[Fixed] Bug Archive

[Gargaroz]

Fixed.

[Gargaroz]

Fixed.

[Gargaroz]

Fixed.

[Gargaroz]

These kind of effects could only be emulated in Manalink, and only for C coded card. If you, for example, Masticore, it won't require any upkeep cost with Eon Hub in play.

[Gargaroz]

Fixed.

[Gargaroz]

Fixed.

[Gargaroz]

Fixed.

[Korath]

Both the cards in his screenshots - Albino Troll and Emberwilde Augur - are in C.

[Althuna]

This has been happening for a long time but I never bothered mentioning it before because it's hard to say when or why it happens. Anyway, when the AI has the land called Desert, often he will get stuck in a loop using the card.

I noticed this last time, that he activated it before I even attacked, and ended up activating it over and over and over, and I had to keep clicking the window (or pressing spacebar or whatever) to close it and it would try to activate again. The game seems to get stuck like that forever. Sometimes if I just hold down enter, it will eventually get past, or sometimes the game will crash, and more recently, it can give a new error message that says AddCard error: no more room in cd to add a card.

Hope this helps.

p.s. It happens with some other activatable lands too. Often the one that can become a 2/2 artifact creature.

[HarlequinCasts]

Describe the Bug:
Haha ok I feel bad reporting this one because I know Coalition Victory has known hard coded issues when you win with it, but this is still a pretty unexpected result even if it is a real corner case.

With a Maelstrom Archangel in play, and one of every basic land type in play, and a Coalition Victory in hand, casting any spell or simply declaring the Maelstrom Archangel as an attacker will cause you to immediately win the game (in the already know bugged out way).

In the attached replay, you may cast either an Urban Evolution, or a Fusion Elemental to immediately win the game. Also you can simply declare the Maelstrom Archangel as an attacker and you will win before you deal damage.

Which card did behave improperly ?
Even though Coalition Victory is know to be buggy, I suspect perhaps it is the Maelstrom Archangel 's "cast a card without paying its mana cost whenever it deals combat damage to a player" code that is perhaps triggering too soon.

Which update are you using?(date,name)Which type(Duel,Gauntlet,Sealed Deck)
June 2013 : the prophecy of dragon's maze nemesis
Both dueling and gauntlet (replay is a duel)

What exactly should be the correct behavior/interaction ?
Maelstrom Archangel only lets you cast a card without paying its mana cost after dealing combat damage to a player.

Are any other cards possibly affected by this bug ?
Uncertain.

[HarlequinCasts]

Worth noting that I have also had unexpected interaction with a Maelstrom Archangel in play, Coalition Victory in hand (but without enough types of basic lands in play due to using Noble Hierarch for white) and activating a fetch land caused me to instantly win...

(If you haven't noticed I've been playing with a Coalition Victory deck a ton...)

EDIT: If you would draw into Coalition Victory with the Maelstrom Archangel on the field you instantly win without the card even in your hand (See screenshot).

[stassy]

It happen also on spell (Lightning Blast) but not on sac (Diamond Valley)

[Korath]

I remember seeing this too.

Theory: lose_the_game() causes an immediate game end even if it occurs during the AI's speculative moves. In the exe code for Lich, for example, the call to lose_the_game() during EVENT_GRAVEYARD_FROM_PLAY only happens if _UNKNOWN728574 != 1, but player's life is set to -99 regardless. I'll experiment some.
----------
OK: On the plus side,

Code: Select all

diff --git a/src/cards/gold_oneshot.c b/src/cards/gold_oneshot.c
index b7fec72..b73e812 100644
--- a/src/cards/gold_oneshot.c
+++ b/src/cards/gold_oneshot.c
@@ -30,7 +30,9 @@ int card_coalition_victory(int player, int card, event_t event){
             if( target_available(player, card, &td) ){ count++; }

             if( count == 5 ){
+              if (unknown728574 != 1)
                 lose_the_game(1-player);
+              life[1-player] = -99;
             }
         }
         kill_card(player, card, KILL_DESTROY );

fixes the reported bug: it makes the game not end until you're prompted for the spell to cast, and you actually pick that one. And it gives a proper you-win dialog, instead of the weird textless one that usually shows up for forced-wins.

On the minus side, I had hoped the life change would make the AI chump block, which it never did. And when I gave the AI a Platinum Angel, I won even though it was still in play (no surprise, since I'm still just calling exe functions directly).

[stassy]

That is the problem, I knew from Paradox Haze that only with C code this kind of effect is doable, that is why I choose C coded cards.

[stassy]

Yup, that is a known issue, unfortunately it even happened in the 1st Mok update so as Gargz usually said : "hardcoded is hardcoded"