OD&D Recap Redux

Having spent some time with OD&D and thinking about how I'd like to iron out some of the wrinkles (see below), I think I can express my overall philosophy pretty succinctly:

OD&D had a wonderfully strong kernel in the original white box set, but there were some clear oversights and dependencies that fairly cried out for correction and clarification in a work such as Supplement I. Unfortunately, when Gygax actually wrote Supplement I: Greyhawk he suffered in many ways from an overreaction (overly complicated ability modifiers, weapon details, monster damage, thief mechanics, etc.) Somewhere between the original "white box" and Supplement I there's a nearly perfect hypothetical game for my tastes.

Now, some might argue that this is precisely what the Holmes/Moldvay/Mentzer Basic D&D project provided, but the fact that it stripped out the key options for multiclassing combinations (human-only thieves, etc.) and high-level endgame (mass tactics, naval combat, castle-running), while keeping many of the awkward mechanics (table-based combat, poor figure scale, percent-based thief skills, etc.) are things that I cannot personally overlook.


OD&D Recap

Now that I've spent some significant time thinking about the OD&D rules (and writing up my house rules, below), I do think that OD&D provides perhaps the best basis for a hobbyist to use as a platform for their own, personalized version of the game. However, if you approach the "little white box" for the first time you may be surprised by a few things -- here are the most significant things I felt compelled to "fix" in my Original Edition Delta:

1. Abilities do not modify actions. Generally speaking, while OD&D has the classic 6 abilities, in the main the idea that they should be used to modify actions in the game had not yet been invented. The first 3 (Str, Int, Wis) do nothing except modify XP awards (respectively for the initial 3 classes: Ftr, M-U, Clr). The latter 3 (Con, Dex, Cha) give at most a +/-1 modifier to certain rolls (Con: hit points; Dex: missile fire; Cha: hireling loyalty).

This was rapidly changed by EGG in Supplement I: Greyhawk, where all of the abilities received some sort of action modifiers (identical to what we know from AD&D). However, it's a fair claim that EGG went off the deep end here, with extensive tablature lacking any core system for the DM to remember mentally. Every system since that time has responded by seeking to standardize the various ability modifiers (Moldvay Basic, 3E D&D, Castles & Crusades, my own OED, etc.)

2. Racial benefits are rather scattershot. The benefits accrued to different races (dwarf, elf, halfling) are a somewhat complicated melange of abilities in the OD&D player's book, references back to Chainmail, and variations on abilities indicated in the OD&D monster booklet. You have to hunt a pick and do some interpolation (as from Chainmail) to figure out exactly how these abilities should fit into your D&D game. Dwarf trap-finding, elven combat advantages, and halfling missile fire are not given any numbers in the OD&D books themselves. Furthermore, I feel that the lack of a core system created a proliferation of later add-on abilities, which became highly burdensome to track in AD&D or 3E D&D. The same can be said for the elven multi-class ability.

Again, Supplement I: Greyhawk felt compelled to quickly address these issues. Multi-classing was expanded and formalized (as seen in AD&D), not with complete success. Racial abilities were given additional detail (some in errata), again in ways that were not totally balanced. This is one of the things I massaged in OED; selecting from a limited number useful, relevant, and memorable racial abilities.

3. Weapon effects are not distinguished. In the classic OD&D system, no weapon is different from any other weapon; they all strike at the same rate, with the same to-hit chances, for the same 1d6 damage. If you want different weapon effects (first-strike, chance-to-hit), then you have to make use of the earlier Chainmail system (which was apparently the default assumption for OD&D).

At the risk of sounding repetitive, Supplement I: Greyhawk again saw this as a need clearly requiring redress, and provided a variant system giving different damage, to-hit chances, required space limits, and other details to the different weapon types. Personally, I judge this as a well-meaning project but with limited success; similar to the Greyhawk ability modifiers, this material was perhaps too cumbersome for convenient use (what with separate damage rates for small-versus-large targets, difficult-to-adjudicate space requirements, and an enormous table of weapon-vs-armor adjustment minutiae). With Supplement II this project further expanded into a complete hit-locations system which pretty much collapsed under its own weight. Almost all versions of D&D since have felt compelled to pare down these rules (no size differentiation; no weapon-vs-AC matrix, etc.) In OED I took the Greyhawk project as inspiration and again trimmed it down to its usable kernel.

4. Table-based functions are the rule. Now we get to points that EGG never distanced himself from. All forms of classic D&D (OD&D, AD&D, Holmes/Moldvay Basic, etc.) are largely table-based in adjudicating combat, saves, class hit dice, thief skills, etc. This creates a bit of a slow-down in play as the DM has to look up each of these cross-referenced items. It's additionally odd because in most cases there's a simple one-line formula that could be used to replace the entire table (for combat, it's d20 + fighter level/Hit Dice + AC >= 20, etc.)

Various camps did finally see this simpler formulation, and included things like THACO in the AD&D DMG, or the d20 core mechanic in 3E (albeit at the expense of every target of any action in the world requiring a DC listing). I am greatly in favor of the speed-up gained from a simple formulation, so of course that's what I use in all my games, including the OED.

5. Encumbrance is at the wrong granularity. I've railed about this for a long time, and it's a fundamental error that no version of D&D has ever corrected to date. In OD&D, EGG specified weights in coins (1/10 of a pound), so to find encumbrance you have to add up numbers in the hundreds and thousands of units -- ultimately it requires a calculator in hand, and it's highly cumbersome. 3E uses pounds, and it's little different (still adding scores or hundreds of units). Legions of us have simply ignored encumbrance for decades because of this aggravation.

What you really need is a system whose scale requires adding numbers up to, say, 10 or 20 at most, which we can all do immediately in our heads without a calculating device (on the same scale as the D&D system for class levels, ability scores, hit dice, damage, to-hit scores, saves, and so forth). Basically any granularity on the order of about 10-pound units would suffice. Most fortuitously, there is a real-life Imperial unit of the "stone" (14 pounds) which both satisfies this requirement perfectly, and at the same time provides additional in-world color. All you need is for the core rules to specify weights in "stone" and the whole encumbrance system becomes trivially easy to sum and adjudicate. (See previously here.)

6. Combat scale and sequence are unclear. Perhaps the most troubling oversight in OD&D, the exact relation of distance, time, and actions in a round lives in a hazy gray zone somewhere between Chainmail, OD&D, and AD&D. The OD&D distance scale is specified as 10 feet (a briefly-considered switch from Chainmail's 10 yards), which fails to physically fit miniature figures in the assessed space (resulting in the tortured explanation early in the AD&D DMG). The OD&D time scale is described as "fast and furious" at 1 round per minute, which is not really "fast and furious" at all (and again results in a labored and unconvincing justification at the start of the combat chapter in the AD&D DMG).

Here, the 3E D&D game finally made the required fix, clearly setting the distance scale at 1" = 5' (basically the same as miniature figure scale), and round length at a fraction of a minute each (6 seconds, whereas I prefer 10 seconds after some analysis of real-life archery speeds and the like). Now we finally have "fast and furious" combat, which was an admirable goal from the start. This I've carried into OED, along with additional clarifications gleaned from how the Chainmail and Swords & Spells combat sequences were intended to function.


Original Edition Delta

I've collected the various house rules discussed on this blog for OD&D, and compiled them under the title "Original Edition Delta" (currently referenced as version 0.3). You can download either a PDF or Open Document Text format (the latter with designer note sidebars) from the following link. Comments welcome:



Monte Cook's Dungeonaday.com

Monte Cook has started a subscription site called Dungeonaday.com, in which he plans to present a megadungeon, one new room each day. Today it was reviewed at Grognardia.com, which motivated me to likewise check out the free sample parts of the site.

I must admit that some of what I just read at Dungeonaday.com raised my hackles a bit. I think a decade after the design of 3E D&D, I see that Monte's need for systematization is perhaps enormously misguided. I sometimes half-jokingly call myself OCD, but I honestly think there's a compulsiveness to Monte's system-building which leaves me a bit stunned.

For starters, the very first line of the first sample room I saw (#6) had me sighing. "In this temple to an evil god, everyone has the barbarian's ability to rage..." Mostly discarding any consideration of thematics, the basic instinct seen here is to reach for a pre-written mechanic and just reference it. There's an aspect to the verbosity of 3E, that gets you handcuffed into looking for ways to avoid actually making up new mechanics, and this is one example. As enormous as is the text for this room, the primary mechanic here is actually not new, it's just a reference to something else in the game already.

Honestly, I think that's uninspiring game design. I remember seeing this at a few of the computer game companies I worked at. A crown-jewel example is the 3E proliferation of spells that raise your Int/Wis/Dex/Con/Cha... "Hey, there's a spell to raise Strength; just change the ability name and we can act like we created 5 new spells". Design by expanded systematization I definitely don't like.

This leads me to look warily at his "Dungeon Design Assumptions". Item #1 says this:

1. Things get more dangerous as you go deeper... 3rd Edition created a system that used Challenge Ratings to match relatively appropriate encounters to a given group of player characters (the key word being "relatively"). Matching monster toughness with PC toughness has always been in the game in one form or another, of course. But in dungeon design, this isn't that important, because the dungeon level dictates (or at least suggests) the difficulty of the encounters. Things too easy? Go down. Things getting pretty dicey? Go back up...

Sounds good, right? Well, my concern with Monte is that this italicised dictum will itself turn into yet another set of handcuffs. I recall purchasing Monte's Fane of the Demon God and at some point realizing that every encounter in the entire module was at exactly the same, fixed Encounter Level. My suspicion here is that Monte will be specifying an EL for every dungeon level, and get hung-up on every single room being the exact same EL throughout. If the "go up/go down" is meant to dependably let PCs choose their own EL, then he will have to do this, leading to predictability and a lack of surprises in the general sense of "is this really dangerous for us?"

In Item #2 he says, "In a standard campaign, the DM controls the level of challenge for the players. But in a dungeon like this, the players can choose to seek encounters that might be too challenging for them in order to get bigger rewards, or stay and face easy challenges for low rewards." It's a bit hard to swallow this characterization of a "standard campaign" for someone trying to assert an "old-school" stance on these matters.

In Item #5 he says, "The player characters are not the first adventurers to explore this place, and they won't be the last. As they explore, your PCs will find the remains of previous adventurers. They will hear about other parties coming to the dungeon to test their own mettle. They may even encounter them while delving into the depths themselves. This contributes to the dynamism of the dungeon environment."

In the broadest theory that might be nice, but the problem is that what it really references back to is Gygax's original campaigns where he actually had different competing PC groups meeting every night of the week, all racing to plunder the same dungeon mega-complex. I don't believe that you get the same effect when you attempt to mechanically recreate it with pre-placed NPCs. (Much in the same way that computer AI opponents do not trigger my fight-or-flight reflex the same way that real, human opponent minds do). Something never smells right when I read a prepublished adventure with one "this room was sacked by earlier adventurers" sitting by itself in the middle of a dungeon level (as occurs in Monte's room #6).

Maybe this project will be a raving success. Monte certainly has a productivity level which I can only look at with envy. But when I look at the enormity of the text written for a single room (even quasi-empty, as with Monte's sample room #6), I feel like all the effort is put on exactly the wrong end of ship. I'd like my core rule system to be finely systematized, robust, and short. I'd like my dungeon designs to be large in room-count and atomically brief in room description and statistics. This work seems like pretty much the exact opposite of that.


The Joy of Six

One of the great joys I get from running OD&D games, as a mathematician and quasi-OCD numerologist, is getting to play the game almost entirely with just one d20 and a fistful of d6's and nothing else. (Well, that and the opportunity to come up with clever blog titles.)

In OD&D, hit dice are d6. Weapon damage is d6. Spell damage is d6. Search and hearing checks are d6. Surprise and initiative are d6. You simply don't have to keep track of any other dice types (is that a d8 or d10, or are you just happy to see me?). The d6 is of course the most widely available type of die, so it's really easy to get a big batch of them for all your hit dice rolling needs (as opposed to the big batch of d8's I had to keep together for AD&D; and don't get me started on the big fistful of everything you need in 3E).

Now, if you read my last blog post you'll notice a few exceptions to that. I use variable weapon damage in my OD&D games (from Supplement I, snip out the extra L-size damage column). I use variable PC hit die (again from Supplement I). In general, I'm happy to have reasons on the players' side to have some variety in their dice, as long as it doesn't effect me on the DM's side of the screen.

So does that mean that I'm being inconsistent when I still use d6's for all the monster damage, including humanoids using actual weapons? (This being the self-criticism that demanded this particular blog post.)

I think I can make a pretty reasonable defense of that. First, let's say that most monsters' fang/claw routine is about as damaging as a sharp spear or arrow. Well, both of those are d6 even in Supplement I, so we're fine.

Secondly, let's assume that humanoids are generally too primitive to make the highest quality arms and armor. Spears and hand axes and arrows are easily made in the field without complicated tools (in fact, that might be a primary advantages of such weapon types). But when they try to fashion more advanced armament like swords, polearms, battle axes, etc. (normally d8), the results are sufficiently crude as to lose one die size in damage (that is, d6). Under this assumption, we maintain our consistency.

But there's a little wriggle room. If you decide to arm your orcs with crude but ultra-heavy two-handed swords or lochaber axes (halberds), go ahead and give them +1 on damage rolls (equivalent to d8). If they have managed to arm themselves with better human or dwarven weapons, again give them a note for +1 damage.

On the flip side, consider the smaller OD&D monster types. In my game, halflings take a one-die reduction in damage for weapons cut down to their scale. Presumably I should really do the same for the smaller goblins and kobolds. If I did that, I'd have two die steps down (one for size, one for crudity) resulting in d4 damage by the PC rules (which, parenthetically, matches the variant damage they get in Supplement I). I think I'll just go ahead and give them -1 damage from their d6 (basically equivalent to d4), and again be able to use the d6 for everything in sight; in fact, it sort of provides a nice symmetry for the larger ogres getting d6+2. Ah.