Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Snaaty]

Hello Forum,

first off, I'm sorry if this has already been discussed, or I post something wrong, or the idea is just silly. Please be gentle with your criticism. 

In the current game, some creatures have [TRAPAVOID] and others don't. That leads to the situation (in my opinion), that traps (esp. cage traps) are very powerful against many (most) enemies, but fail enitrely against others.

Now, as I don't know that much about the possibilities of dfhack or the vanilla code, I'm just throwing this out there:

Make the [TRAPAVOID] also demand a number from 0 to 100 (something like [TRAPAVOID:60], for example), that stands for the propability, that the particular creature can avoid traps.
Edit: To clarify, I guess this shoul dbe checked every time a creature steps on a trap-field.

Some megabeasts or Titans probably should still get 100% chance to avoid traps.

I believe this could make sieges and defending your fort more interesting, as you can't entirely rely on traps, yet they are not totally useluss against some enemies. Also, doing this in a fashion that could be easily changed in the raws by modders is probably a good idea.

Tell me what you think!

[Sergarr]

Maybe instead of percentage chance, you could give creatures TRAPAVOID strength, which would get compared to the trap's TRAPCAPTURE strength.

[Dyret]

Or just base it on observation like with adventurers, which will probably happen eventually.

[GavJ]

My initial thought is just "long hallway, two bridges, ping pong them back and forth over the traps and even a 99% trapavoid will be completely reliably caught."

However, I guess if you're going to be that lame about it, you could just mod out trapavoid already. So meh.  Couldn't hurt to add it for the roleplaying folks who won't abuse.

[Snaaty]

Maybe instead of percentage chance, you could give creatures TRAPAVOID strength, which would get compared to the trap's TRAPCAPTURE strength.

Hm, that does sound interesting, also allows for "upgrading" your traps, as it is currently done in MDF.

Edit: This would also allow balancing within traps, so Stonefalltraps could have a high strength, weapon medium and cage low. That way you have to make the decision if you want to hurt the enemy with a higher chance or neutralise them totally, though with a lower chance.

Or just base it on observation like with adventurers, which will probably happen eventually.

I didn't know the way it works in adv. mode. Is it just one check of your observation against the trap and then you either trigger it or not?

My initial thought is just "long hallway, two bridges, ping pong them back and forth over the traps and even a 99% trapavoid will be completely reliably caught."

Yeah, I can see, that this could be abused by min-maxing players. (Although you're right about the modding out ov trapavoid, too.)

Maybe to avoid that the numbers could just be tweaked respectively?

[GavJ]

How about just 2 numbers:

[TRAPAVOID:X:Y]
X = % likelihood avoiding the trap
Y = % likelihood jamming the trap, rolled only on successful avoidances

So like [TRAPAVOID:50:50] would mean 50% avoidance, and 50% of THAT (25% overall) they jam the trap as well.
Vanilla behavior right now would be [TRAPAVOID:100:0]

[Sergarr]

How about just 2 numbers:

[TRAPAVOID:X:Y]
X = % likelihood avoiding the trap
Y = % likelihood jamming the trap, rolled only on successful avoidances

So like [TRAPAVOID:50:50] would mean 50% avoidance, and 50% of THAT (25% overall) they jam the trap as well.
Vanilla behavior right now would be [TRAPAVOID:100:0]

Still abusable. And doesn't simulate better traps which are harder to avoid.

[GavJ]

I doubt Toady wants to implement trap upgrades, since his long term plan is to do away with pre-built, 1-tile traps altogether anyway, if I recall correctly.

But if he does, they would still work in conjunction with the above suggestion of 2 percentages, just like the 1 percentage.

Also, you could make it "break the trap" instead of jam if you want, that way you could still guarantee capture via some crazy device, BUT at least it couldn't be automated while you grab a sandwich, since rebuilding buildings requires direct overseer micromanagement.

[Neonivek]

Trapavoid seems like a place holder created mostly because of how overpowered and unavoidable traps are.

I'd rather see it just thrown out as a whole then "making it percent based"

[Larix]

I'm not convinced percentage-based trap avoidal would really change traps for the better - as GavJ said, you can just spam more persistently. And if kobolds were reduced to an under-100% avoidal, they'd lose their only point of interest.

I think traps as they currently exist should remain as a typical early-game defence: easy to set up, reasonably efficient with little management, but with limitations. I think observation-based trap avoiding (and maybe area avoidance), a dodge chance for cage traps and lower efficiency for weapon traps (less weapon speed, a substantial "cooldown" time, better dodge/block chances) would do more to balance traps. Of course, you could still overcome many of these limitations by massive spam, but if features like "disarming" or deliberately jamming traps were added, goblin grinders and even ping-pong trap halls would run into real limits.

[dudlol]

It can always fall to the reasoning that no matter how good you are at something you will always occasionally fail.

We're all pretty good at eating. We probably chew a mouthful of food dozens to hundreds of times a day. Once every few months we all still bite our tongue hard enough to want to check for blood.

Also,  this should (if implemented) be combined with a failure rate on traps themselves. If you use a soft stone to make low quality mechanisms and then have an unskilled mechanic throw a cheap wooden cage that some peasant made into a trap it should almost never trigger, and then often miss, break, fire late, jam, or otherwise fail.

[Dirst]

I like the idea of a trap strength, done in a way that works as a skill vs. skill contest.  Traps don't actually have skills, per se, but quality levels factor into how well it does what it does.  This can be extended to as we move toward roll-your-own traps in future versions, but for now I'll describe it in terms of pre-baked traps.

The trap should have some chance to surprise the target, and a separate chance to affect the target.

Step 1: Surprise
If a creature notices that a trap is present, it can path around it or take its chances trying to get through.  Good Observation skills would alert the creature to the trap's presence, and perhaps Mechanic skill would give a rough idea of how dangerous a detected trap is.  Semi-intelligent and non-intelligent creatures might be good at detecting traps but not so good at deciding what to do with that information.

"Sniff... sniff... traps all the way across the hallway.  Damned dwarves and their machines!  Now, can I dart past them or should I just go home?  I'm a lizard with an IQ of 8, decisions aren't really my strong suit."

The stealth "skill" of the trap should be based on the quality of the trap components including embedded mechanisms.  Maybe the skill of the mechanic who installed the trap could play a role as well.

Step 2: Affect
Once a creature enters the trap's tile, the trap springs into action and attempts to affect the target.  If the creature was truly surprised, it would be at a disadvantage to dodge/parry the trap.  The attack "skill" again depends on all of the trap components, perhaps including the installation as a component.

For pre-baked traps, things can be play-balanced on a spectrum from "easily hidden" to "reliably effective when triggered."  Maybe the player can explicitly spend quality points on the stealthiness and effectiveness of the trap.

Later, roll-your-own traps will involve many more components, but the same principles can apply.  The only issue is deciding early on which path to take:

1. The game recognizes player-built traps and resolves things in a simple skill-vs-skill manner.
2. Creatures have a chance to observe any machinery at all (big bonus if it actively powered) and depend on AI to attempt to avoid being in the wrong place at the wrong time.  This puts a premium on trap designs that spring quickly, but it should have the side-effect of keeping dwarves from getting mowed down by minecarts.

[GavJ]

Also your traps should all be known if anybody detects them AND GETS HOME alive, by anybody else visitng from that civ later. If you kill them all, then no.

[Dyret]

I didn't know the way it works in adv. mode. Is it just one check of your observation against the trap and then you either trigger it or not?

It rolls against your observation to spot the trap when it is within sight range, I forget if it actually lets you walk over it safely after though.

[Sutremaine]

Yeah, I can see, that this could be abused by min-maxing players.

A nomination for the 'DF tagline' topic right there.

Any starting percentage for trap avoidance could be for the first time the trap is triggered. Subsequent trips over the trap would each provide a bonus to avoidance, with the bonus being based on the creature's relevant mental stats. For good measure, those mental stats could be trained by the experience.