Collectible Card Game Headquarters

Describe the Bug:
Quicksand can't be activated because if you haven't got another card that activates or can be played instantly fast effects menu doesn't open.

Which card did behave improperly ?
Quicksand

Which update are you using?(date,name)Which type(Duel,Gauntlet,Sealed Deck)
2014/05/06: Journey into Nyx

What exactly should be the correct behavior/interaction ?
Quicksand can be activated,fast effects menu appear in attack phase even if there isn't another card to activate them.

Are any other cards possibly affected by this bug ?
-

This also affects Kor Haven, Tectonic Edge, Strip Mine, Kher Keep and Wasteland.

I bet there's more.

Do you want me to post all of them separately? Maybe someone should compile a list.

This list should be fine here but we better here from devs in the rare case that this can be fixed "globally" without needing a list and you waist your time.

Damn, it seems an hardcoded problem, I'll ask Korath and let you know

Any similar lands (i.e., with both mana and non-mana activated abilities) that do work?

Are they still broken if you flag them as an artifact? (Type: Artifact column in ct_all.csv)

How about if Flags: Mana Source is removed?

(I can't conveniently run the game here.)

1) It's a global bug, even the manlands are affected.
2) Added "type: artifact" to Mishra's Factory : it still cannot be used during combat.
3) Removed "flags: mana source" from Kor Haven: manastripes gone but it can be used during combat.

I was wondering: isn't there any flag the we could forcefully set during combat to make them activeable (TENTATIVE_LCBP_DURING_COMBAT ?) ?

These only fail during combat, then? Is it only that the phase skips past Fast Effects/Assign Attackers and Fast Effects/Assign Blockers, or are you prevented from responding to the AI in those phases too?

(TENTATIVE_LCBP_DURING_COMBAT is already set, so far as I can tell, during combat; which is why I named it that.)

Do non-land mixed activateables like Mind Stone work?

And is the problem recent - that is, anyone have a pre-Theros version around to test with? Strip Mine and Mishra's Factory make this testable all the way back to the original MicroProse versions.

^ Yes, exactly. The game skips Fast Effects/Assign Attackers and Fast Effects/Assign Blockers, unless you have other (not land-based) fast effects you could play/activate like instants, creature abilities etc, in which case they can be activated during said steps.

Mind Stone can't be sacced to draw a card either (fast effects steps skipped).

Not sure how recent of a problem this is, though.

It is working with the previous update:Manalink Update 2014/04/17: Apr14
I tested Strip Mine and others,so something changed with the last update.
I don't know if it helps but if you manually put stops in the attack steps for fast effects you can activate them.

Could someone test the non-ASM coded card like Kor Haven too in a JOU / pre-JOU patch ?
I tried to revert the minimal changes I made to 'produce_mana.c' for implementing strange stuff like Quarum Trench Gnomes / Deep Water, but no luck, that was not the culprit...

Tried also to revert the insertions I made to make stuff like Staff of the Ages codeable: still no luck

I already have tested with ASM and non ASM cards including Kor Haven and both work with previous update = April ,and don't work now under Joy update.I didn't mention as Korath mostly asked for Strip Mine.

Since this used to work, the right tool to find out exactly how it broke is git bisect. Reverting changes at random isn't likely to help.

In the meantime, you should be able to work around the problem by adding stops to the pre-block and post-block phases.

Edit: This is the culprit. Since no other patch to the EXE was done after the April release, I tried taking that version of the EXE and updating it with the new CSVs.
Result: lands with activated abilities works back at instant speed during combat.
-----

I found an interesting commit, I report it here for you Korath, but since is all ASM, I really can't help with it :

Code: Select all

diff --git a/Magic.exe b/Magic.exe
index bdacf00..d9466be 100644
Binary files a/Magic.exe and b/Magic.exe differ
diff --git a/src/patches/patch_charge_mana_can_activate.pl b/src/patches/patch_charge_mana_can_activate.pl
index 4393be0..6d2de87 100644
--- a/src/patches/patch_charge_mana_can_activate.pl
+++ b/src/patches/patch_charge_mana_can_activate.pl
@@ -3,6 +3,8 @@
 # Updates Magic.exe in-place.
 # 1. Always sends EVENT_CAN_ACTIVATE when manually drawing mana during charge_mana(), rather than assuming tapped permanents and summoning-sick creatures can
 #    never activate, and (otherwise) lands always can.
+# 2. Allows cards flagged both EA_MANA_SOURCE and EA_ACT_ABILITY to be activated whenever one with just EA_MANA_SOURCE can, by essentially gutting
+#    is_mana_source_but_not_act_ability().
 
 use strict;
 use warnings;
@@ -29,6 +31,7 @@ patch("Magic.exe", 0x42e913,
 ################################
 # Skip check against TYPE_LAND #
 ################################
+# Previous contents:
 #42e9b9:   8b 45 f0      mov   eax, dword [ebp-0x10]      ; eax = iid
 #42e9bc:   c1 e0 03      shl   eax, 0x3         ; eax <<= 3
 #42e9bf:   8a 8c c0 38 70 7e 00   mov   cl, byte [eax+eax*8+0x7e7038]   ; cl = cards_data[iid].type
@@ -39,3 +42,13 @@ patch("Magic.exe", 0x42e913,
 patch(0x42e9b9,
       "eb 10",   # jmp 0x42e9cb
       0x90);
+
+###################################################################################################
+# Allow cards flagged both mana-source and act-ability to be activated whenever a mana-source can #
+###################################################################################################
+# Previous contents:
+#435c58:   f6 84 c0 4c 70 7e 00 01   test   byte [eax+eax*8+0x7e704c], 0x1   ; if (cards_data[iid].extra_ability & EA_ACT_ABILITY
+#435c60:   74 02         je   0x435c64         ;     == 0) goto 0x435c64
+patch(0x435c58,
+      "eb 0a",   # jmp 0x435c64
+      (0x90) x 6);


The bug that was put in to fix.

is_mana_source_but_not_act_ability() is called from dispatch_event_to_single_card() and two places in sub_475A30() (which is part of the user interface). I suspect, based on absolutely no evidence whatsoever, that we'll need to call something equivalent to the current (changed) version from the former, and something equivalent to the old version from the latter.