Wednesday, October 03, 2007
What Makes a Great RPG - Mechanics
We've talked about setting, story, characters, interactivity, immersion, putting the player into a role as aspects of a great computer RPG. But we're missing a key ingredient... the actual mechanics. Or "gameplay." What that means is that, when you boil it all down and you strip away all that groovy context and story and all those other cool things that make a great RPG, is it still interesting and fun to play?
If not, you've got mechanics that might cripple an otherwise stellar RPG. Arguably, more than any other element, the gameplay mechanics must stand on their own. However, I've got at least one counter-example in the "Examples" section that might indicate that mechanics might not be as important as I think.
So what makes good RPG mechanics? Much of that is a matter of taste. As much as the holders of the D20 license would like it to be otherwise, there's no "one, true" RPG system. In general, the desirable features include:
The system should provide "interesting decisions:" The system should be hard to min / max down to a single "winning strategy." The player's choice of actions shouldn't be obvious.
The system should provide a "reasonable" simulation of the genre: A gritty Cyberpunk adventure shouldn't have rules that allow the player to take on a dozen lesser opponents at a time through his martial skill, nor should a bronze-era Superheroic adventure deal with gory details of maiming and instant-death shots.
The system should provide rewards to the player at appropriate intervals: The definition of "appropriate" varies by flavor and genre, but in the game system should provide some sort of regular schedule of player improvements that are neither too long and frustrating nor too short and meaningless.
The system should be easy to learn, hard to master: This mantra is the holy grail of all gaming systems. But it remains true - the best game systems are ones that anybody can figure out and feel like they are playing competently within a few short minutes, but have the depth that allows hardcore players to still improve their performance even after investing dozens of hours into it.
The UI should be clear, easy to use, and provide useful feedback: The goal of all UI's.
The system should be balanced in risk versus reward: Measuring that balance is, of course, the real trick. This is a major factor leading to "interesting decisions," but is worthy of note by itself.
Some more contoversial elements of mechanics seem to be more of a matter of player or designer taste and appropriateness to the story:
- Random versus fixed loot / reward distribution
- Skill-based versus class-based systems
- Skill growth by deliberate advancement versus "natural" improvement by activity
- Turn-Based versus Real-Time
- Player Skill versus Character Skill balance of importance
- NPC parties versus Solo characters
Unfortunately, the nuts and bolts behind the mechanics of many RPGs are rarely very transparent. This isn't actually bad thing in general, but it makes talking about the details more difficult.
Early RPGs were frequently based on the original Dungeons & Dragons game. The core rules were very simple. Unfortunately, they also lacked depth. A later version of the rules, Advanced Dungeons & Dragons, was used in many officially licensed titles, particularly the so-called "Gold Box" games by SSI. While the rules themselves were a little deeper (and harder to understand by new players), the most interesting challenge came from the tactical combat game. Positioning of characters, hit locations of area-effect spells, and use of terrain features all came into play during combat. It was a hardcore RPG fan's dream. The otherwise straightforward (and, dare I say it, uninteresting) combat of the Dungeons & Dragons game came alive when returned to its natural element - a virtual "miniatures table."
Final Fantasy X used the series' traditional abstract positioning of characters in turn-based combat. There were no terrain features or movement (other than swapping members in and out) to worry about, but the game did offer some interesting rock-scissors-paper challenges for the player. The challenge came in identifying the strengths, weaknesses, and patterns of new monsters, and then adapting your strategy to fit. Sure, several monsters could be brute-forced to death without much in the way of tactics, but it would lead to a slower game with lots more time spent running back to a save node to heal up. Unfortunately, the twentieth time or so you encounter an identical grouping of monsters, the battle had become pretty mechanical and tedious. But as in all Final Fantasy games (that I have played, at least), there was a tremendous variety of monsters throughout the game to keep things interesting from chapter to chapter.
At some point, Richard "Lord British" Garriott opted to vastly simplify the rules system of the Ultima games. While computers could crunch numbers much better than a human player or DM, he opted for simplicity. In fact, it is rumored that the "Dexterity" attribute in Ultima VII didn't actually do anything - it remained in the game for legacy purposes. However, in spite of the lobotomized game mechanics in terms of actual character attributes, there was still an interesting interplay between spells and melee, and control of the party members (even in real-time combat). But while I consider it to have weaker mechanics than many of its predecessors, it remains my favorite. Does this mean I overrate the importance of solid gameplay mechanics? Maybe.
However, in direct counterpoint, the Virtue system of Ultima IV pretty much made that game. While the virtue system had no impact on combat, it had a huge impact on the non-combat parts of the game.
Most people don't mention the game mechanics much when describing their favorite RPG, but Fallout is an exception. Many players bring up the SPECIAL system. As a system, it never struck me as anything - pardon the pun - that special. But it was one of the more transparent systems for the players, spelling out in numerical format exactly the chances and percentages were. And you had that one trait that allowed you to always see the goriest death, which seemed very popular...
Vampire the Masquerade: Bloodlines inherited (and highly modified) then pen-and-paper system from its license. The "Humanity" and "Masquerade" attributes are very interesting additions to the game. Both offered soft restrictions to player behavior, much as the Virtue system of Ultima. As a blood-sucking monster in a game where it is always night-time, your potential actions are pretty unrestricted. However, if you decide to choose the path of monster, you'll find your humanity rating dropping very quickly. As humanity drops, so does your control over your character (the beast takes charge). Going into such a frenzied state is bad news where caution and careful tactical approaches is best. And it can also cause you to lose control in public, where you can be seen. This impacts your Masquerade score.
Vampires survive by making their presence secret, and pretending to be humans. Your efforts to maintain this pretense is represented by the Masquerade score. As it drops, Bad Things happen to you, particularly in the form of Vampire hunters - and will eventually result in losing the game. The game would be VERY different without Humanity and Masquerade, as simple and subtle as they they are.
What Players Say
"I’m a mechanics guy... I find bad mechanics interrupt my immersion the most. Examples of bad mechanics include instagibs, universal strategies and poorly scaling abilities." - Alexis, at Twenty-Sided
"I remember the first time I found the Cruisader's Ax in Wizardry 7, what a reward for exploration. In my opinion, constant exploration rewards water down the effect, while a handfull of well placed "killer finds" is much more preferable. NWN really burnt me out quick with the whole one or two treasure chests or desks per room. I wound up wasting all my time picking locks based on a stat roll. After I killed my second dragon in that game I called it quits, I was tired of the camera, and tired of locks and traps (overkill)." - drslinky1500, Rampant Games Forums
"Since most RPGs are characterized by character growth/advancement, you have to feel like your development is meaningful and worthwhile. If your character doesn’t acquire awesome powers, significant stat changes, or fantastic equipment, then there’s no reason to bother with such matters. Alternatively, if the only way to achieve such uber-pwnage is to level grind for 200 hours, then the sense of accomplishment is also somewhat diminished. Goals should be palpable but not require a commitment that forgoes real-life." - Hal, at Twenty-Sided
"Great games don't have repetitive elements and introduce lots of new ideas. Combat/Gameplay systems need to constantly introduce new elements throughout the game. Great games always are always dangling the next reward in front of you. This also means letting the player know where to go or how to get them (a steady supply of quests are good for this). Great games allow multiple approaches to a problem. Great games are the appropriate difficulty for the person playing them so that rewards aren't trivial but are still possible to obtain." - Ezin, Rampant Games Forums
"To make a game more real, I would much prefer a system that allows my character to learn anything no matter his trade or profession. This seems more real to me than an enforced set of skills with the inability to even dabble in anything else. In any adventure setting the ability to defend ones self to physical attack is common sense in my mind and I think anyone should be able to learn basic self defense, and how to use the more common weapons. The classic 1st lvl wizard out of D$D would lose a fight against a house cat 3 out of 4 times if they couldn’t use their magic. That doesn’t make sense to me." - Leopardeternal, at Twenty-Sided
"In all reality a player should not have to worry with saving the game constantly, they should be able to trust that the developers are not out to get them, and that a loss might loose them some ground in much the same way as Ultima 7, you are not going to loose your shirt. What I am getting at is that the game should help out the player so more focus is on the game, not on deciding whether or not to save." - drslinky1500, Rampant Games forums
"With the class/skill discussion, I think that is tied to solo/party. If your solo, you need to be a generalist, so a skill system allows you to survive. If your in a party, then each member can specialize to become more than the sum of the parts. Random loot is good, but needs to be sensible. Personally, I would randomize everything (plot included) and have the loot fall out of that, but thats just me." - Primogenitor, at Twenty-Sided
"Randomized worlds are very good. If you can’t have random scenery, then at least the loot should be randomized. Having the exact same items in the exact same locations every time though the game just kills the replay value... Skill-based leveling system. This Rampant Coyote post talks about skill-based vs. class-based systems. I’m a big fan of classless systems for sure. For me the ideal system was the one used in Fallout." - Shamus Young, Twenty-Sided
"With (Diablo 2) Blizzard created a almost perfect balance of giving players rewards and then make them want and struggle for more. It's one of the most addictive games if one likes to get and improve equipment." - Hajo, Rampant Games Forums
Links:
* What Makes a Great RPG?
* What Makes a Great RPG - The Answer?
* What Makes a Great RPG - The World
* What Makes a Great RPG - Playing a Role
* What Makes a Great RPG - The Story
* What Makes a Great RPG - Everything Else
* What Makes a Great RPG (Twenty-Sided)
* What Makes a Great RPG, Part II (Twenty-Sided)
* What Makes a Great RPG, Part III (Twenty-Sided)
Discussion on the Forum! Woot!
Labels: Game Design, Roleplaying Games
Comments:
Links to this post:
<< Home
Mechanics are definitely important for a CRPG. Most of the time they're just an uninspired knock-off of D&D. Instead of trying to create a system that fits the computer environment, they're just shoving a system into it and duct-taping over the cracks.
Tabletop RPG's and CRPG's are different environments, and what works in one won't necessarily work in the other. For example, the reagent system in Ultima is a great limiting factor in the game, and works in the narrow confines of the engine. In a tabletop RPG, you need a lot more variety; eight just isn't enough.
I think the most important thing to do, as a developer, is always try and fashion something that fits the context of your engine and your design well, and don't let "other" systems influence you, only inspire you. The game's the thing, not adhering to some fanciful standard of CRPG design.
Tabletop RPG's and CRPG's are different environments, and what works in one won't necessarily work in the other. For example, the reagent system in Ultima is a great limiting factor in the game, and works in the narrow confines of the engine. In a tabletop RPG, you need a lot more variety; eight just isn't enough.
I think the most important thing to do, as a developer, is always try and fashion something that fits the context of your engine and your design well, and don't let "other" systems influence you, only inspire you. The game's the thing, not adhering to some fanciful standard of CRPG design.
True. In fact, I think Shamus did a write-up about a year or so ago about why D20 is a lousy game system for CRPGs. He's not wrong (though I personally believe that CRPGs influenced part of the D20 system's design... they were planning for it making the jump again to computer games).
One reason people keep going back to those systems, though, is that it is understood by players. Or it used to be. Back in the day, if you threw "Strength, Intelligence, Wisdom, Dexterity, Constitution, Charisma" on the screen (or reasonable synonyms) with values between 3-18 on the screen and labels like "Fighter" or "Magic User," players immediately recognized it and were able to understand what they were doing during character creation.
Nowadays, however, these games are being targeted to people who have never once rolled a twenty-sided die around the kitchen table. So what might have been important and useful back then is pretty much useless today.
One reason people keep going back to those systems, though, is that it is understood by players. Or it used to be. Back in the day, if you threw "Strength, Intelligence, Wisdom, Dexterity, Constitution, Charisma" on the screen (or reasonable synonyms) with values between 3-18 on the screen and labels like "Fighter" or "Magic User," players immediately recognized it and were able to understand what they were doing during character creation.
Nowadays, however, these games are being targeted to people who have never once rolled a twenty-sided die around the kitchen table. So what might have been important and useful back then is pretty much useless today.
Character creation alone is a worthy topic but as you say: Nobody today wants to spend too much time in creating a character let alone four or six.
Most party CRPGs today let you discover your party members later in the game anyhow.
What I did in the "old games" was rolling the dice (using the random number generator button ;-) until the data was to my liking (= I maxed it out).
This isn't real gameplay at all - this is trickery or even cheating.
Why not give the player the best values at the start and let him try the game?
He can't blame the game (or the "dice") if he fails, then. ;-)
Funny example with Oblivion (again):
The character data was quickly customized but I've spent half an hour to design his looks...
take care
Most party CRPGs today let you discover your party members later in the game anyhow.
What I did in the "old games" was rolling the dice (using the random number generator button ;-) until the data was to my liking (= I maxed it out).
This isn't real gameplay at all - this is trickery or even cheating.
Why not give the player the best values at the start and let him try the game?
He can't blame the game (or the "dice") if he fails, then. ;-)
Funny example with Oblivion (again):
The character data was quickly customized but I've spent half an hour to design his looks...
take care
I did like the point-buy system of Champions / Hero system (back in the day when non-random character generation was a novelty). Frayed Knights assumes a certain distribution of values for human characters, too.
But there is something to exploring random generation and seeing how lucky you get. I do miss it sometimes.
But there is something to exploring random generation and seeing how lucky you get. I do miss it sometimes.
I was researching PnP RPG systems for my personal project, and I found almost all modern systems are point-buy and level-less. The "random dice throw until you get what you want" shtick is pretty well gone. Even DnD3 did away with it. Creation is still a long process, but it's more of a choice of what you want your character to be within the bounds of the system; everyone has the same investment of points, so all initial characters are as powerful though varied by the players' choices. This is as opposed to pushing the initial stats as far as possible and ending up with an initial character 3x as powerful as another with crappy rolls.
Additionally, almost no systems start you off as weak as DnD. Your character begins at what would be about level 5-10 in most point-buy's, and doesn't really get 20x as powerful through the course of his game-life. They improve significantly, but they don't become gods. Almost all CRPG's, regardless of the system they use, follow the DnD model of starting as a complete wimp and ending up as the single most powerful thing in the universe.
Additionally, almost no systems start you off as weak as DnD. Your character begins at what would be about level 5-10 in most point-buy's, and doesn't really get 20x as powerful through the course of his game-life. They improve significantly, but they don't become gods. Almost all CRPG's, regardless of the system they use, follow the DnD model of starting as a complete wimp and ending up as the single most powerful thing in the universe.
Just do everything Richard Garriott did. There is no discussion. The problem with modern RPGs is they are too far removed from the basis of their genre. They have become somewhat pointless.
Just do everything Richard Garriott did.
Hmmm... the "Super Avatar Brothers" gameplay of Ultima 8? ;)
almost all modern systems are point-buy and level-less.
True enough. What you lose with that, though, is the inherent "spikiness" of random generation. You end up with characters that are all "a little above average," with stats that are all between "slightly sub-average" to "good."
What I did in for my current D&D campaign (3.5) was randomly generate several sets of stats... I think I had 10 or 12 sets altogether... and had my players choose from any of those sets. Some were pretty average (10,10,12,12,14,16, that kinda thing), and some had both really good stats and really bad stats. I think one of the best sets... with the highest average... also had a 5 in one category. (And yes, someone DID choose that).
This was far more interesting to me. We have some characters that are generally "above average"), and then a few that have taken greater strengths in exchange for greater weakness.
That's obviously a feature of point-buy systems as well, but with wildly different weights assigned to values. A very low score greatly offset a similarly high score. This just seemed to be an easier presentation than forcing the players to deal with a complex mathematical system with scaling weights in either direction and all that junk. It also sidestepped (a little bit) some of the min-maxing efforts point-buy systems lend themselves to.
Post a Comment
Hmmm... the "Super Avatar Brothers" gameplay of Ultima 8? ;)
almost all modern systems are point-buy and level-less.
True enough. What you lose with that, though, is the inherent "spikiness" of random generation. You end up with characters that are all "a little above average," with stats that are all between "slightly sub-average" to "good."
What I did in for my current D&D campaign (3.5) was randomly generate several sets of stats... I think I had 10 or 12 sets altogether... and had my players choose from any of those sets. Some were pretty average (10,10,12,12,14,16, that kinda thing), and some had both really good stats and really bad stats. I think one of the best sets... with the highest average... also had a 5 in one category. (And yes, someone DID choose that).
This was far more interesting to me. We have some characters that are generally "above average"), and then a few that have taken greater strengths in exchange for greater weakness.
That's obviously a feature of point-buy systems as well, but with wildly different weights assigned to values. A very low score greatly offset a similarly high score. This just seemed to be an easier presentation than forcing the players to deal with a complex mathematical system with scaling weights in either direction and all that junk. It also sidestepped (a little bit) some of the min-maxing efforts point-buy systems lend themselves to.
Links to this post:
<< Home


