S.L.Acker's page

It does not take an action for a versatile weapon to change between damage types. You're thinking of the modular trait. With a versatile weapon you simply declare what kind of attack you're making when you attack with it.

In that case my bad. Though it still doesn't make sense that something like a morningstar doesn't do both types at once.

The way Pathfinder 2 has decided to implement versatile damage makes little sense for most of the weapons it applies to. For instance, what exactly is a character doing to switch a morningstar between bludgeoning damage and piercing damage? Why can't you use polearms, especially spear-like polearms, to deal bludgeoning damage? What is gained by removing the tempo advantage that a cut-and-thrust sword traditionally used to its advantage?

It simply doesn't make sense how the trait was assigned and why it takes an action to change between damage types.

Quote:

It demonstrably cannot. You cannot play as a commoner as happened in a hilarious thread on the old D&D forums or the story of a troll blessed with a ring of sustenance that enabled him to pursue pursuits other than sating his hunger.

I am confused. I regularly play commoners.

Sure, PF2E Cannot pull the same RULE Interactions as 1st Edition, That is not the same as being able to tell the same stories?

There was a rather memorable thread back on the old D&D forums where a player used the commoner NPC class and the rules for the game to play a fully supported campaign where they just did daily commoner tasks. Deciding where your 2+int skill points went at each level was interesting because you should probably level up your profession but a couple of points in craft could save you some coppers in the future. PF2 can't do this without major house rules, 3.x and PF1 could handle it from the outset.

Think about the roleplaying afterwards, "You ran away and left us to die when the tank went down. You are out of the party!"

Maybe that happens or maybe the player has played up this fear well enough that the party was just waiting for it to happen so the character could progress their character growth.

"I saw the fighter go down and went running for help." Is a valid counterpoint.

What would have happened if Rand ditched Matt when the dagger started to make him a liability to be around?

Because those vaster numbers of stories are about a teammate being a jerk to the team. The other players hate those stories.

I've run entire evil campaigns and Cyberpunk games where everybody was just waiting for who would betray the group first. Where dying party members were stabilized and sold for parts so the rest of the group could be combat-ready faster. Those were fun and memorable games that PF1 could support but that PF2 seems unable to support.

This is a solid opinion. Optimizing PF2 characters to be better than average for their level is so difficult that it requires great expertise and focus to achieve minor success. If everyone has to fit the mold of Superman, then everyone is the same.

In contrast, superhero teams like the X-Men or the Legion of Super-Heroes have individual specialties that they combine to neutralize enemies stronger than them. They don't have to all be Superman when one is Phantom Girl, one is Lightning Lad, one is Cosmic Boy, and one is Saturn Girl. That is how my players defeat difficult PF2 encounters. And they love non-combat skills, too.

I feel like my issue with PF2 is that the divide between classes is too sharp and artificial and that too much of it gates things a skilled warrior should just be able to do behind feats or other hoops. It doesn't make sense to me that, for example, a high-level fighter can never figure out the ranger's knack for hunting prey or that only a thief rogue can use dexterity as a damage modifier when the more combat-focused swashbuckler cannot. The niches seem like invisible walls in a video game, a frustrating design choice that should be avoided when possible.

Plus, it doesn't help that aside from a single stat, Phantom Girl, Lightning Lad, et al. will all rank up Dex, Con, and Wis at every opportunity while being likely to have the same general feats. There's little room for expressive character building with PF2 because often a concept can only be done with a single class and if you don't like how that class was built, that's too bad.

In 3.x/PF1 a new player could describe what they wanted or name a character from fiction and there's a good chance you could build it for them with little compromise. That's just not the case in PF2 and I feel the system is less for this lack.

S.L.Acker wrote:

Squiggit wrote:

That seems less like merely RPing in combat and more like one player actively refusing to participate and then the GM engineering a scenario to sabotage the rest of the group for it.

Is the cowardly rogue not a common enough character trope for running away to be a valid way to RP?

Pathfinder is a team game, so no.

So is Cyberpunk, but that game encourages cutting your losses and cutting the group's dead weight out of the reward. I don't understand why this most balanced possible game* can't support a thing that happens in the genre fiction surrounding it.

*According to its fans.

That seems less like merely RPing in combat and more like one player actively refusing to participate and then the GM engineering a scenario to sabotage the rest of the group for it.

Is the cowardly rogue not a common enough character trope for running away to be a valid way to RP?

The notion of half the party not having great odds of handling a boss designed for the whole group isn't really a PF2 specific phenomena either (especially with the added layer of an extra ambush on top of that).

I'd argue that in PF1 a party, even at level 3, has better odds of avoiding that TPK than the same party in PF2 would.

Not really. Warpriest has some mechanical issues, but the former two are fine. Which turns this less into a systemic issue and more "this particular option wasn't balanced well", which is a fair complaint and one I've raised before, but a far cry from "If you RP you will suffer a TPK"

I consider how you build a character as an extension of your RP. Mechanics should be a part of how you express your character and that means. I would like it if a pure blaster wizard actually got to function and feel good rather than relying on the failure effects of their best spell slots when the going gets tough. Or for the single element focused, bender, sorcerer to feel like there's a path for them to be viable.

PF1 had a lot of bad content to be sure, but the content that was good was often good enough to allow even the most niche build to function and pull off their gimmick. In PF2 if there isn't a class or archetype for it (or if you just got one of the undercooked classes or archetypes) then you're SoL.

So...

Lunge (reach)
Knockdown (combat maneuver)
Intimidating Strike (combat maneuver)
Fighter's +2 proficiency (increased critical range with no confirmation roll needed)

And those are all on whatever weapons you want to use them with.

You are still looking for: bracing against charges and versatile damage. Yes, those are weapon traits that you need to pick specific weapons for. Both of those traits do exist on some weapons, though not a Katana.

Then no weapon is special in PF2 and you may as well just re-fluff that Maul into a Katana and enjoy better combat outcomes.

Sandal Fury wrote:

What I meant was there seem to be very few options that both require a two-handed weapon and do something more elegant than "knock dude down," "push dude," or "hit more dudes."

Well, I can't think of much else to do with a weapon other than hit dudes with it. The same could be said of 1-hand weapons. Besides "hit dudes, possibly with some added effects like extra damage or status effects", what are you going to do that requires the weapon - Juggle it?

Maybe you could give us some examples of what you want to do with your 2-hand weapon.

If we're looking at other editions it could brace against charges, enjoy reach, grant a bonus to a combat maneuver fitting your character, and have versatile damage that doesn't take an action to switch between. Yes, that is a polearm and not a sword but even swords got increased critical range and versatile damage as almost baseline abilities in PF1.

It's also a good idea to make "the rules for skills, saves, armor class, difficulty class, and combat are all basically different applications of the same rules."

PF2 doesn't really do that though. If it did you'd have a classless skill-based d% system where you can earn points used to bump skills or buy feats for skills that meet a specified threshold. As is, there's a big difference in how the rules treat attacking or casting a spell and how they treat using Recall Knowledge.

I don't want to be an optimally effective min-maxed hammer man.

I want to be a cool samurai.

Then you're going to pick a sword and propulsive bow, start with an 18 in strength while making sure you bump dexterity, constitution, and wisdom. Congratulations, you've completed your build.

I was agreeing with you that it all depends on what the AI is trained on.

If you train it with something, then it will be good at that one thing.

My bad. It can be hard to tell with so many people pointing at the AI equivalent of a toddler and mocking them for not being able to do things a fully functional adult takes for granted.

You're leaving effectiveness on the table with a Fighter by taking anything that doesn't have knocked prone as its crit spec. There simply aren't many optimal builds in PF2 that use a sword and forget making a proper Samurai that uses a horse, bow, and sword to full effectiveness.

There was that one military surveillance AI that was trained to detect movement of humans. So the soldiers testing the system used everything from just jumping to going full Solid Snake to bypass it.

AI are great, but they are not miracle machines.

Machine vision is orders of magnitude harder than language fluency and knowledge of a finite set of ordered rules. There's also AI that has been trained in gait detection that is extremely hard to fool by behavior that wouldn't draw unwanted attention from human security, so...

It's interesting to me how "the game has no room for improv" has become a new complaint about PF2, when when the system was new it was common to hear people complaining about the exact opposite and the system wasn't nearly strict and explicit enough.

PF2 doesn't define enough out of combat but is overly prescriptive in combat and in exactly how class features work. Look at the debate about familiars for the former and the entire melee Magus action economy for the latter.

I feel like this one is only half true. PF2 does, for better or for worse, expect a minimum degree of optimization, but hitting those notes is also relatively easy

It also leads to characters feeling the same as the vast majority of characters will increase exactly, primary stat + saves, every fighter will have the exact same proficiencies, there's no granularity to skills (and little enough use for having low ranks in lots of skills even if you could), no room for proper multiclassing.

It makes building a good enough character easy but remove the ability to ever build anything more personalized.

Not sure why RPing would lead to a TPK, that seems like another weird GM quirk, though.

Picture a level 3 party facing off against a severe threat. The tanky character drops, perhaps for the first time, and the cowardly rogue decides to cut his losses and flee. The rogue isn't strong enough to fight off even a fairly easy encounter if they run into a wandering patrol and probably isn't sneaky enough to avoid that patrol either. The party isn't strong enough to win the fight now being two characters down and everybody dies.

In a less extreme case building a pure evoker, an elemental sorcerer, or an offense-focused melee Warpriest are all options that are likely to weaken a party such that published modules become an issue if you have one or two players in a party who want to play this way but who also object to the DM making the game easier for them. The game doesn't leave a lot of room to deviate from its core assumptions.

Except "bad at math" only covers one of the problems answered wrong by ChatGPT, for which yes, a math sub-system could be created.

For at least two of the problems it got the math correct, but was so off on its interpretation of the physics concepts the prof said he'd give that answer a zero despite the bot getting the correct final answer. With RPG, "correct math answer, but wrong understanding" wouldn't help gaming, as what could be built on/extrapolated from such a shoddy foundation?

And one would expect such a GM-bot would think it was correct, even if totally wackbards, maybe even using its correct math as justification. And how would players reason with it? And if it does solicit input, how might we protect it from being manipulated by wordplay?

ChatGPT is also a new AI trained on a lot of unvetted data with a massively broad scope. If you made a focused AI that only had to understand a relatively tiny subset of ideas I would expect accuracy and perceived comprehension to rise.

The Sixty-Symbols video also had a lot of, "Wow this is really impressive." mixed in with explanations of where it has a lot of room to grow.

PF2 is the definition of if everybody is Superman nobody is. Its balance comes only from painting everybody into a box; hyper-focusing on combat balance while ignoring exploration, crafting, and social encounters; and removing the ability and desire of a GM to say, "Sure, roll for it." because such off-script moments ruin the balance. Your character is more mechanical than PF1 because the game expects optimization and you'd better not RP in combat or you could lead to a TPK.

Oh, and half the modules are unbalanced and present themselves as time-sensitive dungeon crawls while the game rules want lots of rest between fights and downtime for retraining.

You could base the Shaman around trances where they beseech the spirits for aid when they refocus or take a long rest. The long rest version would grant a long-term buff that lasts all day while the short rests grant different focus spells. This could make the Shaman a very flexible primal caster.

If you granted them a lot of access to divination spells from all traditions they also take on a very advisory role.

Their secondary feature could be built around mage hand and unseen servant-type effects. Not a fully functional familiar as such, but a fickle helper that can be cajoled into performing tasks.

A primal counterpart to the bard that supports the party by being exceptionally flexible with their frequent selection of focus spells with a large once-per-day buff that rewards scouting and divination to ensure it gives the correct bonus for the situation ahead.

Let me put another lens on this and see how it looks.

If the next book that comes out has a class specifically based on a stereotype of African tribal warriors and mashes them into a class that can only use specific weapons and armor in exchange for a set of class features with very limited flavor options, would we be okay with that? If we wouldn't be, why does the monk get a pass?

Lol. I meant representing the obvious American-inspired character by wearing a fat suit would be mean. That’s the equivalent of the awful Asian makeup in that other picture.

Clearly, that's the thing that's wrong with yellowface...

TheRabidOgre wrote:

The problem is that it treats Chinese culture not as a culture you might have yourself and thus see the whole game through the lens of, but as something special to a specific and singular class.

That I see it repeated over and over in asian media makes it seem fine to me. Seriously, almost every isekai story these days has a monk class.

So a white guy doing a rain dance would be fine with you as long as native groups had enough media presence to put their stories and culture in front of a mass non-native audience?

Part of the issue with the Monk is that there is too little respect for HEMA and the training it took to be a skilled knight. People forget that their training started as children with the page, to squire, to knight progression. The monk really shouldn't look all that special next to what a European knight or fencer can do.

<snip>

I think it may be possible to solve some of these issues with a multi-layered AI. ChatGPT is the interface and compiler as well as the module that chooses what to present but there could be a math module, a conversational memory module, etc. This feels like how AI is going to grow in the future.

It's also a bit like how our minds work with different sections processing different tasks with our consciousness attempting to select the most useful/appropriate outputs from these competing systems.

Um... To clarify, I am not talking about tweaking the algorithm into a better algorithm. I'm talking about the computer science concept of an algorithm.

Every computer program is an algorithm. At least currently. They all have the computing power of a Turing Machine. That's the Church-Turing Thesis for those who want to look it up.

So unless we can create an AI that isn't an algorithm - or a computer program as we currently know them - they are going to be limited by the limits of computability of an algorithm. There are things that an algorithm can't do.

We don't need AI to think or to question the answers that it gives to vastly reduce the number of confidently stated wrong outcomes. There's still a ton of room to improve on the systems we have now for even better results.

It's also odd to assume that AI needs to be perfect when we don't expect the same of another human. Just look at those stupid math memes that go around for examples of "intelligent" beings confidently being wrong about math.

S.L.Acker wrote:

It's also extremely new so we should expect these kinds of issues to improve over time.

I don't think that is the problem.

The problem is that algorithms have no understanding. They can't decide if something 'makes sense' or 'is reasonable'. By definition they always give the correct answer and so have no way of verifying their own work.

I expect that the baseline algorithm will be updated based on current errors and unwanted behaviors. That or data could be selectively pruned to improve results. Both are a lot of work, but so is building ChatGPT in the first place so I expect it to happen.

I got into a conversation with ChatGPT last night. I asked it to produce a well-ordering of the real numbers. It provided (<) and we ended up in a situation where the Axiom of Choice was equivalent to its negation.

So the future of AI is bright.

Is ChatGPT designed around mathematical accuracy? It seems to me like its primary goal is more around answering general knowledge questions conversationally. It's also extremely new so we should expect these kinds of issues to improve over time.

10 rounds, you mean.

Well, yes.

It's because the Barbarian is basically the Ultimate Warrior. They're tough and shrug off damage exceptionally well but get gassed if forced to fight matches longer than 10 minutes.

Turns out ChatGPT made such a big leap only because it hired Kenyan moderators at 2 USD / hour to weed out all the paedophillia porn stories that sweaty American nerds tried to make it process. So much for "Artificial" intelligence.

That doesn't make it any less intelligent. You have to train children and pets to avoid unwanted behavior, so why should an AI be any different?

I like GMing. I like creating. AI won't change that anytime soon.

However, AI will eventually surpass the wildest speculation on this thread. Imagine the most brilliant and inspired GM you've ever played with who can run a campaign for you and your friends with zero prep time, god tier maps, guides, and collateral, and is always on and primed to continue the campaign 24/7.

Post-Turing test AI will deliver. Today, "AI" is a buzzword thrown about by every company's marketing teams. It's not the gauge to estimate the potential of AI to GM for us. True AI will fundamentally change the human experience, and I can't wait to RP with it.

I'm hoping we get there sooner rather than later. Imagine your favorite work of fiction but without wait times and in endless detail always focused on the characters you enjoy with plot lines you weren't interested in neatly pruned. It'll be like endless social media scrolling but for any medium that you might ask for. Seductive and dangerous.

Why in all that is creative would you want this? All I can see from this (or the poster using it to “generate” “content” their campaign) is laziness. Or arguing that you are time poor. This isn’t some jumped up ammanuensis, it’s replacing your own creativity.

Why in the world would you use a calculator, you CAN do math right? Why use a vehicle when you can just walk, you aren't too lazy to walk, are you?

The big leap that makes Chat GPT better than what came before it is the AI's ability to remember 15 messages worth of context. That number will likely grow as we learn to make this kind of AI more efficient alongside the inevitable upgrades in the hardware running the AI.

The Gleeful Grognard wrote:

In the end I think it would be a mistake to look towards Ai GMing over Ai augmented GMing.

What would AI augmented GMing even look like? What could/would the AI be doing that the GM wouldn't?

Filling in details for areas of the game that aren't as deeply prepared as might be preferable. Like that one NPC the players decided is now their best friend, you can quickly generate a few AI-made passes for details about them and suddenly you have a fully fleshed NPC in seconds. It can also be an idea generator, you toss in a few plot ideas, let the AI cook, and see which one seems like something worth cleaning up and running.

Not everybody is a constant font of creativity and lots of GMs prep way less than they should an AI can help them focus on the parts of running a game they enjoy by filling in the bits they don't.

S.L.Acker wrote:

Captain Morgan wrote:

I feel like Devise a Strategem does the RDJ thing? I guess it doesn't do a string of hits but I don't think "unleash four attacks with 100% accuracy" was ever going to be a realistic expectation.

I feel like just hitting in melee doesn't really capture the feel of those scenes that well. There should probably be a feat path that allows the Investigator to add extra effects on attacks made while using DaS.

So like Shared Strategem, Didactic Strike, Athletic Strategist, Scalpel's Point, or Strategic Assessment?

With the exception of Athletic Strategist all of those also work with ranged attacks. An Investigator needs more incentive than that to get into melee because they're significantly safer and have a fuller suite of options at range.

I feel like Devise a Strategem does the RDJ thing? I guess it doesn't do a string of hits but I don't think "unleash four attacks with 100% accuracy" was ever going to be a realistic expectation.

I feel like just hitting in melee doesn't really capture the feel of those scenes that well. There should probably be a feat path that allows the Investigator to add extra effects on attacks made while using DaS.

If AI-GMing can get beyond what Wrath of the Righteous can deliver it will have value. If you can interact with it using natural speech and a VTT it could easily be the first way people experience TTRPGs.

As somebody who mostly GMs for friends, I would probably enjoy blowing off steam by running 4 min-maxed characters in an AI-run game.

those scene are amazing

was thinking exactly that the minute start reading investigator

tried for hours to figure out how to make martial artist investigator work

one inch punch require expert strike which is a specific monk class feature so investigator can not take them raw

grievous blow seem to be the best with a d10 stance strength monk at the time

There's definitely room for an improvised weapon and unarmed strike build to be created for the Investigator.

I think the basic issue with the Investigator is that it's difficult for some people to weigh "combat abilities" against "non-combat abilities". Pathfinder 2e kind of sequesters combat abilities into "class feats" and non-combat abilities into "skill feats" but the Investigator breaks that assumption.

But the right way to design a class called "the Investigator" is "whenever there's a thing that needs figuring out, that guy/gal/etc. is the one you want to have around."

I think there's room for an Investigator to wreck people like the RDJ version of Sherlock Holmes did and that the team missed the mark on making melee Investigators satisfying to play.

Still waiting for Inquisitor. And more customization in general I know the limited options serves as a break for power creep but at the same time I feel 2e lost a lot of that concept of being able to create characters in several different manners as you were able in PF1.

Yeah, it would be nice to have many paths that reach the same goal the way PF1 did.

I would also like some Book of Nine Swords-type classes that do martial arts but magic. Kind of like a 5e Battlemaster Fighter crossed with a cantrip-focused class that regains focus in a similar way to a swashbuckler. I'm aware that probably sounds OP in the current PF2 paradigm, but I like swiss army characters that fulfill the fantasy of being self-sufficient outside of the specific circumstance of facing threats large enough to be worth gathering an adventuring party to defeat them.

If I were running a game where disguises and intrigue were integral to the campaign I'd use the rules below.

I'd set a fixed DC to create the disguise and then use the 4DS system to grant a +1, a working disguise with no modifier, a -1 penalty, and a complete failure that requires you to not only start over but to observe the target a second time. From there the disguise would work as normal and checks to stay in character would be run using the usual social skills modified by how well you've been disguised.

I'd also let others aid the check with with a broader range of skills than I would in a normal game so everybody feels involved even if they aren't as skilled in disguise or might not have as large a role in the upcoming scene. I'd also make sure the rules changes are making things at my table more fun and see if my players like the rule or want to ditch it for RAW starting next session.

In a normal game, I'd use the rules as written as I wouldn't expect it to come up overly often.

QuidEst wrote:

A disguise might be set up well indoors, but not account for the hot weather, or strong winds outdoors. Eye color is tricky to change through mundane means before modern plastics, but also hard to notice. And so on- there are plenty of technical details that can go wrong without a mistake in behavior.

But I've never seen a GM going into such details. In general, if the character is "disguised", the GM will ignore all but the most extreme situations.

Some GMs also allow familiars to scout without rolls too but that doesn't mean we should assume that all GMs are going to nudge odds in their player's favor.

S.L.Acker wrote:

This ignores that a Centaur would be able to use its mass to add a ton of power to strikes ... A realistic Centaur with a lance or spear would be terrifying to a degree that a man riding a horse simply can't be.

Would also be dreadfully prone to severe spinal damage from the force of most any lance or spear charge, sadly, but then realism has never been much of a deal-breaker in fantasy.

Depending on how exactly the connection between the human torso and the equine torso is handled that could 100% be the case, though my headcanon is that there's probably a lot of thick muscle there that mitigates the risk. Beyond that they'd also be prone to breaking their legs in ways that need magic to heal due to how little vascularity a horse's leg has, they'd have issues with a lot of spaces a humanoid can comfortably adventure in, require armor that would cost many times what a humanoid fitted set would cost, etc. These are issues, but no more so than people playing aquatic species in games that are mostly on land.

D&D5E making centaurs Medium is apparently controversial, but I will staunchly defend the choice. Here's the thing, while a centaur's whole body might be large (though that can vary massively depending on the breed of horse you're using for reference) every practical way they interact with the world is based on their "human" half.

They wield human-sized weapons and have human-length arms. In most situations they are effectively a human that simply can never dismount their horse.

This ignores that a Centaur would be able to use its mass to add a ton of power to strikes, has built-in natural weapons that can (depending on the species of equine) one-shot lions, and that weapon size has less to do with damage than you'd think. There's also the added carrying capacity, stability, and movement speed over flat ground to consider. Heck, they might even be able to sweat which would make them even more terrifying endurance hunters than humans are.

A realistic Centaur with a lance or spear would be terrifying to a degree that a man riding a horse simply can't be.

The word genocide doesn't appear once in Lost Omens Legend so no, it is not "literally" text in it.

I would argue mass relocation at swordpoint should count as genocide due to the cultural destruction that comes from being forcibly relocated away from ancestral homelands and the cultural and religious sites contained within.

"Skeleton Crew" is a great name for a adventuring party. I should have made all my Blood Lords player pick skeleton ancestry.

Boney M is both the bard that the party needs and the one it deserves.

You could look at what you do with stats and start from there for a rebuild.

I tend to think in terms of these general uses for stats:

-Force, the ability to influence a challenge in a straightforward way.

-Resist, the ability to resist a challenge in a straightforward way.

-Manipulate, the ability to influence a challenge with finesse.

-Avoid, the ability to resist incoming challenges with finesse.

Then you figure that you can do those things physically, mentally, socially, and with magic and you have four stats with four uses. Characters might be straightforward physically but cunning mentally or could lean into defensive uses of their talents while being weaker on offense.

Tie these stats into skills, figure out which dice to roll, and you have the bones of something with a lot more freedom and a lot less baggage than the traditional array of six stats.

S.L.Acker wrote:

I prefer a more gritty lethal game than many people here and often wish that Paizo had taken more risks in game design with PF2,

You have used this argument in more than one thread. We have very different understandings of the words being used. See, the way you're saying it, "I wish they had taken more risks" seems to suggest "I wish they had made the specific product I want, that isn't very much like the product they made at all. I will imply that they wanted to make the game I wanted, but chickened out and played it safe."

Game design is not a linear progression toward the game you imagine, and failure to evolve in that direction is not cowardice. They made the game they wanted to make. You can like it or dislike it, but insisting that the game would be specifically what you would like, if only Paizo were bold enough, is just... weird and disrespectful.

The Devs have literally said that they couldn't make certain changes they had planned for fear of alienating a vocal segment of their audience. The entire way shields work and the rune system weren't what the devs wanted to do. The Witch's development was stunted by another dev leaving and never really got finished which leads to an odd class. The Alchemist had to get a large amount of errata before it was on the same lap as the rest of the classes.

If you think the game we got is the game the devs wanted to make, I'm not sure what I can say to you. I also admit that the game they wanted to make wasn't that much closer to the game I want, but I would have respected the choice to stick to their intent and release what they wanted to.

How do the rules of PF2e inhibit hexcrawling or sandbox style games? Or to flip it on its head - what rules do other systems have that enable hexcrawling?

The tight encounter math and strict rulings on spells and out-of-combat actions mean that running into a higher-level encounter is much worse for a party in PF2 than it would be in say D&D 5e. Plus, PF2 is really best suited to tailored encounters designed either for the environment they take place in or, better still, tailored to the party's strengths and weaknesses to create a series of knife-edge balanced encounters that threaten doom but never quite get there. PF2 has a massive strength in these kinds of well-crafted combat puzzles but often feels lacking when played in a more freeform fashion where players don't treat combat as an optimization puzzle.

There's a certain appeal to D&D 5e's looser encounter math so where the party is rarely put into a situation that creative use of spells and abilities can't at least let them escape from. This can lead to 'dead' encounters where the party isn't challenged by something that, in theory, should challenge them but also allows for the DM to ramp things up to where a party is fighting enemies that feel better because the players know their [insert party level] group beat an encounter that is, supposedly, [x+party level] by the game's default math.

There's a reason knives have never been a dominant weapon on battlefields across history or altered the nature of warfare like even early firearms did. Even with something like a knife bayonet, it is more effective attached to the end of a rifle (functioning like a spear) rather than using it as a knife.

Knives (or more accurately stone handaxes) were likely among the first manufactured weapons used by humans against other humans. Along with the sharpened stick and thrown rocks; knives absolutely changed warfare. It just happened so long ago that we take for granted that people use weapons to kill other people.

Guns also don't deal more damage than a spear or a knife that imparts the same energy. A .38 round will do less damage than a 9mm which will do less damage than 5.56. It's just easier and safer for the attacker to use a gun and bullets can do a better job penetrating armor and clothing than a knife.

This all said I don't think this forum is the right place to discuss terminal ballistics and studies of harm caused in cases of IRL trauma. I prefer a more gritty lethal game than many people here and often wish that Paizo had taken more risks in game design with PF2, but even I can see that this conversation can't go any further without getting too graphic for this forum.

Yeah lol. Frankly, I think the reason people expect guns to be so lethal is because of bias from irl. I certainly have no idea why anyone would think the capability of peasants to use something easily would allow for greater ability to take down "high level" things so to speak.

IRL firearms scale really well. There's very little that should stand between the party and taking the idea of a gun and scaling that up into something like a 4-bore or a punt and using that to take down high-level threats. Or taking an air repeater, enlarging the tank, and playing with the chamber and hose geometry to get higher muzzle energies without losing the number of shots between reloads. There is no reason for these weapons to stay balanced in the hands of intelligent beings who see them as a good way to defend themselves with minimal training.

PF2 is not the simulationist game it sounds like you want.

I'm well aware of that, what did you hope to accomplish by saying this?