Quote from: Bloody Stupid Johnson;698330I also thing that there are other instances where making a roll 'active' has a time cost that hasn't been considered - the primary one being Perception checks. If a player has to roll to notice something, they roll in every room.
If a monster or secret door has to roll to stay concealed, then you roll only when there's one present. Consequently you can save a huge amount of dice rolling by having the GM roll Hide checks (aka using passive perception), instead of the PCs rolling Spot checks.

This is an issue that comes up in a lot of systems - because even systems that aren't player-only roll still often call for an opposed roll or a skill check for PCs to notice something (as opposed to only a GM roll).

I don't think it is a big deal to ask for a Perception roll. It does telegraph a little - but so does rolling dice behind your screen. Just a few red herrings is enough to stop the players from drawing too many conclusions from this. I've never seen any GM demand rolls for every room.

If it's important, my usual solution is for players to roll a series of Perception checks at the start of a given leg of exploration. I note these down and use them in defined order later for points of interest.

Quote from: jhkim;698384This is an issue that comes up in a lot of systems - because even systems that aren't player-only roll still often call for an opposed roll or a skill check for PCs to notice something (as opposed to only a GM roll).

I don't think it is a big deal to ask for a Perception roll. It does telegraph a little - but so does rolling dice behind your screen. Just a few red herrings is enough to stop the players from drawing too many conclusions from this. I've never seen any GM demand rolls for every room.

If it's important, my usual solution is for players to roll a series of Perception checks at the start of a given leg of exploration. I note these down and use them in defined order later for points of interest.

For passive rolls, I can see this as being a problem where the GM either "tips their hand" or players have to roll a lot of "nothing happens" rolls.

Although, the concept of a passive roll is kind of an oxymoron. The idea of a passive skill or trait is that your character doesn't need to do anything for it to trigger. Why have the player engaging in an active mechanic for a passive "always on" skill. If the character is some kind of observant or paranoid or keenly attuned to a particular kind of clue, and that clue comes up, why not just go "this is your thing, you notice X"

If the character has to actively search or closely study something to uncover it's secrets, then the player rolls.

Quote from: Soylent Green;697879Ah, I see the misunderstanding now. Player only rolls is about action resolution. Sorry, I thought that was implicit. It has no bearing on rolling random encounter or other tables.

Right. I mentioned Legendary Lives earlier, which may have been the first published game to take the approach (called "semi-diceless" by the authors). It's really just a matter of inverting a subset of things, such as having players "roll to be missed" (mathematically equivalent to a DM rolling "to hit" for monsters). IIRC, 4E D&D does something like this in having casters roll for spell "attacks" vs. Fortitude, Reflex or Will.

Legendary Lives is pretty nifty mechanically (whereas the "fluff" seemed in 1990 like a flashback to 1975, an even more light hearted T&T).

I rather wish I still had my 1st ed. books, as looking at the rules set again got me hankering to run a game (and I'm not likely soon either to print out the 2nd ed. PDF or to buy a copy).

Quote from: Adric;698080Systems built around players only rolling say something about how the world is represented. In most cases, the only randomness or uncertainty that the game is interested in is when the PCs act. For the rest of the world, what the GM says happens, happens.

I think that accurately represents the ethos in Legendary Lives. Joe and Kathy didn't put much into simulating "off stage" processes, preferring to focus on the action at hand, the decisions of the players. A procedure for rolling for purely NPC-NPC interactions may have been mentioned, but I reckon they regarded it as usually a waste of time.

Quote from: Adric;698080Dungeon World isn't interested in using dice or rules to emulate a world, it's interested in using the GM's imagination to emulate the world, and the rules and dice to describe how the PCs interact with it. It's got rules for creating stuff before the game that you can draw upon so you don't have to rely on just your improvisational skills, but that prep is still mostly interested in what happens when the players either interact with your prep, or willfully disregard the danger you present them with.

Instead of rolling to see if something that has no player input happens, the GM just decides if it happens or not.

Emphasis mine.

It strikes me that GMs don't do much relying on dice to decide what happens, even in 'emulative systems'? At least as a proportion of decisions made.  Maybe we're not thinking of the same systems?

In an 'emulative' system, I might roll on a wandering monster table, find d8 goblins - but then I use my imagination to decide what they're doing there, what they want, how they react to the players (e.g. what the 'Hostile' reaction result means in this case, like skulking away to get reinforcements, attacking immediately, tailing them to learn more about them, pick off the wizard first, or wait to ambush their at night), what their goal is (capture the players, drive them off, hurt them a bit and then run away to rally for another surprise attack later), where they move in combat, who they focus on, etc. etc.

Stars Without Number seems an extreme case of 'world emulation', with its highly quantified factions and the procedures for how they evolve and affect one another.

EDIT: I do agree that DW is much less/not at all concerned with using randomness to suggest to the GM what happens.

Quote from: fuseboy;698462It strikes me that GMs don't do much relying on dice to decide what happens, even in 'emulative systems'?

Depends on the GM.


"Hi, I'm Gary Gygax, I'm..." [rolls dice] "...pleased to meet you!"

In the 1970s, venues such as The Dragon, Judges Guild installments, Chivalry & Sorcery, The Arduin Grimoire, and Advanced D&D included tables for all sorts of procedural content generation. The C&S Sourcebook, for example, provided (IIRC) one for determining what your randomly encountered goblins were up to.

That's hilarious.

So, I think it's worth differentiating between the GM's role of "emulating a world" (by which I mean, rendering a world that seems self-consistent, has verisimilitude and a sense of independent existence from the spotlight shined on it by the players' actions and choices), and procedural content generation and situation evolution.

The latter supports the former, but I think there's a subtle false cognate. The point of using random tables includes a) getting a boatload of ideas from other creators, and b) insulating the GM from the temptation to adapt happenings too tightly to the players' actions.

Using a hypothetical 100% procedure-bound approach (where even actions in combat flow from a process the GM follows) has turned the GM into a sort of biological 'emulator', which is different.

A GM might have prepared the tables himself (an emulator prepared by imagination), and they might suck (by which I mean the players find the world unconvincing - the emulator fails at emulation).

Quote from: fuseboy;698467That's hilarious.

So, I think it's worth differentiating between the GM's role of "emulating a world" (by which I mean, rendering a world that seems self-consistent, has verisimilitude and a sense of independent existence from the spotlight shined on it by the players' actions and choices), and procedural content generation and situation evolution.

The latter supports the former, but I think there's a subtle false cognate. The point of using random tables includes a) getting a boatload of ideas from other creators, and b) insulating the GM from the temptation to adapt happenings too tightly to the players' actions.

Using a hypothetical 100% procedure-bound approach (where even actions in combat flow from a process the GM follows) has turned the GM into a sort of biological 'emulator', which is different.

A GM might have prepared the tables himself (an emulator prepared by imagination), and they might suck (by which I mean the players find the world unconvincing - the emulator fails at emulation).

I think a 100% procedure bound approach would indeed be hypothetical, since nothing can give you 100% responses and prose to respond to your players. Eventually the GM has to use their own imagination to respond to the players or make the world feel alive.

I don't really have a problem with dice (or any other mechanic) being used to procedurally create the world and it's events, so long as the process feels seamless to the group. I've experimented with writing dungeon rooms on a bunch of index cards and shuffling them to create a new dungeon layout, which isn't a far cry from a few systems' way of doing the same with room tiles and monster tables.

I've discovered I like a gaming model that might give me a prompt every now and then, but generally relies on my imagination and improv skills to create the world. Then most of my thought and activity is spend creating and maintaining this imaginary space with my players directly, only referencing the rules when a player triggers one with their characters action.

Random tables are a crucial element of OSR play, as far as I'm concerned. Just like randomly generating a character allows a player to think outside his usual box, randomly generating world-content does the same for the GM.

RPGPundit

Only the players roll?

Hmmmmm.....

Quote from: Justin Alexander;698129Even if we completely ignore the process of determining range, referencing the other modifiers, etc. you've literally just doubled the amount of time you originally claimed. When you've already conceded a 100% error rate I think the overly simplistic nature of your original "analysis" has been sufficiently confirmed.

Err ... in your rush to claim that other people are overly simplifying analysis, you don't seem to have lingered over the "It goes a great deal faster in my head" bit, did you?  Tell you what.  How about you go back, reread my post, and try again?

(Never mind that, eesh.  Don't know about you, but some of us have played these games for a while.  We don't need to look the modifiers up, and we're pretty good at gauging range in a fraction of a second.)

Quote from: RPGPundit;699093Random tables are a crucial element of OSR play, as far as I'm concerned. Just like randomly generating a character allows a player to think outside his usual box, randomly generating world-content does the same for the GM.

RPGPundit

Numenera, which is set up for "Only Players Roll" still has random tables for the GM to roll on, such as the table for generating Cyphers (one use magical items, similar to potions, etc.).

The main point of only players roll is for player interaction with the setting such as combat or negotiations, etc. Not that the GM can literally never roll.

You could literally change D&D to player's only roll with just a few small tweaks that wouldn't change the probabilities at all. (the only part that would be a bit wonky is NPC/Monster damage rolls.)

I love the idea of it speeding things up but I just can't wrap my head around it.  I keep imagining it as hands-off GMing regarding NPCs, though that's probably incorrect.

One thing it's good for asymmetrical scales. For example, in LL, player-characters have skills rated on a 25-point range; NPCs are rated on a scale of just 10 categories.

How well this works depends on setting up the right matrix. Just 4 categories (the number of colors on the MSH table) would be too few for Marvel Super Heroes, but 10 might work.