Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[faust]

This topic is inspired by a discussion from the Weekly Design Contest. It point of concern is: What happens if you design a cost-increasing fac card when you also have cost decreasers like Highway around, in particular with the "cards never cost less than $0" rule.

@faust's Laborer: I'm skimming through the discussion. I hope my comments on this are not unwelcome. What if we say that cost-increasing happens before cost-reduction, as a general rule? I'm not too versed with strange costs (Wayfarer mostly), but this seems like it should solve problems.

I guess it's not my card / not my call, but picture what this looks like in terms of the dominion dot games log - all the sudden you're inserting events before the card is played. Likewise Gubump's proposal - it creates this weird inconsistency where the whole turn needs to be evaluated in one go (despite not being complete),

You don't need to evaluate the future. You only need to evaluate everything that's happened prior in the turn...just like you have to do with +Actions, +, and +Buys, among other things.

you do if you play enough cost reduction and then play cost increases. Everything else does not require this evaluation filter of "is the cost less than zero? if so, its now zero, but if you increase the cost, it was not presently zero"

Lets consider possibly the only canon example we have for what I'm describing vs what you're describing: poor house and wot chameleon, plus some source of coins on action cards.

Like a moron, I played two peddlers before drawing into my poor house. I've got $2 at the moment. Lets say other than that, I've got a hand full of Copper - 4 coppers, one poor house.
I play Poor House using chameleon, draw four cards (lets say it was a Bad 4 cards - all curses - unhelpful) and reveal my hand - four coppers there. How much money do I have at the moment? $0. $2 - $4 => $0.

Now lets play our treasures. I say we end up with $4 - we were at $0, each one contributed $1.

Your reasoning on this puts us at $2 - we were secretly at $-2 (filtered to appear at $0 before more coins were present) but then played 4×$1 worth of treasures. But we know that's wrong.

I don't see why costs should work differently here - you evaluate things in the order they're played, once you hit a floor, that's the new value.


Likewise, we know that order-of-treasures does matter with things like Bank and Fortune - we can't insert the +$ amounts before all the -$ amounts in the order things are played - time travel evaluation is not how the game works. Again, cost adjustment should be consistent with every other tool we've got.

At the core of this analogy seems to be a conflation of two different types of effect: ongoing and on-play. Of course things that the on-play ability of a card happen when that card is played; this was never in question. But this is not the same as discussing ongoing effects. My previous example was bad, so let's consider this instead:

You play a Haggler and then a Talisman. Now on buy, I get 2 effects. But I get to choose the order, even though Haggler was played before Talisman. This is how ongoing effects are resolved in Dominion, this is the precedent.

I understand if the hypothetical negative amount of coin seem counterintuitive, so here is another way to think about it: If I played 6 Highways and then had the Laborer in hand (for context: Laborer is a $5 fan card that increases in cost per Laborer in play), 5 of the Highway effects trigger to modify its cost, until it costs $0. Then the 6th effect can no longer trigger because the card costs $0. Now I play the Laborer. By Laborer's rules, it increases its cost to $1 but now that ongoing effect sees that the card no longer costs $0 and thus can trigger, again lowering its cost.

I have noticed that there is actually a real-Dominion example that contradicts you. Consider Fisherman. I play 3 Highways with an empty discard pile. By your rules Fisherman's cost reduction triggers first, then 2 or the Highways trigger, then the third Highway fails. Now what happens when I discard a card? By your logic, Fisherman's cost reduction would just stop working and the Fisherman would end up costing $3. But we know that's wrong; in this situation, Fisherman actually costs $2.

[spineflu]

you're right; my mental model of how this stuff happens wouldn't have caught this.

[GendoIkari]

I'm sure I've said this before but it's been a long time.

To me, the clear way to deal with it is to not care when cost-changing events or cost-changing ongoing effects actually start and stop. Rather, any time we care about what the cost of a card is, we start with the printed cost and then apply any applicable effects. It's fine to define an order for those effects, such as "cost increasers first" or "ongoing effects first", but the important thing is that we don't consider the calculated cost of a card to be a new baseline upon which future effects or calculations can be applied.

This is once again borrowing from the rules of MTG. In Magic, if you want to know a creature's power or toughness, you don't constantly track how a creature's power or toughness has changed over time as each new effect changes it. Rather, you always recalculate from scratch; starting with the printed numbers and then applying all the different effects in a defined order.

613.1. The values of an object’s characteristics are determined by starting with the actual object. For a card, that means the values of the characteristics printed on that card. For a token or a copy of a spell or card, that means the values of the characteristics defined by the effect that created it. Then all applicable continuous effects are applied in a series of layers in the following order:

The order then clarifies things such as how if an effect sets the power to a specific number, that happens before effects that just add 1 to the power.

This system deals well with the Fisherman issue; when your discard pile gets a card in it again, we don't just "undo" the Fishing Village effect. Rather, we recalculate from the beginning, and Fisherman's effect is no longer there.


All that being said, I prefer the route of simply never having cost increasers. The main thing that it does was already solved by Donald with the - token. It causes issues with things like Livery, and confusion such as this discussion. It requires an arbitrary rule about whether to apply increasers or decreasers first.

[The Alchemist]

I second "solving" this issue by just never having cost-increasers. Livery has made it very clear this is not a direction Don X. wants to go. I don't think we'll ever see one, and I think that's reason enough to not try to make one.

[spineflu]

I second "solving" this issue by just never having cost-increasers. Livery has made it very clear this is not a direction Don X. wants to go. I don't think we'll ever see one, and I think that's reason enough to not try to make one.

I don't think that one broken card interaction (we don't stop using Bishop because there is one broken combo either; if that combo occurs you can simpl redo the setup) is a sound argument against cost increasers. While I would argue that there is nothing inherently wrong with cost increasers, the more natural way to implement them as Attack (well, that is kinda obvious, if they hit you theyare a kind of self-attack) is via Debt (something like "when another player gains/buys a card, they take 1D").

yeah if they affect other players, use the -$ adv token, if they affect you, debt; because of pin potential, debt should always be a voluntary (although at times coerced, maybe) affair for the player taking it.