|
RP: Poor Lini - she got captured and thrown into a damp cell. Hoping to be rescued by Valeros, Ezren and Merisiel she spots a mouse in her prison cell.
Checking her pockets she notices a crumbled fortune cookie. Carefully feeding it to the mouse she attaches a ring of hers to the mouse and imprints the image of the tavern they stay into the small animals mind.
This is theory craft and a serious rules question - but I thought I start with a real possible situation.
This question got inspired by Ascalaphus and MathMuse in a different thread. How do we actually count actions for flat checks.
Now to the rules part that are important.
Primal Caster
Casting Animal Messenger - a 1 minute casting time
The spell has material, somatic and verbal components
The caster has her wrist bound by manacles (just standard) incurring a DC 5 flat check on manipulate actions
Now the question - how many manipulate actions does casting the Animal Messenger need?
This actually isn't that simple. I requote CRB
There are four types of actions: single actions, activities, reactions, and free actions.
So counting 'actions' can be done in different ways - depending what qualifies as action according to RAW when doing a flat check against a manacle.
Sensible answers by RAW seem:
1 check - I treat all as 1 activity/action
10 checks - the activity is split across 10 rounds
20 checks - there are 10 somatic and 10 material actions done during the whole cast
30 checks - there are 30 actions total (3 actions/round times 10 rounds in a minute)
Chances of success are
1 check: 80% - 4 in 5 (16 in 20)
10 checks: 10.7% - approx 1 in 9
20 checks: 1.2% - approx 1 in 87
30 checks: 0.12% - approx 1 in 800
All numbers are rounded slightly. I included them as they show that depending on a ruling the result goes from a very good chance to an outside chance to near impossible.
How would you rule - and why?
| 3 people marked this as a favorite. |
It's one activity, so it's one check.
The other options on your list of "sensible answers" are not sensible (so much so such a GM would garner a lot of disrespect, likely enough to never return). RAW makes no mention of how many actions/sub-actions of a type there are when casting and the correspondence of # of components to # of actions went away after the playtest. So there is no other answer, no other metric to determine how many such actions occur. May as well roll a d30 which shows the absurdity of trying to apply legalistic rigor to what's also ultimately a matter of narrative & gaming.
If my player had bothered to actually take that spell and fell into such a circumstance (where the guards were also dumb enough to let them keep spell components?), it'd be a cruel deed to rule otherwise (while still maintaining the one fail chance since it is a game w/ ups and downs). So RAW has no other footing, and since PF2 explicitly says RAW is no longer king anyway, why even go there?
And yes, I'm aware of awkward exceptions, namely reloading over separate rounds, yet unless there's a more distinct break in their activity, I'd allow the creature to continue fine if it began w/ a successful roll.
Of course, if time's a factor, there is the pertinent question of "When does the roll occur?" which could matter. Example, if there's a second casting available and they only have a "just over a minute" before the guard will return from their pee break. I'd likely add a roll of some sort, but more of a luck roll than "that's exactly when that component occurs for that spell" roll.
| 1 person marked this as a favorite. |
One check. Use the same result for the entire duration of the casting time.
Like Thod already pointed out, requiring multiple successful checks with no failures allowed causes an exponential decrease in overall probability of success. So it becomes really hard for the GM to set the DC to the appropriate level to make the outcome of the dice roll interesting.
That is also why we don't make different numbers of checks due to different action costs of various spells or even different amounts or types of casting components.
RP: Poor Lini - she got captured and thrown into a damp cell. Hoping to be rescued by Valeros, Ezren and Merisiel she spots a mouse in her prison cell.
Checking her pockets she notices a crumbled fortune cookie. Carefully feeding it to the mouse she attaches a ring of hers to the mouse and imprints the image of the tavern they stay into the small animals mind.This is theory craft and a serious rules question - but I thought I start with a real possible situation.
This question got inspired by Ascalaphus and MathMuse in a different thread. How do we actually count actions for flat checks.
Now to the rules part that are important.
Primal Caster
Casting Animal Messenger - a 1 minute casting time
The spell has material, somatic and verbal components
The caster has her wrist bound by manacles (just standard) incurring a DC 5 flat check on manipulate actionsNow the question - how many manipulate actions does casting the Animal Messenger need?
This actually isn't that simple. I requote CRB
There are four types of actions: single actions, activities, reactions, and free actions.So counting 'actions' can be done in different ways - depending what qualifies as action according to RAW when doing a flat check against a manacle.
Sensible answers by RAW seem:
1 check - I treat all as 1 activity/action
10 checks - the activity is split across 10 rounds
20 checks - there are 10 somatic and 10 material actions done during the whole cast
30 checks - there are 30 actions total (3 actions/round times 10 rounds in a minute)Chances of success are
1 check: 80% - 4 in 5 (16 in 20)
10 checks: 10.7% - approx 1 in 9
20 checks: 1.2% - approx 1 in 87
30 checks: 0.12% - approx 1 in 800All numbers are rounded slightly. I included them as they show that depending on a ruling the result goes from a very good chance to an outside chance to near impossible.
How would you rule - and why?
To be clear, this is an out-of-encounter activity, not an in-encounter activity, making the concept of actions irrelevant to the task at hand, meaning the Grabbed condition's "Manipulate Action" sentence doesn't make sense here. At best, there might be perception checks of guards outside to hear you casting a spell (or, at the very least, talking to rats like a crazy person), otherwise the odds of triggering combat from this activity isn't plausible. And if, for some reason, it does, then the spell will fail no matter what, manacles or not. But let's say the guards are overconfident or incompetent and are simply slacking off, for simplicity purposes.
Assuming no outside interference, the concept of intensive activity required in an encounter scenario doesn't apply, meaning it's not relevant to determine if you failed the check in Round 2 or Round 7, etc. As such, the character isn't as pressured or as concerned with flailing arms and talking, meaning requiring flat checks for each round doesn't make sense, especially given the rules for Long Cast-Time Spells already provided.
That being said, it is still a spell, with components, each of which still has requirements that all need to be fulfilled, lest the spell is a failure to cast. Each component must be separately accounted for, not unlike casting in encounters, and failing to provide any one of these means the spell fails to cast. As such, that leads me to believe that 2 flat checks are required, 1 for each component that requires manipulation on the character's part. The reason being is because each component requires activity on the character's part, and they are separate, distinct sub-activities combined into the singular activity being described (not unlike subordinate actions).
|
How would you rule - and why?
So spells are actually a bit of an unusual case, because the spell components look a lot like actions from a distance, but they're not. Spells tend to require the same number of actions as they have components but that's just a correlation, not a causation. Some spells need more or fewer actions. And using metamagic to get rid of a spell component doesn't change the action cost.
Also unusual is that in the case of spell components, the traits bubble up to the main activity. ("The spell components, described in detail below, add traits and requirements to the Cast a Spell activity.") This is different from typical activities where subordinate actions keep their own traits. But then, spell components aren't subordinate actions, because they're not actions at all.
So for the animal messenger spell, it has the manipulate trait because it gets them from multiple components. But it can't have the trait twice; it either has it or it doesn't.
So, Lini only needs to make one check.
Spell components are provided with a spellcasting feature ('you can supply material, somatic, and verbal components when casting spells') and specific requirements like:
That being said, it is still a spell, with components, each of which still has requirements that all need to be fulfilled, lest the spell is a failure to cast. Each component must be separately accounted for, not unlike casting in encounters, and failing to provide any one of these means the spell fails to cast.The spell gains the manipulate trait and requires you to have a free hand to retrieve and manipulate a material component. That component is expended in the casting (even if the spell is disrupted).The spell gains the manipulate trait and requires you to make gestures. You can use this component while holding something in your hand, but not if you are restrained or otherwise unable to gesture freely.Traits of a whole spellcasting activity are binary, they either exist or not, actually all traits are like that, they aren't counted. There can't be 2 manipulate traits, so all checks for manipulate trait are done only once.
The reason being is because each component requires activity on the character's part, and they are separate, distinct sub-activities combined into the singular activity being described (not unlike subordinate actions).They are not actions and are not activities, you completely invented this yourself and needlessly complicate situation.
This is normally true, but it creates poor dichotomies by contrast. If I have a 1 action activity that lets me perform two interact activities for a single action, I would indeed have to make two flat checks if I am Grabbed or similarly inhibited, because the two subordinate actions are processed in a semi-specific order. Not unlike Sudden Charge, where I would trigger two reactions if I am still in range of an enemy's reach through both Strides (such as going around an enemy to provide flank), since you are told you Stride twice in the activity. In both cases, the same concept can apply to spellcasting, simply because providing components can be linked to a comparison of subordinate actions.
An activity typically involves using multiple actions to create an effect greater than you can produce with a single action, or combining multiple single actions to produce an effect that’s different from merely the sum of those actions. In some cases, usually when spellcasting, an activity can consist of only 1 action, 1 reaction, or even 1 free action.
Casting a Spell is a special activity that takes a variable number of actions depending on the spell, as listed in each spell's stat block.
A spell description lists the components required to Cast the Spell. For most spells, the number of components is equal to the number of actions you must spend to Cast the Spell. Each component adds certain traits to the Cast a Spell activity, and some components have special requirements.
So, with the above, we have confirmation of:
-The definition of activities
-Spellcasting as an absolutely confirmed activity
-Unless shown otherwise, spell components for a spell are equal to the number of actions the spell requires to cast
All from the Core Rulebook. But don't worry, I apparently chose the Inventor class and chose to invent RAW to support my claim. /sarcasm
There's really only one other piece of information that supports your claim, but I still disagree with it for the previously stated fundamental reasons.
Some spells take minutes or hours to cast. The Cast a Spell activity for these spells includes a mix of the listed spell components, but it's not necessary to break down which one you're providing at a given time. You can't use other actions or reactions while casting such a spell, though at the GM's discretion, you might be able to speak a few sentences. As with other activities that take a long time, these spells have the exploration trait, and you can't cast them in an encounter. If combat breaks out while you're casting one, your spell is disrupted
So, the rules do state that the components required to supply are listed, but that you don't have to track the rounds or turns when you supply them, merely that they are provided. This tells us two things: 1. The components don't need to all be provided each turn at all times throughout the duration of the activity, compared to an Encounter Mode spell (or the Repeat a Spell activity, for a more apt comparison), only that they are provided once (presumably as part of the entire activity), and 2. You don't need to track what round or action you (fail to) provide the component(s), only that they are provided at some point throughout the activity's elapsed time. Even if, for some reason, you need to to track it (such as by determining what everyone else is doing in the meantime), this can be easily arbitrated by the GM (such as by rolling a D10 to determine which round the flat check caused to fail).
Honestly, the fact that you have to process each component separately, even abstracted, should mean that things which affect each component should apply only to those components. Otherwise this means you only make one flat check while both Grabbed and Deafened, and I can assure you that people would be up in arms about this.
But fine, I can see where the interpretation comes from, and while I can understand it from the binary perspective (having multiple identical traits is effectively redundant), the problem I have with it is that subordinate actions and other activities aren't simply "binary" as we're stating; they have consequences when each are applied independently, or in sequence, and as such should be treated independently.
| 1 person marked this as a favorite. |
Otherwise this means you only make one flat check while both Grabbed and Deafened, and I can assure you that people would be up in arms about this.
Isn't that exactly what the quote from the CRB that egindar provided literally says?
If more than one flat check would ever cause or prevent the same thing, just roll once and use the highest DC.
In this case the 'thing' being prevented is casting the spell. Two flat checks are required - one for Grabbed, one for Deafened. But you only roll one flat check with the higher DC.
Otherwise this means you only make one flat check while both Grabbed and Deafened, and I can assure you that people would be up in arms about this.Isn't that exactly what the quote from the CRB that egindar provided literally says?
If more than one flat check would ever cause or prevent the same thing, just roll once and use the highest DC.In this case the 'thing' being prevented is casting the spell. Two flat checks are required - one for Grabbed, one for Deafened. But you only roll one flat check with the higher DC.
Interesting; I must have skipped over this part, though I think I missed the post mentioning this while typing up a response that I curtailed quite a bit; biting the tongue is hard, sometimes.
While "thing" can refer to the activity as a whole, as you state, which is reasonable, it could also refer to the individual components/traits as well, or negative effects stacking to the same flat check, also equally reasonable from my perspective. In my opinion, the lack of specification is its biggest downfall.
I do stand by my interpretation still, but at least I can better see the opposition, and I won't be as confrontational when it's proposed.
|
I think this rule was made to facilitate ease of play. It looks very similar to the rule that, when you suffer from several sources of the same condition, removing one of them removes all of them.
And verisimilitude be damned. Which is a good thing in my book, because I favor a fluid game over a "realistic" one.
|
|
| 1 person marked this as a favorite. |
This is normally true, but it creates poor dichotomies by contrast. If I have a 1 action activity that lets me perform two interact activities for a single action, I would indeed have to make two flat checks if I am Grabbed or similarly inhibited, because the two subordinate actions are processed in a semi-specific order. Not unlike Sudden Charge, where I would trigger two reactions if I am still in range of an enemy's reach through both Strides (such as going around an enemy to provide flank), since you are told you Stride twice in the activity. In both cases, the same concept can apply to spellcasting, simply because providing components can be linked to a comparison of subordinate actions.
The difference is that spell components are NOT subordinate actions. They look like it from a distance but they aren't the same thing. It's just typical that a spell takes the same amount of actions as it has components, but it's not a causal link.
You can't say "the third action of this spell is clearly the material component, if I quicken the spell I won't need to do that anymore" or "if I no longer need to do the somatic component of this spell, it takes one less action to cast".
Also, spell components push their traits to the spell, whereas subordinate actions don't push their traits to the parent activity.
... the problem I have with it is that subordinate actions and other activities aren't simply "binary" as we're stating; they have consequences when each are applied independently, or in sequence, and as such should be treated independently....All connections of spell components and actions are invented by you, irrelevant to the base game and simply needless. You definitely can use these homerules, but that is just what they are.
No, they're not. I literally showed you RAW saying otherwise. Unless you're trying to call me Paizo, saying I invented it is plainly false.
| 2 people marked this as a favorite. |
No, they're not. I literally showed you RAW saying otherwise.
???
No you didn't.
This is normally true, but it creates poor dichotomies by contrast. If I have a 1 action activity that lets me perform two interact activities for a single action, I would indeed have to make two flat checks if I am Grabbed or similarly inhibited, because the two subordinate actions are processed in a semi-specific order. Not unlike Sudden Charge, where I would trigger two reactions if I am still in range of an enemy's reach through both Strides (such as going around an enemy to provide flank), since you are told you Stride twice in the activity. In both cases, the same concept can apply to spellcasting, simply because providing components can be linked to a comparison of subordinate actions.The difference is that spell components are NOT subordinate actions. They look like it from a distance but they aren't the same thing. It's just typical that a spell takes the same amount of actions as it has components, but it's not a causal link.
You can't say "the third action of this spell is clearly the material component, if I quicken the spell I won't need to do that anymore" or "if I no longer need to do the somatic component of this spell, it takes one less action to cast".
Also, spell components push their traits to the spell, whereas subordinate actions don't push their traits to the parent activity.
It's a default assumption, yes, unless the statblock says otherwise. The point being made there is that the RAW supports that ideal, and wasn't just a fabrication on my part.
You wouldn't need to specify which action is the component, merely that you are selecting the action taking the component, and is especially true with spellcasting. And I probably wouldn't permit removing the material component if the spell had a costly material listing, obviously. But this is an issue with Quickened Casting not specifying it does or does not remove the associated component, when that previous RAW implies that it would. The only other RAW is with Touch spells, they specifically must have Somatic components, meaning most Quickened Casting Touch spells would be forced to remove the Verbal component, if the rule functioned that way.
There is fundamentally no difference in that respect; that's why the rules had to go out of their way to specify things like "you only trigger 1 reaction per action spent moving, not per square."
No, they're not. I literally showed you RAW saying otherwise.???
No you didn't.
A spell description lists the components required to Cast the Spell. For most spells, the number of components is equal to the number of actions you must spend to Cast the Spell.Each component adds certain traits to the Cast a Spell activity, and some components have special requirements.
Yes I did. This is twice now that I referenced the RAW.
| 2 people marked this as a favorite. |
Yes I did. This is twice now that I referenced the RAW.
Nothing in the quoted section says that spell components are subordinate actions.