Stealth Playtest

To summarize the main points made in this thread so far:

• There's a nearly (entirely?) unanimous sentiment that the condition granted by successful stealth needs to be something other than invisibility.

• Sneaking from one point of cover across open sight to another point of cover - provided you can do it in a single action - no longer auto-fails, and it seems several people already houserule it this way anyway.

• Wording needs cleaning up (but of course that's to be expected at this point).

• Most people bothering to share like the overall "meat" of this re-write and support the changes.

Suggestion: Whenever you guys get to the point where you think you have a publishable new rule,

besides releasing it as Errata, ALSO have a Blog/FAQ explaining the DIFFERENCES to the old version, or, `what is new`.
If the aim is a rule-set that is easily understandable, that will also be helpful for that goal.

Agreed.

The key difference between the new Stealth rules and the old ones is that your stealth does not expire immediately if you enter an open area. Even in bright light, your stealth does not deactivate until the end of your turn - and if you get to a shadowy area or behind cover before then, you can keep using stealth.

I suggest that, rather than monkeying around with new states or making stealth use the invisible state, we merely add the above clause to the stealth rules.

Rephrased clause:

When using Stealth, creatures cannot become aware of you until the end of your turn, even in areas of bright light or in the absence of cover. If you end your turn in an area with cover or concealment, you may continue using Stealth.

(of course, this wording will have to be changed or expanded to account for full-attack sneak attacks not being possible)

Occam's Razor — if adding this effect is the only reason for the use of the invisible state, just add the effect as a clause and you sidestep the problems with the invisible state.

I will be playtesting the "just the clause" version. I'll report back on my results.

Suggestion: Whenever you guys get to the point where you think you have a publishable new rule,

besides releasing it as Errata, ALSO have a Blog/FAQ explaining the DIFFERENCES to the old version, or, `what is new`.
If the aim is a rule-set that is easily understandable, that will also be helpful for that goal.

On the one hand, that kind of clarity is always admirable... on the other hand, if they preface the rules with the intent of the rules, most people will read only the intent, and assume it works. I think we're better off having to parse the rule text for ourselves, at least at this early juncture.

'Rixx wrote:

The key difference between the new Stealth rules and the old ones is that your stealth does not expire immediately if you enter an open area. Even in bright light, your stealth does not deactivate until the end of your turn - and if you get to a shadowy area or behind cover before then, you can keep using stealth.

I suggest that, rather than monkeying around with new states or making stealth use the invisible state, we merely add the above clause to the stealth rules.

Rephrased clause:

When using Stealth, creatures cannot become aware of you until the end of your turn, even in areas of bright light or in the absence of cover. If you end your turn in an area with cover or concealment, you may continue using Stealth.

(of course, this wording will have to be changed or expanded to account for full-attack sneak attacks not being possible)

Occam's Razor — if adding this effect is the only reason for the use of the invisible state, just add the effect as a clause and you sidestep the problems with the invisible state.

I will be playtesting the "just the clause" version. I'll report back on my results.

HOLD ON A SECOND!

"...end your turn or take an ineligible action." If you successfully stealth over from one pillar to the other, then cast a spell (for instance), you lose stealth when you start that standard action, not at the end of your turn. Don't forget that part in your playtesting. :)

I'll refrain from commenting directly as others have voiced my concerns. Instead I would like to propose certain systematic approach to four important, nay, cardinal aspects of using stealth.

ACTIVATING STEALTH
The way I see it, this should be a free action. Tying it to specific conditions creates hermetic subsystems riddled with exceptions.

DEACTIVATING STEALTH
Either an immediate action, or state change caused by certain circumstances. Copy n'paste from losing invisible condition.

HANDLING MULTIPLE CHECKS
The way I do it currently: the stealthy character keeps his last check. He maintains his last check as long as he keeps doing the same action, or takes an action which is sufficiently unobtrusive not to warrant another check.

HANDLING GROUPS
Again, this is the way I do it now:

Stealthy group uses check of the creature with lowest check result. If circumstances allow, more skilled characters may use Aid Another action to allow less skilled characters to take ten or take twenty.

A group of creatures who use Perception passively, like guards or a patrol, take ten, use best score, with Aid Another from the rest of the group limited to +4.

A group of creatures using Perception actively, like guards searching for escaping criminal, take ten or take twenty (depends on time they want to spend searching), use best score, with Aid Another from the rest of the group, limited to +10.

Regards,
Ruemere

HOLD ON A SECOND! "...end your turn or take an ineligible action."

I will play with Jiggy's amended version instead. Thanks Jiggy!

When you are hiding, creatures who fail to perceive you remain unaware of you until the end of your turn unless you take an ineligible action. This is the case even in areas of bright light or in the absence of cover. If you end your turn in an area with cover or concealment, you may continue to hide.

Here's some food for thought on being "hidden"

If a person is hiding behind a bush, some people may see the person, and some may not, depending on their senses and positioning.
That said, Hidden is not really a status or condition. Hidden is a relationship between a creature/object and some subset of all observers.

So, in choosing how to describe "hidden", this relationship needs to be pointed out.

Hidden
A creature or object which is "hidden" is unable to be perceived by one or more observers. Any observer which cannot determine the location of the hidden creature or object, treating them (mechanically) as fully concealed, and also cannot target, nor attack the hidden creature or object.

Special senses, such as tremorsense, scent or blindsense are similarly incapable of pointing out the location of the hidden creature, though such senses may still be able to detect that the hidden creature or object is nearby. A guard dog can use scent and know that someone is in the vicinity, but can't determine an accurate direction or distance.

Note that a creature or object which is hidden, may be perceived by some observers, and not perceived by others. For example, if you are hiding behind a bush, it may hide you from some observers on the other side, but hides you from nobody on the side which you are hiding (due to lack of concealment).

I agree stealth could use to an overhaul. I do not want any system that allows easy indefinitely stealth looping or creates a situation where stealthy characters are impossible/improbable to detect.

How do the new rules interact with stuff like scent, blindsense, and tremorsense?

Quandary wrote:

Suggestion: Whenever you guys get to the point where you think you have a publishable new rule,

besides releasing it as Errata, ALSO have a Blog/FAQ explaining the DIFFERENCES to the old version, or, `what is new`.
If the aim is a rule-set that is easily understandable, that will also be helpful for that goal.

On the one hand, that kind of clarity is always admirable... on the other hand, if they preface the rules with the intent of the rules, most people will read only the intent, and assume it works. I think we're better off having to parse the rule text for ourselves, at least at this early juncture.

Well, this kind of goes with my 1st post here in this thread, that the rules text ALREADY (blog version) feels VERY non-standard as to rules tone/style, because it is trying to be a FAQ at the same time. Sure, the rules should definitely be 100% usable on their own... But trying to also serve the function of a FAQ, especially a FAQ detailling the difference to an older version, is an aim that is contrary to tight, consise, functional rules text. I agree that a short intro, as rules often have, is definitely a good idea also... But for the players who AREN`t prone to going over Errata twice and de-constructing how it differs from both A) the previous rules and B) how they were using the previous rules, simplying laying out the differences in a Blog Post / FAQ* is the most direct and efficient means to the desired end.

*FYI: there are still other Blog Posts that should be linked the FAQ per se, most people aren`t going to search the Blog history when they hear from a friend that Paizo has updated/explained the rules in some manner. All Rules Blogs should really also be included/linked from the FAQ page.

Stealth rules are considered ineffective because logically, when someone is able/isn't able to stealth doesn't match up to in-game situations. These rules don't fix that, just change it.

The current rules say that when Rogue A enters combat with Guard B, he's essentially NOT going to be able to use Stealth effectively, because of the lack of facing rules and B's basic common sense to stand in an area not close to any form of cover. The pillar will hide A momentarily but won't allow him to do anything else. Facing and whatnot would be invaluable to A in this scenario.

The proposed rules here allow Rogue A enter combat with Guard B in broad daylight. Rogue A can simply step behind a single pillar, granting him concealment. Stealth roll is made, Guard B fails to meet - B is aware of A's presence and likely location, but A is considered invisible to him. A can now simply walk up to him, a straight line from behind his pillar to his face, and despite the fact B is actively looking for him in the that direction in broad daylight, he is surprised to look down at himself being caught flatfooted by A who suddenly pops back into existence as he plunges a dagger into him. Facing truly doesn't even matter in this scenario.

I think the overall lesson in all this stealth hooha is that while simplicity is desired for the game, complexity actually allows for more logistical rulings (i.e. facing rules, Perception being the 3.x version of multiple skills, etc. etc.)

I'm glad that Paizo is making an attempt, though. I know that they're a work in progress, and they were clear to state that these are not anywhere finished, but these make less sense than the current ones. It's wanting to go left but taking steps right.

A'course, maybe I misinterpret the rules. I know in my games, we just hand wave the rules and do what makes sense to us.

I guess I'm in the minority on this, but I really, really like using the Invisible state with stealth. Invisibility increasing that then makes a lot more sense mechanically (to me, at least)... rather then creating another state for essentially the same thing (in regards to combat someone is either aware of you or they are not)this really simplifies things.

I am a little concerned about action types and accounting for actions.

There are a variety of actions that are labeled as "free actions" or "not an action" that we historically haven't worried about the specific delineation because they don't mess with the action economy at all. Under the new rules, we have to figure out each distinct free action, because we have to make a new stealth check for each one.

Until now, it didn't matter if activating Power Attack was a free action or part of another action; either way, you can do it for free. Now it matters, because if I want to get my Power Attack damage off along with my sneak attack from stealth, if it requires a free action, I have to make an additional Stealth check.

An attack of opportunity is "not an action", correct? We have the rules for AoO's specifically, but aren't there other "non-actions" that we now have to classify?

Also, I'll reiterate the question others have asked -- under these rules, is there any benefit at all to trying to stealth while under the effects of an invisibility spell? IMO, there should be -- the description should be clear that succeeding at stealth makes you invisible and unnoticed, or something like that.

I'll refrain from commenting directly as others have voiced my concerns. Instead I would like to propose certain systematic approach to four important, nay, cardinal aspects of using stealth.

Hooray for proactive comments! Good call, this far into the thread.

ACTIVATING STEALTH

The way I see it, this should be a free action. Tying it to specific conditions creates hermetic subsystems riddled with exceptions.

Er, I'm not sure this makes sense. If "activating" is its own (especially free) action, then you also have to have rules about how often you can try it. Also, this opens up the possibility of using a free action to activate (intending to sneak from one pillar to the next - perhaps we should name this "pillar hopping"), then evaluating your check result to see if you want to actually pillar hop or just wait until next round. Also, there's a major flavor disconnect here: you don't "start being sneaky" and THEN pillar hop; you attempt to pillar hop sneakily - this is why the proposed rules make the stealth check be part of the action that you're trying to do sneakily (i.e., make your check as part of your move action to pillar hop).

DEACTIVATING STEALTH

Either an immediate action, or state change caused by certain circumstances. Copy n'paste from losing invisible condition.

Why do we need a special action to deactivate? Just use an action that's not eligible for stealth, or use one that is but decline to be sneaky. Or just scream "HEY I'M RIGHT HERE YOU GUYS". Also, do you really mean "losing invisible condition", or "ending the invisibility spell"? Very different things.

HANDLING MULTIPLE CHECKS

The way I do it currently: the stealthy character keeps his last check. He maintains his last check as long as he keeps doing the same action, or takes an action which is sufficiently unobtrusive not to warrant another check.

When you say "keeps doing the same action", do you mean he doesn't make checks for each square of movement in a single move action, or that if he takes the same action over and over again he only uses a single check? If the former, that's how the proposed rules work: one check per action. If the latter, that's kind of silly, because it's very easy to do something well a few times and then slip up and make a noise.

HANDLING GROUPS

Again, this is the way I do it now:

Stealthy group uses check of the creature with lowest check result.

Doesn't necessarily conflict with the proposed rules. Sounds fine.

If circumstances allow, more skilled characters may use Aid Another action to allow less skilled characters to take ten or take twenty.

This isn't even related to stealth; the rules do not allow for a character to cause another character to be able to take 10 or take 20 if they couldn't already. Might want to go read the take 10/20 rules.

A group of creatures who use Perception passively, like guards or a patrol, take ten, use best score, with Aid Another from the rest of the group limited to +4.

A group of creatures using Perception actively, like guards searching for escaping criminal, take ten or take twenty (depends on time they want to spend searching), use best score, with Aid Another from the rest of the group, limited to +10.

Sounds reasonable, though not really related to stealth rules.

The proposed rules here allow Rogue A enter combat with Guard B in broad daylight. Rogue A can simply step behind a single pillar, granting him concealment. Stealth roll is made, Guard B fails to meet - B is aware of A's presence and likely location, but A is considered invisible to him. A can now simply walk up to him, a straight line from behind his pillar to his face, and despite the fact B is actively looking for him in the that direction in broad daylight, he is surprised to look down at himself being caught flatfooted by A who suddenly pops back into existence as he plunges a dagger into him. Facing truly doesn't even matter in this scenario.

The best way to fix this is to give a penalty to re-stealth against any observer that has already noticed you.

For the problem of a Full Plate fighter able to hide sucessfully when creating a diversion to hide with a bluff check, I have no problem in imageing the situation. During the round you created the diversion to hide, it's more the opponent not looking at you than you being great at hiding. After that to maintain your Stealth you need as a free action to make a Stealth check. So a fighter could only hide himself for one round.

The problem I see with this though is that there is no Stealth DC if someone try actively to find you during his turn since you only made a bluff check and will make the Stealth check on the following round.

A can now simply walk up to him, a straight line from behind his pillar to his face, and despite the fact B is actively looking for him in the that direction in broad daylight, he is surprised to look down at himself being caught flatfooted by A who suddenly pops back into existence as he plunges a dagger into him. Facing truly doesn't even matter in this scenario.

The guard isn't flat footed. His dexterity bonus to AC is negated. While both condition allow a sneak attack, they are very different.

The proposed rules here allow Rogue A enter combat with Guard B in broad daylight. Rogue A can simply step behind a single pillar, granting him concealment. Stealth roll is made, Guard B fails to meet - B is aware of A's presence and likely location, but A is considered invisible to him. A can now simply walk up to him, a straight line from behind his pillar to his face, and despite the fact B is actively looking for him in the that direction in broad daylight, he is surprised to look down at himself being caught flatfooted by A who suddenly pops back into existence as he plunges a dagger into him. Facing truly doesn't even matter in this scenario.

Nope, you've got it wrong (at least, I think you do - you weren't very clear on actions/turns). Under the proposed rules, you make your stealth checks at the start of an action. Thus, when the rogue goes behind the pillar, he's going to need to spend some time there (i.e., his remaining action for the turn) in order to make his stealth check.

Then the guard gets his turn. Knowing what direction the rogue went, he can pursue and actively look for him, forcing a new check that might reveal the rogue. If the rogue makes it to his next turn without being discovered (guard's thinking "crap, where'd he go?"), then he can spring out from his hiding place and sneak attack the guard (guard's thinking "where's... OH SH-").

Now, I will concede that according to the proposed rules, the guard could move to where the rogue's hiding place no longer provides cover (i.e., the other side of the pillar, where the rogue is) and potentially fail to spot him. I guess this represents the "dumb guard" who walks a couple of steps too far, looks left and right (but not behind him), and scratches his head. I'd prefer a contingency for when it's utterly obvious that the rogue is RIGHT THERE.

But overall, I think the proposed new rules cover most situations pretty well.

So just to illustrate the badness of equating stealth to invisible I am going to bring up a fairly common argument at my table regarding stealth and Yee old tower shield.

I used to have players argue that because a tower shield can be used for cover that they could then stealth behind said tower shield and that the tower shield would magically disappear with them. I would usually argue that no such thing occurs and to try something else you silly munchkin.

Now, with these rules, the tower shield literally disappears with the rogue post stealth check. Gear disappears along with the character during invisibility. Sorry...but this breaks belief. If nothing else can we please rule out the tower shield as a form of cover for stealth.

So just to illustrate the badness of equating stealth to invisible I am going to bring up a fairly common argument at my table regarding stealth and Yee old tower shield.

I used to have players argue that because a tower shield can be used for cover that they could then stealth behind said tower shield and that the tower shield would magically disappear with them. I would usually argue that no such thing occurs and to try something else you silly munchkin.

Now, with these rules, the tower shield literally disappears with the rogue post stealth check. Gear disappears along with the character during invisibility. Sorry...but this breaks belief. If nothing else can we please rule out the tower shield as a form of cover for stealth.

I was actually thinking about this earlier, but not in the sense of going invisible with the shield. That's silly. Actually, my idea is silly too, but this sounds fun (if a bit impractical and mostly useless), so bear with me. May not even work per RAW. Spoilered because it's only inspired by stealth discussion.

I think a clause should be added.

"Any creature currently engaged in melee combat or engaged in any skill besides stealth is automatically treated as distracted."

This allows a rogue to sneak up against an opponent too busy defending himself to notice somebody sneaking up for the sneak attack.

For the record this is pretty much how I treat stealth. Hide in Plain Sight means you have an additional trigger for when you can begin stealth. (I want to stealth:
Do I...
Have cover?
Have concealment?
Have dim light (shadowdancer)?
Have favoured terrain (ranger)?
Have invisibility active?

IF yes to one of the above then make a roll. If not, meet the above criteria.

I like the rules change and will say that my experience of using a similar method has made for more satisfied rogues.

Do the above rules allow you to:

Hide in a crowd?
Hide in a pile of leaves?
Sneak up behind someone distracted by conversation?

If yes: success!

If no: please revise.

:-)

The guard isn't flat footed. His dexterity bonus to AC is negated. While both condition allow a sneak attack, they are very different.

Don't confuse them.

My bad. My overall point was to allow sneak attack, but yeah, I used the incorrect term.

Nope, you've got it wrong (at least, I think you do - you weren't very clear on actions/turns). Under the proposed rules, you make your stealth checks at the start of an action. Thus, when the rogue goes behind the pillar, he's going to need to spend some time there (i.e., his remaining action for the turn) in order to make his stealth check.

Then the guard gets his turn. Knowing what direction the rogue went, he can pursue and actively look for him, forcing a new check that might reveal the rogue. If the rogue makes it to his next turn without being discovered (guard's thinking "crap, where'd he go?"), then he can spring out from his hiding place and sneak attack the guard (guard's thinking "where's... OH SH-").

Now, I will concede that according to the proposed rules, the guard could move to where the rogue's hiding place no longer provides cover (i.e., the other side of the pillar, where the rogue is) and potentially fail to spot him. I guess this represents the "dumb guard" who walks a couple of steps too far, looks left and right (but not behind him), and scratches his head. I'd prefer a contingency for when it's utterly obvious that the rogue is RIGHT THERE.

But overall, I think the proposed new rules cover most situations pretty well.

If within one move Rogue A gets behind the pillar, he can start a second move action, thus allowing him a Stealth check. It's not start of a round, it's the start of an action. He could spend the movement either staying in his concealed spot (by moving one square away and immediately back - after all, he would be treated as invisible).

Heck, as A Man In Black wrote already, the rogue (if having the ability to) could use a Swift action to grant himself a Stealth check by casting Quickened Spell. Technically, he could use his first action to get behind the pillar, Quicken [anything] to use Stealth, and then even use a second move action to get another roll, if he screws up his first.

This goes back to my overall point that while complexity is annoying, in certain situations it allows for a more believable setting.

Also note that by using the proposed Diversion rules, Rogue A could simply use Bluff (albeit at a -10) to become invisible to Guard A, -without- a Stealth check even involved here, move half his speed and then allowed another move action at half speed for another Stealth check, since he's treated as invisible and thus allowed to use stealth.

Now, with these rules, the tower shield literally disappears with the rogue post stealth check. Gear disappears along with the character during invisibility. Sorry...but this breaks belief. If nothing else can we please rule out the tower shield as a form of cover for stealth.

I was actually JUST about to bring that up! I've had the horrible feeling that folk aren't listening to Feral just because he doesn't seem to enjoy typing in full sentences. His concern is perfectly legitimate. If there is an area with "concealment", then you can turn invisible really easily... what if I take a sheet and hang it from two ioun stones? I now have a spinning blanket of concealment. 51 gold gets me a floating personal screen that grants breaks line of sight several times a turn... even better, attach a rock to the bottom so it naturally spirals.

Have you ever heard of the peasant railgun? You get 500 peasants, line them up 10 feet from eachother and tell all of them to ready an action to pass an object from the person behind them to the person in front of them. You then hand a 10-foot pole to the person at the end of the line. When it arrives at the last peasant, it has gone 5000 feet in less than 6 seconds. It makes NO sense and shouldn't work. The difference between the peasant railgun and the always-invisible tower shield/blur spell is that nothing in the rules says "check the pole's current position as opposed to the beginning of the turn" but these stealth rules say concealment+check=invisible.

I'm really happy that paizo wants to keep improving the game, but I'm afraid we are reaching into a territory where we are trying to make it READ the way most of us have been subconsciously just DOING it and, in the process, we are just messing it up even worse. I am 100% against these rules as they currently are, I'd rather use the old ones.

Just popping in to say that is an AWESOME formian queen!

(Forgive the MtG lingo, but I'm a Vorthos, not a Melvin. ;P)

Thank you for working on a fix on the Stealth rules (and asking for our feedback, too ) !

+1 :-)

No. When you spend a swift action in this way, you still need cover or concealment...unless you are using the hide in plain sight class feature, which is another matter entirely. Unless you have a class feature or some other ability that says otherwise, you always either need cover or concealment to hide, no matter the action you are taking.

By raw this is actually not true all the time. At least not when it comes to the rogues HIPS. This is another tricky rule that is unclear. Check out this link

If you wish to fix the stealth rules you might as well fix the issues with:

I also think this is a good point to address the issue about eligibility to use stealth:

Current:
You cannot use stealth without at least partial cover or concealment. So you cannot use it in the middle of a courtyard in broad daylight.

The intention is obvious enough, but it creates special cases, for example, a dwarf can see you stealthing in the night, you don't have concealment from him due to dark vision.

Instead:
You can use stealth at any time. Creatures against whom you don't have at least partial cover or concealment automatically succeed on opposed rolls to perceive you.

Putting the onus on the observer opens up the stealth skill for practical use without a big list of special cases.

I don't like using invisibility. I don't think we need a new status either.

If you succeed at your stealth roll, you remain unobserved until you take a standard action, or until you attack, or until the start of your next action, whichever comes first. That's very simple and straight forward, and it's how I do it in my own games. There is no confusion of invisibility or anything else.

It let's Jack's chicken get stolen, because the rogue makes his stealth check while under cover, and so long as he moves from cover to cover before his next action, and he doesn't yell at farmer Bob, or attack farmer Bob, and as long as Farmer Bob's dog doesn't make his perception roll either, then Jack is able to sneak past the 10 foot gap in the hedge. No invisibility needed, he just remains unobserved by both Farmer Bob and Bob's Dog.

If the dog makes his perception, but Farmer Bob doesn't, then Jack's hosed, because doggie barks and draws Farmer Bob's attention.

If neither make their perception roll, then Jack could instead sneak up on Bob, and next round, he'd start his round still unobserved, until he does an action, any action. That action could be to sneak attack Farmer Bob, in which case he'd get sneak attack for his first attack, but lose it after that (he's now observed by Farmer Bob, and thus his ally, Dog).

Staying invisible does not cause the spell to not end - anyone with English as their primary language can see that. The spell being active and the creature being invisible are not the same thing, any more than standing in the rain and getting wet are the same thing; sure standing in the rain will cause you to get wet, but that doesn't mean that if you walk inside and jump in a pool that you're somehow still in the rain just because you're still getting wet. In the same way, the spell being active causes you to be invisible, but that doesn't mean that continuing the effect in spite of the spell ending will cause the spell to not end.

(edited, slight qualifications and clarifications attempted)

And no, the language we're dealing with is not English here; we're talking about pure rules text, a fantastic argot which bends the fabric of reality, at times defies mere common sense and drives the incautious stark raving mad.

And thank-you for the rain/swimming-pool example. It was a rather fine one, and I feel that it precisely makes my point that it's a bad idea for any revised Stealth rule to try and do things with the wording presented in this particular blog... ;)

I also think this is a good point to address the issue about eligibility to use stealth:

Current:
You cannot use stealth without at least partial cover or concealment. So you cannot use it in the middle of a courtyard in broad daylight.

The intention is obvious enough, but it creates special cases, for example, a dwarf can see you stealthing in the night, you don't have concealment from him due to dark vision.

Instead:
You can use stealth at any time. Creatures against whom you don't have at least partial cover or concealment automatically succeed on opposed rolls to perceive you.

Putting the onus on the observer opens up the stealth skill for practical use without a big list of special cases.

I'd go a little farther. I'd just give anyone who can observe you a +30 to notice you. I've actually seen people get surprised, even when they are watching for someone. Just because they start to get bored and daydream. So if you roll a stealth 38 check, and the bored guard rolls a 1 and get's 37 (+6 perception), he might still not notice you, cause he's imagining the bar girl and his ale when he get's off duty.

Diego Rossi wrote:

The guard isn't flat footed. His dexterity bonus to AC is negated. While both condition allow a sneak attack, they are very different.

Don't confuse them.

My bad. My overall point was to allow sneak attack, but yeah, I used the incorrect term.

It happen so often that it has become an habit to say correct when the term is used the wrong way. No offence meant.

Jiggy wrote:

Nope, you've got it wrong (at least, I think you do - you weren't very clear on actions/turns). Under the proposed rules, you make your stealth checks at the start of an action. Thus, when the rogue goes behind the pillar, he's going to need to spend some time there (i.e., his remaining action for the turn) in order to make his stealth check.

Then the guard gets his turn. Knowing what direction the rogue went, he can pursue and actively look for him, forcing a new check that might reveal the rogue. If the rogue makes it to his next turn without being discovered (guard's thinking "crap, where'd he go?"), then he can spring out from his hiding place and sneak attack the guard (guard's thinking "where's... OH SH-").

Now, I will concede that according to the proposed rules, the guard could move to where the rogue's hiding place no longer provides cover (i.e., the other side of the pillar, where the rogue is) and potentially fail to spot him. I guess this represents the "dumb guard" who walks a couple of steps too far, looks left and right (but not behind him), and scratches his head. I'd prefer a contingency for when it's utterly obvious that the rogue is RIGHT THERE.

But overall, I think the proposed new rules cover most situations pretty well.

If within one move Rogue A gets behind the pillar, he can start a second move action, thus allowing him a Stealth check. It's not start of a round, it's the start of an action. He could spend the movement either staying in his concealed spot (by moving one square away and immediately back - after all, he would be treated as invisible).

Heck, as A Man In Black wrote already, the rogue (if having the ability to) could use a Swift action to grant...

In the guard scenario there is a strange effect:

....

The Rogue make a fizzling sound when his pseudo invisibility end? :)

In the guard scenario there is a strange effect:

- the rogue hide behind the pillar after attacking
- the guard move in a location where the pillar don't conceal the rogue but fail to beat his stealth roll, so he don't see the rogue
- the rogue turn start, he need to move to attack the guard, but he can't hide as he is not benefiting from concealment. As soon as he do anything he is spotted automatically.
only if he has a way to attack the guard as his first action (without drawing a weapon or any other action that require him to do a stealth check) he can make a sneak attack.

....

The Rogue make a fizzling sound when his pseudo invisibility end? :)

Agreed. It doesn't make sense. And this situation is simply two people and a pillar involved in bright light where the guard has perceived the rogue at the start of it all. There are almost always MUCH more things involved, and just so much stuff can be abused by loosely worded rules.

My roommate and I have been arguing over this. He keeps saying that simple common sense on the GM's part negates any poor wording, but my point is that the better the system is before anyone touches it, the less arguments will come up.

LoreKeeper wrote:

I also think this is a good point to address the issue about eligibility to use stealth:

Current:
You cannot use stealth without at least partial cover or concealment. So you cannot use it in the middle of a courtyard in broad daylight.

The intention is obvious enough, but it creates special cases, for example, a dwarf can see you stealthing in the night, you don't have concealment from him due to dark vision.

Instead:
You can use stealth at any time. Creatures against whom you don't have at least partial cover or concealment automatically succeed on opposed rolls to perceive you.

Putting the onus on the observer opens up the stealth skill for practical use without a big list of special cases.

I'd go a little farther. I'd just give anyone who can observe you a +30 to notice you. I've actually seen people get surprised, even when they are watching for someone. Just because they start to get bored and daydream. So if you roll a stealth 38 check, and the bored guard rolls a 1 and get's 37 (+6 perception), he might still not notice you, cause he's imagining the bar girl and his ale when he get's off duty.

That's an interesting approach. On page 1 somewhere I advocate the use of Sense Distraction: the ability to use Sense Motive to figure out that a particular target is not paying attention to you right now, allowing you to be considered concealed vs that target for your next action.

Jiggy wrote:

Staying invisible does not cause the spell to not end - anyone with English as their primary language can see that. The spell being active and the creature being invisible are not the same thing, any more than standing in the rain and getting wet are the same thing; sure standing in the rain will cause you to get wet, but that doesn't mean that if you walk inside and jump in a pool that you're somehow still in the rain just because you're still getting wet. In the same way, the spell being active causes you to be invisible, but that doesn't mean that continuing the effect in spite of the spell ending will cause the spell to not end.

(edited, slight qualifications and clarifications attempted)

Standing in the rain may not cause a person to get wet if that person has an umbrella.
However, assume for a moment that standing in the rain is certain to get a person wet. Assume also for a moment that jumping in an indoor pool will get that same person wet in exactly the same way. Being wet, here, is a state, and (unless you count minor chemical variations, such as possibly chlorinated pool water) the person is in precisely the same state if they had jumped in the pool as if they had stood in the rain.
To extend that example further, you could be quoted as saying that a person who makes successfully a Stealth check gains a state of being invisible and that that state of being invisible as a result of a spell is exactly the same, down to the last bending of a light ray as if a magical spell had made them that way (and presumably with same same result for anyone attempting to detect magic in the vicinity.

And no, the language we're dealing with is not English here; we're talking about pure rules text, a fantastic argot which bends the fabric of reality, at times defies mere common sense and drives the incautious stark raving mad.

And thank-you for the rain/swimming-pool example. It was a rather fine one, and I feel that it precisely makes my point...

To be clear, my rain analogy was not intended to suggest that there was nothing wrong with stealth providing invisibility - I'm in the camp that says we need some other result (either a new "condition" or just a description of what happens). Rather, it only to specifically point out that "staying invisible" does not imply "preventing the spell invisibility from ending". You had said that a strict or rules lawyering reading of the proposed stealth rules could be construed to mean that, and it can't. That's the only point that my rain analogy was meant to refute.

That's an interesting approach. On page 1 somewhere I advocate the use of Sense Distraction: the ability to use Sense Motive to figure out that a particular target is not paying attention to you right now, allowing you to be considered concealed vs that target for your next action.

Yeah,

Diego Rossi wrote:

In the guard scenario there is a strange effect:

- the rogue hide behind the pillar after attacking
- the guard move in a location where the pillar don't conceal the rogue but fail to beat his stealth roll, so he don't see the rogue
- the rogue turn start, he need to move to attack the guard, but he can't hide as he is not benefiting from concealment. As soon as he do anything he is spotted automatically.
only if he has a way to attack the guard as his first action (without drawing a weapon or any other action that require him to do a stealth check) he can make a sneak attack.

....

The Rogue make a fizzling sound when his pseudo invisibility end? :)

Agreed. It doesn't make sense. And this situation is simply two people and a pillar involved in bright light where the guard has perceived the rogue at the start of it all. There are almost always MUCH more things involved, and just so much stuff can be abused by loosely worded rules.

I think this line of discussion is one of the more important ones going on at the moment. The proposed new rules solve the pillar-hopping issue (i.e., you can sneak across gaps between two sources of cover without auto-failing now), but replace it with the issue of successfully hiding but then having your cover removed without losing your stealth (i.e., guard walks to your side of the pillar and fails his perception check... somehow). Similarly, you could be hidden behind the curtain, then have the guard spend a move action to walk to the curtain and another move to pull it back, now staring you in the face - but since he doesn't have another move to search your square, he doesn't notice you.

I think the crux is this:
There needs to be a method of verifying whether or not you're still hidden that allows for pillar-hopping but still reveals you if conditions change such that you're now right in front of the guard's face.

Perhaps (starting from the rules in the blog post) we could add that any creature that takes an action which causes you to no longer have cover or concealment from that creature automatically notices you?

That would cover it, right?

Yeah,

the specific instance I'm thinking of when I posted that was, me in the airport, riding down an escalator. My wife is at the bottom of the escalator, looking up watching for me. She's got that 'I'm thinking of something else' look on her face while she's watching. I literally walked up to her, got 2 feet away before she suddenly jumped and nearly screamed as she realized I was staring at her from 2 feet away, with a grin on my face (while she was still looking up the escalator for me to exit the gate area). :)

Personally, I think this should best be handled by a GM adjudicating a perception penalty due to daydreaming. Probably not something to try to cover in the Stealth skill description, for simplicity's sake.

I have three major concerns about this rewrite:

1) Using invisibility as the "successful stealth state" has quite a few potentially nasty rules consequences.
2) Disallowing any standard action is overkill. There's no reason we shouldn't be able to use skills like Sleight of Hand stealthily.
3) Scent still needs to be addressed with respect to its interactions with unseen characters.

To sum up a few thoughts...

The invisible condition specifically only covers vision. If this is all Stealth does, then Stealth is, for most practical purposes, useless.

Blanket-banning action types is a simple, but hardly an elegant, solution. There's lots of thing which require a standard action, for example - drawing a concealed weapon shouldn't automatically negate Stealth.

With no rules for facing, Stealth becomes of very limited use. You can't sneak up behind people unless they happen to be standing within half your move of cover. Halfling Rogues everywhere will be crying with their '10ft Stealth range'.

Ruling that Stealth can only be 'activated' at the start of a related action quickly becomes weird: moving behind a pillar doesn't let you go into Stealth... unless you take a free action halfway through to drop a pebble...

... In general the blog rules seem to be more of an attempt to pin down how Stealth works in combat situations, rather than in general situations. However, in the game, I'd guess that a great deal of Stealth is about not having to get into a combat situation in the first place. Maybe this one's a misperception on my part, but it seems this way to me.

I guess I'm in the minority on this, but I really, really like using the Invisible state with stealth. Invisibility increasing that then makes a lot more sense mechanically (to me, at least)... rather then creating another state for essentially the same thing (in regards to combat someone is either aware of you or they are not)this really simplifies things.

You're not alone. I like Invisible being used as the condition. The wording for Invisibility isn't terribly complex. Being hidden is just another way to be invisible, but thats just how it read to me.

With no rules for facing, Stealth becomes of very limited use. You can't sneak up behind people unless they happen to be standing within half your move of cover. Halfling Rogues everywhere will be crying with their '10ft Stealth range'.

Under the current rules as written, sneaking up behind someone is impossible.

To summarize the main points made in this thread so far:

• There's a nearly (entirely?) unanimous sentiment that the condition granted by successful stealth needs to be something other than invisibility.

• Sneaking from one point of cover across open sight to another point of cover - provided you can do it in a single action - no longer auto-fails, and it seems several people already houserule it this way anyway.

• Wording needs cleaning up (but of course that's to be expected at this point).

• Most people bothering to share like the overall "meat" of this re-write and support the changes.

Just saw this. Should have read to the end of the thread before making that previous post replying to the earlier one. Ah well. I had fun writing it anyway...

I don't like using invisibility. I don't think we need a new status either.

One thing to keep in mind as the discussion continues (and thank you for the great discussion, everyone!) is that we are using the invisible condition (Core Rulebook page 567) not the invisibility special ability as described in spell descriptions and on pages 563-564, though it is a goal of mine to make sure that in the final Stealth fix that those rules work together in clear and meaningful ways.

One of the things that we considered early on is either changing that condition to hidden or creating an entirely new hidden condition. We decided against that for this playtest so we could see how it plays just using the existing condition and what people thought of it.

You continued input on that is much appreciated.

The reason why we need to use a condition, is that the "use Stealth" as a condition in the game is, well...a stealth condition. It is not defined. That is one of the chief things we need to change in the Stealth rules. What does "use Stealth," mean other than in the natural language sense of the term. The need for this came from the many, many, many FAQ questions we have received on the subject.

Now for some other notes and observations based on some of the discussion here.

For those of you concerned about hiding in crowds, as it states in the Core Rulebook page 436, crowds provide you cover, not soft cover, and it already says that you can use a crowd for to enable a Stealth check.

Soft cover never allowed you to use Stealth (Core Rulebook page 196) and we are not going to change that.

As far as the action economy part of the equation, we will be reviewing some of the corner-case standard actions, and may make more exceptions to standard action you can use Stealth with, but I can almost guarantee that using spell-like abilities will not be on that list. That would be a big boost to Stealth using monsters, and I don't even want to think about what that would do to some encounters. ::shudder::

We have taken into account the various hide in plain sight abilities, and yes, it turns out that hide in plain sight is and always was really good.

Just some things to keep in mind as the playtest and the discussion moves forward.

ProfPotts wrote:

With no rules for facing, Stealth becomes of very limited use. You can't sneak up behind people unless they happen to be standing within half your move of cover. Halfling Rogues everywhere will be crying with their '10ft Stealth range'.

Under the current rules as written, sneaking up behind someone is impossible.

Also, fast stealth just became a pretty potent rogue talent.

One thing to keep in mind as the discussion continues (and thank you for the great discussion, everyone!) is that we are using the invisible condition (Core Rulebook page 567) not the invisibility special ability as described in spell descriptions and on pages 563-564, though it is a goal of mine to make sure that in the final Stealth fix that those rules work together in clear and meaningful ways.

Even there, though, invisible is defined as being specifically sight-based. (The exact phrasing is "visually indetectible.")

Someone needs to speak up for the insects.

"A creature larger or smaller than Medium takes a size bonus or penalty on Stealth checks depending on its size category: Fine +16, Diminutive +12, Tiny +8, Small +4, Large –4, Huge –8, Gargantuan –12, Colossal –16."

This seems to me to mean that a Fine creature finds it really, really easy to hide from just about anything. And, that's okay. But, it seems to me that some Fine creatures hunt other Fine creatures, and some of the Fine prey use Stealth but aren't nearly as successful when trying to hide from other Fine creatures as they are from trying to hide from, say, Medium creatures.

I understand that the skill needs to work, first of all, for human-sized creatures. And, one (ugly) way of dealing with this is to also modify Perception so that Fine creatures are really amazing at spotting things.

Alas, I think there might be something else going on here.

Oh, the thread is fantastic, and I'm glad the designers are taking the issue of revision of even a single skill very seriously.

Halflings, being small, also get a +4 size bonus to Stealth, so they can sneak more reliably. I think the reduced movement speed is a fair trade-off, and gives incentive for medium-sized races to play sneaky characters.

Also, keep in mind just how far 10ft. really is in real life. To cross that distance in about three or four seconds without your target hearing you coming, and quick enough so that your target doesn't chance looking around - all while being about half the size of a human, is quite a feat in and of itself.

Under the current rules as written, sneaking up behind someone is impossible.

I'm pretty sure I didn't claim otherwise... but it is something the Stealth rules need to be able to simulate. You don't even need combat rules on facing, where everyone is assumed to be looking about as part of the fact there's a big fight going on; you just need a note in the Stealth rules about being able to do it in some way, shape, or form.

Someone needs to speak up for the insects.

"A creature larger or smaller than Medium takes a size bonus or penalty on Stealth checks depending on its size category: Fine +16, Diminutive +12, Tiny +8, Small +4, Large –4, Huge –8, Gargantuan –12, Colossal –16."

This seems to me to mean that a Fine creature finds it really, really easy to hide from just about anything. And, that's okay. But, it seems to me that some Fine creatures hunt other Fine creatures, and some of the Fine prey use Stealth but aren't nearly as successful when trying to hide from other Fine creatures as they are from trying to hide from, say, Medium creatures.

I understand that the skill needs to work, first of all, for human-sized creatures. And, one (ugly) way of dealing with this is to also modify Perception so that Fine creatures are really amazing at spotting things.

Alas, I think there might be something else going on here.

Oh, the thread is fantastic, and I'm glad the designers are taking the issue of revision of even a single skill very seriously.

Alas, using Pathfinder to model my insect hunting based campaign will never work! My players will weep when I tell them they will never be able to play as rhinoceros beetles and cicadas.

(Just being obnoxious. Ignore me!)

Someone needs to speak up for the insects.

"A creature larger or smaller than Medium takes a size bonus or penalty on Stealth checks depending on its size category: Fine +16, Diminutive +12, Tiny +8, Small +4, Large –4, Huge –8, Gargantuan –12, Colossal –16."

This seems to me to mean that a Fine creature finds it really, really easy to hide from just about anything. And, that's okay. But, it seems to me that some Fine creatures hunt other Fine creatures, and some of the Fine prey use Stealth but aren't nearly as successful when trying to hide from other Fine creatures as they are from trying to hide from, say, Medium creatures.

The more technically correct way of doing it from a simulationist standpoint would be to base it on differences in size. So, for each reduction in size category between observer and observee, then give the observee a +4 bonus, for example. And each increase gives a -4 penalty.

However, that becomes a nightmare to keep track of. Probably the easiest way to do it would be to add a line that says 'size modifiers do not apply if the observer and observee are the same size'.

To sum up a few thoughts...

The invisible condition specifically only covers vision. If this is all Stealth does, then Stealth is, for most practical purposes, useless.

Blanket-banning action types is a simple, but hardly an elegant, solution. There's lots of thing which require a standard action, for example - drawing a concealed weapon shouldn't automatically negate Stealth.

With no rules for facing, Stealth becomes of very limited use. You can't sneak up behind people unless they happen to be standing within half your move of cover. Halfling Rogues everywhere will be crying with their '10ft Stealth range'.

Ruling that Stealth can only be 'activated' at the start of a related action quickly becomes weird: moving behind a pillar doesn't let you go into Stealth... unless you take a free action halfway through to drop a pebble...

... In general the blog rules seem to be more of an attempt to pin down how Stealth works in combat situations, rather than in general situations. However, in the game, I'd guess that a great deal of Stealth is about not having to get into a combat situation in the first place. Maybe this one's a misperception on my part, but it seems this way to me.

I think you're right on the money. Stealth is more than just being hidden, and invisibility doesn't do anything for you to be quieter.

Banning standard actions wholly seems to me to be an afterthought tacked on the entire concept of using Stealth as part of the beginning of an action if you meet the correct conditions - otherwise you'd have rogues running around and using standards that don't make sense to grant Stealth; though it does still come into play with the current blanket bans, it's much less so than the current write up. This is why it probably is simply best to rewrite Stealth rules entirely, though that is obviously a large amount of work Paizo has already stated they won't do. I don't really mind that so much, I'm just glad they're up for the changes.

Facing is, of course, a stealther's best friend. After all, any time you see people trying to be sneaky, they try to take advantage of humans' inability to perceive in all directions at once. A firm "No." to these rules will always hurt anyone who wants to use Stealth as effectively as possible. It also makes the game much less complicated, which is always nice. I know in most of the games I've played, though, we use facing for out-of-combat scenarios when someone is sneaking, simply because it makes no sense not to.

Using Stealth in combat is quite complicated. I've spent quite a bit of time thinking about it (we had a Shadow Dancer in a campaign that ended in a banning of the class, due to the pure amount anti-logic involved with Hide In Plain Sight), and I think there needs to be a separate set of rules for such situations. Kom Bat McRogue who's spent all of his time learning to sneak around a battlefield effectively should be able to do so, but those skills are not necessarily the same as Jack the Hungry who simply wants to thieve a dinner from Farmer-o con Shotgun-o and Sleeping Dog. And I use the term "skills" not in the concept of what's on our character sheet, but more on how what's on our character sheet is used.

I don't understand why you want to make things so complected. Seriously, how do you get through all the action when you have to be concerned with things so obvious they don't need rules (like a Colossal modifier? Seriously, that should not be able to sneak past something that is relevant to the scene!)