Mass Combat 1: What Makes a Good System for a Roleplaying Game?

In many an adventure roleplaying campaign, there comes a time when the heroes' struggle gets caught up in larger politics and warfare. They may find themselves helping villagers to defend against a marauding army; get embroiled in a street fight between rival gangs; or, at higher levels, use their hard-earned treasure to hire a mercenary army and rid the land of a nasty orc warlord.

Scenarios at this scope and scale would be tedious to fully play out, rolling for the hits and damage output of each single figure. This is where mass combat comes in. 

Photo by Thom Quine, CC BY 2.0

Mass Combat Desirables

The irony of developing a mass combat system for D&D was fully apparent to the authors of the first such rule set, AD&D Battlesystem. In the introduction to the first edition core book (1985), Douglas Niles points out that the D&D game developed from the Chainmail system for fighting medieval battles at scale, so this product represents a return to the roots of the game. 

Chainmail itself contained rules for fantasy creatures as well as castle assaults involving individual figures. Indeed, early games at Gygax's and Arneson's tables used the dicey, insta-kill Chainmail man-to-man combat system, before Dungeons and Dragons adopted the more heroic wearing down of characters' hits derived from naval combat. 

However, Chainmail mass combat in relation to Original D&D skirmish combat lacks one property: the ability to directly transfer statistics from the roleplaying game to the combat procedure. Indeed, if mass combat bears no mechanical resemblance to the roleplaying rules, you may as well use any tabletop wargame system, be it Warhammer, De Bellis Antiquitatis, or whatever. 

What we are looking for is scaleability: the ability to simply and uncontroversially convert stats from characters and monsters to the stats of the units they make up. This goal is compromised when complicated equations and statistics intervene between one layer and the game and the next. Ideally, you want your units to have Armor Class, hit bonuses, hit dice/points, and movement rate on similar scales to the stats for their constituent figures.

A particularly tricky part of scaleability is the mixing of individual characters such as leaders, wizards, and heroes with masses of troops. You want your system to not just handle battles as a spectator sport, but to actively involve PC participation. The role of individuals in a mass wargame is likely to be abstracted, as a commander token with +2 morale and 3 command points, or an embedded hero who gives +1 to the attack roll. Players, however, will want to see their characters embodied, with access to the familiar set of combat moves and tactics. 

This simulational seam can be finessed, somewhat, by playing out the wargame abstractly when PCs remain in a purely command role, but then zooming in to standard D&D rules when the PCs themselves make contact with the enemy. However, using this method, it might not be clear how the PCs' kill count transfers back into the play of the larger battle. How does eliminating eight knights out of a unit of 100 affect the morale and cohesion of that unit? A satisfying mass combat system for D&D will integrate individual and mass levels organically.

There are other traits that you want the mass combat system to share with other wargames. You could adopt a "power grinder" approach where you feed in troop totals, armament, morale and other factors influencing the battle, then make a single roll that tells you which side won and with how many casualties, like the venerable old computer game Civil War. But that would lack the desirable trait of incrementality - the feeling that the battle is evolving in stages, going back and forth, and you can see exactly when a command decision, a heroic charge, or a well-placed fireball turns the tide.

Another desirable trait is bounded chance. War is chaos, and you don't want the results when two forces meet to be completely predictable. At the same time, you don't want 100 scrawny kobolds to have any kind of plausible chance to defeat 300 brawny gnolls in melee. If seen as a collection of individual combat outcomes, the law of large numbers gives rather predictable casualty figures. For example, if 100 orcs meet another 100 identically armed orcs in D&D combat, with a 25% chance for each figure to kill its opponent on the first round, then 95% of the time the casualties on that round will be between 16 and 34 inclusive, per side attacking.

However, it's on the larger scale that chance reasserts itself. In premodern warfare the question is not how many soldiers will die, a low number compared to the carnage that 0-level D&D figures can wreak on each other in the space of 30 game seconds. Rather, the question is when one side will break and run. This is where the higher-level resolution of the morale check comes in, even more applicable to mass combat than to skirmish. Included under this heading, too, are factors like the formation of the group, which has important consequences for casualties taken and received but cannot be easily equated to the traits of individual combatants. As with any wargame on this level, unit games need to model the emergent group-level properties of morale and formation alongside raw casualties.

It might be tempting to simply treat a hundred gnolls like a single gnoll, having them swing, take damage, and have a chance to run when wounded using the D&D rules. Eric Diaz at Methods & Madness broached this technique recently, but while it is eminently scaleable, it trades away bounded chance and needs its own rule for morale. At the other side of the range are overly complex or wargamey systems that don't scale to D&D stats or don't allow for PC-level input.

Next post, I'll talk about some of the systems out there that I've looked at or tried in practice. Post after that will be my own offering, recently tried out in a street battle in my ongoing campaign.

Failed Monster Designs

Although I have written about bad monsters in RPGs, you can identify another type I have sometimes written about: the failed monster, whose basic idea is OK but whose mechanics are off. Either it's too weak or too strong relative to expectations, or just not a good join-up between concept and implementation.

Some examples, from AD&D first edition, with my writings about the first three:

• Ghosts are super-powered, with “zap” effects like aging and possession, but aren’t really true to the variety of their source material. It's a similar failure to golems. There should be lower-level hostile living statues (as in Basic D&D) and lower-level hostile hauntings.
• The gelatinous cube as a 4 HD bag of hit points just doesn’t, er, gel.
• Piercers can be much improved. 
• The carrion crawler with eight attacks is an overpowered paralysation machine. It hits plate and shield on a 14. Good luck!
• The slithering tracker is an undescribed, underdeveloped monster – really, more of an effect -- of consummate unfairness.It tracks you invisibly, paralyzes you in your sleep, and kills in six turns. There is no way to set watch for it unless you're willing to prod sleeping comrades every hour for a reaction.
• AD&D dragons are borderline failed. Certainly their implementation in 70’s-80’s D&D, with fixed HP and breath weapon damage, substitutes “special” for “scary,” and has been repudiated by every edition since and even some retro-clones.
• Harpies are mixed-up with sirens. There should be just normal shitbird harpies. Charm-harpies should be more powerful than they’re given credit for, with squads of charmed minions.
• The oddly specific horror story of the night hag is hard to use in actual adventuring. Like the Fiend Folio’s penanggalan and revenant, the hag’s description is focused on her threat to a lone civilian. It's assumed the party is supposed to barge in upon and rectify the haunting and draining by the hag, even though the victim by definition has to be exceedingly evil.

A common theme in these and other failures: indecision about combat encounters. There's a desire to make monsters about more than a line of stats, to make fighting them a matter of strategy and decisions as well as lining up and whaling on them. But these work better as rules than as haphazard monster effects. That way the strategy can generalize, and be inverted, working for both sides. Just some examples, some of which I've written on:

• Monsters can be scarier, and true to life, by coming into close combat where your weapons are less effective.
• Little monsters can, and must, climb you. You can also, and must also, climb big monsters.
• Flying monsters should be annoying as hell.
• Immunity/resistance to weapon effects. No flesh, can't slash. Nothing hard, can't bash. No vital points, can't pierce.
• Monsters with that one weakness. Puzzle monsters, in a word; murderous locks with murdering keys.

The Awesome Pain in the Ass That Was DragonQuest

Continuing my trio of bargain-bin rescues from Glasgow (actually a gift from Paolo, in lieu of buying an adventure from the system) I present to you DragonQuest, 2nd edition!

Why does this RPG system set my teeth on edge? My nostalgia should be all for SPI and the days of punching out wargame counters, all for Deathmaze and Citadel of Blood and War of the Ring and Sword and Sorcery. But good boardgaming chops do not guarantee a good roleplaying game.

Let's judge a book by its cover. Sorry guys, RPG players are not just dreaming of being Conan. D&D art got that right more often than not. They are in a fantasy-hero world, but team players; just like they're in a horror world, but not doomed, and in a science fantasy world, but stone cold medieval. The Frazetta muscleman hoisting up the results of his DragonQuest like a trophy bass is someone else's idea of "sword and sorcery".

The writing style of the game is a 180 degree reaction against the fast, loose but evocative D&D writing of the time. No gaps and confusing terminology here! DQ is buckled and strapped into the case law structure of an SPI wargame's rules (see 3.7.5.1 and apply the Rules Writing Procedure). If the GM has leeway, we'll tell you exactly where that leeway is. It's meant to be clear, but it's mechanizing and alienating on the page. The wargame influence also shows in the tight regulation of combat on a hex grid.

Maybe case-law would work if the mechanics were more elegant, as in Metagaming's contemporary offering The Fantasy Trip. But they're standard Rolemaster-type fare, a percentile skill system with "RPG 2.0" features like separate fatigue and physical damage, damage-reducing armor, critical hits, background packages, custom advancement ... Determining target numbers might have you multiplying 39 by 2.5. Damage involves frequent table lookups to see if a crit and physical damage happen. God forbid you should have a d6 laying around the house, here, roll one of your d10's and take half for a d5 instead. And roll four of those d5s to determine your character's stats.

What's a Satanic panic?

But it's funny how often interesting magic systems come attached to clunky base mechanics, while elegant systems like TFT or RuneQuest have difficult or prosaic approaches to magic. Certainly, something was possessing SPI around the turn of the 80's. They had a boardgame about the demons from the Lesser Key of Solomon, and worked some of those names into their Citadel of Blood adventure game. And yes, there's a whole school of DQ demon summoning that ramps up to 16 pages of fully powered Goetic infernal royalty. That menu is clearly where all the love lies, and the other magic schools suffer collectively by comparison. Some are solid, some near-unplayable like the Water Magic school with its Aquaman-style restriction.

By the way, there is a lot of cribbing from D&D, especially in the monster list. And in the kind of rules that compel game balance. Wizards can't cast near cold iron or while being distracted by damage. Player characters who poison their weapons might nick themselves. This points at the heart of the problem, that Dragonquest isn't built around a compelling setting (implicitly, as in D&D, or explicitly, as in Runequest). So much of it is generic that the special stuff fails to stick.

For example, instead of alignment, your characters get a quasi-astrological Aspect which gives them bonuses and penalties for very short periods of the day or year, or around a birth or death. Sounds cool, but it doesn't really resonate with any other social or magical structure, mostly boiling down to an optimum time and place to do housebound skill tests. Only the death aspect has any impact on the typical adventurer, with a +10% bonus just after a mammal dies near you.

At least the 2nd edition book concludes creditably, with a tight little sample adventure in a bandit oasis. It maybe shows, though, that DQ doesn't really know what kind of fantasy game it is. The journey to the camp is described last of all, oh yeah, you might encounter a sand golem. The real detail is put into the characters at the camp, their secrets and intrigues. It's not really necessary that one is a halfling and another is a hobgoblin. The magic, too, is subtle, pulp-story stuff. There are other consequences of aping the pulp era (the camp is run by one "Alla Akabar," and roles for NPC women comprise jealous wife and sex victim). Perhaps the game is more suited for would-be Conans after all?

Anatomy of an Unused Rule

What prompts a game writer to include rules that they don't even use in their game? It happened to Gary Gygax, writing the AD&D Dungeon Master's Guide, putting in all those rules about weapon vs. armor and contracting diseases on the random and grappling.

It happened to me - the decision to give experience for sites, not monsters, in the 52 Pages  rules never came about in my own games.  I kept on with a version of the 100xp/ hit die rule from the oldest version of the D&D, the one that has worked well and given a decent rate of advancement of the last four years of play (together with xp for treasure, carousing and occasional story awards for momentous acts that deliver none of these) .

You could argue that writing a rule you don't play reflects a failure of playtesting; in my case, more like a mistrust of playtesting. Why deviate from something that works? Experience, and a look at the AD&D DMG, gives two main reasons.

1. Misguided realism. The driving force of blogs and heartbreakers - the need to install a system that more accurately models some process. These fail for a number of reasons.

• The GM finds it hard to use the rule on the fly. See AD&D grappling and any number of "Rulemaster" procedures. 
• The rule makes the game more difficult and less transparent for the players. Many rules that give combat options have this fault. Tactics shift to knowing which attack option or power move to use at which time, rather than common-sense ideas based on movement and positioning.
• The rule takes away focus from the main action of the game. In 52 Pages I found that recording where on your body you kept each piece of equipment, although consequential for play, was not appropriate to an adventure game. Encumbrance is just fine as a list-based "you're carrying too much stuff" consideration.



There are other ways to save the peasants.

• The rule misunderstands the way the players want to interact with the setting. If you want nasty element like disease or sexism in the game, it's better to present examples in the game (plague-ridden towns to be avoided, female NPCs having a hard time) rather than mechanically restricting or punishing the players with random disease chances or gender limits.

2. Misguided control. This kind of rule tries to use carrots and sticks to motivate player behavior, but ...

• To encourage something that players do anyway (like giving benefit points for roleplaying).
• To discourage something that really isn't a problem (like players killing monsters "just for the experience points")
• To discourage something that really should be handled by better campaign management or interpersonal skills (like player misbehavior towards others, or characters gaining excesses of treasure from the system.)

My experience points rule fell into the latter category. It treated my players like cow-killing, rat-farming murderhobos when they really weren't --and when in any case there were easy fixes available, like awarding xp only for hostile attackers and awarding less xp for "overkill" of much weaker foes. More on that next time.

The advantage of having experience come from concrete achievements rather than abstract geographical goals? It's adaptable to using other people's adventures -- I don't have to go through Temple of the Iron God or Castle of the Mad Archmage putting little stars in key rooms. It also feels less arbitrary to give out experience through things that exist in their own right. The only feature I would consider adding would be to give some kind of reward for bypassing traps, and that is easily enough figured - say, 50 xp for each 4 points of average damage, with more for unavoidable or lethal designs.

So next time I'll explain the one innovation that I used in coming up with my new experience rule, to deal with a specific problem implementing it. Will it be destined for the dustheap? Let's see ...

Osgiliath Is Brasilia

There is something not quite right about this argument against the design of an unattributed map of the Tolkien city, Osgiliath (the art, if you look it up, is by Dan Cruger but I'm not sure how much of the design is his). That something comes from the hidden assumptions when we - meaning people in the English-speaking world - talk about "fantasy" anything.

* Merrie Olde England is the quintessential medieval kingdom. (Cue Robin Hood, jousts, chicken legs, friars, & c.)
* London is its capital.
* Therefore a realistic medieval city will follow the same planning logic (or lack thereof) as London.
* Etcetera.

First of all, this isn't even true of the medieval world, where most urban dwellers by numbers lived in Chinese cities built on a grid system. All right, it isn't even true of medieval Europe, where the Mediterranean world had Roman city planning to build around -- although as this lecture points out, the original grid often got clogged and complicated for reasons very different from cows needing to find water. None of these reasons enter into Tolkien's world.

Another totally unrealistic fantasy city.

And if you look carefully at Tolkien's Europe-by-analogy, Gondor is the successor state to Numenor's Rome-cum-Atlantis. Osgiliath, too, was a purpose-built capital founded by scions of an advanced civilization. In other words, there is every reason for it to look more like Rome, Washington DC, or Brasilia than London.

The city, in fiction, endured for a thousand years without any of the economic burgeoning or social decay in Tolkien's world that the real European Middle Ages experienced. Again, you can bust Tolkien's world for its assumption that vital goods are dutifully brought by silent farmers to the real scene of the action, where kings and knights rule steadily and wisely unless interrupted by Sauron's evil meddling. But Osgiliath's design in that Middle Earth Roleplaying product, its rational avenues and small size, fits perfectly with those assumptions in Tolkien's very low population density world.

The military critique is another thing, and you can surely see how Osgiliath fell --five hundred orcs in canoes floating downstream would be enough to cause serious havoc. Consider, however, that the place was founded as a refuge on a seemingly peaceful continent, and there might be room for the kind of planning hubris that plagues purpose-built capital cities ("But you must live on both banks of the river so I can build the BIGGEST BRIDGE THERE IS!")

Summary:

• Yes, there is room for a materialist analysis of fantasy literature and gaming.
• That analysis, however, has to proceed from the terms of the fantasy world itself. 
• So, internal contradictions raise the most problems. 
• External contradictions are problems only if people think they are not -- like the popular view that Tolkien's world embodies "the medieval" (instead of what it does, which is to completely skirt around the medieval with an 18th-century bucolic society's view of the sanitized sub-Roman Dark Age in which it finds itself)
• Analysis from original sources will always beat received ideas. I mean, the next fantasy city I design is at high risk to be, like Florence, a rational avenue-planned city now dominated by the blocked-off compounds of warring clans, arrow towers and all.

Strong Magic Cursed

I forget where in the multitudinous blogoland, but someone posted a comment to the effect that high-level characters in Lamentations of the Flame Princess (think basic D&D with player character power dialed a bit down) don't do well in high-level AD&D modules made for more souped-up characters.That may be so - and is doubly true of my own 52 Pages rules where you can't come in with 4 magic missiles and 2 fireballs prepared at the same time.

I mean, great. The less overwhelming your high-level characters, the more you approach the ideal where big scary monsters are actually a threat to them rather than resorting to blind-tiger gimmicks ("you wake up naked and bereft in an anti-magic zone") or stupid dumb munchkin armies of 4 beholders and 20 frost giants, etc. The fewer bonuses pile up on them, the less you feel you need to compensate in an eternal treadmill of armor class and hit bonuses.

Now, another assumption of high-level AD&D, that grittier referees may balk at, is the ubiquity of magic items. A 7th level character in AD&D is likely to be tricked out in at least +2 everything, or a couple of wands and many lesser items if a spellcaster. One answer to this, of course, is the low-magic campaign where items are rare. The answer I prefer is "keep Fantasia weird" and filled with magic, but balance strong magic - most +2 and definitely all +3 and above - with the cosmic revenge of the universe on tools that so defiantly flout its laws - with the karmic debt paid for owning a sword that cleaves bronze like butter - with, quite simply, the human energy cost of a stick that shoots death. Something's got to pay somewhere.

(In effect - a shallower and broader implementation of the 1st edition AD&D "random drawback" approach to artifact-level magic.)

30 Random Drawbacks for Magic Weapons and Armor (d20, -5 at +2, +5 for each plus above +2)
"Armor" here includes shields.

0 or less: No drawback
1: -3 to a random saving throw.
2: -1 to all saving throws.
3: Accentuates your worst personality traits. -2 Charisma
4: Constant whispering sound makes it hard to concentrate. -2 Intelligence
5: Estranges you from God and nature. -2 Wisdom
6: Exhausting to wield. -1 Strength if an armor, -1 Constitution if a weapon.
7: Makes unexpected, clumsy, confining moves. -1 Dexterity.
8: Take 1 hp damage when you equip it.
9: Take 1 hp damage when you un-equip it.
10: If you die while wielding/wearing it, you rise immediately as an undead creature of hit dice appropriate to your level, and attack the party immediately.
11: Has minuses instead of pluses when fighting one creature type (reptiles, undead, humans, etc.)
12: Animals are unfriendly to you while carrying it.
13: +1 of its enchantment vanishes for the day when you hit (weapon) or are hit (armor) on a natural 13.
14: You need to eat five times as much on any day you use it in combat.
15: While wielding or wearing it, unintelligent enemies attack you by choice.
16: When you are aware of an enemy, you have +3 move to go towards them, and -3 to go away.
17: To enjoy its magical benefit, requires you to forswear your religion and follow an obscure, nearly-dead god, wearing its symbol and following its strange customs.
18: Can't heal HP while you're carrying/wearing it.
19: Glows visibly when enemies are near, within 60'... but only if you already know they're near.
20: Jealous ... drops from your grasp if you're carrying another weapon (weapon), falls off your body if you're carrying any other magic item (armor).
21: Each time you do (weapon) or take (armor) 8 or more HP of damage in one blow, you lose 1 HP.
22: Fogs your vision, you can only see 30' in dim light.
23: Makes an audible screaming sound when it hits (weapon) or when you are hit (armor).
24: Only has its magical powers each day if exposed to the rays of the dawn.
25: You must rest and not attack one round out of five while wielding or wearing the item.
26: You lose the ability to speak while wielding/wearing it.
27: Gives -1 to hit (if armor) or 1 worse armor class (if weapon).
28: You hit your nearby friends (weapon) or your nearby friends hit you (armor) on a natural to hit roll of 3.
29: Any NPC's who see it and are able to wield/wear it must save (Will/WIS/spell) or become covetous and scheme to take it from you.
30: Powerful wizard/demon/undead creature thinks the item is theirs and begins pursuing you d4 weeks after you acquire the item.

Kicking the Cleric Out of the Niche

Here and here are Talysman's posts where he hates "niche protection." It's a question of who-can-do-what in a role-playing game. I agree that the rules should be based on a view of what characters can do that has its own integrity, not manufactured to create some outcome at the player level.

And yet - especially when it comes to magic - there's such a variety of plausible ways of doing things that we can't help but choose among systems. In that case, why not choose wisely? As I argued a while ago, a good game will strike a balance between making a given skill set useful while not making it strictly necessary.

One example: the cleric. Many experienced players, depending on their D&D-like system, consider the cleric to be an indispensable member of the group. Let's look at the most extreme version of this, in AD&D, where it is easy to get a cleric who starts with two first level spells(1 bonus spell due to 15+ wisdom), almost always Cure Light Wounds.

So you get a slightly less effective fighter with one less hit point on average, worse attack and limited weapons, BUT:

Average HP restored from those two Cure Light Wounds: 9
Average HP of a first level fighter: 5.5

That's right, a character who not only fights in their own right, not only turns undead and can do miscellaneous other things with spells, but provides a force multiplier in the course of play (assuming damage is not taken in huge lots) equal to almost two fighters.

The decision in OD&D and B/X to not give clerics ANY spells until 2nd level in hindsight seems reasonable, although Labyrinth Lord climbs down from this with 1 spell at 1st level and so on. Even without bonus spells this is still a pretty nice force multiplier, that lags behind the fighter's hit dice at middle levels but perks up with the ability to cure serious wounds and raise dead at higher. The choice, still, is not ideal; between a game where clerics have to prove their gumption by spending their first level as a sucky fighter who can turn undead, and a game where clerics are seen as mandatory to the point where people will play them even though they don't really want to.

I've seen first-hand in my gaming group the dismay with which players face the prospect of adventuring without a cleric (my 52 pages version has a healing power that likewise is a pretty big force multiplier). The obvious fix is to do what I did with wizards to stop the "sleep/magic missile" fixation; allow only one example of each spell to be cast a day. Because my system lets spells become available at every character level, I could even give a very minor healing spell at cleric character level 1 and the Cure Light Wounds equivalent only at cleric character level 2.

This might be the last major change to the 52 Pages, but I think it achieves the goal of letting clerics feel useful at Level 1, but not indispensable at higher levels. Hell, I might even give them their at-will turning back.

Clever Rules Fade Away

After a languid summer I'm ready to enter the home stretch and put the final touches on my 52 Pages house rules. Scrap Princess' review reminded me that there is always a gap between the clever system you think up at home and what actually goes down at the table. Indeed, I had some stuff in there that bore little resemblance to the way I actually play, like the "encounter start" matrix that was my substitute for surprise rolls.  So I managed to boil it down to be more like guidelines than rules, and more like what I'll actually do in play - figure out the surprise status by common sense, with maybe a roll for alertness if I'm unsure of the disposition of the defenders.

Likewise for my magical treasure table, which caused some puzzlement when it first came out. I decided to make it more straightforward and more geared toward low levels - appropriately for the "Basic" style levels 1-3 focus in the 52 Pages. I might make the main treasure table more straightforward too.

Finally, I had an insight about combat where I could get it down to fewer phases if I realized that combat should go with the most urgent stuff first - not in the order that you might think things happen. So, melee first and disengaging, then shoot, magic and move, and miscellaneous stuff at the end of the round. To handle the weird gamesmanship and panzerbush situations that might arise I allowed "overwatch" to happen so you can shoot the charging guy at close range while he is charging you. In a surprise situation, by the way, you can move first then melee.

Oh and yeah, I got rid of the grid. I still play that way but I'm pretty happy with a system that looks more spacious on the page and asks people to think about the dimensions of the fight rather than necessarily making them plot it all out.

And I've made a start on the example campaign, dungeon, and play session that round out the last four pages. So some of that soon, I'm hoping.

It's Not Uncool, It's Normal!

One of the ways to deny the fundamental uncoolness of D&D is to strenuously insist that it's cool. This fuels a perennial struggle to find ways to bend role-playing games into this fiction. And this struggle has gotten harder over time.

I mean, there was a brief time at the start, before it got out in the mass media, when they could make like D&D was some kind of edgy activity that all the beautiful people were doing - something like Fletcher Pratt's or Norman Bel Geddes' chic Manhattan wargames:

And then ... nope, nope, nope...

Floundering ever since the stigma descended on roleplaying games in the 1980's, two escape routes have generally been tried. One is to fight against the shallower definition of "uncool," to try and purge away the fantasy content from a role-playing game. "This game is cool because it's about real things" - or about socially acceptable forms of escapism, anyway. At this task, roleplaying murder mysteries succeed; roleplaying games about rock bands fail (because only 11 year olds imagine themselves as members of a cool rock band); and some other attempts just merit a stunned silence.

Even in the art of roleplaying games we can see the struggle between the desire to go full-on socially uncool and pimp out your game like the ultimate stoner van or Manowar album cover, or to be "mature" and "restrained" and bring it in like a Blue Note sleeve. DCC RPG, in other words, versus Old School Hack or mine own 52 Pages. This may even apply to game mechanics, wherein lengthy tables and baroque calculations represent a certain disregard for the novice, while stripped-down and simple play holds out the hope that you might get Aunt Tillie to play with you yet.

The second escape route is to accept the fantasy scenario but to deny the truly "uncool" thing about roleplaying, the immersion and identification with character. So, we get the comparisons to chess or to improvisational theater. The former implies that it's just a game of strategy played with funny pieces, the latter implies that all the play-acting is not self-indulgent or self-threatening but done with professional control for an enraptured audience.

"We're storytelling!" No you're not. You're collaborating on a story through role-taking and the closest normal-people analogy is when a bunch of drama show writers get together in a room and work out what happens next ... but each of them takes on a part from the show ... and these parts continue year on year ... Really, the only truthful answer is:

"There is nothing else like it."

Fiasco: We All Had Fun In Spite of the Rules

Last weekend's game of Fiasco, like the other three I've been in, was a crazy blast featuring a preposterous mix of meth dealing, Narcanon, Cthulhu phenomena and Alaska. But I also had the strong, recurring feeling that we were having fun in spite of the rules - that the dice mechanics were inelegant and largely had to be worked around. I have always had the strong urge while playing this game to either Ignore the Dice or Respect the Dice.

Case 1: the rolling for relationships and objects/locations/goals by taking dice from a large pool, which gives a bewildering amount of choice for people who go first and an unsatisfying lack of choice for those who go last.

Ignore the Dice: Just pick.

Respect the Dice: Just roll d6/d6 (and make sure every option on that table is AWESOME, and if you can't think of AWESOME, then have 12 or 20 instead of 36 options)

Case 2: the unclear gamesmanship of taking and granting dice during play. Black or white dice are granted to other players in the first round, then to yourself in the second round, either by yourself (if you let the others set the scene) or the others (if you set the scene yourself). At the end, the only people whose characters have positive outcomes are those who have a large imbalance of black vs. white dice (either way). There's also a tension between players who have been gaming this aspect of the game, and others who have just been playing to make a fun story. You could argue that the negative outcome feeds the grimness of the tale, but in the source material of botched-caper films, there's a space for characters who against all odds come out all right.

Ignore the Dice: Just wager for black and white tokens, with outcome based on the imbalance or some other means.

Respect the Dice: Just roll for outcome, regardless of what happened during play.

The larger point about game design is that Fiasco's flaws are very much traceable to an ideology of player agency. An ideology is distinguishable from a technique when it starts getting in the way of fun. Forget GNS or anything else, here are my three necessary ingredients for fun in a game- PLAYER-centered instead of DESIGNER-centered:

1. agency - ability of players to make meaningful choices
2. surprise - results can come up that shock and galvanize the whole table
3. representation - the opposite of an abstract game, where you can visualize how the game mechanics create a compelling situation or story

My point is, the dice element in Fiasco adds very little to point 2, which dice are there to do. Either go full-on storygame (Ignore Dice) and get surprise from the interaction of players, or respect the dice, and open up to the full ranges of twists and turns that they are there to provide.

Another way to put it perhaps: the choices involved in selecting from dice, granting dice, and reading dice at the end are not representational. "Representational" in this game would be picking some sort of theme or strategy for your character, seeing it through and watching it influence play, and having it inform the outcome at the end.

Design Aesthetics: Simple, Detailed, Procedural

I'm stuck at a point in the 52 Pages houserules where I'm confronting two tasks that seem unpleasant.

Remember, the One Page Rules idea is about being simple, elegant, or at least enough for each chunk to fit on a Powerpoint slide using 18 point font. I'm coming up against two tasks where I really would prefer to do things differently  in actual play, so it's hard to honor the spirit of the 52.

One is my beloved carousing table. I want to fit a whole set of rules onto about 40% of a page. But it seems just too violent to chop down the many odd occurrences to a set of generic outcomes.

The other is the treasure page. Yes, I have an idea for a really super-simple approach to treasure that goes very well with the 52 Pages aesthetics, but again, I would prefer rolling up treasure on a big ornate multiple d100 table where I can get marten fur capes or octagon-cut chrysoprases or Dust of Sneezing and Choking.

See, there are three ways you can handle game mechanics. You can go Simple, where combat is: each side rolls d6, add modifiers, higher side wins. You can go Detailed, where combat means, each side rolls on a d100 table with modifiers and gets a result of what you do to someone else or what someone else does to you. (that would actually be pretty cool - an all-critical-hits approach to combat). Or you can do what most systems do and go Procedural, where there are complicated interactive rules that give the players the feeling they're making decisions.

Each aesthetic also has a standard.

• The Simple should be really simple - easy access, basic results. It should involve as few rolls, lookups, and choices as possible.
• The Detailed should be really detailed - easy access, complex results. Each table should really brim with spirit, options and creativity.
• The Procedural should, by that standard, make the players using it feel like they are making real choices, or should make the GM using it  feel like the world is making real choices. Difficult access, but complex, interactive and realistic results.

This map also lets us identify a no-man's land, where a procedure is complicated and at the same time estranges players from the feeling of choice. The pinnacle of this: the unplayable hyper-realistic kind of system that has flourished throughout the history of role-playing, where your hit on an opponent unleashes all kinds of weapon shear and armor abrasion and bone fragmentation, but ultimately it's all as boring as watching a pachinko ball cascade down. By the same token, offering too many options can also rob the players of the feeling of choice, as they thrash blindly about in the long period before they gain "system mastery."

How would this apply to treasure rules?

• The Simple would just dictate a random amount of treasure value by level, and what general form it is in.
• The Detailed would be a d100 x d100 table with all kinds of rare goods and magic items in there.
• The Procedural would take "treasure types" to the next level, basing treasure on what a creature would accumulate, or what would be found in the particular ruin; you could almost roll through the history of each room ("the tomb had a golden funeral mask; it was removed by robbers but the hidden sceptre was never found, and then in came the gnolls with five barrels of wine and a sack of silver coins.")

By this standard, the 52 Pages is about half simple and half procedural. And yes, as you'll see soon, that also applies to the treasure rules I'm working on. I'm pretty sure that when I finish the 52, the return of the repressed is going to compel me into blurting forth something really Detailed, like Pergamino Barocco.

High Numbers Should Mean Something

From the halcyon days of 2008, and the workshop of 4th Edition, comes this testimonial to the Ever-Rising Statistical Treadmill as design principle:

For example, we strongly disliked the inability of 3rd Edition D&D’s negative-hit-point model to deal with combat at higher levels—once the monsters are reliably dealing 15 or 20 points of damage with each attack, the chance of a character going straight from “alive and kicking” to “time to go through his pockets for loose change” was exceedingly high; effectively, the -1 to -9 “dying” range was meaningless. 

Okay. You've already survived two or three blows that would have pasted a lesser character. You are still on your feet and alive. And you are irked that the next blow certainly would kill you dead. You feel entitled to more, somehow, as a hero.

Think of it this way instead:

• When, as a low level fighter, you are one wimpy blow away from permanent death - you are at -5 hit points or something, unable to move or do anything else.

• When, as a high level fighter, you are one mighty blow away from permanent death - you are on your feet, able to flee, distract, negotiate.

yawn

This is known as a qualitative difference between low and high level play. It's what makes a giant's tree-limb club truly scary, rather than being just a force-multiplied kobold shillelagh.  Or to be (urk) simulationist about it - if negative hit points represent the wracking of your body instead of the wearing away of your heroism, how the hell does your twelfth-level body attain the durability of titanium, as in 4th Edition with its negative hit point threshold based on half the positive total?

Let's take another failure of imagination:

Ask any high-level fighter whether he’d prefer the second-to-last attack from a monster to leave him at 1 hp or -1 hp; I’d put odds on unconsciousness, and how lame is that?

Hold on. This preference has to be based on monsters who you know will rationally leave alone the fallen heroes to go after the living. In the world of Wizards D&D, are no creatures sadistic, hungry, bent on kidnapping, or mindlessly corrosive? Shouldn't you be just as worried about ending up on the floor in a dungeon fight as in a bar fight?  Don't some monsters save their second-to-last attack to see what your head looks like when it pops like a coconut?

What's more, we are also imagining heroes who at 1 hp, or even 10 hp when fighting an average 20 damage dealer, irrationally fight on, instead of realizing they are near death and they need to go home NOW. If being unconscious is so great? Then fall down and fake it after the blow that turns you into a 1 hp fighter. The monster will go on to the next guy automatically and you probably won't even have to make a Bluff check.
Whatever system you're in, as I've mentioned before, approaching combat like a Rock'em Sock'em Robots toe-to-toe battle game is to blame for these absurdities. Fighting should be deadly, cruel, guileful - and it should not feel the same when you're battling orcs at level 1 as when you're battling giants at level 8.

Dungeons and No Dragons; Or, RuneQuest Wins Again

On G+ Jez Gordon remarks:

you know, I've been playing D&D for.... 28 years? never once fought a dragon.

Huh. But he's right - DMs do tend to hold back on the iconic monster of the game. . And this all just illustrates how sometimes "iconic" can be a curse, and how RuneQuest won in the end.

3rd edition D&D represented a wholesale capitulation to RuneQuest design principles - stats for monsters, skills, unified resolution system, damage to objects. Even now RQ is mopping up on the silver coinage standard,apparently to be incorporated in D&D 5th. And another thing it won on was dragons.

In AD&D your first baby dragon would have less than a dozen hit points and you could bag it at 2nd level or so with a few lucky arrow shots, like those Elmore guys I unloaded on last year. Even high level dragons were distinguished mainly by giving and taking large amounts of damage, rather than anything truly amazing.

RuneQuest reached in and grabbed D&D by its second D. Dragons in RQ were demigod-level power mongers and the best you could realistically do was to fight their larval form, the humanoid dragonewts.

Following suit, second edition D&D made dragons more badass, although some of the new abilities backfired - I dare you to think about wing buffets without thinking of Buffalo wing buf-fays. Third edition had hurk gurk colossal great wyrms with their RQ serial numbers filed off.

Not Russ' finest hour.

When the ultimate dragons are Galactus-level challenges, you feel cheap giving the players anything less - kind of like the Firedrake and Ice Lizard in the Fiend Folio weren't popular because they dialed down the value of the dragon brand even further, leaving aside the "Pete's Dragon" influence in their art.

And back to the curse of the iconic. Dragons set up the DM in a double bind. You want your players' first time with a dragon to be special, whereas a minotaur or a manticore can just pop out from behind any old bush. But at the same time there's the nagging doubt that the dragon is also the ultimate cliche, like the skull mountain or idol with giant gem eyes. Both these factors work together to inhibit the dragons from coming out and playing in a lot of campaigns.

The final excuse is that lot of campaigns hover in the low levels, where dragons aren't seen as appropriate foes. The one time I put a dragon in the dungeon it came out of a random room stocking roll on the fly as a party explored the upper works of the Mad Archmage's castle. I think the way I used that dragon - as a self-aware philosophical dilemma - was both suitable to low level players, and broke out of the cliche mold, "the great worm asleep on its pile of treasure" &c. What dragons have going for them is intelligence, which means that old school DMs can drop them in at every level, assuming players are smart enough to grovel abjectly or bargain desperately rather than fight.

The Two-Class Rule

In a class-based RPG, who should be able to do what? Let's divvy up the job of the dungeon crew into:

• Frontline fighting
• Ranged fighting
• Control and utility magic (charm, hold, sleep, etc.)
• Healing magic
• Trickery and troubleshooting (traps, locks, overcoming obstacles)

And introduce the basic class types from Advanced and Basic D&D (left to right, fighter, magic-user, cleric, thief, then dwarf, elf and halfling):

The chart should be self-evident. The thief is a bad ranged attack class, in AD&D anyway, because it's restricted to the one-a round sling rather than the two-a-round bow. Yes, the cleric has some troubleshooting spells, but in practice they're overwhelmed by the impulse to stack multiple heals and Hold Persons. You could also see the dwarf class in Basic as a kind of troubleshooter, although limited, thanks to stonework abilities, dark vision and the ability to fit into small places.

If I could graphically show each class/job as a profile over levels, you'd see the thief starting with weak and unreliable abilities, while the magic-user eventually far surpasses the fighter as a ranged shooter. For now, it's enough to comment on the things that this chart reveals about the "original game":

1. There's a fairly obvious bottleneck where the cleric (or variant classes)  is absolutely necessary for healing. This combined with the other abilities of the cleric make it a golden class, yet often derided as "boring" because of its support role. I'll get into why healing is so great in another post, but I'll leave it for now with an observation that experienced players know the cleric is a must-have member of the party.

2. The thief also has a fair monopoly, with the wizard's troubleshooting powers at low levels being one-shot and limited. This doesn't tell the whole story across levels, though, and this is why the A/BD&D thief is generally considered weak. At low levels the thief's abilities are pretty terrible, while by the time that  thief gets up to reliable powers, the magic-user is starting to learn troubleshooting spells of a whole new class - invisibility, fly, ESP - that completely overshadow what the thief can do. The only drawback is that to memorize those spells means to forego more combat-oriented ones. Adding insult to injury, a wizard throwing 3 darts a round is a better back-rank shooter than the thief with 1 sling bullet, even leaving aside the wizard's ranged spells.

3. Dwarves and halflings in BD&D are pretty much redundant with the base classes, but elves are their own thing.

Just for grins, here's the chart for the Wizards versions, showing how little 3rd edition changed this picture (except for the welcome step of making rogues competent ranged attackers finally) and how much 4th edition changed it, to the well-remarked "anyone can do anything" soup.

Now, what do I want for my game? Some designers see it as a virtue to specialize each class further, so that the optimum party contains all four. Others see it as a virtue to allow variety in party composition, so that no one class is strictly essential. James Raggi's Flame Princess rules are built on the former assumption, though in practice his improvements on Basic D&D has more to do with the profile of ability across level; making only fighters improve in hitting, and making specialists much more viable in what they do early on than rogues, for example.

Recent experiences in play have shown me how dependent on the prophet class and its healing abilities my parties are. This is a little embarrassing because prophets are supposed to be rare, semi-outcast members of society. The urge is to stick in another class that can heal; thus, recently, the druid. I don't think every class should do everything, but I do think everything should be doable by a choice of two or more classes. Call it the two-class rule.

So here is the chart for my 52 Pages system, and more info can be found on each class under this search. The system assumes that characters' hit points are more like "heroic courage and luck" that gets worn down in combat, so "healing" hit points is actually a matter of giving a sufficient pep talk.

With this in mind, two classes beyond the prophet and druid might qualify for HP restorative powers. One - a commander-type based on the fighter, kind of a secular paladin.

The other ... I shudder to mention it ... is the bard. Now, I want to have this class in my game, it makes a lot of sense, but please let it be anything but the bard, or the other awful alternatives (minstrel? gleeman?). Seriously, I would rather have a jester class doing this than a bard, and that's saying something.

D&D Next: Two Worlds of Gaming

Slowly through conversations with players I'm getting a feeling for what happened in gaming around 1980 and around 1995, and how the legacies of those watersheds affect people's reactions to game innovations now.

Neither of these dates is exact, but they each represent a middle point of a process of change. There are plenty of intentional throwbacks after, some unintentional look-forwards before. But in general, games have evolved (devolved?) toward:

I have to roll a 6 to get out? What?

Low death, low frustration. Before 1980 or thereabouts frustration was an acceptable part of a game's modeling of life. This was just as true for computer games, as board games, as role-playing. OD&D's 1 hit point characters = Nethack's play-until-character-death = Dungeonquest's brutal, random death at every corner. "Stay in jail" mechanics were OK in board games back then, too (Source of the Nile, Mystic Wood...).

The AD&D Dungeon Master's Guide removed much of death's sting through a "death's door" rule covering survival from zero and negative hit points (not at all evident in the earlier Player's Handbook). Save-or-die poison took longer to go, being finally purged for good in 3rd Edition. Computer games have shown a similar exorcism of frustration, moving from no saving of game state, to save points, to continual saving. State-of-the art design today for many casual games even gives progress for failed attempts, so that eventually the game will be defeated. Time is the commodity here, as always, but the message is "you didn't succeed" not "you failed."

"Activate Mrs. White's Rolling Pin Flurry!"

Character powers for all. This broadly refers to things a character can do distinct from other characters, but more specifically combat moves that can be customized. The move toward this feature in 3rd and 4th editions usually gets blamed on computer gaming, but plenty of computer games in the 1980's and 90's used the standard D&D paradigm, where classes without spells only get to bash, shoot and hack using the standard combat procedure.

The origin of feats and the like in tabletop gaming can probably be traced to the prevalence of special character moves in fighting console games, and to increasing modularity and customization in the gaming world under the influence of collectible card games.

On to the present day.

D&D Next, as much as it draws from the old school, also has strong representation from these two concerns in the system. It's obvious that hit points are numerous, the margin of safety for low-level characters is great, and healing remains as available as in 4th edition. It's also obvious that classes, races, and "paths" each come with an addition to a suite of powers that can leave a 3rd level character looking at 3-5 different combat traits and moves.

And now my confession: My campaign players expect these features to some extent, steeped as they are in MMOs. And so on both counts I have gone about half the way that D&D Next appears to have done. I used a "death & dismemberment" system at 0 hit points or below that is scary, but in practice merciful. Save or die poison is a bridge we haven't had to cross, but I'm inclined to offer multiple saves on poison (one to be seriously incapacitated and one to die.) Meanwhile, class powers and fighting feats in my game are running about half the count of D&D Next, but still give enough of a sense of things to do.

To put it shortly: In order to compete with my homebrew, D&D 5th Edition is going to have to give me the option to cut down on the bones it throws to millennium-era gaming, and let me run a game at my desired level of "old world charm and new world convenience."

Iron God: Taking Secret Doors Beyond the Letter S

Level 2 of the Tomb of the Iron God module is the actual burial place of the corpses consigned there, and it has more secret doors than a season of Scooby Doo cartoons. There are other things that aren't ideal about it and I tried to fix ... which I'll get to in the last post of this review. But the secret doors were the toughest thing to work out on the fly.

(This post is not as spoilery as the last one, so no cut.)

Anyway, we are not talking about dwarven or magical workmanship here. I ended up deciding that most of the doors could be seen; either in passing with a d6 skill check if you're on the same side as the door (exploration movement is slow going as it is, and most party members have a 1 or 2 in 6 chance), or automatically if you're giving that area special attention. No, you cannot give every area special attention - or at least I'm going to make that process player-boring, and eventually you as a player are going to make the choice that it's better to die a hero than live an obsessive grind. But an exception can be made for the niche-lined walls that parties have an interest in searching for loot.
 
The secret doors show as outlines of fine grooves in the limestone corridor walls. In the burial chambers, they appear as seams down the middle of a row of stacked niches, the halves sliding back both ways when opened. Niche doors usually open by pressing a button at the back of one of the niches, and these buttons also open the corridor doors, though they may be some distance away from their chosen door. Part of my map notes involved working out which niche button would open which secret door. There was a good moment when I combined one of those buttons with a poison needle trap that was already in the key. "Okay, you push the button in ... with your finger ..." Well, the dwarf had a great save and made it, but from then on, used his axe handle.

What was hard was squaring the many, many secret doors dividing the level into sections with the appearance of wandering skeletons and the need of nonintelligent monsters to rove around the dungeon. Some of it I just handwaved, other times there would be floor plates on the inside of a secret area that could let skeletons out. The players quickly grew paranoid and started whispering about a "skeleton factory."

As with pit traps, it's all-to-easy to answer "Why did the monks put so many secret doors in their burial ground?" with "Because it's a dungeon, duh!"  But if I had it to do over again I would have reduced the number of S'es on that level. Make some of them plastered-over concealed doors, which is more plausible if you're talking about a burial ground where some sections eventually get full. Just erase the unnecessary ones, like the ones that represent only a shortcut between two areas you can reach anyway, or the ones that hide one down-stair area from another.

This obsession with secret doors - on every level! by the half-dozen or more! - has apparently left its mark upon a whole generation. You see, as D&D players grow up, they get certain ideas about how their family lairs should be constructed ....

High-Level Combat: Designing Feats

In the Wizards editions of D&D, much is made of the tactical depth that the plethora of feats and skills gives to combat. This is part of a conscious decision to make fighter-types as interesting to play as classes that must make tactical and strategic choices among spells and skills.

I say "Bah to that!" and "Yay to dumb fighters!" Well, not exactly dumb; even when doing little more than tracking positioning in a fight, knowing how to use that can mean the difference between life and death. But in my view of the class archetypes, fighters are intentionally the straightforward ones. They don't have to make as many hard choices that are tied in with their class powers, and that is a good thing that serves the diverse needs of players.

Of course, this doesn't mean that playing a fighter has to be boring. In my house rules, awesome and interesting things happen to fighters without any choice. I use Arneson's "chop till you drop" rule to grant extra attacks when an enemy falls, as well as compensating low damage rolls  by having them inflict a critical effect on the enemy with some weapons (feats of force) and a fumble on the enemy with others (feats of finesse; I now reward low damage for both types of weapons, unlike the linked rule). Over some 8 sessions of play, these rules have proved to be fun and empowering without slowing down play.

It's the rogue - not just thieves in my view, but all kinds of light fighter - that I envision as using optional feats more effectively. But even for this goal, math-crunching can really take away from immersion in the game. An all-out attack where you get +2 to hit and -2 to armor class, for instance, can be wildly more effective depending on the relative to-hit ability and armor class of yourself and your opponent. Against someone you're hitting only on a 19, that doubles your effective damage. But if you're hitting on an 11 and your opponent hits only on a 19, that gives you only 20% increase in potential damage while doubling your opponent's. I want you to use your Piercing Stab not because you have calculated it increases your Actuarial Expected Damage Coefficient (AEDC) by 24%, but because you observe that the plates of the dragon are thick and scaly.


Which feats to take should mainly be a strategic, character-level rather than tactical choice. It should be back-loaded onto higher levels, so it doesn't encumber character creation. Also, the mathy figgering-out should be done by the rules designer rather than the player, trying to keep things as balanced as possible. For instance, do you want each feat on average to give the equivalent of +2 to hit (MATH CORRECTED)? This means, assuming a 50% chance to hit the typical opponent, the feat improves the chance by 20% (50% / 50%); 20% of the average damage per round, 2.5, is a half point of damage per round. It also means that an extra chance to hit (including gaining initiative or opponents being incapacitated from a critical hit effect) is worth 2.5 damage, so to be balanced it should happen about 1/5 (20%) of the time. This is roughly in line with the 1/6 to 1/4 chance of a crit/fumble my "feats of force and finesse" rule gives.

Defensive bonuses are harder to balance out, depending as they do on the relative hit probabilities and hit point totals involved. Anyway, I'm ruling these out as feats because a straight AC bonus prolongs fights, and this is anathema to our goal of making them more fun. Any defense will come as a side-effect of feats such as being able to disengage from combat without penalty, or maneuver into a position where fewer foes can hit. I'm also not ruling out hit point recovery mechanisms mirroring the bonus damage amount (about 1 hp every other combat round), making the most of the concept of hit points as player confidence and morale.

Next: OK, OK, some actual feats.

Does Combat Feel Different Across Levels? And Is That Good?

A parody of adventure game design - almost like the old Dungeon! game, or the present-day Munchkin game - would base combat and skill outcomes on a die roll, factoring in the character's level, the challenge's level, and different outcomes for different end numbers.

Well, I say parody, but such a super-abstracted combat actually shows up in other games that emphasize exploration and interaction over the minutiae of combat, starting with Chainmail, proceeding through Tunnels and Trolls, and following through in a myriad narrative-oriented games that keep a long arm's length from wargaming.

And this is how 3rd edition D&D handles skills. As a result, using skills at high levels feels just like using them at low levels. At level 10 you are picking the super-titanium-elder-dwarven-riddle-lock with exactly the same odds and procedure that you used to pick the rusty old peasant lock at level 1.

But in all versions of D&D, combat at high levels works qualitatively differently than combat at low levels. Skipping over the math and simulation aspect, the executive summary is that low level combat is short and deadly, high level combat is long and grindy, and it relies heavily on magic to keep it from being even more long and grindy. This is because:

• Monster chances "to hit" pretty much keep up with frontline PC armor class, assuming steady acquisition of magic shields and armor
• Monster damage rises more slowly than PC hit points (example, a 1st level fighter vs an orc compared to an 8th level fighter vs. a hill giant; the giant does twice as much damage but the fighter has 8 times as many hit points)
• PC chances "to hit" rise more quickly than monster AC, especially factoring in magic weapons, as shown here.
• PC combat damage rises much more slowly than monster hit points (example, from 1st to 8th level a fighter might get at best +3 on his or her average-5.5-damage sword from magic, so damage increases 55% while monster HP increase by a factor of 800% or so.)
• As shown here, the only thing keeping pace with monster HP is damage from spells. Additionally, spells like sleep, hold, charm, polymorph other and so forth have a chance to speed up combat by taking opponents entirely out of the combat. 

Two things are clear from this.

First, without a wizard-type, high-level combats risk being lengthy grinds. This is doubly true of the situation where higher-level characters face a horde of low-level monsters. I remember in my first year of playing AD&D how the party - by then third level and equipped with magical gear - faced a conga line of about 50 kobolds coming through a 5' doorway. Needing a 20 to hit our frontline fighter, the kobolds provided possibly the least fun ever seen in a D&D combat, even applying the multiple fighter attacks rule (which presumably had been put into AD&D precisely to speed up this kind of situation).

Second, while the PC chances to hit rising are more or less balanced by monster damage rising, the leeway available before character death leads to another source of qualitative differences. Low-level combats are deadly because a lucky monster damage roll can knock characters down to zero HP easily. In high-level fights, standard damage is more easily dealt with over the course of many rounds, and deadliness comes from save-or-die effects and the occasional massive damage source like dragon breath.

Both of these qualitative differences, I think, work to the detriment of higher-level combat, taking the basic workings of the system from quick, deadly and exciting to long, slow and predictable. Combat needs to be goosed up with blast 'em spells or deadly monster powers, and this means that when you don't have those elements it turns into a slog.

Next: solutions for high-level combat, old and new.

The Graph Paper Architects

"One final enigma surrounding the architects of the Mythic Underworld: their complexes, catacombs, tombs, strongholds and hypogea were all built to more or less completely fill a 340 by 440 foot rectangle. Perhaps akin to the mining extravagance of their corridors built to a width of either 10 or 20 feet, it is clear that some strange concern with geomancy impelled the rigid proportion of their constructions."

One of many dungeon maps by Tim Hartin.

In the graph-paper shaped dungeon, after a while players figure out to look in corners for secret doors, and have a pretty good idea when they've explored the whole level. And yet so many megadungeons - even less ambitious adventures - are mapped this way. Castle of the Mad Archmage jams four sheets together for each level. Stonehell uses the one page dungeon template, a square about two-thirds the size of the standard graph paper, but otherwise does the same, and makes each square a stand-alone section. World's Largest Dungeon uses 16 blocky maps, each about twice as big as the standard graph paper but still uniformly rectangular.

Standard maps at first glance seem to solve more problems for the DM than the player. The DM doesn't need to wrestle with large or odd-sized pieces of paper. Players, though, don't know exactly where they enter on the graph paper, so even in the most orderly dungeons their maps will tend to go off the edge.

But in actual play, how much usefulness does the graph paper dungeon add? After all, the DM doesn't usually need to see all of the 440 x 340 foot area - only the immediate area of 5-10 rooms is useful to a session of play.

This argues for the modular megadungeon to be portioned out in smaller sections, such as Talysman recently demonstrates. The overall strategic map can then be as irregular as it wants to be. In fact, the fitting together of the modules in a whole ten-level dungeon can be easily mapped on a 1 square to 100' scale, where each section, approximately 7x7 or 6x8 squares, is a separate level.

A related issue on a smaller scale is the way of dungeon geomorphs, as cool as they are, to create an overly dense and connected "wallpaper" of rooms and passages. The standard format seems to be a 10' x 10' square with exits in the middle of each 5' length of side, 8 in all.  This has been the aesthetic barrier to my using geomorphs, up until now.

One solution would be to just arbitrarily say that 1 in 3 geomorph squares is blank. You deal them face-down from your stack on a d6 roll of 1 or 2, and erase or block off the connecting doors and passages.

Another idea is ... well, I'll show you what I mean next time.

Why Does A Dungeon Look Like A Dungeon?

Everywhere from the Zork logo to Hirst Arts adventure scenery, the standard architecture of a "dungeon" - meaning, underground medieval fantasy adventure setting - is:

• walls of regular or irregular gray stone blocks, anywhere from shoebox size to two or three times that
• wooden doors fitted with iron hinges and reinforcing bars
• floors of gray flagstone
• torches in wall sconces

More on CRPG dungeons

It's shorthand for our conception of what a fiendish antique maze stretching many, many levels underground would look like. But what does it really represent?

Certainly, high medieval castle architecture used those big stone blocks for aboveground walls, as we see from Alnwick Castle in Northumberland and similar fortresses. When delving into earth, it might also be a good idea to reinforce the walls with stone, right? Except if the idea is just to reinforce and insulate and make it look neat and nice, then your typical medieval cellar made do with much smaller, brick-like stonework, as seen in this undercroft in Norwich.

Or the stonework might be plastered over, or it might be made of bricks, or packed earth might suffice.

Another problem for the "dungeon look" comes when, digging down, you hit bedrock. Granted, in agricultural plains this might not happen for a hundred feet or so, but castles, wizards' towers, and other such dungeon-toppers tend to be built on hills with rock not far below. In that case the dungeon tunnels are best dug directly into the rock, faced and decorated by planing the rock itself. There's no need to emulate the look of Garden State Brickface by carving fake mortar crannies into the stone.

Could it just be that the big-block idea comes from artists who wanted to draw fewer lines in their medieval settings? Maybe, but there's another reason an architect might want to divert those hard-to-move, yard-long stone blocks from building the important castle walls to facing the underground cellars.

If you are keeping prisoners down there - you know, in your dungeon - then it's important to make the blocks big, so it's hard to get past them to the diggable clay beyond.

So the answer to "Why does a dungeon look like a dungeon?" may very well be: "Because it's a dungeon..."