[confirmed]Camel, Desert Nomads vs AKH deserts

[Korath]

Describe the bug:
Both Camel and Desert Nomads only prevent damage from cards named Desert, not from cards with subtype desert.

Also, Camel only protects itself and attacking creatures banded with it, not blocking creatures banded with it. (Whether from Desert or subtype desert.)

Which card behaved improperly?
Camel and Desert Nomads.

Which update are you using? (date, name)Which type? (duel, gauntlet, sealed deck)
Dev 4d5f02b.

What exactly should be the correct behavior/interaction?
Any damage from a source with subtype desert should be prevented. If it were sources named Desert, these cards would be templated "from sources named Desert".

Are any other cards possibly affected by this bug?
No, and Desert Nomads is correctly unblockable if the defending player controls any lands with subtype desert. (It isn't actually landwalking, though, for the purposes of effects that copy or remove landwalk abilities or let creatures be blocked despite them. Neither are any other landwalk abilities except for the five basics. But that's a separate bug that's very difficult to deal with and that I think has already been reported anyway.)

I'll be the first to admit this is nitpicky, since there aren't any cards with subtype desert in release Manalink other than Desert itself, and none of the new ones in Amonket and only one of those spoiled so far in Hour of Devastation have abilities that can damage creatures. The others can all be made to do so with the help of other cards, though, with animation effects.

Theoretically, this is testable in release, by seeing if they prevent damage from a Desert that's lost its subtype. I don't think there's any way for that to happen except for effects like Jinx's, and those all incorrectly change cards' names and supertypes too.

Damage against blocking creatures banded with Camel actually does manifest in release, and can be tested by animating a Desert.

[Aswan jaguar]

I confirm that Camel not preventing damage when blocking to itself and others.

I confirm the Jinx bug.

I also confirm the subtype bug if my test is correct that is.I animated Desert with Genju of the Realm (doesn't that override only subtypes? it certainly keeps Desert name) and damage was prevented to Both Camel and Desert Nomads. wrong test.

[Korath]

Genju of the Realm shouldn't remove subtypes - the "It's still a land" language means, not only is it still a land, but it's also still all other types and subtypes it already was.

205.1b Some effects change an object's card type, supertype, or subtype but specify that the object retains a prior card type, supertype, or subtype. In such cases, all the object's prior card types, supertypes, and subtypes are retained. This rule applies to effects that use the phrase "in addition to its types" or that state that something is "still a [type, supertype, or subtype]." Some effects state that an object becomes an "artifact creature"; these effects also allow the object to retain all of its prior card types and subtypes.

Example: An ability reads, "All lands are 1/1 creatures that are still lands." The affected lands now have two card types: creature and land. If there were any lands that were also artifacts before the ability's effect applied to them, those lands would become "artifact land creatures," not just "creatures," or "land creatures." The effect allows them to retain both the card type "artifact" and the card type "land." In addition, each land affected by the ability retains any land types and supertypes it had before the ability took effect.

Example: An ability reads, "All artifacts are 1/1 artifact creatures." If a permanent is both an artifact and an enchantment, it will become an "artifact enchantment creature."

Even if it were correct to remove them, I'd be leery of just assuming that it did. It's not straightforward to test that independently, and adding and removing subtypes is awkward in the underlying engine. Most effects that do so are individually special-cased.

I think the most straightforward way to test damage prevention vs. subtypes in release would be to change Manalink.csv to remove subtype desert from Desert (5000h) and add it to some other handy land like Balduvian Trading Post or Rath's Edge. Regardless, it's crystal clear looking at the code - a test against SUBTYPE_DESERT is for the subtype, and one against CARD_ID_DESERT is for the name.

---

Now that I read Camel more closely, it shouldn't prevent damage while in a defensive band. Unless it's somehow banded to a blocking creature while it's itself attacking, which I'm pretty sure is illegal and which the engine couldn't support anyway. (The card's oracle text read "Banding. Prevent all damage that would be dealt to Camel or to creatures banded with Camel by Deserts." at one point, and at the time I pasted card text into the Magic.exe decompilation from Manalink.csv, the csv hadn't been fully updated to current oracle text in a very long time.)