I discovered this site today. I think it's way cool that Hex is available here, but I am concerned about how the pie rule has been implemented.

The pie rule is an opening protocol designed to make the game more balanced and fair. Without it, the advantage of the first move is quite strong, and the game is not nearly as interesting. After the first stone is placed on the board, the next player has the option, at that point in the game only, to swap sides. It is called the pie rule because it is like when two eaters share the last of the pie. One cuts the pie into two slices, and the other decides which slice to eat.

There are two ways to implement this on a server. Suppose the first player is black. After the first stone is placed, the opponent may choose to play black. The player who played the first stone as black is now white, and makes the next move.

The other way is for both players to keep their colors. After black places the first stone, the second player may remove that stone from the board and place a white stone in the equivalent location. This is a mirror image cell, reflected along the long diagonal. See the diagram below. A black stone on A4 is swapped to a white stone on D1.

I have not played Hex here yet, but apparently a black stone on A4 would change to a white stone on A4. This would result in a much weaker position white than a white stone on D1 would provide. This is NOT equivalent to swapping sides.

I appeal to all you clever board designers. How can the pie rule be implemented correctly here? How easy would it be for players to swap sides? How can a black stone on A4 be changed in one move into a white stone on D1?

Thanks very much for your attention.

EDIT: maybe black a4 changes to white a4, and all the border cells swap colors, so the white borders become black borders and vice versa?

Hey Peyrol,

Welcome to the site. Interesting idea. The extra board Designer (Designer with a capital 'D' is the program) features are geared (pun intended) toward the mechanics of Risk.  Games like Hex (and there are a few of them here) are about as far from the original intents of the Designer as it gets.  As you have alluded to, the fact that these games exist at all speak to the power of the Designer and the creativity of the people who design these boards.  Generally speaking, new Designer features that are coded by Tom would be those that augment the "Risk-like" experience.  There is a LOOONGG list of features on designer's wish-list, and swapping sides is not among them.

That said, there may/could be a way to do it with the existing tools, but it would involve quite a bit of 'coding' on the part of the designer and really, I'm not at all confident that it's possible in the first place. Also, I'm not sure if Amidon37 is even active here anymore, much less willing to spend the time to try and make it happen.

Essentially, there would have to be a set of Auto-Capture and Auto-Neutral commands for each starting stone position, and all of them would need to expire after the first turn.  Off the bat, the only way I can think that it might work would be if the second player either plays a stone normally or plays to a holding territory, which would Auto-Capture the correct position and also Auto-Neutral the starting player's stone.

Neutralizing the holding territory would be necessary to prevent cheating (players playing to the holding area on any turn but the first). This is the kind of feature that's very difficult to implement. It's pretty much an unwritten rule that board designs must not require 'voluntary/in good faith' play by players. 

FWIW, if you are looking for an authentic Hex experience, surely there are other sites out there with dedicated programs for play, right?  As a designer, I generally avoid trying to duplicate an authentic non-Risk experience for this reason.  So for instance, while the board Go-Geared (also here) was inspired by the capture mechanics of Go, it plays nothing like Go.

Wow,  you are correct Peyrol.  I am going to have to think of how to implement it as you say.

But I am not sure when i will be able to get to it.  The real world has gotten in the way of my Wargearing the last couple of years and I am not nearly as active as I have been.

Hey A37,  are my thoughts about implementing it in the ball-park? or are you thinking of doing it another way?

I can't remember how the mechanics of the board work, but once the 'holding' territory is captured and it somehow autoNeutrals with 1 unit - that's a good thing, right? - With 0-sided attack dice, It can't be 'recaptured' anytime later in the game.

It's been a while since I made the board so i can't quite remember how I did the implementation -

So, a quick look -

Looks like 0 sided attacker, 1 sided defender, and a modifier from the false-pie rule territory to allow that one to win.

I am thinking that the sides would start neutral and allow the 2nd player to choose which borders they are playing from.   Something like that -

I could also do pey's 2nd method with a whole set-of off board territories and factories.  Would prefer something cleverer.

Amidon37 wrote:

I am thinking that the sides would start neutral and allow the 2nd player to choose which borders they are playing from.   Something like that -

Here's a similar idea:

First Move: A plays to a miniature "First Move Board." (This would conveniently fit on the map because the boards are parallelograms. :)  B then chooses a side.

Second Move: Factories allocate A's first move based on B's choice.  On B's turn?, factories shut down the Starter board with Auto-Neutrals.

One potential problem is that A must be prevented from playing to the main board during the second move if B isn't able to play to the board during the first move (perhaps this could be remedied by somehow giving B two stones for the first move).

Wow, thanks for responding so quickly. I'm probably being redundant, but for the sake of other readers of this thread, Here's a diagram to demonstrate this third way of implementing swap which I briefly described. The left side is after black's first move, and on the right is what the board would look like if white chooses to swap that move.

At least this way, you would not have to build a "swap map" which pairs off every cell on the board. That might be somewhat tedious for a 19x19 grid. So hey, look on the bright side. Easy for me to say. Speaking of which, I should try to learn this design process myself. I found the Workshop page.

M57 wrote: ... FWIW, if you are looking for an authentic Hex experience, surely there are other sites out there with dedicated programs for play, right?  As a designer, I generally avoid trying to duplicate an authentic non-Risk experience for this reason.  So for instance, while the board Go-Geared (also here) was inspired by the capture mechanics of Go, it plays nothing like Go.

Oh absolutely. We have Little Golem, with probably the strongest field of Hex players on the planet. That's for turn based play. We have Game Center for real time play. We have Ludoteka for insanely fast real time tournaments every Wednesday, 1 minute to make all your moves plus 5 seconds increment per move. We have Board Game Arena for both real time and turn based play. And there are several other servers. All this is free for participants. Some of them provide extra features if you pay, but you don't have to. Like this site I guess. As far as I'm concerned, the more servers that offer Hex, the better.

M57 wrote:

One potential problem is that A must be prevented from playing to the main board during the second move if B isn't able to play to the board during the first move (perhaps this could be remedied by somehow giving B two stones for the first move).

Maybe you could make it so that A has to attack a side B hasn't selected? You could have a timer to know if turn is 2 or 3. A could get the 'choose side territory'(autocapture) on turn 2. A get eliminated if doesn't choose side till turn 3.

So it will look like this:

TURN1: A plays a move. B gets the' choose side territory'(autocapture), and attack a side.

TURN2: A gets 'choose side territory' and attack a side. B plays a move.

TURN3: Not having a side causes an auto elimination. A plays a move. B plays a move.

i think it might be a bit easier (maybe? i'm not a factory expert). 

he's already got a factory to fill that pie-rule territory only on the 2nd player's first turn; which the 2nd player may or may not use. 

he could add a factory that if the pie-rule territory is empty at the beginning of the 1st player's second turn, then a string of auto-capture/neutral factories take place to turn the chosen territory to it's mirror location. 

but what if...the 2nd player takes the 1st territory, but doesn't want the mirror location? 

this isn't going to be easy to explain the first time around so bear with me...it's been a few months since i played hex. is there a way to have 2 attacks that first turn for 2nd player?  i thought it was limited to 1 attack a turn. we need to have 2 attacks possible for this to work. (to make 2 attacks possible it might need that we have an auto-fill factory place 1 unit to the playable bucket beginning of each turn. then you can attack up to X times, but only 1 unit there, so all done!) ok now that's fixed and we have two attacks possible:
there's two off-board territories that come into play for the pie-rule. one that attacks to capture the territory in question, and another that triggers the mirror-flip factories.   

@peyrol

It would be great if you could give it a try.. Of course, graphically, it's a simple board.  But as far as working with graphics is concerned, it's anything but..  Probably the last thing I'd recommend to a novice designer.  The good news is that as far as the (not-so-basic) mechanics are concerned, you can study Amidon37's board.  There are very few extant designers who still visit this site, but of those that do, I'm sure they check the boards frequently, so you'd get a lot of help if you need it.

On the other hand, Amidon37 may very well figure out a reasonably easy edit to get things on his board working in the way you describe.  You could PM him and find out.

Sounds like a tricky problem. I'm not sure if it will be possible to properly implement the pie rule with the existing game engine.

Wow again! I'm very grateful for all this active interest. Yes this certainly does look like plunging into the deep end of the pool from this newcomer's perspective. I will PM Amidon37 after I take some more time to learn what's going on.

Is there an XML file which completely describes this implementation of Hex? Along with the images of course. I found the board description in the Boards tab. Is that list of menu driven choices really all the Wargear engine needs to create a Hex game? Amazing.

I get that the server is supposed to enforce the rules, and players should not be allowed to deviate from them. I'm particularly impressed that Amidon37 found a way to define the game object.

If Hex works here, maybe some other abstracts would also. For example, I started a design for Pex, which has the same object as Hex, but uses pentagons instead of hexagons:

That's certainly doable.  It's these Pie rules that are tricky.

Check out this board also - (which I like better than Hex)

http://www.wargear.net/boards/view/Octagons

Also - this is great implementation of a different game -

http://www.wargear.net/boards/view/RoShamBo

M57 - I came here just to remark that this would be easier with "Real time factories".  If I could trigger factories at the end of the turn. 

Wait, maybe I have it.  I need to make a demo board and try it out- shit I wish I wasn't working so much.

And Pey - we don't need to PM on this.  Discussing it right here is fine. 

See if you can look at this:

http://www.wargear.net/boards/designer/6646

Mmmm.. Real Time Factories :))

The thought warms my heart. This is the kind of thing RTAs are perfect for - the ability to get stuff to happen 'within' a turn. 

Seems like an AutoCapture-N (only autocaptures the territory if it is not neutral) might be helpful also.

Korrun wrote:

Seems like an AutoCapture-N (only autocaptures the territory if it is not neutral) might be helpful also.

I was going to ask if we had those.  I guess not -

Maybe it would be worthwhile to wait for AutoCapture -N or something else to become implemented in the WarGear engine, in order to minimize the effort required to make the fix.

M57 wrote:

Peyrol wrote:

In the meantime, while the swap method remains the way it is, here is a suggested policy for the second player.

If the first move is in an obtuse corner cell, swap it.

If the first move is anywhere else on the edge of the board, don't swap it.

If the first move is not on the edge, swap it.

I haven't put any thought into how to do it, but it might be easier to design it such that you don't offer the swap as a choice - simply force the options described above

TL;DR: This would be a huge kludge, totally at odds with the spirit of simple rules and emergent complexity that the game of Hex embodies. I advise against it, but it's not my decision of course.

I put the above in the general forum, but now M57 has turned it into a design issue. I should explain my reasoning. As of May 2016, as far as I know, computers have effectively solved Hex grids up to 9x9. Here are the "swap maps" for 9x9, 8x8, and 7x7:

In all of these, black makes the first move. A black token in a cell means that, using the normal pie rule, the second player should swap that move. A white token means this is a losing first move for black, and the second player should not swap.

This is the data I used when I came up with my crude suggestion. Nobody really knows, just yet, what the swap maps look like for the grids used here, 14x14 and 19x19. But it seems possible that there will be some similarities. In the first diagram, the cells with small dots mean the computer took a long time to figure those out. That suggests to me that, whether the token gets swapped or not, the game that would arise from such a first move on a larger grid would be a deep game, and approximately fair for two humans. All of the initial moves along the edge of the 9x9 grid are either white, meaning the second player should not swap, or they have a dot, meaning that a close game should result whether the second player swaps or not. I made an exception for the obtuse corner cells, because of personal preference. I believe an initial move in an obtuse corner should always be swapped. As it happens, if an initial token there gets swapped, the swap method works correctly, because of the symmetry of the Hex board. So that's why I suggest a move there could be swapped.

You might notice a big discrepancy, however, between my suggestion and reality for the interior cells one row away from the black border rows. Many of those are white, meaning that if swap were working correctly, if black places a first token there, the second player should not swap it. But since swap is not quite right here just yet, a player who moves first could exploit my suggested policy. For example, on the 9x9 grid consider the initial move b5. This cell is colored black and has no dot, which means this move should be swapped, and it didn't take the computer very long to figure that out. If the second player attacks that black token, he winds up with a white token on b5, which is equivalent to a black token on e2. That cell is labeled white on the map, meaning it's a weak initial move which should not be swapped, and there is no dot, suggesting this would be not necessarily a very balanced game. So, the first player gets a decent advantage whether this flawed swap move is used or not.

Maybe some of those initial moves could be prohibited to the first player. But great googly moogly, what a mess. The whole appeal of Hex to me is all about the emergent complexity that results from simple rules.

I would also like to point out that you all managed to survive somehow before I came along and preached the Gospel of Proper Balance to you. There are plenty of servers and plenty of opponents where one could play proper Hex if one wants. It would be nice if a way could be found to get the Pie rule right, or some other simple protocol, but IMO this ain't it.