SVN Bug Reports
Post MTG Forge Related Programming Questions Here
Moderators: timmermac, Blacksmith, KrazyTheFox, Agetian, friarsol, CCGHQ Admins
Re: SVN Bug Reports
by friarsol » 13 Mar 2011, 22:35
Oh I only tested with Lurking Skirge. It's trigger kept hitting the stack after it was a creature which is why I noticed.
- friarsol
- Global Moderator
- Posts: 7593
- Joined: 15 May 2010, 04:20
- Has thanked: 243 times
- Been thanked: 965 times
Re: SVN Bug Reports
by slapshot5 » 13 Mar 2011, 23:13
Ok. That's fixed now. I double checked all the rest.friarsol wrote:Oh I only tested with Lurking Skirge. It's trigger kept hitting the stack after it was a creature which is why I noticed.
-slapshot5
- slapshot5
- Programmer
- Posts: 1391
- Joined: 03 Jan 2010, 17:47
- Location: Mac OS X
- Has thanked: 25 times
- Been thanked: 68 times
Re: SVN Bug Reports
by UnderFlow » 14 Mar 2011, 03:58
Revision 7586:
The Computer controlled a Nightsky Mimic enchanted by Edge of the Divinity (so that it was a 5/4). It played a white & black spell and the mimic became 4/4 (should be 7/7).
Update:
My opponent cast Duergar Hedge-Mage with 2 Sacred Foundry on the battlefield. When it etb, the triggers for its abilities said "[...] (Targeting Font of Mythos)" and "[...] (Targeting ERROR)", respectively. The computer had a Squee's Embrace, so it should have had to target it.
Another Update:
The computer has a Balefire Liege and casts Belligerent Hatchling. In response to the Lieges triggered abilities I cast Cryptic Command in counter & tap modes. It taps the opponents creatures, but does not counter the Hatchling.
The Computer controlled a Nightsky Mimic enchanted by Edge of the Divinity (so that it was a 5/4). It played a white & black spell and the mimic became 4/4 (should be 7/7).
Update:
My opponent cast Duergar Hedge-Mage with 2 Sacred Foundry on the battlefield. When it etb, the triggers for its abilities said "[...] (Targeting Font of Mythos)" and "[...] (Targeting ERROR)", respectively. The computer had a Squee's Embrace, so it should have had to target it.
Another Update:
The computer has a Balefire Liege and casts Belligerent Hatchling. In response to the Lieges triggered abilities I cast Cryptic Command in counter & tap modes. It taps the opponents creatures, but does not counter the Hatchling.
Last edited by UnderFlow on 14 Mar 2011, 05:11, edited 1 time in total.
Re: SVN Bug Reports
by slapshot5 » 14 Mar 2011, 05:07
It was losing it's colors. That's why the pump wasn't correctly applied while it had the higher P/T. Fixed in SVN.UnderFlow wrote:Revision 7586:
The Computer controlled a Nightsky Mimic enchanted by Edge of the Divinity (so that it was a 5/4). It played a white & black spell and the mimic became 4/4 (should be 7/7).
-slapshot5
- slapshot5
- Programmer
- Posts: 1391
- Joined: 03 Jan 2010, 17:47
- Location: Mac OS X
- Has thanked: 25 times
- Been thanked: 68 times
Re: SVN Bug Reports
by UnderFlow » 14 Mar 2011, 05:30
Revision 7586:
When Jedit Ojanen of Efrava attacks, two triggers are put onto the stack. They have different texts, but both put a Cat Warrior token on the battlefield.
Update:
I stole al Multani's Acolyte from the computer, for which it had alrady payed the Echo costs. I was not asked to pay the Echo costs during my next upkeep.
When Jedit Ojanen of Efrava attacks, two triggers are put onto the stack. They have different texts, but both put a Cat Warrior token on the battlefield.
Update:
I stole al Multani's Acolyte from the computer, for which it had alrady payed the Echo costs. I was not asked to pay the Echo costs during my next upkeep.
Re: SVN Bug Reports
by Sloth » 14 Mar 2011, 08:54
This is fixed now! Thanks UnderFlow.UnderFlow wrote:Revision 7586:
When Jedit Ojanen of Efrava attacks, two triggers are put onto the stack. They have different texts, but both put a Cat Warrior token on the battlefield.
-
Sloth - Programmer
- Posts: 3498
- Joined: 23 Jun 2009, 19:40
- Has thanked: 125 times
- Been thanked: 507 times
Re: SVN Bug Reports
by Corwin72 » 14 Mar 2011, 19:25
r7592
An easy one and a hard one.
Brimstone Mage should have a level cost of R3 not R2.
Venser, the Sojourner is in play and I just fired off his ultimate against the Gimli 2 deck.
I cast mind control on his Darksteel Gargoyle. I select Stuffy Doll to remove from the game. The Stuffy Doll has at least 1 Phylactery counter on it. There are 2 copies of Phylactery Lich in play.
Game Crash with a Null error.
An easy one and a hard one.
Brimstone Mage should have a level cost of R3 not R2.
Venser, the Sojourner is in play and I just fired off his ultimate against the Gimli 2 deck.
I cast mind control on his Darksteel Gargoyle. I select Stuffy Doll to remove from the game. The Stuffy Doll has at least 1 Phylactery counter on it. There are 2 copies of Phylactery Lich in play.
Game Crash with a Null error.
-
Corwin72 - Posts: 793
- Joined: 15 Sep 2009, 13:26
- Location: Grayson, Ga
- Has thanked: 25 times
- Been thanked: 9 times
Re: SVN Bug Reports
by Replika » 14 Mar 2011, 20:09
r7599
Proliferate is weird. The way it should work is this:
During proliferating, the proliferator chooses all permanents/players he wants to put counters on and chooses the type of those counters if a permanent has more that on type on it. Then, after that is done, all counters are put onto their permanents at the same time. State-based actions are not checked until now.
However, when I proliferated a second -1/-1 counter onto a Treasure Mage, the Mage died while I was still proliferating. I proceeded to put the 10th poison counter onto my opponent, then put two charge counters on my Everflowing Chalice and Sphere of the Suns. Then I finished proliferating, but the AI didn't die even though it had 10 poison counter. I won the game after passing through the combat phase and the second main phase however.
Proliferate is weird. The way it should work is this:
During proliferating, the proliferator chooses all permanents/players he wants to put counters on and chooses the type of those counters if a permanent has more that on type on it. Then, after that is done, all counters are put onto their permanents at the same time. State-based actions are not checked until now.
However, when I proliferated a second -1/-1 counter onto a Treasure Mage, the Mage died while I was still proliferating. I proceeded to put the 10th poison counter onto my opponent, then put two charge counters on my Everflowing Chalice and Sphere of the Suns. Then I finished proliferating, but the AI didn't die even though it had 10 poison counter. I won the game after passing through the combat phase and the second main phase however.
Shameless Doublepost
by Replika » 15 Mar 2011, 00:58
r7599
Arc Trail is incorrect. The card says "to another target" - it can't deal 3 damage to one target, the targets must be different. That also means that you can't deal 2 to your opponent's planeswalker and 1 to him.
Controlling a creature with protection from red and toughness 2 or less causes a NPE during each of the AIs second main phases if the AI has an Arc Trail in its hand. This also prevents the AI from successfully casting that Arc Trail.
Inferno Titan has a small text error: When entering the battlefield, the damage distribution says "2 more to deal, 1 more to deal, 3 more to deal".
Equipment that becomes a creature does not fall off, but it should.
Raid Bombardment needs "PlayMain1:TRUE", and so does Flame Slash.
Ichor Wellspring, Kithkin Zephyrnaut and Oran-Rief Recluse are fine when used by the AI. As far as I can tell, they do not need "RemAIDeck:True".
The AI always deals 3 to the player with Inferno Titan, even if it could kill two or three creatures.
The AI attacks with Steppe Lynx only to have it blocked, shrinked and then killed by my Phyrexian Crusader. I know that the AI checks whether its creature can block something (it can't), can be killed by a blocker (4/5 vs. 2/2) and then attacks. This is pretty bad against wither, infect and multiple blockers.
Arc Trail is incorrect. The card says "to another target" - it can't deal 3 damage to one target, the targets must be different. That also means that you can't deal 2 to your opponent's planeswalker and 1 to him.
Controlling a creature with protection from red and toughness 2 or less causes a NPE during each of the AIs second main phases if the AI has an Arc Trail in its hand. This also prevents the AI from successfully casting that Arc Trail.
Inferno Titan has a small text error: When entering the battlefield, the damage distribution says "2 more to deal, 1 more to deal, 3 more to deal".
Equipment that becomes a creature does not fall off, but it should.
Raid Bombardment needs "PlayMain1:TRUE", and so does Flame Slash.
Ichor Wellspring, Kithkin Zephyrnaut and Oran-Rief Recluse are fine when used by the AI. As far as I can tell, they do not need "RemAIDeck:True".
The AI always deals 3 to the player with Inferno Titan, even if it could kill two or three creatures.
The AI attacks with Steppe Lynx only to have it blocked, shrinked and then killed by my Phyrexian Crusader. I know that the AI checks whether its creature can block something (it can't), can be killed by a blocker (4/5 vs. 2/2) and then attacks. This is pretty bad against wither, infect and multiple blockers.
Re: SVN Bug Reports
by SoulStorm » 15 Mar 2011, 06:02
: Sword of Kaldra
Had the Sword of Kaldra attached to Ronin Cliffrider. When Ronin Cliffrider attacked, he dealt 1 damage to each creature the AI controlled correctly, but only one of the many creatures dealt damage was exiled.
Had the Sword of Kaldra attached to Ronin Cliffrider. When Ronin Cliffrider attacked, he dealt 1 damage to each creature the AI controlled correctly, but only one of the many creatures dealt damage was exiled.
Re: Shameless Doublepost
by Chris H. » 15 Mar 2011, 12:10
`Replika wrote:Raid Bombardment needs "PlayMain1:TRUE", and so does Flame Slash.
Ah, good catch, thank you. I added the SVar to these two cards.
`Ichor Wellspring, Kithkin Zephyrnaut and Oran-Rief Recluse are fine when used by the AI. As far as I can tell, they do not need "RemAIDeck:True".
Some cards work great in a constructed deck, Kithkin Zephyrnaut can be very powerful in a deck full of Kithkins and Soldiers. These types of cards are rarely useful in a randomly generated deck.
We probably could have used a more descriptive term for this SVar. This SVar prevents these cards from appearing in any randomly generated deck. The card might be buggy, the AI might not be able to use it or it could just be one of those cards that work best when used in combination with other specific cards.

-
Chris H. - Forge Moderator
- Posts: 6320
- Joined: 04 Nov 2008, 12:11
- Location: Mac OS X Yosemite
- Has thanked: 644 times
- Been thanked: 643 times
Re: SVN Bug Reports
by slapshot5 » 15 Mar 2011, 14:24
I've often thought we need to spilt this into multiple SVars. Or 1 SVar with different values. The values would override.
SVar:AI:RemoveRandom - remove it from random decks (like the three cards above)
SVar:AI:DontUse - this is buggy, and should be used by AI even in constructed (also removes from Random decks)
Something like that.
-slapshot5
SVar:AI:RemoveRandom - remove it from random decks (like the three cards above)
SVar:AI:DontUse - this is buggy, and should be used by AI even in constructed (also removes from Random decks)
Something like that.
-slapshot5
- slapshot5
- Programmer
- Posts: 1391
- Joined: 03 Jan 2010, 17:47
- Location: Mac OS X
- Has thanked: 25 times
- Been thanked: 68 times
Re: SVN Bug Reports
by Sloth » 15 Mar 2011, 14:32
I second that.slapshot5 wrote:I've often thought we need to spilt this into multiple SVars. Or 1 SVar with different values. The values would override.
SVar:AI:RemoveRandom - remove it from random decks (like the three cards above)
SVar:AI:DontUse - this is buggy, and should be used by AI even in constructed (also removes from Random decks)
Something like that.
-slapshot5

-
Sloth - Programmer
- Posts: 3498
- Joined: 23 Jun 2009, 19:40
- Has thanked: 125 times
- Been thanked: 507 times
Re: SVN Bug Reports
by Chris H. » 15 Mar 2011, 17:06
`slapshot5 wrote:I've often thought we need to spilt this into multiple SVars. Or 1 SVar with different values. The values would override.
SVar:AI:RemoveRandom - remove it from random decks (like the three cards above)
SVar:AI:DontUse - this is buggy, and should be used by AI even in constructed (also removes from Random decks)
Rob put together a simple message dialog explaining that the deck selected for the AI contains cards using this SVar. Splitting this into several SVars would allow us to provide a more meaningful message at this dialog.
-
Chris H. - Forge Moderator
- Posts: 6320
- Joined: 04 Nov 2008, 12:11
- Location: Mac OS X Yosemite
- Has thanked: 644 times
- Been thanked: 643 times
Re: SVN Bug Reports
by Rob Cashwalker » 15 Mar 2011, 17:25
If anyone wants to refine these Svars, go ahead, I'm knee-deep in the middle of revising draft, and I could use this kind of filtering too.
The Force will be with you, Always.
-
Rob Cashwalker - Programmer
- Posts: 2167
- Joined: 09 Sep 2008, 15:09
- Location: New York
- Has thanked: 5 times
- Been thanked: 40 times
Who is online
Users browsing this forum: No registered users and 7 guests