Tuesday, November 22, 2011

The Articles of Dragon: "Falling Damage"

And so it begins.

Issue #70 of Dragon (February 1983) saw the appearance of "Falling Damage" by Frank Mentzer, the first of what would turn into several articles discussing this strangely contentious subject. I say strangely contentious because, until this article appeared, I don't think the "right" way to adjudicate falling damage was ever a topic of serious conversation, at least not among the gamers I knew. The LBBs provide rules for falling damage hidden away in the section on aerial combat in Volume 3, where it's stated simply that
for every 1" of height a rider must throw one six-sided die for damage occurring from the crash, i.e. a crash from 12" means twelve dice must be rolled and their total scored as points of damage
That passage is the basis for what was the standard interpretation of falling damage in every form of D&D -- 1d6 damage per 10' fallen. That is, until this article, where Mentzer claims that the rules in AD&D were hastily written by Gary Gygax and were, as such, unclear as to his actual intent. Instead of 1d6 damage per 10' fallen, the claim is advanced that Gary actually meant 1d6 damage per 10', with the dice being cumulative in effect. That is,
1d6 for the first 10' feet, 2d6 for the second 10' (total 3d6 for a 20' fall), 3d6 for the third 10', and so on, cumulative. The falling body reaches that 20d6 maximum shortly before passing the 60' mark.
According to Mentzer, this new system -- which in fact Gygax had "always used" -- is "definitely more realistic." (emphasis mine) There's that dreaded word, the hallmark of the Silver Age. It's something that, at the time, meant a lot to me, but that, as the years have worn on, I find myself caring less and less about. In a game where people can throw balls of fire from their hands and adventurers become tougher to kill as the result of slaying monsters and looting treasure, fretting over whether a 60' fall or a 200' fall deals 20d6 damage seems bizarre. More to the point, after nearly a decade of "doing it wrong" (Mentzer's words), did the difference matter enough to make the change?

Regardless, the claim that Gygax had "always used a geometrically increasing system for damage in AD&D games" strikes me as somewhat suspect. I suppose it's possible that, sometime after the LBBs were published, Gary changed the way he dealt with falling damage in his home campaign. But, if so, I find it surprising that he never noticed that in every other D&D product published after 1974, the 1d6 per 10' rule is the norm. Indeed, I'd hazard a guess that, if one were to look through the various modules and articles Gygax penned between 1974 and 1983, we'd find instances where the 1d6 damage per 10' rule was in fact used. There's a fun project for an enterprising soul out there!

47 comments:

  1. Also interesting that Mentzer's own edition of D&D uses the straight 1 die per 10' of the fall.

    ReplyDelete
  2. You want realism? How about this: you fall more than about 50', you suffer broken bones and internal injury. Roll a system shock check. If you fail, you die. If you don't fail, you pass out from the pain. Unless someone gets to you PDQ with some super high level magic, you die in, I dunno, 10d6 rounds.

    ReplyDelete
  3. You want realism? How about this: you fall more than about 50', you suffer broken bones and internal injury. Roll a system shock check. If you fail, you die. If you don't fail, you pass out from the pain. Unless someone gets to you PDQ with some super high level magic, you die in, I dunno, 10d6 rounds.

    What's interesting is that Mentzer out-machos Gygax later in the article, claiming that he'd "be as tough ... [as Gary] and then some." His proposal is to require a system shock roll for any fall of 60', plus a save vs. death, modified by -1 per 10' fallen (plus a bonus for every level of the character).

    I don't know. Maybe I'm weird but this all seems like a lot of effort to model something "realistically" that didn't need it.

    ReplyDelete
  4. Requiring a save or die past a certain point seems like a cleaner and simpler rule to me- beyond a certain point, you stop worrying about the exact height ('bottomless pit', as it were).

    Whenever the number of dice being rolled gets above five or six, whatever the cause, I start to suspect something has gone wrong. No situation should require Sasquatch-sized handfuls of dice to resolve.

    ReplyDelete
  5. You want realism? How about this: you fall more than about 50', you suffer broken bones and internal injury.

    That was the point of massive damage rules in 3e -- to get around these types of artifacts caused by the hit point model. And it was nice because it was a generalization of what you just said; any time you get above threshold X in a single round, role a system shock (actually a Fortitude save, but same difference).

    The "realism" issue is actually a separate issue from balance and mathematical modeling that was growing around this time in D&D. I am all for the latter, but not a fan of the former. Indeed, many competitor RPGs at this time realized that damage modeling was less about realism and more about the theme of your campaign. Did you want a gritty feel or a more cinematic feel? Or even an over-the-top comic book feel?

    The advantage of mathematical tools like probability and expected value is it gives you a gauge of how a mechanic like this effects your system so that you can choose the proper rules for your campaign feel.

    ReplyDelete
  6. I agree, James. Make it simple: cap falling damage at 5d6, 1 die for each 10 foot fallen. You fall further than 50 feet, you're dead, have a Wand of Fonzee's Feather Pillows in hand, which brings forth 6d8 cloth sacks filled with goose down.

    ReplyDelete
  7. I never understood the obsession some had for these rules, but some in my group back in the day were scientists and engineers (well, students in science and engineering majors) who'd love to whip out the calculators to show why all these rules were "broken." (That's a feature of every hobby, it seems: passionate arguing over minutia.) Meh. I just stuck to "1d6 per 10'," though I do love "buckets o'dice" moments. :)

    ReplyDelete
  8. I think what you are seeing here is the beginning of the "corner case theorists" (as I call them) that is most commonly associated with the 3.x base of the game. These are the guys that crunch the numbers and say something like "If I fall 120' I will take a maximum of 72 points of damage and an average of only 42 points of damage! So I don't need to fly, I'll just jump off cliffs and buildings since I can easily survive the falling damage." The game authors response to this is to amp up the rules to cover this corner case (by adding the cumulative dice rule in this example) which only leads to another corner case, repeat ad nauseum.

    This still happens today with theoretical corner case issues like "golf bag syndrome" and "stealth rules are broken" in 3.5 and Pathfinder.

    For people that actually PLAY the game rather than just sit around and THINK about it these issues usually are all tempests in teapots that the game designers compulsively try to "solve". Sometimes because the game designers themselves have stopped actually playing the game.

    ReplyDelete
  9. I find it funny in respect to realism, or 'realism', that the average damage from that 20d6 damage is 70 hp, which is something a high-level fighter can shrug off, if not easily then at least with effort.

    The maximum damage isn't *that* big, either: just 120 hp. With that many dice you'd get amounts near the average, of course.

    So, if I'd go for realism in D&D type games, I'd use something else, like the proposed system shock things. Then again, D&D isn't a game very set for realism...

    ReplyDelete
  10. The whole falling damage debate was always odd.

    There is however a certain unpalatable quality to high level fighters leaping ridiculous heights, surviving the fall and continuing on their jolly way.

    ReplyDelete
  11. Didn't anyone read the two case studies in the 2nd Edition PHB about the fall survivors (Vesna Vulovic and the other guy)? Obviously, both of the them took 20d6 points of damage, not 190d6!

    ReplyDelete
  12. I understand that the falling damage rule also affected Gary's plans to write a thief acrobat class; maybe one of their class features is falling damage reduction from the quadratical progression to the lineal progression.

    ReplyDelete
  13. 4e uses 1d10 per 10 feet but you have more hp

    ReplyDelete
  14. For people that actually PLAY the game rather than just sit around and THINK about it these issues usually are all tempests in teapots that the game designers compulsively try to "solve". Sometimes because the game designers themselves have stopped actually playing the game.

    I agree. In fact, one of my commonest refrains of late has been that fun play trumps "good" design every time. By that I only mean that, if the rules produce fun at your table, then why worry about any other considerations?

    ReplyDelete
  15. I understand that the falling damage rule also affected Gary's plans to write a thief acrobat class; maybe one of their class features is falling damage reduction from the quadratical progression to the lineal progression.

    Supposedly, it was only when Gary published his thief-acrobat that he realized that AD&D's falling damage system wasn't written the way it was "supposed" to be, so you're absolutely right that the acrobat is an important piece in the puzzle of why this suddenly became an issue. My feeling is that Gygax just liked the idea of the acrobat class and then retroactively changed the way falling damage worked because it gave the split class a genuinely useful ability.

    ReplyDelete
  16. Falling damage for some reason brings out weirdness in how people see fantasy realism. Caught in a blazing ball of fire, no problem, but if you fall, you better get hurt.

    I think the strangest example is Villains & Vigilantes where, as far as I can tell, nuclear bombs do anywhere from 5d20 to 20d20 damage, and falling from a tall building will do the equivalent of a large nuclear bomb; falling at terminal velocity will be far worse than a nuclear bomb--for some characters, doing tens of thousands of points damage.

    I don't understand it; while on the one hand falling is something we do see happen in real life (as opposed to fireballs and nuclear explosions, knock on wood), fictional characters certainly survive some amazing falls in our source material. Movies are notorious for it.

    ReplyDelete
  17. "Falling damage for some reason brings out weirdness in how people see fantasy realism."

    It's not just falling damage that does this. Grappling is the same. Hit a moving and dangerous target with a flail or pole arm? No problem just roll a single dice. Want to punch someone or try to grab them? Oh boy, break out the intricate and byzantine grappling rules.

    ReplyDelete
  18. Interesting 0d&d fact. If only 1" fallen, pc takes damage only on a 1-2 d6

    ReplyDelete
  19. "for every 1" of height a rider must throw one six-sided die for damage occurring from the crash, i.e. a crash from 12" means twelve dice must be rolled and their total scored as points of damage"

    I interpret this as being 1d6 damage per 30 feet, since when outdoors (as in this example) 1" = 30 feet.

    In the dungeon, damage should also be 1d6 per 30 feet, so as to line up with the amount of damage from the pit trap in area 8 of the sample map:

    "Falling into the pit would typically cause damage if a 1 or 2 were rolled. Otherwise, it would only mean about one turn of time to clamber out, provided the character had spikes or associates to pull him out, and provided the pit wasn't one with a snap-shut door and the victim was alone."

    In my mind, this is a 10 foot deep pit, not a 3 foot, four inch deep one!

    This is also common sense. A 10 foot fall is really not much. I don't think it would have a 50% chance of killing someone. But a fall of 30 feet having a 50% chance of killing someone sounds realistic.

    In real life, real people have fallen insanely long distances and survived.

    "There is however a certain unpalatable quality to high level fighters leaping ridiculous heights, surviving the fall and continuing on their jolly way."

    An 8th level fighting-man is a superhero. And a superheroine can be thrown off the roof of a building and land on her feet, completely unhurt. I just witnessed this this morning watching the first episode of Futari wa Pretty Cure on YouTube!

    In my house rules, anyone wearing armour takes 2 hp less damage from any fall (I assume most armour includes some form of padding). Also, thieves falling more than 30 feet who make a climb roll take only half damage (to a minimum of 1d6 damage). I assume thieves are "good at falling". They must have had plenty of practice while they were learning to climb walls, right?

    That doesn't mean falling into a pit won't be dangerous - the first pit trap in the dungeon I'm currently working on is only 10 feet deep - but it has both iron spikes and poisonous snakes at the bottom!

    ReplyDelete
  20. Considering how many of us poor li'l 1 HD humans have fallen ten feet and have limped away grinning, I'd be inclined to give the first 10 feet free (or if you're feeling grouchy maybe a 1D4-2). That's a nice offset when you smash into jelly at 60 feet or more.

    ReplyDelete
  21. Rules for falling damage are easy to dismiss until someone falls. Then, unless there are clear and BELIEVABLE rules for damage, the arguments begin. The rules don't need to be complex or even founded in solid physics, but they must be consistent and believable. One of the most acrimonious RPG moments I can recall was when a DM ruled that a character took 3d6 damage from jumping down a ledge while the orcs pursuing him took only 1d6 because they were toughened to such things. Whether or not the DM's position had any merit, the apparent unfairness of it nearly destroyed that campaign. Falling damage engenders this type of reaction, I believe, precisely because it's governed by v=1/2 a tt. There's no such formula for halbard damage, so we accept a lot more abstraction when resolving it.

    ReplyDelete
  22. Plenty of people have survived surprisingly long falls, but more have been killed, paralyzed, or knocked unconscious from falling off a garage roof or a horse. The lesson I take from that is there ought to be a lot of randomness and probably a saving throw vs. breath weapon involved.

    ReplyDelete
  23. There is however a certain unpalatable quality to high level fighters leaping ridiculous heights, surviving the fall and continuing on their jolly way.

    Yes, & the semi-mythical case of the PC shooting himself in the temple to intimidate an NPC.

    If the DM can't handle this sort of player behaviour it is the campaign that is broken, not the rules.

    That being said, I tend to reserve diced damage for combat, using saving throws for most other cases.

    ReplyDelete
  24. Not only does the pit example in Vol 3 (cited by Akiyama) provide a slightly different interpretation of the falling rules, but there's a third set of falling rules in the naval combat section, p. 31:

    "Those falling must make saving throws, one chance out of six for every level fallen that damage will be sustained, i.e. a fall from 40 feet will require a 5 or a 6 to save. Damage is determined by rolling a six-sided die for every level, one die for every two levels if the fall is broken by water or some yielding substance. Note that any figures struck by a falling figure must also make saving throws and are subject to damage, just as if they had themselves fallen."

    These are closer to the pit rules than to the aerial mount falling rules, plus they seem to confirm that it's one die per ten feet, not ten yards. The switch from "sometimes take damage from short falls" to "always take damage from short falls" is maybe connected to the fear of the edge case. Personally, I'd rather go back to the original and reduce the chance of damage for short falls, rather than using Mentzer's suggestion.

    I've got my own simple solution to the edge case which doesn't require changing the damage dice at all: look for matches on the dice. Doubles mean an injury (double 6=head, double 1=leg,) triples mean a severe injury, quadruples mean permanent injury. Injuries cause penalties to actions until taken care of, but are otherwise mostly freeform description.

    ReplyDelete
  25. There's a fun project for an enterprising soul out there!

    I suspect our definitions of "fun" may be wildly divergent from one another. :)

    ReplyDelete
  26. @Pat Henry: Considering how many of us poor li'l 1 HD humans have fallen ten feet and have limped away grinning, I'd be inclined to give the first 10 feet free (or if you're feeling grouchy maybe a 1D4-2). That's a nice offset when you smash into jelly at 60 feet or more.

    Agreed, and I'd go farther than this. I find that returning to the original "to hit = to kill" combat rules of Chainmail/OD&D always clarifies these issues for me. Hit Points are a latecomer to the D&D combat system, as a way to model the slow attrition of the stamina and spirit of a large monster or an extraordinary hero over the course of a long battle. An ordinary human being in Chainmail was either alive, or dead. Accordingly, I'm a big fan of using Saves rather than Hit Points wherever possible. It might be possible for a hero (i.e., levelled character) to survive a fall that would have killed a lesser man, and to suffer HP damage in the process. But the first step should always be to throw a save (at an appropriate bonues or penalty) to handle the ordinary case of what happens to an ordinary human when falling from a height of X feet.

    ReplyDelete
  27. "Yes, & the semi-mythical case of the PC shooting himself in the temple to intimidate an NPC.

    If the DM can't handle this sort of player behaviour it is the campaign that is broken, not the rules."

    I think the problem here is the idea of 'Hit Points' - If I go back to D&D I'd want to explain HP as something more akin to Hero or Fate Points - heroes and villains have greater capacity to avoid being run through with a sword, whether through ability and experience, the blessings of gods, or simple luck. Conan might be able to take a few more blows than an ordinary man, but run him through with a sword and he's dead. But you only run him through with a sword when he's fatigued and used up all his luck, i.e. when he's down to less than 8HP.

    http://drbargle.blogspot.com/2010/06/characters-saint-sebastian.html

    If a player wanted to shoot himself in the head in a game I was running, it'd be a Save or Die situation, with anything less than a natural 1 (which would be a misfire or something) being a permanent crippling injury.

    ReplyDelete
  28. Thought: perhaps the 1d/10' was intended only for pits, where there would nearly always be a wall close enough to scrape along to shed velocity? Then the 1d/10' cumulative variant could be used for falls where there's no convenient surface to use as a brake.

    Worst of both worlds, possibly, but...

    ReplyDelete
  29. What happens in a game: Player realises that his character can easily survive the 60' fall to the water below, so jumps off. I've known gamemasters to look askance at this. But consider...

    What happens in the reality of the game: Hrofgar the Barbarian, running from the pursuing city guards, finds himself atop of the walls of Old Arris, with guards approaching from both ends. Praying to his god Kohnaan that the moat is deep enough he takes a swan dive off the wall just ahead of the guards. The guards are stunned to see Hrofgar pop up to the surface and swim for the far side (just ahead of the moat monster). "Surely the Hand of Kohnaan eased my fall," he thinks and sends up a prayer of thanks.

    Unfortunately the argument about falling comes afoul of the argument of what exactly is a hit point. In reality a serious fall is likely to result in broken bones and ruptured internal organs. But then, so is getting hit by a mace, and we don't see a "you were hit on the pelvis, make a system shock roll to avoid your hip breaking." {well, OK, some people used critical hit tables [ala Arduin and later Arms Law] to get these result on a critical hit but that was only in the case of a critical hit).

    I use the 1d6/10' because it's more heroic. I also don't use rules for blood infection (unless you are fighting someone who actively spreads filth on their weapons), or diseases because it's not heroic. No one wants to hear of Hrolfgar Who Died Of Septicaemia.

    [Incidentally my house rule for damage in AD&D: Hit points are the ability to avoid damage by whatever means, but they must be actively used. Actual physical damage to the character is represented by taking CON damage. This replaces the "live to -10 hp." Critical hits and sneak attacks do damage to CON directly. So actually managing to drive a dagger into someone hurts. I used each 6 on a d6 for big damage (spells and falls) caused a point of CON damage but no hit point damage.]

    ReplyDelete
  30. I had a chat with Gary at one point, and he said something along the lines of "the text should have been 1d6 per 10' per 10' fallen." Notice the repetition of the "per 10'" part, which can easily pass off as an editing error, when it's not. Actually, the above models the acceleration due to the fall, which is definitely more "realistic" (and lethal.)
    About realism: even in a game when you have tens of hit points, if these model the skill to avoid real damage, when you fall like a brick there is little skill which can save your skin.

    BTW, 2e also had massive damage rules; if you got 50hp damage you could die on the spot.

    ReplyDelete
  31. The witty JA Holmgren did an entire series of comics about this and other "broken rules" called, "Joe Genero, Adventures of the Common Man." Loved it, and I still have my Joe Genero tee shirt!

    ReplyDelete
  32. That's how Castles & Crusades handles falling damage.

    ReplyDelete
  33. The revisionist history on this issue was truly bizarre, and frankly troubled me on a more fundamental issue about the status of "what Gygax said", and how it was possible for him to directly contradict what he'd written on multiple occasions and examples. I suppose he's not the first artist for whom that's true.

    That said, for several years I was a proponent of cumulative-damage-dice, and wanted to use this for both falling and any other environmental factors (heat, cold, hunger, thirst, air, etc.) More recently I again became convinced to use the original, simpler mechanic. (Discussed here.)

    ReplyDelete
  34. This comment has been removed by the author.

    ReplyDelete
  35. Indeed, I'd hazard a guess that, if one were to look through the various modules and articles Gygax penned between 1974 and 1983, we'd find instances where the 1d6 damage per 10' rule was in fact used....

    Two such examples, both from 1978 (recalled from when I ran these modules and they caught my eye):

    Module G2, re: slipping over the glacial rift ledge 50-300 feet down: "Each member falling to the bottom of the Rift takes 1 six-sided die of damage for every 10' falling distance, 10 dice maximum due to the cushioning effects of snow drifts." [p. 2]

    Module D1, re: sink holes 10-80' deep: "[I]f the party is running in haste or traveling without light, treat sink holes as pits, with a 1 in 3 (1-2 on d6) chance of falling in. Damage thus sustained is 1d6 per 10'" [p. 3]

    ReplyDelete
  36. The revisionist history on this issue was truly bizarre, and frankly troubled me on a more fundamental issue about the status of "what Gygax said", and how it was possible for him to directly contradict what he'd written on multiple occasions and examples. I suppose he's not the first artist for whom that's true.

    No, he's not, but it's one of the reasons why, much as I liked and respected the man, I didn't always put full stock in his recollections a quarter-century or more after the fact. I think Gary in the years before he died had a lot less reason to be revisionist in his answers than in, say, the '80s, but it still made me wary. This goes double for other TSR employees who report on what they remember Gygax saying.

    ReplyDelete
  37. Two such examples, both from 1978 (recalled from when I ran these modules and they caught my eye):

    Somehow, I knew you'd come through with some examples to cite! Yes, these make Mentzer's assertion that Gary had "always" used the 1d6 per 10' per 10' for falling damage even more dubious.

    ReplyDelete
  38. I wonder how many people will make contradictory statements if they live to be 70 years old.

    In any event, my guess is Gary just changed his mind, or made a house rule official. To note, in both Mythus and LA he uses the newer exponential damage system. I'll bet he read a science article on falls and then changed his mind forever after reading that.

    ReplyDelete
  39. I wonder how many people will make contradictory statements if they live to be 70 years old.

    There's a difference between changing one's mind later and denying that one is in fact changing one's mind, isn't there?

    ReplyDelete
  40. In any sort of later recollection, the word "always" takes on the additional meaning of "this is not true." George Lucas uses it in this sense, too. :-)

    ReplyDelete
  41. I just hope people don't start thinking all people are deliberately lying. I think the only way we can find the truth is examine the perspectives of everybody, and then not letting our own biases influence as well.

    ReplyDelete
  42. I just hope people don't start thinking all people are deliberately lying. I think the only way we can find the truth is examine the perspectives of everybody, and then not letting our own biases influence as well.

    Deliberately lying? No, not always, even most of the time, but, when there's a history of unreliability, I don't think it unreasonable to be skeptical. Plus, in this particular instance, we're talking about what Mentzer said in an article published in 1983, not even a decade since the publication of OD&D, let alone AD&D. If memories were failing even then, it calls into question much of what's been said in the nearly 30 years since then.

    ReplyDelete
  43. My guess is EGG had been running or witnessing higher level campaigns long enough to realize linear falling damage just didn't cut it for all the reasons cited here. And the modified rule more closely mimicked the 10 feet per second per second acceleration dynamic of gravity, which (who knows?) he either knew or did not know or did not care about at the time he was writing the original rules.

    Clearly, a fall of 10 feet might just kill a second-level magic user. In fact, it seems almost cruelly lethal overkill for that class/skill. Not until you're running 10+ level characters does the silliness of players laughing off the linear damage of falls from towering heights begin to manifest.

    ReplyDelete
  44. The thing is, velocity (and therefore energy) increases exponentially with TIME fallen, but linearly with distance fallen. The result being that a linear 1d6 per 10 feet is actually the more "realistic" of the two models.

    ReplyDelete
    Replies
    1. Almost entirely correct. Velocity actually increases with only the square root of the distance fallen. But energy increases with velocity squared. So the square root and the squaring cancel each other out and energy does increase linearly with distance fallen.

      Delete
  45. I think we just recreated that debate all over again! Truly, the classics last forever. I look forward to seeing it one more time when we hit Dragon #88. :-D

    ReplyDelete
  46. And one more in the DMG Aerial Combat section: "it is 1-6 hit points of damage for every ten feet they fall" (p. 53)

    ReplyDelete