D&D General D&D isn't a simulation game, so what is???

[dave2008]

@DND_Reborn , I found this over at UA reddit and thought you might like it (it came with a host of other standard actions, but this one made me thing of you):

You could obviously use this with "monsters" too and adjust the multiplier depending on the animal.

 

[clearstream]

Sure. I'm not saying it's uncommon as a trope. Just that I find it fairly bizarre: before we get to the energy involved, and the mechanism of delivery (look at the amount of energy humans have hard to burn to increase global temperatures by 1 or 2 degrees!), what about the maths of the necessary weather forecasting? How would Classic Traveller's computers even handle that?

Whereas there is at least a surface-level consistency between having air/rafts with their grav modules, having starship with anti-grav manoeuvre drives and grav plates, and having jump drives.

To me, if I am worrying about energy for weather control, I would also worry about computation for the maths to solve anti-grav. There's a strong connection between being able to frame and solve advanced equations, and physics-suspending drives. Or I'd just suspend physics equally lightly for drives and weather.

I guess I am back to the obvious point that Traveller is space-opera not futurology. There are many equivalents of hawks flying too slow in Traveller. Which raises doubts about the OP's concept of simulationism, once Traveller is accepted as an example. (One could simply characterise it as a flawed example... which might be somewhat confounded if we can't agree which parts are the flaws!)

What really makes Traveller simulationist, if it is in truth very unrealistic? Is it the attempt toward realism, even if failed? Is it consistency, even if Traveller has large inconsistencies. Is it just sufficient granularity? I don't know that this thread has yet robustly defined simulationism.

Back to - The key feature of simulationist mechanics is that they model/express the in-fiction causal processes. A slow hawk is still modelling an in-fiction process. I suggested that simulationism can take any world as its reference, including worlds where hawks fly slow! The definition still feels elusive.

 

[Hussar]

To me, if I am worrying about energy for weather control, I would also worry about computation for the maths to solve anti-grav. There's a strong connection between being able to frame and solve advanced equations, and physics-suspending drives. Or I'd just suspend physics equally lightly for drives and weather.

I guess I am back to the obvious point that Traveller is space-opera not futurology. There are many equivalents of hawks flying too slow in Traveller. Which raises doubts about the OP's concept of simulationism, if Traveller is accepted as an example. (One could simply characterise the example as flawed or not ideal... which seems somewhat confounded where we don't agree which parts are the flaws!)

What really makes Traveller simulationist, if it is in truth very unrealistic? Is it the attempt toward realism, even if failed? Is it consistency, even if Traveller has large inconsistencies. Is it just sufficient granularity? I don't know that this thread has yet robustly defined simulationism.

Well, I'd look at the planet construction mechanics first off. Before I get too deep though, I'm only passingly familiar with the rules of Traveler, so, take what I'm saying with a grain of salt.

But, if you follow the planet creation mechanics, you get a pretty workable planet. There's a pretty clear line between the mechanics and what you'll find in the game.

Note, compare that to the random dungeon generation mechanics in D&D. There's no real rhyme or reason there. Running the random generation, you get a funhouse dungeon without any particular logic to it. Mostly because the random dungeon generator isn't meant to create something that could actually exist - it's only there to create something to play in. You don't have to worry about pesky things like ventilation or waste removal or why there's fifteen orcs in this room with a pack of ghouls down the hall. We need an adventure for level X PC's, and poof, here you go.

I'd say that the point of simulations is that when you run them, you get results that make some degree of sense to the observer. Randomly pulling stuff out of a bag isn't simulating anything. Running a series of interconnected mechanics that create a final product where you can follow the process A to B to C is what a simulation is. In other words, the mechanics have to generate information beyond win/lose.

 

[clearstream]

Well, I'd look at the planet construction mechanics first off. Before I get too deep though, I'm only passingly familiar with the rules of Traveler, so, take what I'm saying with a grain of salt.

But, if you follow the planet creation mechanics, you get a pretty workable planet. There's a pretty clear line between the mechanics and what you'll find in the game.

For instance the donjon Traveller System Generator.

Note, compare that to the random dungeon generation mechanics in D&D. There's no real rhyme or reason there. Running the random generation, you get a funhouse dungeon without any particular logic to it. Mostly because the random dungeon generator isn't meant to create something that could actually exist - it's only there to create something to play in.

Having a "logic" to it makes sense. Isn't that what others mean by "consistency"? I don't think randomness prevents a game being simulationist. The Traveller System generator randomly pulls stuff out of a bag, rolling against tables. The DMG dungeon generator randomly pulls stuff out of a bag, rolling against tables.

You don't have to worry about pesky things like ventilation or waste removal or why there's fifteen orcs in this room with a pack of ghouls down the hall. We need an adventure for level X PC's, and poof, here you go.

Simulations are always incomplete. For every - "I care about waste removal", another can respond with "I care about the reproductive behaviour of natural creatures". Nor I think is convenience of executing the simulation at issue, otherwise the donjon Traveller System Generator would prevent Traveller being simulationist on grounds of convenience of execution. Poof, a UWP.

I'd say that the point of simulations is that when you run them, you get results that make some degree of sense to the observer. Randomly pulling stuff out of a bag isn't simulating anything. Running a series of interconnected mechanics that create a final product where you can follow the process A to B to C is what a simulation is. In other words, the mechanics have to generate information beyond win/lose.

I was thinking along the same lines - "make some degree of sense to the observer". Seeing as observers disagree about what makes sense, they disagree about whether a simulation is good. But do they disagree on what is a simulation at all.

Folk here have counted RuneQuest, the ICE system, and Traveller in as simulation, and 5e out. However, mechanics in all those games generate information beyond win/lose, and there are mechanical chains creating final products in every case. Either 5e is simulationist, or our definition isn't there yet. All simulations S have a reference R. We can't just say that when R is our real world we are simulationist, because Glorantha is not our real world. Not now, not in the past or future. Glorantha was created by its deities from the Primal Void of Chaos. That's not a belief that some folk living there have about their world, it is a fact about the creation of their world.

It's not that I dislike the sense of what you are saying. I just don't see how your requirements for a simulation yet separate what we want to call simulationist from what we don't want to call simulationist.

 

[Hussar]

There is an additional point to think about though too. When someone says this or that system is a simulation, I doubt they mean 100% of the game. There are all sorts of fuzzy, grey areas in between. When people say 5e isn't a simulation, for me, it's because the main component of the game - combat - doesn't actually simulate anything. Combat starts, everything exists in this sort of quantum cloud of possibilities, until we finish combat and we can say A lived and B died. But, as we go through the system, we can't really determine anything about what is going on. The system doesn't tell you anything, really, about how B died any more than chess tells you how that pawn died when the queen took it. All we know is the pawn is gone and the queen remains.

Now, for some people, this is a feature, not a bug. It allows the players to narrate pretty much anything they want and, so long as the table is happy with the narration, fine and dandy. Which pretty much precludes the system from being a simulation. You can't take a simulation, make up any narrative you like and not contradict the simulation.

Now, as far as planet generation goes in Traveller, I'd call this a decent simulation mechanic - not of the creation of a planet but, at the end of the process, you have a planet that is pretty plausible. Granted, this mechanics aren't really player facing, so, there's no narrative going on here really. It's telling you what is, right now, rather than how you got there. But, it creates something that is reasonably plausible, so, good enough. Whereas the D&D dungeon generator doesn't give the slightest toss about plausibility. All of the inputs are based around making an interesting game - monsters are restricted by CR ranges, DC's for various hazards likewise are governed by the level of the dungeon, a purely game concept that has no meaning in the context of the game world.

 

[clearstream]

There is an additional point to think about though too. When someone says this or that system is a simulation, I doubt they mean 100% of the game. There are all sorts of fuzzy, grey areas in between. When people say 5e isn't a simulation, for me, it's because the main component of the game - combat - doesn't actually simulate anything. Combat starts, everything exists in this sort of quantum cloud of possibilities, until we finish combat and we can say A lived and B died. But, as we go through the system, we can't really determine anything about what is going on. The system doesn't tell you anything, really, about how B died any more than chess tells you how that pawn died when the queen took it. All we know is the pawn is gone and the queen remains.

That's really not the experience at our table. Our combats evolve continuously, with much back and forth, precarious moments and so on. I can't recall a death in which we din't know exactly how B died. For example, my players still recall the sequence of events leading to M being bitten to death by giant ants, and E still feels bad about their actions in that combat.

But I believe you that 5e can be played in the empty way you describe, provided one employs the available mechanics sparingly and refuses to read the game state as having import in the fiction.

Now, for some people, this is a feature, not a bug. It allows the players to narrate pretty much anything they want and, so long as the table is happy with the narration, fine and dandy. Which pretty much precludes the system from being a simulation. You can't take a simulation, make up any narrative you like and not contradict the simulation.

For me, not a bug, but an obdurate refusal to grasp and make use of the riches on offer Say X lacks web in their spells known or taken, then it will contradict the simulation to narrate that X has web. Suppose Q wields a rapier, then it will contradict the simulation to narrate that Q deals bludgeoning damage.

Now, as far as planet generation goes in Traveller, I'd call this a decent simulation mechanic - not of the creation of a planet but, at the end of the process, you have a planet that is pretty plausible.

I don't agree as to plausibility. For example, almost all planets will have starbases. I don't agree that it will be plausible in the future for most planets to have starbases. At the end, you will have planets suitable for use in a space opera.

Granted, this mechanics aren't really player facing, so, there's no narrative going on here really. It's telling you what is, right now, rather than how you got there. But, it creates something that is reasonably plausible, so, good enough. Whereas the D&D dungeon generator doesn't give the slightest toss about plausibility.

Can you define what you mean by plausibility? Because the dungeon generator aims to produce a plausible dungeon, where the definition of plausible perforce already compasess the existence of dungeons.

All of the inputs are based around making an interesting game - monsters are restricted by CR ranges, DC's for various hazards likewise are governed by the level of the dungeon, a purely game concept that has no meaning in the context of the game world.

You mean the guidelines that suggest what will be a balanced challenge for a group? How does that restrict monsters? Do you take those guidelines to mean that the world's ecosystem is determined by the current sum levels of one PC party? And, I suppose, shifts tectonically when they level.

 

[Hussar]

That's really not the experience at our table. The combat evolves continuously, with generally much back and forth, precarious moments and so on. But I believe you that 5e can be played in the empty way you describe, provided one employs the available mechanics sparingly and refuses to read the game state as having import in the fiction. I can't recall a death in which we din't know exactly how B died. For example, my players still recall the sequence of events leading to M being bitten to death by giant ants, and E still feels bad about their actions in that combat.

I disagree. Nothing in the game tells you that M was bitten at all. Maybe M died of embarrassment for slipping in Ant guts. Since the mechanics don't actually tell you anything, any narrative you choose to use is acceptable. That doesn't mean it can't be impactful or good narratives. That's fine. But, the narrative you chose at that point in time was in no way actually supported by the mechanics of the game because there are an infinite number of other narratives that would also fit the results of the mechanics and none of them are more true than any other.

You mean the guidelines that suggest what will be a balanced challenge for a group? How does that restrict monsters? Do you take those guidelines to mean that the world's ecosystem is determined by the current sum levels of one PC party? I can see how you might decide that was the implication.

The random dungeon generator will ONLY produce a dungeon based on the level that you set for the dungeon. So, if you decide to randomly generate a 3rd level dungeon, it will never include anything over about CR 5 or so. The only reason that this dungeon will never have a dragon in it is because it's meant to be played by a party of X level. So, this obviously restricts monsters.

IOW, the level of the dungeon, ie. the game, decides what range of monsters can occur in that dungeon. It's not based on anything else. There's no simulation there at all. It's no more simulationist than shuffling a deck of cards. The choice of what cards are in that deck is dependent on things that have nothing to do with the world of the game and everything to do with what you're supposed to face as threats based on the level of the PC's.

Imagine how that's different from the Traveller world generator. The world generator doesn't give the slightest toss about the PC's. Whether you have a group of newbie characters straight out of chargen, or a group of very experienced Travelers with much greater resources at their command, that world will be exactly the same.

 

[clearstream]

I disagree. Nothing in the game tells you that M was bitten at all. Maybe M died of embarrassment for slipping in Ant guts. Since the mechanics don't actually tell you anything, any narrative you choose to use is acceptable. That doesn't mean it can't be impactful or good narratives. That's fine. But, the narrative you chose at that point in time was in no way actually supported by the mechanics of the game because there are an infinite number of other narratives that would also fit the results of the mechanics and none of them are more true than any other.

Our fictional position was that the ants were intent on biting M, and M indeed died of a series of rolls for said biting. To go from there to "died of embarrassment" is deliberately perverse. What follows is always up to the group.

Suppose in RQ a broo rolls a hit to M's head and deals enough damage to equal or exceed the HP there. If we wish, we can narrate that M falls unconscious and dies two turns later from humiliation at the way the broo looked at them. We don't do that, because we are committed to a gameful narrative, where we will employ a kind of extended language - with a prior assumption of sincerity - and associate M's death with the ant bites or broo hit to head because that follows from our conversation.

How much of our narrative do we need prescribed for us by written game rules? The FKR movement would argue for very little. What you are saying amounts to a denial that FKR play can produce valid narratives, even though FKR adherents self-report that they find their narratives more intense and just as valid as those dictated by pre-written rules. The presence or absence of rules doesn't alone amount to success as a simulation: is it realistic that every time a person has their chest HP equalled or exceeded they will fall and be too busy coughing blood to do anything? No exceptions? Every. Single. Time.

Addy "Oh, there's Jo again, down on the ground busy coughing blood."
Fitch "Yeah, that always happens. Weird no one ever collapses to their knees, swaying and gasping."
Addy "Yeah, that's weird isn't it? I wish I could sometimes go to my knees and..."
GM "Well you can't, so let's move on."

The random dungeon generator will ONLY produce a dungeon based on the level that you set for the dungeon. So, if you decide to randomly generate a 3rd level dungeon, it will never include anything over about CR 5 or so. The only reason that this dungeon will never have a dragon in it is because it's meant to be played by a party of X level. So, this obviously restricts monsters.

Once one has committed to a world consisting of a single dungeon, any following constraints cannot mar the plausibility. But 5e does not envison that the game world consists of one dungeon; nor even that said dungeon be generated randomly. It would be better to read the many pages of material in the earlier sections of the DMG to grasp the facets a GM is guided to consider for their world.

IOW, the level of the dungeon, ie. the game, decides what range of monsters can occur in that dungeon. It's not based on anything else. There's no simulation there at all. It's no more simulationist than shuffling a deck of cards. The choice of what cards are in that deck is dependent on things that have nothing to do with the world of the game and everything to do with what you're supposed to face as threats based on the level of the PC's.

That missapprehends what a simulation is. So long as the dungeon is predictably like a dungeon - itself a highly artificial "world" - the simulation is successful. For background, a simulation exists in two parts: the model and the rules. A simulation is successful if when executed it predicts (produces outcomes) that map to a reference well enough to say useful things about that reference (things what will often-enough prove to be true).

A random dungeon generator is a successful simulation if it reliably produces dungeons of the expected type. That means that the model contains the features desired for the dungeon, and executing the rules results in a dungeon usefully alike the reference (of being a dungeon of that type.) You can probably see that for something artificial like a random dungeon, it's trivially and tautologically true that the simulation is successful. It would only be unsuccessful if the reference were dungeons in some other world. There are no dragon-containing dungeons in our world, and the dungeons in at least some possible worlds will be successfully predicted by the generator.

I hope it's clear that my point isn't to dismiss your arguments, but to point out that there is something else going on, and a risk of picking-and-choosing what we accept as a simulation based on our preferences and experiences, and not based on anything objectively right about what counts as a simulation.

Imagine how that's different from the Traveller world generator. The world generator doesn't give the slightest toss about the PC's. Whether you have a group of newbie characters straight out of chargen, or a group of very experienced Travelers with much greater resources at their command, that world will be exactly the same.

The world generator is predicated on PCs living in a space-opera of the Traveller type. Hence the roll for a scout base for every planet in the galaxy. Because the planets are tailored to fit the envisioned play for the envisioned PCs.

 

[pemerton]

There is an additional point to think about though too. When someone says this or that system is a simulation, I doubt they mean 100% of the game. There are all sorts of fuzzy, grey areas in between.

Right. I've given examples of this upthread: manoeuvring in a vacc suit, in Traveller, is not resolved in a simulationist fashion. Nor is using Streetwise to find a corrupt official. In Rolemaster, PC build is not simulationist in the way it is in BW or Traveller (lifepaths) or RQ (cultures and occupations).

In Burning Wheel, although PC build is simulationist, and setting obstacles for action resolution is, framing and narrating failure - which together drive the game - are not.

What follows is always up to the group.

<snip>

How much of our narrative do we need prescribed for us by written game rules?

What follows, in RM, is not always up to the group. Often it's determined by a roll on a table. The contrast with (say) Dungeon World, or failure in Burning Wheel, is marked.

How much fiction does a successful simulationist system have to prescribe? The salient fiction. What is salient is up for grabs, but there are some traditions as far as RPG combat is concerned: hit locations, damage to armour, etc.

But the other feature of a successful simulationist system is that it doesn't have "free-spinning" mechanics, or mechanics that are triggered by events in the fiction but don't settle what is happening in the fiction. Spending artha in BW, or calling on a trait, is the first sort of thing: it is free-spinning relative to the fiction. @Hussar's example of D&D combat is an instance of the second: characters move, and attack one another, and (presumably) hurt one another, but the hit point loss doesn't model that hurt in any obvious sense. T&T combat is similar.

 

[Hussar]

What follows is always up to the group.

No, it really isn't. In many systems, the system will tell you directly what happens. You can certainly claim that the character was killed by bites, but, that's entirely on you, not the system. The system didn't tell you that. All the system told you was that the character lost enough HP to die. Since HP don't actually mean anything and don't map onto any specific result, any result is acceptable.

You ask "How much of our narrative do we need prescribed for us by written game rules?" In a simulation system, the answer is ANY. You need at least a tiny, tiny sliver of an answer if you're going to claim that the system is a simulation. If the system cannot give you any answers (or, in the case of D&D, any answer you care to narrate is acceptable) then it's not a simulation.

IOW, I say that your character M died of embarrassment. Use the mechanics to prove me wrong. You can't. And that's the point.

Which doesn't mean that you must have narrative follow mechanics. Obviously you don't need that. You DO need that for a simulation though. If you want to definitively answer what happened, then you need some form of simulation. If you cannot definitively answer any questions, then it's not a simulation.

Take a system I do know better - Battletech. In BTech, you roll your attack, then you roll your location for damage. Then there are a number of potential follow up effects depending on the damage done. I can definitively say that your Mech fell down because that PPC blew off it's hip actuator. That's a (very simple and not terribly accurate) good example of a simulationist system.