| Author |
Message |
Mats W (Kålroten)
| | Posted on Tuesday, November 22, 2005 - 11:27 am: | |
I have implemented Egyptian Siga, a very old game which uses unique "multiple interception-capture." Perhaps it's the origin of alquerque, latrunculi, and many other games. It will be published on the Zillions site later but you can read about it and download it already now on my homepage. Comments are welcome.
http://hem.passagen.se/melki9/Egyptian_Siga.htm
Mats |
Greg Schmidt (Gschmidt)
| | Posted on Tuesday, November 22, 2005 - 8:41 pm: | |
Hello Mats,
Thank you for the advance notice of Siga. I began playing it and found myself in a situation where there were two possible captures and the program allowed me perform one of them. Is there something about the rules that I don't understand? I can send you the .zsg file if you can provide an email address.
Regards,
-- Greg |
Mats W (Kålroten)
| | Posted on Wednesday, November 23, 2005 - 1:11 am: | |
Hmm...that sounds strange. Yes, please send me the zsg-file: mlwi#swipnet.se
(substitute the #)
Mats |
Mats W (Kålroten)
| | Posted on Wednesday, November 23, 2005 - 2:16 am: | |
I realized that my description of the rules was somewhat unclear. I wrote that "if there are more possibilities of capture then one must continue capturing."
It should be: "If the piece with which one began capture has more possibilities of capture then one must continue capturing."
So when you have begun a capture sequence you must follow out the capture sequence with this very piece. Could this be the problem? |
Mats W (Kålroten)
| | Posted on Wednesday, November 23, 2005 - 5:50 am: | |
I fixed another(?) bug which has to do with the tweaking I am forced to do when playing on the two bigger boards in order to encourage the engine to play so he avoids stalemate (otherwise one could win easily). I thought I could use dummy-positions, but for some reason this causes a bug. So I have now exchanged these with real positions, which are used only for tweaking.
So if you've downloaded my program you ought to download it again.
Mats |
Mats W (Kålroten)
| | Posted on Wednesday, November 23, 2005 - 6:40 am: | |
Judging from L. Lynn Smith's implementation of Latrunculi
(a very interesting game) I am becoming even more convinced that Egyptian Siga actually is "The Mother of all Board-games." The kinship is obvious.
Mats |
Greg Schmidt (Gschmidt)
| | Posted on Wednesday, November 23, 2005 - 8:05 am: | |
Mats wrote:
It should be: "If the piece with which one began capture has more possibilities of capture then one must continue capturing."
Yes, that is what I had encountered, thank you for the clarification.
-- Greg |
Mats W (Kålroten)
| | Posted on Wednesday, November 23, 2005 - 12:46 pm: | |
Sigh! When I made that bugfix, I had to introduce a dummy-piece. But then I introduced another bug as the piece number was greater and loss-condition for the two bigger boards was thus altered. But now I have downloaded a new version of the Mother of all Board-games, which hopefully is the final one. |
Mats W (Kålroten)
| | Posted on Friday, November 25, 2005 - 4:16 am: | |
Uwe Wiedemann says here that "siga is not implementable in Zillions yet, because of a changing move order (if you capture for example two pieces you can additional move two times with anyone piece. If you capture again in these moves you get additional moves etc."
Does anybody know where the source for these rules is? I have implemented Parker's rules and these make sense. But Wiedemann's rules seems hopeless. How can one foresee the consequences of a move with such rules? It's hard enough to come to grips with "multiple interception capture."
(BTW, again I made a little bug fix yesterday.) |
Mats W (Kålroten)
| | Posted on Friday, November 25, 2005 - 5:46 am: | |
I just realized that the 5x5 version of Siga is already implemented on the Zillions CD under the name Seega. This implementation uses the rule that a piece on the center square cannot be captured. But as this rule makes the game futile my implementation of the 5x5 version was necessary anyway. I searched Zillions for "Siga" before I started to implement the game, to make sure that it wasn't already implemented. Obviously, one must search after the variant names, too. |
Greg Schmidt (Gschmidt)
| | Posted on Saturday, November 26, 2005 - 8:21 am: | |
Mats wrote:
[Uwe Wiedemann says here that "siga is not implementable in Zillions yet, because of a changing move order (if you capture for example two pieces you can additional move two times with anyone piece. If you capture again in these moves you get additional moves etc."]
I'm frequently amazed by the ingenuity of the game developers. Often, someone finds a way to represent something in the zrf that initially seems impossible. This usually involves some form of hackery, but nevertheless, a workable solution is found. In the above case, I wonder if additional moves can be simulated by forcing the other player to pass. I'm not sure if this would work or not as I haven't tried it, but each move would have a precondition based on a hidden piece. When the hidden piece is present, no moves of that side can be applied resulting in a pass condition for the respective side.
-- Greg |
Ken Franklin (Kenz)
| | Posted on Saturday, November 26, 2005 - 12:25 pm: | |
(post capture Macro1)
Maybe Capture-Counter pieces (not dummy) could be added to spot 1 or 2. I used such hidden "off-board" counting pieces to the right side of Plinko. Drag the Moves-Played window away is see it.
Given two Capture-Counter pieces are present:
Could a move-type pass or "non-move action like a flip, flip of a C-C piece" of the opponent be forced if two or three such pieces were present at the start of each move. This would be done as a 1st Priority move-type.
(post capture Macro3)A 3rd piece could be added if a 2nd piece exists directly next in line.
(post capture Macro2)
Before Macro3, a reset IF motion could be tacked on (2nd move post double capture) to eliminate all three Capture-Counter pieces. |
Ken Franklin (Kenz)
| | Posted on Saturday, November 26, 2005 - 12:31 pm: | |
p.s.
Maybe zone named rows of three capture-counter positions could be applied for continuing move-orders.
Possibly all in a 2nd Grid. I used two visible grids in DoubleAlternate chess & Switch (different pixel sizes). |
Keith Carter (Keithc)
| | Posted on Saturday, November 26, 2005 - 10:46 pm: | |
I am about to give Egyptian Siga a try. Very nice graphics. Did you do them yourself? |
Mats W (Kålroten)
| | Posted on Sunday, November 27, 2005 - 1:11 am: | |
Yes, I dabble some in this, using Paint Shop Pro. I was inspired by the oldest representations of Siga where the squares are represented by round depressions. Siga is a curious game which might not fit us modern people so well. But it's interesting to implement such games for historical reasons. The ancients often used games in religious rites and for divination. Siga is quite useful for this, I suppose, because it's somewhat erratic. |
Mats W (Kålroten)
| | Posted on Monday, November 28, 2005 - 10:31 am: | |
I realized that the engine, in the 5x5 variant, plays very badly during the drop phase, so I added some tweaking which makes it play much better. Earlier one could secure a four piece advantage already in the opening. One must keep in mind that this anticuated game might have been played by Ramses the Great, so it deserves some work. You can download it already, if you're interested in historical games. |
Mats W (Kålroten)
| | Posted on Tuesday, November 29, 2005 - 2:32 am: | |
Ah, well, (cough, cough), I added a new version today (a slight bug fix). |
Mats W (Kålroten)
| | Posted on Thursday, August 24, 2006 - 5:42 am: | |
I have finally realized that the 5x5 version of Siga on the Zillions CD (under the name Seega) is not a faulty Siga, but really a variant deriving from Somaliland. It is called Shantarad, or Bub.
However, my variant Siga is probably faulty because I have not implemented the rule that if a party cannot move a piece then his turn is passed and the opponent must remove one of his own pieces so that the other party can move. Thus, one cannot get stalemated. I have fixed this and this implementation will be published during the weekend, or you can download it now from my homepage. The above rule derives from Murray, but is not mentioned in Parker. That's why I missed it. But it seems a very good rule.
Mats |
|