Page 1 of 2
SudoCue Release 3.1.0
Posted: Wed Aug 22, 2007 7:39 pm
by Ruud
All reported errors have been fixed.
Several new features have been introduced.
No massive rewrites of the code were done this time, so I'll skip the beta stage. I hope I will not regret this desicion...
Ruud
Posted: Thu Aug 23, 2007 10:09 am
by rep'nA
Ruud,
Excellent job on the new version. So far no bug issues to report. Three requests, though. First, could you add keyboard shortcuts for switching between the RC, RN and CN-spaces? When trying to perform Denis' 3D-superchains, it is useful to switch back and forth without taking my eyes off of the grid. Second, it would be nice to be able to copy the marks from the 3 different spaces. Third, I really like the look of the new candidate filter buttons (oh and thank you so much for the pairs button. Now I may never have to use simple sudoku again), but could they be slightly larger? I keep hitting the wrong one because they are smaller than my mouse pointer icon.
Again, thanks for all of the hard work put into this. You've taken the best "product" on the market and made it a lot more versatile and a lot more fun to use.
Posted: Fri Aug 24, 2007 8:30 am
by rep'nA
Okay, strange bug here. If I open the program. Open a new puzzle of sufficiently high difficulty (say, by opening a New Random Unfair), open the solver log and then proceed to select
>Solve all Easy steps
>Solve upto Medium steps
>Solve upto Tough steps
>Solve upto Hard steps
I get the following error message:

Posted: Fri Aug 24, 2007 8:45 am
by rep'nA
Another obscure bug:
Open a puzzle.
Switch to RN-space and remove a candidate.
Copy Marks
Paste
The candidate will not return in normal view, but does return in RN-space.
The same problem occurs in CN-space.
Posted: Fri Aug 24, 2007 5:45 pm
by Ruud
Thanks. I will look into this.
Ruud
Posted: Fri Aug 24, 2007 9:04 pm
by Para
Hi
I noticed these RC-, RN- and CN-spaces mentioned here but what exactly are these RC-, RN- and CN-spaces and what are these used for?
greetings
Para
Posted: Fri Aug 24, 2007 9:32 pm
by Ruud
RC space is what you are normally looking at. Rows and columns, with the pencilmarks showing the remaining numbers.
You can transpose the grid in 2 ways:
RN (rows-numbers)
The rows remain in place, but the horizontal axis now shows the numbers and the pencilmarks represent the available columns for each Row+Number combination. In this view, you lose sight of the box constraints, but instead you can search for hidden singles as naked singles, hidden subsets as naked subsets and the hidden XY-chains recently described by Denis Berthier as if they were "naked" XY-Chains.
CN (columns-numbers)
In this transposition, the vertical axis shows the numbers and the pencilmarks represent the available rows for each CN combination.
It is a nice addition to the toolset. Nobody asked for a "Copy RN" or "Copy CN" function yet, but that would be useful for forum members. Ronk and Rep'nA have already been experimenting with RN/CN views, so they might be able to tell you more.
Ruud
Posted: Fri Aug 24, 2007 9:48 pm
by Glyn
Hi Para
Ruud beat me to this answer. The spaces had been used previously but have really come to attention since Denis Berthier started posting on them in the Eureka forum and the Players' Forum.
If you have ever looked at Bob Hansons graphics on Sudoku Assistant you may have seen the problem of solving a grid expressed as filling a 9x9x9 cube subject to certain constraints. The RN and CN planes are the orthogonal views on this cube to the familiar RC plane view. I recommend having a look at his site as his on-line program shows a lot of the process of building Medusa chains as well.
All the best
Glyn
Posted: Sat Aug 25, 2007 10:10 am
by rep'nA
Ruud wrote:
Nobody asked for a "Copy RN" or "Copy CN" function yet, but that would be useful for forum members.
If I didn't ask before, consider it asked now.
Ruud wrote:
Ronk and Rep'nA have already been experimenting with RN/CN views, so they might be able to tell you more.
Try
here and
here for some more info and examples.
Posted: Sun Aug 26, 2007 6:58 pm
by Ruud
Service Pack 1 for release 3.1.0 is available. The version is 3.1.0.1.
Same installer package:
http://www.sudokuvault.com/SudoCueV310.msi
Fixes:
- Sue de Coq is back in action. Sneaky bug exterminated.
- Grid returns to normal view after pasting a puzzle.
- No more error after repeated Solve upto (level).
- Shortcut keys Ctrl+F5/F6/F7 for view switching (shown in the menu)
- Copy RN view and CN view to clipboard.
- Faster random puzzle generation for lower difficulty levels. Program no longer tries to solve the remaining puzzle when a higher level solving technique is encountered.
- Less memory usage for puzzle collections. If you still get an out-of-memory error, install more memory.
Check the new
Extra / Tags for string Copy menu option. When set to
phpBB,
[code] tags are added to the copied text. When set to
Discuss, the program adds fixed size formatting tags for the
Sudoku UK forum.
No larger filter buttons yet. Maybe next time.
Ruud
Posted: Sun Aug 26, 2007 9:27 pm
by rep'nA
Thanks Ruud. I really like how the RN and CN spaces are auto-formatted for posting on the web. That will save me loads of time. Is there a reason the RN-space uses more space than the CN-space when the marks are copied? I prefer the RN-space spacing, in case you're interested.
Also, in the Extra>Options>Solver menu, the techniques are now all called Sudoku.Tricks.<Technique>.
It's fantastic how fast you are able to fix bugs and add new functionality. Kudos!
Posted: Sun Aug 26, 2007 10:24 pm
by Ruud
rep'nA wrote:in the Extra>Options>Solver menu, the techniques are now all called Sudoku.Tricks.<Technique>.
Murphy strikes again, but it's already fixed. I uploaded a new file with the same version number. No-one else needs to know this bug ever existed...
Is there a reason the RN-space uses more space than the CN-space when the marks are copied?
The spacing is dynamic. The program looks for the longest possibility set and adds a space before and after that set. This ensures that there are at least 2 spaces between 2 adjacent number groups. If the RN-space used more whitespace then it must have a bigger set. I also added this extra space in the normal pencilmark grid, to enable room for markings.
About the next release:
My first priority is to prepare the program for translation into different languages. Once I have all the texts in an external file, I'll ask for volunteers to translate them.
The next big thing will be to look at some new solving techniques to implement. The xyt- hxy, hxyt chains look promising. BUG-Lite is also on my wishlist. I've done a lot of research in that area.
Posted: Mon Aug 27, 2007 1:56 am
by nj3h
Hi Ruud,
Great new features.
I am having one issue. When I select the Clover-X variant and then select New Random, Moderate or Tough (those are the only difficulties I checked), the program continues to show 0 attempts and then after 30 seconds ctrl-alt-del indicates that the program is not responding.
When you have a minute, can you check that you are getting this on your end?
Regards,
George
Posted: Mon Aug 27, 2007 8:44 am
by rep'nA
One minor typo and one (hopefully) minor request:
Typo: The fishing hint repeats the word digit twice in a row, e.g., " Try fishing for digit digit 5."
Request: At the bottom where it lists Current Variant, Current Mode and Current Helper Tool, could you add Current View, which would tell you which of the RC, RN, CN -spaces you're viewing?
Thanks Ruud.
Posted: Mon Aug 27, 2007 10:49 am
by Glyn
nj3h wrote
I am having one issue. When I select the Clover-X variant and then select New Random, Moderate or Tough (those are the only difficulties I checked), the program continues to show 0 attempts and then after 30 seconds ctrl-alt-del indicates that the program is not responding.
I have found the same problem which may related to the difficulty of finding valid puzzles rather than the program itself. The greater the number of constraints the more difficult it is to populate the grid and then to remove givens to create a puzzle. It appears that the set of available puzzles between Easy and Unfair is small. Try generating puzzles with those levels and you will see what I mean, it's much easier (ie you sometimes get one). You may also find that some of the moves required merit a high tariff in Sudocues' rating system and this may be pushing them out of your requested zone.
Ruud. Is stalling on Attempt 0 an indication that current solver methods are not cracking the puzzle. We have already seen some Clover-X that are doable, but require techniques that any one program does not have.
All the best
Glyn