MArcomage

surely interesting, but the store-recall system it's not supported.

hmm? could the game not go through the history, find the most recent played Capture, and (this is where the problem probably is) bring back the cards from that turn?

(..this would also be really hard to balance)

If the history module is sophisticated enough for that, I'm guessing it would be a lot for the system to calculate quickly.

Something like this would be better handled by specifically coding it to the game's engine, but there are other things that should take priority for being added to the game than a new card effect.

It's a cool idea, though. (And yes, probably a beast to balance. If Recall is in the Captured hand, you could restore a few key cards indefinitely.)

Probably implementable, but I don't really like the idea. It could possibly require players to browse through the whole game history because of this card.

I'd like to know what are the actual limits of the system.
is it possible to have cards like: "attack N, N=#Alliance played so far"?
what cannot be done?
about this card, if really is's implementable, it should focus only on one card.
With capture you choose a card in hand to store, with recall you get it back.

@Fithz - In this case it will be better to change this two cards into 1 - Select a card in hand to store, if a card was stored, replace a card in hand with the stored card.

imho lasting effect are a bad idea, paving road to other lasting effects, poison attacks, charging cards and etc. Concept numbers will increase exponentially and it all will be impossible to balance. I like current 'simplicity' of marcomage, with strongly defined rules.

From the implementation point of view the system has access to standard game data of all turns played so far in the game. Note, however that traversing this data costs extra server resources. The exact complexity of a card would depend on the traversing effect itself.