Change paralysis

[Sandman25]

Some characters are extremely unlucky with gear and literally have no items with MR, intrinsic MR is not enough for many species.
There is nothing a player can do when it turns around a corner and gets paralyzed so I suggest to either remove paralysis or make it similar to Petrify (new "Paralyzing" status which can be cancelled by a potion of cancellation, or just gives you last action which can be reading scroll of summoning or whatever). Petrify at least decreases conventional damage, with Paralyze you are doomed if your character relied on EV/SH.

[1010011010]

I think this is an MR issue not paralysis.
Quick solution:add MR to ?res.

[esran]

Quick fix: Make paralyze not reudce SH or EV, or at least not reduce them as much. That would make paralysis bad, but not the instant one shot killer it currently is.

[XuaXua]

Quick fix: Make paralyze not reudce SH or EV, or at least not reduce them as much. That would make paralysis bad, but not the instant one shot killer it currently is.

SH though is the ability to move your shield into place to defend.

EV is the ability to get out of the way.

If you can't move, it makes sense to reduce EV to 0.

If you can't move your shield, chances are it's at best stuck partially blocking and could have a major reduction, potentially down to 0 for small shields like bucklers.

[esran]

Quick fix: Make paralyze not reudce SH or EV, or at least not reduce them as much. That would make paralysis bad, but not the instant one shot killer it currently is.

SH though is the ability to move your shield into place to defend.

EV is the ability to get out of the way.

If you can't move, it makes sense to reduce EV to 0.

If you can't move your shield, chances are it's at best stuck partially blocking and could have a major reduction, potentially down to 0 for small shields like bucklers.

But if you're wearing a large shield, it really shouldn't reduce it down to zero. There are way less places to hit you with a giant shield as big as you are in the way. Just something to consider, but what if it only halved SH for large shields, thirded it for shields, and reduced it to zero for bucklers?

[khalil]

muh flavour

I'd prefer a game that is fun to a game that is flavourful, thank you very much.

[Jeremiah]

The game already has 2 effects that work quite similarly to paralysis: sleep and petrification. Both are less arbitrarily fatal than paralysis because you wake from sleep when you get hit, and petrification has a slowdown period during which time you can take emergency measures.

I'd say paralysis could just be removed and replaced by one of these other effects.

[Sandman25]

I think this is an MR issue not paralysis.
Quick solution:add MR to ?res.

The potion already grants MR.

[bananaken]

I think this is an MR issue not paralysis.
Quick solution:add MR to ?res.

If you could just prepare before any situation where paralysis is involved it would be a non-issue, it'd just be petrification. Paralysis causes stuff like this to happen:



(ignore the turn count, was just experimenting with mummies + Xom )

It's unlikely, but it's not a fun way to die.

edit: You could have paralysis kick in gradually (maybe faster than Petrification), and after a couple turns you'd be paralyzed the same way it currently works. This gives Rupert/Ancient Liches/etc. new scenarios where they have time to clean up summons or close the distance on you before the full effect kicks in, being potentially *more* dangerous in cases where you don't have the means to escape, while allowing you to always take at least one emergency measure before being fully paralyzed.

[phobetor]

What about replacing paralysis with stun. Make it more likely to effect you but have it only last 1-2 turns. This would be less likely to outright kill you but could still have a large impact on combat.

[duvessa]

Sleep as currently implemented is a really dumb effect (iirc it doesn't reduce EV and it gives monsters 2.5x melee damage).

I don't think petrify is an improvement over paralysis - in fact, I think it's a much worse-designed effect because it has some very non-obvious effects like the damage reduction. There's not very much you can do with the 1 or 2 actions you get, just cancel it or teleport or slightly improve your position, so it doesn't really do much to offer players more options. Of course most player hexes are similarly boring; confusion is 1-turn paralysis if you are carrying curing and really long paralysis if you aren't, then there's the long list of effects that just block off some group of actions (moving, reading/casting, melee attacks, etc). The only one I can think of that I actually consider interesting is slow.

[Wahaha]

The point of hexes and effects that can be resisted by magic resistance, the reason they exist, is to sometimes affect the player. Players should not expect to always be able to resist every hex, or else what's the point? By the way the opposite is also bad, looking at mark in particular. So the premise of this thread doesn't make sense: "sometimes characters are affected by paralysis!". If paralysis is too strong then that's another thing. I think it could last one or two turns less on average. Alternatively, make paralysis take 1 turn to activate like petrify does now (like what is suggested in the op). Petrify can activate instantly, have a significantly shorter duration, and after unpetrifying the player can be slowed and immune to petrify for a few turns. I think an immediate effect for petrify is ok because it gives AC or something like AC? and if it gets a shorter duration.

Sleep as currently implemented is a really dumb effect (iirc it doesn't reduce EV

If it reduced EV to 0 it would be a good effect in my opinion.

[bananaken]

The point of hexes and effects that can be resisted by magic resistance, the reason they exist, is to sometimes affect the player. Players should not expect to always be able to resist every hex, or else what's the point? By the way the opposite is also bad, looking at mark in particular. So the premise of this thread doesn't make sense: "sometimes characters are affected by paralysis!". If paralysis is too strong then that's another thing. I think it could last one or two turns less on average. Alternatively, make paralysis take 1 turn to activate like petrify does now (like what is suggested in the op). Petrify can activate instantly, have a significantly shorter duration, and after unpetrifying the player can be slowed and immune to petrify for a few turns. I think an immediate effect for petrify is ok because it gives AC or something like AC? and if it gets a shorter duration.

The OP is not whining that he can't be omniresistant to hexes, he's just pointing out that despite paralysis becoming a potential instadeath you sometimes just don't have the means to resist it or even mitigate the risk. Sometimes you're just unlucky and get paralyzed as you run into an enemy that can cast it, and you die without a chance to react.

Paralysis taking at least a turn to kick in would be enough, that or paralysis kicking in at the end of your next action. The feeling of danger is still there, and you'll avoid the "instadeath by auto-travel" scenarios like in my screenshot (which I also think the OP is referring to). There's other ways to fix this, but it would be the easiest change, I imagine.

[njvack]

Paralysis causes stuff like this to happen:

Well, really, autoexploring into melee range with Rupert can cause this, paralysis or no. Autoexploring into some other monsters (Nikola, anyone?) can be an instant death sentence, too.

My main beef with paralysis is that I find extended loss of control of my character — as in, I can't send any input at all — to be more frustrating even than the death that may follow. I've never minded confusion on mummies in the same way, even though it's nearly as deadly. Well, maybe more, as it can be chained and lasts longer. And it's not quite identical: you can draw from decks (and maybe use other evokers?). You can randomly blast wands. You can try disto unwields. You can even try moving.

I'm not saying those are actually practical lifesaving tricks for most confused mummies. But being able to take an action, even if a probably useless action, subjectively feels better than just scrolling through --more-- while monsters spend six turns pounding you to death.

The sleep-EV thing sounds like a straight bug to me, though it's pretty funny.

[bananaken]

Paralysis causes stuff like this to happen:

Well, really, autoexploring into melee range with Rupert can cause this, paralysis or no. Autoexploring into some other monsters (Nikola, anyone?) can be an instant death sentence, too.

Ok yes, hypothetically it could happen. But let's be honest here, it's paralysis that makes these instadeath scenarios that much more likely. I could replace Rupert with an 8-headed hydra and I would like my chances at living. If Rupert did anything except Paralysis I would live, period (I even had an amulet of clarity equipped). What would he do, Berserk? Ok I'll just blink away. Hit me? Without berserk his +0 battleaxe of chopping was not going to kill me instantly, ever. Nikola at the very least appears later in the dungeon, once you've had a chance to loot and level up thanks to Lair and Orcish Mines.

Another, more sweeping solution could be to simply prevent monsters from getting the first action upon encountering (think XCOM: Enemy Unknown when the enemy first spots you). Maybe they can get additional energy to act next turn to compensate.. but speed randomization feels like a touchy subject in itself. In any case, I definitely think autoexploring shouldn't ever lead to an enemy killing you instantly, and I don't think anyone would go like "well, you deserve it!" for hitting "o" and dying in Lair, or anywhere for that matter. Maybe you can get into an exceedingly shitty situation for auto-exploring sure, but that's normal and even expected depending on the branch.

I agree with extended loss of character control being frustrating, but really, if you can realistically trace back the events leading to your death-by-paralysis and go like "d'oh, I should have done *this* instead", then Paralysis would be fine to me.

[crate]

There is a topic in the past suggesting replacing parlysis with petrify. I go through a long explanation about why paralysis is actually a most excellent monster effect in that topic. I'll try to go find my post.

edit found it: viewtopic.php?p=150027#p150027

[Sandman25]

There is a topic in the past suggesting replacing parlysis with petrify. I go through a long explanation about why paralysis is actually a most excellent monster effect in that topic. I'll try to go find my post.

edit found it: viewtopic.php?p=150027#p150027

I think the fact that you cannot do anything after getting paralysed is interesting, precisely because it forces you to pre-emptively avoid getting paraylsed in a situation where becoming paralysed will kill you.

The spell is great because it does not give player a chance to react, players cannot correct their mistakes by subsequent actions. Use of LoF blockers, fog, blink etc., it can be interesting indeed

I acknowledge that getting paralysed without giving the player any way of avoiding it (see my post above) does happen sometimes from monsters who have the spell. If you can think of a good way to eliminate this let me know.

The spell is not great because it does not give any chance to react (or there is nothing a player can do to avoid being paralyzed due to lack of items, Trog, summon spells etc.).

Probably a good solution would be to make monsters unable to use the spell first 1-3 actions after noticing player. Players will be responsible for their death.

[damiac]

Well, that's nice and all, but it doesn't address instadeaths to autoexploring into a paralyzer.

Plus, I completely disagree that pressing 'more' a few times and watching my health go down is interesting. On the other hand, getting hit by petrification leads immediately to interesting situations, at least, more interesting than 'press space until RNG is satisfied'.

The suggestion of a special case in monster AI to just not cast paralyze (I'd also include banishing here) on the first turn you see them seems like a good idea. I cannot do a single thing to control what happens in the unexplored dungeon as I autoexplore into it. Even if I don't autoexplore I could run into the exact situation in this topic. I typically have as much MR on me as the dungeon will provide, and that's usually nowhere near enough to be 'safe' from paralysis and banishment.

Having +MR on potions of resistance is useless in this scenario, unless you're also suggesting to generate so many potions of resistance that I can constantly be under their effect while exploring.

[twelwe]

Sleep as currently implemented is a really dumb effect (iirc it doesn't reduce EV and it gives monsters 2.5x melee damage).

i can punch and kick in my sleep, and if someone is throwing at me i could probably dodge while sleeping too no problem

[crate]

Your "The spell is (not) great because it does not give any chance to react" is not being used correctly.

You never get a chance to react to paralysis, and that's fine. That's literally the entire reason paralysis exists. What you do get, in most cases, is a chance to "pre"-act; you pre-emptively position yourself such that paralysis will not kill you. You get this chance (or it happens automatically, since things at edge of los are typically not dangerous) in almost all paraylsis encounters.

I do not think that "sphinx appears right around a corner 2 spaces away and immediately paralyses you" is good gameplay, but I am willing to say that it is an acceptable tradeoff for the good that paralysis does in the large majority of situations where it exists, since any thoughts I've had about ways to limit that behaviour don't really work.

("Sphinx comes into view 8 spaces away and immediately paraylses you" is fine. Also, squarelos would help out here since then LOS would be the same in every direction, at least.)

I don't like solutions that say "the monster can't cast this spell right away". For one thing, it promotes either: 1) you need to actually track which particular ogre-magi you have seen (if it is a one-time-only thing); or 2) it promotes breaking LOS to reset the paralysis immunity (if it's not one-time-only). Neither of those seem acceptable to me, and I don't like limiting spells in that fashion anyway (though the other reasons I have are not examples of things that are objectively bad so I do not bring them up here.)

[crate]

Plus, I completely disagree that pressing 'more' a few times and watching my health go down is interesting. On the other hand, getting hit by petrification leads immediately to interesting situations, at least, more interesting than 'press space until RNG is satisfied'.

See I think the exact opposite: petrification is wimpy because I can undo it so I just don't even care if it happens to me (it's weaker than slow a lot of the time!); paralysis forces me to do things so it doesn't happen in the first place. If you do not like paralysis that is fine, but I have pointed out that it does something that no other status effect in crawl does. It is not objectively true that having paraylsis is better than not having it, but it is unique (and there are indications that at least some of the devs agree with me that paraylsis is worth having). Petrification, not as unique (as my linked post indicates).

edit: again, it's entirely possible to not die to paralysis, unless you think for some reason that I am tremendously lucky. Other than grinder paralysing me (and I guess some desperation quaff-ID situations that aren't paralysis-caused deaths) my paralysis-related deaths were all situations where I knew I was doing something dumb, did it anyway, and died for it. Can it cause unpreventable deaths? Ok, yeah, in theory it can, but in practice it's a lot rarer than lots of other things, and I think this is perfectly acceptable.

I guess vp knights could end up working like Grinder, I'm unsure how I feel about vp knights (they're later in the game so they're not quite the same as grinder). I don't really like grinder but the problem there is that he has blink more than anything else.

edit again: I do agree with the OP that it's a bit silly how much stronger paralysis is against EV than AC, but the easy fix there is to make it not affect your EV or something (it's a bit unclear to players that it has an effect on EV at all, since I'm not even sure when the EV value gets re-displayed on the screen if it does at all.) But that's not a problem with the idea of paralysis but instead the implementation.

[bananaken]

I don't like solutions that say "the monster can't cast this spell right away". For one thing, it promotes either: 1) you need to actually track which particular ogre-magi you have seen (if it is a one-time-only thing); or 2) it promotes breaking LOS to reset the paralysis immunity (if it's not one-time-only). Neither of those seem acceptable to me, and I don't like limiting spells in that fashion anyway (though the other reasons I have are not examples of things that are objectively bad so I do not bring them up here.)

1) doesn't have to work this way, you may simply compromise and limit this change to uniques. Uniques with Paralyze are the main culprits I think; they tend to appear the most when you simply lack the levels and MR equipment to do something about it (Grinder, Erolcha and Rupert come to mind). Rupert's main gimmick is to Paralyze/Confuse you immediately and go berserk, so he's more likely to cause an instadeath. I don't feel like we need to be "fair" and affect every enemy including non-uniques. Ogre Magi thankfully appear later than Rupert does (if you never do Orc before Lair) and have 5 different spellsets to "dilute" the chances even further of a silly instadeath. Once you move on to Sphinxes, Liches, Ancient Liches, etc. I feel like the chances go down further of an actual unpreventable death with increasing HP, MR and better equipment.

Blocking the "free action" upon an initial encounter (because I imagine this has some hidden energy cost), then giving some extra energy for their next actions seems like a better solution, but just blocking just Paralysis would work too.

I don't think anyone wants or has suggested 2), it'd be very obviously abusable through spoilery knowledge, while knowing about 1) wouldn't change anything about your tactics, you'd just be alive in some cases where you'd be unavoidably dead.

[duvessa]

I don't like paralysis on things like ogre magi since they are really common, easily avoidable, and it just makes it tedious to fight them safely (same for the wand - and monster consumable usage in general), but I see no problem with paralysis on Rupert. Sure it can very, very rarely cause an autoexplore death, but there are much more common autoexplore deaths all over* - if you actually just want to fix those you should propose something like making shouting use up energy for monsters. It sounds more like you dislike paralysis for some other, not well specified reason.

*the only unique that I'd change because of autoexplore deaths is the Lernaean hydra, and that's because the Lernaean hydra is completely harmless except for causing autoexplore deaths which is really dumb

[bananaken]

It sounds more like you dislike paralysis for some other, not well specified reason.

Why is that? I literally just suggested enemies not getting free actions (not just Paralysis). Blocking only Paralysis was not my idea. I don't like the idea of auto-explore deaths in general, but the one with Rupert + Paralysis was the only example I had in hand to show you guys and make a case for.

Sure it can very, very rarely cause an autoexplore death, but there are much more common autoexplore deaths all over* - if you actually just want to fix those you should propose something like making shouting use up energy for monsters.

Aren't we trying to suggest the same thing, then? Upon an initial encounter, I'd like to prevent the enemy from getting a free action. If they use up energy when shouting (i.e. upon an initial encounter), then this would effectively achieve the same thing, no? (minus giving them extra energy for the following turn)

[duvessa]

It sounds more like you dislike paralysis for some other, not well specified reason.

Why is that? I literally just suggested enemies not getting free actions (not just Paralysis). Blocking only Paralysis was not my idea. I don't like the idea of auto-explore deaths in general, but the one with Rupert + Paralysis was the only example I had in hand to show you guys and make a case for.

Several people in this thread suggested changing paralysis itself.

[bananaken]

Several people in this thread suggested changing paralysis itself.

I think it's an ok suggestion.. even if there's great points against it, it's thanks to the discussion it spurred that I learned other people dislike auto-explore deaths in the first place. That said, if the playerbase + developers see Paralysis as a "sacred cow" (for lack of a better term), then probably the only way to go about this is avoiding the "auto-explore death" scenario. I can see stuff like bumping into an Ancient Lich and getting hit by an unlucky crystal spear or TLH being a thing, but I didn't want to assume those things happened without experiencing them, for all I know their AI could be handled differently.

My question then is: would making the enemy's reaction to initial encounters more deterministic make the game objectively better or worse? I'm guessing from previous posts that the act of an enemy "shout" is synonymous to an initial encounter with the player, in which case you could add a few lines of code there to deduct enough energy to always allow the player to act first, basically removing the chance of instadeath. From there on it's up to the player to not die in stupid ways.

I think it would be fine and make the game better. Even if it's something that won't occur in 100 games or even 1000 games to certain people, it's clear there's people here unlucky enough with RNG to have this happen to them. Maybe I'm not seeing some adverse consequence to making this change though, so who knows.

[n1000]

I vaguely think that reworking MR would be more fruitful than removing paralysis. I don't have any idea how to do such a thing but somehow the current implementation doesn't feel perfect. Maybe it's the either/or nature of hexes: You cannot partially resist paralysis, shortening its duration, by having higher MR, for example. I am not sure whether such a system would be an improvement. Indeed, the fact that hexes work this way is what makes them unique.

[Quazifuji]

I vaguely think that reworking MR would be more fruitful than removing paralysis. I don't have any idea how to do such a thing but somehow the current implementation doesn't feel perfect. Maybe it's the either/or nature of hexes: You cannot partially resist paralysis, shortening its duration, by having higher MR, for example. I am not sure whether such a system would be an improvement. Indeed, the fact that hexes work this way is what makes them unique.

I'm definitely in favor of making hexes less binary. For most hexes just having the duration affected by resistance would help, the main exception is banishment. For me, one of the most frustrating parts of hexes is that there's no amount of MR where you can truly feel safe for most hexes. The standard solution to encountering a banisher or paralyzer is to load up on MR, but there's a chance that'll all turn out for nothing and you'll just get banished or paralyzed anyway. I find that frustrating.

Another possibility is one I've mentioned before, but never seen a response to (although maybe that's just because it's a bad idea no one felt was worth responding to): make resisted hexes give a temporary MR reduction (like a weaker ?vuln). Possibly reduce the effectiveness of hexes against a given MR value to compensate. So it would be much easier to resist a hex the first time, but every time you do it becomes more likely to hit next time. Makes hexes a bit less luck based. Player hexes could also be changed this way (while having player hex duration vary by MR doesn't work as well because for some hexes the player often only needs to get one or two turns out of them anyway (if they're using them to stab).

[Sandman25]

Another possibility is one I've mentioned before, but never seen a response to (although maybe that's just because it's a bad idea no one felt was worth responding to): make resisted hexes give a temporary MR reduction (like a weaker ?vuln). Possibly reduce the effectiveness of hexes against a given MR value to compensate. So it would be much easier to resist a hex the first time, but every time you do it becomes more likely to hit next time. Makes hexes a bit less luck based.

I think it would make hexes more luck based. Having Ogre Magi use banishment 3 times in a row would be similar to having an Orc Priest use smite 3 times in a row.

[crate]

Fwiw the chance of you dying to edge-of-los (in a cardinal direction, at least, you can blame circlelos for diagonal ruperts) rupert doing para + zerk as his first two actions is still pretty small. Crawl los is, what, 8 spaces? Max para dur is 7, so 6 left after rupert uses his zerk, it then takes him 7/1.5=4.6 turns to get to you, so he gets 1-2 attacks on you before you get to act. And he only gets that something like 20% of the time. Even at 7 spaces away it's very unlikely for you to die to Rupert without getting actions even if we assume he starts with para (success) + zerk.

(Blame circlelos being really dumb for diagonal deaths being a thing here; I suspect changing to squarelos of 7 spaces would prevent quite a few of the "unavoidable" paralysis deaths that aren't Grinder simply by making the minimum edge-of-los bigger.)

Grinder and vp knights are much more relevant examples if you want to complain, since they both have blink. Grinder might actually even be more dangerous without blink, to be fair (he also happens to be a big wall-of-stats and appears at depths where you are reasonably likely to not have escape consumables). Paralysis instadeath problems are very much not limited to uniques.

edit: fixed bad math

[Quazifuji]

Another possibility is one I've mentioned before, but never seen a response to (although maybe that's just because it's a bad idea no one felt was worth responding to): make resisted hexes give a temporary MR reduction (like a weaker ?vuln). Possibly reduce the effectiveness of hexes against a given MR value to compensate. So it would be much easier to resist a hex the first time, but every time you do it becomes more likely to hit next time. Makes hexes a bit less luck based.

I think it would make hexes more luck based. Having Ogre Magi use banishment 3 times in a row would be similar to having an Orc Priest use smite 3 times in a row.

There's always luck when spellcasters are involved because their AI might randomly actually do something really smart and dangerous. But my thinking is that right now, it's just every turn within line of fire of an Ogre Mage you have a chance to get banished unless you've determined they don't have banishment. With the change, that chance would be low or nonexistent until they'd tried it a couple times, at which point you'd be well aware you're in heavy danger of banishment and can flee. Right now, I find that tiny but not zero chance of getting paralyzed or banished by an ogre mage when you have moderate MR frustrating. Thinks with a tiny but not zero chance of happening are the most tedious and luck based, because while it's technically correct to play assuming they might happen at any time, it's also mostly unnecessary. Things are much more interesting if they're more likely so you know you have to play around them, or completely impossible so you can ignore them. That's why I prefer the idea of banishment being, say, a 0% chance to succeed the first time, 20% the second time, 50% the third time, etc, instead of just a 5% chance every time.

[damiac]

If you look at the screenshot in this thread, the issue isn't LOS size, it's that you can be right next to a guy when you first see him. If you always saw enemies at the edge of LOS then it's true that first turn paralysis wouldn't be such a big deal. Paralyze and banish are just really terrible things that can happen to you in a single turn, so they're the biggest focus of complaints of out of LOS threats.

I cannot have more MR than the dungeon has provided. I cannot know what's outside of LOS. I cannot control what the enemies outside of LOS will do when I first spot them. I will not stop using autoexplore, and the design doc is with me there, as I shouldn't have to perform tedious manual exploration to gain a slight advantage of survival.

[crate]

Paralysis is a lot different from banish: banish shares the no-reaction thing that paralysis has, but it also has the effect of completely not caring about your character's positioning, so the pre-action you can take to mitigate paralysis is pretty much completely negated. I do not like banishment at all. There's no interesting tactics involved with it. (Of course this is certainly partly because of my enormous dislike for the abyss, since if you replace banishment with something like smiting you get the same position-ambivalent behaviour but I think it's acceptable in that case.)

I acknowledged the corner thing earlier in the topic but like I said there's not anything I personally see that solves the problem in an acceptable fashion other than removing paralysis entirely which I do not personally think is an improvement. It is true and unfortunate that paralysis and banishment favor manual exploration over autoexplore, but then so does literally everything dangerous in crawl and I don't feel that paralysis is noticeably different in how autoexplore affects it (compared to e.g. other things with ranged attacks. Autoexplore does affect melee-only things noticeably less, of course).

[Siegurt]

Note however if you are *not* using autoexplore, this situation is completely avoidable, Moving in such a way that you'll reveal undiscovered squares that are adjacent to you is risky, and it's only because autoexplore doesn't know any better that you would get in this *particular* situation..

That's actually why I rarely autoexplore except to pick up ammo or on floors where I've mostly explored and I'm just checking to make sure i didn't miss something. Even without this particular extreme, autoexplore has no way to be "smart" about how to make sure the squares you reveal are going to be at the edge of your LOS.

[Wahaha]

I don't think that an instant cast of paralysis is a problem, monsters should be allowed to perform any action they have available at any time. If one of their actions is too strong then it should be changed instead of adding weird cases. Instant paralysis deaths are so rare I don't think they matter.

The reason why I think paralysis might be a little too strong is that getting paralyzed for 5+ turns in a row by a top-tier enemy (Rupert, ancient lich, sometimes Norris, sometimes vp knights) might be too much, regardless of whether it happens on LOS entry or not. Arguably only ancient liches because the others are easy to avoid and are a more optional fight. Although having a monster like vp knight be 3 times more dangerous solely because of paralysis is not great because it's already a dangerous monster without paralysis.

On the other hand I think paralysis on ogre mages, sphinxes and orbs of eyes is good. So I don't necessarily think paralysis should change but maybe some monsters could get a shorter duration version or get a delayed paralysis.

One thing to note, and this is pretty "meta" but teleport scrolls are pretty close to "extra lives". If a cast of petrify makes the player use a teleport scroll, it might seem like petrify was easy to deal with because the player stayed at high hp the whole time and escaped safely. But teleport scrolls are so valuable. If a single cast of a spell forces the player to use one, and at high hp no less, it means the spell is really strong. Petrify doesn't always force a tp but it sometimes does. For comparison, think of spells like iron shot and LCS. I don't think that petrify is a weak spell at all, maybe it is with potions of cancellation, I don't play 0.15 and don't know how common they are. This is why I think making paralysis take 1 turn to activate is fine. If the player uses a life-saving consumable to escape delayed paralysis that makes that cast of paralysis arguably stronger than not escaping it would. At least I don't think delayed paralysis is at risk of being too weak. Even if delayed paralysis is not as strong as instant, it could be made more common (either with more monsters that use it or with an mr check buff) with a shorter duration to compensate. Then it wouldn't be as "full hp to almost dead" deadly but it would make a fight harder more often. I don't really care about changing paralysis to delayed though because instant is good on monsters that I named above. Maybe petrify could be changed to fit this role because currently petrify is used pretty badly aside from catoblepas clouds.

[crate]

"Delayed" paralysis is an ok effect but it's not paralysis.

[duvessa]

One thing to note, and this is pretty "meta" but teleport scrolls are pretty close to "extra lives". If a cast of petrify makes the player use a teleport scroll, it might seem like petrify was easy to deal with because the player stayed at high hp the whole time and escaped safely. But teleport scrolls are so valuable.

I...

But...

...ok

[Wahaha]

Maybe if you read the following phrase. Having to use a life-saving consumable to deal with a spell means the spell is not exactly easy to deal with. Well I'm not sure what you disagree with or don't understand.

[Galefury]

Unless the level you are on is mostly cleared, teleport is not very likely to put you into a great situation. So "pretty close to extra lives" is not a very good description of teleport.

Also, reading teleport on getting petrified is often not good. If you read teleport on getting petrified, you have no influence on the situation after the teleport. If you do something else, for example read fear, summon something, use wands or other consumables to kill some monsters, or many other things, you can improve your situation and very likely make it a lot less dangerous. So the situation has to be pretty bad to make giving up all control over it worthwhile.

[Wahaha]

Unless the level you are on is mostly cleared, teleport is not very likely to put you into a great situation.

In my experience the likelihood of teleport putting you in an acceptable situation makes them pretty close to "extra lives". 0.75 per scroll? You can disagree since this is only based on my experience. This description becomes more true with better play, because the player sees death coming earlier, teleports earlier, and is therefore more likely to land in an acceptable situation. The player is also less likely to need to teleport when the level is not explored because they are near the stairs with a low chance of something being between them and the stairs.

If you do something else, for example read fear, summon something, use wands or other consumables to kill some monsters, or many other things, you can improve your situation and very likely make it a lot less dangerous.

It's hard to evaluate the usefulness of these actions because there are like 3 monsters in the game that use petrify. I'd say that against a dangerous enemy most of the actions you listed and didn't list but implied are not useful. Fear is good vs low mr. With summoning you need a guarantee that the summon moves between you and the enemy. So it has to be something that summons a lot of stuff like scroll of summoning or sack of spiders. Or it has to be a summon in a narrow area. Wands and damaging consumables, or attempting to kill enemies in other ways is close to useless because there's no guarantee of killing anything in 1 hit. It might help a little to kill some weaklings but presumably those aren't the threat anyway. I'd say teleporting is better than all common actions except blinking away (except vs Jorgrun), a blocking summon (except vs Jorgrun) and possibly invis. Of course not all cases of petrify are dangerous enough to warrant a serious response but they regularly are. So in my opinion teleport is one of the best answers if the situation is dangerous enough, and almost always the best one vs Jorgrun. One more minor point is that when petrify activates, there are still a few turns left (1-3?) before the teleport triggers, so the petrify at the landing spot lasts a few turns less.

[crate]

The strength of petrify as an effect is pretty much irrelevant to this topic. You're repeating the mistakes I made in the previous topic discussing paralysis.

The important thing is to decide whether paralysis--which is implemented for a reason as "you get no actions after it affects you"--is a desirable thing to have. As I have said, I believe personally it is, and there are reasons that other players do not. At least dpeg (among the devteam) thinks that paralysis existing is a good thing.

Petrification is not that kind of effect, so discussing how strong it is is irrelevant here, unless you have already decided that paralysis is not a good effect and are discussing how strong its replacement should be, though I think you'd be better off attempting to convince the devteam to remove paralysis in ##crawl-dev instead of trying to do so here.

[Wahaha]

My last post is offtopic, it was only a reply to Galefury's post, it doesn't argue for anything related to this thread. Sorry for the confusion. Originally I mentioned petrify because it's similar to a delayed paralysis effect that was proposed in this thread and I wanted to argue that delayed paralysis is still a very strong spell (partially because you said petrify is wimpy). I don't think that paralysis should be changed. It's good on some monsters, like I said previously. I do think that paralysis might be too strong on ancient liches and maybe on a few other monsters so I think a delayed paralysis effect, or simply a shorter duration paralysis might suit them better. That doesn't mean changing paralysis for other monsters. It should probably be petrify that is changed a little to fit that role or a new delayed paralysis effect could be added (this will never happen).

I don't care about instant paralysis on LOS entry, and it's not used as an argument for anything, but I care about paralysis possibly being too strong on some monsters and this is kind of a thread about paralysis in general. Coincidentally, giving some monsters like vp knights either shorter or delayed paralysis would remove the chance of them instantly killing the player on LOS entry, but that's not part of my argument.

[Sandman25]

I don't care about instant paralysis on LOS entry, and it's not used as an argument for anything, but I care about paralysis possibly being too strong on some monsters and this is kind of a thread about paralysis in general. Coincidentally, giving some monsters like vp knights either shorter or delayed paralysis would remove the chance of them instantly killing the player on LOS entry, but that's not part of my argument.

Sorry, I don't understand. Do you mean you don't care about instant paralysis on LoS entry, but then you approach the enemy for killing in melee and start caring about the paralysis being too strong? For me getting a paralyzer into LoS means I am abandoning the level if I can't kill it safely and I think it is expected effect of the spell.

[Wahaha]

When a monster like ancient lich or vampire knight is dangerous relative to other monsters at their depth, even WITHOUT paralyze, and then because they have a chance to paralyze you for 5+ turns in a row they become a further 2-3 times (not a real statistic) more dangerous, I think that's kind of bad design. They're already dangerous and the chance of paralyze makes their difficulty too out of line with other monsters. So yes if you want to play carefully they should be avoided. I can understand that on uniques like Rupert and Norris because they're not too hard to avoid, but on normal monsters not so much. At least that doesn't seem to be the intended design. A vp knight already hits harder than a stone giant (they have a weapon on top of base damage) and with a little vampiric effect, can haste, invis, blink close, sinv and undead resists. To be fair they have less hp. It seems like a good strong enemy... but then they also have a chance of paralyzing you for 7 turns in a row. It doesn't happen often, but is this really something they should have a chance of doing at all? A shorter average duration paralyze would be more appropriate, probably by reducing the max duration by 1 or 2 turns. As for delayed paralyze now that I think about it, it's a lot worse design-wise than instant on vp knights, so I changed my mind a little.

[Sandman25]

Late game monsters should be more dangerous, I mean not only higher HD but synergy with other monsters. This is why we have packs of spriggans (air mages, defenders, berserkers), draconians (shifters, annihilators, monks), different dragons. They are much easier when split in smaller groups (ideally 1 monster in each group). Also this is why Vaults 5 is so dangerous, we cannot split monsters there.

[njvack]

autoexplore has no way to be "smart" about how to make sure the squares you reveal are going to be at the edge of your LOS.

I know this is getting a little OT, but this is not true. Just because autoexplore doesn't prioritize safe positioning doesn't mean it can't.

I'm not saying the change would be easy (and it might be very hard to keep it from being irritating) but it's certainly possible.

[hannobal]

Getting paralyzed once is bad but we need bad things in the game
However chain-paralyze is too bad effect, because there is nothing that can be done about it once you're in the loop.
So let's give paralyze effects 3 turns maximum duration (which of course can still kill you), and equal duration of immunity to paralyze afterwards, so player has at least small chance to avoid getting paralyzed immediately again.

This should also affect monsters paralyzed by player of course.

That would still maintain paralyze as a very dangerous effect, but chain-paralyze would not be an insta-kill anymore.

[duvessa]

Paralysis already makes players immune to further paralysis for a few turns. Chain paralysis has been impossible for years.

Monsters don't get the immunity though, possibly they should. (it's also funny that ensorcelled hibernation has the exact opposite situation; monsters get immunity to chain sleep but players don't)

[crate]

rip red wasp instakill melee

[PleasingFungus]

(it's also funny that ensorcelled hibernation has the exact opposite situation; monsters get immunity to chain sleep but players don't)

Actually, they do. (As of slightly over a year ago.)

e: assuming that this code actually works; I haven't tested it myself, and it doesn't really come up very often now that spen are gone (RIP)