Propose a method that is completely free (no mtgo) and allows players to have a game of Magic with the same advantages of correspondence chess. Ideally, neither player needs to own any of the cards involved, but deck lists would have to be posted, as well as a mediating random number generator for drawing, coin flips, etc. Another Issue would be the stack. Waiting for a response at each and every priority shift would take too long, so how could action be consolidated?
|
0
|
|||||||||||||
|
|
3
|
Basically, as ripper said, Magic isn't a perfect information game. Each players knows things the other player doesn't (for example, the cards in his hand) and thus to prevent cheating you'll need a mediator with perfect information. This means that every action that involves luck or chance (drawing cards, flipping coins) has to go through the mediator. The process would then be something like this:
As you can see, this can get very tiresome with a human mediator. A computer mediator such as MTGO or even the one provided by MWS would do (note that the mediator doesn't have to understand the rules of the game, only keep track of which cards are in which zone). As for priority passing, there are actually rules for that:
These happen implicitly all the time in face-to-face games, but will need to be formalized in games played over distance. Again, player one might say "I propose to move to the beginning of the next turn" to which player two might respond with either "I accept the shortcut; I untap my permanents and draw a card" or "I shorten the shortcut to the beginning of your end step; then I play X". Basically, any message exchanged in the protocol acknowledges the previous shortcut and proposes a new one; any subsequent action by the other player assumes the shortcut has been approved, unless explicitly stated otherwise. This allows two players to conduct a game almost as if they were face to face in the sense that it conserves messages for uninteresting events ("I pass priority" - "I pass as well"). Players would have to be careful so as not to divulge any information in their shortcuts: for example, if a player says "I attack with creatures A and B" and the other player responds with "I shorten the shortcut to the declare combat phase and play X on B, preventing it from attacking" then some player one "gave away" some information he didn't have to (maybe player two wouldn't have played X if he didn't know B was going to be declared attacker). Note however that this kind of information leak happens all the time in face-to-face games as well (when was the last time you heard someone say "I declare combat and pass priority, do you want to do anything?"), and players just have to choose their own path between letting the game flow or being as secretive as they can. The problem is increasingly apparent in online MWS games where players usually skip phases all the time (with the notable exception of the "end my turn" convention, allowing players to play end-of-turn moves). So basically, yeah, you can play Magic over a distance, given an unbiased, fast mediator, a quick way of exchanging information and/or keeping reference (most probably internet-based). You basically just need something like MWS, only with players being able to keep a connect, make a move, then disconnect whenever they please (where the game state would persist over time on some server until explicitly shut down). |
|||||||||
|
|
0
|
I'd like to suggest some middle-thingie between a client-server-based mediation and a client-only game which depends on trust about their resources. I'm trying to implement such thing but it's far from finished so don't expect too much. you can of course look on my blog. The Idea is that both clients have full information and can therefore verify if some action is possible. however, the application prevents the players from looking at ressources they shouldn't look at. This of course requires some trust, but instead of "I trust that you really have the card you play in hand", it's shifted to "I trust that you don't write some software that tells you parts of the game state you shouldn't know", which is much easier to accept. Now the idea is that everything a player does (or doesn't - pass priority) is an action that modifies the game state and that action is transferred to the other player(s). Every player has a duplicate of the game state and can perform the action on his local copy. The clue behind this is that even the random number generators - which aren't random but only compute a hard-to-predict number sequence - are synced between all players. That means that the initial game state only contains the ordered deck lists and the seed for the generator. how the decks are shuffled is a result of the seed. |
|||||||||||||||
|
