Haha. I picked 10 just to be easy.
But, if we want to go all even farther on this thing, a 100 sided Resplendent Mansion would have an area of 71,000,000+ square feet. Per floor. That's 2.54 square miles. Per floor.
And, my calculation above for the 10 sided mansion was a slight mistake. The 69,000 figure is also per floor.
And, I actually hadn't considered that the height of each floor was not specified. I mean, it's a mansion, so it's probably not going to have 8' ceilings, right? That would hardly be "resplendent."
Honestly, the more I think about it, the more hilarious this spell becomes.
There is no demi-plane with the Resplendent Mansion, though. It differs from the Magnificent Mansion, in this respect.
The spell only has the Conjuration keyword, while Magnificent Mansion has both Conjuration and Extradimensional. Magnificent Mansion's text also specifies that it is an extradimensional space. Resplendent Mansion makes no mention of extradimensional spaces. That spell actually conjures a freaking mansion.
I kind of think anyone in upper floors would just plummet to the ground when the spell came to an end.
I would think you should be able to. The "Buy Equipment" step for character creation comes after the "Choose a Class" and "Record Class Details" steps.
Under the "Choose a Class" step, in the "Character Sheet" highlighted section, it states, "...then write a '1' in the Level boxy to indicate that your character is 1st level."
Under the "Record Class Details" step, it states, "See the class advancement table in your class entry to learn the class features your character gains at first level...."
Under the "Buy Equipment" step, it states, "At 1st level, your character has 15 gold pieces...."
So, by the time you get to the Buy Equipment step, you are already considered 1st level.
This is half a rules question and half an invitation for fun discussion. Not positive it belongs here, because I don't see anything in the rules that directly addresses this, unless it's in a FAQ or something similar that I haven't seen.
The Mansion created is up to 300' per side and has up to four stories (however, interestingly, it doesn't say how many sides it has). The duration is "until the next time you make your daily preparations."
So, what happens to the people up on the 4th floor when you make your next daily preparations if you forget to tell them to get out first?
And as an aside, if the mansion has four sides, it can be up to 36,000 square feet of living space. Pretty spacious. But, what if you decide to make it a decagon (ten sides) at 300' per side, it would have a bit over 69,000 square feet of living space.
This is how I handle secret checks.
At the beginning of each session, I hand each player a note card with their character name at the top. I ask them to make 20 rolls of a d20 and record them on the card (my players are ones I trust not to fudge the rolls, and some of the epic 1-fests I've seen in the cards I've received backs up that trust).
I keep the cards out of sight of the players, and when I need a secret check, I look at the next number on the card, mark it off, and use that as the die roll result.
I do this for a few reasons. One, my players feel like it gives them more agency; these are their own rolls, not mine. Second, there is no tell-tale clatter of dice when a secret roll is made. Lastly, it's faster.
Ever since I started doing it, my players have really liked it. They even get a laugh out of the times that the secret check hits one of the 1's or 20's they managed to roll. There's an added tension that they know they have a couple of 1's in there that might crop up at a bad time, and a 20 or two that might pop up at exactly the right moment.
In any case, this has worked really well for our group.
When even the core books have literally hundreds of spells, volume of spells is an issue for balancing from the very beginning of each edition. As editions go on, the number of unbalanced spells increase.
Systems like the one I mention with Hero allowed for a very high level of optimization that is terribly difficult to balance.
I would argue that the main reason D&D magic systems have been unbalanced is because of a combination of their sheer size (based on number of spells) and power creep over time during an edition as opposed to a failing in the basic nature of the system.
Like I said, I have not seen any system in a game that I would categorically call superior, but I have not played every TTRPG in the last 42 years of my gaming life, so I would be completely willing to take suggestions.
I will admit I have not played every single game out there, but in my experience, Vancian magic allows for better game balance than other options.
The old Hero/Champions system from back in the 1980's had a system where you could have a point pool that you would set aside from your character point build that you could manipulate periodically to have whatever sort of magical effect that you wanted (limited by the points available and the costs of the effects you wanted). It was interesting, but the entire system almost required a degree in accounting to keep track of everything.
I wanted to dig into this a bit more to explain why switching from a d20 to a d6 based system (using, for example, 3d6 to get a comparable result range of 3-18) would require significant changes to the core system of Pathfinder.
Assume for a moment that a d20 system has a core mechanic where at any given level, the average assumed score for a check has "easy," "average," and "hard" potential levels of difficulty. Easy requires a 5 on the die, medium requires a 10, and hard requires a 15. In this system, the average person will succeed on an easy check 80% of the time, an average check 55% of the time, and a hard check 30% of the time.
If you switch over to a 3d6 system without completely changing the rest of the system, just merely using the same rolls and modifiers, the successes change to 98.16% for the easy difficulty, 65.31% for an average difficulty, and 9.23% for the hard difficulty.
Any way you think of it, using d6's rather than the d20 would require a complete re-work of the entire game mechanic from the bottom up. It would fundamentally become a different game.
Replacing d20 with d6 would result in a dramatically different game, functionally speaking. And, IMO, RPG's that use d6's as the primary random die for checks have always been inferior, mechanically speaking, whether it's a system that uses the sum of the d6 rolls or one that uses multiple d6 checks against a target with number of successes determining the outcome.
The d6 doesn't provide enough variation. Using a number-of-success type model has issues with upward scaling as characters advance because of the low range variable range. Using a sum-of-dice method creates a bell-curve distribution of results rather than an even distribution, which makes the math for creating a balanced core system more difficult.
Choosing the d6 just because it's more common at a time when you can get a d20 for $0.25 USD is an odd concept. It's not like d20's are something new. They've been around for 50 years.
While this probably wouldn't be allowed in Pathfinder Society, the ancestries section of the CRB indicates that a Dwarf could be Half-Orc. This would allow combining things like Mountain Stoutness + Toughness for 6+ Dying saves and the Orc Ferocity feats to continue acting when you would otherwise be unconscious.
Couple this with a high HP class, and you can take a ton of damage and keep on fighting until you're beaten all the way to death.
James Jacobs wrote:
There are a lot of ideas possible herein.
One could be building the guild from the ground up in a city where there has never been a meaningful criminal organization.
Or, the characters could begin as low level members of an existing organization working their way to the top.
Or, the area could be one with a large number of competing organizations where the party are members of a lesser organization that fights to increase in power and assert dominance over the rest.
In my opinion, Sanctuary is meant to work only against Attacks. Otherwise, the ability to completely stop the triggering attack would mean stopping a spell or a dragon breath. For a first level spell, it would be way too good.
I have to disagree.
First, if it only blocks Attacks, then it becomes pretty useless since there are so many possible ways to negatively impact a character without using an Attack action targeting the caster of Sanctuary.
Also, it's not a blanket prohibition. The attacker gets a Will save every time it wants to launch an attack. Only a critical failure results in an inability to attack the caster for the remaining duration of the spell (while a critical success ends the spell for all attackers). Considering that the current math in PF2e makes failing saves far less of a certain thing for enemies, the spell is not remotely some insurmountable obstacle to levying attacks on the caster.
Not sure if this has been done (am not an expert on all the AP's that have been done), but here's one I would like:
An AP focusing on the Mana Wastes and the dwarves of Alkenstar. Maybe a situation where the conflict between Geb and Nex is flaring up again, threatening another war in the Mana Wastes, while the party works with Alkenstar to try to keep the city-state from being overrun by the two significantly larger armies.
It's a great spell to exist so that NPC's can use it to confuse and confound the players.
The usefulness of spells differ for PC's and NPC's as their goals are pretty different. There have always been lots of spells that, on the surface, don't really have enough obvious usefulness for a PC to use them, but help give a mechanical explanation for how an NPC can accomplish his or her ends.
Like in this case, the party keeps meeting a variety of different women of varying ages that appear like they might be all related to one another. They first meet a motherly matron and have a role playing interaction. Later, they meet a young girl who says, "Oh, Aunt Bee told me all about you!" and have an interaction with her. Later on, they meet an old crone who calls herself Granny Ceecee and have further interactions.
Little does the party know that they are all the same Witch using the different guises to tease out information, deceive the group, and advance her hidden agenda.
I don't think I can create my fearsome canine warrior concept when I'm stuck with a small statured anthropomorphic pug.
And, seriously, if the art is accurate to the concept, the ancestry should take a penalty to stealth because they constantly snuffle when they breath.
In the real world, wheel lock firearms didn't exist until after Columbus sailed the ocean blue. Rifled firearms didn't exist until the mid-16th century. Firearms didn't become particularly common until the latter half of the Renaissance. The flintlock didn't exist until the 17th century. Percussion caps for firearms didn't appear until 1820.
If your world is fashioned after a combination of Medieval and Renaissance historical elements, it's perfectly reasonable for firearms to be rather rare items that required specific knowledge to use. This is even more necessary when you talk about care, maintenance, and manufacture of firearms, as prior to 1798, all firearms were one-of-a-kind creations by a craftsman; Eli Whitney's factory near New Haven was the first firearm factory of it's kind where every weapon produced was created to a specific standard so parts from one could be interchanged directly with another.
We're just going to have to agree to disagree.
To me, there is zero "neat" about the concept of the flickmace. I find it dumb from every angle. It's particularly so because the game has come up with a bunch of original fantasy weapons that actually are interesting with a high "cool" factor. The flickmace has great mechanical features in game, but the concept is just kind of dumb.
I agreed with you up to the last paragraph. Yes, it's a ridiculous weapon but then again it's a gnome weapon. Gnomes would happily show up with a monofilament yo-yo if they could. I'm imagining Gogo as a gnome
This is one of the several things I hate Dragonlance for. The entire idea that gnomes are weird tinkering mad-scientist experimenters comes from Dragonlance. Hate it.
But, Gogo's weapon is a meteor hammer.
And, a meteor hammer doesn't remotely work like a mace and certainly doesn't move around with a "flick of the wrist." It's far more complicated than that.
My issue with the flickmace isn't the mechanics.
It's just a stupid weapon idea, conceptually.
"More a flail than a mace, this weapon has a short handle attached to a length of chain with a ball at the end. The ball is propelled to its reach with the flick of the wrist, the momentum of which brings the ball back to the wielder after the strike."
Flails with long chains and short handles are more dangerous to the user than the opponent. There is a reason that historical flails pretty much universally have longer handles that the combined length of the head and the chain. If the chain/head is longer than the handle, the user is going to end up hitting his/her own hand or arm with alarming frequency. A longer chain also allows for far less predictable deflections of the head when it hits an opponent's armor. Historically, the ball-and-chain type flail has very little actual evidence of use in warfare, probably because of the issues mentioned above.
Then there's the whole flicking a wrist to send out the ball which miraculously reverses momentum to return to the wielder. Is chain elastic or something? Also, if the chain is that long, how is merely "flicking" the wrist enough to send the ball flying towards an enemy. As an experiment, get a yoyo and let it hang and full length. Then, without any other movement, "flick" your wrist and see how far the yoyo moves.
The description of the weapon and how it works is complete nonsense.
Yeah, yeah, I know, fantasy-blah-blah-blah. It's still a stupid concept from that perspective because it isn't even a cool idea. Rather, it's just a fundamental misunderstanding of how flails work. The Rule of Cool (though highly impractical) works for things like the Dwarven Waraxe (big ax with two bits), Orc Necksplitter (huge ax reminiscent of a bardiche with serrated blade), or Rhoka Sword (sword with two parallel blades). The flickmace is a ball attached to a short stick by a chain.
Apep is not _solely_ dedicated to murder or darkness. He/it has a much broader purpose for existence in Egyptian mythology as a bringer of chaos and enemy to Ra.
The Phonoi were very minor entities, nowhere near on the level of Zeus, Dionysius, Hera, etc.
Both your examples prove my point with historical reference. It's one thing for a major god to have something like "murder" as part of their repertoire, but to have that be their sole area of influence? Just silliness.
The problem is that Golarion is a world where the gods actually, irrefutably exist. It's just plain stupid for there to be multiple gods of the same aspect from different pantheons across the world. It's nonsensical, confusing, and personally I think it only happens because Paizo needs stuff to put in new books.
Let's look at an alternative that would make a significantly greater degree of sense with a larger amount of internal logic. Imagine that there is an actual trickster god here in the real world. In Native American areas, he's worshiped as Coyote. In Nordic lands, he's known as Loki. In the Mediterranean area, he's know as either Hermes or Mercury. In China he's known as Sun Wukong. In Polynesia, he's known as Kaulu. (There are a bunch of examples.) But, ultimately, he's the same, single, individual divine being who is just known by different names in different cultures due to variations in language and local experience.
Paizo, however, has over 70 deities with the Trickery domain. It's poor design. It creates bloat with no benefit. It creates a world with no fundamental internal logic for the gods since the vast majority of those gods have no real explanation for why they exist in the world and how they relate to the other divinities.
To me, the number of deities is less important than the pantheon making sense.
One of the big problems with Golarion's gods (and, honestly, every set of pantheons ever created for D&D before Pathfinder existed) is that the pantheon just didn't make sense. I've written about this in the past, so won't repeat myself.
Make sure your pantheon has internal logic. Have a story for the origin of the universe that supports not only the existence of gods in general, but the existence of those specific gods. Explain the relationship between the gods, who are allies, who are foes, and why. This will play into how their followers react to each other in the world and help justify the world's geopolitical situation.
Some might not agree with me, but avoid the silly practice of having greater gods solely dedicated to some silly fantasy trope like "murder" or "darkness." Sure, a god might have multiple aspects, and one of those aspects might include that, but you don't need an entire divinity dedicated to nothing but darkness. Or, maybe a minor god that serves another greater god could have sway over such a minor sphere of influence.
I personally prefer smaller, tighter pantheons just for simplicity in creation and ease of maintaining internal logic, but the real world definitely has examples of much larger pantheons (such as China's celestial bureaucracy, where if counting every named divinity will get you over 1,000). When I create small pantheons, I will give them multiple aspects that effectively create multiple gods without having more actual gods. From a historical example, look at Zeus; he was the god of the sky, lightning, king of the gods, honor and justice, and creator of the laws the gods had to follow. (Which is all kind of funny considering how much he cheated on his wife.) But, the point would be that in fantasy setting, a cleric could serve in a temple that revers Zeus specifically in his facet as the god of thunder and lightning, and the character tailor himself around that, while another follower of Zeus could revere him in the aspect of the king of the gods.
But, to wrap up, just try to make the pantheon make sense. Don't just create a god to have a god for something needed as a plot point in an adventure, never to be used again.
Captain Morgan wrote:
I didn't use hero points at all in book one (because I had forgotten about them). The addition of using them in book 2 has already proven to be pretty significant, specifically where a couple of potentially fatal natural 1 rolls for saving throws are concerned. For example, this past Sunday, the party discovered the Elephant People village, and promptly critically failed the attempt to calm the elephants. In the resulting combat, one of the elephants trampled 3/5 of the party. The wizard rolled the 1 for the reflex save, which would have put him unconscious in round one of the fight. At that moment, the rogue player reminded the wizard player of the hero point, which saved the wizard from going down.
Tactically - the Paladin soaks up damage by wading into the middle, while the wizard and cleric both free cast as best they can (though the Cleric acts as our second tank). The rogue does buckets of damage, but always manages to be between the boss and a bad place, and take a licking for it. The bard has negligible combat influence, though numerically his inspiring performances do account for a lot of the party damage.
I think this is some of the issue. In P2e, it's really hard to just go in and tank stuff. Enemies' first attacks in higher challenge fights just hit too easily, and the second attack is often better than 50/50 on hitting, too. It also sets up the "tank" to be hit with the bosses' strong 2- and 3-action attacks.
I would suggest in tough one or two enemy fights that the party needs to do things that eat up the enemies' actions. Something as simple as move in, attack, then move away will force the enemy to use at least one action to get into attack range (assuming it's not ranged focused). Other things like shove effects that force the enemy to close distance have the similar effect. Tripping eats up an action to stand up.
Then, there are lots of spells that eat up actions as an effect. At 4th level, Confusion gives Stunned 1 even on a successful save. Hideous Laughter gives Slowed 1 on a failed save and no reactions on a successful save, and is a sustained spell. Things that cause Sickened are good because they give a penalty to rolls, and the enemy can sacrifice an action (also a good thing) to try to remove the penalty. Lots of debuff spells like Fear and Goblin Pox impose a negative state on the enemy even if they pass their saving throw.
If you can find ways to make the enemy only make a single basic attack action each round (rather than multiple basic or any of the 2- or 3-action attacks), beating the enemy with the favorable action economy of the party becomes a lot easier, even for tougher enemies.
I haven't read the whole thread, so others might have already pointed this out.
Instead of having the campaign solely based on the idea that the players mistakenly help the wrong side, how about developing both sides of the plot quite fully. Have neither side be completely virtuous or sinister. Allow the players a chance to know something about the factions involved. Their actions in the initial encounter can give them an opportunity, conscious or otherwise, of which side they wish to ally with.
That way, the players have more agency. Sure, they may fail all the checks to recognize who any of the people are, noble, soldiers, and wizard alike, but if they do fail all the checks, at least they had the chance to figure it out. From a player standpoint, this would be FAR superior to merely hitting the party with a "gotcha" after defeating the wizard.
Kinds of checks:
After a couple weeks' delay due to life issues, I finally got to run the social events section at the beginning of the book with my group.
The party surprisingly enjoyed the event, and the dance contest ended up being their favorite, with the party deciding to have head-to-head dance-offs between each other for the title of party's best dancer. Probably spent 45 minutes just on this event. The monk was the eventual champion.
I decided to present everything exactly as written in the book, including the romance portion that I previously questioned. My group reacted pretty closely to how I expected. Two of them thought it was weird that a someone would invite strangers into their personal family matters. None were comfortable with getting involved, with one specifically stating that they'd be concerned with messing something up with potentially disastrous social fall-out. The wizard was the only one remotely interested in the story, but mostly because he wanted to watch other people try (his character is terrible at social skills).
Second favorite section was the chili eating contest, which was interestingly won by the sorcerer with a combination of lucky rolls and a high deception ability.
I took the advice to pick rewards for the part of 6th level items appropriate to each individual character, and that went over really well. I super like this aspect of treasure reward design in the new AP's. It's a bit more work for me to try to find things that each individual party member would like, but the excitement they all showed when they got something cool that worked well with their character concept was really gratifying.
I'm still not buying it. It's just way too much of a stretch to me for someone to go from never having seen someone before in their lives to engaging them in arranging a romantic tryst with a family member in less than 24 hours.
And, the timing isn't all that wrong. The party will spend a significant amount of the time between arriving at the town and the feast out on the hunt. The time around the dance contest, talking to the lion, etc. is a couple of hours, at most. And the time during which the party is directly engaging Nketiah in conversation is a fraction of that.
The entire scenario is so achingly cringey to me. I might be projecting a bit, but if I were a player in this AP, I'd totally check out of the game until it was over. Romance between PC's or between PC and NPC can be cringey enough, but facilitating a romance between NPC's....ugh.
Our group just finished Hellknight Hill yesterday and began Cult of Cinders. All they've done is start repairs on the castle, repair the gate, and scouted it with a familiar (I went ahead and had the familiar trigger the Dahak remnant thing even though I wasn't sure if the familiar should, but didn't want to spend time digging around the rules, and it seemed an intelligent use of a familiar).
Next week, we should actually get to the social events with the Ekujae, and I'm having an issue with making the motivations in the Matchmaking section make any sense.
The Ekujae are described as being borderline xenophobic, especially with non-Elves. A few weeks earlier, a group of Cinderclaws attacked and killed several of the elves to activate and use the gate wherein the Ekujae's ancient nemesis is trapped, and today, the party just pooped out of that same gate. The party has the opportunity to speak with Nketiah for maybe 30 minutes of in-universe time before she decides to approach these complete strangers and recruit them to engage in a high-school level plot to encourage a romantic tryst involving her father? Why the heck would she have anywhere near enough trust in these people to involve them in her intimate family matters?
I honestly cannot think of any way that this makes any sense. I also cannot find any way to make it make sense. I'm probably going to go ahead and present it as written, but knowing my party, none of them are going to be interested in getting involved because they're either afraid of messing things up or they think it's a stupid idea in the first place (which I also kind of agree with). The average age of my play group is in the 30's, and this story line seems more appropriate to people who watch the Disney Channel.
Sure, but then do you go into detail about the design and quality of every enemy's weapons, armor, personal effects, etc.? (The OP was asking about descriptions of magical items.)
If every room description is like your second example, that's fine because you've established a standard of expectation for your players. But, I'll assert (based on my 4 decades of gaming experience) that many players will start to tune those descriptions out if they are too long. I'm sure you've experienced the same thing, where players seemingly having no knowledge of things that you just read out to them 60 seconds earlier.
I just wanted to put this out there for commentary.
We are still in the middle of Age of Ashes, but our group is already looking forward to Extinction Curse after I described the basic idea of the party being party of a circus.
Now, the group is kicking around the idea of fashioning the party and circus after the Carnival of Chaos from the old GW game Mordheim.
The worst thing about this is that I'd really like to play in that....
Descriptions tend to make players think things are important, so I'd only describe important things. So give a description of the item is magical, made out of a special material, or is particularly valuable.
I agree with this. It's kind of a play on the idea of Chekhov's gun. Giving more elaborate than usual descriptions of items will make the players assume there is something important about the item. If there is not, in fact, anything important about the item, it will waste the players' time trying to make that determination, and then make characters less interested when you give elaborate descriptions in the future.
You don't think you need to? But you did.
You are the one that differentiated between Shoanti, Varisian, Numerian, Chelish, Mwangi, and Osirion, but then lumped the entire continent of Tian into one.
I just suggested that this was silly. You countered that you didn't want "to narrow it down too much," when you had already done so.
I'm not sure why you're defending your decision to be so narrow and specific for one region, but then poo-pooing a similar degree of granularity for another region.
As for Mwangi, it's a tiny region compared to Tian. Mwangi is roughly 20% of the continent upon which it lies. Tian is the largest continent on Golarion. I'll reiterate that lumping Tian into one cultural group would be like lumping all of the Inner Sea region into one group (including Mwangi, Osirion, Absalom, Cheliax, etc. into one group).
Nice, but totally missing the point.
If we're going to make cultural distinctions in this case between neighboring kingdoms in the Inner Seas region (which is just a part of the continent upon which it is located), it is silly to lump an entire continent (containing multiple different nations) into one single cultural group.
Shoanti and Varisian peoples are no more culturally dissimilar than are the various Tian groups, for example. If we're going to say "Tian" is one single cultural group, we should also say "Inner Sea" is one single cultural group.
Tian Xia isn't just one culture, I don't believe. It's a continent with a bunch of different nations. The Tian aren't a single ethnic group or single culture.
Tian-Shu are somewhat analogous to real world China, so weapons like the Meteor Hammer and Nine Ringed Sword are appropriate there, since they are drawn from Chinese martial arts tradition.
Tian-Min are somewhat analogous to real world Japan, so weapons like the Naginata and Tekko-kagi are appropriate to this group since those weapons are drawn from real world Japanese history.
My point is that it's silly to differentiate between groups like Shoanti and Varisian, or Numerian and Chelish, but then lump an entire continent into one cultural group.
MAP maxes at -10, regardless of how many attacks taken after the second.
"The more attacks you make beyond your first in a single turn, the less accurate you become, represented by the multiple attack penalty. The second time you use an attack action during your turn, you take a –5 penalty to your attack roll. The third time you attack, and on any subsequent attacks, you take a –10 penalty to your attack roll."
And, to extrapolate even farther.
If your character had a modifier of +24, the rolling a 1 would result in a 25 on the check. For a DC 15 skill check, this would normally be a critical success, but the 1 on the roll would reduce it to a normal success.
I have to wonder if you're deliberately ignoring the part about the bow being the backup weapon for the WP, so wouldn't be used against the majority of opponents.
Castilliano has the right of it. The square targeting mechanic is just for game purposes. In a "real world" situation, you'd just fire your weapon in that direction, not at a specific volume of space. Honestly, that makes being blind in the game even worse. In a "real world" situation, if I fired in that direction, and they were fleeing directly away from me, I'd still have a chance to hit them. In the game, if I fire at a specific square, and they've moved one square directly away or directly towards me, I have no chance of hitting them, even though the enemy is still exactly in line with the shot.
One thing I haven't seen mentioned, but just want to make sure the OP hasn't missed it (forgive me if I'm being a bit pedantic).
Remember that when blind, assuming no other sense they can use for targeting, all opponents are undetected by the character (the Blind condition says that if vision is your primary sense, you cannot detect enemies).
When targeting an undetected enemy, they can pick a square to target. Then the GM makes a secret DC 11 flat check and makes the attack roll in secret; this is different than targeting a hidden enemy where the player makes the flat check and attack roll. Regardless of the success or failure of the check, the action is carried out, which will increase the MAP of subsequent attacks, use ammunition, expend spell slots, or whatever else is appropriate. If the attack misses, the GM does not inform the player as to why it missed; the player isn't even informed if they are attacking the wrong square.
Now, don't be a jerk and fudge the die rolls because you don't like the player attacking the last place they saw the enemy, but a character attacking in such a situation is going to have a tremendously hard time hitting anything between the negative to hit modifiers for blind/dark and the flat check for targeting an undetected enemy.