View Tags

July 5, 2011

The battle system I wish RPGs would stop using

Dungeons and Dragons, son of Chainmail, is the great granddaddy of the modern role-playing game. Its importance cannot be overstated: not only was it revolutionary for its time, it has also directly inspired many of the early computer RPGs on which the genre is now based. Because of D&D’s power as a brand, early edition D&D rules were imported into a variety of classic computer RPGs, forming the backbone of their combat systems.

So you might be surprised to know that, as a game designer, I get heartburn any time I see a computer game deliberately aping the D&D combat rules. In spite of D&D’s intimidating legacy and marketing power, its rules simply aren’t well-suited for use in a computer RPG. Those rules (and the numerous knock-offs they’ve inspired) have been implemented sloppily in RPG after RPG, reducing what should be enjoyable tactical combat into an opaque and frustrating slog.

Rather than address the issues this has caused, the older generation of western RPG developers has instead collectively jumped ship from the U.S.S. Turn-Based. Accordingly, this article is mostly aimed at indie RPG developers who may feel tempted to build combat systems modeled on the D&D-alikes that came before. Here are the reasons why we shouldn’t.

Reason 1: The D&D combat system is sloppy.

Back in high school, I developed a pen-and-paper alternative to Advanced Dungeons and Dragons Second Edition, which I chose to call (rather uncreatively) “Epic RPG.” The primary design idea behind the Epic RPG system was to streamline combat and focus on role-playing. It dispensed with randomized damage rolls. It ditched saving throws. The few rolls that remained were made uniform: all of them were performed on a 100-point percentage scale using two 10-sided dice.

The system also employed a large stable of skills and spells designed to open up the player’s options for approaching different problems through role-playing. There were no spells that dealt direct damage: no magic missiles, no fireballs. Rather, the spells all had simple, descriptive effects that required players to combine them creatively in order to take out enemies. Gravity Well, for example, was a magical trap that would suck enemies flat onto the ground and hold them there for a certain period. Rubberfoot gave the caster two consecutive bounces where he could soar up to 60 feet into the air, then land with impunity. Crumble would cause a stone surface to break apart into chunks with a touch.

You can imagine how a clever player might prepare for an encounter with a large group of enemies in a cavern. She might set up a Gravity Well trap, then cast Rubberfoot, and once the enemy triggered the gravity well, leap up to the ceiling of the cave and Crumble a portion of the ceiling down onto them, crushing them.

After years of playing D&D, my friends and  started using this system. We quickly found our battles involving less time rolling dice and more time role-playing. I bring this up to make the point that a good RPG combat system should be elegant. It should establish relatively few, simple rules that can interact in interesting ways, thus creating a vast possibility space.

Think of chess: there are six different types of pieces (pawn, knight, bishop, rook, queen and king), each with its own movement rule. There is a rule stating that the white player moves first. There is a rule stating that you can move one piece per turn, once. There is a rule establishing a forced move (check) and one establishing a win condition (checkmate). There are a small handful of other, minor rules as well, but for the most part, that’s it. Out of that small set of rules, you get a highly tactical combat system that exemplifies the old design cliché “simple to learn, hard to master.” That, friends, is what emergent complexity is all about.

Unfortunately, D&D combat utilizes something like the exact opposite of emergent complexity: it complicates even the simplest of actions, bogging combat down with numerous rules that apply only in rare, select instances. Just look at the size of the D&D rulebook: the Player’s Handbook was more than 125 pages thick in its first incarnation, more than 250 pages in its second and more than 300 in its third. The Fourth Edition Player’s Handbook is the largest yet, weighing in at 320 pages; its companion, the Dungeon Master’s Guide, currently contains an additional 224 pages on top of that.

image-890

“But wait: what do I roll if I duct-tape a bec de corbin onto my long sword?”

Not all of those pages are devoted to rules, of course, but a significant portion of them are. Every single weapon has a damage range with its own associated die rolls, and yet there are numerous different weapons and spells which essentially duplicate each other. There are five different saving throws for different types of effects (three if you’re using the latest rules), which are themselves entirely separate from rolls against your character’s six ability scores. Every single successful attack requires at least two die rolls. You need a bare minimum of six different kinds of dice just to be able to play: a four-sided die, a six-sided die, an eight-sided die, a ten-sided die, a twelve-sided die and a twenty-sided die.

There is a word for this: bloat. From a design perspective, D&D’s combat system is an overgrown mess. This leads us directly to our second reason for giving the D&D combat system a pass…

Reason 2: The D&D combat system is inscrutable.

In order for a tactical combat system to work, the player has to be able to figure out the likely results of his or her potential actions. Unfortunately, the D&D ruleset is horrible at this.

To determine a character’s percentage chance of hitting a target, the player has to find the target’s Armor Class, find the attacker’s Thac0, subtract the target’s Armor Class from the attacker’s Thac0, then multiply the result by 5.

image-891

“Wh…what’s a Thac0?”

More recent editions of D&D have changed this mechanic in ways that, if anything, make discovering one’s percentage chance to hit even more labor-intensive:

To resolve an action in the d20 System, a player rolls a 20-sided die and adds modifiers based on the natural aptitude of the character (defined by six abilities, Strength, Dexterity, Constitution, Intelligence, Wisdom, and Charisma) and how skilled the character is in various fields (such as in combat), as well as other, situational modifiers.[3] If the result is greater than or equal to a target number (called a Difficulty Class or DC) then the action succeeds.

Very few RPGs using an D&D-style ruleset gather all the necessary stats, perform the calculations behind the scenes, then just spell out the resulting percentage chance of hitting for the player. (Eschalon is a happy exception to this rule.)

Even those that do spell out a player’s chances to hit, however, essentially only tell the player whether her character is going to do no damage. Under an D&D-style ruleset, once the attack hits, the damage actually dealt can still vary from essentially nothing (1 point) to as much as twelve times that amount, depending on the weapon used! Each level of damage within that range has an equal probability of occurring; thus, unless she is attacking something with very low hit points, the player has no reasonable way to guess what will happen even if her attack connects. This makes planning out a strategy for even a single round of combat difficult.

Compare the situation above with the way Fire Emblem handles information about a potential attack: you select a character, then select an opponent. The game then immediately and clearly spells out for you a) your chance to hit, b) the damage you will deal, c) your chances of landing a critical hit, d) the number of blows you will be able to get in, and e) all of those same stats for the enemy vis-a-vis the attacker. That, ladies and gentlemen, is how you give a player clear information about her tactical options.

Reason 3: Results are too randomized under the D&D combat system.

Before computers, game developers needed ways to model unpredictability. Unfortunately, spin wheels are cumbersome to transport and rely on physics so simple that players may learn to reliably manipulate the results with skill and practice. Playing cards are only as random as the shuffler makes them, and are even more labor-intensive to produce and use than spin wheels. What was a game developer to do?

image-892

The spin wheel–probably not a great way to resolve combat.

Dice were the answer. Dice are fast (compare a die roll with shuffling a deck of cards), portable, and reliably generate randomized results. In short: dice are the original random number generators, and as far as random number generators reliant on real-world objects go, they’re probably the best. Hence, the popularity of dice in non-computer games.

There are big advantages to being able to generate numbers at random in a game. The biggest is that it introduces an extra element of risk to player choices. Risk is exciting. As the Wired biography on Gary Gygax put it:

Gygax was fascinated by the way the rolling of dice affected — and enlivened — the game experience. “Random chance plays a huge part in everybody’s life,” he says. He learned this first hand in his job as an insurance underwriter, which was a game in itself. His work involved evaluating policies and calculating how much to charge in premiums based on salary, age, medical reports, and the potential for long-term disabilities. He did special risk underwriting as well, such as evaluating the payout for a Major League Baseball team that wanted to take out a policy on one of its players. “I wasn’t popular in the home office because I wasn’t chicken,” he says. “I’m just a risk taker. I have gut instincts.”

That said, there are also big disadvantages to using randomized numbers in a game. Every time your game requires the player to roll dice, you take control–and therefore, responsibility–away from the player. You open the door of “that wasn’t my fault, it was bad luck.”

image-893

That’s a bad door to open. Some degree of player control is essential to skilled play of a game. If your game’s battle system ventures too far into Randomized territory, the player is going to feel like she doesn’t have control over the outcome—and she’ll be right.

Reason 4: Mitigating factors in real-life D&D do not translate to computers

For the most part, D&D gets away with relying so heavily on die rolls because it is a role-playing game run by human beings. Players have the flexibility to improvise tactics during a play session—and, just as importantly, the ability to nag the Dungeon Master to fudge the dice in the name of a fun play experience.

If you are designing a computer game, there is no Dungeon Master to fudge the rules for you. There is only a program that is going to execute every line of code you enter with exacting literalness. You do not have the luxury of designing an arbitrary or unfair combat system. Any factors that are going to tilt combat in the players’ favor have to be coded into the game itself.

Unfortunately, most games that use an D&D-style combat system fail to give the player enough tactical options to manage the risks imposed by a highly randomized combat environment. They adopt the Thac0, the randomized damage and the saving throws, but then fail utterly to give the player more than a small handful of real tactical options. The player’s only friends become superior stats and superior loot.

This is poor design. Giving the player better combat numbers merely amounts to weighing the dice in the player’s favor. Weighing the dice does not turn a game of luck into a game of skill. In a game with randomized outcomes in combat, you have to give the player some robust tactical tools to manage her risk, or else she might as well be playing this.

Some game designers have done a better job of introducing outside tactical considerations than others. Compare, just for a moment, two popular computer RPGs with D&D-based combat systems: Baldur’s Gate and Pool of Radiance.

Pool of Radiance implements a wide variety of tactical considerations which allow the player to mitigate the risks inherent to a highly random, dice-based battle system: distance, directional facing, different attacks (fighters can sweep groups of adjacent enemies), and large numbers of characters under your control with different strengths and abilities.

Compared to Pool of Radiance, combat in Baldur’s Gate is an unpredictable mess. The tactics you can use are highly limited due to 1) the small number of characters under your control; 2) the fact that they each have only a single attack (not counting spells); 3) the awkward non-grid-based movement system; 4) the fact that enemies close distance with you almost instantaneously; and 5) the fact that you do not have direct control over your characters.

B-but…but…

“But Craig,” you protest, “Baldur’s Gate 2 and Planescape Torment are widely considered to be two of the greatest RPGs ever created, and they both used the Baldur’s Gate implementation of the D&D combat system!” This is true. However, these games are classics because of their engaging characters, strong writing, and player freedom.

They are not classics because of their combat. So far as I am aware, there is not a single person on this planet who has played Planescape Torment, then decided that it was the best RPG ever made because he just loved using D&D’s clunky combat mechanics to determine the outcome of enemy encounters.

The Takeaway

The D&D combat system barely functions in its original, intended medium; recreating it in a computer game is like creating a calculator program that lovingly renders a pixelated abacus and lets you click to slide the beads around. You might as well just code something new that isn’t tedious and frustrating to use.

In my next post, I’ll talk about various ways to create a genuinely satisfying tactical combat system. Until then, for the love of all things holy: please leave the D&D combat system alone.

UPDATE: Part two has now been posted.

  • Tuco

    That’s crap, considering how even with all its flaws, the D&D battle system easily tops most of the crappy battle systems seen in RPGs today. Baldur’s Gate 2 and Temple of Elemental Evil send their best regards.nnOf course, i think it could be easily bested too… But somehow the competitors always come out with worst solutions, instead of best ones.nBut please, prove me wrong… Point me computer RPGs with better battle systems.

  • Guest

    Oh, no, no, no… it’s /not/ crap.u00a0 Temple of Elemental Evil’s frankly horrid combat mechanics are even a major blow to your rebuttal.u00a0 Moreover, both of your examples are from the era that still used THAC0, a system that I still feel is superior to what followed.u00a0 Take Neverwinter Nights as a more recent example, and you’ll notice such affronts to common sense as the rogue who can easily walk through a flaming corridor with no damage because it involved reflex saves, and the paladin whose massive defensive focus renders him immune to anything but a natural 20 from that dragon’s claws and bite attacks.u00a0 NWN was most definitely not a hit because of its combat system.u00a0 Heck, I killed Moraug without taking any damage when I went through with my rogue; what sense does that make?u00a0 nnYou can look at KotOR for another smashing example of the old Chainmail battle system making the very setting melt down; since when is Han Solo a total idiot for preferring a blaster to a rusty dagger?u00a0 Since the hamfisted attempt at equalizing ranged and melee turned all ranged attacks, without exception, into an idiot’s last resort.u00a0 Even Mission Vao, dexterous and weak-limbed Twi’lek, would actually do far better damage per round if I forced her to use a ceremonial sword instead of her preferred blaster rifle.nnThere are games with very different battle systems, like Golden Sun (referenced in the article) or, say, Battle for Wesnoth.u00a0 Whether they’re better or not is a matter of taste, so proving you wrong is arguably impossible.

  • http://notimetoplay.org/ Felix Pleu0219oianu

    Funny, that sounds not unlike my recent rant against roguelikes (shameless plug: http://notimetoplay.org/2011/04/04/my-love-hate-relationship-with-roguelikes/ ). Now I understand where the problem with those likely stems from.nnAlso, +1 for Wesnoth. It has a completely deterministic combat system — and very simple rules to boot — yet impressive tactical depth. Well worth studying, along with theu00a0 document explaining their rationale.

  • Green Eagle

    Reason number 5:nnRPG’s are all about suspension of disbelief.u00a0 The forcible interaction with the details of what is going on behind the curtain in D&D games is almost guaranteed to fracture any feeling of reality.u00a0 I agree with your comments, but to me this is the worst feature of D&D mechanics, and it has bothered me for years.

  • Opus

    To me, the description of Fire Emblem sounds like it’s removing a big chunk of “adventure” and replacing it with “statistics”.u00a0u00a0 When you round a corner and encounter something you’ve never seen before, you should be in the dark about how to bring down this new beast.u00a0 A good game will give clues that you’re fireball isn’t going to harm the thing, or that arrows won’t bother it (or whatever), but it should be up to you to work out a good strategy, gaining experience and knowledge of the world as you do, not just picking the option with the highest percentage written beside it.

  • Anonymous

    Fire Emblem isn’t really of the “encounter something unknown, find the thing it’s weak to through trial and error, then spam that one attack until it dies” school. Even though it’s up-front with the likely results of any one attack, I think you’ll find that the Fire Emblem games still require a huge amount of strategy. (My personal recommendation: check out Path of Radiance.)

  • Tuco

    “Temple of Elemental Evil’s frankly horrid combat mechanics are even a major blow to your rebuttal.”nnYeah, sorry, I can’t even try to take you seriously after this.u00a0Try to repeat it on every hardcore RPG focused forum, like RPG Codex or No Mutants Allowed, and enjoy the laugh.nAnd both NWN 1 and 2 were shit not because of the ruleset but cause of its cheap implementation.n

  • Tuco

    “Also, +1 for Wesnoth. It has a completely deterministic combat system”.nnYep, totally deterministic. That’s why you can reload a savegame, do the same actions and have a totally different outcome.

  • Anonymous

    This is what you essentially just wrote: “I am not going to listen to your points because people on a select set of internet forums would disagree with your conclusions.” That’s silly. If you’re not going to offer a valid counterargument, why post?

  • Tuco

    I’m not listening to his (your?) point cause there isn’t a point. At all.nIf you pretend to paint the ToEE battle system as horrid we have nothing more to argue about. It isn’t just wrong, it’s a delusional claim.nThe only reason to dislike it is to hate turn-based combat, but then I should add that people who hate turn-based combat systems can just fuck off.nnBy the way it’s very easy to pint the flaws of D&D ruleset, we all know them. And there are at least half dozen of pen and paper better rulesets.u00a0nWhat we don’t have, instead, it’s videogames using something better: with the same staggering amount of tactical options, the same effort to keep things challenging, and so on.nnSo, please, tell me what have you worked on so far. Tell me about your games and how they put the D&D system in shame.u00a0nLet me refresh your memory: Baldur’s Gate 2, Temple of Elemental Evil, Knights of the Chalice.u00a0nThree games, all of them based on the D20, all of them the most enjoyable experiences I can remember about combat in RPGs.

  • Craig Stern

    I’ve never played TOEE myself, so I don’t have a comment on that. As for what sort of combat system would work better, I’ll be posting an article laying that out in just a few days. :)

  • Anonymous

    I haven’t played TOEE, so I don’t have a comment on that particular game. As for combat systems that would work better, I’ll be posting a follow-up article laying that out in a day or two. :)

  • Vorpalone

    I played (A)D&D (2nd Edition) almost every weekend when Planescape: Torment was released.u00a0 I absolutely loved the game, but it was a rather loose interpretation of the combat rules.u00a0 Planescape had some great characters, but the setting was what made the game.u00a0 It allowed for these unique characters to exist in one place.u00a0 Fallout 1 and 2 had combat systems loosely based on GURPS, and again their world was what really set the game apart.u00a0 Sure the characters were great too, but transplanted to another setting they would be far less effective.nnYou mention the newer versions of D&D, but don’t do a very good job of discussing how radically they have changed in the last 10 years.u00a0 4th edition seems more suited for adaptation to a computer game with rules more representing a tactical wargame.u00a0 As an example, there is only one saving throw, it always succeeds on the same number, and there are four defenses (AC, plus the old three saving throws from 3/3.5)nnI rather enjoy random chance in combat, but think it stinks in dialog (skill checks,) so this may well be a matter of personal tastes.u00a0 Random chance encourages you to find every tactical advantage to have the greatest possible chance for success.u00a0 Unexpected events make what might otherwise be boring exciting.u00a0 The unfortunate truth about CRPGs is that they very rarely offer interesting tactical choices that PnP games can as you mentioned above, meaning the system is only half-implemented.nnOverall 4/10nSome good points are in there, but not properly supported.nResearch is lacking on some games.nLack of understanding that the move away from ThAC0 simplified math.nNo commentary on how Real-time with Pause (RTWP) is a terrible method for implementing turn-based rules.

  • Mykolas

    I’d like to go through this point by by point:nn1.u00a0 The idea of allowing for more roleplaying is a good one, but the rolls allow for the NPC’s and enemies of the game world to fight back; combat is indeed about combat, rather than a puzzle. In your example with the cave, theu00a0enemies are completely passive. There’s nothing the enemies can do about what your doing: Gravity Well will always hold them, Crumble will always cause the rocks to fall andu00a0kill them. Every cave encounter has been solved.u00a0I’m not saying one is better than the other, but it can hardly be claimed to be pointless to inject some measure of randomness into the equation.nn2. D&D combat rules are actually simple at their heart. You make a roll, add numbers from things that make it more likely to hit, subtract things that make it less likely. If it’s above a certain number, voila, you hit. Rolling for damage likewise, as you could have a very weak hit or a strong one. Roll a 1 on a longsword? Perhaps that bandit made a desperate parry that caused it to only graze his arm instead of something vital. Roll an 8 instead? You struck true and dug a deep gash into his side. The differing dice sizes are for logic. Logically, a dagger (1-4) will tend to do less damage than a greataxe (1-12). You add modifiers, yes, but they’re largely strength based. If your character is strong, they hit harder. Not all that inscrutable. nn3. As mentioned above, the randomness changes the world from a static one to a living one, where sometimes things don’t go exactly as planned even with the best of intentions. And Playing the game properly, you shouldn’t need to worry about it all that much. A good fighter should be able to hit reliably, and his combat bonuses let him, whileu00a0wizards might not be the strongest, but they can still get lucky. In my view, it’s good that the players aren’t completely in control; who is honestly completely in control of everything that happens to them? No one is the protagonist in a sport; both teams play and take actions. It’s not a case of “Player A decides to score,” it’s “Player A tries to score.” The D&D universe is no different.nn4. There are plenty of tactical options. I honestly have no idea where you’re coming with this. Flanking, stealth, charging, ranged of various kinds, a whole heap of different spells… and gasp! The opponents also use tactics to beat you. As for complaining that enemies close right away, that’s what melee characters are for. Opposing melee characters won’t stand around doing nothing, they’ll charge in swinging; use yor own melee characters to stop them.nLet’s give an example from Baldur’s Gate. The party is gathered outside ofu00a0a door that, from a previous failed attempt, I know leads to a room with 4 drow magesu00a0and a couple slaves. When I went through the first time, they stunned my party with spells while the slaves rushed in, holding off the characters who weren’t, then they threw fireballs until we were heaps of ashes. The second time, after coming back with more experience, I opened the door and then retreated, allowing the enemies to come out single file into my archer’s reach, while the druid cast a spell to functionally incapactiate the mages, leading to my victory.nnIncidentally, that “out of control” thing from Baldur’s Gate? That’s because the game actually does use something resembling turns. It’s not unpredictable in the slightest. Rounds last 6 seconds, in which, you could attack, or cast spell, or drink a potion, or ready a weapon. Each takes varrying amounts of time, but you can largely do one thing a round, like in any turn based game. The “delay” it talks about is waiting for the next round to begin. You can even set the game to auto-pause at the end of a round (among other conditions), turning it into completely turn-based gaming.u00a0Would you complain in a different RPG, say Golden Sun,u00a0that you can’t attack and use an item in the same turn?nnI will agree that Baldur’s Gate isn’t a classic because of its combat, but it’s not held back by it either.u00a0The flaw in games that don’t use D&D well is just that: They don’t use D&D well. It’s a fine rule system, and it works fantastic for RPGs. But any tool can be misused.u00a0

  • Pingback: Roll a D20 to Hit: Using Dice-And-Paper Rules in a Computer RPG.

  • Pingback: 12 ways to improve turn-based RPG combat systems « sinisterdesign.net

  • http://notimetoplay.org/ Felix Pleu0219oianu

    Oops, my bad. Only some aspect of combat are deterministic, such as how much damage a unit does. I was misremembering the discussion here: http://forums.wesnoth.org/viewtopic.php?f=6&t=21317nnIn any event, it feels very fair compared to other games I’ve played.

  • Guest

    “So you might be surprised to know that, as a game designer, I get nheartburn any time I see a computer game deliberately aping the D&D ncombat rules.”nnYou are only allowed to say that, when your games get name recognition of at least hardcore indie-players. Hint : they dont.

  • Anonymous

    I don’t think games need to get rid of randomness entirely; I quite enjoy a little risk in my combat. But D&D goes too far with it, in my view.nnWhen I think of combat in Baldur’s Gate, I mostly just recall hours and hours of low-level fights with kobolds, hobgoblins, gnolls and orcs in open areas with no viable choke points, a highly limited selection of spells, and insufficient number of characters to even think about flanking (flanking, of course, meaning that you have multiple characters on multiple sides of the enemy at once).nnThe first boss fight in the game (the one with the mage soon after Gorion bites it) was a great example of a battle almost entirely dependent on die rolls. He has magic missile, which always hits, and he attacks first. The first time I fought that character, all his magic missiles dealt close to their maximum damage and I died immediately. The second time I fought him, they didn’t–I survived and killed him. That, to me, illustrates a combat system too heavy on randomness.

  • Fool

    Except they do:nnwww.tvropes.org/pmwiki/pmwiki.php/Main/TelepathRPGnnGetting onto TV Tropes isn’t an epic accomplishment, but it does require at the very least a moderately-sized fanbase.

  • FinalSonicX

    Quite frankly, this is poor reasoning and much of it isn’t true or is highly misguided.nn1. The D&D system does indeed have many rules, but it is not necessary to use them all or even under all circumstances. Regardless, later incarnations of the system have removed many of the seemingly obscure tables. The DM may still need to sort through tablees and tables of data, but players largely rely on a few core mechanics (d20 roll plus modifiers against a target number) while the DM handles the world’s interactions.nnEvery single weapon has different randomized damage because otherwise what weapon was being used would be pointless, and frankly it just makes sense that a giant axes is going to hurt an ogre more than a dagger will.nnWhile 2nd edition did have a lot of strange and bloated rules, moving forward into 3rd and beyond these have simplified considerably. Saving throws are now sorted into three extremely simple and straight-forward types. There is nothing bloated about 3 saving throws, HP, AC, and spells. We might agree on the bloat elsewhere, but the core concepts are simple.nnStill, I can admire the desire for a simple, more elegant rules system. I agree that sometimes the books can be overwhelming in size and content.nn2. The system is not inscrutable in the least. In fact, the system is very up-front about everything because by necessity the player is exposed to all possibilities/probabilities and likely knows a handful of the probable outcomes. Your example of Thaac0 is hilarious, because THAC0 is easily one of the most convoluted and complicated mechanics ever used for something as simple as trying to attack. Later editions of D&D made it much simpler. THAC0 was completely backwards: you wanted a low thaco and a high roll, and you wanted low AC. This is highly counter-intuitive, especially since ability checks in 2nd relied on rolling BELOW your stat (thus, you wanted high rolls to attach but low rolls on ability/skill checks. The math with THAC0 was also complicated. You need to take the number, subtract the enemy’s AC (this also helps reduce the level of mystery around monsters if you can know their armor), add/subtract situational modifiers, and then roll and compare against this number.nnAll of this was simplified and improved beginning in 3rd edition which unified many of the mechanics. Instead of all of this confusion about high/low, it was simple. High attack and high armor are the best, you roll a d20, add/subtract modifiers, and compare against the enemy’s AC (the DM could then declare a hit or miss, allowing the enemy’ AC to remain a mystery beyond the first attack). Ability and skill checks were similar: roll, add/subtract modifiers, compare to a target number. Incredibly simple and elegant compared to THAC0.nnOne can calculate the chances of something within reason, but players should not simply know innately what chance they have of any action. They may reasonably request of the DM an estimation of the difficulty of any given task as they perceive it, and make the decision to go forward or not. This is simple to do in computer game or a pen and paper game. Does this lock look like a challenge for me? Yes/No/Degrees of Difficulty. Additionally, each +1 is +5% chance of accomplishing something, and each -1 is -5%. This is incredibly simple if you know the DC/AC of your given challenge/target.nnAll of this is far superior to the incredibly obscure, confusing, and poorly balanced systems used by RPGS today which are invariably cooked in-house and usually undergo little playtesting compared to tried and true systems like D&D. I don’t mind learning a new game system, but when every other RPG on the market is using their own take on the RPG, it becomes tedious to learn them again and again. All of these problems are compounded by the fact that many games do not come with a manual anymore (much less any explanations of the rules and how they work), and the game intentionally hides info from the player (usually in hopes of increasing immersion). In a game, if no numbers are presented to me, how am I supposed to make decisions. D&D exposes all the numbers to the players, there are no mysteries as to what decisions they should make when building their characters, the only mysteries are the ones they encounter during their adventures. That’s the way it should be.nn3. Randomization is necessary to keep the game interesting and exciting. A completely deterministic system is not a game, it’s a simulator that you can optimize to its fullest potential and never be surprised once. Random systems, on the other hand, involve a great deal of risk/reward, plenty of excitement and surprises. Roguelikes, for example, randomize almost everything, which makes replaying them a pleasure, and I’m often surprised by my adventures in those games. Randomness is essential to enduring interest in a game that attempts to model anything more than a mere tactical challenge between two equal competitors (just as Chess sucks when playing against a new player, playing a deterministic game against the AI is boring. It’s too predictable).nnAs for the player saying it’s “not their fault”, that’s nonsense. Rolls are not forced onto the player unless it is a saving throw of some kind. The player decides their actions, makes their roll, and the DM decides the result. Their actions are usually directly related to how high or low the DC is, and thus it determines how likely they are to be successful. Why are players adventuring if they know they can always succeed under certain circumstances? Some element of risk keeps things fresh. Ultimately, players can make good decisions or bad decisions, and they take full responsibility for those. Single actions might fail or succeed, but good or bad decisions in the long term are usually what determine a character’s fate. Save or die spells are exempted, but a good DM knows to use those sparingly and in special circumstances.nn4. The tactical tools used by the player to mitigate risk are their decisions. This is obvious to anyone who has played D&D for a long time. These decisions can come in many forms, there are explicit rule-based options available to players involving positions are special attacks/actions such as flanking, attacks of opportunity/reach, casting/fighting defensively, tripping, disarming, bullrushing, overruning, charging, grappling, called shots, etc. There are tons of options available to players JUST from the rules. Now when you include all the creative options players have in a real D&D game to role-play actions and have the DM decide results, now things get even more diverse and tactical! Spell casters have even more options available to them.nnloot, character development, and decisions all factor into probabilities, and all of these are under the player’s control. The only thing left out of the player’s control is the situation, the DC/difficulty, and the dice. The dice are not nearly as impacting as the other elements which can be handled well by a good DM or designer. Skill is knowing how to work with a given game to play your best. Random elements exist in life, they exist EVERYWHERE, and to toss them out because sometimes we get upset that we failed our poison save is a bit silly.nnIn short, more games should use the D&D ruleset (or at least some kind of playtested and established P&P ruleset) rather than attempting to crate their own black-box systems that are incredibly mysterious.

  • Anonymous

    You might notice that this piece is actually aimed at “Those rules (and the numerous knock-offs theyu2019ve inspired),” so the various black-box systems modeled on D&D which you dislike are explicitly included in my critique.nnThat said, I don’t think you’ve made a very convincing case for D&D not being bloated and inscrutable. “roll, add/subtract modifiers, compare to a target number” is exactly what we used to do with Thac0, only now we count upwards instead of counting down. Big deal. How is that better? It’s still a multi-stage process just to figure out a simple percentage chance of hitting.nnAnd why does the game need separate saving throws at all? What’s the point? Why can’t the DM just roll against a character’s most appropriate stat (Dexterity to avoid physical effects, Intelligence or Wisdom to avoid magical effects)?nnFinally, of course rolls are forced onto the player! Every single combat action a player takes beyond simply moving around involves multiple die rolls. Die rolls to hit, die rolls for damage, die rolls to see if a spellcaster fizzles, die rolls for saving throws, and so on. They’re unavoidable. Nowhere in this piece do I say that it’s bad to have some element of risk, mind you. Some element of risk can be good. D&D just takes it too far, in my opinion.

  • http://profiles.google.com/terraobject Frederick Burmer

    While I support D&D’s superior stats system ( which are the best part of D&D tbh ) by not allowing the initial stats to increase far too much it gives characters a good sense of individuality. Compare the D&D system with the Final Fantasy system for instance. It didn’t matter the slightest if you used Cloud, Berret or Yuffie as a caster as far as their initial stats go as all of them were ( not counting their weapons ) essentially entirely the same character with the same strengths and weaknesses.nnI digress. The D&D system really falls short on it’s arse due thanks to the combat system and the reasons you provided. I remember playing PS: T and BG2 and just standing there in my boredom whilst waiting for combat to finish so I could talk to someone or in the case BG2 listen to my companions. Well written article indeed.nnHowever I can’t say that your idea of spells in “Epic RPG” were any better. The increased amount of complexity are usually far too powerful for the players to use. It’s in it’s essence something that’s difficult to learn and easy to master. I mean: going into a dungeon and casting the same sequence of spells for easy victories are something equally boring to RPGs as it doesn’t present any challenges to the players.

  • FinalSonicX

    My complaint about the black-box systems is that if a game uses a certain rulesystem in an RPG. they need to expose us to the rules for the system at a deeper level than vague explanations like “higher strength means more damage”. If more games used the D&D system, we wouldn’t have to relearn a new and poorly balanced system for every game, and we’d be able to make things more balanced and less confusing.nnTHAC0 is worse than BAB because THAC0 is confusing and inconsistent with other mechanics in the game. With Attacks, you want a lower THAC0 but you want to roll high. You want your AC and saves to be as low as possible, but you want your stats to be high. With attack rolls you want to roll high, but with skills checks you want to roll low. Whether or not you’re actually adding or subtracting the enemy AC depends on whether they pass the zero threshold (negative armor class, anyone?). Not to mention THAC0 is basically calculated on the fly rather than being static, so time and thinking are wasted. hit calculation with THAC0 = take your THAC0, request the enemy’s AC, subtract AC from your THAC0 (could be positive or negative, remember!), add or subtract modifiers, roll, and compare against this number you’ve generated. With BAB, you roll, add modifiers, report to the DM, and get a result. Simple. To hit with THAC0 is determine dynamically whereas with BAB, the target number is static. Plus, with BAB no AC needs to be revealed to the player, only the result (possible but complicated in 2nd).nnI view it as THAC0 being a method of determining who hits by recalculating the target every time someone tries to hit it as opposed to third where the only thing you need to calculate is your roll + mods. As opposed to your roll – ac + mods.nnAs for the different saving throws, Castles & Crusades has the player roll against their stats for thinggs instead of having discrete values for saving throws, but there are disadvantages. The idea of classes is to distinguish roles in the party, but if everything is based on stats then the wizard could theoretically be the best fighter in the group (as opposed to the fighter). By allowing for the class to modify the rolls and increase over time, you ensure that fighters are usually better at certain saving throws than other classes, and vice versa. Same with attack bonuses (it’s the reason THAC0/BAB is different for different classes, and the same for saving throws.nnRolls are not forced onto the player, they make a decisions and thus they decide to make the roll associated with the action. they are in control of which roll to use, under what circumstances. These decisions help them avoid the more risky scenarios.

  • Thewaever

    Hi, nnThank you for the article. It was interesting. nnHowever, I’m going to take your opinion (on this topic) with a grain of salt for a couple reasons. nn#1. You cherry picked your “Why D&D is Bad” examples. The game has been around in one form or another for 40 years. The rules have gone through at least a dozen iterations, some actually pretty good. The most recent iteration is especially good. I think it matches your Four Virtues very well. nnBut, concerning your opinion about THAC0 for example, D&D hasn’t used THAC0 in over 15 years. You could’ve sired & raised a kid in the amount of time that THAC0 has been obsolete. Which leads me to my second point. nn#2. Your interpretation of D&D rules is incomplete. There are not 3 different types of saving throws in the latest edition of D&D. I won’t go into a diatribe about how the rules really work, because, in my opinion, I think if you really wanted to know, you would’ve found out for yourself. This less-than-perfect understanding at the very least colors your opinion on this subject. Also, the implicit lack of a need to get things right doesn’t inspire confidence. An imperfect understanding of the game can’t help but lead to an imperfect opinion. nn#3. I disagree that amassing “better combat numbers merely amounts to weighing the dice in the playeru2019s favor.” For one thing, I have yet to see any RPG that excludes a leveling up system, and what is leveling up if not “getting better numbers”? In fact, I’ve seen arguments saying that “getting better numbers” is the defining characteristic of RPGs. Not to mention the amount of strategy that can go into amassing the right kind of numbers, which I think you are failing to consider in this article. nnnAll this being said, I agree with you that game systems can be improved upon. In fact, I think that 4th edition D&D is a GREAT improvement on the THAC0 system. 4E D&D very clearly has the Four Virtues you talked about in http://sinisterdesign.net/?p=914nnThat's not to say that 4E D&D isn’t without its problems, of course. But, I chalk those problems up to the game developers not being as thoughtful or careful as they might have been, rather than a design flaw in the game system itself. nnSpeaking of http://sinisterdesign.net/?p=914 I think the list of 12 design features is useful & very easy to understand. Thank you for taking the time to write these articles.

  • Pingback: In Defense of “Gamification” in RPGs « sinisterdesign.net

  • Kaiyl_kariashi

    ToEE is the closest to the PnP experience you can get in a video game. It had effectively all the tactical options (trips, grappling, defensive casting etc. etc.), turned-based combat like PnP, and most of the mechanics and monsters were exactly as described in the books. The ONLY fault of ToEE is that Trioka collapsed before it could receive proper patching to address it’s issues (which were later corrected by the community).

    Planescape was not a combat oriented game and was by agreement by all NOT a representative for 2nd edition combat mostly due to how heavily limited the options were BY DESIGN. What is the point of combat in a game where the main character is Immortal and can pretty much resurrect his party members at will? None, Planescape is a social RPG, more then anything else and that is the reason for both it’s low sales and heavily cultish following for the people who actually tried it. It’s a one of a kind basically, no other game since has offered that experience.

    And Baldur’s Gate is a hell of a lot more tactical then you give it credit. Weapon and gear choices are important, using scounting, making best use of potions, scrolls, wands, available spells. The story was great and the characters are memorable yes that is a very strong part of it, probably it’s strongest aspect but the combat is not weak by any stretch of the imagination, Diablo’s combat is weak, Baldur’s Gate’s is about as strong as you can get. The manual devotes a whole chapter to the basics, the first game’s prelude is an outright tutorial while the 2nd game had a tutorial on it’s main menu to explain the few things you need to know. And really ALL you need to know is lower AC is better. The rest the computer handles and can be safety ignored unless you’re obsessed with power gaming in which case the manual tells you all you really need even for that. If it didn’t come with a manual the complete PDF can be found in the main game directory.

    Dice rolls add excitement, plain and simple. Why is Diablo popular? Random maps every time, make each game fresh even when you know where you’re going and how it ends. Baldur’s Gate is similar, the locations are generally the same, but beyond some set enemies most encounters are random as well as the combat itself adds another element on top of that. A game becomes boring and worthless once you hit a point that you’re basically immortal. That’s one of my biggest problems with NWN and NWN 2, they basically throw gear at you that makes you nigh indestructible.

    NWN’s only saving grace was mods and the world editor. People don’t play NWN for the campaigns, they play it for the persistent worlds or PVP worlds. And most of the mods for it are heavily involved for bring it closer to PnP’s implementation anyway.

    NWN 2 is pretty much more of the same except an even worse interface and made by a horrible company who’s games are only slightly less buggy then the garbage Bethseda puts out.

  • stinky472

    Does this article include the original AD&D Gold Box games like Pool of Radiance and Champions of Krynn? If so, I have to disagree with the general premise about the combat system only — I feel like they are still among the finest examples of tactical, turn-based combat — up there with the likes of Fallout and Shining Force.

    That said, I do agree with your overall assessment that D&D rules and stats are not optimally suited for computer gaming. A simple percentage indicating chance to hit and accuracy stats, would be a lot more intuitive than, say, THAC0.

    However, with respect to determinism, I think overdoing it tends to give a synthetic, lifeless feel to the game. Fallout is a fine example of a fairly randomized system done correctly IMO. That game made random criticals capable of turning the tide of a battle (finding a desperate shot at someone’s eye with 65% accuracy actually causes a blinding critical and wins you the battle, e.g.) and, in reverse, sometimes causing you to lose a battle you could have otherwise won 95% of the time. In spite of this, the game doesn’t feel too random.

    Players are addicted to random outcomes to some degree. After all, that’s the whole selling point of Diablo — if it weren’t for randomized drops, people probably wouldn’t still be playing these games over a decade later.

  • Pingback: Unpredictability and control in turn-based combat: an examination « sinisterdesign.net

  • Howard

    Dark Souls > DnD Combat

  • Tuco Benedicto

    Which is as pertinent as pointing that smoked fish tastes better than cheesecake: in other words, it’s completely idiotic as they just don’t compare at all.

  • The Recursion King

    The tactics in a D&D game like Baldurs Gate come from preparation, spell selection and usage and position. Get any of these things wrong and the difficulty of the games is high enough to crush you.

    Pool of Radiance is much better at strategy, though. D&D (in this case, AD&D) has so many hidden ‘in plain sight’ systems that many of the tactics you need to succeed in that series are really not obvious. All the obvious stuff is just the basics. Arguably we could do with more games like them.

  • teddybowties

    oh man, this is an old thread, but I don’t care. …dice rolls add excitement? random-generated maps add replay value? Tell that to everyone who ever hit the ps1 buttons 200 times in FF10… or for a more pitiable and poignant example, all the Zork fans who saw their beloved Underground Empire reduced to a brainless sniveling whore drooling in a backroom when they turned it into a random battle visual online card game. I prefer Divine Divinity ONE over a game that pretends to be a role play but instead merely scrapes it’s own storyline off it’s shoe in favour of an over-used, bloated beyond recognition, antiquated combat system. no need to reply; it’s merely my opinion. But I’m not the only one. but on the other hand, too much dialogue and I run for the hills. i want the story and gameplay to be thought-provoking, so good ON AVERAGE that I forget I’m playing a game; not give me a beautiful but empty enema because someone was still drooling over an 80’s era math problem. I don’t intend this comment for incitive insult, however; i merely point out that the excessive use of such systems in games that could be better off without any system like dnd at all is crippling the gaming genre as a whole, stunting the contextual thinking skills required FOR and therefore the evolution OF the concept which once made gaming great; -innovation with vision and purpose- and A MUCH BETTER STORYLINE THAT MAKES psychological sense, along wiht that old favorite… NO SHOEHORNED morality dialogue trees unless they ADD to the story in a concrete psychological-sense-making way. ;)

  • Cheimison

    All this tells me is that we don’t share the same interests in video games or D&D. Also, don’t treat the versions of D&D as equivalen t. They’re very different, especially between0-2E and 3rd or 4thE. I like resource management games where I can die by being unlucky. What you dislike is a feature, not a bug. Also, no one was even supposed to use all the AD&D1E rules, and the system is less complex than most modern RPGs.

  • Cheimison

    There is no Universe in all of possibility where 3E is simplified from 2e. It’s MORE complex in terms of modifiers and rules you need to track. But it has a more standardized system (d20). However, combat and non-combat play are vastly simpler in 2E than 3E, especially at high levels. You can’t equate ‘less confusing primary mechanism for non-Grognards’ with ‘simpler’. 3E and especially 3.5/PF are rules medium-rules heavy. 2E is the heaviest of the old games, but most of those rules are clearly labled as optional.

  • Cheimison

    Saves aren’t rolled against stats because saves are level dependent, and reflect a class’ unique resistance to difference forms of danger. If we used a stat roll Mr. Wizard could resist poison better than a Dwarf just because his CON was higher. This encourages stat inflation, pointless multiplication of modifiers and gimps low-stat players for no good reason. Here’s my question: instead of criticizing a D&D system you admit you abandoned early and do no understand, why not do a google search so you can easily learn all the reasons for these things? D&D isn’t a perfect system, but its a low-fantasy exploration game and it does that well. If you don’t like that and what Plot Mail for your PC, play another game!

  • Cheimison

    THAC0 is not confusing. Some people just have trouble with subtraction. I’m not a math whiz, but I’d rather do THAC0 than add up 900 circumstantial bonuses from my feats and positioning, etc. And you can easily invert it…THAC0 12 = BAB +8. Now roll, add your BAB, and add the enemy’s armor class. Did you get a 20? IF so, you succeeded. There’s no reason for anyone bu the dungeon master to keep track of more than 1To-Hit number anyway. This whole ‘BAB is objectively superior’ is wankish and lazy.

  • CraigStern

    I DM’d campaigns in D&D for years; I understand the system just fine. I simply think it’s a poor fit for a computer game.

    “If we used a stat roll Mr. Wizard could resist poison better than a Dwarf just because his CON was higher.”

    So what? There are design alternatives to producing a million little tables to try to compensate for an imbalanced–in fact, literally randomized–system of stat generation.

  • Jokey McKegchugger.

    Yes… let’s take the randomness out… let’s let every action “auto-hit… If you do this for the players, you’ll have to do this for monsters too (fair play environments anyone?) Do players want a chance for the monsters to miss? Certainly. Guess what? No matter how well a tactical plan is thought out, it is bound to fail once in a great while. Whichever whiney player boohoo’s and blames dice probably needs to play less dnd, and more non-randomized dating sims.

  • Jokey McKegchugger.

    Btw… The track system was easy to use. Class adjusted thac0 minus armor class equals target number… roll d20. (No need to calculate the percentage)

  • Joseph Hines

    I have to say I agree with this interpretation of D20 Modern/DnD PnP Systems for the most part. My two favorite PnP games are Basic Roleplaying (Chaosium- As a player it’s very easy to gauge success chance to any skill or stat check; as it’s percentile already; and it outright tells a GM that implementation description of a skill’s use is important to determining if it should be easier or harder- As a GM, I replace the clunkier parts with simpler methods anyway, which is easy because of the toolkit approach), and Strands of Fate (All rolls are 4DF+Stat+Advantage Potency, with a resource for adding +2/Reroll that increases as your character represents their faults ;4DF here is a a six sided die with 2+, 2-, and two blank spots … counting +/- and ignoring blank; The weapon/armor damage rules require a single roll but are balanced, and there’s methods of using it for three types of stress or more, the obvious being mental/physical/social).

    As a GM, I’ve found DnD too random, too hard to balance new creations in any of the editions. It’s nearly a science in BRP and Strands, and in D20’s later editions they’ve made too much of the tactics only available with the right feat- If you have an advantages system, don’t make tactics so dependent on them, because they should represent special training, not things you should already be able to do. On top of this, a lot of GMs don’t care about HOW you do something in DnD, just the rolls. It takes a lot more for a GM to convince me to play DnD than any other system that I don’t know is outright stupidly bad initially.