Mailing-List of Abalone-Programmers 1992-1993:

http://www.cise.ufl.edu/~mpf/abalone/abalone-archive.mail


Abalone2000 Java-Programme:

http://www.informatik.uni-bremen.de/~jakob/abalone2000/index.html  


Games.net-Abalone:

http://www.gamerz.net/pbmserv/abalone.html  


The KA-Abalone Handbook:

http://www.kde.gr.jp/help/doc/kdegames/doc/kabalone/HTML/index.html


Results of the 8. ICGA-Computer-Olympiad 2003 (Graz)

http://www.cs.unimaas.nl/olympiad2003/

-> Homepage of the ICGA (International Computer Games Association)

http://www.cs.unimaas.nl/icga/  

--Last edited by Funky-AbaloneTheory-JazzClub on 2005-10-24 02:21:28 --

Yahoo group, open for everybody. (Low trafic)

http://groups.yahoo.com/group/abalone_prog/

hello,

Is "Nacre/abalone3D" open-sourced? (the link on yahoo.groups points only to a Mac binary)

Is there any sourceforge project out there?
More precisely, is there any sfp that is directly related to the outcomes of the discussions on this forum? (I'm thinking about a whole project: crossplatform gui+net+AI.

thanks for the info,
kr.

Short answer: No, but you can make one.

Abalone3D is shareware -- it says so on the web page.

Nacre is private-ware (you cannot get source nor binary)

The abalone_prog group on Yahoo has two more links, these programs have source. The authors are Oliver Thil and Peter Tax.

If you would like to participate in an open source project we could probably set one up based on Peter or Olivers program. I have a small GUI that I could contribute if the license was MIT or something similar. You can mail me on nacre_computer at  hottmailkom (fix spelling errors to get correct mail address)

If you just want to compile some source, get Olivers program. It should work almost out of the box on Windows with DevCpp

Turkish University project: ABLA.win32 & Abalone3d.java .
(fyi,tbi: they states that their 4-ply implementation has been unbeaten by ABA-PRO at level B??)

 http://cse.yeditepe.edu.tr/~eozcan/projects/ABLA/  

--Last edited by kopedito on 2005-10-28 21:45:12 --

ABLA is not a serious effort.

The program uses two features 1) distance to center and 2) neighbour count. It claims that neighbour count is original, even though Peter Tax mentioned it in 1993 on the Abalone maillist (http://www.cise.ufl.edu/~mpf/abalone/abalone-archive.mail)

Further, it claims that ABLA is unbeaten = AbaPro cannot beat it. This is based on 4 games ABLA vs AbaPro where ABLA searched 4 ply, and AbaPro 3. ABLA uses a full tree, wherease AbaPro prunes the tree. This is hardly a fair comparison. Under tournament conditions, ABLA would probably search no more than 5 ply, where AbaPro easily does 8-9 ply.

My own experiments with Nacre & AbaPro suggests that 1 extra ply is enough to win most of the time. 2 extra ply always wins. This is also why my effort turned from evaluation function to search improvement about 1 year ago.

You need the .NET framework 1.1 and a pentium IV.

download "my serious effort" (i.e. My Lovely Abalone v1.5)  at http://www.mogwai.lunarpages.com/lovely/

unzip then setup. the engine tuning is accessible via the display all windows option in the contextual menu displayed when you move your mouse on the small icon on the right of the board.

use drag and drop of one stone for pushing moves and right click on several stone then drag and drop for broadside moves.

Mogwai

Hi all,

I have just finished porting Peter Tax's Abalone program from Mac OS 9 to Mac OS X
(technically, it has been carbonized, as they at Apple). It is available here :

http://cassio.free.fr/abalone/

I have also included a couple of new evaluation functions and accelerated the search
engine, and I think the program is quite strong now (especially using the last two
strategies : Epaminondas and Hannibal). I would be really interested to know how it
performs against other programs...

The souce code is also available, of course. Thanks to Peter for making it public in the first place !

Enjoy,
Stéphane