Actual Play

Castle Xyntillan – Session #2 – Killer Clouds

The Company:

  • Anna (F1)
  • Fernando (T1)
  • Robain (C1)
  • Michelle & Raul (porters)

Casualties: All of the above. That’s right, the session ended in a TPK. Read on for the sordid details.

Loot: None, obviously.


Resuming the action immediately after last session’s demise of bowman Mattia, the company takes a moment to debate where to go next. Ultimately Anna decides to try her luck once more with the locked door in the galloping hallway. She rushes across, bashes the lock to bits and tumbles into the room beyond. An armory!

While Anna takes stock of the armaments on display, the remainder of the company gingerly crosses the hallway, taking care to avoid the horses. The armory contains a generous amount of weapons (although, to Robain’s disappointment, none of them blunt). There are also four well-oiled chain hauberks on display, and a badly pierced suit of plate. From behind one of two closet doors something can be heard trying to bash it down.

They search the armory for exceptional weaponry, but none can be found. Anna inspects the plate armour, and although it is badly damaged, she decides to exchange it for her chain mail. Robain boldly opens the closet door from behind which no sounds can be heard, and is confronted with a bizarre monstrosity!

Rust Monster from AD&D 2e
Real D&D heads know

Robain immediately smashes the thing in the head with his mace and is surprised to see a glass eye fly from its head, stuffing protruding from the empty socket. The monster turns out to be a stuffed specimen. Puzzled and somewhat amused the company proceeds to investigate the thing and the closet it inhabits.

While they are thus preoccupied, the door through which they entered opens and six gaseous clouds, each with an iridescent nucleus, enter the room and proceed to suck blood from the company through remote osmosis. Horrified at the sight of blood streaming from their pores and flying off to the monsters, the company hesitates to engage.

Robain makes for the other closet, and flings open the door. An animated suit of armour emerges and immediately flies off to attack the armour that is now being worn by Anna. Fernando ducks into the closet that previously held the armour, followed by the retainers. Anna, meanwhile, barely manages to jump into the closet containing the monster and shuts the door in the proverbial face of the animated armour.

Panicking while blood continues to be sucked from their bodies, Robain attempts to push over a rack of weapons on top of the clouds but the thing won’t budge. Fernando takes aim with his crossbow and obliterates one of the clouds. The porters remain in their closet, terrified and near being exsanguinated. In the other closet, Anna begins to shed her plate armour, as the animated plate continues to hammer away at the door.

Fernando pushes over another weapons rack on top of the clouds and manages to eliminate a few more. Robain swipes at the evil things with his mace and deals some damage as well. But the clouds slowly but surely continue to suck blood from all the company’s members. The porters begin to collapse.

Having removed the plate armour, Anna flings open the door and dexterously sidesteps the animated suit. It flies into the closet and proceeds to assault the discarded plate in the corner. Fernando proceeds to fling daggers at the clouds. Robain succumbs to the clouds’ attacks and collapses.

Anna emerges from the closet and begins to bash away at the clouds as well. The two remaining companions persist for a while longer, eliminating clouds left and right, but ultimately are overwhelmed. Both Anna and Fernando fall to the armory’s floor, as the last of their blood is drawn from their bodies and consumed by the evil, killer clouds. Thus ends the company’s first expedition into Castle Xyntillan.

Referee Commentary:

Well, there you have it, the first honest-to-god TPK of my refereeing career. When I rolled this encounter I did not expect it to be quite so deadly. But a number of factors conspired to produce the regrettable fate of our player characters.

Classic D&D characters are indeed quite squishy. The party was rather limited as well, with only three 1st level PCs and no combatant retainers. Upon reflection this should have been reason enough for the players to terminate the expedition and return to town. Furthermore, the players found themselves in a cul-de-sac, with the only escape route leading past the clouds and through the hallway with the racing phantom horses. Despite this, again in hindsight, they should have probably still risked running for it, because they were squarely outmatched. The fact that the clouds’ attacks can only be averted through a save (effectively circumventing AC) also probably made a bit of a difference, although the players rolled very well on their saves throughout the battle. So it goes, I guess. We finished the session rolling up new characters.

The joke with the stuffed monster was lost on my players because most of them haven’t been playing classic D&D for that long. It still sort of worked as a puzzling oddity.

The moment when the animated suit of armour emerged from the closet and proceeded to attack Anna produced more than a few smiles around the table. It’s this kind of serendipitous mayhem that CX appears to be engineered to produce. I absolutely love those moments and can’t wait for the players to return to the castle to once again try their luck.


Hackbut – Core Mechanics – Ability Checks

So far in this series on Hackbut core mechanics I’ve described attack rolls and saving throws. That leaves a way to resolve anything else that may come up in the game. Out of these three components, ability checks, general task resolution, or “situation rolls” as Talysman aptly calls them, were the hardest to pin down, and I don’t feel I’ve completely settled on an approach yet. But what follows is what I am going with for the time being. Strap in, this one is a bit longer than usual.

Probably the biggest reason it’s hard to settle on a general way to handle situations is that there wasn’t really any general mechanic as such in the early editions of the game. The closest candidates would be the n-in-6 die roll pattern that frequently occurs (with the 2d6 reaction roll a close second), then we have the bit in Moldvay about using roll under ability checks, and finally there are the various ways thief skills are handled. Let’s tackle each in turn.

We can quickly eliminate roll under ability because, like I mentioned before, I have over 30 sessions of The Black Hack under my belt. This uses roll under for everything (attacks, defense, saves, skills, you name it). The problem I have with this is that it makes abilities too important. I also find it encourages rolling for trivial stuff. Used in moderation I guess it can be fine, but since I have had my fill of this mechanic I decided to not use it at all in my hack.

The thief skills are a different matter. Similar to save categories, I never got on with the granularity of the various skills. In the KéK classes, the separate skills are maintained but rationalized to n-in-6 probabilities. In WBFMAG, the thief has a generic “thievery” skill that is also an n-in-6 chance. I like the latter quite a bit because it affords some flexibility for determining what does and does not fall under the thief’s abilities. However, where these approaches fall short is in being generalizable across all classes for any kind of skill check (or situation roll) that may come up. This also applies in the other direction, so to speak, to those generic n-in-6 rolls you find in classic D&D, such as finding hidden things, forcing doors, etc.

Some searching (using Brendan’s invaluable OSR search engine) turned up an excellent post by Homebrew Homunculus, which outlines a general way of handling any skill check on a d20. The nice thing about this in particular for me is that it allows for applying an attribute modifier if desired, and it also allows for improvement with level if a class applies. The TLDR of it is: if a situation comes up for which a roll is warranted, roll a d20 and try to get a 15 or higher. If an attribute applies, add the modifier, if a class applies, add your level. Done.

Slick, right? I do like it quite a bit and have basically replaced all the typical n-in-6 rolls with this mechanic, as well as replaced the thief skills with this. It’s very easy to grok for players because it’s so similar to Target 20. And similar to roll under ability checks it’s easy to apply on the fly when a new situation comes up. However, abilities make less of a difference with this approach and if desired, improvement with level is baked in. Also, the target number is fixed, so trivial rolls are discouraged.

I wasn’t entirely happy with it, though, and the main reason for it is that I like the chunky feel of the d6, and I worried a bit this made rolling for various things feel too similar. In a way, I wanted to feel like I was refereeing the game more closely to the old ways, as for example described here by Delta:

If a “brand new” thing comes up (say: baking skill, something like that) then I revert back to a d6 roll — like OD&D uses for listening, opening doors, finding secret passages, traps opening, etc. I feel like on an improvisational basis I can estimate a reasonable chance for success out of 6 (but not 20) — as a default I give a 2-in-6 chance to succeed, like: roll d6, add some ability bonus, and a total roll of 5+ is success.

Delta’s D&D Hotspot

However, I wanted to stick to a fixed target number (5 on a d6) and I could not really figure out how to translate HH’s approach to a d6. The main issue being that levels and attribute modifiers quickly overwhelm the d6.

The final piece of the puzzle was the previously linked series of posts by Talysman on situation rolls, as well as a pamphlet on “general abilities”. First of all, John offers some neat guidelines for translating attribute modifiers across the various dice rolls (d6, 2d6, d20). Basically, if you have a +/-2 in an attribute, you get a +/-1 on a d6. (I use Delta’s attribute modifier sequence, so I don’t need to deal with +/-3.)

Second, in response to a comment of mine, John suggests a painfully elegant way of applying character level to a d6 roll: compare to dungeon level or monster HD. If lower, get a -1; if higher, get a +1.

The upshot of all of this is that I can now freely choose between resolving a situation on a d6 or a d20, depending on what I feel like in the moment. I might gravitate to one or the other at some point, it’s too soon to say. And for proper skills (like the thief skills) I expect I will stick to the d20 system. But for anything else, I now feel comfortable using the 2-in-6 roll as well.

Before I close, some of you may be wondering: isn’t a generic task resolution mechanic anathema to old-school D&D? You may be right. I don’t know. But what I do buy into is the insistence on reducing die rolls as much as possible. For this, Talysman also has some excellent guidelines which I’ve chosen to adopt. Because I feel it’s not just important to be able to explain to players when we will roll dice, but also when we won’t.

In closing, I think 15+ on a d20 and 5+ on a d6 are a sufficiently rich palette for adjudicating any situation that does not fall under an attack roll or a saving throw. With the tricks outlined above you can apply ability modifiers and class levels if you so wish, and if you pair this with a doctrine that prioritizes skipping die rolls all together, you are freed up as a referee to run a game at the blistering pace that classic D&D in my view requires.


Hackbut – Core Mechanics – Saving Throws

Continuing on my discussion of Hackbut core mechanics, let’s briefly touch on saving throws. This one did not take a lot of poking around. To begin with, I have no nostalgic attachment to the classic saving throw categories. They strike me as obtuse and needlessly granular. I know the origins of the S&W unified save are primarily legal, but they represent a welcome streamlining in my view. So I decided to port them over into the KéK classes. This was a trivial exercise. If you want to do the same, simply use a base save of 15 minus level for fighters and thieves, and 16 minus level for magic-users and clerics. I also use the +2 save bonuses for the various classes as listed in WBFMAG.

That more or less covers saving throws for player characters. One last thing would be how to handle monster and NPC saves. Here, interestingly, WBFMAG and S&W proper diverge. The former has the basic guideline of using 19 minus HD, whereas the latter has a table listing a save value for each HD.

I had problems with both approaches. While convenient, WBFMAG’s monster save seems high in comparison to those of player characters. S&W’s solution requires a table lookup which is a no-go for me. So I decided to do a quick analysis of the S&W numbers and arrived at an average base save of 16 minus HD. Let’s call it a nice and easy to remember 15, incidentally making them save the same as fighters. This means low-level monsters save a little better than by-the-book S&W, and high-level monsters get off ever so slightly worse. I call it a wash.

As an aside, I could have also gone with Delta’s approach to saves, which in many ways is similar in spirit and mathematically balanced to a comparable level as S&W’s solution. The reason I did not go for it is that it maintains the various save categories which, like I said at the top, are just not something I feel I need in my game.

So that covers saving throws. The point of all of this is basically: a unified saving throw is a convenient and justifiable streamlining, easily hacked into your preferred classic D&D ruleset of choice.


Hackbut – Core Mechanics – Attack Rolls

In the previous post I mentioned using Target 20 and Homebrew Homunculus’s simple d20 skill system. I should probably also mention that although I use those basic KéK classes, I did stick with Swords & Wizardry’s unified saving throw. Between the three of them those cover all the “core mechanics” in the game.

I arrived at Target 20 mostly through a process of elimination. I knew I did not want to do any lookups during gameplay so the traditional attack matrices were out. The logical alternative baked into S&W and also WBFMAG is ascending armor class. However, as HH has pointed out, S&W’s base ascending armor class of 10 is mathematically incorrect. You don’t really notice it when you use it out of the box, but I ran into trouble when I tried to come up with a player-facing defense roll that would be mathematically identical to a referee-facing monster attack roll.

This left me with two options: adjust the AACs listed in CX on the fly by 1 point (violating my rule to remain fully compatible with the module), or use descending AC after all. This is when the appeal of Target 20 really became apparent to me. It is both 100% mathematically identical to the original game, and very easy to use at the table. You just need to get over the fact that lower ACs are better.

Incidentally, it is also trivial to rewrite Target 20 as a player-facing defense roll:

Defend: d20 + your armor bonus + opponent attack value + modifiers ≥ 20

Armour bonus: 9 - AC
Attack value: 9 - HD

However, after using this for one session, I came to the conclusion that rolling to attack as a referee is actually faster (and possibly more fun). I’d gotten so used to The Black Hack’s player-facing defense rolls that I thought they were essential to a smooth-flowing game. But the big difference is that attack and defense rolls in TBH are exactly the same procedure, so there is no extra learning involved for players. With my so-called clever defense roll, players now need to learn two procedures. That’s one too many.

This is getting long-ish so I will leave discussion of saves and skill checks to future posts. I’ll just reiterate that Target 20 is indeed the superior procedure for attack rolls in classic D&D. I recommend using it.


Hackbut – Overview

So, let me tell you about my D&D hack.

I put this together for my Castle Xyntillan campaign. After a 30+ session campaign using The Black Hack I decided I wanted to try a ruleset that would be closer in feel to Original D&D. In particular, I had grown weary of the roll-under ability score mechanic. What I did like about TBH, and continue to appreciate, is its simplicity and accessibility. So whatever I would switch to would ideally be of a similar level of complexity.

The obvious choice would have been Swords & Wizardry, because that’s what the megadungeon is ostensibly written for, although I believe Gabor Lux used another system to play-test it, and there are a few details, like intelligent sword stats, that don’t show up anywhere in S&W. But I wanted something a bit more light-weight, while at the same time maintaining full compatibility with the module’s contents.

I also considered using Old-School Essentials, but although its presentation is incredibly slick, believe it or not, I still felt it comes with overhead that I would then have to house-rule out. I wanted to be able to have one document that would contain all of the rules.

So, I decided to put together a full-fledged hack of my own. The trigger, if I recall correctly, was reading the Kazamaták és Kompániák basic classes. I really liked the level 6 ceiling on those. It nicely matches the level range of CX and I prefer a low-level game anyway.

Not feeling like re-inventing the wheel, I looked around for a game to use as a chassis. Ultimately I went with White Box: Fantastic Medieval Adventure Game. It’s small but comprehensive, and therefore makes for easy copy-pasting and adapting.

So those are the two basic building blocks of what is now titled “Hackbut”, my classic D&D rules kit-bash: WBFMAG provides the rules framework, and KéK provides the classes, although I did overwrite quite a bit of both. Other key elements include Delta’s Target 20 for attack rolls, and Homebrew Homunculus’s simple d20 skill system for, well, most all other situations that need resolving through dice.

There is a lot more to be said about the various components that make up Hackbut, but I’ll leave that for future posts. The point here was mainly to say that if you are in the mood for putting together your own D&D hack, consider picking up WBFMAG. It’s an excellent place to start if you want something that is fully compatible with the old editions but easily adaptable.