Copenhagen Hnefatafl

Tafl rules
Adam
Posts: 117
Joined: Fri Mar 20, 2015 1:28 pm

Re: Copenhagen Hnefatafl

Post by Adam » Sun Jan 25, 2015 7:20 pm

Regarding Q2, the threatening player is defined as the player who is the first to create a repeat in board position in an exchange of repeat moves. Put simply, the one who started it.

Sigurd
Posts: 10
Joined: Sat Mar 14, 2015 10:12 pm

Re: Copenhagen Hnefatafl

Post by Sigurd » Tue Jun 02, 2015 7:21 pm

In connection with Rule 7, something I've been wondering about...
Is this:
[invalid link www.tim-millar.co.uk/section509308_182391.html]
a valid white win exit fort?

sqAree
Posts: 17
Joined: Sat Mar 21, 2015 8:51 pm

Re: Copenhagen Hnefatafl

Post by sqAree » Wed Jun 03, 2015 4:11 pm

Yes it is!

User avatar
Hagbard
Posts: 688
Joined: Sat Mar 14, 2015 6:07 pm
Location: Copenhagen
Contact:

Re: Copenhagen Hnefatafl

Post by Hagbard » Mon Aug 31, 2015 10:40 pm

sqAree wrote:there have been more and more dicussions about the rules in the small chat lately and I'd like to contribute to the game by taking part in them.
A brilliant analysis!! Thanks! I believe that we followed all your suggestions by this rules formulation:
http://aagenielsen.dk/copenhagen_rules.php

User avatar
Hagbard
Posts: 688
Joined: Sat Mar 14, 2015 6:07 pm
Location: Copenhagen
Contact:

Re: Copenhagen Hnefatafl

Post by Hagbard » Fri Dec 18, 2015 6:15 pm

Summary on the Copenhagen Hnefatafl:
http://aagenielsen.dk/copenhagen_summary.php

User avatar
Hagbard
Posts: 688
Joined: Sat Mar 14, 2015 6:07 pm
Location: Copenhagen
Contact:

Re: Copenhagen Hnefatafl

Post by Hagbard » Fri Dec 18, 2015 6:21 pm

The shieldwall rule.

Does somebody know the origin of the hnefatafl shieldwall rule?

The rule is explained in this German page:
http://www.hnefatafl.net/englisch/ehnef ... Schildwall
The Shieldwallrule

Shieldwallrule means the application of the double trap and the capture of the king on the edge of the board to the ordinary taflmen. The shieldwallrule is used in order to prevent one side, especially the attackers, from resting safe on the edge of the taflboard.
So, if several taflstones with the same colour stand side by side on the edge of the taflboard and one opposing taflman is placed in front of each of them, it is possible to capture the complete row at once by flanking the opposing taflstones. In contrast to the so far explained kinds of capturing the refuges do NOT count as substitute taflstones in this case.

Capturing by Using the Shieldwallrule
If there is the same number of opposing taflmen in a row in front of a row of taflmen the whole group can be captured at once by bracketing them.

The shieldwallrule is rarely used. Rather, it enables the breakup of a row of opposing taflmen on the edge of the board by forming a row with one's own taflstones because the other player wants to avoid the loss of several taflstones at once.
These rows facing each other remind of a shieldwall, a viking battle formation, the rule is called after.


A shieldwall rule is also mentioned in this scientific paper from 2007 on artificial intelligence, "Evolving Players for an Ancient Game: Hnefatafl":
http://www.wfg.csse.uwa.edu.au/publicat ... G2007f.pdf
There may be a “shieldwall” rule, allowing the capture of a connected group of pieces by completely surrounding them up against one of
the sides (reminiscent of capturing groups against the side of the board in Go);

cyningstan
Posts: 38
Joined: Sun Mar 15, 2015 8:22 am

Re: Copenhagen Hnefatafl

Post by cyningstan » Sat Dec 19, 2015 10:33 am

Hagbard wrote:Summary on the Copenhagen Hnefatafl:
http://aagenielsen.dk/copenhagen_summary.php
Just a quick aside: although the Swedish sections of the diary were translated by Troilius, Smith says (in his preface) that he had to redo the Latin passages himself. So Smith, rather than Troilius, is responsible for the part of the diary that are relevant to us.

User avatar
Hagbard
Posts: 688
Joined: Sat Mar 14, 2015 6:07 pm
Location: Copenhagen
Contact:

Re: Copenhagen Hnefatafl

Post by Hagbard » Thu Jan 14, 2016 9:10 pm

I did an analysis on the archived games and found that no Copenhagen game
lasted more than 64 (half)moves after the latest capture of a piece (i.e. 32
full moves). I would suggest to simply transfer the chess fifty-move rule into
the Copenhagen hnefatafl. Wikipedia says:
"The fifty-move rule in chess states that a player can claim a draw if no
capture has been made and no pawn has been moved in the last fifty moves (for
this purpose a "move" consists of a player completing his turn followed by his
opponent completing his turn). The purpose of this rule is to prevent a player
with no chance of winning from obstinately continuing to play indefinitely
(Hooper & Whyld 1992:134), or seeking to win purely by tiring the opponent
out.
All of the basic checkmates can be accomplished in well under 50 moves."
This goes for Copenhagen as well, since games were ended after 32 or less
moves.
With such a rule there's an end to any game, and no game can continue
indefinitely.

August 2015 this rule was added to the Copenhagen Hnefatafl:
10. The game is a draw if no capture has been made in the last fifty moves (for this purpose a "move" consists of a player completing his turn followed by his opponent completing his turn).

User avatar
Hagbard
Posts: 688
Joined: Sat Mar 14, 2015 6:07 pm
Location: Copenhagen
Contact:

Re: Copenhagen Hnefatafl

Post by Hagbard » Thu Jan 14, 2016 9:40 pm

As for the two Copenhagen rules for repetitions, 7c and 8, they cover two sorts of repetition:

1)
The side stepping repetition. Often an eternal check repetition, in any case a repetition to break out from some position, but in vain, since the break out attempts are all the time stopped by the opponent's blocking moves. The pressure for an open path could be done by black as well as white.

There are two questions to this repetition:

First question: Who is the guilty party?
Answer: The guilty party is the player, who seeks an open path out. You can identify him by observing that he's all the time side stepping towards an open path. So it is he who is "the threatening player". The opponent is characterized by all the time blocking these attempts.
It is "the threatening player" who has the free will to move somewhere else, but the blocking player has no choice, he has to react and block, otherwise the "threatener" breaks through and probably wins immediately.

Second question: When is it enough? When has so many repetitions occurred so that the game must end with "the threatening player" losing?
Answer: When the identical, total board position is reached for the third time, the game ends. No matter which player actually created the first instance of that board position, because we've already identified the guilty player who must lose.


2)
In the end game, sometimes black has confined white to a limited space. Now, if it's the full enclosure, we have rule 7b and do not have to watch for repetitions.

White edge draw fort.
If it's the white edge draw fort (the illegal sort), then white is normally confined to a very small space and cannot avoid to do a lot of white position repetitions. Perhaps white has some loose men outside the fort, in which case black kills those, and then come the white repetitions. Normally in such a situation, only a few blacks are tied up in surrounding the white fort, so normally a lot of free black men move freely around on the full board.
Now, if we in this case look for total board repetitions, we won't find any for months, because the black party moves freely through the whole board. And who is the "threatener"? White doesn't threaten, he just stays within his fort, and black doesn't threaten either, he just maintains his surrounding.

First question: Who is the guilty party?
Answer: The only one who can break this situation, is white, and therefore we forbid the draw fort, and white is the guilty party who must lose if this continues.

Second question: When is it enough? When has so many repetitions occurred so that the game must end with "the guilty player" losing?
Answer: When the identical, white board position is reached for the third time, the game ends and white loses.


Other sorts of general draw forts.
In the Nath-game diagram which illustrates this rule, there's *not* an actual draw fort, but black and white stand in a trench position which no one of them wants to break. Also in this case, we won't find total board repetitions, because black has many free men to move freely around on the full board.

First question: Who is the guilty party?
Answer: Either of the players could break the situation, but no one wants to. White however has the duty to escape, whereas black has no other duty than to prevent the escape, which he apparently did. So white is the guilty player.

Second question: When is it enough? When has so many repetitions occurred so that the game must end with "the guilty player" losing?
Answer: When the identical, white board position is reached for the third time, the game ends and white loses.
Last edited by Hagbard on Thu Apr 14, 2016 7:05 am, edited 1 time in total.

User avatar
Hagbard
Posts: 688
Joined: Sat Mar 14, 2015 6:07 pm
Location: Copenhagen
Contact:

Re: Copenhagen Hnefatafl

Post by Hagbard » Thu Jan 14, 2016 9:48 pm

The draw in Copenhagen is still an interesting subject. I remember that it was Adam's hobbyhorse to get rid of as many draws as possible in the Copenhagen game. And a very useful hobbyhorse it was too, since it turned out to be possible to do so and still have a well functioning game, probably even better functioning.

I think it was mentioned at some early time of the discussion, that the zero-order, maximum simple way to get rid of all draws would be to say:
White has to win. If white doesn't win, white loses. No draws.
If this were the rule, black would not have to actually catch the king, but only one way or the other to prevent the king's escape.

Such as the Copenhagen rules are written, this is very nearly where we are today. With only two exceptions left:
- black can lose on repetitions too and not only white.
- if both parties have too few pieces left to continue the game, it is a draw.


In the eternal check, it's "the threatening player" who stalls the game, because he refuses to do an alternate move, being he white or black.

In the traditional white draw fort, it's white who stalls the game, because he refuses to try to escape.

In the Nath draw fort, it's both players who stall the game. White refuses to try to escape (because it's impossible), and black refuses to try to catch the king (because it's impossible).
If black had to catch the king as well as white has to escape, this would actually be a draw. Only when we don't require black to actually catch the king, we avoid this to be a draw.

We have a precedent for this in rule 7b: "If the attackers surround the king and ALL remaining defenders, then they win, as they have prevented the king from escaping." This rule actually comes all the way from David Brown, and we reinvented it facing the Two-move Monstrous Fort problem.
Without rule 7b we have the Monstrous Fort.
So, logically black must not be required to actually catch the king, but only to prevent his escape, because otherwise we'll get the Monstrous Fort.
This is a problem for any king type, not only the strong king, because the Monstrous Fort can be built in any game variant on boards >= 11x11.

Post Reply