[Oz] The Oz Minesweeper

[Peter Van Roy]

> Nice, finally some games in Mozart. Do you have the source available?
> I'm interested to see how you used QTk and when start talking of lines
> of code, I want to see them.

Raph can give you the code.  I think that his plan is to put the code on
the Web page, together with some explanation of how it works.

> I got the impression that the rules lacked something. There where
> occasions where the auto-player stopped, but I could continue.

Did you try the 'Search' feature using the right mouse button?
This does a global search.  Otherwise, the Constraints propagation
does only local propagation, which is clearly not enough.  The game
separates the two to see the effect of local vs. global deduction.

> An interesting problem would be to place the mines such that the "sweep"
> is always possible. The most annoying thing is when you're almost done with
> a large map just to find a corner where the information is too scarce. You
> end up guessing instead of solving.
> 
> /Erik
> 

Yes, it seems that there are certain patterns of mines that make deduction
impossible -- guessing is needed.

There also seems to be a critical percentage of mines - 15% to 20%.  Above
it, you almost always lose, since the mines block deduction.  Below it, a
very simple deduction is enough to win.  The percentage seems to decrease
slightly with increasing board sizes.  I have tried boards up to 2000
squares -- do not try this size board manually unless you want to go
crazy :-) !

Peter

-
Please send submissions to users at mozart-oz.org
and administriva mail to users-request at mozart-oz.org.
The Mozart Oz web site is at http://www.mozart-oz.org/.