Author: azabaro

Odd Attribute Modifiers in d20 games

When Wizards of the Coast published the 3rd edition of Dungeons and Dragons (and thus set down the foundation for d20 games) they adopted a new, standardized, open-ended schedule of attribute modifiers (as compared to the fixed-range and sometimes attribute-specific schedules from earlier editions). In order to be open-ended the modifiers had to be defined by a formula – mathematically it’s

Floor(attribute/2) – 5

In plainer language, it’s +1 for every 2 points above 10, -1 for every 2 points below 11. Simple enough, but there’s a drawback: the “every 2 points” bit (necessary to keep attribute modifiers from becoming completely overwhelming differences between characters, overriding most other differences in ability) means that odd-valued attributes are basically useless – a 13 is worth exactly as much as a 12, a 5 is just as bad as a 4. These values seem like vestigial bits from prior editions, and while the designers of 3E tried to ameliorate that by giving them some utility (defining attribute requirements for feats in terms of odd numbers) they weren’t very successful – for example, substandard attribute values don’t end up differentiated, and feats a player wasn’t considering can’t make their character’s odd attribute values meaningful.

To fix this without upsetting the apple cart (all the other design work in 3E/d20), we’d really like to give these odd attribute values a modifier that’s halfway between that of the even values that bookend them, so for example, a 12 => +1 and a 14 => +2, so a 13 => +1.5. Fine, except what does that mean? How do you add half a point to a roll? You can say something about using it as a tie-breaker, but that won’t come up very often, and even when it does there’s no guarantee the opposition won’t have an odd attribute and the accompanying tie-breaking modifier on their end of things.

How about this: let’s get to that “.5” modifier statistically. That is, instead of adding half a point to a bunch of rolls, let’s add 1 point about half the time. We don’t even need to do any extra work for this – the rule could be

When your die roll is an odd number, add 1 to it for every odd-valued attribute that would modify the roll.

As an example let’s look at a Paladin making a Fortitude save. They have a Con of 13 (+1) and a Charisma of 15 (+2), and because they’re a Paladin their Charisma bonus applies to all their saves. When they roll a Fortitude save, they’ll add +3 when their roll is even, and +5 when the number they roll is odd. On a roll based on Charisma alone (Diplomacy, perhaps) the Paladin’s attribute bonus will be +2 half the time and +3 the other half, averaging out to +2.5. This makes their 15 Charisma worth exactly half the difference between a Charisma of 16 and a Charisma of 14, and doesn’t require us to rewrite the rest of the game.

Posted on by azabaro 0

Problems with setting skill system numbers in D&D

D&D has never had a skill system that worked numerically as well as it should. Oh, there were plenty that good DMs could make it work by fudging numbers and outright ignoring the rules, but out of the box it’s always kind of a mess. There’s a reason for this – a mathematical tension that hasn’t been resolved.

Here are the conditions I see D&D skill systems attempting to satisfy:

  1. Resolution is handled by a d20, or some random number generator with a comparable range (e.g. 3d6)
  2. A 1st level character is basically a normal human, or at least close enough to one that they ought to be challenged by the same sort of things (that is, climbing a wall or even a rope shouldn’t necessarily be significantly easier just by virtue of being a 1st level PC instead of a level-0 or unleveled NPC)
  3. The contribution to a character’s skill from accomplishments (leveling in a relevant class, advancing the skill itself somehow) should generally dwarf the effects of ability scores (which are generally set at character creation), at least by a few levels in.
  4. Many of the classic D&D spells are assumed to become available at more-or-less the same levels as they traditionally have.

Condition 4 brings in the most obvious problems: many skills are almost immediately obviated by spells (when available); for example, traditionally the ability to Climb suffers in comparison to the 1st-level Spider Climb, the 2nd-level Levitate, and the 3rd-level Fly (with the latter two spells offering capabilities that no amount of mundane climbing generally permits). This can’t easily be brushed aside by referring to a limited spell payload or even by house-ruling away certain spells – there really is a transition around 3rd level spells (so generally around 5th level) where characters tend to break away from normal humanity in terms of their capabilities. And that’s fine, except conditions 1-3 tend to keep these borderline superhumans very much on par with normal humans, except when magic comes into play.

Ways to address this? Well…

  • We can shrink the random number range (as well as target number for skill checks). An easy test for this is how you want arm-wrestling contests to work – if you handle them as a single opposed set of ability checks, how much of a Strength difference does it take before the resolution system becomes “don’t even bother rolling”? If an average-strength character (Str 10) takes on the strongest human possible (Str 18), how often should the Str 10 guy win?
    • Ability checks based on the d20 system give the contest to that guy 30% of the time – so about 1/3 of the time.
    • If you use 1d10 for these checks instead it happens 15% of the time
    • on a 1d6 resolution system less than 3% (1/36 times)
    • and on a 1d4 it never happens (to even have a chance the weaker guy would need at least a Str of 14, at which point their odds would be 1/16 – about 6%).
    • You might be tempted to use several small dice to give a curved distribution (for example, 3d6 instead of 1d20). That does make outcomes like Str 10 beating Str 18 less likely…but since the range of the distribution is pretty large (16 points) they’re still possible – indeed, even a Str 3 character can sometime beat the Str 18 character at arm wrestling.
  • We can negate conditions 2 and 3 just assume that PCs are substantially better at everything than normal folks. This could operate gradually, like 4E’s bonus of 1/2 character level to rolls like skill and ability checks, or could only kick in at certain tiers (so you get a +2 or something just for being a PC, and another big kick around 5th level when wizard-types can fly, and again around 9th level when death goes from existential terror to inconvenience, etc.). If the boost is a simple level-based (or tier-based) bonus it will tend to mean that PCs are hypercompetent – that they don’t have any real weaknesses compared to normal humans (not necessarily a flaw in a game like D&D); it can also reduce skill differentiation between PCs to being mostly cosmetic (if this is a flat bonus; if it instead depends on whether or not the character has some training in the skill we might just end up with a hypercompetent PC standing next to a helpless PC)
  • We could, I suppose, negate a hidden assumption built in above – that modifiers accumulate 1 point at a time. If instead they come in large chunks (say, 5 points where we currently give 1) we could get the benefits of using a small die while continuing to actually use 1d20 for resolution (we might even be able to directly add an attribute score directly to skill and ability checks rather than using an attribute modifier). The downside is that this does mean dealing with larger numbers – a lot more 2-digit addition, which will slow things down if the skill system sees any real use.

I suspect smaller random number ranges are the easiest way to address this, with the smallest side effects. In fact, I specifically think using 1d4 for the skill check random number generator is the way to go – it clearly illustrates the idea that the same character is capable of entirely different things at level 5 than at level 1 (which reflects the differences that already exist due to condition 4).

Posted on by azabaro 0

Runnin’ Bread

A loaf of Runnin’ Bread (possibly a corruption of Run-in Bread) appears to be a typical loaf of a mundane bread (e.g. white, wheat, sourdough, rye, etc. – though white is the most common variant). Although typically found in loaf form, up to a dozen of the first slices of the bread (1d12) will carry its enchantment.

When the loaf or one of its enchanted slices is held by the mouth, the character (and their party) are empowered to run at twice their normal movement rate; however during this movement their chance of random encounters and of being and inflicting surprise are also doubled. If the mouth-borne bread was toasted first these values are instead trebled.

Posted on by azabaro 0

Basic magic item creation rules

D&D 3rd edition introduced a well-defined process for creating magic items. It seemed like a good idea – lots of folks wanted something beyond the ask-your-DM-and-prepare-to-quest-like-crazy-to-make-a-shield+3 guidelines of 1st and 2nd edition, and I was certainly among them. Too bad none of us – players or designers – anticipated the effects of the standardized system we wanted or got: magic items being taken for granted (because they could almost all be produced reliably), severe disruptions to the way people played the game (because certain items could be produced quite cheaply, thus altering assumptions about the power available to characters, parties, and NPCs), and mundane wealth being devalued (because both could be acquired with gold, any mundane object was just some fraction of a more powerful or directly useful magic item). So we probably want to dump the 3E item creation rules. But that still leaves us wanting some sort of replacement. What form should that take?

Continue reading →

Posted on by azabaro 0

Settlement Sizes for D&D

When describing settled areas in D&D (and maybe fantasy RPGs and medieval settings in general), I often see terms like Hamlet, Thorp, Village, Town, and City thrown around without a clear distinction between them – and if Wikipedia is to be believed there is no clear dividing line.

So I came up with the following qualitative descriptions of settlements sizes, something that would hopefully allow players and DMs to look at a term and understand what that group of habitations is capable of doing easily or at all.

  • Village (including Hamlets, Thorps) – communities too small to support significant specialized labor. To some degree all households have to participate directly in maintaining their own lives – farming, fishing, hunting, or animal husbandry (though the degree may vary between households), maintaining dwelling structures, making their own clothes, etc. Any specialized laborer (a priest or doctor) is likely to operate in that capacity part-time at best, unless they have some ability to outsource that labor (slaves, golems, etc.)
  • Towns – Have significant specialization of labor and the consequent need for markets to exchange products and services, but does not have enough economic activity to sustain standing markets; as a result the town will generally have market days (depending on the economic status of the town these could be once per month, fortnight, or week, perhaps even twice a week – more often than that is more of a standing market). Market days may be associated with minor festivals or other occasions that bring people together, such as church days.
  • Cities have enough sustained economic activity to require standing markets (markets that operate almost all the time and have stable locations). They have significant labor specialization and large enough concentrations of wealth that there will be noticeable expenditures on various types of overhead: infrastructure such as roads, bridges, aqueducts, and arguably even temples (in settings with active gods), as well as defenses such as standing armies and town guards, and fortified city walls (smaller settlements can have walls of course but the wealth of cities mean they almost certainly need them, and will tend to have better walls than any village). Because of their investment in military strength and infrastructure and the trade advantages of standing markets, cities will tend to be regional power centers – if they are surrounded by smaller settlements or control access to a vital resource (sea routes from a river, for example) they may even be power brokers.
Posted on by azabaro 0

Why Watching Evangelion 3 at a Convention is Frustrating

I attended Fanime over the Memorial Day weekend; aside from the registration process (clearly designed for a con half the size) it seemed to run well enough, and I’d certainly consider returning for a 3rd year. One of the events I attended was the screening of Evangelion 3.0 or 3.33 or whatever the official number is supposed to be – “You Can [Not] Redo” is the subtitle. Fansubbed, of course – Fanime actually has a video room openly labelled as the Fansub Room, right down a short hallway from the Industry Video Room. The movie itself was fine – I liked it when I saw it with my (small) anime club, and I still liked it. But there were a few issues watching it at the con:

  • The video was prefaced by the Giant God Warrior Appears in Tokyo short from the Japanese Blu-ray release. However, that hadn’t been subbed along with the movie, and nobody in the room (including the person in charge of running the show) seemed to know what was going on; that lead to fiddling with movie settings early on trying to enable non-existent subs.
  • We were right across the hall from the Karaoke room, so there really wasn’t any such thing as a quiet moment (but you could at least tell when the movie was aiming for that, and everything important was audible)
  • The movie wasn’t in a huge video room, which means it started off packed and there wasn’t space for a light trap at the entrance; people kept leaving only for more people to file in and take up the space. I suspect this only bothered those of us at the back of the room.

All these nits aside, I did have one significant complaint about watching Eva 3 with this crowd. See, when Shinji and Kaworu have their meet cute and their subsequent interactions, the audience went crazy for all the subtext there. Which is great: I remember the premier of the Gundam Wing movie at AX 2000, and fangirl screaming at that one brought enough energy to make up for a movie that was otherwise not worth the time. The Eva audience did even better in that cheering for innuendo was apparently just as popular with men as women. And they went wild for every little bit of it, even stretching to find innuendo where I really don’t think any was intended.

And – seriously – that’s all fantastic; it’s great that the fans were such an active audience during those parts. But that’s also why it’s such a bummer that they weren’t as active throughout the rest of the movie – not just being quiet (Eva 3 is almost intensely hushed at points, so that’s actually what I’d hope for), but just…passive. They didn’t seem to be trying to read into the film, and that impression was only reinforced when the showing let out and the collective impression I heard afterward was “?”. Of course, to some degree that movies have set themselves up for this – I heard people opining on Shinji as action-movie-hero based on movie 2, and I spoke to at least one person who hadn’t watched the first 2 movies on the mistaken impression that they were just straightforward remakes of an early part of the TV show (reasonably accurate for movie 1, dead wrong for movie 2). But even setting those aside, even when we discard the significance of Eva otaku details like the track numbers as played on the Ikari family Walkman, I feel there should’ve been some engagement with the images in that movie. Because it communicates a lot visually (seriously, the shot of Shinji curled up on that bed, back up against the wall; huddled on a white square in a sea of red, his Walkman the only barrier between himself and reality…barriers everywhere throughout the movie…Kaworu descending to the scene of near-3rd impact dressed normally as Shinji follows tentatively in some sort of Apollo-era spacesuit…Misato and Gendo wearing the same strange wraparound glasses…), so much so that while I understand people coming out of that movie confused I don’t understand people who react to the movie as merely incomprehensible, as if it wasn’t stuffed full of undeniable (if cryptic) visual messaging.

So that’s the disappointing part I guess: a movie that begs for interpretation only seems to get it from the fans in the scenes that already meshes with their expectations; as soon as it goes off the reservation they appear to lose focus. That’s probably related to sequelitis – once we as an audience have seen a movie we develop expectations of it, unconsciously taking at least partial ownership of it; and any piece which fails to conform to our internal fanfiction outline is wrong or suspect or just uninteresting. It’s the sort of problem that makes me wish I could show the Evangelion movies to people fresh, without having them watch the TV series…but of course the movies are significantly meant as reactions to the original show and don’t really work the same if you don’t know how all that went down well enough to be struck by the differences. “It’s the curse of Eva,” I suppose.

Posted on by azabaro 0

The Shears of Atropos

These right-handed scissors are a leaden gray; their size varies on the cosmological stature of their wielder: in the hands of one of the Moirai they appear to be of a normal size for cutting threads (though perhaps a bit large for Clotho and Lachesis), in the hands of a mere god they appear to be over large, as though a child were attempting to use an adult’s tool. For demigods and mortals of such power as to be their equals (archmages etc.) they’re more like gardening shears in a child’s hands, and must be used two-handed. Lesser mortals may also use them two-handed; in this case each blade appears to be as long as a sword.

Regardless of the bearer’s prowess, the known power of these frightful instruments are usable in full. They include the ability to

  • bring death by severing the thread of an individual’s life. In this respect, the shears act as a vorpal two-handed sword, except that the death thus inflicted is not necessarily decapitative, and creatures which can survive without their heads are in no way immune to this effect. For the purposes of attacking the Shears not only gain the +3 attack bonus typical of vorpal swords (representing their ability to cut a creature’s thread with directly cutting the creature’s body), but also the ability to ignore any armor the creature is wearing (because they can easily cut through it – see below). This ability is presumed to be ineffective against the Fates, and those who are wise know better than to speculate whether gods can be thus cast down…at least out loud.
  • end even magic, by cutting a spell at the center of its effect or origin point; this has the effect of a Disjunction (an automatically successful Dispel Magic) on the targeted spell only.
  • cut through all material objects as though they were cloth. Note that this doesn’t allow large volumes to be cut in any single motion – although it’s possible to tunnel a path through a mountain in this way, it will still take a long time.

In addition, the Shears are rumored to have at least some of the following abilities, in addition to a number of other, miscellaneous abilities:

  1. A dim, perhaps merely empathic intelligence coupled with a special purpose: to destroy undead (likely by either disintegrating or disrupting them)
  2. The ability to operate against distinct individuals, objects, or magical effects at a distance assuming the bearer can directly cut a proxy object with a sufficiently sympathetic connection to the target.
  3. The power to foul water, making it unfit for any sort of consumption as it becomes polluted, metallic, and reddish in color, reminiscent of rust and of blood (treat this as a reversed Purify Food and Drink).
  4. Some capacity to sever even the relationships between individuals.

The side effects and other drawbacks of the Shears of Atropos are generally unknown, though it is widely assumed that the Fates (and Atropos in particular) wants them back; and as ending lives is generally within her purview anyway she probably won’t be shy about leaving a trail of bodies – or shattered empires – behind her as she attempts to reclaim her tools.

Posted on by azabaro 0

Removing Level Limits (But Not Their Benefits)

Older editions of D&D limit what classes non-humans can take and limit how far they can progress in those classes. The idea was that if you didn’t have these in place, demi-human characters would quickly overwhelm human ones, both because they have advantages at first level and because significantly longer lifespans meant that they should be able to continue advancing well after their human contemporaries have died, causing humans of significant level to be completely outnumbered over time. Of course, there were still players interested in demi-human characters and as far as I can tell these level limits were never that popular. So at first glance we’re left with a dilemma: retain level limits and the dissatisfaction they cause, or remove them completely and subordinate humans to other player character races.

Now, there’s no shortage of other options – perhaps the best-known is 3rd edition’s decision to give humans bonuses at first level to make them more competitive with demi-humans. That still didn’t solve the problem of elves continuing to level up at age 500, though. I think we can solve all that easily enough, though (and as a side-benefit the rules are short). Note that I came up with these with OD&D in mind, but I don’t see any reason they wouldn’t work with editions as recent as 2E – basically, anything with a notion of prime requisites (though in editions where race = class you’ll have to work up a progression beyond the listed levels, or let non-human characters switch classes on that – maybe I can get to that later).

The central idea here is that characters can be of any class regardless of race, and there’s no preset level limit for a character based on race; advancement is only limited by their ability to continue gaining XP.

  1. Typical demi-humans have a 10% XP penalty (races with greater abilities than the classic dwarf/elf/halfling trio may have a larger penalty).
  2. Any class outside those “allowed” (i.e. typical) for the character’s race imposes an additional 10% penalty to XP applied toward that class.
  3. Once a character of any race has lived longer than a human can (100 years, say) they face an additional 80% XP penalty.
  4. XP modifiers for high or low prime requisites apply as usual.
  5. Multiple XP modifier percentages can apply at the same time, and are added up rather than multiplied. So an Elf over 100 years old faces an XP penalty of 10% (elf) + 80% (outside human lifespan) = 90% in general; for classes other than Fighter or Magic-User (only allowed Elf classes in OD&D) it will be 100% before adjusting for prime requisite scores.

The consequences of this are that demi-humans may still be slightly superior to humans in their earliest adventures (when simple class differences like Fighter vs. Magic-User means that balance isn’t too precise in the first place, and at any rate low HP totals place everyone pretty close to an early grave) and pay for their advantages over time with slightly slowed advancement. Characters that want to continue advancing after their human colleagues have aged out will best do so by sticking to the career paths archetypal for their race and those for which they have any particular talent (XP bonus due to high prime requisite scores); even then their advancement will be sufficiently slowed that the world shouldn’t be overrun by super-elves or what have you.

Posted on by azabaro 0

Don’t Track Charges

Charged items are an interesting sort of PC find. Because they have a cost, there’s generally some real benefit to using them, but because there’s a cost there’s also some motivation not to use them for every little thing. Also, because charges are generally non-renewable even unbalancing items tend to go away eventually. And they’re usually a lot more interesting than “+3 to some roll or other”. This all seems like a win except for one small detail: keeping track of charges is a pain.

It’s a pain because it’s easy to forget to do it, and because there’s a question of who should be tracking (players? do they know the exact number of charges?). It’s also a pain because different numbers of charges remaining can cause real changes in the way players use the items: a wand with 50 charges left feels almost inexhaustible; one with 3 charges left may be hoarded, even at the cost of PCs’ lives. To some degree this reflects the should-I-use-this-or-not interesting choice mentioned above, but it generally fails to emulate the way a character would act when feeling threatened, the tendency to keep pulling the trigger until ammo runs out. And of course it’s impossible for the PC to run out of ammo without the player knowing it’s about to happen, unless the DM is the one actually tracking charges remaining (as if the DM needs more numbers to juggle!)

Solve this by not tracking a definite number of charges, but by assigning each item a percentage that the next use will be its last. The percentage can be calculated by dividing 100% by the number of charges you’d normally assign the item, so for a wand with 50 charges that’s 100%/50 = 2%; for a staff with 8 charges it’s 100%/8 = 12% (actually 12.5, but let’s round down). There are of course a few items that should still have charges tracked explicitly, generally those in which charges are iconic and small in number (e.g. a Ring of Three Wishes).

The way this looks in play would be that a player would have their character use the item, then roll against this depletion chance; as long as they roll over the percentage the item remains usable (note that the percentage doesn’t change as the item is used); otherwise the item is used up. You can probably just tell players the depletion percentage for an item (or wait until they use something like Identify to get a sense of how charged it is), since this number doesn’t provide any definitive information on how many times it can be used – an item with a 4% depletion chance may fail after the first use, and one with a 25% depletion chance may be useful after a dozen invocations.

In short, this change reduces the amount of tracking that has to be done during the game, increases unpredictability, and tries to make players more likely to use items as charges dwindle rather than bypass them on character sheets until they’re forgotten.

Posted on by azabaro 0