10/18/2010 MM: "I Want To Threshold Your Hand (Or Possibly My Artifacts In Play)"

52 posts / 0 new
Last post
This thread is for discussion of this week's Making Magic, which goes live Monday morning on magicthegathering.com.
Apparently R&D's definition of a threshold mechanic is so broad as to be essentially meaningless.  I maintain that the definition should apply only to mechanics with exactly two states.  I'll buy that Kird Ape qualifies but all of the subsequent examples are things that shouldn't qualify.  My definition would be that a threshold card should have a state that is an on/off switch defined by the board state - Halcyon Glaze ends itself at a predetermined time, so it is something other than threshold.  Meanwhile Warlord, Maro and Priestess are all "counting" cards; they don't need to also be lumped in under threshold.  A narrow definition is best in this case.

As usual I think Maro is vastly overstating the difficulty of tracking multiple levels of a threshold, and it irritates me to know that they straight-jacketed the design as much as they did, especially given how boring most of the metalcraft cards ended up being.  Overall I have come to regard metalcraft as a bad mechanic; it pointlessly attaches a condition to things that you want to have happen, just so they can be costed more aggressively, reading to blowout games when you're successful in meeting the condition and miserable failures when you aren't.  Dull, random, and bad.

"There is no reason to open up future design space we won't use" - tsk tsk, MaRo.  Throwing away perfectly good design space forever, just to get the game to play a little better now?  The design team for Magic 2079 will hate you for this.

Getting three artifacts out is absurdly easy if you build your deck right.  As long as the metalcraft cards aren't things like Fog and Llanowar Elves that you must have in the early game, and as long as there isn't much of Shatterstorm going around, you should have no trouble at all meeting metalcraft four games out of five.  Which only further reinforces the stupid borkenness of cards like Carapace Forger and Auriok Sunchaser.  It's not that they might hit for 3-4 on turn 3, it's that they will virtually always be that large when  you play both of them at once on turn 4 and then bash for 7.

Scars of Mirrodin is linear?  Other than infect, I wouldn't have said so.  Proliferate is vastly open-ended, equipment lets one creature play many roles, artifacts by their nature can go in any deck of any color, and metalcraft can use any artifacts.  If it's the most linear set since Lorwyn (and I'm not sure that's true, Alara was pretty linear too with its shard themes), that's only because you've done a pretty good job of avoiding excess linearity in most of the recent sets.  Zendikar's Allies and Vampires and Landfall were fairly narrow, but it felt to me like there was a fair bit of room left; ROE as well had some decent flexibility.  So overall I think you just plain told an untruth here.  Infect aside, SOM seems pretty modular.

Overall, a rather disappointing column.

And finally, next week we get some answers about that stupid test.  Were its questions fuzzily-worded on purpose to weed out all but the designers that think exactly like MaRo, or was it just written poorly?  I'm eager to know.
My New Phyrexia Writing Credits My M12 Writing Credits
As far as the benefit of the rest of Magic is concerned, gold cards in Legends were executed perfectly. They got all the excitement a designer could hope out of a splashy new mechanic without using up any of the valuable design space. Truly amazing. --Aaron Forsythe's Random Card Comment on Kei Takahashi


As usual I think Maro is vastly overstating the difficulty of tracking multiple levels of a threshold, and it irritates me to know that they straight-jacketed the design as much as they did, especially given how boring most of the metalcraft cards ended up being.  Overall I have come to regard metalcraft as a bad mechanic; it pointlessly attaches a condition to things that you want to have happen, just so they can be costed more aggressively, reading to blowout games when you're successful in meeting the condition and miserable failures when you aren't.  Dull, random, and bad.

"There is no reason to open up future design space we won't use" - tsk tsk, MaRo.  Throwing away perfectly good design space forever, just to get the game to play a little better now?  The design team for Magic 2079 will hate you for this.




Maybe he is maybe he isn't, but I'd rather it be a little too simple than be too complex and have all my game's go to time if the board state gets complex and each player needs to keep recounting artifacts to make sure the various metalcrafters are active.

And what gain is there? terribly little, the difference between four and three artifacts isn't terribly much, and once you start going to five and six you start making cards that are just terrible in 90% of the games they appear in as sort of shattering will make sure they never activate fast enough to be above curve.

Besides if Magic 2079 really cares that much they can always go back and change it. Nothing is ever set in stone with magic. Keep things simple today and if tomorrow needs more variations they just rewrite the rule.





Scars of Mirrodin is linear?  Other than infect, I wouldn't have said so.  Proliferate is vastly open-ended, equipment lets one creature play many roles, artifacts by their nature can go in any deck of any color, and metalcraft can use any artifacts.  If it's the most linear set since Lorwyn (and I'm not sure that's true, Alara was pretty linear too with its shard themes), that's only because you've done a pretty good job of avoiding excess linearity in most of the recent sets.  Zendikar's Allies and Vampires and Landfall were fairly narrow, but it felt to me like there was a fair bit of room left; ROE as well had some decent flexibility.  So overall I think you just plain told an untruth here.  Infect aside, SOM seems pretty modular.




Not linear? Proliferate is linear! It says "play me with cards that have counters!". Just like how Goblin chieftain wants gobbos. If you don't have counter based cards than its pointless. Sure you don't need every card in your deck to be counter based but its proliferate definately encourages you to play with certain cards. And thats what makes it linear not necessarily that you need to play proliferate with other cards that have proliferate but simply that it encourages you to build a deck in a certain way. Metalcraft DEMANDS you play with artifacts, again like the Goblin chieftain. If a cards asks you to build a deck a certain way that its a linear mechanic.  So yeah this block is very very linear. A truly modular mechanic is rebound from ROE. It makes no demands of the rest of the deck. Or clash... from Lorwyn.


..."jsCall();" id="jsProxy" type="hidden" />
I thought this was a great column. "Surely," I thought, "this will demonstrate to the naysayers that they really did think hard about these types of decisions and the decisions they made were in fact correct for the overall health of the game." Apparently not.

I did think it got a little silly, how wildly threshold could vary. But you need a name for the mechanics that are similar to Threshold but aren't, and treating the whole thing as a sliding spectrum makes more sense than excluding things with only two states. (Although I can't think of any real examples of a card with more than two states)

I don't know that I'd consider Mirrodin all that linear compared to other recent sets. It's a bit more linear than Zendikar (artifacts = not as common as lands), but it's significantly less linear than Lorwyn, which was only a few years ago. 

For one thing, MaRo and the gang are overthingking it.


If (condition)
  result


It's that simple.

And what gain is there? terribly little, the difference between four and three artifacts isn't terribly much, and once you start going to five and six you start making cards that are just terrible in 90% of the games they appear in as sort of shattering will make sure they never activate fast enough to be above curve.



Quick example of a card that they aren't making because of this rule:

Floating Fortress
Artifact Creature - Wall, 3, 2/8
Flying, defender.
Metalcraft 6 - As long as you control six or more artifacts, Floating Fortress gets +8/+2 and loses defender.

10 power is a very important number since it ends the game in two hits (or one if you give it infect).  Therefore, simply powering this card down until it's balanced with Metalcraft 3 (which would utterly break it) makes it dramatically less viable than in its current state.  This card offers a strong possibility of an instant win when you get five other artifacts out, making it possible for you to create a lot of tension by sitting with four out and leaving your opponent to wonder, do you have two Scale of Chiss-Goria in your hand?  Does he need to Revoke Existence now, when it might leave him without a kill spell for one of your later guys?  Obviously this example is a little out-there, but I'm sure that you'd find it possible to do a lot of design work at Metalcraft 6 that just plain doesn't work at 3.

Besides if Magic 2079 really cares that much they can always go back and change it. Nothing is ever set in stone with magic. Keep things simple today and if tomorrow needs more variations they just rewrite the rule.



Resulting in old cards that don't do what they say and are hard to use around new players because of that.  This should be avoided whenever possible.  Making it "Metalcraft 3" today, even if they never printed any other metalcraft numbers ever again, would have left the door open just in case, and at absolutely no cost to the present.  The number does not make it any harder to figure out, because it's just repeating what's in the text.  And replacing "three" with two instances of "3" (so they matched, and making comprehension for non-English natives easier as a side benefit) would not take any more space on the card.

It was ALL-upside.  There is no excuse for their not doing it.  None.

Not linear? Proliferate is linear! It says "play me with cards that have counters!".



That's "modular".  There are lots of cards that use counters in lots of different ways, therefore proliferate has lots of potential uses.  If Proliferate isn't modular, nothing is.

A truly modular mechanic is rebound from ROE. It makes no demands of the rest of the deck. Or clash... from Lorwyn.



Clash is linear because it demands that you engineer your deck to be as commonly able to win the clash as possible.  You need lots of high-CMC cards to win clashes frequently; you need library manipulation to win them regularly.  Not to mention the cards that triggered off clashing; those were insanely linear, on top of the linearity of needing to be able to rig the clash if it was to be winnable.

..."jsCall();" id="jsProxy" type="hidden" />



WTH?
My New Phyrexia Writing Credits My M12 Writing Credits
As far as the benefit of the rest of Magic is concerned, gold cards in Legends were executed perfectly. They got all the excitement a designer could hope out of a splashy new mechanic without using up any of the valuable design space. Truly amazing. --Aaron Forsythe's Random Card Comment on Kei Takahashi
Quick example of a card that they aren't making because of this rule:

Floating Fortress
Artifact Creature - Wall, 3, 2/8
Flying, defender.
Metalcraft 6 - As long as you control six or more artifacts, Floating Fortress gets +8/+2 and loses defender.

Watch this:

Floating Fortress
Artifact Creature - Wall, 3, 2/8
Flying, defender.
As long as you control six or more artifacts, Floating Fortress gets +8/+2 and loses defender.


..."jsCall();" id="jsProxy" type="hidden" />

WTH?

It's just a glitch in the WYSIWYG post editor.  Nothing to worry about, unless it becomes far more common.

Quick example of a card that they aren't making because of this rule:

Floating Fortress
Artifact Creature - Wall, 3, 2/8
Flying, defender.
Metalcraft 6 - As long as you control six or more artifacts, Floating Fortress gets +8/+2 and loses defender.

Watch this:

Floating Fortress
Artifact Creature - Wall, 3, 2/8
Flying, defender.
As long as you control six or more artifacts, Floating Fortress gets +8/+2 and loses defender.



So, if they printed this alone (let alone others), why should players have to talk about "all the Metalcraft cards plus Flying Fortress" when they could just talk about "all the Metalcraft cards"?  And if they wanted to exclude Flying Fortress, they could just say "all the Metalcraft 3 cards".
My New Phyrexia Writing Credits My M12 Writing Credits
As far as the benefit of the rest of Magic is concerned, gold cards in Legends were executed perfectly. They got all the excitement a designer could hope out of a splashy new mechanic without using up any of the valuable design space. Truly amazing. --Aaron Forsythe's Random Card Comment on Kei Takahashi
Because there's plenty of ways to preserve the coolness of the idea without creating a multitude of cards that make metalcraft confusing:

Floating Fortress
-
Artifact Creature - Wall
Flying, Defender
Metalcraft - Floating Fortress gets +8/+2 if you control three or more artifacts
Sacrifice three artifacts: Floating Fortress loses defender until end of turn. 
2/8

I realize this isn't exactly the same card, but it accomplishes a lot of the same cool stuff, plus combos with Molder Beast and others.
R&D never lets brain freeze-inducing sets out the door. (I'm sure there are those that would argue on this point.)



Really, what about Scourge? Laughing
The end is always nigh.
I'm surprised that this is the distinction that R&D makes. It seems to me to be a meaningless categorisation.


My distinction would be that there are two types of mechanics here:

"threshold" mechanics, which have a specific target condition, at which an ability "turns on". (Like metalcraft and Kird Ape.)

And

"Compounding" mechanics, which just get better and better, without there being any target condition, or any "steps". (Like Maro, and Priest of Titania.) I wouldn't regard the smallest unit of something to be a threshold.

\:>
I think there's some value in having a name for the broader category that both of those fit in, but I agree that Threshold makes more sense as the name for the first one. That said, I can't think of a good name for the super-category that includes Threshold and Compounding.
I think metal craft is stupid personally, it limits you to mostly artifacts. One of the disigners tweeted that there will be a mass artifact destroyer in the later sets of the mirrodin block. Infect could be good if there was an infect planeswalker, as currently any t2 archetype requires $50 planeswalkers that have no answer besides hexmage.
I also agree that extending the term "threshold" to include counting cards would seem to make the categorization pretty meaningless. Halcyon Glaze I can sort of see, but not counting cards. Is Beacon of Immortality a scaling threshold card? Storm Herd? Biomantic Mastery?

I'd personally limit my use of the term to only those cards with a set, finite number of distinct states, so landfall instants and sorceries, yes, but landfall permanents, no. (Save maybe the Quests.) (BTW, Raemon, Homarid and Tidal Influence from Fallen Empires could be an example of three-state threshold cards, assuming "time" is considered an outside influence.)

So, if they printed this alone (let alone others), why should players have to talk about "all the Metalcraft cards plus Flying Fortress" when they could just talk about "all the Metalcraft cards"?  And if they wanted to exclude Flying Fortress, they could just say "all the Metalcraft 3 cards".

Then watch this:

Floating Fortress -
Artifact Creature - Wall
Flying, defender
Metalcraft - As long as you control six or more artifacts, Floating Fortress gets +8/+2 and loses defender.
2/8

Metalcraft is an ability word, not a keyword. There is absolutely nothing preventing them from doing metalcraft again in some later set but using a different number that time. They probably won't, but they could if they wanted to.

Come join me at No Goblins Allowed


Because frankly, being here depresses me these days.

(Although I can't think of any real examples of a card with more than two states)



Mul Daya Channelers =)

That's "modular".  There are lots of cards that use counters in lots of different ways, therefore proliferate has lots of potential uses.  If Proliferate isn't modular, nothing is.



It's a hybrid ^^ It's linear in its requirements, in that you are encouraged to play as many counter cards alongside it as you can. It's modular in the way explained above.
My distinction would be that there are two types of mechanics here:

"threshold" mechanics, which have a specific target condition, at which an ability "turns on". (Like metalcraft and Kird Ape.)

And

"Compounding" mechanics, which just get better and better, without there being any target condition, or any "steps". (Like Maro, and Priest of Titania.) I wouldn't regard the smallest unit of something to be a threshold.



Yeah, that's the fundamentals.

I think Devin Low's article explains things better than MaRo; why R&D feels the need to delve a little bit deeper into this and differentiate between different kinds of tresholds and scaling effects. 

It's not only about how and when the mechanism triggers, but also what the effect is. Cenn's Heir and Spellstutter Sprite are both "compounding" (or "scaling", as R&D calls it) cards, but Cenn's Heir pretty much just gets better and better the more kithkin you play, whereas Spellstutter Sprite pretty quickly caps out. Therefore, you're much happier about playing Spellstutter Sprite if you haven't got enough fairies to build your whole deck around it (or don't want to!), and that is very important, both for limited and for keeping a set from being too linear and parasitic.

One could argue that Spellstutter Sprite was a threshold card, though: "Either you have enough fairies to counter target spell, or you don't - it's a treshold mechanism!". I believe that MaRo is trying to make us think about the possibilities of such cards; those that exists somewhere in-between Scaling Count-me's like Goblin King on one end, and Treshhold one-in-deck-cards that are wide-open in their modularity but lacking in focus.

----

Me, I love Metalcraft; and I really dig threshold mechanisms like that, because they keep your decks focused without being too parasitic - finding the optimal balance between artifacts and cards with Metalcraft is a lot of fun; and these decks are a blast to play. In GDS2; I'm most excited about those sets that have simple and clever treshold mechanisms.  

Willpell thought that now that metalcraft is counting three artifacts, they can't make a card that sets the treshhold at six; but that is wrong. After all; Scute Mob counts five lands, Dragonmaster Outcast counts six lands, Avatar of Fury counts seven lands, and Budoka Gardener counts ten. No problem. It just gets a lot simpler and better if they keep all midrange numbers for counting artifacts to three. If they ever wanted to make a big, crazy effect on one individual card for having lots of artifacts, that would be fine. 

A truly modular mechanic is rebound from ROE. It makes no demands of the rest of the deck. Or clash... from Lorwyn.



Clash is linear because it demands that you engineer your deck to be as commonly able to win the clash as possible.  You need lots of high-CMC cards to win clashes frequently; you need library manipulation to win them regularly.  Not to mention the cards that triggered off clashing; those were insanely linear, on top of the linearity of needing to be able to rig the clash if it was to be winnable.



In order to make the most use of clash, yes you need to engineer the deck to be able to manipulate the library.  However, clash cards can be used in any deck since it doesn't have specific requirements (such as playing artifacts) and you can occasionally win the clash in a low CMC deck.  This is similar to Threshold in that to make the most of it you need to be able to quickly fill your graveyard, however you can use threshold cards in most decks since you can get seven cards in your graveyard.  This makes it modular.
IMAGE(http://pwp.wizards.com/1205820039/Scorecards/Landscape.png)
I just came here to post that I also find MaRo's definition of threshold pretty unintuitive and unnecessarily broad.  To me, a "threshold" mechanic should be one where a card functions differently dependent upon the current amount of some countable element of the game state.  

Primarily, this should be a state A vs. state B thing.  However, the scaling mechanics like Keldon Warlord and Maro seem similar philosophically, so I could see how design insights from the two state threshold card could carry over to the scaling ones and thus that it might be useful to group them under a larger umbrella.

I think the way MaRo drops Halcyon Glaze and the landfall mechanic on top of the two state and multi-state threshold cards just muddles things though.  These cards are based off of game actions rather than game states.  They require the player to play in a certain way in order constant trigger them rather than to sculpt the game state.  To me that seems like a big distinction because it not only influences how players interact with the card (you have to remember whether the card was triggered each turn rather than being able to just look over the board at any point for the threshold condition) but also how the player thinks about building a deck around a card (rather than "make as much of X as possible" where X is artifacts or creatures or cards in hand, the incentive is do "do X each turn"). 

I liked MaRo's explanation of the sweet spot for threshold mechanics though.  I found it pretty convincing.  If you think metalcraft should be "metalcraft X," I'd recommend rereading the article.  Getting three artifacts out is not something that typically happens outside of a dedicated artifact deck so it is a reasonable goal.  Once you make a player set out to meet that goal though (by filling his/her deck with artifacts), are there really novel cards you could make that couldn't just have their A and B states scaled to work with metalcraft 3?

Re: Clash: As executed Clash was modular with a very minor linear subtheme.  Clash easily could have been linear with decks built around cards like Sylvan Echoes.  However, the only repeatable clash card was Sentry Oak which was really underpowered.  With dedicated library manipulation, you could also use the repeatable clashers like Recross the Paths, but even with library manipulation it's difficult to get the odds of winning the clash above 50%.  It also didn't help that the Clash card effects drew from the broad range of typical abilities given to each color (draw a card, do one damage, pump a creature, etc) and so had little synergy with each other, making a "linear" clash deck still somewhat scattered and unfocused.  So in practice Clash was a modular mechanic with cards like Broken Ambitions and Lash Out being played because they were fine cards even when you lost the clash.
I wonder: would the current R&D team ever see a card like Multani, Maro-Sorcerer see print?
Metalcraft is an ability word, not a keyword. There is absolutely nothing preventing them from doing metalcraft again in some later set but using a different number that time. They probably won't, but they could if they wanted to.

I stopped being annoyed by ability words right after I finally realized the cards would be exactly the same when the ability word is omitted.
They provide a mechanism for block-specific mechanics, in a reprint-safe implementation.   That's pretty cool.
Poison = Linear. First thing that came to mind when I saw the word.
As for how linear Scars is, you all seem to be missing the huge, overarching theme of the set:  Mirrans vs Phyrexians.  Each side has cards that get better when you play them with more cards of the same side - hence the watermarks.  I'm assuming this is a big part of what MaRo was referring to in his article.
I know one questions that's going to be on the GDS3 test. :D

Man they really dig deep into the definitions of threshold here huh. Some of them seem like they'd be better served having thier own names.

But as a whole, a good read.

As for how linear Scars is, you all seem to be missing the huge, overarching theme of the set:  Mirrans vs Phyrexians.  Each side has cards that get better when you play them with more cards of the same side - hence the watermarks.  I'm assuming this is a big part of what MaRo was referring to in his article.



Scars is linear because only 1 mechanic (imprint) is modular. Nothing to suprising here.


… and then, the squirrels came.

I think it's important to point out here that "Linear" is not a binary quality.  It's not "yes/no".  A set or mechanic can be more or less linear in comparison to other things.


For example, every Magic card ever made is Linear because they are meant to be played with other Magic cards.  The won't interface well with Pokemon or Pinochle cards.  But that's a very low linearity because of all the options.


So you could correctly say that Infect is highly linear because there are few infect/poison cards in existence.  Proliferate is less so due to the wider array of sources, but still sets linear requirements.  Metalcraft still less so, because there are many many artifacts.  And yet it does push your deck toward more.


So think of a score meter from 0 to 100


 



 


Modular                                     Linear


<---------------------------------------------->


0                                                 100


 


 


And everything is in there somewhere.  But it's not just a matter of 0 or 100.

If you're on MTGO check out the Free Events via PDCMagic and Gatherling.

Other games you should try:
DC Universe Online - action-based MMO.  Free to play.  Surprisingly well-designed combat and classes.

Planetside 2 - Free to play MMO-meets-FPS and the first shooter I've liked in ages.
Simunomics - Free-to-play economy simulation game.

Good article.  I'm happy to see something I noticed in Magic design be explicitly stated.  (That such mechanics work better with a "sweet spot" number, rather than having a variety leading to brain freeze.)

(No, it's not super hard to guess, but a lot of designers tend to overcomplicate their mechanics.)
Magic Judge Quis custodiet ipsos custodes? Rules Theory and Templating: "They may be crazy, but they're good." --Matt Tabak, Rules Manager*

I think it's important to point out here that "Linear" is not a binary quality.  It's not "yes/no".  A set or mechanic can be more or less linear in comparison to other things.


For example, every Magic card ever made is Linear because they are meant to be played with other Magic cards.  The won't interface well with Pokemon or Pinochle cards.  But that's a very low linearity because of all the options.


So you could correctly say that Infect is highly linear because there are few infect/poison cards in existence.  Proliferate is less so due to the wider array of sources, but still sets linear requirements.  Metalcraft still less so, because there are many many artifacts.  And yet it does push your deck toward more.


So think of a score meter from 0 to 100


 



 


Modular                                     Linear


<---------------------------------------------->


0                                                 100


 


 


And everything is in there somewhere.  But it's not just a matter of 0 or 100.




What's also interesting to note is whether the mechanic is affected by other cards or affects other cards. metalcraft requires artifacts to do something itself. Proliferate requires cards with counters to let them do something. This inherently makes proliferate more opne-ended than metalcraft, because there are 22 cards with metalcraft but hundreds of cards with counters. Smile
I wonder: would the current R&D team ever see a card like Multani, Maro-Sorcerer see print?



Sure.  What about Lord of Extinction?
It also depends on the set it's in.

Modular in the Mirrodin block was less linear than Modular in Zendikar.
… and then, the squirrels came.
From the article:


...this [brain freeze-creating] line varies from player to player.



*Literally* the next sentence:


Most players never have to think about concepts like brain freeze because R&D never lets brain freeze-inducing sets out the door.



That strikes me as Orwellian.  :-)  In the first sentence, the line is fuzzy.  In the second, it's distinct, and R&D (Guardians of Justice And Perfection(tm)(R)) never let anything across that line.

Me, I don't think "Brain Freeze" is such a bad thing.  (Although "Brain Freeze" the card was a pretty horrifying mistake...)  Yes, gamestates can get complicated.  That's ok, as long as it's not a constant thing.  Complex gamestates are GOOD.  They create complex decision trees, which reward strong play, and punish poor play.  That's why games have decision trees to begin with.

At a minimum, I think R&D has pressed too far back against the "Must not let little Timmy get confused!" side of the building.  That's why 8/8s with Annihilator get reminder text that tells you to attack with them, and that's stupid.
The example of the wall that becomes a 10/10 can work with regular metalcraft simply by increasing the mana cost. Then the opponent still has to be just as worried about keeping you off metalcraft, but the card is not broken.

Proliferate is a linear mechanic. "If I have some of this, do I want more of this?" If it's a modular mechanic like cycling, the answer is "not necessarily". But you wouldn't use proliferate in a deck that only has one or two cards that put counters on things, any more than you would run Relentless Rats as a four of in a deck. You would use it when you put lots of counters on things. And then, since you have lots of counter-putting things, you want more proliferate. Two different buyback cards go in completely separate decks. But you wouldn't say that Contagion Clasp and Contagion Engine go in completely separate decks.
Apparently R&D's definition of a threshold mechanic is so broad as to be essentially meaningless.  I maintain that the definition should apply only to mechanics with exactly two states.  I'll buy that Kird Ape qualifies but all of the subsequent examples are things that shouldn't qualify.


They are all, at their heart, the same thing though. They rely on the same principle - he's talking about something which is a very basic mechanic. Kicker is another example of something which underlies a very large number of other mechanics.
My definition would be that a threshold card should have a state that is an on/off switch defined by the board state - Halcyon Glaze ends itself at a predetermined time, so it is something other than threshold.  Meanwhile Warlord, Maro and Priestess are all "counting" cards; they don't need to also be lumped in under threshold.  A narrow definition is best in this case.


They are threshold cards, they just have multiple states, as he points out. Its worth noting the level up mechanic is itself essentially a threshold mechanic with multiple possible states.
As usual I think Maro is vastly overstating the difficulty of tracking multiple levels of a threshold, and it irritates me to know that they straight-jacketed the design as much as they did, especially given how boring most of the metalcraft cards ended up being.  Overall I have come to regard metalcraft as a bad mechanic; it pointlessly attaches a condition to things that you want to have happen, just so they can be costed more aggressively, reading to blowout games when you're successful in meeting the condition and miserable failures when you aren't.  Dull, random, and bad.


Its actually very interesting, because there's a battle over activating it. That's why three is the "right" number - you have to include quite a lot of artifacts in your deck to do it, which puts constraints on what else you can have in your deck, and the fact that your opponent can easily turn it off (especially if you use artifact creatures) is a big deal. To run a deck like that, you want to ensure you run ehough artifacts (or whatever other threshold) to maintain your threshold condition, while your opponent battles you over that condition. It adds another interesting dimension to the game.
"There is no reason to open up future design space we won't use" - tsk tsk, MaRo.  Throwing away perfectly good design space forever, just to get the game to play a little better now?  The design team for Magic 2079 will hate you for this.


You're missing the point here. The point is that the design space isn't there in the first place. Metalcraft 2 doesn't give you enough benefit, while metalcraft 4 is too difficult to hit.
Getting three artifacts out is absurdly easy if you build your deck right.  As long as the metalcraft cards aren't things like Fog and Llanowar Elves that you must have in the early game, and as long as there isn't much of Shatterstorm going around, you should have no trouble at all meeting metalcraft four games out of five.  Which only further reinforces the stupid borkenness of cards like Carapace Forger and Auriok Sunchaser.  It's not that they might hit for 3-4 on turn 3, it's that they will virtually always be that large when  you play both of them at once on turn 4 and then bash for 7.


Not at all. Its easy to get three artifacts, but its far more difficult to get three worthwhile artifacts in play. Mox Opal is easy, but what are the other two? A piece of equipment is an obvious one. So what's the last one? Memnite dies to bolt, so oops, you lose your metalcraft, while another piece of equipment or a similar non-threat card suffers the penalty of making your deck too threat-light - I don't care if your carapace forger is a 4/4, frankly, because if I can trade my two removal cards for your two creatures (and I can) then oops, you lose because you have no threats.
That's the problem with metalcraft - you are either forced to play bad cards, play cards awkwardly (Mass for 0 or 1 is not very awesome), or wait. Sure, sometimes you get a god draw like Glint Hawkx2, memnitex2, mox opal, carapace forger, plains or forest, but a lot of the time you don't, and the mox opal looks quite terrible when you only have it and one other artifact in your hand.
Its very easy to get out three artifacts, but it doesn't make for a good deck necessarily.
Scars of Mirrodin is linear?  Other than infect, I wouldn't have said so.  Proliferate is vastly open-ended, equipment lets one creature play many roles, artifacts by their nature can go in any deck of any color, and metalcraft can use any artifacts.  If it's the most linear set since Lorwyn (and I'm not sure that's true, Alara was pretty linear too with its shard themes), that's only because you've done a pretty good job of avoiding excess linearity in most of the recent sets.  Zendikar's Allies and Vampires and Landfall were fairly narrow, but it felt to me like there was a fair bit of room left; ROE as well had some decent flexibility.  So overall I think you just plain told an untruth here.  Infect aside, SOM seems pretty modular.


A lot of the stuff in the block is actually fairly linear; anything which cares about artifacts tends to be that way. While there are a ton of artifacts, how many good ones are there?
Quick example of a card that they aren't making because of this rule:
Floating FortressArtifact Creature - Wall, 3, 2/8Flying, defender.Metalcraft 6 - As long as you control six or more artifacts, Floating Fortress gets +8/+2 and loses defender.


This card wouldn't cost 3 mana to start out with (2/8 flying wall is going to cost more than that, especially given it is colorless and ignores intimidate as well), and is it really so cool when it is costed properly? No.

Not that its very cool this way either. Darksteel Juggernaut is a better execution.
R&D never lets brain freeze-inducing sets out the door. (I'm sure there are those that would argue on this point.)


Yeah, like... Yourself, Mark. Remember how you talked about Time Spiral block confusing the heck out of new players?

Wasn't that long ago.
Time Spiral problem isn't the same thing as brain freeze. Brain freeze occurs while playing a game, even when you know how everything works because there are just too many things to think about at once. I have seen this in complicated board states in chess and in Starcraft even though I played those game all the time. In TSP, once you know what the mechanics do, it's not that hard to keep track of the game state. The TSP problem was that the game was hard to learn how everything works in the first place because there were a lot of mechanics. After some play experience this problem went away completely.

Also I notice you're discussing the metalcraft question in terms of constructed, which is not entirely bad, but I think that discussing Limited would be more relevant (Since I think it's what willpell was referring to. Also, when MBS comes out, constructed metalcraft decks will be much easier to make because there'll be more good artifacts. And there is other constructed than standard).

A player will almost never get metalcraft 6 in limited, so if a card required it, players would just ignore that card rather than try to build around it. It would only be for use on rares, and in that case you can use scaling cards instead like Darksteel Juggernaut. Even metalcraft 5 is hard to get, and it would pretty much come down to the luck of the draw and we don't want that kind of swingyness around here. Basically changing metalcraft to a number would only let you choose between metalcraft 3 and metalcraft 4. Save the designers of Magic 2079 from having to reinvent the wheel and just stick with the number we all know works.
I think there's some value in having a name for the broader category that both of those fit in, but I agree that Threshold makes more sense as the name for the first one. That said, I can't think of a good name for the super-category that includes Threshold and Compounding.



Externally-Dependant Variance cards?
My New Phyrexia Writing Credits My M12 Writing Credits
As far as the benefit of the rest of Magic is concerned, gold cards in Legends were executed perfectly. They got all the excitement a designer could hope out of a splashy new mechanic without using up any of the valuable design space. Truly amazing. --Aaron Forsythe's Random Card Comment on Kei Takahashi
You /never/ let brain-freeze out? It happens. You can try to avoid it, but, well, Magic is a game that /will/ eventually grind to a halt in frozen brains, somewhere. It's just something that shouldn't happen *often*... but I mean, get a dozen or two dozen cards out on each side and people /will/ lock up, the game is just too big to hold in memory at that point.
Time Spiral problem isn't the same thing as brain freeze. Brain freeze occurs while playing a game, even when you know how everything works because there are just too many things to think about at once. I have seen this in complicated board states in chess and in Starcraft even though I played those game all the time. In TSP, once you know what the mechanics do, it's not that hard to keep track of the game state. The TSP problem was that the game was hard to learn how everything works in the first place because there were a lot of mechanics. After some play experience this problem went away completely.

Also I notice you're discussing the metalcraft question in terms of constructed, which is not entirely bad, but I think that discussing Limited would be more relevant (Since I think it's what willpell was referring to. Also, when MBS comes out, constructed metalcraft decks will be much easier to make because there'll be more good artifacts. And there is other constructed than standard).

A player will almost never get metalcraft 6 in limited, so if a card required it, players would just ignore that card rather than try to build around it. It would only be for use on rares, and in that case you can use scaling cards instead like Darksteel Juggernaut. Even metalcraft 5 is hard to get, and it would pretty much come down to the luck of the draw and we don't want that kind of swingyness around here. Basically changing metalcraft to a number would only let you choose between metalcraft 3 and metalcraft 4. Save the designers of Magic 2079 from having to reinvent the wheel and just stick with the number we all know works.


Well, the threshold is different for different people, and some people are immune entirely, as their brains don't function that way, or they can simply see past the complexity and simplify it. This is actually a well-known process.

Magic can definitely induce the state if you think about what they COULD draw and what COULD win you the game. Even relatively simple game states can put you in that situation, especially in limited where realistically they COULD pull any random card of their colors out of their deck. 

At a minimum, I think R&D has pressed too far back against the "Must not let little Timmy get confused!" side of the building.  That's why 8/8s with Annihilator get reminder text that tells you to attack with them, and that's stupid.



That isn't reminder text on Ulamog's Crusher.  You are forced to attack with the creature every one of your turns whether you want to or not.  Would you want to attack with Ulamog's Crusher into a Progenitus or Krosan Cloudscraper ?  Sure you make the opponent sacrifice two permanents but then your creature dies.

The "attacks each turn if able" is to add a certain flavor to a card.  It makes the creature into an all offence, no defense dumb brute trying to use force to get at you.  The creature isn't going to hold back just because you have a bigger creature than itself.  That doesn't alleviate confusion, it may actually create a little confusion because Timmy wouldn't want to use an 8/8 to attack into a 10/10 or 13/13.
IMAGE(http://pwp.wizards.com/1205820039/Scorecards/Landscape.png)

The "attacks each turn" was put in specifically to tell people they needed to attack with it. Not for flavor reasons.


www.wizards.com/Magic/Magazine/Article.a...

It's important to get a sense how people react to the set when they play it for the first time, so we like to hold playtests with fresh eyes and a more casual sensibility. During the first such playtest, we noticed that when players got an Eldrazi out, they wouldn't attack with it. R&D had quickly realized the power of the annihilator mechanic, but the less experienced players didn't seem to feel safe throwing their giant creatures into the fray.

We took notes and talked about it, but changed nothing. The next playtest the same thing happened, and the playtest after that. During one development meetings we had a discussion about what we could do to make less experienced players realize that you should be attacking with the Eldrazi. (Quick aside—Less experienced players: When you get an Eldrazi on the battlefield, attack with it.) How about, I suggested, we take the cheapest common Eldrazi and add the text "CARDNAME attacks each turn if able." Once players saw how much damage the Eldrazi did when they attacked, hopefully that would encourage them to attack with the other Eldrazi. And that is how ..."OpenTip(event, "Ulamog's Crusher")" class="nodec">Ulamog's Crusher got its second line of text.


The "attacks each turn" was put in specifically to tell people they needed to attack with it. Not for flavor reasons.


www.wizards.com/Magic/Magazine/Article.a...

It's important to get a sense how people react to the set when they play it for the first time, so we like to hold playtests with fresh eyes and a more casual sensibility. During the first such playtest, we noticed that when players got an Eldrazi out, they wouldn't attack with it. R&D had quickly realized the power of the annihilator mechanic, but the less experienced players didn't seem to feel safe throwing their giant creatures into the fray.

We took notes and talked about it, but changed nothing. The next playtest the same thing happened, and the playtest after that. During one development meetings we had a discussion about what we could do to make less experienced players realize that you should be attacking with the Eldrazi. (Quick aside—Less experienced players: When you get an Eldrazi on the battlefield, attack with it.) How about, I suggested, we take the cheapest common Eldrazi and add the text "CARDNAME attacks each turn if able." Once players saw how much damage the Eldrazi did when they attacked, hopefully that would encourage them to attack with the other Eldrazi. And that is how ..."OpenTip(event, "Ulamog's Crusher")" class="nodec">Ulamog's Crusher got its second line of text.




Nothing wrong with it.
Nothing wrong with it.

It assumes the average player is a freakin' idiot.  That's what's wrong with it.

Um, it assumes they had several playtests wherein people consistently didn't attack with it. Which they did. So they did something that helped newer players, and which doesn't hurt experienced players in the slightest. Why is this a big deal?
Because, it merely reflects on R&D consistent inability to accept any use for a card than exactly the one intended.  Until, of course, something they never really considered ends up mauling a Standard environment and nine cards all get banned at once.