If I was playing a game where them was a complex combat system like that, how would I understand what was going on, how much damage I could do, whether I could beat the Ugly Ogre or if I needed to run away. If there are no reasonably simple stats I can look at to make these decisions, I either run scared from everything because I''m not sure if I can win, or I die every other battle because I''m getting in to fights I can''t win. I agree that the system doesn''t make a whole lot of sense sometimes but it is very playable. It is easier for me to see that I have more hitpoints and strength than him so I can probably win, rather than seeing he''s using a sword three inches longer than mine in his left hand and that he wife just broke up with him and he''s going on a killing spree. At some point, I need to see numbers or meters or something that allows me to form these decisions, otherwise, I''m not in control because I don''t understand what''s happening.

I not saying that I like the system or that I think there''s anything wrong with changing it or that it''s better. I''m just saying that I need to be able to understand what''s happening.

tj963

I love this. I''ve had 3 discussions about exactly this in the last 2 weeks.

I was talking to a friend of mine (actually lead programmer) who had just picked up NWN the previous week or so and was chatting with him about the game. His character was an 18th level monk. I sat there thinking about it. I''ve played D&D since 1981 or somewhere thereabouts and never have I gained 17 levels in less than a years worth of playing. This made me think that something was inherently wrong with the game on computer, then it hit me.

D&D is balanced for a half dozen people sitting around a table spending half of their time bullshitting and getting about a level every month or so. How do you incorporate that into a computer where the DM doesn''t have to roll a single die or figure out what bonuses apply to a situation? How do you adjust for the fact that an entire portion of the group is ''inactive'' while waiting for the other portion to finish talking to the GM and working something out? You don''t.

75% (or more) of the role-playing aspect of these games is gone, which leaves you with what? Combat and a little plot intrigue. So now your game is 80% combat and 20% plot development instead of the 50/50 balance that so many games strive for. What happens because of this? The characters level WAY too fast, the minimized plot is finished in a week or two instead of a year and there isn''t the real appreciation for the characters that you get when you bust your butt for a year. So what if your character just died after playing for a week... it''s not like you spent a year building that character. You don''t ''know'' the character the way that you would if you had actually played him for all that time and cared intimately about this character that you had built from the ground up.

A friend of mine who is in RPGA went to GenCon a number of years ago and played in a grandmaster game. The scenario was for level 18-25 characters, he brought his thief that he had played for 4 years to the game, while almost everybody else that was there actually sat there and made their characters on the spot. The game was supposed to be cooperative and eventually the GM changed it to competitive because the thief was so much better played than the other characters, he knew his character inside and out, knew when to run and when to fight.

I think that the biggest problem with these games nowadays isn''t that the rules sets don''t fit the computer and that they should be more complex, it''s that the players don''t have to really bust their butts to keep their characters and have to learn them inside and out. You pull that character outside of the computer and play him in a PnP game with people who have played their characters PnP the entire time and your character will get wiped out every time.

This is the biggest loss that RPGs suffer nowadays, they don''t force you to actually know your character. That''s one reason that I like Everquest to some extent, you really have to know your character and how to play the class. Even a player with a level 50 warrior can''t get a level 50 Enchanter and have a prayer of playing it correctly, they''ll get wiped out and vice versa. People complain about ''exp sinks'' in the game all the time and I have to say that it probably helps the game and the players.

This is something that could probably be used in NWN, exp sinks to force the players to work harder to gain their levels. Force the players to do things in different ways to learn. Hopefully more games will include the ''exp sink'' in the future. It''s not about suckering them into playing longer, it''s about suckering them into knowing how to actually play, whether they like it or not.

a bit ''off-topic'' but anyway...

quote: I''ve seen worse, I''ve seen games with both a "hit" check and a "damage" check, where if you fail either, you don''t hurt the other guy at all.

I don''t see any problems with that Everything is a question of balance and need to be tested over and again.
It seems normal that if your agi is far higher than your opponent agi, he won''t be able to touch you so often because of your high speed (and his limited speed due to a heavy armor for example).

But in some games (like Morrowind) it''s not really good implemented. You see your axe bashing the enemy, and most of the time nothing happens if you''re not at a high level. It might look odd if not handled correctly.

Since my name was mentioned, I guess I should add my 2 euro-cents to the simmering fire...

I've played various forms of D+D and AD+D in pen and paper form for some time. I've also played various computer games based on their rules, from the original Pools of Radiance to Baldur's Gate. (I've not done any more than read the manual to Neverwinter Nights, however.) I've also played pen and paper RPGs that had far better rules than AD+D, and CRPGs with better rule systems than AD+D.

Dungeons and Dragons was originally based on a set of wargame rules called Chainmail. It has kept most of its wargame roots - numerous look-up tables, formulae, arbitrary numbers, quick deaths for 'average' soldiers, and so on. Playing as individual heroes was tacked on as an afterthought. And the notion of 'roleplaying' in the way that most of you think of it was also tacked on as an afterthought. Through being the first roleplaying game, and subsequently the most popular and best-supported, it is the system that most people know of and are comfortable with. But it was built upon a rules system that was never designed for 'roleplaying', even in the loose Diablo sense. Just as many of MS Windows' problems over the last 5 years have stemmed from backwards compatibility with DOS and Windows 3.1, Dungeons And Dragons has awkward, unintuitive, and counterproductive rules as a result of its wargame heritage.

Now I should make myself clear - I am not against RPGs with levels, or character classes, or experience points, or even the "kill to progress" game mechanic. What I am complaining about, is that the D+D rules are a horrible way of going about it.

One example, which is one that MadKeithV has repeated for me, is the whole 'hit points' concept. A level 1 fighter will have somewhere between 5 and 10 hit points in most AD+D games. (They have 'd10 hit die', which means they get, on average, 5 or 6 hit points per level.) Most weapons deal 3 to 4 points of damage in a hit. This means that most level 1 fighters will die if they sustain 2 hits in a fight - not uncommon. When you're sat around the table, the Dungeon Master who makes the roll to hit the player may pretend to have missed, in order to spare the fighter. In Baldur's Gate, there is no benevolent Dungeon Master interested in the flow of the game, so you end up reloading every half hour or so. Despite being a game about heroes, the AD+D hit point system translates appallingly to computer form at 1st level.

Of course, when you hit 2nd level, your capacity for taking damage has doubled, and your chance of dealing damage has risen too, not to mention any additional skills or abilities. So you are now at something like 250% of your previous fighting capacity. Most of the encounters that were too hard at level 1 are now a walkover as you pass the magical lv1 -> lv2 mark. Again, this is very evident in Baldur's Gate. This doesn't lend itself well to game balance as you have to either have very distinct Level 1 and Level 2 zones, which means that the player loses access to a lot of content, or you need to dynamically tweak the encounters, which feels artificial as someone already stated. Worse of all, subsequent levels do not confer the same kind of advantage for mathematical reasons: level 3 is something like 170% of level 2, level 4 is something like 140% of level 5, and so on - the relative benefits of each level slow down, while the amount of experience needed each time goes exponentially up. The system's progress curve starts off too sleep and levels out too quickly to make for good game balance or use of content.

And there's more wrong with hit points - in the pen and paper form they are a necessary abstraction of physical health, ability to dodge and deflect blows, the ability to ignore pain, and so on. Keeping track of all those on paper would be a nightmare (although it hasn't stopped AD+D trying to bolt those sorts of features on afterwards to placate the 'realism' camp). But in a computer game, we don't need to abstract numerous stats down into one stat. The computer is more than capable of handling them all with a few trivial calculations. You can have separate scores for Dodging, Parrying, Ignoring Pain, Absorbing Damage, and everything else. If you want your players to see all these stats, you can show them. If you want a simplified interface, just perform the relevant mathematics that will get used in the attack roll and show that score as a 'Defence' stat or whatever. If you don't want to show any stats, as a lot of you keep saying, then show none - just hint in the manual that your game responds to a lot of real factors and name a few so that the players take them into account. Either way, the principle is clear - you can make the interface as abstract or as detailed as you like, but there's no reason for the system itself to be so abstract it's a poor representation.

The same goes for Armor Class, THAC0, and other AD+D statistics... the system they use is an oversimplification, good for pen and paper, poor for computer games. Why does armour affect how often you are hit, rather than how much damage you take each time? Simple - because keeping track of individual damage levels in a wargame would be very fiddly, so they didn't want that. You either got hit and died, or got missed and lived. It just happened that as AD+D evolved and they started keeping track of the individual health levels for heroes, it would have been too much work to change the armour rules. So we're stuck with them. But they're a poor abstraction, and we shouldn't keep them. The weight and quality of the armour can affect the chance of dodging, and the type and quality of the armour can affect the amount of damage reduction per hit. This is trivial to do on a computer, and the player doesn't need to know the details. It would enhance the game by offering tangible results to the player that match their intuitive expectations of what would happen.

That's not all, of course. The rest of the rules are not exactly simple or intuitive either. AD+D has numerous different resolution mechanics that it uses: attack rolls, saving throws, skill rolls, and so on. In 3rd Edition these have been standardised somewhat, but in the past you had a system where some stats were better when they were lower, some stats were better when they were higher, and so on. All in all, a mass of confusion just to make the dice rolls easier to remember. All this is pointless in a computer game - bigger should always be better, because that is most intuitive, and handle the awkwardness of subtracting one thing from the other behind the scenes.

The class system is horrible too - rather than settling on a sensible system where each character plays a certain class, they have 4 basic classes, each of which has some derived classes, and those classes can be mixed using either a dual-class system (where you learn 1 of 2 classes at a time) or a multi-class system (where you learn 2 or more simultaneously)... yet more added complexity for very little reason. Plus they tend to unbalance the game by being hard to compare against other characters of similar level. If you want character classes, pick a few that highlight memorable archetypes for your game, and stick to them. If you want classless, pick skills, abilities or attributes that allow one character to differ from another and develop those. But don't pick a game that was designed to use classes and had multiple-classing tacked on as an awkward afterthought.

Ok, I could go on criticising the mechanics as applied to computer roleplaying games all day, but you get the point. Instead I'll address a few points made by the rest of you:

Sandman : indeed, the players are familiar with the rules, but the point is that the rules should not need to be complex enough that prior familiarity is a significant benefit. Look at Civilisation: the chance of hitting is Attack divided by Attack + Defence. One simple combat rule, yet with all the potential modifiers for terrain, morale, fortification and so on, there's an ocean of depth there, none of which needs spelling out in explicit rules.

As for the ruleset being playtested and balanced... I don't think it is balanced at all, although I obviously cannot argue about the playtested part. However, I will stress that it has been extensively playtested in a different domain where the game is played very differently - ie. with a human dungeon master's discetion mainly.

Machaira said: "Yes, but most players like to see the numbers. Also, if you make it "make sense" your characters wouldn't last through the first battle. There's a tradeoff between making sense and being entertaining. Having to roll up another character every 10 minutes would not be very entertaining."
The thing is, in most AD+D computer games, you usually do have to roll up another character every 10 minutes until you hit level 2 or get a load of people in your party. And whether you hide the numbers or not is not really the point, as you can hide or expose as much or as little as you want of a complex system. But you can't make a simple and overly abstract system fit your needs as easily.

Silvermyst : you ask a philosophical question that I don't think can be answered here. It's true that some people like hack-and-slash RPGs, whereas others want to write something a bit more cerebral. But no matter which type you want to write, statistics will come into it if you are writing what most people consider 'role-playing' to be about. Perhaps what is being overlooked is that a good statistical system extends far beyond just combat. The system I use for my game uses the same mechanics for combat, stealth skills, political skills, and anything else you want resolved according to stats. In AD+D, the non-fight elements are tacked on as an afterthought, which is perhaps why a lot of people make the assumption that statistical resolution is unsatisfying for non-combat activities in these games.

tj963 : as I pointed out above: the system itself could have 100 variables for working out whether you hit or miss, yet it could still present a nice and simple "Attack" score on your character sheet if the programmer wanted. On a computer game, the complexity of the system can be far higher than the complexity of the interface. The problem is that we're using a system designed for humans where the interface is the system and it therefore had to be kept very abstract. This limits the gameplay and balance possibilities.

Sorry for the long post.

[ MSVC Fixes | STL | SDL | Game AI | Sockets | C++ Faq Lite | Boost | Asking Questions | Organising code files ]

[edited by - Kylotan on July 24, 2002 9:47:54 PM]

Since stats are simply a generic, abstract representation of what happens in the game''s reality blow by blow, it stands to reason that (heh, given enough art resources) you can hide stats and express them solely through your character''s actions and movements (as Project Ego plans to do).

If you''re encumbered, you might dispense with numbers (305lbs/310lbs) and instead animate a character that is extremely slow and weighed down.

If you''re not very good with a sword, the arc that the sword swings is often wild, too high, and hits the dirt alot.

If your number of attacks are slow, this could be expressed as a delay between hit and ability to hit again.

If you''re severely damaged and close to death, your character might stumble and hold his stomach; or even bleed profusely.

If you have something like "opportunity attacks" which are a part of 3rd edition, you can represent this as your character or an NPC leaning and dropping their guard.

If you strike someone and do significantly more damage, they might spin around faster or fly back farther (or maybe the sound effect of metal clanging on metal is louder or deeper).

If your hitpoints are greater, your character might automatically block faster, or blocks might be much more successful.

The animated expression doesn''t actually have to affect or detract from gameplay. Just because your character stumbles doesn''t mean that he falls over cliffs or away from an enemy you''re trying to attack. The character can express stats and states without fundamentally affecting controls and gameplay.

Essentially, the computer gives you a chance to actually play out all the things that a GM describes as a result of rolls ("What, you rolled a 1? You throw your sword!")

You still have to handle the case of those players who want to see their numbers. You also need a reminder of capability for returning players who haven''t played in awhile.



--------------------
Just waiting for the mothership...

My problem with my character''s vital statistics being wholly manifested with numbers is that you end up with the excessive micromanger present in computer D&D games. Unlike most everyone else who has posted thus far, I myself haven''t played much pen&paper D&D, so I don''t know the ins and outs of the system as well as those who have. Every level brings with it an assult of new numbers to figure in with the old so that I can determine my character''s ability not to get its ass handed to me. On top of this, I only get 20 or so "magic level-ups", so each level is exceedingly crucial. In summery, the story and the game''s fun usually end up being lost in a sea of "1d6 + 1/level of caster" numbers.

Morrowind at least took a step (albeit a wobbly one) in what I consider the right direction. I enjoy the fact that you slowly gain power by using and practising your skills. It''s majors problems are, however, that it''s unbalanced (to say the least) and the lack of evidence of progress. You practise and practice, but about all that changes are the proverbial "behind-the-scenes" numbers. That''s not to say that all of its skills are that way, because there are a couple that do manifest themselves over time.

The only other aspect of gameplay I really want to mention is health (as everyone else has). It seems to me that as someone is damaged, their body usually begins to show signs of wear and tear. What happened to the days of Wolfenstein 3D when you at least got a picture that gradually became more bloody and gruesome? My character shouldn''t just go from top physical condition to dead when they lose that last hit point. I''ll end with this: I will get down on my hands and knees and bow to the first game that allows me server every limb on my enemy so that I can hear him scream that he''ll bite my bloody head off.

Wavinator I had a discussion like this elsewhere last week and my position was pretty much the same as yours here. I salute you. And this is the first I''ve heard of Project Ego, just looked it up and I''m fascinated. Thanks.

I''m rehashing Wavinator''s post a bit but here''s my take on the whole stats thing. We roll dice in PnP RPGs because they''re an easy way of simulating chance, not because there''s anything special about them. But in CRPGs computers can simulate chance by generating random numbers for us. Voila, no dice rolls required. It seems logical to me to extend this idea into other areas.

I agree that stats are just an abstraction of what happens in the game reality but they also serve another significant purpose. They give you a way to judge your character''s ability to accomplish a task without attempting it. This helps prevent you, for example, from constantly trying to take on far more powerful opponents. So you need not only an alternate representation for the outcome of a character''s actions, you need to make sure that the player can make an educated guess about their chances at succeeding in some task. Playing experience is part of this, but not all. In real life I don''t have to try and kill someone to see if I''m strong. I have other means - I can look at my muscles and see how big they are (or hehe I can try lifting something heavy - didn''t someone here post about lifting weights in an RPG to test your strength?).

quote: Original post by Sandman
I don't know if you meant it like that but the way you describe it still sounds somewhat combat-centric.

Yes, it is, I guess. Mostly because the discussions I've had have always stemmed from the hopeless combat rules and resolutions in CRPGS. As Kylotan said, in pen-and-paper, there's a reason for simple rules, and I don't mind, but when I see those simple awkward rules badly translated onto a computer just for the sake of more-or-less sticking to those rules, I get a shiver running down my spine.

[edit: I just realised that my rant below doesn't really apply much to what I've said above, but I'll let it stand I'm completely scatterbrained today ]

Yet, there's no reason that good solid basic rules couldn't be translated into other sections of the game. I'll take Vampire - The Masquerade's abstraction as an example (because I happen to like it an awful lot - leaning more to the story side than the combat side).

Firstly, there are three basic categories of abilities and skills: physical, mental and social. These are further subdivided in the game, but for challenge purposes in a computer game, lets just stick to these three. Hence, you can be faced with three kinds of challenges: a physical one, a mental one and a social one. These could also be the main "abstracted" stats that the player can see all the time. Basically, let the player know how good his character is at each of these three, generally. They also show the character's general knowledge in these areas, but no specifics - giving the "if nothing else applies" score on a challenge - i.e. if you find a challenge for which you have no particular skills.

Now, I've mentioned skills, so they must be in there somewhere right? Well, yes. Again, skills are divided into the same three broad categories. Seduction is a social skill, brawling is a physical skill, electronic engineering is a mental skill. Now, this is where it starts getting really interesting on a PC. For instance - Seduction is a social skill, but I've mentioned that social may be subdivided into sub-abilities. Lets take subvisisions such as physical attractiveness, leadership and charisma. Instead of saying "seduction is an attractiveness skill", lets say instead "seduction can be attempted with any social skill". If you seduce with attractiveness, your character struts around the person you're trying to seduce, showing off his or her "assets". If you seduce with charisma, you start a suave conversation bowling the other over with your witty remarks and romantic poetry. If you seduce with leadership, you radiate an aura of power and ability intentionally, perhaps flashing some gold or diamond or green here and there...
Basically, the underlying roll doesn't change from ability+skill, but there's a lot of depth there already, and a lot of potential for characters that might be weak in one area to still succeed by being better in another area. Plus, certain opponents could be especially susceptible to one particular field, for instance, a "golddigger" girl might be very impressed by your leadership seduction, but couldn't care less about your charm and wit.


It seems to me that there could be a lot of potential in there, without overcomplicating things.

[edited by - MadKeithV on July 25, 2002 3:36:59 AM]

I agree with some of what has been said, but strongly disagree with some statements.

First, without going off on a rant, let me say that the entire concept of NWN is immensely stupid. It is foolish in that it attempts to be way overly literal in the translation from D&D to PC. Imagine playing D&D over a conference call - pretty sucky! Now imagine even worse and you have NWN. Rather than realize that PC gaming is actually different from D&D gaming and adjusting accordingly, they just totally ignored what we like to call "reality." For example, you can only have one character in the single player campaign. Yes, you can only have one character in D&D...how often do you play single-player D&D? Have you ever played an extended D&D campaign where it was just you and the DM...BORING! Rather than try to capture the FEEL and spirit of pen & paper gaming, they chose to capture the literal rules. Dumb.

Anyway, that said, some of what you people are suggesting is way off. For example the idea that we don''t need all these abstractions, we could have tons of different stats like "Pain Tolerance" and what have you. There is an obvious danger in this, and if you need proof look at the Moo3 disaster. Gamers need to know WHY things are happening. If the player swings at a goblin similar to one they have killed 100 previous times and they can''t kill it, they will be confused. Even if you have some good reason like their fatigue is too high, or some nagging injury is hurting them. Moo3 suffered from that problem, where things would happen but the cause was somewhat unknown to the player.

Once again, you cannot be overly literal. If in real life I am tired I KNOW I am tired. If some nagging injury is holding me back I can feel the pain and dizzyness. In a game how do I know that my dizziness is now less than kDIZZY_LEVEL or that isCritical(bloodLoss)==true. Unless you can SHOW me I don''t. Unless there are little dizzie birdies flying around my head or my character is flopping around all over the place I have no idea what is going on. Even then, WHY did I get dizzy? Which injury is the one that lost so much blood?

I am not saying that the players need to understand every rule in the game, but there should be a pretty clear cause and effect. Let me give you a good example - in Street Fighter Alpha 3 you can get dizzied at very odd times. You get hit a whole bunch and don''t dizzy, wait a few seconds without getting hit by anything, then get hit by a low damage fireball and you are dizzy - huh? It isn''t random, there are rules governing it. But the rules are too complex for me to understand even vaguely or to predict them. Whereas in old SF games the rule was "get your ass kicked a lot over a short period of time and you dizzy."

So, if the damage you do or your chance to hit is a combination of your current fear level, fatigue, confidence, weapon skill, etc etc etc, how do you communicate that to the player? Or do they just throw their hands up in the air and think "who the hell knows why I can''t hit this thing"?

Furthermore, I would point out from a practical standpoint the more complicated you make the rules the harder it is to debug and make sure they actually work the way you intended.

I don''t believe that "complexifying" rules adds much to a game, and if you look at what games are popular I think they illustrate that. Depth of rules and breadth of rules are not the same thing. Abstractions are good.

Using the D&D rules is dumb, but not because the D&D rules are not complex enough. D&D rules are made for D&D, not for PC games. DMs have discretion in D&D, your PC doesn''t. In PC games things happen at a different pace, certain things become more or less obvious, players fall into different patterns, etc.

If you are creating an RPG the FIRST question is what do I want my RPG to feel like and play like? Base your rules on that. Is the RPG a horror style RPG where any encounter can mean death? Is it a power-levelling game? Is it based on swords, or guns, or vehicles? Choose your game first, THEN your rules.

quote: Original post by AnonPoster
Using the D&D rules is dumb, but not because the D&D rules are not complex enough. D&D rules are made for D&D, not for PC games. DMs have discretion in D&D, your PC doesn''t. In PC games things happen at a different pace, certain things become more or less obvious, players fall into different patterns, etc.

Just take this observation:
the, currently, most advanced implementation of the D&D rules on the master-of-all-calculations, the Personal Computer, OMITS a lot of the rules that D&D has.

What does this say about D&D? I think it says a lot. You can argue all day about how complex rules don''t make things better (and you''d be right!), but while NWN is complex, it''s LESS complex than the paper version, and still doesn''t really work.
So you have a system, that''s probably not really a shining example of simple rules. Then it''s translated to a PC. Then they decide to strip stuff ''cause it doesn''t make sense. It just SCREAMS "design mistake". Start from what the PC is good at, and work from there. And don''t make the mistake of "we need this or that stat to make it realistic", because realistic is not what RPGs are after. Otherwise we wouldn''t have elves and goblins up the wazoo anyway

BUT - if you have an easy-to-understand rule, such as "as you progressively study harder, you progressively get less benefit from it", and apply a floating-point inverse logarithmic function to it - the PC doesn''t have a problem. It''s still a really simple rule. Yeah, the math is complex, but the result is easy (still just a single number). This is how it can be leveraged - gradual advancement, a granularity that is a LOT finer than what pen-and-paper offer, a sense of continuity, and also, not abstracting things that should not be abstracted, just because they are "too hard to calculate".