Character Progression and the Minimal Elements of an RPG

[galsiah]

Personally, I see two aspects as essential to RPGs in this area:
(1)The player must have a wide choice of character options to start with.
(2)The character must change over the course of the game.

I don't think anyone disagrees with (1). Some might not agree with (2), but I think that is necessary. What is not necessary is achieving it through stats. It is also not essential in my view for the character to be making "progress", or to be getting better.

Increasing stats is an easy and fairly inoffensive way to achieve (2) without needing to think about it much. It's tried and tested, it works and it's safe. It's not the only way though. What is necessary (IMHO) is that the player's idea of his character changes in interesting ways. Stat increases provide an easy cue to change a player's perception of his character - i.e. he's becoming a better swordsman / mage, or becoming stronger... Stat increases aren't that interesting though. Each one usually has fairly small, predictable implications.

A player can be shown the changes in his character in different ways. E.g. through dialogue decisions, story events... It's probably hard to keep this up for long though. Stat increases might not be that interesting, but they can be constantly present - preferably in the background.

There's also no necessary reason that stats / skills need to increase. An RPG could be based around a story where your character is slowly dying. Things start off easy, but as your character becomes less capable, problems become more challenging and require more creative solutions... It could be done, but would be hard to get right: players will more easily forgive a reward happening at an odd time for no predictable reason, than a penalty. I think it could be interesting, but it would have to be well designed - it's certainly not a safe premise.

Stat progression is useful as a clear (though uninspired) indication to the player how his character is changing, and as a gameplay device to provide more options (or fewer). Both these things can be achieved in other ways, but stat development is an easy way to do it. Personally I like character progression to be part of an RPG (usually in any case), but I want it very much in the background, or connected to the game world. If I make a character progression decision, I want it to be a character decision:
E.g. Gaining the potential to learn a new spell from a powerful mage, and picking one - fine.
Allocating my 3 points for this "level" to strength because I want to be stronger - stupid.

 

[AlanC9]

Don't most game characters undergo such changes? Gordon Freeman and April Ryan don't finish their games in anything like the place where they start.

Freeman actually undergoes a kind of progression, as his weaponry upgrades. According to the story Ryan does as well, though in an adventure game it's hard to tell.

 

[Voltare]

what i would like to do one day is a stat de-proggression rpg.you start off a high level, with max stats, and have to de-level to get back through the game.not sure yet how it would work....just a thought...

 

[galsiah]

Don't most game characters undergo such changes? Gordon Freeman and April Ryan don't finish their games in anything like the place where they start.

Sure, but for an RPG there needs to be a wide variety of possible changes and destinations. Things should be very different according to the type of character played. Whether this is accomplished through stats, story or something else is less important.

 

[Diogo Ribeiro]

Personally, I see two aspects as essential to RPGs in this area:
(1)The player must have a wide choice of character options to start with.
(2)The character must change over the course of the game.

I don't think anyone disagrees with (1). Some might not agree with (2), but I think that is necessary. What is not necessary is achieving it through stats. It is also not essential in my view for the character to be making "progress", or to be getting better.

Increasing stats is an easy and fairly inoffensive way to achieve (2) without needing to think about it much. It's tried and tested, it works and it's safe. It's not the only way though. What is necessary (IMHO) is that the player's idea of his character changes in interesting ways. Stat increases provide an easy cue to change a player's perception of his character - i.e. he's becoming a better swordsman / mage, or becoming stronger... Stat increases aren't that interesting though. Each one usually has fairly small, predictable implications.

A player can be shown the changes in his character in different ways. E.g. through dialogue decisions, story events... It's probably hard to keep this up for long though. Stat increases might not be that interesting, but they can be constantly present - preferably in the background.(...)

Stat progression is useful as a clear (though uninspired) indication to the player how his character is changing, and as a gameplay device to provide more options (or fewer). Both these things can be achieved in other ways, but stat development is an easy way to do it. Personally I like character progression to be part of an RPG (usually in any case), but I want it very much in the background, or connected to the game world. If I make a character progression decision, I want it to be a character decision:
E.g. Gaining the potential to learn a new spell from a powerful mage, and picking one - fine.
Allocating my 3 points for this "level" to strength because I want to be stronger - stupid.

Not sure where I'm going with this but... I always thought that one of the problems with simply and arbitrarily assigning experience points or upgrades troughout the course of a game to a character relies on how it's used, not with the underlying system itself. For most computer roleplaying games, character development is much faster than in a pen and paper session; and the gameworld is also much more finite and contained than in pen and paper.

I think character development doesn't need to be necessarily tied to skills or statistics - or at least to have these fully noticeable. A PC may perform a deed for a land owner or noble who will then provide the PC with free access to the land's higher social circles or some other previously inaccessible parts of the region. In this case the character has advanced in the game, has changed, and has grown in terms of what he can do. No statistic necessary to demonstrate this.

On the other hand, a game like Torment rewarded a player with knowledge and abilities that would be useful later on by simply interacting with NPCs. For instance the Nameless One could not get into the Siege Tower before he discussed its presence with the son of a weaponsmith. When he gains entry to it he does so without any special ability - just character knowledge. Of course, this was also tied with the setting where belief is power which is aptly demonstrated in this example, but similar things can be applied to other settings.

Since I'm working on a NWN2 module, I've had to try to handle things a bit differently when it comes to the outcomes of certain situations. Several quests just give information to the PC, a couple are meant to give special abilities and such, but the underlying system still belches out experience points so players will level up and spread out their points across statistics which they might want. So I'm trying to balance both things.

 

[Wysardry]

I think character development doesn't need to be necessarily tied to skills or statistics - or at least to have these fully noticeable. A PC may perform a deed for a land owner or noble who will then provide the PC with free access to the land's higher social circles or some other previously inaccessible parts of the region. In this case the character has advanced in the game, has changed, and has grown in terms of what he can do. No statistic necessary to demonstrate this.

If a game keeps track of when the PC gains access to someone or something, it's still supporting a form of character progression via a statistic, even if it only has two possible values.

Using statistics and/or character progression doesn't require them to be explicitly shown to the player.

 

[DarkSign]

Yes but you still havent addressed certain problems that either get played out with statistics or frustration.

Say you're a rogue/theif and you come upon a lock.
Now with a no-stat system, you have no idea whether you should attempt opening it or not.

Lets say this lock is the Whizomatic 3000 - meaning it needs a functional lockpick skill of 3000, but you're only at a low functional level.

Do you just keep trying until you give up...do you get a little text line that says "your skill is not even close to opening this"...or does something graphical in the world happen (your lockpicking tools break)?

Stats are also a way of saving players frustration...if you dont give them...then you have to replace them with something graphical, in my opinion.

 

[Section8]

Yes but you still havent addressed certain problems that either get played out with statistics or frustration.

Say you're a rogue/theif and you come upon a lock.
Now with a no-stat system, you have no idea whether you should attempt opening it or not.

Lets say this lock is the Whizomatic 3000 - meaning it needs a functional lockpick skill of 3000, but you're only at a low functional level.

Do you just keep trying until you give up...do you get a little text line that says "your skill is not even close to opening this"...or does something graphical in the world happen (your lockpicking tools break)?

Stats are also a way of saving players frustration...if you dont give them...then you have to replace them with something graphical, in my opinion.

And how often are stats like that exposed to the player? Does a good DM say "Okay, this lock has a DC of 27, and you have a skill of 8, so you take twenty and the lock opens," or does a good DM let the player figure it out for themselves?

You should try to save the player any unnecessary frustration, but there's nothing at all wrong with frustrating the character. Anyone who thinks otherwise needs to "man the fuck up" and accept that not everything is handed to you on a platter.

Besides I fail to see how displaying "Your skill is not even close to opening this" is any different to saying "This lock requires 3000 skill and you only have 3."

 

[galsiah]

You don't need to display any stats to get around that. You only need to give the player a way to tell an easy lock from a hard lock, and feedback to indicate how likely it is he'll be able to open it. If the feedback was terrible (i.e. no chance) for a similar lock last time, and the player has no reason to think he's improved, he won't frustrate himself trying a lock he can see is similar.

You do of course need a feedback mechanism more involved than success/failure. That feedback can be direct graphical feedback (e.g. text / game world graphic / HUD graphic), audio feedback, or game world feedback (e.g. picking a particularly difficult lock with little skill might break/damage tools, set off alarms, set off traps..., or a minigame which provides clear feedback indicating the chance of success - perhaps none).

Feedback is required whether or not stats are used though. The benefit of stats is that you know when a relevant stat changes that your chances to pick a lock may have changed. Without stats, you just need to use common sense (so long as the game system makes sense) - if there's a reason to think you've got any better at lock picking, then your chances have probably improved. Otherwise they won't have.

With stats, the system relies less on being intuitive and sensible, since the player can easily see changes even if they are odd. I'd say this is a bad thing. Remove the stats, and you put pressure on designers to make sure the system is intuitive and makes good sense - something that should be true in any case. [this applies less to turn based games of course - some stats really are necessary there]

Take Morrowind for example. If the designers had wanted to hide the stats from the player, they'd have been forced to create a fluid, intuitive system that made good sense. As it was, no such pressure existed, so they came up with an unfortunate mess of counter-intuitive nonsense.

 

[DarkSign]

So let's roll with that.

In character creation you take a JA2 approach and answer questions about your character to be, but dont attribute stats to the guy.

When you get in the world you basically find out how strong you are based on what you can pick up, how charismatic you are from ease of dialog choices, and constitution based on how long it takes you to die. In fact, you dont even have hit points per se, but you know see that your left arm wont raise your shield and your hobbling on one leg.

As you progress through the game, different actions grow and change your character and, as in real life, you'd just remember what you'd learned.

 

[galsiah]

I think it's doable, but you'd ideally have quite a few feedback systems. For instance, as you say, a character should be aware if he's too weak to use his shield at all properly without needing to get in a fight first - it should be clear as soon as he picks it up.

Equally, a character who is strong, but has never used a sword, should know as soon as he picks up a sword that he can't use it effectively. He probably remembers he can't in any case, but ideally he should get feedback as he would in real life - pick up a sword for the first time and it doesn't feel at all natural. It might be tricky to display this well - perhaps have the sword held in a clearly awkward way, or have it sway...

Preferably, the player shouldn't need to remember how skilled he is in every area (though usually he would remember). Trying the activity, or preparing to try it should always give feedback: I know I can't use a sword as soon as I pick it up; I don't need to try fighting first.

The trouble is that while feedback for some skills is fairly easy - e.g. weapons / shield / armour can cause clear graphical differences to a character's body position / movement... -, not all skills would be so easy to cover. As a designer, you'd probably want either to display all skill statistics, or none - half and half would seem a bit odd.
Ideally then you'd need intuitive feedback for every skill. That would be hard to do well. (though possible I hope)

 

[Solik]

How fun would that be to play, though? It could lead to a feeling of helplessness. Sure, I can make choices about my character, and I may have some idea of what those choices might cause in the game by drawing some parallels to reality, but the fact remains that the game is far, far removed from an accurate simulation of reality. As such, my guesses may be entirely off-base; that could be my fault, the fault of the developers, or neither (attributed merely to differing cultures and experiences). Choices around stat development, when the stats are visual, creates a global game language where, assuming sufficient documentation of the effects of stats exists, players are able to make informed decisions about how they want to play the game.

Feedback is great, and can assist subsequent play-throughs, but by definition, it always occurs after the choice has been made. This can be pretty damaging if the effects aren't felt solidly until after you've already rolled your character, or spent four hours developing a skill, etc.

 

[galsiah]

How fun would that be to play, though?

It's easy to do it badly, and hard to do it well, but if done well it'd take the player's focus of the stats and put it permanently on the character and on the game world. That's good if it can be made to work.

but the fact remains that the game is far, far removed from an accurate simulation of reality.

RPGs are not usually accurate simulations of reality, but they should be "accurate" models of some reality. It might take a while to learn about the game world, but once the player is familiar with it, it should usually feel intuitive (about as intuitive as reality anyway), and it should make sense. Getting rid of the stats makes it very important that this is done well, but as I say, that's a good thing in some ways - it prevents many bad stat systems early.

Character creation is a difficulty though. The player hasn't had time to get to know the world before he needs to make important decisions. I don't see a problem with making character creation a special case though. When the player is creating his character, he is setting up the game, not playing it. He is making decisions as a player, not as a character. Therefore, I don't see a problem with presenting stat screens etc. at character creation (not necessarily using numbers, but possibly). A question and answer character generation could be given as an option.

As such, my guesses may be entirely off-base

This should be very rare if things are well designed (hopefully). It's an issue that needs dealing with though. Can you think of examples?

Feedback is great... but by definition, it always occurs after the choice has been made.

True, but the feedback can usually be given early enough not to harm the character. For example with a sword or shield, the character should get feedback on how skilled he is as soon as he equips such items - not after he uses them in a fight. He can still suffer by pulling out a weapon he's never used, for the first time in the middle of a battle - but then that's his own stupid fault.

As I said above, giving feedback before the character makes the important decision might be difficult for some skills. It's a realistic goal though. Feedback just needs to occur as the character prepares to use the skill (as often as is possible) rather than after the skill is used. Most of the time feedback won't be too important in any case, since the player will usually have a fairly good idea of his character's capabilities. Where it is necessary, it needs to happen as early as possible.

 

[Norfleet]

In character creation you take a JA2 approach and answer questions about your character to be, but dont attribute stats to the guy.

Untrue. You answer some initial questions which determine what character traits he picks up, but then you go and attribute stats to him in the next part of the process.

 

[DarkSign]

You're correct about how JA2 does it, but I was coming up with a new stat-less method for our discussion here.