wade wrote:
> adak wrote:
> > What else do we need to know and agree on?
>
> Do players know the number of mines?

Normal would be 40 mines on the intermediate (16 x 16) board.
>
> What is the scoring system?

Your score would be the number of mined squares you succeeded in
marking correctly during the game. 40 points would be a successful
(perfect), game.

The number of squares you uncover is irrelavant, since many of them
just "flop" open, as the game progresses, and depend on the geometry of
the mines, which you can't know in advance.

> With the scoring above, programs could pick their desired board density
> (a denser board is worth more if successfully cleared, but is harder to
> clear).  You'd need to either limit the density (to, say 50%), or
> eliminate the first-move-is-always-safe rule.  Otherwise a board with
> only one safe square is both high-scoring, and trivial.

You've been really pondering this matter, haven't you? I'll sign you
up!! :)

Unfortunately, I don't know of any Minesweeper programs that have any
AI to make these kinds of decisions!  How could we compare their
performance with a program like you describe?

> Variations are possible (eg. you can pre-spend 10 points to safely
> clear any one square, even if it previously had a mine).

OK,  I'll expect your program entry no later than next week!! You're
hooked! :)

> You answered this, but perhaps some discussion would be interesting:
> Are boards random, or are they created by an intelligent adversary (one
> who knows the scoring rules).

I haven't heard anything about an "intelligent adversary" setting up
the mines. Very interesting (and would definitely wreck havoc with all
the good programs I've heard of), but for this testing situation, I
can't see it fitting in.

That's one of our big challenges - to get several programs to all
accept the same game parameters - size of board, number of mines
(needed to figure the odds properly), loading successive games from a
file that would give the same challenge to every program, playing with
exactly the same rules, and having the same scoring system.

We'll have our hands full just doing this, for a bunch of programs.

Is it too early to start gathering up a few programs? I finally gave in
and started coding up mine.

Adak


--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups 
"Algorithm Geeks" group.
To post to this group, send email to algogeeks@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at http://groups.google.com/group/algogeeks
-~----------~----~----~----~------~----~------~--~---

Reply via email to