PrimeJunta

Nope, I want to get rid of kill XP as a systemic reward. (In fact, I would get rid of all systemic rewards. Rewards should always be explicitly placed.) You can still assign it to specific critters, of course, if you want to incentivize killing them. You would probably want to let the player know that you've done that though, even if it isn't through a formalized quest.

Yes, dear. That's because if the Baron of Derpwood offers the bounty, it becomes an in-game objective. Therefore, because whacking them aligns with designer intent, by definition, it is not degenerate behavior. This really shouldn't be that hard to understand. What you, Raszius et al. are proposing is providing a systemic reward, then identifying situations when it will produce perverse incentives, and then patching it up by removing the possibility to get it, such as by magically disappearing the critters, locking you out of areas, or making them exceptional 0 XP kills. Conversely, what P:E is going to have is no systemic reward, but situational rewards wherever and however the game designers feel they're appropriate. They can even put in kill XP for specific creatures by designating them objectives and manually connecting XP awards to killing them, if they feel so inclined. The former means that you have to anticipate degenerate player behavior, and then actively dis-incetivize it. The latter means that you just have to incentivize non-degenerate player behavior, and let the player do whatever he wants. The latter is fundamentally easier, because it's very, very difficult to anticipate everything a player might want to do.

"I don't mind that it's not going to be D&D, as long as it's exactly like D&D in every detail." Right.

I've no doubt you could make an un-abusable learn by doing mechanic. Haven't played JA so can't comment on that. But it would no longer be XP based, and that IMO would take things too far from the IE archetype.

What's wrong with having the Baron of Derpwood offer a bounty of 10 zorkmids per dead orc, payable upon presentation of an orc snout? I'm kinda interested to see how the P:E team does decide to handle it, since on the face of it The Endless Paths sounds very Diablo-esque (except not randomly generated) and as such a good candidate for an area where kill XP would work without creating perverse incentives. My guess is exploration XP -- they have said that that's in. So for example n * (dungeon level) XP for every room you enter. Yup, you can certainly work your way around the problems. My point is that with a kill XP system, you will have to constantly work around questions like this, whereas with an objective XP system you don't -- you just decide what you want to reward, and reward it, end of story. Systemic XP is problematic precisely because the bigger your game gets, the harder it gets to manage, and the more exceptions or special cases you have to add: locking areas (which always feels artificial to me), disappearing former foes (also artificial unless there's an in-game rationale for it), switching off kill XP for certain foes (e.g. those spawned by a spawn-o-mat, à la that BG2 temple). This is poor design IMO; you're taking something that doesn't really work very well and then adding patches to sort of keep it afloat. I'm curious to hear it.

There are certainly ways of making combat-XP less "degenerative," but since there are fundamentally less problematic alternatives available, what's the point? It's fiddly, involves more scripting, more testing, and more bugs, and every decision like that comes with its own downsides. For one thing, I think a world that behaves consistently is inherently more enjoyable to play in than one where the rules change all the time. I know I felt a bit gypped when I was suddenly getting 0 XP for those respawning shades, for example. Objective-XP is easily understandable and transparent; kill XP that you sometimes get and sometimes don't would likely just feel random and frustrating.

So the reward for doing all objectives and no objectives in a given quest would be exactly the same? See above. If I was the dev, I'd probably playtest this and see how players respond. Note that there's no way to complete a quest without completing any objectives, since completing it is an objective in itself, and I would expect it would have the highest individual XP reward. From a certain POV you will have completed all the objectives in any case, even if you completed some of them before they were active. It depends on how you look at it -- if you're a Communist you'll want to reward the player for work done whether he did it before being hired or after; if you're a capitalist, you'll only want to reward him for work done after you hired him, and consider anything he did before that point a freebie.

@Razsius: if those critters really are a deadly threat, why not make cleaning them up a quest? Suppose you sneak your way through, avoiding combat, then mention it to that freaky Nosferatu guy at the other end. He goes "Oh carp, that sounds nasty. Somebody really ought to do something about them. Would you...?" Then you could go back and splat them, then come back for the quest XP. Or if you already did on your way there, you can go "Well, I already smeared them all over the walls akshually," and he'd go "Good boy, Raszius, here's your XP. DING!" Point being: if something really needs killing, then presumably there's someone in the game world who wants it dead, in which case it makes perfect sense to make it a quest. On the other hand if something is there just as an obstacle to getting spit done, it doesn't matter how you got around it, in which case there's no point in favoring one particular approach (e.g. killing) over another (e.g. sneaking). Take that adventure staple: the guard. The poor sod is just doing his job. You need to get to whatever he's guarding. There's no compelling reason to kill him if you can just, oh, sap him, or sneak past him, or pickpocket his key. Why automatically grant extra XP for killing him?

@Osvir, lots of questions there. Some thoughts about them. First, I'm not a big fan of the "improve by doing" mechanic (improving your stealth by being stealth, improving your combat by fighting etc.). Realistic in a sense, perhaps, but also hugely prone to abuse. The Elder Scrolls series works this way, and it's by far the most exploitable cRPG system I've played; the incentives are totally wacky in it. Specifically, "learn by doing" encourages pointless busywork even more than action-XP, whether from combat, stealth, lockpicking or whatever. When I was playing Morrowind, I was jumping whenever traveling anywhere, 'cuz that's how I improve my athletics skill. Must've looked like an idiot. XP is a completely contrived mechanic; it's a high-level abstraction representing "stuff you've learned," which allows you to decide specifically what you've learned, regardless of how you got there. It really doesn't make much sense from a simulationist/realist point of view, but it makes for good gameplay. So as I said earlier, I'm all for XP-less cRPG's, but Project Eternity isn't it. So I wouldn't want to split up XP into "sneaky XP," "fighty XP," or "musical XP." However, I like your idea of associating affiliation with advantages in specific subsystems (e.g. thieves' guild members gain advantages in sneaky kinds of things, warriors' guild in fighty kinds of things etc), but I don't think XP is the way to go here. Perks or access to special equipment would be more appropriate and less fiddly I think. So the "no stealth experience?" question doesn't arise under this system, since I'm only awarding XP for reaching objectives, regardless of how you reached them. You get your XP however you got past the guards -- by fighting, sneaking, intimidating, or bribing, if that's what's allowed in that particular quest. "Do you get experience... because you saw the windmill?" See above: I addressed this in my reply to Hormalakh. In brief, I think the least problematic compromise would be to award XP for active quest objectives only, but then bump up the bonus when completing a quest to make up for any objective-XP you missed by e.g. killing the bandits before the quest was active.

/me raises hand I don't understand. Why is it bad design not to reward players for clearing those sewers instead of sneaking through them?

Osvir, it was funny. And I did get the references. I'm sorry my attempt at humor in response fell flat. I honestly didn't intend to mock. It's an amusing idea in general (in a good way), but I don't really have much else to say about it, as things like titles depend so much on the cultures that give them, and we don't know a whole lot about those. Once again, I apologize for causing offense. I tried to make a joke and failed. I'll try to do better next time.

I didn't intend to mock. I just thought Barbazon, Wizitch, and Palaest sounded wonderfully like something you'd think up in a certain very particular state of mind. :cheers:

I was thinking about that actually. There are basically three options: (1) Award XP every time you trigger something marked as [Objective] for a quest. -> upside: rewards exploration, which is a Fun activity -> downside: doesn't make sense in context ("Hu? I got 200XP for walking up to this windmill? WTF?") -> downside: if there are mutually exclusive quests in the game, encourages XP farming by tripping Objectives of quests you're not going to accept (2) Award all "retroactive" XP belonging to a quest when completing it, even if you tripped some of the Objectives before it was active -> upside: fair in a Communist kind of way (you're rewarded for your benefit to society) -> upside: makes devs' jobs easier, since player XP totals will vary less (3) Only award XP for "active" objectives -> upside: encourages focused gameplay, more strongly aligned with in-game goals -> upside: fair in a capitalist kind of way (you're only paid for work you're hired to do) -> downside: discourages exploration -> downside: makes devs' job a bit harder since player XP totals will vary more (4) Some compromise between (2) and (3) Consider the farmer's daughter quest again. Suppose we made the daughter herself also a questgiver. So if our party was just dickin' around, beat up the bandits, broke open the windmill door, and killed the orcs, they'd find her at the top of the mill. She could ask them to please escort her to [Farmer]. In this case, they'd only trip the objectives [ExitWindmillWithDaughter] and [ReturnDaughterToFarmer], worth, what, 600 XP. That missing 400 XP that [ReachWindmill] and [EnterWindmill] would've given will have to be dealt with somehow. My inclination would probably be to go with (2), although on a gut level (3) appeals to me more, but that may be because I wear such nice shiny jackboots and want to force everybody to play My Way.

@Gfted1, funny then that you're still not even able to use the term correctly, given that you've put so much energy into railing against it. #strawman

Waiter? I'll have two of whatever Osvir's having. With extra bitters. Thanks.

@SqueakyCat - Yeah, that's a nice example. I'd expect things to work more or less that way. Also, my cat says hi to your cat. Maybe they're long-lost cousins. (Also condolences to Valorian's cat about that accident. I trust it'll grow out.)

@Heresiarch, that's a completely valid way of approaching stealth mechanics as well. In a game that's more IW-esque (i.e., relatively straightforward, combat-heavy dungeon crawler), that's probably the best way to handle it; it is useful without being a dominant strategy. I certainly won't complain if P:E decides to go that way, and executes it well. That would certainly be better than a half-arsed implementation of both combat and stealth. But if there's resource enough to pull it off, I personally would prefer good implementations of both, with at least some cases where combat is not the only, or the best, approach. It would make for more variety in gameplay and better replay value.

No, it isn't. It's just better in most ways than kill-XP, action-XP, or obstacle-XP. XP itself isn't the perfect system. It's a serviceable system. Other character development systems exists. Some of them are arguably better. I would like to play a well-made XP-less cRPG too, but P:E isn't it; it would take it too far from its IE roots.

@Amentep -- Okay, let's take this thought experiment a bit further. At this point, let's scratch the alternative path (ladder + window). I just put that in to demonstrate how I could yank Valorian around like a rat in a maze with obstacle-XP. So we're back at the original setup: [Daughter] in [Windmill] guarded by [Orcs], with [bandits] on the road to the windmill, and a [Locked Door] to overcome. To get past [bandits] you can fight them, sneak past them, intimidate them [if you meet the prerequisites], or bribe them. To get past [Orcs] you can fight them, sneak past them, or pay the ransom. And let's still assume objective-XP only: [ReachWindmill], [EnterWindmill], [ReachDaughter], [ExitWindmillWithDaughter], [ReturnDaughterToFarmer]. Same XP reward no matter how you accomplish it. Let's further assume that [Farmer] will reward you with 400 ZM if you return the daughter alive, and [Orcs] guard a [Chest] which contains 1000 ZM and a Sword of the Munchkin. The upshot is that the player who manages to accomplish all objectives (and get the treasure from [Chest]) with the least resources spent gets the biggest reward (since net reward is [XP + loot] - [resources spent]. Now, let's look at what the various approaches mean. Let's also assume that we have a stealth system somewhat like the one in the beginning of the thread in place. * [bandits]. ** Fighting them means expenditure of Health and other possible resources. The better you are at fighting, the lower the resource cost. Upshot: better fighters are rewarded. ** Intimidating them successfully has zero cost. Upshot: if you have previously spent strategic resources to make yourself more intimidating (e.g. by killing lots of bandits so you've got a scary reputation for it), you now get payoff for that investment. ** Bribing them costs zorkmids, and is not obviously more advantageous than fighting them. ** Sneaking past them carries no cost if you succeed. *** If you fail, you will be in a worse position to fight them than if you had chosen that path to start with, and the ensuing fight will be that much more costly. * [Door] ** Picking the lock costs a [Lockpick]. ** Using Key is pure benefit. That's a reward for being more thorough preparing for the quest -- you talked to the Miller and convinced him to lend you the key. ** Bashing in the door carries no resource cost, but will alert [Orcs], making the fight against them more difficult, and making it impossible to sneak past them. * [Orcs] ** Fighting them carries a resource cost. The better you are at fighting, the lower the cost. ** Paying Ransom costs ZM, and is not obviously the cheapest way to go. ** Sneaking past them successfully: see Bandits, except this will be much more difficult since you're indoors in close quarters. Now. Sneaking. Let's assume that sneaking has not been made too easy: if you've done nothing to improve your stealth, the bandits and the orcs will both spot you, and you will pay in the form of a more difficult fight (and possibly dead hostage -> quest failure). That means that to do it successfully, you'll have to expend resources: * [invisibility Potion], cost 400 ZM. * [invisibility Spell], uses up combat spell slot. * [boosted Sneak Skill], uses up skill points that could have been used for something else. This is known as an opportunity cost. * [be A Rogue], which means you're not, for example, a fighter or a wizard -- that's another opportunity cost. The bottom line? In this system, sneaking is a high risk, high reward strategy. If you're successful, you'll (probably) expend less resources than if you had taken on the bandits+orcs head-on. If you fail, you will take more damage in the ensuing fight. If you want to boost your odds of success, you will have to spend resources. Finally, this is just one quest. Not every quest has to be perfectly balanced between approaches. It's perfectly fine to have a quest where stealth is, indeed, the most efficient option. It's also perfectly fine to have quests where combat or diplomacy are the most efficient options. It's precisely this kind of variety that makes the game interesting. If you know that sneaking (or fighting, or diplomacy) is always the best option, things get boring, and there's no point even trying to look for alternative approaches. (Of course, Valorian would just save, sneak, and reload from the save every time he's spotted. Or, perhaps, fight the bandits and the orcs and reload from the save every time he thinks he took too many hits. Assuming the save system allows that. Which is another degenerate strategy, and a reason I think most savegame systems suck, but that's a whole 'nuther thread.)

It doesn't. That's not what it's for. However, the resource-consuming high risk/high reward stealth system described in the first message of this thread does. I'll be happy to discuss that with you once you indicate that you actually want to engage in discussion rather than just blowing raspberries.

Hey, good. This discussion is going places. I like that. I also like the term "obstacle XP." But I still think the concept isn't as good as "objective XP" (that's a better term than "quest XP" perhaps). Here's why. If you go with "obstacle XP," you're stuck with two options: (1) Always award XP for resolving anything defined as an "obstacle." (2) Only award XP if an "obstacle" is resolved in the context of an "objective." Let's consider what kinds of incentives this gives in the context of our little save-the-farmer's-daughter quest. And let's complicate it just a little, by adding an alternative approach: you can also get to [Daughter] by climbing through [Window] at the top of the windmill, which you can find out about by talking to Miller (with right dialog choices), and reach by using a Ladder you can borrow from Innkeeper. We have [Orcs], [bandits], [Windmill Door], and [Windmill Window] as Obstacles, and [Daughter] [Exit Mill with Daughter] and [Return Daughter to Miller] as Objectives. Suppose we pick option (1), and award Obstacle XP for each Obstacle. Always, but only once. Now, if Valorian was playing the game, what would he do? Something like: (1) Talk to Miller, get Key, find out about window. Talk to Innkeeper, get Ladder. Go through Fields, use Ladder, climb to Window [DING! Window Obstacle XP], get to Daughter [DING! Objective XP]. Then climb back out, go to Road, Intimidate Bandits [DING! Obstacle XP], go to door and open it with miller's key [DING! Obstacle XP]. Enter, sneak past the Orcs [DING! Obstacle XP]. Then sneak back out, go up the ladder and to the [Window] and [Daughter] again, carry [Daughter] out of window, down to the fields, past the bandits, to the Farmer [DING! Objective XP]. As you can see, this sequence had a whole unnecessary loop with the bandits and the orcs -- the most efficient way to rescue [Daughter] would have been to carry her out of the mill through the window as soon as Valorian got to her. Your Obstacle XP will incentivize players to seek out Obstacles whether it makes sense or not. In other words, this is the same problem as with kill-XP, lockpick-XP, or any other action-XP: it creates an incentive that is not aligned with the in-game goal, and therefore causes poor Valorian to run around like a rat in a maze seeking out levers to push, instead of being swept away by the epic tale of the kidnapped farmer's daughter. Poor Valorian, I wouldn't want him to humiliate himself that way. So, bad. Now, what if we picked option (2), and switched off Obstacle XP if it didn't make sense? Say, you would only be able to get Obstacle XP for [Window] or [Door], but not both, and for [GoingThroughFields] or [intimidatingBandits] but not both? First off, this would be fiddly. You'd have to do the extra work to connect the different obstacles, so that resolving one switches off XP for the other. More scripting, more testing, more work, more bugs. This is a downside. And second, you can represent the same thing (or close enough not to matter) simply by defining two more Objectives: [ReachWindmill] and [EnterWindmill]. [ReachWindmill] goes DING! whenever you're within touching distance of the windmill, and it perfectly encapsulates [DealWithBandits|GoThroughFields]. [EnterWindmill] goes DING! once you're inside the windmill, and it perfectly encapsulates [Window|Door]. And it's simpler. So, for my revised Rescue the Farmer's Daughter quest, I'll have * ReachWindmill: 200 XP * EnterWindmill: 200 XP * ReachDaughter: 200 XP * ReturnDaughterToFarmer: 400 XP To complete the quest, you'll have to complete all of the objectives. If you're smart and play well, you can reach them with less resource expenditure -- whether it's because you're good at talking and investigating and find out about the window and the ladder and the fields, or you're good at sneaking and lockpicking and get past the orcs and the bandits, or you're such a badass fighter that you can slaughter the orcs and bandits without taking a scratch on your shiny armor. OTOH if you're not so clever you'll end up bribing the bandits and paying the ransom to the orcs [OK, I didn't have that one in, but why not add it?] or get badly beaten up by them and have to spend lewt to replenish your stores and heal up. Good, smart players are rewarded, poorer, dumber players... not so much. Just the way it should be.

She's most definitely my cat. Here she is with my dog. You can tell who's the boss maybe.

We're getting closer. I do think that I think that a lower level of granularity would be better. The higher you make the granularity, the more exploitable and trickier to manage it gets. If it was me, I'd probably design this quest something like... Objectives: Reach the farmer's daughter 250 XP Exit the windmill with the farmer's daughter alive 250 XP Return the farmer's daughter alive to the farmer 500 XP Obstacles: Windmill door (locked) * Can be picked (requires lockpicking skill, expends lockpick) * Can be bashed (alerts the orcs inside) * Can borrow the key from the miller, who is currently at the village inn with the other refugees from the orc raiders Bandits on road * Can be fought * Can be avoided by crossing through the fields rather than the road * Can be intimidated (if your reputation as all-around dangerous guy is high enough) * Can be bribed (100 ZM) Orcs in windmill * Can be fought ** but once you're down to the last one, he'll grab the daughter, hold a knife to her throat and you'll have to negotiate with him to stop him from killing her * Can be bribed (1000 ZM) * Can be avoided by sneaking, if you're good enough at sneaking, do it in the daytime when they're mostly inactive (orcs being nocturnal), and take care to avoid the patrols ** if you're spotted or they'll jump you, and try to kill the farmer's daughter first if she's with you. This way, fighting the orcs and the bandits and picking the lock would be the most obvious way to go, and also the least risky assuming you're tough enough to beat them. It would cost resources you expend in combat. The best way to deal with the bandits is to avoid the road and cross through the fields; to do this, you need to do some scouting ahead to find that alternative route. You can bribe the orcs if you're really rich, but you probably have a better use for the 1000ZM; successfully sneaking past them is least costly, but you risk getting jumped and failing the quest (high risk, high reward).

Here's my cat!

I'll keep that in mind and write out each iteration separately instead of putting it in a for loop then, the next time I'm working for you. Many more lines of code...