Collectible Card Game Headquarters

I think I started a topic like this awhile ago... anyway, there's a lot of cards the AI (or the player in a generated deck) cannot use effectively (counterspells, Aether Vial, Tromp the Domains, etc.).

Feel free to post some completely useless cards you've received during a constructed match when randomly generating a deck, or have witnessed the AI receive some cards he cannot use effectively at all, and I'll add them to the list of cards that shouldn't show up during these kind of games.

I think Terramorphic Expanse should be on the list.
Compy had it in play in 2 games and never used it.
Compy used the 4-color deck I downloaded a while ago from here.
I actually never saw the card in a generated deck match although I have played plenty of those.

Id say exclude any cards that aren't 100% correct. This way you won't have people asking awkward questions about enchantment auras that are coded as sorceries.

I'm looking through the CardFactory code and I am investigating the cards with Buyback. It appears that the computer can not use Buyback. As an example, the Capsize spell states:

public boolean canPlayAI() {return false;}


I guess that these five buyback spell cards should be added to the GenerateConstructedDeck.java file at the "//cards the AI cannot play (effectively):" section:

Capsize
Elvish Fury
Lab Rats
Sprout Swarm
Whispers of the Muse

Ah no, actually the AI currently *CAN ONLY* use the full buyback option of those spells.
This should be changed I guess (since sometimes it's actually smart to cast Capsize without the buyback portion), but it's tricky to check these conditions.

DennisBergkamp wrote:Ah no, actually the AI currently *CAN ONLY* use the full buyback option of those spells.
This should be changed I guess (since sometimes it's actually smart to cast Capsize without the buyback portion), but it's tricky to check these conditions.

`
Is there a way to determine how much mana the computer has untapped? And tie this in with the number of cards in the computer's hand?

if computer can afford buyback and has has less than 4 cards in hand
set buyback to true
else
set buyback to false

The SpellAbility spell_one and SpellAbility spell_two code would follow and would need to check the buyback flag.

Actually, this is how it would happen already. AI will always try to cast its largest spell/ability.
But I'm not sure if that's much of an improvement, I guess it also depends on the spell, but oftentimes it's a good idea to save up for a buyback...