It is currently 23 Apr 2024, 23:40
   
Text Size

[fixed mostly/closed]Manalink.csv/manastripe inconsistencies

Moderators: BAgate, drool66, Aswan jaguar, gmzombie, stassy, CCGHQ Admins

[fixed mostly/closed]Manalink.csv/manastripe inconsistencies

Postby Korath » 27 Jul 2013, 05:54

A handful of land cards have different mana source colors specified in Manalink.csv than what eventually shows up in their mana stripes. Granted, the only thing I can find that actually currently looks at the manalink.csv value (which ends up in card_ptr_t::mana_color) is sub_43F800, which seems to be part of the sealed deck AI. It's the only thing I have to look at to determine land color for lands in the deckbuilder or a graveyard or library, though, so it'd be nice to get the ones that are clearly bugs fixed. Plus, in a couple cases, it's the in-play or in-hand version that's wrong.

Behavior as of OtIv2 and recommendations:
LandIn printDeckbuilder/graveyard/libraryIn handIn playNotes
Shadowmoor/Eventide filter lands (Cascade Bluffs, Fetid Heath, Fire-Lit Thicket, Flooded Grove, Graven Cairns, Mystic Gate, Twilight Mire, Sunken Ruins, Wooded Bastion, Rugged Prairie)U/R, etc.correctcorrectnothingAlready fixed in next version due to interaction with Reflecting Pool, IIRC
Gemstone Cavernscolorlessgoldcolorlesscolorless or WUBGR, depending on counterA case could be made for leaving this alone for sealed deck generation, but it's probably closer to a Glimmerpost than a City of Brass. Recommend change manalink.csv value to colorless.
Lorwyn vivid lands (Vivid Crag, Vivid Creek, Vivid Grove, Vivid Marsh, Vivid Meadow)per primary colorgoldper primary colorWUBGR or primary color, depending on counterRecommend leaving unchanged.
Tendo Ice Bridgegoldgoldcolorlesscolorless or XWUBGR, depending on counterIn-print version contradicts vivid lands! Recommend leaving unchanged.
Simic Growth ChamberG/UB/GG/UG/Uclear bug in manalink.csv
Sungrass PrairieG/Wnothingcard is green/white, but no manastripesame as in handIn-play already fixed in next version, IIRC
Torment tainted lands (Tainted Field, Tainted Isle, Tainted Peak, Tainted Wood)W/B, etc.nothingnothingnothingseparate bug report filed
VesuvacolorlessgoldcolorlesscolorlessColorless manastripe in-play even after copying; separate bug report filed. Recommend manalink.csv gold->colorless.
Urborg, Tomb of YawgmothblacknothingblackblackRecommend manalink.csv colorless->black
Shimmering GrottogoldgoldcolorlesscolorlessTentatively recommend changing in-hand and in-play colorless->XWUBGR.
Mirrodin artifact lands (Ancient Den, Great Furnace, Seat of the Synod, Vault of Whispers)per colorper colorcolorlessper colorThese four show up as colorless mana when in hand, only. The fifth, Tree of Tales, correctly shows as green everywhere.
Cavern of SoulsgoldgoldcolorlesscolorlessHalf isn't even approximated; and the approximation on the other half is very inconsistent. Recommend disabling the card.
Invasion sac-lands (Ancient Spring, Geothermal Crevice, Irrigation Ditch, Sulfur Vent, Tinder Farm)goldgoldper primary colorper primary colorSimilar argument to Gemstone Cavern, though more tentatively, and despite their in-print color; i.e., recommend changing manalink.csv values to primary color only. These currently don't show up with Reflecting Pool, but changing their manastripe to all three colors would cripple the AI.
Phyrexian Towerpredates nonbasic land coloringblackcolorlesscolorlessThis doesn't get its black manastripe even when there's a creature to sac to it. Recommend changing in-hand/in-play colorless->XB, preferably only when its controller has a creature also.
Last edited by Aswan jaguar on 22 Feb 2014, 08:15, edited 2 times in total.
Reason: fixed/closed
User avatar
Korath
DEVELOPER
 
Posts: 3707
Joined: 02 Jun 2013, 05:57
Has thanked: 496 times
Been thanked: 1106 times


Re: Manalink.csv/manastripe inconsistencies

Postby Aswan jaguar » 27 Jul 2013, 17:46

Professional work to say the least. =D>
About AI that I want so badly to help him as much as possible,has a lot of problems caused by lands:
- AI not playing a single card if some lands are in AI hand,
- AI stalling playing a land for few turns if certain lands are in his hand or
- just miss the first land drop,
- and other weird behaviours (like AI doesn't miss land drops with many lands IF AI PLAYS SECOND/DRAWS FIRST).
- Strange but AI has land problems even with some lands made by the original Microprose developers.

I had made a lot of tests about lands-AI and wrote a survey(never finished though) maybe you want to take a look:

viewtopic.php?f=86&t=5726&start=15#p85404
---
Trying to squash some bugs and playtesting.
User avatar
Aswan jaguar
Super Tester Elite
 
Posts: 8078
Joined: 13 May 2010, 12:17
Has thanked: 730 times
Been thanked: 458 times

Re: Manalink.csv/manastripe inconsistencies

Postby Aswan jaguar » 28 Jul 2013, 07:29

It is just me,but for all land cards except BASIC in the deck builder they have all this "gold" more of a "Light Brown" manastripes.Am I missing something?
Also you got me confused at first as I thought you were talking about the mana stripes on the top right corner of the card, of what mana a land can produce.(but they are not visible in deck builder or library,graveyard,exile).
---
Trying to squash some bugs and playtesting.
User avatar
Aswan jaguar
Super Tester Elite
 
Posts: 8078
Joined: 13 May 2010, 12:17
Has thanked: 730 times
Been thanked: 458 times

Re: Manalink.csv/manastripe inconsistencies

Postby Korath » 28 Jul 2013, 08:17

You're not missing anything, no - right now, they're all drawn with CardBk_Land, and the only way the differences will manifest is in the sealed deck deckbuilder AI miscounting its mana, and by clicking the Stats button in the deckbuilder. It's frankly pretty amazing that there's only nine that are even arguably wrong there. (The in-hand and in-play columns do refer to the diagonal mana stripes that appear in these two zones, which will be identical to the mana in most cases.)

Here's one way to confirm with a standard drawcardlib, besides looking directly at the mana source color column of Manalink.csv:
Attachments
stats-sgc.jpg
User avatar
Korath
DEVELOPER
 
Posts: 3707
Joined: 02 Jun 2013, 05:57
Has thanked: 496 times
Been thanked: 1106 times

Re: Manalink.csv/manastripe inconsistencies

Postby HarlequinCasts » 29 Jul 2013, 10:12

Not sure if related but on the subject of artifact lands, I've been noticing they don't work correctly with Felwar Stone. Specifically Tree of Tales seems to be acting weird. I'll make another post, but wanted to mention since Felwar Stone is maybe a good test for mana stripe related things.

Posted here
User avatar
HarlequinCasts
 
Posts: 922
Joined: 07 May 2013, 14:33
Has thanked: 68 times
Been thanked: 30 times

Re: Manalink.csv/manastripe inconsistencies

Postby Sonic » 03 Aug 2013, 04:14

I've checked back to the 2000_no_more_beta file and some of these errors have existed since the original construction of the manalink3.xls file. Although, I'm partially to blame for not checking the Mana Source Colors field when completing the data for a card activation.

Simic Growth Chamber = 10 - Should be 12

Sungrass Prairie = 0
Tainted Field = 0
Tainted Isle = 0
Tainted Peak = 0
Tainted Wood = 0
Krosan Verge = 0

I'll go through the file and do a thorough check.
Working On: Life, the Universe, and Everything.
User avatar
Sonic
Apprentice
 
Posts: 827
Joined: 27 Feb 2010, 00:37
Has thanked: 3 times
Been thanked: 161 times

Re: Manalink.csv/manastripe inconsistencies

Postby Korath » 03 Aug 2013, 05:11

If you're going to tweak Krosan Verge, then all ten of the Onslaught/Zendikar fetchlands are set as colorless, too; all eleven cards the fetches and modern reprints of Krosan Verge were printed like two-color lands (the Judgment printing of Krosan Verge was colorless, despite two-color lands being printed as hybrid by then). Setting their color in manalink.csv will color them in libraries/graveyards/deckbuilder, but they'll be drawn colorless in play and in hand, since they have no manastripe.

I thought about an option to special-case them, but didn't want to deal properly with Magical Hack, was afraid to check if the fetches themselves dealt properly with Magical Hack, looked at the consistently colorless printings of Terramorphic Expanse, and decided not to bother.
User avatar
Korath
DEVELOPER
 
Posts: 3707
Joined: 02 Jun 2013, 05:57
Has thanked: 496 times
Been thanked: 1106 times


Return to Archived Reports

Who is online

Users browsing this forum: No registered users and 92 guests


Who is online

In total there are 92 users online :: 0 registered, 0 hidden and 92 guests (based on users active over the past 10 minutes)
Most users ever online was 4143 on 23 Jan 2024, 08:21

Users browsing this forum: No registered users and 92 guests

Login Form