TRPG General Chat - 2019-05-06 (page 2 of 2)

« first day (3140 days earlier) ← previous day next day → last day (1824 days later) »

7:34 PM

Tried a different angle on the "how much does AC actually benefit a character?" question, I think I've come up with a better way to chart the data.

╔═══════════════════════════╤═════════════════════╤════════════════════════════╗
║                      Name │ Greatsword (STR 10) │ Greatsword (STR 16 + Prof) ║
╟───────────────────────────┼─────────────────────┼────────────────────────────╢
║  Avg. Rounds Alive @ AC 2 │               2.130 │                      1.637 ║
║  Avg. Rounds Alive @ AC 3 │               2.245 │                      1.637 ║
║  Avg. Rounds Alive @ AC 4 │               2.373 │                      1.637 ║
║  Avg. Rounds Alive @ AC 5 │               2.517 │                      1.637 ║

This table plots the average number of rounds a defender with the given AC will survive against an Attacker with the given weapon.

Note: each of these cells has roughly 4.806x10^220 possible outcomes. It also has a margin of error around the fact that it only iterated up to 50 rounds of combat (meaning all possible outcomes of 51 rounds, 52 rounds, 53 rounds, etc. are all treated as 51+ rounds) so these numbers are inconsequentially inflated over their real, mathematically true values.

I feel that it is likely that the rounds after 51 are not statistically significant

lol

@goodguy5 Depends on the AC. For low ACs, the difference between 25 rounds simulated and 100 rounds simulated is trivial. For high ACs, the difference can actually vary by quite a lot.

I'm still trying to figure out a good system for determining the minimum number of rounds that need to be calculated to arrive at the true answer; I think I may need to apply some Linear Algebra instead.

2309402969223664982470059021038277867632036617291686237437284890065201564055108565729636735744987978694671350965526224601098898473204370021617288223651149177842586734168519702960547039709473688421416898416657906334890561024163835653711024019572866346735729097801882632996545714089069687891325746797154175658567113724609018294075400159766773760000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000

@Xirema Is a number my program spat out as "the number of possible outcomes of this one simulation".

that is absurd

Pro-tip: if you ever see a number that big, you should quit your job and live as a hobo, because you've discovered the utter meaninglessness of the material world, and you're not benefiting anyone by pretending you can still participate in society after being exposed to this terrible truth.

7:48 PM

Isn't that fairly simple to solve without monte carlo?

@kviiri This isn't Monte-Carlo in the technical sense.

@kviiri I do believe, like I implied, that a proper application of Linear Algebra would at least produce the objectively correct averages.

What are you actually plotting there?

rounds alive at 10hp

@kviiri Each cell is the average of a table that looks, more-or-less, like this:

Ooof, that does not render, hang on...

Wait, a character with 2 AC would, at average, survive a bit over two rounds? That sounds a bit questionable to me

Hm, then again the chance of dying on the first round is only around 16%...

7:56 PM

On average requires two hits. Slight change of a miss

@Someone_Evil ^

user image

What does the "trials" column represent?

@Xirema This is for 10HP, 24AC, against a Proficient Greatsword wielder

I think the value is 'Average round the are killed on' which is subtly different. The median should be them being felled on the second round/attack

@kviiri The number of unique dice rolls that result in that outcome.

7:58 PM

Oh ok

The numbers on the trial column goes of the screen, but I think you've made whatever point you were trying to make

I'd think you could get some neat closed forms out of this using Markov chains or perhaps even some linearity of expectation tinkerin'.

@A.B. You can squash a lot by switching to the mobile interface (bottom right corner).

That table in particular is giving me an idea about how to solve the "how do we know if we've done enough trials?" question: When the odds of the last row drop below a certain threshold, we stop. In that case, the odds dropped below 0.1% by round 79, so we could have just stopped there and arrived at a semi-accurate average.

Conversely, for AC2, that happened at round 6; no need to iterate further.

@Xirema When you say "odds dropped below" do you mean the change thereof?

8:05 PM

@Someone_Evil The "Pass Odds" column, which didn't break from the floating-point overflow. It is "the odds that we'll roll this number or higher".

(If my BIGNUM implementation weren't rubbish, the "Odds" column would be working correctly)

@Xirema OK, I don't understand enough about the method you've used and/or the data you've given to say anything intelligent about it

@Xirema Have you tried scientific notation for numbers greater than a billion?

@Someone_Evil I run into that with these all the time.

@GcL That's what it does; but my BigRational class I had to borrow from some university's Java course because I couldn't find any Open Source implementations for Java.

@Xirema This is an improvement over the previous table.

So its doubleValue() method is hacked together and bad.

8:10 PM

@Xirema Yet another reason to use an existing, established, and tested framework.

@Xirema I can add those descriptions to nearly all the code I've ever written or seen.

@GcL Which is what I would do—IF THERE WERE ANY!!! Mr_Turner.jpg

@Xirema R or Numpy to name a couple

@GcL Am writing code in Java. There's an excellent implementation in C++ I use when I revise the code, but I don't have access to my C++ workspace right now.

75

A: How to build an evil campaign

I've played in and run evil campaigns of various sorts in both 3.5 and 4e (though not 5e, I think my learning will transfer), and run into a lot of problems: My Guy Syndrome comes up a lot, as does a tendency to default to a regular D&D storyline only with more stealing of spoons and kicking of p...

Yet another reason I prefer C++ over Java.

8:13 PM

@Xirema What's the stats library for C++ ?

@GcL I dunno. I've just been implementing everything using the boost.multiprecision library since it gives me the bigint and rational types I need for this program.

Specifically, boost::multiprecision::cpp_int and boost::multiprecision::cpp_rational, which both work basically flawlessly and are very fast.

@A.B. I'm having a hard time getting a bead on exactly what your family's RPG needs are yet; try reading Lady Blackbird and see how it feels?

Even if it's not right, your response to it can help calibrate what we look at next.

@BESW My personal answer is I want to play, but no one will play new games with me :'(

@BESW Ooo... good suggestion.

@GcL It's got romance and swashbuckling and space magic, and clear rules that specifically push storytelling moments, and GM advice that encourages collaboration, and a setting which means the group can crib from a lot of popular culture when they can't think of what's happening next.

8:28 PM

@BESW Like firefly

And Star Wars.

@A.B. How are you at swearing in Mandarin?

And Pirates of the Caribbean.

@BESW Well sure if you like guidelines.

user15026

@BESW those are all fun sounding things!

8:33 PM

@Ash We should play it, too!

@BESW Cribbing from pop culture is like my go-to strategy for anything.

Meeting new people? Hide my insecurities by making obscure references to early 00's web culture.

@Yuuki I thought that was hoisting a subtle flag for others of the same group?

@GcL Ah but explaining the reference means more time talking about anything other than myself.

Usually local sports teams are the smoke screen to hide behind. A quick google for sports on the good old pocket computer yields enough material to nod along.

@Yuuki Sure, but now you're stuck talking.

user15026

@BESW yes! I would like that.

8:38 PM

Talking is fine. Going out is already going to spend a lot of mental energy so I might as well go home exhausted and have a good night's rest.

@Yuuki Same rationale as at the library, might as well read some stuff, or at the gym might as well lift a thing?

Pretty much. I'm already in front of people, I might as well pretend to be a functioning member of society while I'm here.

Hot RPG Questions

3

Q: Can the attacker of a Sanctuary'd target choose not to attack anyone else?

Sanctuary reads [...] any creature who targets the warded creature with an attack or a harmful spell must first make a Wisdom saving throw. On a failed save, the creature must choose a new target or lose the attack or spell. Imagine a case where two orcs (allied to each other) are standing...

Hot RPG Questions

9:02 PM

1

Q: Do the Zhentarim fire members for killing fellow members?

A PC became a member of the Zhentarim and two weeks later killed another one. I don’t know if there are rules within the Zhentarim against killing fellow members. Would the Zhentarim fire a PC member who killed another Zhentarim? He hid it from the other Zhentarim but after that was arrested by...

dnd-5e forgotten-realms

@GcL Sometimes I feel like a one-note record, just suggesting Lady Blackbird for everything, but it's often genuinely a good place to start.

9:26 PM

@HotRPGQuestions speaking of Sanctuary, I think there's quite a few answers to questions about "does X break Sanctuary" that need to be updated in accordance with the 2018 PHB errata that added "[the caster] deals damage to another creature" to the list of things that ends the spell

@V2Blast I thought the Kali thing was in bad taste, but I don't think Sanctuary really broke until the Hollow Earth plot... oh wait.

@BESW wat

Sanctuary (TV series)

Sanctuary is a Canadian science fiction-fantasy television series, created by Damian Kindler and funded largely by the Beedie Development Group. The show ran for four series from 2008 to 2011; a fifth series was planned but never made. The show is an expansion of an eight-webisode series that was released through the Internet in early 2007. Seeing the success of the web series, Syfy decided to buy the broadcast rights to the series and pay to re-stage the series in a season of thirteen episodes.The show centers on Dr. Helen Magnus, a 157-year-old teratologist (born August 27, 1850), and her team...

ah

never heard of it

I won't say it's good, but it's often a lot of fun.

It would've been much better if they'd given Magnus a clearer moral arc on realizing that views which were compassionate and progressive in Victorian London are... not, anymore.

But Nikolai Tesla is an electric-powered vampire and the IT guy is a werewolf, so I can't complain too much.

9:35 PM

@BESW I remember that seeing trailers and posters about that.

Never watched it though.

The people making it were clearly having a lot of fun, many of them were friends from Stargate: Atlantis and it was a low-budget "not getting typecast" vehicle for Amanda Tapping after she left that show.

lol

it was a little weird that she basically just grabbed weird animals from their natural habitat and stuck them in her little personal zoo

in retrospect

like, I can remember one thing that was super intelligent and really didn't want to get snagged and they still basically had a whole episode of fighting it to drag it there with them, it was like a Super Octopus or something something that lived in water had tentacles and was super smart anyway

which to me screams Super Octopus anyway

but like, it took down a plane because it really didn't want to be in a cage

The supposed principle was that the creatures in the Sanctuary were ones which would be miserable or die elsewhere--often because they'd be hunted by humans, either for sport or resources, or because they were dangerous and called attention to themselves.

yeah I know but like

In practice, either the writers or the characters (probably both) tended to forget that.

9:48 PM

what is she going to do, take literally all the supernatural weird animals out of all their habitats?

that's what it felt like sometimes

that episode especially as I recall

yeah, like I said, the show would've been a lot better if it'd made that question into a significant arc.

like this Super Octopus literally seemed like it was perfectly capable of taking care of itself and outright hated captivity

@BESW yeah I agree

Same thing with the Flash and the Almost-Definitely-Crimes-Against-Humanity-Makeshift-Jail-I-Mean-Repurposed-Particle-Accelerator.

ah yeah that

That particular episode of Sanctuary was... not quite as you remember it, but the general principle stands.

9:52 PM

well they still put people in a secret prison without a trial

10:21 PM

@BESW ah well it was a while ago

My recollection goes off of imperfect memory from several years ago

Oh I completely forgot about the other thing it was fighting

Literally only remembered that now

Like the sort of scorpion thingy

That raises a whole new question for me, why the heck are those two animals deadly enemies when they don't even live In the same environment?

It was highly contrived.

yes indeed

10:44 PM

newbie DM again. When you're writing the plot for the session, what are some things y'all do to balance linearity vs the 'adventure' aspect of a mission? Outside of dungeons, of course, where the whole point is gameplay

@Regress.arg I find that play-to-discover is much better than play-to-uncover. Prepare NPCs with plans, prepare locations with interesting stuff going on. Prepare what is happening and then play to discover what happens next.

what's the difference between discover and uncover?

Discover is where the GM doesn't know what's going to happen. You set things up, and work with your players to during the game to improvise stories forward from that setup.

Uncover is where the GM has planned what will happen and the players are just trying to ask the right questions, go the right places, solve the right puzzles, to uncover your pre-made story.

In other words, I'm saying that you don't write a plot. You write a setup, and the players come in and turn it into a plot in which they're the main characters.

If you know your NPCs--what they want, how they plan to get it, what they won't compromise to achieve their goals--then when the PCs barge in like a bowling ball that's jumped its lane you don't need to have a plot for what happens next. What happens next is that the PCs exercise their agency and the NPCs respond appropriately.

(I'm assuming you're not using a game system which has much support for complex non-combat interactions, and you're just going to be using mostly nonmechanical free-form play.)

11:05 PM

Puts me in a difficult situation, since the players have also bought into a campaign with a strong plot-this isn't a "learn about the evil ring you have to escort to the fiery volcano" kind of affair

Ah, they're expecting to be led by the nose but you don't want to make it too obvious?

11:22 PM

if you've DM'd in the past, how have you planned out how your sessions are going to go? What was your process?

Specifically D&D? It depended on the campaign and the group. My first campaign used a very complex world from a fantasy novel, and I just dropped the players into it. For a given session, I'd prep the mechanics for the area they were visiting, and refresh my memory about the NPCs' plots and motives and what sorts of things the PCs would run into. I didn't plan what would happen during the session, though.

For other campaigns, sometimes I'd use a "you're working for an NPC" structure to give the players a diegetic reason to follow a pre-made plot; that way I could influence their goals and motives for an adventure by giving them instructions from their boss, which made it easier to predict what would happen and plan how the world would respond.

But if there was something I needed to have happen, it either had to be geographically inevitable (like entering a particular room in a dungeon where you can't go around it) or be included in my setup of the first scene of the session.

For campaigns where I wanted a complex plot to go the way I predicted, I needed to get clever with Xanatos gambits: that's where the players have control over their choices and can gain some kind of independent success, but the things which need to happen for the plot to move forward are unrelated to what the players can control.

The trope-naming example is from the TV show Gargoyles, where the villain arranges for a prison break and the heroes spend the whole episode catching the escaped prisoners. So they feel like they did something useful. But the real goal was to get a different prisoner released legally on parole, which was granted because she stayed in the prison when she could've escaped with the others.

So the villain's goal had already been achieved no matter how the heroes responded to the prison break.

Over time I combined these processes, so that prepping for a session involved identifying what the players' current goal is, and how the world would complicate that goal based on their previous choices (does an old enemy show up for revenge, or a grateful town throw a parade in their honor which attracts unwanted attention on a stealth mission, etc).

The best sessions now, I find, happen when the PCs have a simple, straightfoward goal but the world keeps putting obstacles in their way which the PCs have to respond to. So prep involves defining the path(s) to the goal and preparing a set of obstacles to use as needed.

Hot RPG Questions

11:38 PM

5

Q: What’s the interaction between darkvision and the Eagle Aspect of the beast, if you have Darkvision past 100 feet?

Directly based off of the question, What’s the interaction between darkvision and the Eagle Aspect of the beast?, what happens if your Darkvision extends past 100 feet? This question is focused around fact that the Eagle totem features states "You can see up to 1 mile away with no difficulty, ab...

dnd-5e class-feature vision-and-light barbarian

Recently my prep is less about making all that up beforehand, though, and more about preparing leading questions for the players which guide them to collaborate in making it up as we play.

One of the reasons I have really liked running GSS so far is because it's a good showcase for me from a GM perspective of how PC creativity can make the same story significantly different

GSS is also just great but because it's been easy for me to run it, and because I already like it a lot, it's been the first system I could do that with without even trying

Whereas my 4e experience GMING has mostly been flop after flop of me trying and failing to make the story work while balancing the mechanics

But yeah anyway, the point is I really see value in that GM style

Of making the outline of the story and pushing it when the PCs don't know what to do, but when they do know what they want to do letting them be a very active voice in what the story dynamicly becomes

It's also way less prep work to let them actually drive story and only stepping in when you really have to

You still need to provide structure, and it's maybe hard sometimes to balance that with giving so much agency to the PC's, but it's not nearly impossible

@Regress.arg Generally, I want to start a session by introducing a clear obstacle or challenge which the party cannot ignore. The obvious choice is "someone is trying to kill you," but context can make it almost anything.

@Regress.arg Like BESW said, don't try to plan out the session. As DM for a D&Dish game, you can control:
1. The first minute of the first scene
2. Who the NPCs are
3. What the NPCs want to do
4. Some premade environments

Assume that if you put something in the same scene as the player characters, the players will then assume they can interact with it

11:54 PM

An example from one of my last D&D campaigns:

> You're guests at an international diplomatic party hosted by the Dragon Pope, and you hear rumors that the fighter's home kingdom is being threatened with civil war by someone who wants to de-throne the rightful prince and put the fighter on the throne instead. You see the prince (whom the entire party considers personally loathesome and suspects is unfit to rule) walking toward you angrily from across the room.

The early parts of the session are important, that's where you use your limited DM powers to set the pace. Establish the rules and goals early on.

That particular session ended with helping the prince de-escalate the rebellion before it became a civil war, by having the fighter and the rebel leader become advisors to the prince in the hopes he could become more worthy of the throne.

But I had no idea that would happen; I was 60% sure the fighter would punch the prince in the nose in the middle of the Dragon Pope's ballroom and shout "LONG LIVE ME!"

Kamola the Prince-Maker

Aug 28 '13 at 10:46, 19 minutes total – 55 messages, 4 users, 0 stars

Bookmarked Nov 15 '13 at 17:39 by BESW

« first day (3140 days earlier) ← previous day next day → last day (1824 days later) »

Transcript for

May '196

TRPG General Chat

Main chat room for tabletop role-playing games. Anyone can ask...

dice pen-and-paper roleplaying

join this room
about this room

00:00

06:00

12:00

18:00

all times are UTC

site design / logo © 2024 Stack Exchange Inc; legal

mobile