WotC_BenFinkel

WotC_BenFinkel



31 May

Comment

Originally posted by fiskerton_fero

But it's digital-only design space yeah? Does it matter if the computer handles it for you? Assuming you're thinking to yourself "which one would this kill among this stack of tokens", it doesn't matter which 1/1 is killed. If they had any distinguishing characteristics like a counter or a perpetual text, they shouldn't be stacked with the rest.

Maybe an implementation of the arrow system could've been used? If you hover in your hand it will tell you what it will target.

What if:

  • you suspect your opponent has a Herald of Vengeance

  • you attacked them with two otherwise identical creatures, but only one dealt damage to them

  • you have some reason to distinguish them, e.g. the ability to put a shield counter on one of them

We do not have a good way of communicating to players "this is the one that would be destroyed by Herald of Vengeance". In fact, we don't even "break up the stack" at all - they'll be presented to you in a pile. #wotc_staff

Comment

Originally posted by fiskerton_fero

Very meh. I wonder if the name thing is just a side effect of how they had to word the effect (for coding reasons?) or an intentional decision.

I mean, obviously the design started out as "destroy each permanent that dealt damage to you last turn.". But the real obstacle was UX, not implementation.

If we went with that text, players would have to be able distinguish otherwise identical permanents that dealt damage to players the previous turn - they'd have to unstack, and you'd have to be able to somehow tell which stack was which.

And this would have to happen in EVERY game of Arena, lest you get improperly informed as to whether Herald of Vengeance is in the game! #wotc_staff


20 Apr

Comment

Originally posted by Equivalent_Ad_8413

Sad.

Imagine me on the shoreline waving goodbye to D&D as it sails away......

I hope Alchemy Horizons: Baldur's Gate will help scratch that itch! #wotc_staff


28 Mar

Comment

Thank YOU for the game! A nailbiter to be sure. I think it was your [[Showdown of the Skalds]] that sealed the deal for you there - so much value! #wotc_staff


16 Feb

Comment

Unfortunately, yeah, it's a bug (one which will with good luck be patched this week). The script that converts the mana cost to mana value did not adequately account for the "hybrid-phyrexian" mana symbol. #wotc_staff


11 Feb

Comment

Originally posted by mrbiggbrain

Autotap will now do its best to deal with abilities that discount the first time you do something, but not subsequent times. There will be times where Autotap THINKS you can afford something and gives you a stop, but you actually can't. We'd rather give you a stop that you don't need than not give a stop you do need. Autotap is doing its best.

Wait are you saying solving complex backpack problems in real time with multiple vectors is difficult?

Who woulda thunk it? #wotc_staff


26 Jan

Comment

Originally posted by FutureComplaint

Cool.

Any particular reason we have to wait so long for a bug fix?

Essentially, shipping releases is a pretty complicated process that involves a lot of coordination. We keep them pretty tightly to a schedule. The December holidays were certainly a big gap there. #wotc_staff

Comment

This issue should be fixed with the Kamigawa: Neon Dynasty release, by the way. #wotc_staff

Comment

Originally posted by thomar

Most problematic card for the software team? Most problematic card for the QA team?

Probably the most notorious card for us to implement was [[Explore the Vastlands]], the backside of [[Wandering Archaic]]. It literally took weeks to work out the kinks on this card - and to add insult to injury, it's such a strategically worse card than its other face that it feels like it was a waste of our time. Managing the control flow of the "each player" loop was very much like pulling teeth for our semantics processing...

For QA, I'd say something like [[Thief of Sanity]] and/or [[Underworld Breach]] would be the biggest problem child. Essentially every action permission mechanic (e.g. MDFCs, Mutate, Disturb, etc) needs to be vetted through "well what happens if you get to cast your opponent's cards, and what if they're face down?". And there have been many bugs with conferring an alternative cost with non-mana payments over the years; I think we're finally through the woods with them though. #wotc_staff

Comment

Originally posted by thomar

Hey, your team does a good job maintaining Arena's game logic. The infinite complexity of MTG card interactions is insane, it's impressive the game works as well as it does.

Any chance we could get a rundown of how the software and design tools work under the hood? Or anecdotes about some particularly squirrely bugs? It sounds like working on Arena would make for a really fun programming story.

Why thank you! While we still have never found a good opportunity to give an in-depth article about its workings (and it's one of those topics where's it's very hard to avoid going down endless digressions!), we did do a video with Gavin Verhey's Youtube show, Good Morning Magic, a few months ago that discusses this very topic, featuring yours truly! Here it is. Also I'm usually open to answering ad-hoc questions (at least ones that I have good answers to!). #wotc_staff

Comment

Looks like the predictive discount from Vine Gecko is acting up, possibly in combination with the Skydiver's minimum-1 kicker. That can definitely cause you to lose priority of it's your only meaningful action. I'll look into it tomorrow. #wotc_staff


01 Sep

Comment

This was probably introduced with AFR, it seems to be some interaction between "choosing your hybrid mana symbol" and choosing an X value. We'll investigate it soon. #wotc_staff


15 Aug

Comment

Originally posted by LiteralFan

https://www.reddit.com/r/MagicArena/comments/p01y49/bug_druid_class_tier_3_emergent_sequence

Is this a bug? Similar situation that involves [[Emergent Sequence]] and [[Druid Class]]. I didn't think it was but one of the comments had a good argument for it being a bug. Sorry to bug you but that post got buried pretty fast and it's been bugging me that there wasn't anything conclusive.

I'll double check with rules manager Jess Dunks on Monday, but I'm pretty sure the ability conferred by Druid Class is a characteristic-defining ability, and thus is applied in layer 7a, before Emergent Sequence is applied in 7b, meaning this is not a bug but certainly unintuitive. #wotc_staff

Comment

I assume you cast [[Emergent Sequence]]? The permanent is affected by a layered effect turning its power and toughness to 0/0 (which was created by Emergent Sequence). This means that at layer 7b, the permanent becomes a 0/0, far later than where it becomes a 2/4 in layer 1. #wotc_staff


13 Jul

Comment

Originally posted by freelutri

Before the first post-release patch, the final level, instead of giving a creature +1/+1 for each other attacking creature, gave it +1/+1 for each attacking creature, so the bonus was 1 greater than it was supposed to be

In a way, it was giving +1/+1 for each other attacking creature - each attacking creature other than Paladin Class! #wotc_staff


08 Jul

Comment

Originally posted by mickeybod

I assume Field Trip is facing the same bug? Won't pull snow forests

Anything that talks about a "forest card" is affected by the bug (and the fix). #wotc_staff


30 Jun

Comment

Originally posted by Filobel

Definitely a bug. I'm surprised by this though. They actually fixed that interaction a little while ago when [[set up the stage]] had the same bug. The fix was global, not just on set up the stage, because it had the unintended side effect of breaking [[Thallid Omnivore]] (don't ask). One possibility is that because Dominaria draft was back on Arena, they reverted the fix so that it didn't disrupt it too much (omnivore is a key card in that draft format). Another possibility is that the fix wasn't as general as they thought and doesn't work on prismari command for one reason or another.

As always, precise words matter. Expressive Iteration reads "the exiled card", Light Up the Stage reads "those cards". This means EI ends up using the machinery that linked abilities use, whereas LUtS uses recorded information from the resolving spell. It should be reasonable for our anaphora resolution to also use the recorded resolution state information for EI. #wotc_staff

Comment

Thanks for the report, we'll get on it. #wotc_staff


21 Jun

Comment

Lemme guess: the Auspicious Starrix I see in your opponent's graveyard had been mutated onto an Old-Growth Troll, and this Troll Warrior token came from sacrificing the Starrix-as-an-aura? MTG sure can be ridiculous sometimes. We'll look into getting the right token identifier for this bananas case... #wotc_staff


19 Jun

Comment

We actually do have a fix for the loop detector to identify this style of situation and cause a draw now. It should come with AFR. The bug was that we didn't identify that spell copies might be mandatory, so the loop detector thought that the copies were being proactively cast (and that therefore something like running out of time would cause this loop to end, as the autoresponder would rather pass than cast anything). #wotc_staff