Follow along with the video below to see how to install our site as a web app on your home screen.
Note: This feature may not be available in some browsers.
Welcome to rpgcodex.net, a site dedicated to discussing computer based role-playing games in a free and open fashion. We're less strict than other forums, but please refer to the rules.
"This message is awaiting moderator approval": All new users must pass through our moderation queue before they will be able to post normally. Until your account has "passed" your posts will only be visible to yourself (and moderators) until they are approved. Give us a week to get around to approving / deleting / ignoring your mundane opinion on crap before hassling us about it. Once you have passed the moderation period (think of it as a test), you will be able to post normally, just like all the other retards.
Yeah I was thinking about DX:HR earlier -- I didn't play it all that much and was a while ago but I remember this system being in the game. Not sure if it is any good but at least they tried something different.
Just a little something I noticed in Arcanum (that I am currently playing), related to conversations. In the elven city Quintarra, there are a couple of quest-solving conversations that I found very satisfying.
For example, the discussion with William Bench at Falcon's Ache. You have been sent there in order to force him to leave. If you have the skills, you can use various lies to persuade him. Only one of the lies works, and only through a very specific conversation path (various paths exist within this specific lie). On my first playthrough, I reloaded this conversation 5-6 times and still I didn't get this to work, so I concluded that I probably didn't have the skills. I then found an alternate way to solve the quest.
Later, I read in a walkthrough that the peaceful resolution was there for me to take. I had actually reached the correct node but through the wrong path. So while my PC's skills were up to par, my own were not. Very impressive.
To sum up: In addition to social skill checks, make the player work hard through the conversation tree in order to manipulate/manage other NPCs. Challenge him to reach the right node through the right path, while he does not even know what is the best he can hope for. This works well in conversations with long trees and short nodes.
One thing that is tricky is balancing player skill vs character skill. In combat they both are important, because the player gives orders and the characters must execute. Poor direction, weak characters, and/or bad luck = you lose.
It might be even harder to get right in non-combat though. Really thinking about it, the basis for this thread is that the balance is currently shifted too far in the direction of character skill. If the character has the right skills then all the player has to do is click a button to win. In other words the game is essentially playing itself, kind of like an auto-resolve system.
But if you give the player too much agency then the game can become trivial -- people get good at mini-games, learn meta-knowledge about the plot and characters, etc. Without skill checks or randomness then it's just me, the Player, interacting with the gameworld -- my PC has been totally removed from the equation.
That is one way to approach it I guess but it doesn't feel all that satisfying. I think you have to involve the PC in there somehow, even if it's something like -- your character picks up a history book and reads, roll Perception to see if he notices the important information. If so, a beneficial dialog option unlocks in conversation.
But how does a perception check make the PC involved? If the player reads a book in game, so does their character; they still have to find that book first. And the trick is making that non-trivial. Sure, on subsequent playthroughs (or by simply looking it up on internet) the player will find it easily. But that is why you build up mutually exclusive plot lines: perhaps if you become friendly with one NPC, another will become your enemy, perhaps in order to advance a different quest you need to burn down that library before you can access it.
I think designing the game around "how to stop the player from metagaming" isn't a good strategy. If someone wants to cheat, they will, even if they have to hack the game. And personally I don't think there is anything wrong at meta-gaming, or cheating, or hacking a character in single player games. Multiplayer, yes, but that's the difference between taking a sneak peek at cards when playing poker and when playing solitaire. And it is not like a perception check will stop any player from meta-gaming. If they know there is something in that book, they will reload until they get a good roll, same as with dialogue checks. But if finding that book is not easy, then I'd rather reward the player who doesn't cheat, and give them the reward for figuring it out, instead of having a quest outcome reduced to getting a good roll.
If you want to make it so that not every character can read a book, that is doable without skill checks. Make it so that a character has to be proficient in a given language (and perhaps make it even race-dependent, in case of fantasy or sci-fi settings). E.g. SitS had the language skill and, as far as I know, it didn't use random rolls (it also had a nice dual use; you could communicate with other races using it, and you could do word magic with it). If you need to obtain a certain item to bribe an NPC, make it so it requires thievery, and don't put any "convenient keys" in the game.
Random rolls work in combat, because there is such a multitude of factors involved (and those include meta-gaming too; after you've played through a battle once, you know the enemies, their number, abilities and positioning), that getting even several bad rolls in a row, while memorable, doesn't necessarily mean you'll loose the fight.
Exploration, mapping. Less spoon-fed details and instead requiring the player to take notes. These are non-combat gameplay elements that need to be reintroduced and re-emphasized.
When I asked myself, what would be the gameplay built around communication, my answer was Bard: The Game.
No, not some kind of rhythm game.
It goes like this.
There is a long list of nouns - Love, War, Sea, Taxes, Christianity, Feminism, Josh Sawyer, etc. And a short list of verbs for them - talk about, approve/disapprove, joke, show knowledge about etc. Verb + noun is a phrase.
Phrases can be sequenced into a "tale" or used one by one in a conversation.
There is an auditory of one or more people who are characterised by their relation to themes. Basically, each person is a tale of his life and a tale of his current mood. When they hear some phrase, it is compared with their me-tales and a reaction is produced.
It can change their mood (me-now-tale), or make them do something, like throwing money or knives at you.
Gameplay is knowing your auditory (from context, by observing what they are talking about etc) and firing phrases or tales that are likely to give desired reaction. You may just want to get generous tips for telling people what they like, or something more complex, like change their opinion about something.
Ideally, I suppose, persuading (or intimidating, or tricking) someone would be more than just a single Speech/Charisma/Whatever check. Like, maybe after passing an initial check, then the player has to choose from a list of responses while thinking about which of them would be most effective in manipulating who they're talking to. Maybe that'd be tedious, I don't know, but it sounds like a good way to involve the player more rather than just going through check after check.
Of course, in a dream game scenario, it'd never really be clear in the player's mind whether they should use combat or find an alternative in any given situation, and killing people or letting them live would have unexpected positive or negative consequences later on. Non-combat solutions would also come down to being way more than just a skill check in place of combat - like, completely different branches through a quest for non-combat solutions.
Other non-combat systems (stealth, magic, etc) being robust helps to encourage the player to explore alternatives to direct combat, too. People find really cool non-combat ways through games like Morrowind using spells and enchantments in ways that the developers probably didn't anticipate.
Non-combat also needn't equal non-violence, so there could be creative ways to kill people that incorporate the player character's skillset without entering combat mode.
Had an idea that persusion, or charisma, would sort of act like Lore in AoD, but unlock various npcs who would be more willing to talk to you and give you more information that might open up new locations (underground bars, high society balls etc) and/or new contacts that a non-talker would never even see or hear about.
So a talker trying to convince other more important npcs would then have access to more sensitive information and more allies to refer to - or call upon - to help resolve situations.
Another feature thats been given spanish archer from RPGs now is reconaissance, usually wi Thief character, who you could use to case joints, break and enter, gi company maps, lay out guard patrols and generally be a smart fucking player. He could create weaknesses an all, bribe an off duty guard, poison wi a laxative the guardhouse well, stick a sleeping draught in a freshly delivered barrel o wine. Smart Thief character were fucking deadly, and dint even hav to engage in combat, ideal situation were that nobody knew he'd ever been there. Wi that and quick levelling mechanic o AD&D Thief you had a monster o a character on your hands if he was played smart, thing is a lot o players approached him like any other class and forgot about all those special skills.
Once had a character called Fingas, ad is little fingers cropped for petty thievery as a nipper hence name, anyway he portrayed imself as a locksmith to party and always pretended to be as such, went out of his way to convince em. Course players knew e were a tea leaf, but their characters didn't cotton on for longest time. Anyway by night Fingas'd put on a mask, cloak and black painted elven mail and assume his alter ego o the Masked Avenger, regular little superhero stealing from rich and givin to is fence to sell on. He had a little amoral superhero act going on, and fuck me did e play it smart, used all his skills to never get caught or put himsen in danger that he couldn't handle, and he made a fucking fortune.
Had to play solo sessions wi im just to keep up wi his nocturnal activities, and all characters in group reckoned e were a lazy wino bastard who slept late and caroused all night. This obliged him to do so ocassionally an all. But he were invaluable when it came to contacts, info, lay o land and secrets suddenly being discovered or obstacles disappearing.
Anyway party were waging a battle wi this decrepit wizard at time, really old cunning bastard who ruled a ruined part o city they were in, they'd been basically warring wi each other and finally party had gotten upper hand. Anyway on way to put an end to this wizard Fingas disappears, party slog their way through ruins, smash their way up the levels of the wizards tower and finally get to his chamber. Fingas is allready there, climbed in through window thanks to some anti-magic dust he sprinkled over wardings there and further on, crept up on sleeping mage and killed im in his sleep, while this wizards summoned guardian creatures stood outside door utterly unaware. Thief were taken stock o all goodies and preparing to place them in his different bags of holding.
Thing is that wizard woulda kicked Fingas arse, but e needed sleep knowing what was coming, so that's when Thief struck, using is skills in stealth, climbing and common sense to catch the fucker unawares. Got all experience from mage, shared only stuff e dint want wi group and though his secret were blown he had saved party from a big risky battle, and they still ad all monster and guard xp, as well as loot from rest o tower.
Wish they'd bring solo runs for Thief back in to games, mind you he's now just another Fighter, not half as interesting or unique.
But how does a perception check make the PC involved? If the player reads a book in game, so does their character; they still have to find that book first. And the trick is making that non-trivial. Sure, on subsequent playthroughs (or by simply looking it up on internet) the player will find it easily. But that is why you build up mutually exclusive plot lines: perhaps if you become friendly with one NPC, another will become your enemy, perhaps in order to advance a different quest you need to burn down that library before you can access it.
I think designing the game around "how to stop the player from metagaming" isn't a good strategy. If someone wants to cheat, they will, even if they have to hack the game. And personally I don't think there is anything wrong at meta-gaming, or cheating, or hacking a character in single player games. Multiplayer, yes, but that's the difference between taking a sneak peek at cards when playing poker and when playing solitaire. And it is not like a perception check will stop any player from meta-gaming. If they know there is something in that book, they will reload until they get a good roll, same as with dialogue checks. But if finding that book is not easy, then I'd rather reward the player who doesn't cheat, and give them the reward for figuring it out, instead of having a quest outcome reduced to getting a good roll.
If you want to make it so that not every character can read a book, that is doable without skill checks. Make it so that a character has to be proficient in a given language (and perhaps make it even race-dependent, in case of fantasy or sci-fi settings). E.g. SitS had the language skill and, as far as I know, it didn't use random rolls (it also had a nice dual use; you could communicate with other races using it, and you could do word magic with it). If you need to obtain a certain item to bribe an NPC, make it so it requires thievery, and don't put any "convenient keys" in the game.
Random rolls work in combat, because there is such a multitude of factors involved (and those include meta-gaming too; after you've played through a battle once, you know the enemies, their number, abilities and positioning), that getting even several bad rolls in a row, while memorable, doesn't necessarily mean you'll loose the fight.
Requiring the PC to pass a check makes it so the Player can't just win based on their own knowledge. People can memorize conversations and interactions pretty easily -- sure they can memorize combat encounters too, but there are many more variables involved for the player to manage, and then the RNG on top of that.
Here's an example, if you've played through Fallout New Vegas then you know at the end you have to confront Legate Lanius. If your Speech skill is maxed out then you have the ability to talk him out of fighting -- unfortunately it's the same type of boring skill check we've been criticizing, where as long as you have the required skill you can just click the Awesome button and win.
But the skill check is an important piece because it locks non-talky PCs out of that path. You can't smooth talk him if your PC is not a smooth talker, even if you the Player know which dialogue options to take. This is no different than if your PC is not a good fighter and you try to take him head on, you will probably get your shit pushed in. It's not a cheating thing -- it's about consistently enforcing a system where the PC has to actually be good at the thing the Player tells him to do in order to succeed.
So if the non-combat side of the Lanius encounter was more nuanced, your talky PC would have to do more than just raise their skill to 100 to beat him. He is the game's final boss and is described as being one ferocious mofo, so it would extreme cunning to keep him from killing you. So you could design content around that, things the Player has to do -- and PC has to interpret -- in order to put together a master plan. Take the actual encounter as Obsidian implemented it:
So basically the way the PC talks Lanius down is:
1) Remind him that the Legion has struggled to take Hoover Dam in the past. Even if the Legion wins, it will be a costly victory.
2) NCR is weak because they are overextended.
3) If the Legion tries to conquer the West it will become overextended too, causing it to risk losing the territory it already conquered or even implode altogether.
4) Lanius agrees to retreat, bide his time, and return when the Legion is stronger.
There is a ton of gameplay you could design around that to make it more interesting and challenging. Things the Player might have to do -- steal documents, spying, bribery, read books, etc. And things the PC would have to be good at -- interrogation, disguises, understanding military strategy, etc.
And the approach the PC takes in the game is just one of many potential possibilities. You could design a betrayal plot, covert assassination, psychological intimidation, and who knows whatever else. All ways to pacify Lanius and win the game without having to actually fight him.
Requiring the PC to pass a check makes it so the Player can't just win based on their own knowledge. People can memorize conversations and interactions pretty easily -- sure they can memorize combat encounters too, but there are many more variables involved for the player to manage, and then the RNG on top of that.
Here's an example, if you've played through Fallout New Vegas then you know at the end you have to confront Legate Lanius. If your Speech skill is maxed out then you have the ability to talk him out of fighting -- unfortunately it's the same type of boring skill check we've been criticizing, where as long as you have the required skill you can just click the Awesome button and win.
But the skill check is an important piece because it locks non-talky PCs out of that path. You can't smooth talk him if your PC is not a smooth talker, even if you the Player know which dialogue options to take. This is no different than if your PC is not a good fighter and you try to take him head on, you will probably get your shit pushed in. It's not a cheating thing -- it's about consistently enforcing a system where the PC has to actually be good at the thing the Player tells him to do in order to succeed.
So if the non-combat side of the Lanius encounter was more nuanced, your talky PC would have to do more than just raise their skill to 100 to beat him. He is the game's final boss and is described as being one ferocious mofo, so it would extreme cunning to keep him from killing you. So you could design content around that, things the Player has to do -- and PC has to interpret -- in order to put together a master plan. Take the actual encounter as Obsidian implemented it:
So basically the way the PC talks Lanius down is:
1) Remind him that the Legion has struggled to take Hoover Dam in the past. Even if the Legion wins, it will be a costly victory.
2) NCR is weak because they are overextended.
3) If the Legion tries to conquer the West it will become overextended too, causing it to risk losing the territory it already conquered or even implode altogether.
4) Lanius agrees to retreat, bide his time, and return when the Legion is stronger.
There is a ton of gameplay you could design around that to make it more interesting and challenging. Things the Player might have to do -- steal documents, spying, bribery, read books, etc. And things the PC would have to be good at -- interrogation, disguises, understanding military strategy, etc.
And the approach the PC takes in the game is just one of many potential possibilities. You could design a betrayal plot, covert assassination, psychological intimidation, and who knows whatever else. All ways to pacify Lanius and win the game without having to actually fight him.
I haven't played New Vegas, and I don't intend to, but what you have described is the system I dislike. Because at its core, it is still reduced to passing a single skill check: speech. As you wrote: "People can memorize conversations and interactions pretty easily", so they can memorize the sequence that leads to winning the conversation with the NPC. But here is the thing: they can also use meta-gaming knowledge to learn the sequence that is necessary to have those options: "steal documents, spying, bribery, read books, etc.". Nowadays it is a simple as googling it. If you look at it from the point of view of players who do that, then your entire set up is still reduced down to passing that single speech check. The difference is in the first case the player needs to know the correct dialog choices, and in the second they need to know the sequence of events, but in both cases in the end a roll or a skill check decides the outcome.
That is what I mean by not designing a game around players who might use cheats, hacks, or meta-knowledge to win, because they will still win even with the speech check in place. It is as simple as googling how much speech you need and either developing your character based on that knowledge, or cheating and changing their skill points. I'd rather have the game designed around those players who try to find out by themselves, by playing the game. You don't need speech checks at all for this. The challenge is in figuring out what you have to do in game, to convince an NPC. Repetitiveness can be avoided by making quest lines and quest outcomes mutually exclusive.
King of Dragon Pass has great non-combat gameplay, even if it is not truly an rpg.
A lot of those non-combat elements are very rpg-ish and could be used in rpgs or roguelikes (I am working on something like this).
One thing that is tricky is balancing player skill vs character skill. In combat they both are important, because the player gives orders and the characters must execute. Poor direction, weak characters, and/or bad luck = you lose.
It might be even harder to get right in non-combat though. Really thinking about it, the basis for this thread is that the balance is currently shifted too far in the direction of character skill. If the character has the right skills then all the player has to do is click a button to win. In other words the game is essentially playing itself, kind of like an auto-resolve system.
But if you give the player too much agency then the game can become trivial -- people get good at mini-games, learn meta-knowledge about the plot and characters, etc. Without skill checks or randomness then it's just me, the Player, interacting with the gameworld -- my PC has been totally removed from the equation.
That is one way to approach it I guess but it doesn't feel all that satisfying. I think you have to involve the PC in there somehow, even if it's something like -- your character picks up a history book and reads, roll Perception to see if he notices the important information. If so, a beneficial dialog option unlocks in conversation.
That is why intelligence as a concept in rpgs always kind of bugged me. If I roleplay as a dumb character—let me make dumb decisions, and reinforce that through core game mechanics. Such a leap in game design this could be...
An easy way to do that would be to copy JA2 1.13. Number of rounds in a weapon was hidden from the player if the character was stupid or unskilled with weapons, you could just expand on that. Hide all kinds of different information from dumb characters, maybe they don't even know how stupid they are.
The only way I can think to make non-combat different is to introduce more randomness and make skill checks more than just pass/fail. Make a long list of different outcomes, the one you get depends on skill + some modifiers + randomness. For speech, modifiers could be like what you're wearing, Charisma, knowledge you gained from other Npcs, whatever. And they could be negative too. Show up naked to the King's court and get thrown out no matter how high your speech is.
The only way I can think to make non-combat different is to introduce more randomness and make skill checks more than just pass/fail. Make a long list of different outcomes, the one you get depends on skill + some modifiers + randomness. For speech, modifiers could be like what you're wearing, Charisma, knowledge you gained from other Npcs, whatever. And they could be negative too. Show up naked to the King's court and get thrown out no matter how high your speech is.
Fair point, but I disagree. Randomness is the opposite of the right direction, at least on its own. IMHO the problem lies in a lack of agency. In combat you normally have a fair amount of options, while in dialogue you are forced to stick to a script.
I'd say new mechanics need to be introduced, that would make you more proactive in dialogue. Special skills, unique strategies. Turn talking into a game of chess.
Wouldn't it be awesome if the PC could, say, pretend to be drunk or stupid? Or bluff about what he knows? Or dare the opponent into doing something? Or propose a quid pro quo? All of that in any conversation?
I would love to see something like this, and tbh ten years ago I thought this would be the exact evolution of RPGs
Funny, while typing the thing about "the exact evolution", I was the thinking exactly about the Master dialogue. But imagine this being procedurally rich, instead of being hand-crafted.
It's just, how do you do that without turning it into some kind of mini-game? It could be done, but it would be shifting the core of the gameplay over to dialogue rather than combat... I don't know. Randomization just seemed like a easy way to change it up.
It's just, how do you do that without turning it into some kind of mini-game? It could be done, but it would be shifting the core of the gameplay over to dialogue rather than combat... I don't know. Randomization just seemed like a easy way to change it up.
Speaking of dialog gameplay, there was an obscure RPG/Advnture Arcatera: the Dark Brotherhood. It was a terrible game overall, but one thing it did right was the dialog system. Your relationships with every NPC were measured by three meters: general disposition, anger and patience. Those were affected by your stats and reputation and could be somewhat improve with spells. Most of the time "win" options were only availible if the NPC's disposition was high enough, but sometimes they on the contrary had to be sufficiently angry (but not too angry, as he could attack you instead). There were the typical compliment/offend options, but some dialog topics could affect those meters too - e.g. if you spoke to a superstitious barbarian about magic, he'd immediately get angry and loose all his patience. Patience was the most brilliant part of the system, as it only gave you a limited number of things you could say befor the NPC shut down the conversation, and you could only speak to him again after some time (and the main quest had a time limit). Together with a ton of topics to choose from (having main quest be investigative rather than heroic helped) this made you think very carefully about what you were doing.
Unfortunately, the rest of the game didn't quite live up to its dialog system.
It's not that it can't be done, it's just that gameplay=combat in RPGs. When you shift focus to non-combat it's... not bad, but it also isn't a Tru RPG anymore.
It's not that it can't be done, it's just that gameplay=combat in RPGs. When you shift focus to non-combat it's... not bad, but it also isn't a Tru RPG anymore.
Well, yeah - currently. But Pathologic exists. DXHR exists. LA Noire exists.
Mind you, by no means are these particularly good examples. But they are a sort of proof of concept that demonstrates a new universe of untapped gameplay potential. Seriously, they all look like an early iteration of something new and great rather than being something by themselves, like there is a whole new universe that needs to be constructed and put into a new convention of mechanics and technique.
Speech "mini-games" are not desirable because few RPG gamers actually want to play a speech simulator, despite the emphasis on story and characters. It's the difference between visuospatial "fun" and verbal "fun." The former is deductive, while the latter is creative. We might both enjoy solving puzzles and writing poetry, but we enjoy them differently. It's hard to conceive of speech as a "game" with strictly defined rules and attributes.
To this end, I'm not a fan of making speech into a "mini-game" ala Oblivion, and don't think Bethesda's failure is unique - developers have had similar ideas for a long time, and they never worked. I do think, however, that the reliance on combat can be reduced by making exploration a bigger focus for RPGs. As several people said, make the player work for a diplomatic solution, don't just hand it to them with an attribute check. It makes sense that, to avoid a fight, players have to look outside of the box, because people who are about to fight have already exhausted the "obvious" diplomatic solutions.
Thus, in terms of making out of combat improvements, I'd improve exploration as a priority. The main attributes of a diplomatic protagonist ought to be "Perception" and "Knowledge", instead of "Charisma." You could then build quests and encounters around those attributes. For example, Perception could result in detective skills that find hidden objects & clues in certain locations. Knowledge could result in information gathering skills that allow the player to obtain additional facts & useful lore from people & places of knowledge, and to examine the longer term consequences of decisions. Together, they would allow the player to put together an effective solution that combat characters wouldn't be able to reach. You could even make $ an important ingredient in all this "diplomacy," as a replacement for the gear that a combat character buys.
That is one way to approach it I guess but it doesn't feel all that satisfying. I think you have to involve the PC in there somehow, even if it's something like -- your character picks up a history book and reads, roll Perception to see if he notices the important information. If so, a beneficial dialog option unlocks in conversation.
As no developer is going to write a bunch of books for the player to read in their entirety, high Literacy could be used to simply reveal useful snippets vs. not so useful snippets, and then you could even involve a random roll for which snippet to show. The player is still involved because he needs to comprehend what he's reading and decide whether they are useful. The "burden of design" is actually not to use this to unlock a hidden conversation option, but to distill it into a series of facts/clues, which have to be acted upon. Thus, the book might hint at where to find another, more important clue. Else, it might hint at what to say to a NPC to achieve the best effect. The player has to decide what to do with the new information, not an automatic trigger; this makes it similar to combat, where gaining a new combat ability doesn't equate to automatically winning battles.
When I first thought of successful non-combat gameplay, which makes me think primarily of dialogue, I think of the system used in Fallout and FO2. The system seems simplistic, which isn't a bad thing, but has a lot going for it under the surface that modern games could take notes from.
Lets consider one of the more ornery talking-head NPCs, such as FO2's Overseer First Citizen, Lynette in Vault City. A lot of behind the scenes action goes into the PC's conversations with this lovely woman that the player is likely never totally aware of. Use of the Empathy perk helps understand this better by color coding dialogue choices to see how she reacts. Lynette's (and most of the other talking-head NPCs in the early FOs) reactions are what makes this gameplay brilliant and way ahead of mostly anything in the 15 some years since.
Your dialogue choices with Lynette are assembled from a variety of sources. Some are standard vanilla questions and responses. Some come from gameplay experience, like whether you've learned about Vault City's policies regarding slaves, er, um, servants, or seeking a diplomatic solution between Vault City and the neighboring town, Gecko. Others are gated behind the PC's statistics, charisma and intelligence as well as speech skill and are usually identifiable as the more verbose responses unless the PC has exceptionally low intelligence, where she handles the PC much differently. Lynette even has checks based on player inventory, such as showing her the vault 13 jumpsuit and canteen. Most of these options either raise or lower your standing with Lynette, which is all handled behind the scenes except for her talking-head changing to one of 3 states (pleased, neutral, angry). Make her angry enough, she'll throw your ass out of town, citizen or not.
This conversation is an excellent example on account of how naturally it plays out. It doesn't encourage save-scumming or scouting for statistical requirements and returning later after a few level ups. It's a complex interaction between the PC and an NPC drawing from a variety of factors with the player mostly unaware of how awesome and revolutionary (for its time and... well, still) it all is because it's handled elegantly and believably. Non-combat RPG gameplay at its finest.
