Collectible Card Game Headquarters

First of all, hail to all forum users.
I used to collect and play Magic in its first days, but the lack of friends to play with and the absence of retailers near the place where I live eventually drove me off from the game.
However...
Recently I discovered that an imaginative, organized and programming-proficient community is working really hard to create new electronic iterations of this amazing game (and I'm really impressed that the game itself is just like a good wine: older it grows, better it become), and it sparked my curiosity once again.
I've tried all of them (beginning with Manalink) and everyone has its own merits, but MAGE is a different beast. Really, MAGE is by far the best one out there, at least in my humble opinion!!!
Its merits:
- fantastic GUI and overall graphics;
- HQ cards support without major impact on performance, and an appropriate working function to download all of the card art;
- cards from all latest sets (not complete yet, but I'm sure they are working really hard to complete the various blocks);
- essential Magic "experience", without useless orpels: view card collection by blocks, construct your own deck or generate it randomly, play against AI or players in single matches or tournaments;
- fantastic implementation of game mechanics.
All in all, my best wishes to all people programming this amazing piece of software and just a plea:
PLEASE, PLEASE, PLEASE, KEEP MAGE ALIVE MAKING IT EVEN BETTER!

Stop it, you're making me blush. But honestly, thanks for the praise. I too started playing paper magic back in the good ole days but I was more interested in the strategy than the collecting. Which is why I started MAGE. I wanted to be able to play any deck I can dream up without needing a huge stockpile of cards.

MAGE really has become a community effort though. The entire team is filled with talent and dedication. MAGE wouldn't be half the app it is today without them.

As for myself I've been working on MAGE in various incarnations (Visual Basic, VB.NET, C# and finally Java) for almost 10 years and plan to for another 10. So you don't have to worry about MAGE disappearing.

BetaSteward

I think the card by card ai is a wrong approach and a more general card text apprach is better but hey I am not a mage developer YET
the mass of 8000+? cards makes a card by card approach somewhat strange

@hans2

I may be misinterpreting what you are saying so please forgive any misconceptions on my part. We don't have a card by card AI but rather a generic AI that is aware of Abilities and how to potentially use them. A recursive algorithm (minimax) has been implemented to evaluate the game state, any potential actions that could be taken and traverse this decision tree and attempt to find the best usage of its resources before actually committing its action. Basically, it plays a game in the background and once it finds the best move, it does said move and then lets the player go.

To help illustrate why a text based approach could potentially give you some MAJOR headaches is in Arc Trail. According to the rules one could misinterpret it so that this card should be able to target the same target twice due to the fact that the word 'Target' is used twice. Rule 114.3 tells us this but yet Arc Trail states 'Arc Trail deals 2 damage to target creature or player and 1 damage to another target creature or player'. From a raw text parsing perspective if one implemented rule 114.3 this card may potentially be implemented wrong and could potentially be hard to get the fix in place to make the engine recognize the usage of the work another. Now, I'm not saying that a text based approach is impossible, just that it itself will also have its own drawbacks.

well I did not clear tell what I want to say
atm mage does implement card to cardpool based on cards not on text so basically you must make 8000 logic for cards (max if all would be different)
i think more reasonable is make a text apprach card support becuase there are far less text logic samples compared to different cards

lets assume the logic is 100% implemented and a new set comes out without new logic so you could basically implement the new set just by adding the card text file

The card text approach isn't a viable option in my opinion. There are cards that do the same thing but have different wordings. Example effect: you gain 1 life for each <something> vs you gain life equal to the number of <something>
Blunt the Assault vs Golden Urn

Also, as it is right now, many common effects and abilities are already implemented. Many cards are just a combination of those. A couple of cards have some tricky abilities that must have custom implementations and no matter the approach they would have to be implemented separately.

The idea I got from this thread is to write a piece of code that generates cards whose stats can be interpreted.

I just felt the urge to pipe in here and give my 2 cents. Mage is great. Period. I think the main reason some people give up on it is due to its client/server set up, not being able to find the card images and symbols(!!), etc. They simply don't take the time to check out the entire application and push all the buttons. Please keep up the fantastic work.