warning though: this view will not make a distinction between valid and invalid locked sets, the latter having the option to cover 2 base cells with a single tile.
Finally, these puzzles do contain strong and weak links, so it should be possible to introduce AIC techniques. With so many "forbidden squares", these chains could contain several of them.
Ruud
Posted: Mon Dec 03, 2007 7:39 pm
by nj3h
Ruud,
If you do have options available, then it would be nice if the program remembered them when you exit. Even now, the program always defaults to level 9 when opened.
Sudoku has "fish". I don't like fish, so maybe since DominoCue creates pens around the dominoes, perhaps the new techniques can be named for cattle and/or chickens. Just a silly thought.
Regards,
George
Square isolation bug
Posted: Sat Dec 08, 2007 5:30 pm
by unkx80
Code: Select all
.-------------------------------------------.
| 6 2 3 0 4 5 2 | 4 8 | 9 5 |
| + + + + + + +---+---+ + |
| 7 9 8 5 0 1 5 | 2 2 | 9 7 |
|---+---+---+ + + +---+---+---+---+---|
| 2 | 6 | 4 3 7 | 1 2 0 1 | 3 3 |
| + + + + + + + + +---+---|
| 4 | 8 | 9 2 6 4 3 0 2 0 6 |
|---+---+ + + +---+ + + + + |
| 2 8 | 1 3 9 0 9 3 6 0 3 |
|---+---+ + + +---+ +---+ + + |
| 4 1 5 7 6 5 9 | 6 | 0 2 1 |
| + + + + + + + +---+---+ |
| 4 9 1 8 9 1 4 | 6 | 0 8 | 8 |
|---+ + + + + + +---+---+---+ |
| 8 1 6 | 5 8 1 | 7 7 9 3 5 |
| + +---+ + + + + + + + |
| 8 3 0 5 8 9 | 2 7 4 4 5 |
| + + + + + +---+ + + + |
| 3 5 7 6 7 7 1 7 6 0 4 |
'-------------------------------------------'
For the above puzzle-in-progress DominoCue suggests square isolation for the upper-right square, but this square has already been isolated.
BTW, in this forum's login screen, the link "forgotten your password" is broken.
Posted: Sun Dec 09, 2007 12:17 am
by Para
DominoCue can't seem to move past a square isolation move. Even if it actually does do something, it keeps referring back to this move.
Posted: Sat Dec 15, 2007 1:30 pm
by Pete
well ive been playing for quite a while.
so some more thoughts
I'm Clueless(tm) about how the initial singles numbers are generated.
the solution seems to have a lot due to pariety.
The badtile into oddity as ruud calls it.
ive been pretty sucessful making unplaced tiles into bigger tiles (tetrinmos or heximos) .. these have a lot less places they can fit on the board.
the puzzle solves more like a jigsaw puzzle with this idea.
Any plans to go to 12 dominos? you can buy plastic(wood) sets in 6,9, 12 or 15 ..
if you go to larger puzzles i would suggest the digits 0123456789ABCDEFG...
these fit nicer in the squares espically when candidates are listed
Pasting grid with borders
Posted: Wed Dec 26, 2007 7:05 am
by unkx80
Hi Ruud,
While you are at it, may I also ask that: when I copy and paste ASCII grids like the following, paste the borders as well. (Currently only the digits are pasted but the borders are not shown.)
Code: Select all
.-------------------------------------------.
| 6 2 3 0 4 5 2 | 4 8 | 9 5 |
| + + + + + + +---+---+ + |
| 7 9 8 5 0 1 5 | 2 2 | 9 7 |
|---+---+---+ + + +---+---+---+---+---|
| 2 | 6 | 4 3 7 | 1 2 0 1 | 3 3 |
| + + + + + + + + +---+---|
| 4 | 8 | 9 2 6 4 3 0 2 0 6 |
|---+---+ + + +---+ + + + + |
| 2 8 | 1 3 9 0 9 3 6 0 3 |
|---+---+ + + +---+ +---+ + + |
| 4 1 5 7 6 5 9 | 6 | 0 2 1 |
| + + + + + + + +---+---+ |
| 4 9 1 8 9 1 4 | 6 | 0 8 | 8 |
|---+ + + + + + +---+---+---+ |
| 8 1 6 | 5 8 1 | 7 7 9 3 5 |
| + +---+ + + + + + + + |
| 8 3 0 5 8 9 | 2 7 4 4 5 |
| + + + + + +---+ + + + |
| 3 5 7 6 7 7 1 7 6 0 4 |
'-------------------------------------------'
Many thanks for this fine program!
Posted: Wed Dec 26, 2007 9:01 am
by Ruud
Thanks for your suggestions.
After I completed the new release of SumoCue, I will build these fixes and enhancements into DominoCue.
Ruud
Posted: Fri Jan 04, 2008 5:08 pm
by unkx80
One more thing. When a puzzle is loaded from a saved file or pasted from the clipboard, please update the difficulty rating and the number of initial singles!