Witch of Miracles's page

I've had issues with hazards and haunts since I started running the system (and my first foray into that was Malevolence, so I've had more than a few of said issues). The OP has the right of it.

The biggest issue in my experience is that most solo hazard encounters don't seem to consider that the party can just... avoid them. There's usually no reason to interact with them unless you're forced to, and you're rarely forced to, so you don't. The party is informed it's there (or isn't and someone triggers it), and then they just walk around it or something. It's silly.

And like the OP says, even if you don't avoid them they often have no long-term cost. (Malevolence at least does do something about this, to its credit.) They feel like bug bites in too many cases instead of an actual scare, for all the reasons the OP describes.

Hazards as part of another encounter are a different story, obviously. But hazards just in an otherwise empty hallway or room, or along your travel route, just are not the stuff unless they're really really really scary.

I wouldn't allow people to Avoid Notice if there was no cover or concealment to hide behind, and place them at the nearest cover point where they would be able to peak around and see the enemy when initiative is rolled.

This is a sensible fiction-first way to handle it, but I can already think of at least one case where AP text is written as though the party members are allowed to use Avoid Notice while traveling along what look like open roads or similar on the map—and it has the mechanical effect of reducing the flat check for random encounters.

rules snip

That is a good catch, though that admittedly doesn't say what to do if the map lacks reasonable hiding spots. The "you're just seen" interpretation is clearly consistent with the other rules, though, yeah.

I do think there's a bit of an intuitive conflict here, insofar as your ability to use Avoid Notice, to subsequently roll Stealth as initiative, and parse that by following the rules in "Initiative with Hidden Enemies" isn't really dependent on the map... but whether you're in plain sight is pretty clearly dependent on the map.

I'm not sure there's a clear answer to this one, honestly. I think it's just an artifact of transitioning from loose, abstracted exploration rules (where the game just assumes you have a way to hide or stealth regardless of the actual terrain) to tighter grid-based combat system with a fixed map (where that assumption can quickly be broken). The GM is kind of left to decide how to square that difference.

Honestly, the prevalence of multiclass archetypes, in particular, is a symptom of the game both failing to scale in certain ways (early class abilities poached via archetyping scale worryingly well) and the game failing to provide in-class solutions to obvious problems (granting strong 1A abilities and reactions).

Too much of the game's mechanical payload is frontloaded. The result is that archetyping grants you features on par with in-class feats of a much higher level in terms of real, in-play value. It's admirable to want to make the stuff you do at level 1-6 remain relevant to a degree throughout the game, but this is a consequence that wasn't really designed around.

And we've talked at length already about how desirable it is to get good third actions and reactions on classes that might natively lack them.

I would also like to say that even in FA games, no players at my tables or the tables I play at have known to take psychic on magus without being informed by a GM. Even then, one of those magi never bothered with it and is still very happy with their build, being a more RP-first player.

This is not particularly true in practice, at least not from level 6 onwards when the Magus gets imaginary weapon and becomes able to deal slot spell-grade Spellstrike damage 3 times per encounter. This, and not the Magus in a vacuum, is the standard I'm operating on, especially as I don't think the Magus is too strong a class even with that combo. This is similarly why I don't believe this would meaningfully affect the Magus's damage output overall, even if it absolutely would massively buff the Magus's ability to output single-target debuffs and crowd control, because the Magus already operates at a level where using slot spells instead of imaginary weapon would not constitute a major damage increase.

This is one thing if it's for personal use at tables with a certain expected level of optimization; you don't need to worry about it. But if it's for general release, you need to be aware that classes that have high built-in floors can cause issues at low optimization tables.

It can be true that this magus does less damage in most encounters than IW magus. It can also be true that this magus does significantly more damage in most encounters when built by a player that doesn't know much about PF2E optimization as compared to stock Magus. And that can make it a problem.

Easily optimized classes and subclasses (i.e., ones that you can optimize with little system knowledge, usually just by picking class feats that fit what the class/subclass seems to want you to do, or ones that have mechanics that give them a good baseline regardless of feat selection) are often the strongest classes in practice at most tables. I'm worried that this class proposal has a very, very high baseline in virtue of its mechanics.

Some things I'm immediately leery of:

-Both the magus players I've run for have enjoyed having the option of using slotted arcane spells available for utility or damage, and saw it as a significant perk of playing the class. I do not think they would be happy to lose the ability to, say, cast invisibility on half the party if they felt it would be a good plan for the day. This is especially true for magi in parties without two full casters.

I don't /personally/ think this is optimal and wouldn't play magus this way, but it is clearly a play pattern people ask for and would want to preserve.

-Comparing your attack roll against a fortitude/reflex/will DC to see the outcome of a spellstrike spell is a 4E takeaway that makes some sense, but I think this could result in some worrying interactions. Saves aren't designed with martial accuracy progression (especially the to-hit bonuses from items) in mind.

-Cantrip spellstriking is still pretty okay damage, especially with Force Fang as a recharge. None of the magus players I've had felt like they underperformed when using gouging claw spellstrikes, even if slotted spellstrike or amped IW spellstrike are more damaging. The buy-now-pay-later nature of spellstrike (i.e., you get 3 actions of damage for 2 and pay the third later) is a fairly significant tempo advantage at times, even just with cantrips. It is frontloaded burst at MAP-0! Spellstriking with cantrips still needs a recharge cost to account for this tempo advantage.

-You're effectively letting magus pick a slotted spell to use as a focus spell with spellstrike, no? If you're doing that, the spell rank should be capped to one or two lower than the max a 10th caster gets at the same level, to keep it more in line with other focus spells. Even with this limited casting, though, I'm certain there will be some extremely wonky abuse cases.

I know you're not mechanically making them focus spells, but as a renewable cast, I think the same power level guideline should apply.

I also think the number of slot spells available is both unspecified and incredibly important. Frankly, I cannot see this being balanced unless it's either just a single top-rank spell per encounter and that's it, or you get maybe 3 top rank-2 spells if you'd prefer a larger quantity.

The issue is that you've taken slotted spellstrike from a daily resource that might not be spent into a per-encounter resource that is guaranteed to be spent. Magus damage now has however many slotted spellstrikes per encounter as a baseline.

-Between the last two changes (guaranteed slotted spells for use in every encounter and removing spellstrike recharge on cantrips), you've massively increased Magus's damage floor. Even unarchetyped, magus damage is acceptable right now because they deal massive damage on crits. Removing spellstrike recharge on cantrips increases the damage floor substantially by removing action inefficiencies and granting magi better mobility. It also has a massive knock-on effect: a ton of archetypes that were terrible on Magus because it was action-starved are viable choices, because it has 1A open far more frequently. This is neat from a build perspective, but it hugely increases the baseline powerlevel of the class.

I feel like these changes have way too much of an eye towards the current power ceiling of magus when archetyped for focus spells, and don't take enough account of the floor.

-I think the general flavor of magus that attracts people is "INT caster with a sword." Things like recharging spellstrike with more typical martial skill actions (athletics, intimidate) don't feel as on-flavor unless they're attached to a conflux spell. There's a reason the main skill-action-plus-recharge is Magus's Analysis. And generally, I think this entire overhaul risks alienating people who come to the magus to get half a spellcaster and want to feel like a spellcaster some of the time.

I feel like the proliferation of alchemical items that solve common problems is far more of a nail in the coffin than trick magic item, which eats a skill feat and doesn't work with a lot of useful spells.

Like, if you haven't looked at the alchemical item list in a while, you really, really should. Some of the stuff on there is more than a little good.

What if we are looking at this all wrong? What if Psychic Dedication is not Overpowered but all other Dedications for Spellcasters are Underpowered? Psychic gives you the one thing you'd want most from a Dedication. What is all Dedications did that to some degree? Psychic grant focus spells which is better then ANY cantrip (Yes AMP'd Cantrips are both Cantrips and Focus Spells).

That feels much like the problem here, Psychic is frontloaded because it feels amazing. All other Spellcaster Dedications feel weak and like you're wasting a class feat to get a under performing feat to say, here have 2 cantrips AND 1 skill, which is better then the lackluster Cantrip Expansion feat.

Yeah, you can say Psychic is not broken. Magus is not broken but once they unite their hands together they become something more powerful then their individual parts.

As a reminder before the Remastered a lot of classes were grabbing Dangerous Sorcerer just for the flat 1-10 damage on thewir slotted spells.

As long as we agree that casting dedications allow you to use items like wands and scrolls, I think every casting dedication has a pretty significant frontloaded benefit. I feel like I remember seeing in another thread a long time ago that PFS locks all that behind the basic spellcasting benefits, though? Unsure.

The issue that a remastered magus has to deal with is that the focus spell space of the class was given over to conflux spells that recharge the spell strike. When you MC to get spell attack roll spells, you do lose out on the recharge focus spell option. That doesn’t really matter much with the starlight magus but the rest of them do have action economy issues with having to spend an action just to recharge. Just adding in conflux spells that use an attack roll ihas more complications that it might seem like.

You don't lose the option at all. You're still free to spend your focus points as you see fit. And psychic even gives you focus points more quickly than just going plain Magus, considering you get good focus spells at 2 and 6 instead of just at 2 (Force Fang) and then possibly never afterwards (depending on your opinion of Hasted Assault). If you really are in a situation where using a conflux spell multiple times is beneficial, you can use it more often for having taken psychic archetype—not less.

shroudb wrote:

agree on the fireball (because it's visible as soon as it leaves your person), disagree on the "the direction the pain comes from"

Witch of Miracles, Deiven, shroudb:

That's not how fireball works. The explosion of fire simply appears; there is no "trail back to the caster."

You do always create manifestations unless the spell is cast with subtle spell, but... yeah, I definitely got got on tacitly implementing the classic fireball flavor.

It's a weird one because people are reading it in two ways:

1. The spell creates a tree and makes it do stuff. Since it's a spell effect that is doing it, the caster is the source point for who is an "ally".

2. The spell creates a tree and the tree does stuff. Since the tree is an actor that has some of its own stats, can be targeted independently, and can react without the caster doing anything, the tree is the one doing the blocking and thus the tree is the source point for who is an "ally".

Inside Protector Tree itself, we don't have any guidance in terms of descriptive text or anything to suggest what was intended. So people reading it one of those ways will see it as making total sense and not understand the other way.

I guess the thing that's weird to me specifically about Trip's reading is that it hones in on a convention that seems like coincidence at best—that spells like protector tree have a subject+intransitive verb or a passive voice construction up front, and don't say "you do x"—and uses that convention to argue that the point of view of "ally" changes for reading such spells. The idea that the spell would change its reading if it said "You cause a medium tree to suddenly grow in an unoccupied square within range..." strikes me as prima facie absurd.

The tree can't change locations, but it is able to move its branches based on what is happening around it.

So, define "animated tree creature". :) Sometimes I hate how imprecise English is.

Yeah. Words are inherently vague, and it's not always for the best.

In my book, it just doesn't need to be more "animated" than a dancing weapon to perform the function being requested. It could be "alive" in the way a construct creature is alive, as well, but that doesn't feel necessary.

...It's a bit weird, too, because a tree is already alive in the colloquial sense anyways. It's a tree.

The semantics get really silly if you dig deep enough. According to Player Core's glossary, one could argue that the Tree is in fact a creature while it's active:

Player Core Glossary wrote:

Creature: An active participant in the story and world. This includes monsters and nonplayer characters (played by the Game Master) and player characters (played by the other players).

While the tree is interposing itself to block damage, it's active and is a participant in the story. Since it's a creature that does stuff without someone else spending actions to make it do that, it makes sense that the tree is the one doing the thing and thus it's the trees allies that count.

Now I'm certain this isn't how the definition is meant to be used, but if you take RAW to its extremes, you get outcomes like this. Which is why I don't tend to do that when I'm making rulings for my table.

Okay, that got one heck of a good laugh out of me. Point taken.

FWIW, my usual understanding of the game is that it mostly has two classes of "thing" at the top of its "ontology," and those are creatures and objects (with objects further being split into attended and unattended objects). The difference between creatures and objects is mostly is how you can interact with them, based on what rules text references each. The lines between them are sometimes blurry (particularly in the case of constructs and intelligent items), and sometimes, you need to override rules text that distinguishes the two when it probably shouldn't. (Strike only targets creatures by RAW, which is often a problem, though I can see why they would want you to need GM permission to target attended objects.) But I see those as being the main categories of thing in the game.

The game mainly delineates between the two, in my view, to try to rule out some "unfun" interactions from the start. E.G., it rules out having enemies target your gear in most cases, and also keeps people from litigating what happens to a metal table in the area of a burning hands spell (because it affects creatures).

Yeah, Timber Sentinel is the issue. Were I to decide to nerf it, that's where I'd be aiming.

The issue is that if you do it at the Protector Tree level, you've also nerfed the spell when its cast by a spellcaster at a substantially higher cost (since casting this from a slot 5 ranks below max is basically a waste of a turn). That's part of what makes this a weird situation: the spell as a spell really isn't a big deal. It was just repurposed as a class ability where it got a lot stronger as a result.

Yeah. I agree.

As an aside, Timber Sentinel is also very strange in that it has a kind of narrative power the game has desperately tried to curb. Create Water isn't a cantrip anymore, so that you can't spend all day making a lake... but a kineticist can absolutely spend all day repopulating a forest. That's kind of silly, isn't it?

I don't really know where I'm going with this post, heh. Just musing on the thought process of the two viewpoints, I suppose. This feels like one where it'll be hard to change anyone's mind because what's different is the thought process involved in getting to a conclusion.

I definitely enjoyed reading it. I don't think I'm likely to agree with Trip on this, but it is a fun discussion.

I know you're trying to target low level casters without buffing higher level casters, but making an option do less as you level just doesn't feel good. I'm not a fan of that.

What kind of save penalty are you trying to apply? If it's a circumstance bonus, that's alarmingly strong for a skill action—but if it's a status penalty, now it doesn't stack with frightened, clumsy, etc. and looks much less attractive. The typing really matters here.

Buffing RK like this doesn't buff INT characters; it buffs RK characters. Those are very different categories. INT characters may not be able to usefully RK on a creature depending on the skills they choose. And as someone else pointed out, this can also make Trip/Grapple builds stronger, as well.

How're you handling skills with RK as a rider, like Magus's analysis? Does that still provide a save penalty?

If this is the kind of approach you want to take, I honestly think you're better off just giving every caster in the game Vision of Weakness for free, or at least adding it as a class feat for every caster and possibly removing the cursebound condition from it. +2 status to spell attack rolls is incidentally pretty good at buffing low level casters while not affecting high level casters very much, as well.

===

This also won't improve the low level caster experience as much as you might expect. The biggest problem with the low level caster experience is that everything dies instantly and you have less time to affect fights. A slight adjustment to saves won't change that.

The "enemies make their saves too often" experience is also usually a result of throwing too many higher level enemies at low level parties. It's an low level encounter design issue as much as a caster issue.

Yeah I've been to this rodeo before. ;) Of course, it's not really about Protector Tree at all: back when it was just a spell, people didn't really think that hard about how it worked because it has to be heightened to keep up so it feels fine (or even rather weak considering what that spell slot could be doing). A max rank spell slot designed to protect people that doesn't protect the caster is... not good.

But then Timber Sentinel happened, and that's when these discussions all picked up steam. People largely actually want to nerf Timber Sentinel but don't see a way to do it without house rules, so instead Protector Tree gets targeted on what amounts to a technicality which makes an already meh spell pretty terrible.

I just never saw it as particularly strong past the early levels to begin with. A kineticist spending 2A on timber sentinel at level 11 to prevent 60 damage is not good; that's usually worse than using a higher level impulse for utility or AoE damage. A kineticist spending 2A at level 1 or 3 to prevent 10 or 20 damage, though? That could stop someone from hitting the floor the next time they get hit, so that's pretty good. Archetyping for Timber Sentinel also usually means that you're either using Timber Sentinel instead of casting a higher rank spell (which is usually bad) or instead of just doing your job as a martial (which is also usually bad because you're important to winning the damage race). My read has been that most people play at low levels, so people have an inflated opinion of Timber Sentinel.

EDIT: Protector Tree also falls off really fast because its AC is 10 and doesn't scale, and you just start being able to obliterate it with easy crits.

This whole thing only bothers me because "it means the tree's ally, not you!" is such a strange reading. If someone thought that Protector Tree was itself just written wrong, or was poorly balanced and wanted to change it, that'd be something else entirely. But I just don't see how to get to "the spell protects the caster" by RAW. And Trip seems to be arguing that the RAW is that it's talking about the tree's ally. So I'm left scratching my head. That's the part that especially bothers me.

I guess I'm also kind of baffled by the assertion that Protector Tree creates a animated tree creature. That's just... not necessary for the spell to work, and not said by the spell.

This stuff is more "I'm really nitpicky about semantics (both as part of who I am and because my degree trained me to be)" than it is that I think it's essential to keep Timber Sentinel from protecting the caster. I do think it's too strong at early levels if it protects the caster, but it falls off pretty fast. If it turned out I was wrong and Timber Sentinel protected the caster, it wouldn't be the end of the world.

Which is a waaay worse pain than that of just reading some text, getting the wrong conclusion, and taking the ego hit of being corrected later.

...Haven't I already conceded I was probably wrong once already in this topic, in addition to elsewhere in other conversations? For example, I already said that I probably value single feat dip amped guidance less than is warranted because of both choices from other players in games I've played in (like aggressively pursuing status bonuses) and some house rules at tables I've played in and ran at (like giving out hero points more often, and usually running with FA) leading me to misjudge its utility. And the topic about subordinate strikes also had me change my rules interpretation via the discussion in the topic.

I do take people's arguments seriously. There wouldn't be a point in participating in these conversations otherwise. I care a lot more about having a well-supported opinion than I care about being the one who "wins" an argument. I'm just pressing really hard on this reading of Protector Tree because it doesn't click for me at all.

Avenging Wildwood is templated from Protector Tree, sure. But I'd also say Avenging Wildwood is miserably poorly written and misunderstands how Protector Tree functioned—or is at least wholly disanalogous to Protector Tree. Avenging Wildwood acts like some kind of psuedosummon that doesn't follow the standard summoning rules, and it's left in this horrifically defined space where it's not really a creature and not really an object and not really a summon either despite acting like most of the above in some way. If the spell were written normally, the tree would either be an unattended object (like I think protector tree is), or would have the minion trait and act like a summon, or wouldn't have any statistics at all and wouldn't be something enemies can interact with (like the "swarm" from animated assault, or something with the Incarnate tag). Instead, it's none of the above, and is its own arbitrary rules object that works in ways implied only by the statistics it has. Given this, it's pretty easy to believe the spell is poorly written and just says "but allies can pass through its space" when it should say "but you and your allies can pass through its space." That's small peanuts compared to the most fundamental parts of how the spell works not fitting into any established framework for how an effect like this should function.

On top of that, it's pretty clear from the lack of statistics other than AC and HP in Protector Tree that the tree is supposed to be treated as an unattended object. And the lack of AC scaling on the tree makes this even clearer. You'll note that Protector Tree also makes no mention at all of animating the tree, either. Avenging Wildwood is... something else entirely.

===

I'll also note that basically every time you should be included in an effect that targets your allies, the game says "you and your allies." As an example of a similar effect to protector tree that explicitly includes (or in this case excludes) the caster alongside their allies, you have Bridge of Vines: https://2e.aonprd.com/Spells.aspx?ID=2294

Bind Undead, you'll note, is an even better case for arguing that the "tree's allies" reading would be a nonstandard stretch; it could choose to refer to the minion's allies, but instead uses the standard "you or an ally" phrasing. The same goes for Summon Deific Herald, which summons an incarnate entity and then refers to "you and your allies."

There's just no reason to think the game really meant to refer to the tree's allies without the game explicitly telling you. It goes against the way basically every other well-written spell is worded.

If the spell was intended to exclude the caster, it would not say "protect you from" and instead would say "protect your allies from"

A Medium tree suddenly grows in an unoccupied square within range. The tree has AC 10 and 10 Hit Points. Whenever an ally adjacent to the tree is hit by a Strike, the tree interposes its branches and takes the damage first. Any additional damage beyond what it takes to reduce the tree to 0 Hit Points is dealt to the original target. The tree isn't large enough to impede movement through its square.

There is nothing in the text of protector tree that implies it protects the caster. There is nothing that implies "ally" means "ally" of a magically animated object here. There's no reason to think the reading deviates from the normal reading of ally, and a whole lot of reason to think it doesn't because of how strong the spell is if it also protects the caster.

The Ghoran ancestry feat is more evidence than the person who wrote the Ghoran feat runs Protector Tree wrong and it slipped past QA than anything else. It wouldn't be weird for it to; I've seen more people who assume you count as your own ally in pf2e than people who know you don't.

EDIT: To be extra clear, I think it would need some wording like this for me to think it protected the caster on the grounds of it checking the tree's ally:

"...You and your allies count as the tree's allies. Whenever an ally of the tree is adjacent to the tree and hit by a strike..."

There is nothing even remotely close to that, so it makes more sense to assume it uses the default syntax and reference for a spell, where "ally" means the caster's ally.

I do not like ramp up abilities. Too much movement, too many other PCs doing their stuff, and targets often don't last long with the crit rules whether a spell crit or melee crit from multiple party members launching attacks.

Yeah, same. Ramp-up works if fights are long enough that the thing you're ramping up to still has a significant impact; however, the early turns of an encounter are the ones that matter most here because fights don't go on long enough for it to be otherwise. It's not great to spend a few turns getting ready for your big attack, only to realize that the fight is mostly decided by the time it goes off—and that's exactly what'd happen with backloaded abilities in this game.

Dragged out fights can exist in this game, but they're so far outside the normal course of play that designing backloaded abilities for them is a rough sell. And sustained spells kind of fill that backload niche better anyways; you do get progressively more value from them the longer a fight drags out.

Right, so I'm a GM of multiple games and get to see how much people are failing by when things actually matter. Someone's estimations are off here, but they're not mine. This is a popular ability for a reason: it makes a huge difference when it matters and it matters frequently.

I'm running once a week right now, and it's barely coming up more than once a session. I can't argue with it coming up more often for your table, though.

I will say that I'm running below level 11, which skews the result a bit.

Aid has a massively higher cost in that it takes both an action and a reaction, and it often makes absolutely no difference to the outcome whatsoever despite all that investment.

Investing an action and a reaction to do nothing is far worse action economy than having a good third action option in the first place that always makes an impact and having a reaction left available for whatever reactions you have, of which Amp Guidance might be one.

There are very few third actions that have guaranteed outcomes, and the ones that do aren't usually that large unless you spend resources. Glimpse Weakness is the first one that comes to mind in context, though I also brought up force fang earlier. It's mostly 1A casts that have significant reliability.

A typical 1A skill action will give a -1 to an enemy on a success and a -2 on a crit success for about a round, and then that penalty may or may not be relevant. Given the high bonuses available from aid and the low DC, plus the bonus typing, I think aid is pretty competitive with a lot of non-cast uses of a third action and reaction, as well.

We're also comparing aid to the options on a character that lacks a strong reaction economy anyways. If your options are committing to aid and floating your reaction for a chance at an amped guidance that might not come, that's pretty different than trying to aid on a character with stronger reactions. (Also worth noting you aren't required to aid if you spend an action preparing and have something better come up, even if it makes the third action worthless.)

You seem really fixated on "it can cause crits", while somehow completely ignoring the flipside of "it can prevent crits." Turning a crit fail that would shut down/control/drop/kill an ally into a failure that doesn't is a massive swing to a battle. You also get to apply this where it matters, when it matters, instead of having to guess ahead of time what the enemy might do and who is going to roll a 2 on their save.

Preventing critical failures is good, but that's the primary use of hero points, and a lot of critical failures aren't fixed with a +1 anyways because they're the result of a nat 1 success downgrade.

I do think this points out something I hadn't considered, though, which is that the value of amped guidance is actually somewhat lower the more often hero points are given out. I do give out hero points at a rate that's on the higher end.

Counter Performance is much more specific on what it can be used on and also requires Performance, which for most characters is not worth investing into at all. It's a great ability when its relevant, but hanging an entire argument on "this doesn't actually come up that often" only to turn around and offer an example of something that comes up even less often doesn't really make sense.

I think for me, this was more about contrasting the party utility of a larger ~3 feat investment into bard (which is usually high) vs. the drought of usefulness for psychic after your initial investment. I think it's relevant for comparing the value of the archetypes overall, but you're right that if we're focusing solely on the dedication, it's a moot point.

Definitely a blind spot. Any class that has good feats but has a dead/weak level or even just a level with something you can live without benefits immensely from an archetype like Psychic. Magus is the obvious one, but it works on a lot of classes. Like it's a great pick on Oracle which has a bunch of stuff you likely want but it's easy to find a single feat level to go pick up an Amp and never think about it again (like on any mystery with a lousy Advanced focus spell or where the level 10 Cursebound feat is poor).

Wouldn't you just opt into Quickened Casting on Oracle at 10 instead? I'm also a bit lost on the magus point, because you need at least two feats invested into an otherwise weak conscious mind to get amped imaginary weapon, and the earliest you can get it is 6 (which competes with reactive strike).

Still, point taken.

An aside, but I was asking myself if Oracle benefitted that much from the additional reaction, and saw that battlecry recently did add an odd amount of strength to the reaction options on the divine list. Curse of Recoil is a much less situational filler reaction than schadenfreude, and while Helpful Reload isn't useful in every party, I'm shocked that spell was even printed; the effect it has on gunslinger action economy is profound. (Martyr's Intervention remains quite strong if allowed as well, of course.) Oracle also doesn't seem to have as easy access to some of the better reaction domain spells as one would've hoped, since they seem to require mysteries that look less attractive. Battlecry really did just buff the divine list a bizarre amount, in that regard.

There's a few dedications like this. Exemplar Dedication is one of the single best feats for most of the martials in the game. Anyone who wasn't going to archetype otherwise can find a feat to invest into that for far more return than they'd otherwise get. Alchemist is the same because of just how much raw utility Quick Alchemy gives, even if you never put another feat into it. Psychic is in this situation too.

Point especially taken wrt exemplar. Literally forgot about that one because it's just banned at my table.

Looking over these feats, it seems like a throughline in a lot of the outlier dedications is that they give one skill training (or none) instead of two, and often give it in a less valuable skill for combat. I can sort of see the logic at low levels, but the value of skill training falls off a cliff at higher levels, and the value of alchemy benefits, ikon damage, and amped guidance does not.

You're ignoring other Amp options and also now saying that what you said was an easy to get bonus is now super high value.

Something can be both easy to get or common, and also high value. E.G., the first way to heal to full out of combat is extremely high value for a party, and there are a lot of ways to achieve it. Additional out-of-combat healing drops off sharply in value after you have it.

By the same token, amped guidance is less valuable the fewer rolls it can proc on. If your party prioritizes getting status bonuses, then amped guidance loses rolls it can target.

===

Looking over all this, I think you've generally made fair points. I think a lot of how my tables and the tables I play at run does devalue amped guidance, somewhat. That's leading me to be more down on the option than might be warranted.

===

Cantrip expansion sucks. It’s a bad feat and shouldn’t be used as a power yardstick here.

I brought it up earlier just to point out how bad some class feats are compared to archetype feats. I don't think anyone was really using it as a benchmark for where power "should" be.

snip

The flipside of being "unable to waste reactions" on amped guidance is that it also doesn't actually give you a use for your reaction unless someone fails by 1 or 2. The amount of times it's useful is lower than I think people give it credit for. Between it not stacking with other status bonuses and the limited opportunities to use it, you can go an entire night without amped guidance actually giving you a way to use your action economy. This a downside, not a benefit. If you're trying to fix your action economy via archetyping, an inconsistent reaction isn't actually that good. You want something you can consistently take advantage of.

In a lot of cases, spending an action and reaction to aid every turn would actually be more effective than sitting back and waiting to use amped guidance. That gives a larger bonus with a typing that's much more difficult to come by and can cause crits, even if it also uses your third action. It's a bit apples and oranges because of that, I admit, but aid is an option everyone has to use reactions without spending any feat slots.

Amped guidance is very visible (because you only use it when it changes an outcome) but also not very effective (because it can only change one outcome between turns and can never upgrade a success to a crit). I think the visibility tends to make people overestimate how good it is and how often the ability actually procs.

I agree that it's particularly clutch when it works on saves, but there are other reactions you can poach later on in other archetypes that are good for that, and they can be more effective within their niche. (Bard counterperform comes to mind.)

I think I'd also note that I can't think of any character I've made that would only want to dip into an archetype for a single feat, outside of a character taking multitalented as their level 9 ancestry feat. This could be a personal blind spot, but most characters I've seen either have class feats they don't want to skip; have pretty specific things they want to get out of archetyping (e.g. reactive strike or crashing slam) and need to spend multiple feats to get them anyways; or have bad enough class feats that they're glad to take two more psychic archetype feats before jumping ship to another archetype. Psychic dedication becomes less and less competitive with higher level class feats and higher level dedications as your character levels, as well. I just can't think of a character that has their level 2 class feat free that also has the rest of their feats tied up so tight that a single dip into psychic dedication is slam dunk.

Imagine if just taking the Bard Dedication gave you courageous anthem and Lingering Composition in a single feat basically. That's kind of what Psychic does.

This is wildly overestimating the value of amped guidance. Spending a single action and a focus point for 3 turns of +1 status to hit and damage and +1 status to saves against fear across the entire party is far higher value than being able to upgrade a single failure to a success (or critical failure to failure) per round at the cost of one reaction per round if the miss is within 1/2 of hitting depending on level.

I think the major problems with psychic dedication are more general game design issues over issues with the archetype. I see three big issues:

1) Too many classes just get poor reaction options, and psychic is a bandaid fix that can be shoehorned onto any class that uses INT or CHA for anything.

2) Too many class feats fail to be competitive with archetype dedications and archetype feats in general. Basically any caster dedication is better than cantrip expansion, for example. The 3-feat lockout on other archetypes doesn't counterbalance the power of most dedication feats at all.

I also don't think psychic dedication is significantly stronger than a lot of other INT/CHA archetype dedications—it's just one that gives a benefit that a wide array of characters can make use of. I'd also argue that while amped guidance is good, it's honestly not as strong as a lot of people make it out to be, since it never lets you crit and also applies a status bonus. Getting a party-wide +1 status bonus to hit is already one of the the most common party-level optimization goals, and a lot of the best buff and prebuff spells (e.g., heroism) give status bonuses as well.

3) Psychic itself is weak, so poaching the things that are good about it is just better than playing it. Psychic promotes a playstyle that is bad in the context of PF2E's design, and doesn't offer enough to compensate; psychic has poor class feats that often ask you to make your teammate's lives harder and class features that make your own life harder, so taking the good parts of the class via archetyping is quite attractive; amps don't compensate well enough for a lack of spellslots, but sure are great backup options for casters that already have a lot of spellslots; and so on.

I consider the other issues below—the ones the thread zeroes in on—to be more minor, personally.

1) It lets you accumulate focus points unusually quickly on a lot of classes. I think this is fair. It tends be very beneficial early, but significantly reduces the value of focus spell feats later.

2) Magus benefits disproportionately from psychic archetype. I consider this minor because it's frankly more of a problem with magus's design than a problem with psychic archetype in and of itself. As per why:

2a) IIRC, unamped imaginary weapon is only slightly better than gouging claw. The amp specifically is the problem, because it makes focus spells extremely competitive with slotted spellstrikes.

2b) I believe Force Fang is still a competitive use of focus points and gets more competitive the higher enemy AC is. The problem is moreso that the other conflux spells aren't great—and being able to spellstrike with amped IW then force fang to recharge (provided you contort your feat selection to do so, which isn't trivial) allows for an extremely potent one-round burst without spending permanent resources.

2c) There are still other good focus spells for magus to use, even if you disallow amped IW. The largest benefit of psychic is that it's an INT class and a lot of people like to have INT on their Magi to increase their spell saves anyways. The other favored option (champion to pick up fire ray or winter bolt, along with better armor proficiency to mitigate dumping DEX) requires CHA instead.

2d) Magus is itself a bit of a problem child, design-wise. It's a great class from a pure damage perspective, but its action economy is stilted and it has a hard time taking advantage of its own class features because spellstriking is so attractive. I suspect that amped IW would be a less attractive pickup for Magus if the class were designed better.

This rule is written notoriously poorly, as are other features. You'll note the rules example tells you that your damage type changes, but the rule itself tells you to substitute the resistance value and doesn't tell you to change the damage type. It's unfortunate.

In this case, though, you should note that void healing doesn't mean you heal when you take void damage—you still only heal from a spell if it says it heals. An enemy dhampir doesn't heal from wails of the damned, for example, and living creatures don't heal from the vitality damage portion of a Fire Kineticist's Solar Detonation. (Neither will take damage from the respective damage type, but likewise, neither will heal from it.) The confusion is understandable given how Harm and Heal 3A casts are worded, IMO. But my understanding is that "deal void damage to living" and "heal undead" are two separate functions and both don't occur unless the spell says they do.

Does feel really weird that you can just carry a whip to grant reach flanking, even though you're not proficient with it. I'm gonna remember this if I ever play another kineticist, though. It's pretty good if you're not using a hand to hold an item (or a shield, at lower levels).

tch. So you did the stick in your own bike wheel meme. How much of the table's time did you waste on this thing?

Unless you're shouting "Furious Strike!" when you do it, what is it about starting an activity that would break hidden? The GM can determine that other actions don't break hidden, and I have never encountered a GM who would make such a troll ruling that just starting an activity breaks hidden. That breaks practically everything a rogue can do and also cripples way of the sniper gunslingers so it is surely not the intended reading

I think almost everyone (though not literally everyone in every test case—see comments wrt spellstrike) feels the RAI is that you should get it. The problem is that the RAW -really- makes it seem like you don't unless the GM gives you permission, and there's no good arguments to the contrary that we've found under RAW.

I think at bottom, I at least feel that it feels weird that something so obvious requires the GM to go "yeah, you're good." That does seem to be the case, though.

Thinking about it, a lot of this is technically solved if "your next action" or "your last action" is always parsed as "the next thing you spend actions to perform" or "the last thing you spent actions to perform." As in, it only checks for named actions you paid actions to perform, not those you perform at no action cost as subordinate actions or in response to triggers as free actions or reactions.

I think this aligns with most people's intuitions here, and also doesn't require you to litigate much else about how to read the rules. It's incongruent with Trip's reading (because there's no "activity end" tags on Trip's reading), but it seems congruent with most other readings. It captures the elision of "action (cost)" and "action (activity)" that seems to occur a lot in the rules, and I can't think of any case where it produces an unclear result. It does produce the result that double-slicing while hidden won't proc off-guard, but I think most of us also agree that it proccing off-guard requires GM fiat anyways.

I feel like shroudb is making the same mistake now that they cautioned against earlier in the thread: confusing actions (things a character does) with actions (costs to perform things). This is pretty bad semantic design from the game, to be sure, even if it's inherited from older d20 systems. I wish the "actions (cost)" were called stamina or maybe even just action points instead; AP for action points is definitely a common convention in some other games I've played, and it would be much clearer to say that you get 3 AP (action points) and 1 RP (reaction point) at the start of your turn instead of three actions and a reaction.

I don't think Hide is referring to "next" or "previous" actions at all. You could easily Hide, then Step, then Strike and still benefit from off-guard.

You're correct that I probably should have said it is equivalent to something like "if you are hidden and your next action is a Strike, then..." But I figured that was implied from context.

I also figured it wasn't implied that the if statement required your last action to be sneak or hide, because the rules as written already don't require that.

I further didn't think it was necessary to translate that entire section into a string of conditionals covering all the action buckets (step/sneak/hide, strike, anything else with GM approval, anything else without GM approval) to get the point across, but I suppose this is a topic where we're already being very nitpicky about semantics and it could've spared confusion.

Witch of Miracles wrote:

...(Sneak Attack is an easy problem case for this reading, as are abilities like Debilitating Strike ...

Just as a point of note, Debilitating Strike is unaffected as it is a Free Action with a trigger that does not have the next/previous action language.

"Trigger Your Strike hits an off-guard creature and deals damage."

Per the rules on actions reactions and free actions with triggers (a la Debilitating Strike) my be used at any time, including in the middle of an activity, that the trigger is met.

That was in the context of a rules reading where you're not actually doing the Basic Action Strike as part of an activity that tells you to strike, but instead something like following the instructions of strike without actually using Strike. That would mean that you couldn't use debilitating strike because the condition isn't met (as it requires you to Strike), nor would you proc Sneak Attack (because it again mentions Strike by name).

I don't agree with the reading I was outlining here. I agree with what you're saying.

In that case, your last/previous action was a Strike, not Sudden Charge.

Yeah, that's a consequence of Trip's reading. I think it's an acceptable cost, though. I don't think the game explodes if Strike>Exacting Strike>Reap the Field is an acceptable sequence of actions for an exemplar with fighter archetype, or Skirmish Strike>Reap the Field is an acceptable sequence of actions provided that you step first. I'm not going to get annoyed if someone says those don't work at their table, though.

This reading also spares you from trying to divine if there's a difference between the requirement on followup attack and something like Reap the Field, which is nice.

If you want to think of it as [Sudden Charge starts> Stride > Stride > Strike > Sudden Charge ends] instead to let sudden strike be the last action, you can. I don't like it as much right now. But the most important thing is just that everyone agrees the game actually tells you to use Strike, so you get Sneak Attack damage and can proc Debilitating Strike and so on when an activity tells you to strike.

===

All this being said, though, there is one action I can find that doesn't play completely nice with the "the last action is the last subordinate action executed" reading: Cratering Drop. https://2e.aonprd.com/Feats.aspx?ID=5540 It explicitly asks for your last action to be the activity "Pluck from the Sky" using a melee strike. Sure, the last thing Pluck from the Sky asks you to do is have the enemy make a reflex save if the strike deals damage, so maybe there's an argument "pluck from the sky" happens after the strike or something. But then we end up with the unintuitive and seemingly wrong result that if you use Pluck from the Sky and miss, your last action was a missed strike; but if you hit, your last action was Pluck from the Sky. I don't think this is really acceptable.

Weirdly, though, this ability would also violate the containerization reading. It cares about the kind of strike used inside the activity, which doesn't square with how containerization is normally explained. It's not like anyone is going to parse what the ability does incorrectly because of it, mind.

Anyways, this ability—one-off though it may be—supports the assertion there's an activity end tag. Going to think on if this is illustrative or an outlier.

Witch of Miracles wrote:

...So Double Strike should proc off-guard, because I get all the effects of a named action strike, including effects that ask for Strike by name, even though using an activity isn't the same as using its subordinate actions and you don't think you ever actually use a strike and just drop in the rules text of strike?

Yes. That is what that sidebar says.

While resolving the subordinate actions treat them as if you used the referenced base action, potentially modified by the activity. While not resolving the subordinate actions only the activity exists.

That's why a subordinate strike gets everything a regular strike gets, but when not resolving it the only thing that you can refer to is the activity, so next/previous action and the like won't see the subordinate actions.

No. This contradicts the text.

Subordinate Actions
Using an activity is not the same as using any of its subordinate actions. For example, the quickened condition you get from the haste spell lets you spend an extra action each turn to Stride or Strike, but you couldn't use the extra action for an activity that includes a Stride or Strike. As another example, if you used an action that specified, “If the next action you use is a Strike,” an activity that includes a Strike wouldn't count, because the next thing you are doing is starting an activity, not using the Strike basic action."

Hide
"You cease being hidden if you do anything except Hide, Sneak, or Step. If you attempt to Strike a creature, the creature remains off-guard against that attack, and you then become observed. If you do anything else, you become observed just before you act unless the GM determines otherwise."

The Hide text specifies Strikes. You are starting the activity Double Slice; Double Slice is in the "anything else" bucket, because Double Slice is not the same as using strike twice; therefore it requires GM fiat for the enemy to be off-guard to the first hit of double slice. I think every GM should rule that the enemy is off-guard to the first hit of double slice, mind, but it is GM fiat by the rules.

If you attempt to Strike a creature, the creature remains off-guard against that attack, and you then become observed.

This part of the Hide text is functionally equivalent to saying, "when hidden, if your next action is a Strike, the enemy is off-guard to that attack and then you are observed." It is effectively identical to the example given in the sidebar. This is, in fact, the problem this whole thread is about.

===

I should also note that what I said there is internally inconsistent, to point out that shroudb's position there is internally inconsistent. It requires Double Slice to both be treated as equivalent to strike and not be treated as equivalent to strike.

My best guess is they mixed up "effect" and "affect." What they said makes some sense if you had read "This subordinate action still has its normal traits and effects" as "This subordinate action still has its normal traits and whatever affects the subordinate action normally still does." That's not what the rules say, though.

How about you answering my question for a change though:

If you can only do an action after an activity completely finishes, if a subordinate action is the same as the base action, how do you start it?

This only makes sense if you think the game isn't telling you to do other actions as part of activities to begin with. But it is! It even says you don't need to spend more actions or reactions to do so (which effectively makes them free actions, one might note).

An action might allow you to use a simpler action—usually one of the Basic Actions—in a different circumstance or with different effects. This subordinate action still has its normal traits and effects, but it's modified in any ways listed in the larger action. For example, an activity that tells you to Stride up to half your Speed alters the normal distance you can move in a Stride. The Stride would still have the move trait, would still trigger reactions that occur based on movement, and so on. The subordinate action doesn't gain any of the traits of the larger action unless specified. The action that allows you to use a subordinate action doesn't require you to spend more actions or reactions to do so; that cost is already factored in.

Note the wording is literally saying you use the basic action.

And since you get ALL the effects of stuff that affect "the named Action Strike" with a subordinate action Strike, you get it.

...So Double Strike should proc off-guard, because I get all the effects of a named action strike, including effects that ask for Strike by name, even though using an activity isn't the same as using its subordinate actions and you don't think you ever actually use a strike and just drop in the rules text of strike?

Witch of Miracles wrote:

shroudb wrote:

Less than a shorthand, and more like building blocks, yes.

That's why activities often change how the base actions behave (adding dices, speed, removing stuff, adding and removing traits from them, and etc).

The game has a handful of said base actions, and uses them to build the more complex actions (activities). All the rules are easily referenced this way without having to repeat them all the time, and stuff that are suppossed to influece the base parts of the game (like weapon specialization adding damage to your attacks as an example) can be easily accounted for in current and future content with extreme ease.

The thing is, either the game is telling you to use a strike when it says to strike, or it isn't. You are saying that the game is not telling you to use a Strike when it says to use a strike inside of an activity; it is saying this in lieu of copypasting the rules for Strike into every activity.

So you would say Rogue does not get Sneak Attack on any attack but a Basic Action Strike. This is a necessary consequence of your position. Do you agree with that statement?

No, I don't, because thankfully the game tells us to

Quote:

This subordinate action still has its normal traits and effects

And sneak attack is an effect of a Strike.

This is not logically available to you. Strike is a named action. Sneak Attack specifically refers to a Strike by name. It does not refer to any trait or effect of Strike; it specifically says Strike.

When your enemy can't properly defend itself, you take advantage to deal extra damage. If you Strike a creature that has the off-guard condition with an agile or finesse melee weapon, an agile or finesse unarmed attack, a ranged weapon attack, or a ranged unarmed attack, you deal an extra 1d6 precision damage. For a ranged attack with a thrown melee weapon, that weapon must also be agile or finesse.

Less than a shorthand, and more like building blocks, yes.

That's why activities often change how the base actions behave (adding dices, speed, removing stuff, adding and removing traits from them, and etc).

The game has a handful of said base actions, and uses them to build the more complex actions (activities). All the rules are easily referenced this way without having to repeat them all the time, and stuff that are suppossed to influece the base parts of the game (like weapon specialization adding damage to your attacks as an example) can be easily accounted for in current and future content with extreme ease.

The thing is, either the game is telling you to use a strike when it says to strike, or it isn't. You are saying that the game is not telling you to use a Strike when it says to use a strike inside of an activity; it is saying this in lieu of copypasting the rules for Strike into every relevant activity.

So you would say Rogue does not get Sneak Attack on any attack but a Basic Action Strike. This is a necessary consequence of your position. Do you agree with that statement?

You do not have to invent anything though?

That's the whole reasons that paizo gave us the building blocks called base actions. To make more complex maneuver without hving to have a page describing each one of them.

In short:

My position is that when I use "Act together" is use the action called "Act together". That action may tell me to attack...

So, on what you are saying, Skirmish Strike says to Step and Strike. ("Your feet and weapon move in tandem. Either Step and then Strike, or Strike and then Step.") But what you should read is not what is written, but something like this: "Your feet and weapon move in tandem. Either [perform the rules text of Step] and then [perform the rules text of Strike], or [perform the rules text of Strike] then [perform the rules text of Step]."

This is literally what I mean when I say that you have to be saying the game uses Strike "as a shorthand."

At the end of the Sudden Charge, did you finish a Strike, or did you finish a Sudden charge? Because you need to finish the Sudden Charge to do your next action.

Performing Sudden Charge is paying the 2A to invoke sudden charge and then performing the subordinate actions of stepping twice and then attacking. Once you've attacked, you've done everything sudden charge told you to do, so you've completed it. There's no need to add a "sudden charge finished" step. Doing the last thing the activity tells you to is finishing the activity in my book.

That is the rule. That is the containerization. An activity contains its subordinate actions and using the activity is not the same as using any of its subordinate actions: not its first one; and not its last one.

That alone doesn't imply any containerization. It just says that activities are not identical to the actions they ask you to perform, that using Flurry of Blows is not identical to striking two times in a row. Containerization is an entirely separate assertion from the identity assertion made here.

snip

So here's the thing. To understand your position at all, I need you to answer one question. Do you believe you actually perform a Strike when the game tells you to in an activity like Skirmish Strike? Or do you think the game just tells you to perform a strike "as a shorthand" and you never perform any action that is named "Strike"?

I am taking the game at face value. The abilities tell you to perform a Strike; I say you perform a Strike. Do you think you are not performing a Strike when activities include a Strike? Do you think that a Summoner using a Strike as part of Act Together is not actually performing a Strike? Do you think Rogue gets sneak attack on Skirmish Strike if they have flanking?

The way Trip parses this, for better or worse, gives an extremely clean answer to these questions: yes, you do perform a strike, and it is a strike just like every other strike. This just doesn't. It requires you to invent the existence of Strike* and divine when the rules mean to include strikes-in-activities when they say Strike, and when they don't.

The rules clearly separate the effects of a subordinate action from the subordinate action actually counting as said actions.

This is, unfortunately, a misunderstanding—one I shared. The exact wording is "using an activity is not the same as using any of its subordinate actions." All it ultimately means is that using Skirmish Strike, for example, is different from doing Step and then Strike without using it. It is natural to further read into this that "using Skirmish Strike means you don't actually do a step or a strike," but that would contradict the plain rules text of Skirmish Strike that says you step and strike.

Using an activity can modify its subordinate actions, as per the rules text. The correct way to see this is that you use Skirmish Strike for 1A, and it instructs you to step and then strike or strike and then step, modifying their costs so you're doing both as a free action, and not allowing you to do anything else until you've used a step and a strike. This is implicit rules text in basically every activity; you pay the cost of the activity, perform the "activity," and then do any further subordinate actions it instructs you to do as a free action without doing anything else until you've finished. (The "without doing anything else until you've finished" part is supported by the statement on combined activities in spellstrike, and also just by common sense—you would look at someone funny if they tried to feint inbetween stepping and striking during skirmish strike.)

On this reading, the reason the free action strike doesn't count as a strike for haste is fairly simple: the Skirmish Strike modified the cost of your strike to be a free action already.

If you don't use this reading, you're stuck with the weird thing I proposed earlier, where there is Strike and Strike*; Strike is the standalone action; Strike* is the one used in activities; and you have to guess if the rules text wants things to apply to Strike and Strike* or only Strike every time it says "Strike." Or your other alternative is a kind of containerization that prevents the game from looking forwards into the contents of activities, but may or may not allow it to look backwards into them depending on your view of the RAW, while executing the game normally inside the container. Both can be made to work with enough ad hoc adjustments, but they just don't play nice with other rules text and require more band-aid fixes.

-The reading where there's Strike and Strike* requires an unacceptable amount of houseruling and makes it impossible to tell what's meant when the rules say Strike. (Sneak Attack is an easy problem case for this reading, as are abilities like Debilitating Strike. Summoner's Act Together would also be a nightmare for this, because it would make it so Stride used with Act Together is not the same as Stride used outside of it.) It also contradicts the plain text of every combined activity: you are literally striking.

-The reading where you containerize doesn't square with the plain text in the subordinate actions bar that says "...the next thing you are doing is starting an activity, not using the Strike basic action." This makes it clear that the activity itself is invoked (that you're "starting an activity"). It is also worth noting what it doesn't say: it doesn't say that an ability that says "your next Strike gains X benefit," like the Nagaji poison ancestry feat, won't work with the strike inside of Skirmish Strike; and it is not stated that backwards referents (like debilitating strike) don't work, though you'd be forgiven for thinking they wouldn't based on context. (One should note, though, that they assiduously avoided the "your next strike" wording in favor of almost exclusively saying "your next attack," probably to ensure no one would ask if "your next strike gains X" and "if your next action is to strike" worked differently. The Nagaji poison feat is the only time I saw "your next Strike" used.)

The "Activities are a flat chain of actions" ruling that actually exists in the text results in the expected behavior. No need to invent mechanical scope voodoo.

That's not clear, because the flat chain of actions reading makes several wordings a nightmare to parse.

-anything that says "your target is off-guard against your next Strike" probably works with activities including a subordinate strike, though I won't guarantee it
-anything that says "if your next action is a strike" does not work with a subordinate strike
-anything that says "if you attempt to Strike a target before the end of your next turn"... might work with a subordinate strike? Unclear if this is asking for the strike basic action or not!
-anything that says "the target is off-guard to your next melee attack" (e.g., feint) definitely works with subordinate strikes

I also really hate this because it makes activities parse differently than basic actions. On your reading, Strike just causes you to perform strike; there is no "starting to strike" and then performing Strike. It's extremely unintuitive to have basic actions work normally, but then have activities like spellstrike produce some phantom action called "spellstrike" with no effect before executing cast a spell and a melee attack.

It also is a real nightmare when abilities like debilitating strike (that refer backwards) interact with subordinate strikes, on your reading. It's not at all clear if the precondition is asking for the Strike action specifically or if subordinate strikes qualify. It's especially bad if there's a way to have that kind of ability proc in the middle of something like double slice; it's not clear that's even allowed, because the rules for spellstrike imply you "typically" can't interrupt the flow of subordinate actions in a "combined activity." (But maybe that's because most, though not all, spellshapes are 1A, even though there are a few that are a free action that a magus might be able to use.) And "combined activity" seems to be mentioned... in spellstrike and nowhere else. Good luck divining the intent here!

Furthermore, I also have no clue why they're called "subordinate actions" if they're not a part of the activity. What you're saying is that flurry of blows is three things, not two things: it is flurry of blows, then two strikes, and the strikes themselves are not in fact flurry of blows, just something you do after flurry as a free action. Or perhaps it is four; flurry, strike, strike, flurry ends? This is just a ridiculous reading to me on every level. It makes much more sense to say Flurry of Blows is just the name for doing two strikes in one action, and that Flurry of Blows is not named Strike, so things that ask for Strike don't work with flurry of blows. And it's completely asymmetrical with any activity that doesn't invoke a subordinate action! It's awful.

===

...Okay, I typed all that, and the more I read how the system is literally written, the more I'm inclined to agree it's actually written the way I hate. What a miserably awful way to design your semantics. Every activity with subordinate actions has a phantom action with the action cost of the activity up front to block forward referents that care about your next action, but only the activities with subordinate actions? ...But it doesn't block backwards reference? ...And Flurry of Blows actually isn't the strikes, which are their own action, just the phantom action before said strikes? Really? Really? REALLY? This is so awful. It feels absurd to read it. And it looks correct as a description of the system as written, especially considering the lengths they go to in order to avoid phrases like "your next strike" from occurring in the rules. Why would anyone ever design something this ugly? I'm gonna scream into a jar and bottle it for my players to open on Halloween.

I have seen zero text that instructs activities to "containerize" and isolate their sub-actions in any way. If that happened, then we would no longer be able to trigger Sneak Attack off of sub-Strikes, etc.

FWIW, I disagree with this assessment. I think you can sandbox things that happen inside activities, apply the usual rules to them as they occur inside the activity, and also have the game act like only the container activity and none of its subordinate actions occurred once the container is completely resolved. It's like a box the game opens only during the activity and shuts afterwards. The main game thread doesn't have permission to view what happened inside the activity.

I also don't see how this prevents sneak attack from happening on skirmish strike, because "strike" would occur in the sandboxed container, and the rules are ran the same in the container as outside of it.

Assuming that activities are names for sequences of actions executed in a sandboxed sort of container also produces all of the results that most of us in this thread think are silly, but seem correct by RAW. Any further divergence depends on whether you think the container itself counts as an action that is executed, or if it's just a name and isn't executed outside of adding steps as needed for the trait tags. I think there's an argument both ways; the difference is trivial outside of cases that care about your next action.

===

I think the main thing that's annoying is that if you don't intuit this weird containerization (that I'm not entirely sure is coherent, though I think it is), you're left with a different, very weird reading of the rules:

Every "basic" action has a "basic*" equivalent, which is the exact same as the basic version except it's the one used inside activities and has a different name. Because of this, actions inside activities aren't technically identical to actions done outside of them. Things that ask for Strike by name, like haste, don't ask for Strike*, so Strike* (the version inside activities) doesn't proc it.

Now, the game -reads- like this should be true. But the game never ever uses the term Strike*, or Step*, or Stride*, and so on. Sometimes the game says strike when it means Strike only (Haste), and sometimes says it Strike when it means both Strike and Strike* (Sneak Attack)—but the only way to guess at which is intended is which makes sense in context. And that's MISERABLE. But it does produce a coherent reading of the game. You just have to determine whether the rules mean "strike only" or "strike and strike* both" whenever they mention strike by name.

===

...I feel like I should stare at this post another hour to make it more coherent or clearer, but I'm just going to move on and hope this makes sense.

Everyone is happy... except for people who care less about the exact rule and more about it being consistent.

I apparently don't speak for everyone, since I don't speak for you and presumably plenty of others. I'll concede that. But saying it'd be wrong to have hardcoded the rule because some tables dislike having it hardcoded is, in itself, ignoring a different preference axis (that of table-to-table consistency). If you claim hardcoding it is wrong because tables disagree on possible rulings and they should have freedom, I'm not sure why I can't claim with equal force that leaving it fiat is wrong because tables will produce different rulings and players should be able to expect consistency. We clearly have different and incompatible values on this front.

For what it's worth, I care more about immersion than consistency in some games—but not in this game, both because it already goes to lengths to define most interactions and also because the game is already poor at being a narrativist game for me. A good table can push past it and have great RP, but the system actively works against it in my experience. So I'd rather have the game invest in what it's already good at (defining interactions, providing a more consistent experience table-to-table, consistent combat balance) than have it invest in immersion to the expense of those things.

Witch of Miracles wrote:

I have a similar issue with reactions before your first turn. There's no default. GM always determines it based on the circumstances, but there is no GM guidance at all about what those circumstances are. And it definitely matters because the new Guardian class has a level 1 feature directly based on this rule that is pure GM fiat.

Yeah, I also strongly dislike that; it really affects the difficulty of some combats. I've seen some DMs default to yes, and some default to no. I tend to run it as "if this combat would have a surprise round in 1E, you only get a reaction at combat start if you would've acted in the surprise round; if it wouldn't have a surprise round, you always start with your reaction." (And I fully admit that, itself, is extremely idiosyncratic and can see people having problems with it.)

This particular activity couldn't work with your ruling, since you need to create a diversion, which is not inherently stealthy (but can be depending on what the player does) before the strike.

The ruling producing a weird result sometimes doesn't make it inherently bad. And this logic applies to spellstrike as well (casting a spell is loud!) but I allow spellstrike to get off-guard from hidden. It doesn't bother me. "GM fiat lets it be more simulationist or make more narrative sense" isn't really an angle I care about in this game at all, to be honest. I'd prefer consistency in table rulings.

This example is also poor, because stumbling feint gives a better result for a monk than create a diversion>flurry would anyways. (Stumbling Feint is also weird for this insofar as it lets you feint as a free action before the flurry, not as part of the same activity as the flurry. But yeah.)

I think it's pretty easy to say "or the first strike of any activity containing a strike" in that rules text and then add a trait you can add to attack activities when they need to be excluded, yes.

The text doesn't need to cover literally everything, and that was never my position. My position is that rules questions of importance to basic combat routines should probably not be GM fiat. We've already had one person disagree that create a diversion>spellstrike should grant off-guard, which is evidence there's room for table variance here on fairly consequential questions.

Also, this "GM discretion is the RAW" thing I keep seeing pop up is driving me nuts. "Rules as written" means "as written." If it gives the GM carte blanche, no rules are actually written about how to handle it, and there is no RAW. It's a semantic thing, yes. But why RAW even matters, and why anyone cares about it, is that playing RAW significantly reduces table variance. It is literally about playing the game "as written," in a way that means people trying the same things get the same results by following what's indicated by the rules text. "GM discretion is RAW" is useless for reducing table variance in this way. It's the kind of response that follows the letter of the question while ignoring the spirit of it.

(And yes, people can dispute and argue about RAW. RAW is always something of an idealized construct, and discussing rules text to get at the RAW is as much an attempt at forming group consensus as an act of exegesis.)

If someone disagrees with that, please provide a practical use case for buying and using Spellstrike Ammunition on a regular basis. Or even for buying it for niche use cases.

Magus only has melee spellstrike by default; starlit span is the only subclass that has ranged spellstrike. The ammunition can be used by other magi with their backup ranged weapon.

A champion (or warpriest, or maybe even a monk) could use it with their focus spells, as well. Champion can pick up Fire Ray, right? It's expensive, but it doesn't seem awful aside from the cost. It's not a terrible trick against a flying enemy, in a pinch.

Rarity in PF2 is used for too many things (which is a different issue), but the idea is a massive improvement over PF1's "an AP published a spell that a Runelord knew 10,000 years ago which means RAW a random level 5 Wizard can just get it at level up" approach.

As much as I miss free access to some utility spells, I am very, very glad that no PF2e GM is ever going to just find out their player took Blood Money because they saw it on Nethys... and then find out oopsies, the wizard ruined the economy at level 3 with blood money and masterwork transformation haha isn't that crazy guys what a fun game c:

Witch of Miracles wrote:

These kinds of commenters aren't unique to PF2E. What's unique to PF2E, though, is the quantity and concentration of them.

It's... really not. Like even remotely. 5e forums are this way. Lancer communities are this way. PF1 and 3.5 were this way when they were more popular. Even more expressively narrative systems like World of Darkness get this kind of talk. "TTRPG company released something that's really good/really bad and people are talking about it" is just like, a standard component of discourse in these kinds of games. The variance you see is mostly a matter of how crunchy the system is, you see it more in PF2 and less in PBTA, because the system rules are more/less weighty. But like... people talking more about the crunch of a crunch heavy system than a system that isn't meant to be crunchy also is kind of a self evident revelation.

Like IDK you're just describing one of the most normal aspects of online discussion for crunchy TTRPGs and then framing it as something special about PF2. The reality is this is just how people talk about games like this. Like people complaining about how bad the 5e ranger is is basically its own meme.

Maybe you've simply had different experiences than me, but the proportion of those players to other kinds of players has been unusually high in PF2E. At no point did I say those posters don't exist elsewhere. I simply feel like they make up an unusually large percentage of participants (emphasis on unusually). My experience is that most of the people who grow really attached to PF2E and talk about it online are attached to the mechanical aspect of the game, and they're attached to it as a game. Not as many Vorthoses in online discussion, to drag out that term from discussing types of MtG players, but a whole lot of spikes.

Another way of putting it is that lot of other crunchy RPGs have commenters who're invested in the mechanics, but like the crunch as a form of simulation or representation or metaphor. Those players are disproportionately absent in 2E discussions.

This is such an oddly antagonistic way to describe the GM-player relationship.

Plus you realize a 1e GM can effortlessly stop you from doing anything they don't want to at any time too, right? You're always constrained by what the table will allow you to get away with in every system.

I want to jump into this just to say that I, personally, feel you're really underestimating the difference between the whitelist-based system in PF2E (where anything uncommon, rare, or unique is off-limits by default, and the GM whitelists options for you) and the blacklist-based system in 1E (where everything is open unless the GM says otherwise, and it's on them to tell you what's off-limits and play whack-a-mole). If you've got a girl who's been running 1E games for ten years, yeah, she'll tell you that Speak with Dead is off-limits in the murder mystery campaign, or you'll only get half-answers when you use the spell because the victim themselves doesn't know what happened. A new 1E GM won't even know what to blacklist, and the easiest blanket blacklist (allowing only PFS legal options) leaves most narratively powerful spells on the table and doesn't even ban stuff like emergency force sphere.

The "system default" is very different between the two, and that matters.

A new GM confronted with speak with dead for the first time in the middle of play in 1E is in the position of having to tell a player they can't use a spell they had no reason to think was off limits, or they're scrambling to improvise solutions to a problem they didn't know existed 30 seconds ago. A new GM in 2E can tell the player it's uncommon and they forgot to ask permission. It's been remarked on many times, but 2E's rarity system allows GMs to sort of make the system the lame hall monitor spoiling their fun instead of having the GM expend the social capital themselves.

I think there's honestly some truth to the cynical take Sigh is expressing—but in my opinion, paizo is as much an Adventure Path company as a game system company, and APs run much better when the players aren't one spell (possibly a spell published in some splatbook long after this campaign was made!) away from derailing the whole thing. To me, it's not malicious; it's a sensical way for a company that makes a lot of money off AP subscriptions to design its system. Why wouldn't you want the system those APs run within to make the APs easier to GM, with less possibility of the players doing something the AP book won't have space to cover? It makes them a better product for the person using them.

GM fiat stuff still goes into my houserule doc, like clarifications on what is and isn't allowed to aid at my table.

And I think I speak for everyone when I say that questions like "does ghost shot benefit from off-guard proc'd from covered reload?" should probably not be up to GM discretion.

I see how this is an issue and I hate it. Thanks for pointing this out. Into the houserule doc it goes, I guess.

Can't imagine they thought that a gunslinger that uses sneak and then does risky reload shouldn't get the off-guard bonus because they reloaded before shooting, but that sure does look like the case by RAW. Also can't imagine they thought a sniper that uses covered reload shouldn't be able to follow it up with Ghost Shot and get off-guard, either. Chalk it up to a case where everybody read what was intended and no one read what was actually written, I guess.

I think everyone has given reasonable advice. To go further than what everyone else has said, it's generally hard to know what to tell you without seeing everyone's characters.

The group is odd enough on a party level that I feel like I really need to know how their individual characters are built to begin to give more targeted advice. My answer to the OP's question varies wildly depending on how the kineticist is built, the level of optimization on the individual casters and their spell selections, and so on.

OP, if you're still around, I think everyone in the topic would benefit from some more detail on the character builds.

I'm really confused as to why you think this is some special feature of PF2. Like I can't really think of any crunch heavy, splatbook based tabletop where people don't talk about new options in a power sense if they're really out of line one way or another. It would be really bizarre if people didn't. The only TTRPGs where this doesn't feature prominently are ones that aren't mechanics forward and/or don't rely on splatbooks (and even then you'll still find mechanically minded discussions on stuff, just less).

These kinds of commenters aren't unique to PF2E. What's unique to PF2E, though, is the quantity and concentration of them. There are far fewer people vocally attached to 2E that will talk about how the mechanics match up with their preconceived notion of how something works in real life, and so on. 1E had these kinds of comments all over the place, and you see these kinds of comments in discussions of most other TTRPGs. They're noticeably absent here.

The game draws in people who specifically like the wargame balance. This has been great when you actually want to talk about how the system functions; very few people will derail a conversation about weapon balance with their experiences in their local HEMA group. But that wargame-focused group -really- cares about the wargame balance, much moreso than the other crowds the product could try to attract. That makes them sensitive to details of a kind that are just harder to perfect on tight product timelines.

.. I also don't really get the suggestion that quality assurance is primarily an issue of staving off angry fanboys. Do you think there isn't any inherent value in trying to deliver a well written product?

Of course there's value in making a good product.

QA isn't a binary pass/fail, though, and there are many dimensions along which you can try to deliver a good product. The specific kinds of QA issues I'm referring to here (mostly balancing errors or horrifying mechanics typos, like the gibbering mouther taking one fewer action to engulf than it should or exemplar dedication giving too much damage) are along a dimension that's much more difficult to spot in a QA pass. It's unreasonable to expect your copyeditors or proofreaders to know that engulf should take two actions, or to compare all the options you could take with exemplar dedication and see that some of them are out of whack. It's normal to miss things like this in a QA pass.

The stuff I'm referring isn't really a matter of making sure the content looks good and reads well, that the book's layout isn't messed up, or that the fantasy is being communicated clearly via the text. That is QA's job. My assertion is that a lot of what people complain about goes beyond QA's normal job, and requires either a proofreader/copyeditor that's also an expert in the game mechanics or an entirely separate QA pass from someone who is versed enough in the mechanics to catch things like this. I'm guessing that's a difficult ask on the current publishing timeline.

.. I'm also not even sure if the idea lines up with reality. Paizo gets dinged in community circles arguably even more often for being too conservative in its design choices, not too out there... Paizo products also come out with a lot of editing mistakes that sometimes get corrected very slowly. Not that they're releasing bad products, but it doesn't really suggest there's some intense hypervigilance in QA and editing that's being implied here.

I'm not trying to say that such hypervigilance exists right now. I just mean to say that the reaction to run-of-the-mill errors tends to be extremely negative, and it's partially because the system is designed in a way that leaves less room for those errors to occur without it affecting the play experience. Typoing +27 instead of +24 as a monster's to-hit while entering numbers on a numpad would be much worse in this game than most other d20 systems, for example—but those kinds of mechanical errors are equally hard to spot in proofing in any d20 system.

My point is just that from a pretty zoomed out perspective, "game with great balance" is a powerful market differentiator—but also one of the hardest differentiators to consistently live up to because of the sheer attention to detail it requires.

I feel like the only meaningful benefit of PF3E being a more polished version of the current product would be possible backwards compatibility.

I'd prefer a more drastic shift. We have PF2E; if I want to play it, I already can. It's more than a complete product, at this point, and has several fairly iconic APs attached to it (such as abomination vaults and season of ghosts, the latter of which polls as one of the best APs the company has ever released). There's enough PF2E content for me to run games for a few years as it stands.

I also think that since a lot of the driving forces behind 2E's system design have left the company, it may well make sense to take a different tack with them gone. It's an extremely opinionated design. I'm unsure whoever paizo assigns or brings on to develop 3E will want to carry the system ethos and design forward.

Personally, I also feel like PF2E is a fairly rigid framework where coloring outside the lines, even a tiny bit, can stick out like a sore thumb. And we've seen how detrimental that can be time and time again in the reaction to new material. The balance-heavy ethos draws in a crowd that's fairly picky about how different classes and options stack up against each other; a single option out of whack really starts to dominate the discussion. (Think about how exemplar dedication was such a topic on its book's release.) If I were paizo, I don't know if I'd want to make another system where I'd have to keep such a tight eye on the content of every single splatbook to keep those more vocal parts of the playerbase from being annoyed—especially not with paizo's release schedule. The game just demands a high level of system mastery from its content designers to avoid breaking things, and a high level of QA on its products (from people with good system mastery!) to catch slipups. And I don't know if that's a good place to be, marketwise. It's more attention to detail, and more system fluency, than I think people expect from any market competitor. It's a good differentiator if you can keep it up—but it inherently draws a fickle audience. You're punished if your quality goes down, even if it's still better than your competitors. There's an incentive to move away from the PF2E formula.

I understand wanting "the perfect version" of something you already like, but after seeing god knows how many tabletop and video games get iterated on for this many years, I've just accepted the perfect version of a thing doesn't exist and you're better off being grateful for what you've got. Something novel and interesting is usually more enjoyable than a really polished version of an experience you've already had, anyways, once you've had said experience enough.

The thing is what ive found in this forum is that people here do not treat anything as a consequential statement if its not hyper specific.

So laughing to them isn't speaking and wont cause any air loss.
I had similar discussions with enthrall claiming you cant choose to take actions inconsistent with your giving your undivided attention to the caster. But they consider it all flavor text and not instructions.

To me, this read of uncontrollable laughter and pointed question is in the same category as physics-lawyering the temperature of a fireball to try to cheat in extra effects not listed in the spell. It is something I, as a DM, am going to shut down immediately without further argument.

Even if you think laughing fit should force a creature to open its mouth —which I again find questionable, because it works on creatures that don't have mouths at all, and the game's only restriction on what creatures it will and won't work on come from its trait tags—or you're happier playing a game where mechanics entail flavor that entails further mechanics in this way, which is your prerogative... I think any DM with half an eye for balance will still push back on you here. At the least, if I counterfactually had a table of players that -all- wanted to play this way, I'd be flavor-lawyering them back and telling them it's functioning as an incapacitation effect and will grant save upgrades. And that's assuming I didn't just ban the strategy outright, in the way I'd be playing whack-a-mole with silly strategies like this in 1E.

gesalt wrote:

ScooterScoots wrote:

Starlit with IW is downright tame compared to disruptive stance with reach and tactical reflexes (and later boundless reprisals), or resentment witch, let alone something outright gamebreaking like instant minefield or anything to do with the drowning rules....

Eh. Disruptive stance doesn't really do anything you couldn't already do with silence 4th + trip and minefield is just another thing to throw into the berms-shove shredder. That said, I am glad that strat keeps getting more tools to use.

Do tell me about drowning though. Not something I've given much thought to outside of the monk suffocation combo.

Because of how the drowning rules are set up anything that makes you lose your breath/speak results in instant unconsciousness underwater. Methods include laughing fit, mask of uncanny breath, and pointed question. Probably a couple other ways too, it’s a problem with the drowning rules not any specific method of invoking them.

Pretty much an instakill on anything that breathes so long as you have two casters in the party or they’re dumb enough to stay in the water for a round.

I think you're playing the RAI game instead of the RAW game if you believe laughing fit mechanically requires a creature to open their mouth. Laughing Fit doesn't even require the creature to be able to laugh in order to work. Honestly, it might be more accurate to say you're playing the flavor game and not the RAI game. I doubt the devs intended to make laughing fit force a creature to open their mouth, if they have one.

If you're playing in that more loose narrative space, I don't see why a creature forced to respond with pointed question can't choose to mime a response or something instead.

If it helps anyone, my list of information categories for my players is as follows:

-One weakness (This includes unique, non-numeric weaknesses; BPS weaknesses are grouped as one weakness)

-One resistance or immunity (BPS resists are grouped together as one resistance)

-One special attack (I prioritize information about reaction abilities unless told otherwise)

-Half of the creature’s innate spells (or 3 innate spells, if 3 would be more)

-Best Save (player can choose to include AC)

-Worst Save (player can choose to include AC)

-Senses

-Movement types

Additionally:
-Identifying a creature gives all type-specific immunities and traits automatically (e.g. Undead traits)
-Identifying a creature may also give certain defining, exceptionally well-known abilities for free—a dragon’s breath attack, for example.
-Players are given no information if they select a category with no information to give.
-When a player rolls a failure with Dubious Knowledge, they are given a true and false statement about the category they inquired about conjoined by “or.” E.G., “The creature’s worst save is either reflex or will.” If the category contains no information, it’ll be something like “the creature either has a reactive strike with its claws, or has no special attacks.” ("It has none" is always a possible false clause, as well.)

This list is from a houserule variant that makes RK a free action, which is why some of this looks stingy.* But the categories themselves should work just fine for vanilla RK. If I were running vanilla RK, I'd adjust the amount of information you get from each category upward, though. E.G., you should get all weaknesses if you ask for weaknesses.

===

*Under the houserule, you can't repeat the action, but you do get an additional piece of information for every 5 you exceed the DC. To prevent everyone trying to RK everything, anything above DC 15 requires you to be at least Trained in the relevant skill. And since it came up earlier, I also don't use rarity modifiers that increase the RK DC on creatures at all; that stuff is anti-player in a way I strongly dislike, especially with this system.

Yeah. Assurance has uses, but it feels like it's supposed to a replacement for take 10—and it's just not. This is mainly a problem because the game has a lot of places where it feels like take 10 and take 20 should probably exist, still. Parties do sometimes just have no time pressure. Failure at some repeatable tasks (like picking locks when there's no one you're hiding from) just can't be made interesting. Assurance doesn't solve these, and it looks like it should.

Instead, the problem is just tossed to the GM to solve. I compress or outright skip checks if it's detrimental to game pacing out of combat at this point, sometimes just substituting things like "PCs with x proficiency rank automatically succeed."

I will say that—as others have mentioned—climbing is, by RAW, the single most egregious offender for this in the game. It really needs fixing.