Traps, again

[Antoine]

So then, how does a warrior avoid falling down a trap door as he walks along a corridor? Or does he just have to accept that this happens sometimes?

A.

[wobbly]

False floors in Sil can be, just risky. It's pits that can't. Mostly I like this new idea though would have to play to see for sure. Will perception be a one time only check like HOM or every turn. If it's the 2nd rest becomes search with all the problems that entails.

[Pete Mack]

Some traps cannot be disarmed. e.g. rookery, web, trap door in Sil.

[Nick]

OK, I've had a good long think about this, and considered all the comments that have been made. A couple of philosophical points first:

• The current, 4.0.5, gameplay boils down (with some exceptions in the early game) to "everyone has magical trap detection, and everyone uses it all the time, unless they forget to". As a consequence of this, the only way a player can hit an invisible trap is if they forget to detect, which amounts to a constant low-level attention tax on the player. I don't think this is interesting gameplay. Mitigating it by allowing some sort of "detection on" setting where detection happens automatically when needed at a constant mana cost is maybe possible, but seems to me to be treating the symptoms.
• Manual searching for traps (or to a lesser extent secret doors) also seems fundamentally flawed to me. Either a search is guaranteed to find a hidden trap, in which case "always search" becomes the optimal play; or it isn't, which seems even more frustrating.
• Despite these problems with the current methods of finding traps, there is a significant fraction of players who want some form of invisible traps retained.

The problem with invisible traps can be simplistically summed up as follows: if you can always detect them, they're not really invisible; and if you can't always detect them, you can be killed by a random event. Which seems to leave us with nowhere to go, but I'm hoping that in fact there's some wiggle room in the middle. And I wouldn't have written all this if I wasn't just softening everyone up for my next suggestion

So here it is:

1. Traps continue to be placed as in the nightlies (at places the player has to walk through, not the middle of rooms), but there should be less of them
2. Some traps are visible, some are invisible; the more dangerous, the more likely to be visible
3. Magical detection of traps (and secret doors ) is available to mages early, and priests late, and no-one else (in particular, no trap detection scrolls, rods, etc)
4. Perception comes back as a skill, with rogues being excellent, rangers good, and others less good; maybe (along the lines of what Grotug suggested) have it as an actual trainable skill which improves with traps noticed
5. Vaults no longer have designated places where traps appear, rather vaults have an individual trap density, and an appropriate number of traps are placed at random
6. Disarming remains as two skills, physical and magical
7. Most traps have a chance to be gone after use (with, as now, that being a certainty for some, like summoning traps); this chance is greater if the trap is triggered by an attempted disarm (imagine a warrior "attempting to disarm" by hitting the trap with their sword)

This is a set of principles; obviously there is a lot of detail to sort out (which traps are always visible, etc).

What is wrong with this? What have I missed?

[Grotug]

If you run, rather than walk, you don't have to worry about stepping on traps.

To run, you hold down shift and press the direction you want to run in. @ then automatically runs until something disturbs him, such as a monster waking up, a choice needing to be made on which way to go or @ discovering a trap in his path. The point Pete Mack was trying to make is if you run everywhere, you will never unintentionally stumble upon a trap in the new proposed system of traps.

I have mixed feelings on the two different systems and can see the pros and cons of both. Personally, I am still not bothered by the supposed tedium of detecting traps. The tedium of juggling inventory is so so so so so so so far and away more of a bother to me that detecting stuff is the least of my concerns while playing Angband. If you are playing Angband, you detect for monsters and you detect for traps. It's what you do. No big deal.

I guess I like that traps don't play a larger role in Angband. I think their niche-place in the game is just right. Monsters should be the much greater danger. I think the way traps currently work as a nuisance and a potential compounder of danger to the monsters around is just fine. The only thing I might suggest is that pit traps do a sliding scale amount of damage corresponding to dungeon level instead of still doing 10 damage on DL60. Something like: DL1-DL10 do whatever the current pit damage is. DL11 through DL20 they should do about 30 damage, DL21-30 ~60 damage, DL31-DL40, 90 damage, 41-50 120 damage, DL51-DL60 150 damage, DL61-DL70 180 damage; DL71 DL80 210 damage, DL 81-DL90 250 damage. DL91-DL100 300 damage.

The overall number of traps currently lying around the dungeons is quite high. They could be made less common. Being less common and difficult to detect would make them more surprising and interesting when experienced. Along this line, maybe trap detection's cast-ability should be tailored to the class you are playing ala Carnevian's post.

To compensate for traps being harder to detect, @ learns from experience. The more @ experiences a certain trap, the better @ gets at recognizing the trap and disarming it, and thus discovering it before he blunders on it again. So, for example, a CL10 warrior might have a 5% chance of detecting a summoning trap that he has never seen before while going down a hallway. Once he's triggered the trap, the next time he encounters a summoning trap he has a 40% chance of noticing it before stumbling on it, and a 70% chance of noticing it after the second time he stumbles upon it, and each time he disarms it successfully his chance to disarm it again in the future goes up as well. For a CL10 Rogue, it might look something like: 60% chance to stumble on a summoning trap he's never seen before, 95% after he's triggered it once, 100% after he's triggered it twice. So by the end of the game traps are acting more like Derakon's idea of highly-visible dungeon obstacles. To incorporate AH's idea, familiar traps could also start showing up one square before you reach them, and super familiar traps could start showing up at end of LoS.

Alternately, @ only needs to see a given trap to increase his ability to detect it in the future. So @ doesn't need to stumble on a trap to improve his ability to see it next time; he merely needs to see it, either by detecting it successfully using his perception skill or by using magic detection. But @ needs to be adjacent to the trap to improve his ability to detect it in the future; seeing it on the map 20 squares from where he is will not help him identify it next time.

To keep things interesting with this new way of handling traps, you would need to have new, more dangerous traps continually appearing the deeper into the dungeon you descend. So while most traps are fairly noticeable by DL85, there would still be a few traps @ isn't familiar with because @ only starts to encounter them at ~DL80 (I'm thinking possibly 5 different trap danger categories: one set of traps that typically show up through DL1-20, another set through 21-40, another set through 41-60, another set... 61-80 and yet another set through DL81-100). And the lower the native depth of a trap is, the better @'s perception/searching would need to be to detect and disarm the trap.

Traps, like monsters, could have a native depth, and just like monsters, they could be generated out of depth.

Since certain classes are more susceptible to the effects of traps (especially early on [ahem, *Mages*, ahem]) and since I am trying to move away from detection spamming, what if there was one stat for both casters? Wisdom could be shared by both Priests and Mages. Intelligence would then be removed and replaced by Perception (ala PowerWyrm's post) which would be dedicated to searching and disarming. Therefore, if you are worried about traps, you can boost your PERception at the start of the game, and items with PER boosts on them would be covetable. Having both casters share WIS simplifies Randart generation, too. All those Holy Avengers can be used by both classes as well.

Why is it setup so that the different casters don't share the same stat for casting? Is it so there is more variedness to the two caster types in that Mages get better searching, but Priests get better saving throw? I have been operating under the assumption that only WIS affects saving throw, regardless if you are playing a Priest or Mage (Angband sure is complicated).

Wouldn't it be better having both casters share the same stat so that there are less unusable items? Either Amulets of Wisdom or Rings of INT are currently useless to about 60% of all players. With my proposed changes all items are usable by everyone (except for Warriors, WIS remains mostly useless) and there are more choices to be made about which items you are going to wear/carry and what you leave behind. This seems like a much more interesting solution than removing traps outright, or removing their essential 'trapness'.

This redoing of the stats also makes what each stat does less convoluted: I currently find trying to figure out the secondary purposes of INT and WIS to be very confusing--my understanding is that searching and disarming are bundled into INT and saving throw is bundled into WIS, putting Priests and Paladins who are trying for WIS boosts at a disadvantage for searching and disarming, while Mages, Rangers and Rogues are at a disadvantage for saving throw, since they are trying to boost their Mana pool with +INT.

This retweaking of stats, detection and traps opens the door to the creation of new items/artifacts, too. For example, it would be handy to find an {ego} Cloak of Perception <+3> PER. Or The Cloak of Aragorn <+5> PER, <+3> DEX. Activation: detects all traps within 20 square radius. Takes 100 turns to recharge. or Detects all traps you have experienced before. Takes 20 turns to recharge.
Now presented with the choice between two {ego} cloaks, does a mage wear a cloak of the Magi or a cloak of Perception?

Maybe there could be a new buff type: Awareness. So a Cloak of Awareness would be significantly better than a Cloak of Perception. A cloak of Perception merely boosts your PER stat, whereas a Cloak of Awareness allows you to see all traps @ has come across before within a 20 square radius (same way ESP sees all thinking monsters). You find the Boots of Elessar: <+5> Speed <+5> PER <+5> DEX. Grants the ability to see all traps @ has come across before up to 20 squares away. Now a helm of Awareness <+5> PER competes with a cap of Telepathy.

Amulets of Searching get replaced by amulets of Perception. Amulets of Awareness show up a little before Amulets of ESP do and get a Perception bonuses (the way Amulets of ESP currently get a Searching bonus). Amulets of ESP lose their searching bonus (not to be replaced by a Perception bonus). Amulets of Wisdom remain. An Amulet of Trickery <+4> <+7> <+2> would look like:

+4 Dexterity
+2 Stealth
+7 Perception
+2 Speed

§

Let me just clarify the problem again: Repeatedly detecting for traps is not interesting, but repeatedly detecting for monsters is, because monsters move around and can harm you from a distance, whereas traps don't move and cannot do any harm to you unless you step on them? I guess what I'm getting at is: why is there a problem with the tedium of detecting for traps, but not with the tedium of detecting for monsters? Or maybe the argument is simply that traps in general are just not that interesting? I like traps. They seem like an integral part of Angband. I guess I like my proposed changes since they incorporate many of the ideas presented in this thread. A happy middle-ground? Traps get to stay, detection spamming is addressed and the concept of highly-visible dungeon obstacles is introduced.

*Of course my perspective on traps is biased since most of my experience in the dungeon is as a warrior.

[Egavactip]

OK, here is a proposal which possibly retains the surprise element of traps while not enforcing detection:

• All traps are detected only on stepping onto them, at which point they become 'active'
• Once the trap is stepped on, the player can choose either to attempt to disarm, or to step off the trap
• A failed attempt to disarm while on the trap sets it off
• Stepping off an active trap is completely safe, but the trap remains active (and visible), so the player can't now retreat over it without disarming or accepting the effects
• There is no trap detection, and no noticing before activating the trap - visible means active

It is also possible to expand this a bit:

• Some traps (which I'll call radius-1 traps) have an effect when the player is adjacent to the trap, and are activated by stepping adjacent
• To avoid an activated radius-1 trap, you have to step out of the radius
• A radius-1 trap can be disarmed on noticing, or from distance 2 if approaching when already activated
• Successful disarm of a radius-1 trap moves the player onto the trap square (or toward it if disarming from distance 2)

Please find the flaws in this proposal, or give opinions, positive or negative.

How is it that every single trap in the dungeon is so cunningly constructed that no player, no matter how eagle-eyed or experienced, or no matter what magical equipment he might have, could fail to detect each and every one of them before stepping on to them?

So I don't like your proposal right from the first sentence.

Being able to step off an active trap without harm pretty much defeats the purposes of many traps. That doesn't make any sense.

This all sounds awful and a damn site worse than what we have now.

[takkaria]

B) without it, ALL traps are detectd AUTOMATICALLY (the rod of detect is BUILT INTO THE TRAP) but the player now has to tediously walk one keystroke at the time to have the same practical effect of detecting via tool.

Are you aware of the run command? It's '.' (or ',' in the rogue-like keyset.) It moves you in the direction you specify but stops if monsters or traps appear. It means that you don't have to hold down keys to move longer distances and saves you a lot of RSI.

[Sky]

i'm sorry guys but you are wrong, in capital letters. there is a massive logic flaw here which i am gonna point out then i'm out, i'm gonne be ok with whatever trap system you chose to go with.

A) with a tool of detect traps, i actively detect trap. this is influenced by various factors.

B) without it, ALL traps are detectd AUTOMATICALLY (the rod of detect is BUILT INTO THE TRAP) but the player now has to tediously walk one keystroke at the time to have the same practical effect of detecting via tool.

thats it.

[Philip]

As long as detection is a viable way of protecting yourself from traps, all characters past the early game will do it. Which makes it tedious (basically just a detection range marker). The only people who use rods are warriors anyway, and they hardly care about weight at all. Making detecting traps more annoying doesn't remove the reason people detect traps, which is that stepping on a trap can kill you or force you off the level.

The damage traps are not the annoying ones. Also, damage is only relevant in combat and in the early game, when traps can kill you. Occasionally screwing up a fight for characters would be nice if characters weren't always detecting because of summon traps and trap doors. Increasing trap damage and also changing the game so that detection wouldn't be guaranteed would be a good idea, if we're keeping damage traps.

Making disarming worse does introduce an interesting problem, since everybody without some form of Disarming might as well give up on vaults unless they have a want/rod. It feels irritating, but maybe vault raiding should have these kinds of costs.

Adding traps sounds great. Any ideas?

I really don't think there's any point in keeping both detection and traps in their current form, and I think nuking detection and then modifying the traps around that is a better bet.

[AnonymousHero]

So lets find a middle ground.
[snip]

There you have it, a tweaked version of 4.5 where traps are now relevant, without annoying the player or breaking the game.

I disagree, I don't think this changes the fundamental problem (boring and tedious mechanics) at all.

I'd suggest trying the LoS-modulated-always-visible-traps-as-terrain idea for a release or two and then deciding if we really want the old traps back. I predict almost nobody will want the old traps back.

(... but seeing as I'm not actually coding on Angband, I guess it's up to whomever who's coding it up. I'm just suggesting that Nick shouldn't be afraid to experiment. The traps-as-terrain feature has precedent and seems to work well in other games, so I don't see why it wouldn't work in Angband. EDIT: Enough words have been spilled over this already, I'll stop talking now.)

[Sky]

So lets find a middle ground.

Detect does not mean disarm. Slow down the rods, forcing more carry weight and more resting, or face potential traps.

Increase the damage from irrelevant to significant.

Lower the disarm rate by class+race.

Add some new swag traps.

There you have it, a tweaked version of 4.5 where traps are now relevant, without annoying the player or breaking the game.

[AmyBSOD]

If you want to try a game with more dangerous traps, try ToME 2.x -- obviously some version before we removed traps . There are traps of "drop every item" and "scatter all items from your character about the level"... and most of the damage traps do enough damage to outright kill just about every character.

It's only forces you into more of a corner wrt. detection. Since traps now can now actually be lethal in many more circumstances, all you do is ensure that players will always get reliable detection and will always cast the Detect Traps spell. This is dull and entirely mechanistic play. (Either that or you just get absurdly roll-the-die deaths.)

In short: I don't think what you're proposing will change anything. In fact, it might actually possibly make the problem worse.

It also has traps of "permanently transform the player's equipped artifact weapon/armor into a heavily cursed junk weapon" and "permanently reduce the + to speed on your high-end artifact to half its original value" traps, which can ruin your entire day. But the grand daddy has to be the trap of silent switching when you're carrying the One Ring and plan to destroy it, because then you're forced to play the "evil" path and cannot get to the postgame, just because you stepped on the wrong square once. In short, ToME 2.3.5 forces you to detect traps at all times or you're dead. At least it has an option that makes you stop when trying to step into a known trap... right until a pet stands on it, and then trying to displace that pet will set off the trap with no confirmation anyway and ruin your character.

Yes, I think it's not very good design, because the traps are so evil that you simply cannot survive without detecting them. But at least there are several easy ways to allow you to detect them, so as long as the player doesn't forget to do so, everything's fine.

[AnonymousHero]

the current traps are not good because, except for trapdoors and summons, you can just walk into them because you are immune to practically everything that traps do.

If you want to try a game with more dangerous traps, try ToME 2.x -- obviously some version before we removed traps . There are traps of "drop every item" and "scatter all items from your character about the level"... and most of the damage traps do enough damage to outright kill just about every character.

It's only forces you into more of a corner wrt. detection. Since traps now can now actually be lethal in many more circumstances, all you do is ensure that players will always get reliable detection and will always cast the Detect Traps spell. This is dull and entirely mechanistic play. (Either that or you just get absurdly roll-the-die deaths.)

In short: I don't think what you're proposing will change anything. In fact, it might actually possibly make the problem worse.

[chknflyrice]

I am with Carnivean on this. Rouges should be the only class that has any reliable way of detecting traps early in the game. Except trap door and summoning, all the other traps after DL50 or so shouldn't have any effect on most players because they *should* have a sustain on their primary stats. But this should only apply to actual traps. Such things as acid or fire spots should be apparent to everyone but the lowest perception % or intelligence while dart or pits should take a more perceptive player to notice. I like playing Trolls and I think that its laughable that I can have (almost) a 90% chance to disarm a trap.

I think that the problem is not in the current way that traps are detected but that they are too common and too easy to disarm. Rouges should be able to detect/disarm, warriors/paladins/rangers should be dexterous enough to have a winning chance to avoid the consequences if they should set one off, and mages or priests should be able to destroy or go around (by means of stone to mud).

Make the traps scale to DL, punish players for not playing their class the proper way.

[Pete Mack]

If you run, rather than walk, you don't have to worry about stepping on traps.