lostcities: An implementation of an adictive two-player card game

[ game, program ] [ Propose Tags ]

Based on the card game designed by Reiner Knizia. In this implementation you play against the computer.

[Skip to Readme]

Downloads

lostcities-0.2.tar.gz [browse] (Cabal source package)
Package description (as included in the package)

Maintainer's Corner

Package maintainers

PedroVasconcelos

For package maintainers and hackage trustees

edit package information

Candidates

No Candidates

Versions [RSS]	0.1, 0.2
Dependencies	array, base (>=3 && <4), containers, haskell98, mtl, wx (>=0.11), wxcore (>=0.11) [details]
License	LicenseRef-GPL
Copyright	(c) 2009 Pedro Vasconcelos
Author	Pedro Vasconcelos
Maintainer	pbv@ncc.up.pt
Category	Game
Home page	http://www.ncc.up.pt/~pbv/stuff/lostcities
Uploaded	by PedroVasconcelos at 2009-11-21T23:38:03Z
Distributions
Reverse Dependencies	1 direct, 0 indirect [details]
Executables	lostcities
Downloads	1876 total (7 in the last 30 days)
Rating	(no votes yet) [estimated by Bayesian average]
Your Rating	λ λ λ
Status	Docs not available [build log] All reported builds failed as of 2016-12-30 [all 7 reports]

Readme for lostcities-0.2

[back to package description]

Lost Cities
-----------

This is an implementation of a short two-player card game
in Haskell using the wxWidgets GUI toolkit.
This program allows you to play against a computer AI using
a minimax algorithm with alpha-beta prunning (based on
Bird and Wadler's presentation in the "Introduction
to Functional Programming"). 

The game is played with a deck of cards numbered from 2 to 10
in 5 colors plus 3 "investment" cards for each color (marked with a 
multiplication sign). The game progressed by playing cards
in columns of same colors of increasing value ("expeditions"); 
starting a new expedition costs -20 points and the played
cards are summed; the investment cards multiply the result
(positive or negative).

Crucially, the computer AI does not "cheat": unknown information
(i.e., the opponents hand and draw deck) is hidden
by considering possible scenarios that are compatible
with the known information. It seems to play only moderately well,
but still better than some online AI opponents I tried.


For more details check the Boardgamegeek page:
http://www.boardgamegeek.com/boardgame/50.


If you have any comments please drop me a line,

Pedro Vasconcelos
pbv@dcc.fc.up.pt
Department of Computer Science
Faculty of Science, University of Porto, Portugal