I have already provided at least seven examples of player Feats or item abilities that do not work. One could claim these are designer mistakes, but considering the number of them and the fact they run the gamut from pre-remaster to relatively new releases, this is unlikely.
It has already been explained, but that's not true. (and this idea-jam is the same blocker preventing this talk from getting through to you)
This is why I bothered to write out the example where a cat can both be "in the center" and "at the edge" at the same time. "Where is it?" at first glance seems to only have one valid match, but it genuinely has more than one correct answer.
The issue is the ask requires a frame of reference, leaving the reader to fill in that gap. As a general rule, the first assumption is the most specific one that's shared by speaker/listener.
ScooterScoots' spaghetti eating exemplified that quite well.
Action(subtype-Activity) [Eat Spaghetti] contains 3 sub-Interacts to [fork the noodles] [swirl noodles] [swallow noodles].
"What was your last action?" has 2 correct answers. Both [Eat Spaghetti] and [swallow noodles].
This is why the devs don't worry about ambiguity when "prev action" is asking for an activity. Activities are actions. That more-specific activity ask is reducing the valid options to check against to 1.
(though the devs 100% should ask for prev "activity" in those circumstances, because...)
This is also exactly why this is throwing some people for such a loop. When the "prev action" check is not an activity, there are 2 different right answers to check.
And instead of allowing that, some folks are instead disqualifying the sub-action categorically to change the logical ask into a more math-like a value-exclusive equation.
Done by adding a rule that doesn't exist:
Because what myself and others are saying is that we read the rules as stating that subordinate actions do not count, specifically for abilities that have a Prerequisite that references the next action you take or the last action you took. That's it. In all other ways, they function exactly the same as their non-subordinate counterpart.
The "problem" is not that you are giving a "container" structure to activities. The problem is you point to that structure as "invisible rules" text that has a big mechanical impact on the game.
On the other side of the ~180 hits for "prev action" abilities, there's a few hundred meta-strike activities that would be rendered incompatible.
Again, I really don't believe the dev who wrote Clawdancer would be happy if you only allowed standalone Strikes to count for all their "prev action" feats. Just because a ruling doesn't render an ability literally unusable does not mean it's a reasonable ruling to enforce.
Banning sub-actions from "prev action" checks is imo blatantly unreasonable.
___________________
We keep circling this issue where some intuition-based trigger is resulting in you removing a perfectly correct match for "previous action."
I'll just repeat that implementing your idea and following through would genuinely block an unknown number of interactions that previously did not even register as something of doubt.
| 1 person marked this as a favorite. |
Yes, well, if stating how trip/grab works following the double attack is not considered allowing you to do just that we are quite frankly being very silly.
Dude... text is not some ambiguous thing you can just call "close enough" and claim truth. If you are going to talk like that, why are you even wasting your time posting here?
... until after it has made both attacks. If the zelekhut subsequently uses the Knockdown action, it affects all creatures it hit with Double Attack.
.
When text is giving you an ~'eligibility override,'
it will either demand that ineligible action: "you do a ___"
or it will directly add a new possibility: "you can ____"
"if the __ uses ___" genuinely has no way to twist that into being an eligibility override.
That phrasing only adds an extra mechanic to the existing Knockdown.
If you want to have any sort of productive discussion, you need to engage in good faith. You can easily still attack the cited text; diminish it's weight by saying that it's just monster ability with less QA, and a dev error, etc.
TripH, back to the original topic How does Barrows edge/ drink of my foes work you make a strike with your immanence drink of my foes that takes 2 actions correct? you deal your damage and heal the same amount, Correct?
Drink of My Foes is a 1A action with the requirement that your prev action was a successful Strike, rather normal. But the HP restore is "half the damage dealt," rather rare language in pf2.
That gives it a lot of flexibility, as keying into the "final damage" of the Strike means that all methods of increasing that Strike damage will translate to more healing.
Your spark does need to be in there before the Strike, and there's a 1 Trans ability per turn limit to Exemplar, so you likely want to look outside of Exemplar for means of boosting damage.
There are a lot of power-attack style meta-strikes all over the system, but you don't even need to get one if that's not your style.
Before-Strike buffs work just as well, even a spell like Organsight would be compatible.
My unplayed Exemplar on the shelf went Fighter for Swipe, and to get Reactive Strike in an archetype slot. It's not really a great option. Swipe is rather contextual, and requires reach upon 2 foes adjacent to each other.
Looking elsewhere, there are genuinely too many Strike-boosters to count. Ancestry like Nagaji can buff a future hit for 1A, spells like Blink Charge make a Strike w/ bonus damage, and plenty of martial archetypes involve meta-Strikes, or buff Strike damage some other way.
I like Drink of my Foes exactly *because* you don't have to know AoN's complex search syntax to actually find a decent pairing; if your PC is going to want to smack foes and deal damage, it'll work.
Even if it's not the #1 whiteroom tested combo, having so many options give a numeric boost helps one to stop min-maxing and instead choose what they think is a cool and thematic match for that particular character.
| 2 people marked this as a favorite. |
Citing text like Zelekut's was done to determine the author's understanding of the RaW on "prev action" compatibility.
The monster text is specifically useful because it is harmonious only if "prev action" matches with sub-actions.
If the author believed sub-actions were incompatible, then this ability would technically be impossible to trigger with the RaW. Meaning an author with that "not compatible" understanding would have written different text; it would have been as easy as swapping from the sterile "if ___ then ___" and into the usual "___ can ___" language of a permission override.
And in the case of GM's reading the Kraken or Zelekhut entries I am confident they will be able to run them as intended and it is totally unambiguous.
This was never the point of bringing in Zelekhut's text. No one discussing it did so in an effort to claim the GM would become fraught with textual ambiguity and unable to proceed.
In truth, it is the exact opposite.The presumed-universal agreement that the ability is intended to function is the entire point of citing that text; the understanding that Zelekhut's ability is supposed to work is why it enables us to gain insight into the author's understanding of "prev action" compatibility.
Dare I say, one might call it "disingenuous" to act as if that irrelevant question was the real reason that Zelekhut was cited, and not that it unambiguously reveals the author's PoV on "last action" compatibility.
You are free to use other attacks, accuse Zelekhut, and all the other monsters with similar text as being "author mistakes" if you wish, but please refrain from whatever tactic you wish to label that prior post.
| 1 person marked this as a favorite. |
If the authors themselves are confused it points towards the issue being ambiguous, and the best way to resolve that is to just use the interpretation that lets everything work.
In this case we even have conflicting elements (e.g. Flensing Slice vs. Zelekhut), so it's more or less useless as a method of analysis.
You have already had it explained that there is no conflict. While it is understandable for you to desire to remove that perceived ambiguity, it is still a mistake to then invent a rule that's not in the text.
It is precisely because of that "container" idea that you end up with 2 correct matches to check against when prompted for "last action?"
Even when operating inside normal English, that still invokes logic with non-math answers like that.
[Eat Spaghetti | cost:2A](Action-Activity) invokes 3 sub-Interacts to [fork the noodles] [swirl noodles] [swallow noodles].
Because the activity is subdivided into actions, both the last sub-action and the activity itself are valid answers for "your last action was: ____" when working with natural language.
I like the "where is the cat?" example because it sounds so contradictory for both "near the edge" and "in the middle" to be correct answers without context.
It does a good job of showcasing that even the logic of natural language genuinely DOES NOT translate to the simplicity of computer logic or math, where things usually have one single answer.
Because the natural language interpretation of the last thing you did includes both the broad thing done and the subcomponent, and there are features that depend on the broad thing done, and features that depend on the subcomponent (Nothing in Zelekhut‘s knockdown grants bypass to the standard last action requirement).
Next actions don’t work this way because it’s specifically said they don’t, but if that wasn’t said sure they would.
| 1 person marked this as a favorite. |
I think we're at a point where this thread is no longer a productive discussion.
It doesn't seem like anyone is interested in changing their mind, just defending their positions (myself included, which is why I've disengaged, because I find no value in trying to convince someone who doesn't want to be convinced).
Barring developer clarification on the matter, everyone run it how you see fit.