DRYH: Car Ride Session [backlog]

In July 2010, three of us played a little Don't Rest Your Head on a long car ride. Mechanically, this worked very well, as there's not much overhead. I used a cup to hold and roll the mini-dice I created previously. I GMed from the back seat. I didn't have a preset story and just planned to improvise.

My two players created some very interesting characters.

Brett: John Hannigan
Surface: Homeless war vet: scraggly, scars, limpy, quiet. 15 years since the war. Has a shopping cart. Carries a book from/by sister Laura (about 28 years old) which he also uses as a sort of journal. It has old letters stuck in among the pages.
Beneath: Memory is slipping away. Feels vague guilt over the death of his brother Tom (fled?). Laura's letters stopped coming recently.
Insomnia: Nightmares about Tom's face. Sees bright/flashing lights while awake. Recently saw a "nice lady" too.
Path/Goal: Find Laura and explain (to her?) about Tom.
Just Happened: A bright light while crossing the street, now shopping cart is gone. Only book remains.
Exhaustion talent: Heightened Senses
Madness talent: Draw something in his book and it becomes real
Respones: 1 fight, 2 flight.

Becky: Eli Fox
Surface: Upper-middle class white architect. Attractive in his business suit. Early 40s. Drives into city for the week. Works long hours. Married for 8 years. (Wife?) married young.
Beneath: Distrusts wife. Has been drinking a lot and has a coke habit. Wants to kill wife and lover. Bought a gun recently.
Insomnia: Walked in on wife and another man (a stranger).
Path/Goal: Revenge and closure on infidelity
Just happened: Hit someone with his car, and ran, taking his gun with him.
Exhaustion talent: Persuasive/smooth-talking
Madness talent: Endurance/resistance/strength
Responses: 2 fight, 1 flight

I was impressed by these characters, but I found it hard to bring them together. They started separately, after Eli crashed through John's shopping cart. John followed a wandering PaperBoy into an old gas-lit subway entrance. Eli got picked up one of Tock's clockwork patrolmen.

Improvising as I went, I had a very hard time bringing these two characters back together. Even after a long roadside break and some pondering, the story never really picked up again. I had some ideas that Eli would get involved with the Ladies-in-Hating--probably through one that looked like his wife--and possibly dealing with Mother When. Mother When would also work for John's story of loss--possibly reached through lost Mad City children and the Finishing School. I also had some interesting custom nightmare thoughts for John: a War Monger and a number of Leaf Tenants (play on the word Lieutenants, in light of the British pronunciation, "leaving", and tenements.) that burn in horrible conflagrations. This would somehow tie into John's war experience.

But these ideas never really picked up steam. I eventually got Eli and John together using a previously conceived nightmare/local of my own design: Benjamin "Button" Bridges, who runs a small warehouse of buttons, each imbued with a different memory or dream experience. As I recall [it's been over a year now], Tock's men came knocking on the warehouse door and John and Eli fled. I think things ended there, because I hadn't yet managed to give them a solid hook to tie them together and drive them into a story.

The main reason I'm blogging all of this now is as a reminder to myself of the agony of improvised GMing. I still think I had some rich characters, willing players, and some decent general ideas here. But I just couldn't pull it all together in time on the fly. Any GM-improvising system/tool should help with this sort of problem: not just generating ideas, but pulling them together too.

Another lesson is a variation on that old chestnut: "Never split the party." In this case, I think starting with the party together and firm goal would have helped a lot. Bouncing narration back and forth between two storylines sucks.

mini20: Tensions in RPG design.

Last week, I stumbled across Microlite20, which is a very pared-down incarnation of d20. I find that, while I'm ever seeking simplicity, I have a difficult time actually hacking large swatches of a game system away. I make only little snips here and there, always minding backwards-compatibility (or at least equivalency) with the source material. In contrast, Microlite20 reminds me of pruning in the tropics: you can cut all the branches off, leaving only the stump behind. And instead of this killing the tree, it grows back as lush and beautiful as ever in 6 months.

That said, there were a few little quibbles I didn't like, so I set about tweaking it a bit. This did turn into putting a few things back in (in reduced form). The result is mini20.

Like I said, a large portion of my week went into this (despite having other more pressing projects!). I even drew up some hand-drawn character sheets and converted S.'s Omri line to mini20. However, we didn't get a chance to play this weekend; instead, we got drunk on pizza, beer, white wine, and 70s TV Wonder Woman. I woke up a few hour later, a little hungover, and started thinking about the project.

As this blog demonstrates, I seem to be constantly hacking on various game systems, yet never seem to reach a satisfied conclusion. These are some of the tensions I've identified in the past 24 hours:

DMing vs Design. DMing is a live, dynamic, social activity. While the game is underway, quick decisions are called for and the game moves ever forward. Design, on the other hand, is a largely solitary activity, where the resulting document can be reiteratively edited until perfect.

I'm coming to realize that DMing actually scares me. I think that's why I keep falling back to design, in this vague belief that the perfect RPG system will make for flawless, perfect DMing.

Rules vs Rulings. On the heels of the previous point, some systems provide tomes of specific rules. The advantage here is consistency and a crutch for the DM; the down-side is that then the DM has to recall or look up the specific rules at game-time. The alternative is to provide a simple system for the DM to make rulings on: to apply a modifier or make a call at game-time, and then move on. There are no lookups involved here. The downside is that the DM needs to manage making fair, consistent rulings, along with all the other game details going on at the same time.

Desiring a lite system, I want rulings; but being a little fearful of screwing up as a DM suggests more established rules. In short, rules compile all the possible rulings into rules before the game starts. But of course it can never be perfect: some situation can always arise not covered perfectly by the rules, requiring adjudication (ie, a ruling).

Game vs Story. On the one hand, an RPG is a game: characters are bought, described, and balanced in terms of points. There's a simulated world with clearly defined possible actions (skills, powers, etc.) On the other hand, especially these days, all of this is supposed to serve only as the underlying foundation for a story. (I would argue that, in the past, dungeon-crawling was more of a simulation/game than a story.)

Character detail vs simplicity. So, in terms of the game/simulation, we need character modeling. The more detail and customization possible in terms of the rules, the more stats, rules, and preparation time needed. Otherwise, character details are left superficial to the rules--which often does already happen with roleplaying details. Also, if too simple, there's no way to differentiate different characters and abilities--like different fighting styles, a smooth-talking con man vs an earnest likable diplomat, etc--in terms of the game world itself.

But in terms of story, it is frequently not the numbers and die rolls that determine the ultimate course of the story, but in-character decisions made by players based on the "superficial" personality traits of their character. This suggests that if story is the goal, simplicity and letting the game aspects take a backseat should work just fine.

Complete System vs Hack Document. This is my personal preference: when hacking on a system, I tend to want a single finished document that describes the resulting game, rather than simply producing a hack/errata-like document that lists the changes to make to the underlying system. I should really try to get over this, since it takes a lot more time to do it this way.


So, in conclusion, I'm not sure if mini20 is a good idea or not. It could just be more mental flailing around on my part, largely due to my failure to resolve the above tensions and decide exactly what the hell I want from my gaming system.

Omri: Myrksog

In the spirit of getting back into the game, S. and I played a bit of my new d20house form of D&D today. It had been 1 week short of a year since our last session!

For session details, check out the "Myrksog" section of The Amazing Escapades of Omri Buckle & Co.


It went well mechanically. The new damage mechanic (the most obvious d20house change) was pretty smooth. My new DM screen worked well, especially when combined with more intelligent use of the SRD's spell index and monster index pages. Chess pieces still make great cheap miniatures.

Story-wise, there were a couple minor snags--mainly, how to balance the power of social manipulation with combat? For example, Jack charmed the lead hobgoblin and told a good story about having an important prisoner... but I couldn't decide whether this should be enough to let the party pass the whole encounter. (Of course, this still may have worked out interestingly since they would have had to come back this way later.) But, if I wasn't going to let them bypass all the guards with it, it seemed something useful still should come of it so that there's at least some value to trying tact before steel.

First, I rolled a contest between the charmed hobgoblin and the orc to see if the orc could talk some sense into his companion: that this rabble shouldn't be allowed into the inner sanctum without Myrksog's express approval. Turns out the orc won. But even in this case, there was some benefit to the charm: it got rid of the hobgoblin. Sadly, this backfired slightly, since this logically left the party locked outside without any lock-picking skills! Turns out the standard DC of 25 to break down a barred door was just barely in range (with a bit of magical aid), so it all worked out decently enough.

So, in retrospect, I think I played it pretty well. And, as a personal note for the future, a spell and a successful skill check should be enough to bypass an encounter, especially if those creatures are still available to serve later as monster re-enforcements (and now behind the party besides!).

Notion: Speeding Play

Since I'm already posting today, here's a few thoughts from the last couple weeks on z20 work to speed of D&D gameplay. In short, how do we make an RPG faster? (I'm just thinking in terms of game mechanics here, not table-talk and other social reasons for slow play.)

Fewer rolls. Rolling the die and then totaling the result and relevant modifiers takes time. Instead, we could just use static numbers. For example, use a static Defense Factor or Armor Class instead of a defense roll. Or roll initiative once at the start of combat, rather than every turn.

Group rolls are also possible: roll the dice once and then apply everyone's modifiers separately to determine everyone's results. So this is good for speeding initiative rolls or doing quick Spot check to determine who sees what.

Rolls just provide randomness, so the average outcome over the long haul is the same as not rolling. It means there are no lucky breaks, but also no unexpected failures. But, since not everyone's looking to go diceless, we just need to decide which rolls really need this randomness.

Quicker rolls. The more dice a roll involves, the more totaling is required. The same is true of modifiers (even in the absence of a die roll). Most of the time this is not a big issue, but consider the math required to determine the difference between normal AC, flatfooted AC, or touch AC if they're not pre-totaled or if some component of those totals has since been modified (such as if the character was subject to a Shield of Faith spell).

Fewer lookups. This is the difference between having a list of specific spell descriptions verses a general magic mechanic. The more rules you write, the more players and GMs feel obligated to look them up and follow them exactly. (Of course, general mechanics can also take time to adjudicate, especially if these leads to social argument.) Still, if GMs know there are only a couple simple mechanics and are empowered to apply those on the fly, it saves bogging down in rule lookups.

Lookups also happen in just recalling what modifiers or effects are currently in play. The fewer possible states or conditions, the less often this has to happen. For example, if a game has stunned, shaken, dazed, disabled, and clobbered conditions, someone's going to have to look them each up at some point. On the other hand, if there's only a single Impaired state in place of all these, there will be fewer lookups required.

Quicker lookups. Even if lookups are needed, how fast are they to perform? For example, how handy and visible is the information needed? This is the realm of DM aids like a screen, outlines, notecards, etc. Player's need to be able to quickly find the info on their character sheet: attacks, damage, AC, saves, skills, etc. And then there's all the stuff that fluctuates during play: spells in effect, conditions of characters (including hp, location, status, etc.), and so on.


The time saved on rolls is pretty minor, but adds up over time. Changing lookup frequency usually means changing the game mechanics. So probably the technique with the most payoff is to figure out how to speed lookups.

Notion: Modifiers suck

Modifiers are the essence of an RPG: some things are easier to do than others, and some people are generally better at certain tasks than others. Modifiers to the roll (or even without the roll, in a diceless game) represent innate skill or external circumstances that may aid or hinder your action. Even if modifiers are applied to a target DC rather than the roll itself, the Game Master still needs to track them.

However, modifiers tend to suck when there's a lot of them. They're easy to overlook: "Oh, wait, I forgot the +1 from the Bless spell! And I would have hit last turn if I'd remembered it; can we back up?" If you remember them, they take time to tally: "Let's see, that's a +4 from the Bull Strength and another +2 because I got Enlarged last turn; I'm raging right now, so that's +4 more..." And then there's issues of which apply when: "Oh, wait, I'm wearing a Belt of Giant Strength, so that means I don't get the +4 from the Bull Strength after all, since they don't stack." And then there's those modifiers that affect the calculations of the numbers already on your character sheet (rather than just adding or subtracting from them): "Okay, so that's a total of +6 to strength. That means +3 to melee attacks, +3 to damage, but only +1 to off-hand damage... oh, but considering my normal strength of 12, that's actually +2... now I can carry a heavier load and, since I was right on the edge, I'm no longer encumbered, which means I can move 10 feet further this round!" And all this d20 headache comes from only three spells, one magic item, and one class ability in effect!

Actually, I'm talking about more here than just a modifier: a + or - number. I'm talking more broadly about all effects--which can also include spell or magic effects, character abilities (including feats, maneuvers, powers, etc), and conditions (flanked, paralyzed, fatigued, etc). Even damage taken--especially to attributes/abilities--can be seen as an effect. Some effects (such as conditions or states) do not affect rolls but instead determine what actions are even possible.

So, with this broad definition in mind, the problem remains: how to streamline the tracking of all the active effects during a game? The first step is to look at what exactly comprises an effect:

Size (or condition)

First is there is usually some positive or negative modifier to a die roll. Alternatively, if a condition, the effect may instead limit or allow certain actions, or else represent a number of different modifiers. For example, being entangled might impose a -1 to all physical actions, reduce speed to half, and prevent all spellcasting.

Applicability

A modifier is usually applied to only a certain subset of rolls, such as only a certain skill or skill group. For example, a modifier might apply to all Jump checks, all saves, all combat rolls, or all STR-based skill checks. (As mentioned above, this applicability can be simple to add on to existing inherent character modifiers, or it may require recalculating other modifiers.)

Type (Stacking/Overlapping)

Modifiers are not always cumulative. Some may have a type or category that determines how they stack or overlap with similar modifiers. For example, in d20, natural armor stacks with mundane armor, resistance bonus do not stack (the highest applies), and damage or energy resistance overlaps.

Duration

Effects can either be inherent/permanent (determined at creation time from things such as skills, attributes, or even worn magic items) or circumstantial/temporary (determined at play time based on the current surroundings, spells in effect, relative positions of the characters, etc.) Temporary modifiers usually expire after a certain number of rounds, minutes, hours, or days. Or (even more of a headache) the modifier may vary over time, such as gradually degrade. (In general, it is temporary effects that require effort to track.)

Source

Effects usually have a source that determines how the effect can be canceled or manipulated. For example, magical spell effects can often be dispelled or nullified in anti-magic fields. Or certain defenses may exist if an ability damage effect comes from an undead creature but not if it comes from disease or poison. (Knowing a source often implies its type, applicability, and duration, but only indirectly.)

Visibility

Finally, there is the question of whether other characters/players can tell if an effect is currently in place. First, there's the question of how apparent effect is in the game world. Secondly, there's a question of how other players track global effects, even if those effects are not apparent to their characters. For example, one character may be exuding some 30-foot healing aura that other players must be made aware of whenever their characters enter the range.

Now that we've had a look at what goes into an effect/modifier, we can start thinking about how to simplify the tracking process.

Dynamo/Fluffy: Lily and Will

Last weekend, S. and I tried out Dynamo, which is a mechanic I've been working on for co-GMing. It's still rather tenuous at the moment, but we tried alpha testing it anyway.

For the gaming system, we used Fluffy, which also needs some testing.

Dynamo (pre-game): On Saturday, while strolling around town, we hashed out the general setting, which was basically film noir, but with slightly updated fashion and Hong Kong action style gunfights and vaguely Lovecraft-like occult powers at work in the background.

On Sunday, we came up with a list of motifs: scenes and elements that captured our desired genre and tone. Our list:

-obscure dying last words
-"damsel" in distress
-all a lie?
-PI gets roughed up
-PI gets dragged in by the cops
-PI offered a bribe
+a dusty, ominous tome
-connection to the decadent element and/or seedier part of town
-tangible evidence of evil (but evidence lost by end of tale)
+a disturbing figurine
+a shoot out
-the good bad guy
+betrayal
-no penalty for shooting a pistol in each hand
-an extra gun somewhere

The +s are those motifs we've used so far (see below). The last couple are almost aspect-like--an idea I want to work on more. It'd be nice to have "themes": general/story aspects you could invoke for a plot point. So, here, any character could spend a plot point to fire two guns without a penalty or to whip out another gun.

Fluffy (char creation): +6 attribute points to spend across 8 attributes. 1 attribute point can be traded in for 4 skill points. A skill or attribute can be reduced to raise another. Later, once the game has started, you can spend a plot point to add a rank in an unbought skill (+1) or to add quirk/hook and invoke it. (We haven't used this last rule yet though.)

LILY McNESSA (S.'s character)
Combat +2
Health: +1
STR: +0
DEX: +1
PERC: +0
KNOW: +0
APPL: +2
SOC: +0
[S.'s didn't bother buying specific skills.]

--Honor (sees things through)
--Doesn't smoke or drink, but has a Tootsie-Pop addiction.
--Weirdness/occult magnet.

Equipment: Beretta, boots, switch-blade, pentacle.

Background: Boxing gym father (died of heart-attack). Catholic maternal grandmother. Apprenticed to a detective, who was bumped off and she had to solve the case of his murder. Owns her own agency now.

WILL STANTON (my character)
Combat: +1 (Guns +3, Brawling -1, Defense +1)
Health: +1
STR: +0
DEX: +0
PERC: +0 (Taste/smell +1)
KNOW: +2 (Occult +4, Languages +3)
APPLIED: +0 (Forgery/docs +1)
SOCIAL: +0 (Disguise +2, Read Person +1)
[1 extra skill point left, which converts to an extra plot point.]

Quirks/Hooks:
--Not respected in occult field (too young; methods too hands-on)
--Needs glasses to read
--Protector of innocent

Equipment: pistol (auto) in shoulder holster, derringer.

Background: Works as office clerk for Lily, both to pay the bills and because occult occurrences seem to follow her.

Dynamo (the game): We then came up with an initial conflict/inciting incident: a job to recover a strange box. This quickly turned into the first scene, with S. taking the lead in narration.

We didn't pause to do the story sketching phase of Dynamo: determine possible conclusion(s) and map out a few NPCs and waypoint scenes along the way. It seemed obvious at first: the story ends when we solve the case. But, though it seems like it might have taken some of the mystery out of things, having a few waypoints added to the motif and character hooks list probably would have helped. Doing so still would have left all the specific details that need to be adlibbed.

We then traded the limelight/narrative control (represented by a large silver coin) after each scene. Scenes seemed to be naturally delimit themselves, and generally corresponded with locations (as I suspected they would).

Here's how the story progressed:

Stolen item job: box, with gold corners and inlay. [S]
--Client: officious/fastidious business man, paying 3x normal rate
--Stolen by hood/crime lord: Citizen Frank Valenti
--Lily knows one of Frank's men from father's gym: Rocko.
Go see Rocko [Z]
--Uncooperative; knows nothing of Frank's "ivory" interests
--Fresh pink scars on hands
--Suggests we talk to Richard Ontario, Frank's art dealer
Break into R.O.'s office [Z]
--Past the lobby security guard by using fire escape
--Odd shrunken head in desk drawer
--Frank's file of purchased: mummified infant+cat, cabinet of curiosities, grotesque Venus-like statue, death-mask of Mata Hari.
--Current file: large cargo (6'x6'x4') coming in from Ivory Coast.
--Invite to Frank's party.

About this point, we took a break. This was tiring work coming up with the whole story on the fly.

Dynamo discoveries: GMs probably won't want to keep the limelight, as I first thought--it's a lot of pressure. And so it should probably regularly pass back and forth in turn (unless someone wants to pay to keep it).

Also, there's probably no need to spend plot points to contribute to the story as a player. The narrating GM should still be able to deny a contribution, but we found that any input was welcome, as the narrating GM was usually willing to grasp at any straw. S. did a great job of contribution, producing most of the items in Frank's file of purchased artifacts. She also was always conscious of placing a good solid lead into the next scene, providing both the connection to Rocko and the party invite. Overall, we largely ignored the exchange of plot point chips in relation to narration control.

The incorporation mechanic worked well. For example, in the Rocko scene, I threw in the "ivory" comment and Rocko's scars in the hopes that this would come to mean something later. "ivory" became "Ivory Coast" in the next scene, and S. built on it again later (see below). This is now becoming a clear thread.

The original idea was to work in a motif whenever possible in order to earn a plot point. However, S. suggested placing the motifs into a hat and drawing one out each scene and then having to work it in. We ended up rolling a die to pick randomly--but with the same result. This was a good idea, as the added constraint actually made scene construction easier, rather than harder.

After our break, we returned to the game:

To Frank's party [S]
--Check out his museum, seeing most of his artifacts.
--See box... but Frank walks in with Rocko [betrayed!]
--Escorted out, back to party
Rival gang arrives (Oscar Calzone) and start shooting up the place [shoot out] [Z]
--Dash back to museum room in confusion [S.'s contribution]
Museum room again [S]
--Box is gone... but do find a [dusty tome]
--Will recognizes it as containing African hieroglyphs, from Ivory Coast region.
--Slip out the French windows with book while shootout rages and alarms blare.


Fluffy discoveries: We never actually rolled the dice during the whole session. I almost asked for a roll to pick the lock of R.O.'s office, but I wasn't really prepared to handle a failure, and in any case, a Good Applied(lockpicking) skill should have been sufficient for Lily to just check (Take 0) on the roll anyway. This suggests that Dynamo might work best with diceless games (though rolling is certainly possible for those who miss it).

Overall, a very fun session with some good design feedback.

D&D: Omri, Cheap Battlegirds, and Player Caution

S. and I played a bit of D&D this weekend. We overlaid some clear Omnigrid quilting rulers over the white board for a quick battle grid, and then used some old chess pieces for miniatures. Worked out quite nicely! Normally I just use the white board, but it always takes an extra minute drawing the grid before each combat, and then I have to erase each character's mark before moving them to another square. There's just something quick and satisfying about miniatures... but also something a bit pricey when doing it right! The chess pieces were a nice compromise.

I was also reminded what a cautious player S. is. While taking out a goblin army outpost that had her party pinned down with crossbow fire, her character, Omri, burst out of invisibility within the enemy dugout and immediately decapitated an orc and a goblin. Half the remaining seven goblins in the room were still flatfooted, and the rest of her party was about 3 rounds away. Omri had both Shield and Sheild Other in effect, and I was envisioning an awesome cinematic bloodbath as Omri tore through the remaining goblins while her friends dashed in, their own crossbows and spells blazing. But S. chose to retreat rather than press her advantage! So now the entire scattered goblin army--not just this one outpost--knows they're coming.

But upon reflection, I think this retreat is not so surprising. For one, Omri--who had never seen either orc or bugbear before and failed her Knowledge check to identify the orc before her--thought that she may have killed the commander bugbear they were coming for. (And S. did a great job of staying in character even after I accidentally let slip that it was an orc.) Secondly, I may not have been clear as GM that the army had little outposts scattered around these mountain valleys, watching all the passes, that this was just one such outpost that could alert the others given a bit of time.

But most importantly, it's just a (nice) surprise to have a player that isn't unconsciously meta-gaming the whole time, that doesn't have a good idea of the challenge rating of the current encounter, and who's worried more about avoiding unnecessary combat than just killing everything that moves and nicking all its stuff. And, I must say, if it were me in Omri's place, I don't know if I'd stand my ground in a cramped dugout when seven armed and outraged goblins turned to stare at me! Overall, it was still an impressive hit-and-fade attack... marred slightly only the by the 3 crossbow bolts the party took in the back as they fled into the surrounding hills.

Anyway, it was just another lesson in that constant GM balancing-act: providing enough challenge that players have to be cautious, but not so much that they're afraid to just rush in--swords swinging--when the situation calls for it. I guess that's the trick: successfully conveying whether or not the situation calls for it.