Rethinking non-combat gameplay

[Zanzoken]

One thing that has been on my mind lately is how simplistic non-combat gameplay is in most RPGs.

Even in the most basic combat systems, outcomes are determined by many factors. Attributes, skills, equipment, abilities, enemy stats, die rolls -- it all comes together in ways that make you think, plan, and strategize.

Non-combat, on the other hand, usually boils down to a simple skill or attribute check. If you possess or roll a high enough number, you pass -- if not, you fail. Imagine how silly it would be if combat worked this way.

Question is, how would you design non-combat gameplay so it rivals the complexity and richness of a combat system? I am curious to hear how games that want to feature stealth- and/or persuasion-focused gameplay might innovate beyond the basic skill check.

 

[Hugh Vagoth]

It's not just RPG's, games in general are terrible at handling non-combat gameplay.

 

[vonAchdorf]

The biggest difference is that combat is often a long series of modified skill checks, while most non-combat is only one (modified) skill check. Though in both cases the checks are influenced by skill level, feats, enemy stats and equipment. The difference is that with just one skill check, there's no reaction, no changing circumstances and therefore few possibilities to plan and react yourself.

Sometimes crafting is modeled after the combat experience, requiring a series of skill checks, modifying the conditions and allowing you to use different gear and skills. For example Final Fantasy XIV (the MMO) uses that formula extensively it its crafting system.

The dialogue mini-game in Oblivion also tried to go a similar route, though it was executed quite badly.

 

[hell bovine]

One thing that has been on my mind lately is how simplistic non-combat gameplay is in most RPGs.

Even in the most basic combat systems, outcomes are determined by many factors. Attributes, skills, equipment, abilities, enemy stats, die rolls -- it all comes together in ways that make you think, plan, and strategize.

Non-combat, on the other hand, usually boils down to a simple skill or attribute check. If you possess or roll a high enough number, you pass -- if not, you fail. Imagine how silly it would be if combat worked this way.

Question is, how would you design non-combat gameplay so it rivals the complexity and richness of a combat system? I am curious to hear how games that want to feature stealth- and/or persuasion-focused gameplay might innovate beyond the basic skill check.

Make you work for it. If you need to "convince" an NPC to do your bidding, a skill check shouldn't be the solution. Have the player do some detective work and uncover a dark secret from the past (or just kinky sex habits at the local brothel), in order to blackmail them. Or use bribery, but instead of having to gather X amount of money, the player has to obtain a very specific item, which that NPC wants. Or - if you really want to persuade that NPC in dialogue - learn what type of person they are, their history and believes, in order to successfully debate them.

 

[Siveon]

It's not just RPG's, games in general are terrible at handling non-combat gameplay.

Do games without combat at all count? If so, I'd have to disagree.

 

[Vikter]

I think non-combat should rely less on variable and easily manipulable stuff. One bad example is Mass Effect's romances: if you talk to someone a few times and act nice, you bang this person and they will be forever in love with you. It's a really deterministic aspect because they are just waiting for you to woo them over.
Maybe in terms of relations with any other NPCs, there should be a RNG-based "chaos theory" happening all the time, or something like that. Instead of only having a single variable from 0 to 100 saying how much they like you, NPCs should have other flags, like doubting the player even if the relationship is at max level, or having erratic behavior based on time, etc. Anything to make conversations less min-maxable and more about actually talking to someone.

To back it up, there should be different sets of attributes for other activities. Maybe reading skills, so that you can read more books in the game, or something like that. It's hard to think of it because while most games tend to focus on a cool balanced combat, all other actions seem trivial or based in attributes that you already use in combar OR are exclusively for non-combat (charisma, usually). It would be nice to have a game that has more attributes that influence your actions outside of combat, indeed.

 

[V_K]

Non-combat, on the other hand, usually boils down to a simple skill or attribute check. If you possess or roll a high enough number, you pass -- if not, you fail. Imagine how silly it would be if combat worked this way.

Question is, how would you design non-combat gameplay so it rivals the complexity and richness of a combat system? I am curious to hear how games that want to feature stealth- and/or persuasion-focused gameplay might innovate beyond the basic skill check.

The approach that seems to have worked best so far is to wrap that single check in a puzzle - i.e. the QfG way. Granted, that limits the scope of the game, but arguably any additional full-blown system would, since resources are finite.
To make some sort of systemic non-combat gameplay as deep as (good) combat you'd need three things:
-different ways to do it (e.g. stealth could be achieved through traditional sneaking, or jumping and climbing through places no one would look, or disguise use);
-level design (and possibly AI) that supports it;
-some sort of resource management involved (smoke grenades, disguises etc);
I'm probably the biggest sucker for non-combat gameplay here, but even I don't believe that's feasible (but let's see what Otherside will come up with, maybe I'll be proven wrong). Puzzles seem a safer bet. On the other hand, an RPG that has only non-combat options would be rather intriguing but would probably be a complete commercial flop.

 

[Reif]

As far as I can see, dialogs don't suit this complexity well because it requires orders of magnitude more effort. Crafting (the actual process, not getting and combining the components) on the other hand is imo boring by principle in games.
As I see it, Exploration is the most likely candidate to become a principle mechanic next to combat, at least in third person 3d games.

If I had to design such a game, it would comprise an "exploration" skilltree next to the combat one(s).
This tree would consist of types of movement like swimming or climbing, e.g. with higher climbing skill, the player would reduce the chance for slipping off different surfaces (with too low skill making it virtually impossible to climb certain slopes).
To make trying to e.g. scale a mountain not a glorified skill check, I would let a small chance to slip off persist even at higher levels so that players naturally try to find the "savest" route to the top.
If this is combined with time-critical elements, like flying/ranged enemies by that one would rather not be spotted in the vulnerable position that is hanging at the side of a cliff, exploration might become as significant as combat.

This would also naturally fit mmorpgs and especially Bethesda games, which's actual focus - exploring - is in a way hilariously underexplored, as it consists entirely of walking around and occasionally trying to scale mountains by jump abusing.

Another example regarding both exploration and crafting: Instead of concealing the alchemical properties of otherwise clearly identifiable ingredients like in those said Bethesda games, I might conceal the type of the ingredient itself. So certain kinds of quite similarly looking herbs would look exactly the same for players with too low alchemy/herbology/whatever skill.
And now that I mention it, this could also work for swimming - visually (subtly but noticably) show currents only to those with high enough skill. Or noticing likely to collapse dungeon floors/walls.

 

[Neanderthal]

Case by case basis is best thing for it I think, for instance Arcanum where farmer asks you to get rid o them pigs, slaughter shouldn't be only solution cos what thick bastards farmer wants to get rid o his swine? Animal charming spells should come into play if you're a mage, for a diplomat character arrange for another farmer to buy em, tech character makes an automated scarecrow, dumb Half Ogre roars and chases em off etcetera.

Manual work has always interested me in games, the next quest in Arcanum for that farmer is to clear his fields of stone, now this presents problems straight away: The stones take up a shitload o room and you're probably carrying around a right load o shit, now you could perhaps have a strength or constitution check here to see if you strain anything while lifting em, though doubt this'd happen for a strength character. After doing this and re-arranging your inventory probably you could have had a burden of fatigue put on you, tired more than usual you get fatigued earlier wi usual penalties. Same wi gathering pumpkins for farmer Brownie in Britannia.

Baking bread though I thought was pretty good in Ultima, and various tasks there, you had to know what to use, what to do and what things'd mix and break down into what. It were a pretty good way o realising this kind o thing, and though it could get boring if repeated, it were still always useful. Some o new content and bits and bobs added by Exult and Keyring mod stand out like this, simple as they should be, but really add to world and what you can choose to do other than grinding monsters.

Think magical experimentation and research should be a thing an all, back in old days all my mage players wanted to make and name their own spells, but now we seem content wi spell lists and magic only being used in bloody combat which renders it just a tiresome, mundane weapon. What happened to mage casting light in darkness, summoning a disc to carry his loot, translating writings, sensing where good or evil came from, raising spirits to advise him, and using a dozen or more tricks to astound and amaze?

Thought the items and their use in Betrayal at Krondor was pretty good in this respect an all, made sense how you could use em, lute for barding, hammer for armour, whetstone for weapon, oil for bowstring, elven boots for sneaking and all the books you could read that would aid the various skills. The puzzle boxes were good an all, few really bloody hard uns I never did manage to open, but most o em were pretty logical. Everything was once again pretty simple and not using a complicated system, but it didn't need to as it was a break from the system heavy combat and recovery.

Personally I think dialogue skills should just gi you options to choose from, and everybody you talk to responds differently, a bluff old soldier might like plain speech and getting straight to fucking point, or you could have Gann like blokes who like wordplay and cleverness, a priest might be on the surface a faithful man but if talked top for long enough will almost admit his interest in arcane lore and the forbidden texts the church protects, and you can either choose what image you want to try and get across depending on how you see the person. Or not and just be yourself, which might have rewards in and of itsen.

 

[Excidium II]

If you're gonna add gimmicks and minigames leave stuff the way it is.

 

[pippin]

Anyone ever tried to have dialogue be combat itself? Sorta like how Shadowrun Returns does hacking.

I think the etiquettes were a good example of both character customization and gameplay diversity, even when those were just another kind of skill checks. I've always believed origin stories should matter more in rpgs, giving you special tools, abilities and dialogue options. For me, this could be a good example of non combat gameplay done well.

 

[Neanderthal]

Weren't there some early alpha footage o Witcher 2 where Polish Steven Seagal is fighting one o kingslayers an holding a conversation at same time, nowt in end game like except for some convos wi Letho, but that weren't mechanic driven.

 

[Zanzoken]

Make you work for it. If you need to "convince" an NPC to do your bidding, a skill check shouldn't be the solution. Have the player do some detective work and uncover a dark secret from the past (or just kinky sex habits at the local brothel), in order to blackmail them. Or use bribery, but instead of having to gather X amount of money, the player has to obtain a very specific item, which that NPC wants. Or - if you really want to persuade that NPC in dialogue - learn what type of person they are, their history and believes, in order to successfully debate them.

This is along the lines of what I was thinking. For dialogue, systemically a few of your attributes might be Perception, Beauty, Humor, Logic, and Intimidation and maybe you have skills like Music, Arcane, Mechanical, History, etc that represent things your PC is knowledgeable about. And maybe even props, like a forged signature or something of that nature.

So let's say you need to talk a scholar into showing you a banned book, so you can steal it and replace it with a fake. The battle is the conversation, the NPC is your enemy, your abilities are things you can say and do to get what you want, and your attributes and skills affect how convincing you are. Doing damage gets you closer to your goal, taking damage is when the NPC argues or resists, you can buff by playing yourself up and debuff by playing the NPC up (i.e. raising his disposition). The conversation ends either with them agreeing to what you want, or telling you no, which is a fail state.

The design goal would be to make the player play to their PC's strengths but also have to switch up their tactics. If you make a character who tries to be intimidating but is short and wimpy, you will fail -- it would be like making a fighter with no strength. But even if you're a big brute you can't just use intimidation on everyone, because some people don't scare, just like for instance creatures who are immune to fire. That's where you get to create complexity and challenge via encounter design.

That's building it analogous to combat anyway, which seems plausible. But who knows, the combat design structure might also completely fail to translate and not be fun in the least.

 

[Zanzoken]

If you're gonna add gimmicks and minigames leave stuff the way it is.

Agree 100%, it needs to be implemented holistically and not just feel tacked on. The system needs to have as much thought and development put into it as combat to really pull it off.

 

[Somberlain]

Multiple skills and attributes that govern non-combat stuff and some additional quest items and actions that are involved are enough. All stupid minigames can fuck off.

If you want to make dialogue and other non-combat stuff as complex as combat in a proper RPG, just make your game completely built around that. Trying to make a game with complex combat and equally complex non-combat stuff will never work with realistic resources.

 

[AW8]

Question is, how would you design non-combat gameplay so it rivals the complexity and richness of a combat system? I am curious to hear how games that want to feature stealth- and/or persuasion-focused gameplay might innovate beyond the basic skill check.

Two things I would do is to make Speech skills less binary, and have the diplomatic approach require more preparation. In most games, your Speech check is just a simple dialogue check whose outcomes are:

1. Makes the enemy leave peacefully/give you the McGuffin/kill himself
2. The check fails and you're forced into combat.

That's boring and also hilariously unrealistic. With a few well-placed words, you can often change an NPC's entire view of the world. It's cheap and not convincing at all.

A Speech playthrough should be more involving than walking up to strangers and talking them into suicide. A combat character buys gear, learns special attacks, and plans out tactics. What could a Speech character do? Well, if the goal is to persuade a diplomat to support your faction, you need leverage. Sneak into his office and steal some embarrassing evidence. The evidence in hand, you might be able to convince him with a speech check. But even if you fail the check and combat starts, your preparation should have some effect, like making him emotionally inbalanced and lower his combat skills. Binary is boring. A "grazing" Speech check that tips the scales somewhat in your favor will make the player's work feel worthwhile instead of wasted.

Just as there are trash combat disposed of with ease, there should be simple Speech checks that doesn't need preparation as well. If a backalley mugger demands money, it's OK if the speech character gets out of the situation peacefully by simply telling him to piss off, convincing him this mugging isn't worth the risk. Other on-the-fly Speech checks could nudge things in the player's favor in small ways - such as getting extra bullets from the quartermaster, opening up a bribing option to get into a party etc. Small stuff that makes the player feel like they're playing a talky character. But major Speech checks should definitely demand some preparation, like planting evidence, stealing material for blackmail, sabotaging defenses/surveillance (make your persuasion target feel unguarded) etc.

Example:
DeBaddie has the McGuffin in Evil HQ. There are a few ways to get it:

Walk in there, start combat, kill DeBaddie, take the McGuffin. (Combat option)
Sneak in there, steal the McGuffin. (Stealth option)
Aquire pardon for DeBaddie's gang, walk into Evil HQ, pass a speech check, get the McGuffin. (Speech option)
Aquire pardon for DeBaddie's gang, walk into Evil HQ, fail a speech check, start combat but get some of DeBaddie's gang on your side or at least out of the way (enticed by the pardon), kill DeBaddie, take the McGuffin. (Speech/combat option)

So basically the backbone is not the player's words, but the pardon. The player's words are applied afterwards and tip the scales a bit but doesn't define the entire conversation. If the player's Speech skill isn't high enough, DeBaddie won't be convinced that it's better to jump ship from the doomed Evil Organisation, instead opting to continue to serve Evil Master for great rewards down the road.
And there's no absurd "What you are doing is wrong! Don't your realize this?"-speech check that changes his entire view on life with a couple of sentences.

 

[Telengard]

The reason non-combat stats isn't as complex as combat is, the output of combat a single number - your damage. The output of non-combat is lines of text for social, or minigames for rogue. If one were to make the output of social as simple as the output for combat - say an icon - then social can easily be made as complex as combat.

Spoiler

Think a mixture of the Sims and Bioware.

To do otherwise means one would needs 1 text output for every fine gradation of your skill roll, in order to make all of those attributes matter instead of just be window dressing. And that's really resource hungry. An average rpg is five novels worth of text. Put a complex series of outputs on top of that, and you just turned it into 25 books of text.

Or, just talk to VD about Age of Decadence development...

 

[Space Insect]

I also would prefer to have the speech checks hidden behind logical dialogue choices. For example, in BG2, when romancing people you would not be told which dialogue options leads to you having sex, like in modern Bioware games. I would like for the choices to be ambiguous and have the player use logic and reasoning to decide which option is the best option. Also, to keep with the skill-based checks, the stupid dialogue options could make the skill check harder, while the logical ones could make it easier.

 

[Bruma Hobo]

Forget about skillchecks and minigames, just give the player a parser-based dialogue system and let him connect the dots. Ultima IV, Wizardry VI and Wasteland did it right. Then you can implement a roll against charisma everytime the player asks something, and if it fails the NPC should not answer the question or even stop the conversation (did Star Trail do that?).

You can also implement a hidden reputation system (like in Ultima IV), assign a "virtue" to each NPC and make them react to the player's reputation. So, if your character is a known liar, honest NPCs will not talk nor trade with him, and to fix that the player will have restore his reputation or live with the consequences.

 

[an Administrator]

I like this kind of persuasion in rpgs

 

[laclongquan]

The hacking and lockpicking minigame in Fallout New Vegas can be pretty hard.

Legend of Mana offer a separate strand of gameplay call item crafting. you can set up the lands of the world (dont ask!) so that your garden produce most interesting kind of produces. From them, you can enchant items of various characteristics to make some interesting (and powerful) items. The enchantment in particular can be quite lengthy.

Triple Triad of Final Fantasy 8. You can play games to get card, materialize cards into items needed for other purposes, manipulate rules of world's regions to your design. It's the most interesting card games of the entire FF series.

Morrowind's alchemy and enchant process. You can use each to booster its own strand, (make more potions to enhance stats to make more potions) or cross-strand (make spell to enhance skill to make potions to enhance spell related stats)

The trade routes of Might and Magic 8.

 

[likaq]

Dear OP.
There is no such thing as 'non-combat gameplay'.
How do I know that?
Just ask mondblut, jasede, porkamorka or MMXI.

For example (porkamorka post ):

A brief tour of the Decline of proper CRPGs (non blob, non action, party based P&P simulators);

Start with Pool of Radiance (1988), for an example of a CRPG that inclined the genre and was largely untouched by Decline.

Next, play Curse of the Azure Bonds (1989), and notice how several interface issues from PoR were cleaned up, but also note that many ambitious design concepts from PoR have been simplified, streamlined and/or eliminated. For example, parlaying with monsters, finite numbers of wandering monsters, hiring NPCs from the training hall, animating party members as undead, interesting stuff to find on the overland map, dungeons that sort of make sense, etc.

Next, play Secret of the Silver Blades (1990). Some of the Decline that I attributed to Curse of the Azure Bonds may actually have occurred here. The world map has been largely eliminated/rendered irrelevant and you're mostly just fighting endless wandering monsters in a linear dungeon that no longer makes much sense at all. Somehow there are hordes of dragons and high level mages hanging out in this abandoned mine. It's worth noting however that this is a low point for the Gold Box games. They would never again be as ambitious as Pool of Radiance (1988), but they would bring back the world map and generally settle into a formula that offered solid gameplay with some P&P style adventure and more of an illusion of freedom than SotSB.

If you're taking the extended tour, it might be interesting to take a look at the Dragonlance games at this point, there was certainly a significant potential for decline as the plot was to some extent driven by NPCs from the official (bad) Dragonlance novels. Many people see them as part of the Decline. However, I would NOT consider the Dragonlance series to be true Decline, as there were a number of additional mechanics added to make the games more complex and interesting and there isn't really a significant change in the amount of freedom that is offered (not much). If Tanis Half Elven hadn't showed up to make you do something, some generic NPC/event would have.

Next, play Dark Sun: Shattered Lands (1993). Aside from the turn based combat, this game should immediately feel familiar to early 2000s CRPG players. Notice how in-level exploration has been streamlined/simplified/reduced to a shroud removing formality by the switch from first person dungeon exploration to zoomed out third person exploration. Exploration has changed from a game into a task. Notice how much more time you spend on reading text boxes and clicking on options from lists. Notice how much less time you spend on tactical combat and adventuring. Notice how the party has been reduced to four members. Notice how there is more diversity in character options and how more spells are modeled, but also notice how the tactical combat has not significantly improved. Notice how much easier the game is if you select certain types of characters.

Next, play Fallout (1997). Fallout technically belongs to its own short lived sub genre of games, but it is also highly relevant to any discussion of Decline. Take note of how much time you're now spending on reading text and selecting options from lists. Take note of how the turn based combat is so simplistic that it barely even qualifies as a game anymore. Realize that nothing of value was lost when they switched this series over to shitty FPS combat. Finally, give Ian a submachine gun and stand in front of him in a fight.

Next, play Baldur's Gate (1998). This game is highly relevant to any chronology of the Decline, but in and of itself, it is not a bad game. However, it is worth noting that the strategic ganeplay involved in party building has been replaced by simply picking from a modest sized list of recruitable characters. BG popularized real time combat in CRPGs, but since turn based combat in CRPGs hadn't significantly inclined since the very early 90s, the combat in BG isn't actually problematic. Recruitable party members will occasionally interrupt gameplay to make you read text and click an option from a list, but it isn't a significant distraction yet. Not everything is positive for this game though, it really drives home how tedious the shroud clearing task can become.

Next, play Baldur's Gate 2 (2000). In absolute terms, this is without a doubt the best CRPG ever created, but if you take the year of creation into account you can't help but feel disappointed at how little things have improved since 1989. Virtually all future decline can be traced back to this game, but in most areas it was a better game than Baldur's Gate (1998), even though you need to click through a lot more text. However, it is worth noting that party building has been simplified even further as there are not very many recruitable characters to pick from anymore. Recruitable characters are interrupting gameplay more often, but it still isn't particularly irritating. Romances appear; they will eventually transform this into a genre for perverts, but they're quite tame and minor here.

Next, play Pool of Radiance: Ruins of Myth Drannor (2001). Don't play for long, but get a feel for the combat in this game. Obviously, a huge step down from Baldur's Gate 2. But amazingly, the combat is not much more interesting than the combat from Curse of the Azure Bonds (1989) and it takes far longer to play. They seem to have forgotten how to design turn based games.

Next, play Knights of the Old Republic (2003). Notice that information is no longer presented to the player in a clear, 2D format, but is now obscured by messy 3D graphics. Party building has declined further, you now have only two recruitable characters as party members. But don't worry, they'll nag you quite a bit.

Next, play Temple of Elemental Evil (2003). Play through Hommlet only, you can stop after that. Sorry, that was cruel, I just wanted you to experience the worst level ever included in a CRPG, three hours of walking around, reading mundane text and clicking on options from lists. This game did finally improve turn based combat in CRPGs above 1989 standards, but was crippled by horrible level design, so it did not have any significant effect at halting or reversing the decline, aside from (probably) inspiring KoTC.

Next, play Knights of the Old Republic 2 (2004). Information is again obscured by messy 3D graphics, but this time combat has been made so incredibly easy that it is almost a formality. Many players report going through the entire game with no need to use any consumables and there is little need to pause to consider your tactics. While this game still has the mechanically robust d20 system behind it, it is on par with Fallout in terms of terrible combat gameplay, simply because the game is so easy. With combat and exploration providing negligible gameplay, most of the "game" is about reading text and clicking on options from lists. We've almost reached full decline here, we just need some romances to attract the perverts.

Next, play Knights of the Chalice (2007). Take note of the many, many ways in which turn based combat in CRPGs is finally improved from the standards of 1989. Take note of the fact that this game did not sell enough to warrant a sequel (Codexers complained about the graphics quite loudly) and that the developer is now doing a real time strategy game.

Finally, play Dragon Age and Dragon Age 2. Just be sure to pirate them.

So, a summary of the decline. Initially, exploration provided gameplay in the form of first person mazes. This was eventually replaced by the task of clearing all the shroud from the map. Eventually, this was replaced with third person 3d levels, but this didn't bring back significant amounts of gameplay to exploration. (Image of completely linear DA2 mini map goes here).

Initially, turn based combat in CRPGs was good. But then it failed to incline (in CRPGs) for like 14 years. In some significant ways it actually got worse. By the time turn based combat inclined again, it was too late.

Initially, real time with pause combat in CRPGs was not bad. But then it got bad.

Initially, building a party offered strategic gameplay. But then you just got to pick your party from an ever shrinking list of RPCs. Eventually these RPCs started trying to molest you.

Initially, gameplay was the focus of RPGs. But eventually, reading (or listening to) words and clicking options from lists became the focus of RPGs. That and molesting RPCs.

As you see gameplay=combat. Combat and only combat can be considered as gameplay, so 'non-combat gameplay' does not exist.

 

[hell bovine]

Make you work for it. If you need to "convince" an NPC to do your bidding, a skill check shouldn't be the solution. Have the player do some detective work and uncover a dark secret from the past (or just kinky sex habits at the local brothel), in order to blackmail them. Or use bribery, but instead of having to gather X amount of money, the player has to obtain a very specific item, which that NPC wants. Or - if you really want to persuade that NPC in dialogue - learn what type of person they are, their history and believes, in order to successfully debate them.

This is along the lines of what I was thinking. For dialogue, systemically a few of your attributes might be Perception, Beauty, Humor, Logic, and Intimidation and maybe you have skills like Music, Arcane, Mechanical, History, etc that represent things your PC is knowledgeable about. And maybe even props, like a forged signature or something of that nature.

So let's say you need to talk a scholar into showing you a banned book, so you can steal it and replace it with a fake. The battle is the conversation, the NPC is your enemy, your abilities are things you can say and do to get what you want, and your attributes and skills affect how convincing you are. Doing damage gets you closer to your goal, taking damage is when the NPC argues or resists, you can buff by playing yourself up and debuff by playing the NPC up (i.e. raising his disposition). The conversation ends either with them agreeing to what you want, or telling you no, which is a fail state.

The design goal would be to make the player play to their PC's strengths but also have to switch up their tactics. If you make a character who tries to be intimidating but is short and wimpy, you will fail -- it would be like making a fighter with no strength. But even if you're a big brute you can't just use intimidation on everyone, because some people don't scare, just like for instance creatures who are immune to fire. That's where you get to create complexity and challenge via encounter design.

That's building it analogous to combat anyway, which seems plausible. But who knows, the combat design structure might also completely fail to translate and not be fun in the least.

Yeah, this would be definitely a better system than the usual one skill check (but: see EDITs). I'd disagree on intimidation being tied only too looks, however. A big muscled guy is going to look intimidating, but sometimes it is not how you look, but how you act; insanity can be a lot scarier than brute strength. (e.g. Heath Ledger's performance as Joker)

EDIT: I would advise against tying everything in conversations to numbers, though.

For example, if you need to convince a lord of a clan to join you in a military campaign, then learning about the history of his clan or family (even if it is something as simple as reading books in game) should give you additional information, that your character can then use to their advantage. After obtaining said information another dialogue option would be open to the player: " but you great grandfather So-and-so has said there is no greater honor than dying with a sword in your hand...".

If you tie everything to skill checks and attributes, then you get an over-bloated version of the persuasion check system.

EDIT the 2nd:
To be honest, If I were ever designing a game - which is not going to happen, unless I win the lottery or marry rich - I would abandon all skill checks in dialogues altogether. Skills and attributes make sense when designing a character for combat. But in conversations, the amount of options open to your character should be related to the happenings in the game: reading books, listening to gossip, paying NPCs for information... Even something as simple as reading a gravestone at the local graveyard could give you clues for the murder quest, making the player pay attention to the world in game.

Maybe I simply have enough of everything solved by the charisma score.

 

[Jedi Exile]

The hacking and lockpicking minigame in Fallout New Vegas can be pretty hard.

Lol no. It's easy as hell, because save/load. If you don't save/load, then it could be slightly more challenging, but that's it. You could say the same about pickpocketing in IE games and other non-combat options, like stealth for example. The developers probably assume that no one use this options anyway (like newspapers in Arcanum which supposedly no one read), so they make them as easy/skippable as possible.

 

[Zanzoken]

snip

Yeah, this would be definitely a better system than the usual one skill check (but: see EDITs). I'd disagree on intimidation being tied only too looks, however. A big muscled guy is going to look intimidating, but sometimes it is not how you look, but how you act; insanity can be a lot scarier than brute strength. (e.g. Heath Ledger's performance as Joker)

EDIT: I would advise against tying everything in conversations to numbers, though.

For example, if you need to convince a lord of a clan to join you in a military campaign, then learning about the history of his clan or family (even if it is something as simple as reading books in game) should give you additional information, that your character can then use to their advantage. After obtaining said information another dialogue option would be open to the player: " but you great grandfather So-and-so has said there is no greater honor than dying with a sword in your hand...".

If you tie everything to skill checks and attributes, then you get an over-bloated version of the persuasion check system.

EDIT the 2nd:
To be honest, If I were ever designing a game - which is not going to happen, unless I win the lottery or marry rich - I would abandon all skill checks in dialogues altogether. Skills and attributes make sense when designing a character for combat. But in conversations, the amount of options open to your character should be related to the happenings in the game: reading books, listening to gossip, paying NPCs for information... Even something as simple as reading a gravestone at the local graveyard could give you clues for the murder quest, making the player pay attention to the world in game.

Maybe I simply have enough of everything solved by the charisma score.

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.