Darryl, somone mentioned hitting a bomb on there first move, and you said that everyone will get their share of bad luck, why not just reset that game like it never happened? More fairness that way.
Darryl, somone mentioned hitting a bomb on there first move, and you said that everyone will get their share of bad luck, why not just reset that game like it never happened? More fairness that way.
Hmmm, I thought I edited that post...oh yea I did, keep reading. Here's the recap...Originally Posted by Cool-August
No one hits a bomb on the first move, Originally I did say tuff luck :-) but then i found out that the windows version of minesweeper guaranteed the first move is not a bomb...so... what happens now behind the scene is that if your first move would had been a bomb, it quickly shuffles it's elsewhere and treats it as if it was blank all along.
*** Sneak Preview ***
I saw a post the other day here about "sliding puzzle", properly called 15 puzzle, created by Sam Lloyd, and I thought that would make a nice contest too, so after this is done. The Kicker is you can use any programming language you want ( or do it manually) . Details to follow at the conclusion of this contest.
Last edited by Darryl; 04-11-2006 at 10:35 AM.
Darryl, just to confirm, did you receive my submission?
yes I did, but havn't tried it yet... i will give it a try within the next few hoursOriginally Posted by Dante Shamest
Err... but isnt that puzzle mathematically unsolvable?I saw a post the other day here about "sliding puzzle", properly called 15 puzzle, created by Sam Lloyd, and I thought that would make a nice contest too, so after this is done.
Look up a C++ Reference and learn How To Ask Questions The Smart WayOriginally Posted by Bjarne Stroustrup (2000-10-14)
I have been solving them by hand since I was a little kid! wait for the details of the contest.Originally Posted by laserlight
Last edited by Darryl; 04-11-2006 at 12:51 PM.
Only if the program uses a completely random method to place the tiles. If you design the scramble method to shuffle the tiles as you would on a real-life tactile board, this shouldn't be a problem.Err... but isnt that puzzle mathematically unsolvable?
Will you be turning optimizations on?Originally Posted by Darryl
Yes, please optimize, or we won't know whether the submissions work for a few days.
P.S. I assume you got mine from last week as well. Feel free to announce how badly it beat's Dante's.
yes I do have yours and mind you this is only a mini run of 100 instead of 1000 which only took about 15 seconds, so scaling up, the full run should take 150 secs unless it doesn't scale linearly.Originally Posted by jlou
*edit*
I added a counter to track wins and here are the results, which are quite interesting.
RandomBot: -188 :wins 0
SimpleBot: 39113 :wins 0
Dante: 339271 :wins 239
jlou: 220378 :wins 30
Last edited by Darryl; 04-11-2006 at 02:24 PM.
If this isn't a fluke, you beat me by -118893.Originally Posted by jlou
edit: updated to reflect new scores
Last edited by Dante Shamest; 04-11-2006 at 02:25 PM.
I must say that is quite impressive that you are winning almost 1/4 of your games
I just found this link for a minesweeper solver that claims 14% for expert: http://www.nuprograms.com/minedefuser.html
Here's quite a scholarly link as well, they ended up with an avg. 25% for expert with a high of 30%, so Dante's is probably all it can be :-)
http://www.dcs.warwick.ac.uk/~kasper...les/report.pdf
Last edited by Darryl; 04-11-2006 at 02:58 PM.
Holy crap.
I guess maybe I have some work to do... I thought you said you were out of ideas, Dante.
I have one and a half more "rules" to put in, but beyond that the only thing I can think of is to mimic the question mark part of the actual game, and that would just be more work than I have time for.
Wow... I just looked at the link and its rules. It turns out that those are the exact same rules I am using or plan on using (even in order). Oh well... I guess it isn't a secret anymore.
Small confession to make: My current Honours project involves writing an AI to play Minesweeper Flags, the MSN version of the game. If you've never played it, it's almost like winmine.exe, except the object is to open mines, instead of avoiding them.
I modified it a bit for use in this contest.
This is the main homepage of my project.
This is a slightly outdated description of the AI I'm using.
edit:
I honestly was. But turns out I had included a stupid bug. I found very safe squares, but since the code was originally meant to be a Minesweeper Flags AI, it avoided them.Originally Posted by jlou
The file I sent to Darryl also includes a "limiter" to stop checking regions with more than 12 possibly mined squares. If the limiter is removed, I think my code would actually do even better, but would take much longer to run.
Last edited by Dante Shamest; 04-11-2006 at 03:10 PM.
I was wondering about this. How can we all play the same boards if it's shuffled like that?Originally Posted by Darryl
There is a difference between tedious and difficult.