SV: SV: hist-games: Zamma

Mats Winther mwi9 at swipnet.se
Thu Apr 27 00:54:29 PDT 2006


Den 2006-04-26 23:11:47 skrev Peter Michaelsen <PMI at KM.DK>:

> Hello Mats,
> I think it would be great, if you would implement the two mentioned variants,
> and perhaps also the variant on the 25 points board: 'kharbeg'/'khreibga', described by Th.Monod and M.
> ould Hamidoun.
> This game is probably identical with the Moroccan 'quireq'/'fetach' described by Abdelmajid Bouhajbane.
> Thierry Depaulis kindly sent me the text by Bouhajbane. The rules of 'quireq' (same word as 'querque' in
> 'al-querque') are too short to be of much use.
> Bouhajbane gives the rules of another alquerque-variant, however, which you might perhaps want to
> implement: 'felli'/'fich', played on a double triangle with 13 crossing points and 2x6 pieces.
> This is perhaps the smallest alquerque-variant with promotion, or, perhaps better: the smallest
> traditional checkers variant, if promotion is regarded as the characteristical feature of checkers,
> distinguishing this game from alquerque.
> Zillions-of-Games is a good tool to test hypothetical reconstructions of games.
> I have tried Ulrich Schädler's reconstruction of Ludus latrunculorum, and let also the program play with
> itself. I am not convinced that Schädler's game was the same as the old Roman game. As far as I remember,
>  the game played by Zillions lasted for several hundreds of moves, and I am not sure that the old Romans
> were patient enough for that.
> Your Damiano is also very interesting. I let Zillions play it against itself, and one game lasted for
> c.160 moves. This is a good game for very patient people, who like to play games like Chu Shogi, or even
> larger Shogi variants. I personally like to play Go, but I do not often have the necessary time for the
> 19x19 game. If I play the computer I normally play 9x9, and sometimes 13x13. Damiano is a very
> positional and strategical game, but I am afraid that most players might find it too slow.
> As I told you, Dr. Arie van der Stoep has some different, and very interesting ideas concerning the
> origin of draughts/checkers. I quote from his newest book, Draughts in relation to chess and alquerque,
> 2005, p.178: "16.7 Summary  In parts of Europe in 1000 AD, two draughts variants were played on lined
> board A: a variety with a short doubleton under a name that derives from the Latin word MARRUS = 'piece',
>  and a variety that derives from Latin speaking people, i.e. people with a Roman civilization. The
> influence of the Romans is also showed by the use of flat round gaming pieces. The variety with the long
> doubleton was introduced in Spain by the Moors. Arab influence is also proved by the gaming pieces that
> were used, pawn shaped figures.
> Draughts has been transferred from the lined board on to the chess board. Where this happened in a
> culture where the chess board was chequered, such as in France, Italy and Spain, draughts became a
> diagonal game. In cultures where people played chess on an unchequered board, such as in Turkey and the
> Middle East, draughts became an orthogonal game."
> A.van der Stoep poses the question on p.140, if the known 20th-c. variants of draughts on a lined board,
> games like Zamma, and the smaller Fetach/Quireq (to which might be added the similar Mauretanian Kharbeg/
> Khreibga) are not in fact relics, survivals from the past.
> If this is correct, the variety with a long promoted piece introduced in Spain by the Moors, is still
> played by the Moors and other peoples in North Africa.
> The other variety with a short promoted piece on the lined 25 points board, played by Latin speaking
> people, seems to have disappeared. One might perhaps use Zillions to test reconstructions of this game,
> or games? Perhaps there was an ancestor of Italian checkers played on the lined alquerque board, in
> which the king could only be captured by another king, and perhaps there was another variant without
> such a rule: the ancestor of Anglo-French, later: Anglo -American draughts/checkers?
> Van der Stoep does also mention an Indonesian game from Java: 'Permainan Tabal', also named 'Dama',
> played on an extended alquerque de doze board with two extra triangles, with in total 2x16 pieces on a
> board with 37 crossing points. Like in the smaller game described by Monod and Hamidoun, and also like
> in the Zamma game played on the same board as the Mauretanian game 'Kharbaga', pieces did move in all
> directions; they were not restricted to forward movement in their unpromoted state. Like in the North
> African game the kings was long. I hope to be able to get more information about this game, which seems
> to be an exception from the rule that promotion in games on the lined board is a phenomenon confined to
> North Africa.
> I will try to ask Dr. v.der Stoep for more information about the latter game.
> I shall try to send you as exact rules as possible of the mentioned games, and look forward to play them
> with your Zillions implementations.
> Best warmest wishes,
> Peter.

Of course, Damiano takes many moves, but the gist is its positional
character. This is easier for the human cogitation, i.e., strategical
foresight rather than concrete calculus. This means that it takes
much less time to decide on a particular move. So the number
of moves is not the sole factor that decides whether a game is slow.
So I suspect that Damiano takes not much longer than regular checkers
games to play.  In a human-human game, the party with king minority
would generally resign, as his opponent would only encroach upon his
kings and surround them, whereas a computer would only attain a win in
a roundabout, time-consuming, way. But it's hard for me to make
judgements on checkers games, because I'm not good at this game type.
I'm specialized in chess-type games (having many piece types). But I'm
attracted to checkers games due to their archetypal simplicity.

I have also realized how valuable Zillions is for historical game
research. For instance, I've "proved" that my rendition of "16
soldiers" is the correct one, due to the fact that the game works so
elegantly, combined with simplicity of rules. Murray's and other's
renditions simply won't work when implemented in Zillions. I also
refuted his rendition of the "Tamil Leopards game" by this method.

However, development of Zillions is discontinued. Obviously, it's not
much money in this area. But since the software is so valuable for
research one would wish that institutions and journals could subsidize
a next release. It's not only valuable for historians, but could also
be of use for other branches.

Mats



More information about the hist-games mailing list