Thursday, 28th March 2024
Puzzles Solved Yesterday: 124
Forum Index
 
Page 2 of 3<123>
Paper vs. Computer
mathmaniac
Kwon-Tom Obsessive
Puzzles: 1293
Best Total: 20m 57s
Posted - 2006.06.13 00:18:59
That is a good strategy I had not thought of. Thanks for sharing
tilps
Kwon-Tom Obsessive
Puzzles: 6482
Best Total: 20m 22s
Posted - 2006.06.13 00:30:26
In my latest versions of my program, you can use shift click click and right click clikc to mark edges which are the same or opposite to eachother with colors.  Its currently a work in progress so it generates colors which are hard to see alot of the time, and hard to distinguish from each other.  In order to see colors at all you have to turn on the use empty edge colors option in the settings.  But Its useful for marking things like 2 in a corner in the previous example.

Speaking of my program, would foilman be adverse to my program having an image recognition loader capable of loading the big puzzles you are trying out?  People (who use my program) can then resize, change font size etc before printing them.
mathmaniac
Kwon-Tom Obsessive
Puzzles: 1293
Best Total: 20m 57s
Posted - 2006.06.13 00:38:45
How do I get to this program?
m2e
Kwon-Tom Obsessive
Puzzles: 607
Best Total: 16m 43s
Posted - 2006.06.13 01:10:28
Mathmaniac: its explained here: Link
m2e
Kwon-Tom Obsessive
Puzzles: 607
Best Total: 16m 43s
Posted - 2006.06.13 01:20:12
Here's an image of some of the patterns i use. lodenkamper explained one of these before. The symbols are self explanatory to me (thats why i use them), so im not sure which ones i would need to explain. So ask away if you want!
mathmaniac
Kwon-Tom Obsessive
Puzzles: 1293
Best Total: 20m 57s
Posted - 2006.06.13 01:40:52
m2e I successfully openned the window but it kept having errors and had to keep closing. Besides I couldn't figure out how it worked.
m2e
Kwon-Tom Obsessive
Puzzles: 607
Best Total: 16m 43s
Posted - 2006.06.13 02:31:01
Sorry, i dont know what to do then
Tilps
Kwon-Tom Obsessive
Puzzles: 6482
Best Total: 20m 22s
Posted - 2006.06.13 03:58:53
The right click click and shift click click are a beta feature, and there is a known issue where if you ctrl-click-click or shift-click-click on the same edge (rather then choosing two different edges) you'll get an error. (Edit: This is fixed in the latest beta - also fixed is the load command, it gives a nice error message if you try to load a bad file rather then blowing up the program)

Other thing to do is click the new button so you get a puzzle at all, it opens up with an empty board just so there is something to show.
If you managed to get the window up, then you at least have .net 2.0 installed, which is where most people get stuck.
I'd love to hear details of any bugs you encounter (other then the known issue I mentioned above) and what you were trying to do that you couldn't work out.  Either post them here, or on the deadofnight forums or email me - smith dot deadofnight at gmail dot com
If you get a box saying an error has occured and there is a continue option, the continue option in most cases will let you keep working without issue.  But I would love to have the details copied out of the dialog (including the details section which expands out) and into an email to me (or onto the deadofnight forums) so I can save everyone else from getting those errors.
Last edited by tilps - 2006.06.13 09:28:42
foilman
Kwon-Tom Admin
Puzzles: 3384
Best Total: 24m 6s
Posted - 2006.06.13 07:48:11
Quote:
Originally Posted by tilps
Speaking of my program, would foilman be adverse to my program having an image recognition loader capable of loading the big puzzles you are trying out?  People (who use my program) can then resize, change font size etc before printing them.
A better solution might be for me to allow an "export" option here which produces a file in a format your pogram would understand? That would be much easier to do!
Tilps
Kwon-Tom Obsessive
Puzzles: 6482
Best Total: 20m 22s
Posted - 2006.06.13 08:03:26
That would indeed be a simpler option. (For me at least...)
My file format for an initial rectangular grids of squares is:
[fixedwidthfont]
+ + + + + + + + + +
                   
+ + + + + + + + + +
3   1             
+ + + + + + + + + +
[/fixedwidthfont]
Reading my code, it should blow up if there are any trailing blank lines, but I must say I don't remember that happening when I have hand edited files in the past.  But each line representing the grid must be the same number of characters long.
A grid with Width W and Height H has 2*W+1 columns and 2*H+1 rows when saved.
A grid in progress uses x for excluded edges and - and | for lines in the horizontal and vertical positions respectively.
Example:
[fixedwidthfont]
+x+-+
x2|3|
+-+x+
| x |
+-+-+
[/fixedwidthfont]

... hrmm ... copy paste into a text editor with fixed width font to actually see things properly.
lodenkamper
Kwon-Tom Fan
Puzzles: 21
Best Total: 47m 58s
Posted - 2006.06.13 18:13:34
Quote:
Originally Posted by foilman
I managed to make a bit of progress on this over the weekend... it's now possible to display a puzzle as an image, so the really big ones are displayed almost instantly. They don't print so well... in firefox it's possible to scale the print by doing a print-preview & changing the size drop-down, but that doesn't seem possible in Internet Explorer.

Anyway, for now (until I remove it) here's a sample Super-Big puzzle as an image which you MIGHT be able to print... (you could try saving it to your local machine & printing it another way - let me know how you get on with that if you try it)...

Super-Big Puzzle Test

Thanks much for posting this.  It was most enjoyable to solve.  I saved the image file and printed it with no problems.
Last edited by foilman - 2006.06.14 14:41:32
foilman
Kwon-Tom Admin
Puzzles: 3384
Best Total: 24m 6s
Posted - 2006.06.13 19:05:58
I'm glad everyone's doing ok with the super massive puzzle... because I'm still plodding my way through it! It's been going two days now... and I've done about half.
astrokath
Kwon-Tom Obsessive
Puzzles: 3258
Best Total: 13m 42s
Posted - 2006.06.13 21:07:20
Top left and bottom right, right?

mathmaniac
Kwon-Tom Obsessive
Puzzles: 1293
Best Total: 20m 57s
Posted - 2006.06.13 21:12:54
most of my work has been on the bottom
foilman
Kwon-Tom Admin
Puzzles: 3384
Best Total: 24m 6s
Posted - 2006.06.13 21:26:03
Actually I've done the entire lower right corner and have just started moving out from there. But I just do bits during breaks at work, so in total probably haven't spent so much time on it!
Acorn
Kwon-Tom Obsessive
Puzzles: 555
Best Total: 42m 58s
Posted - 2006.06.13 21:37:55
I'm plodding through it too.  I worked on it for a while last night.  It will probably take me most of the week to do.
PuzzleLover
Kwon-Tom Obsessive
Puzzles: 1033
Best Total: 38m 17s
Posted - 2006.06.14 04:15:31
Quote:
Originally Posted by astrokath
Top left and bottom right, right?

Yep.  Even getting this far was pretty challenging for me.
m2e
Kwon-Tom Obsessive
Puzzles: 607
Best Total: 16m 43s
Posted - 2006.06.14 12:16:14
Well i finished the big puzzle, and found it very fun. So i'm a big supporter for a big puzzle every-so-often. I also liked it how the puzzle was possible to solve using logic only (as in i didnt need to get a pencil to use brute force).
procrastinator
Kwon-Tom Obsessive
Puzzles: 1083
Best Total: 12m 56s
Posted - 2006.06.14 12:35:36
Quote:
Originally Posted by m2e
I also liked it how the puzzle was possible to solve using logic only (as in i didnt need to get a pencil to use brute force).

They're all possible to solve like that. you've just got to think harder sometimes.
drnull
Kwon-Tom Obsessive
Puzzles: 1053
Best Total: 23m 25s
Posted - 2006.06.14 13:07:29
Quote:
Originally Posted by procrastinator
They're all possible to solve like that. you've just got to think harder sometimes.

I'm gonna have to disagree here.  There's a difference between solving a puzzle using lookahead, and solving it using patterns.  What (I believe) you're saying is that using lookahead (no matter what depth) is still logic.  But it's not.  Logic is systematically applying patterns or rules.  Guessing at one configuration, then seeing if that guess works you to contridiction (whether using fix position or doing it in your head) is not, in my opinion, logical.

Now, we each have our own heuristics for deciding where it would be best to "guess", i.e., where we will most quickly (hopefully) reach a contridiction, but they are just that, heuristics.  By definition, that's not logic.

Oh, and by the way.... is there any way we could have "notify me when a reply is posted" default to unchecked? 
Page 2 of 3<123>

Forum Index