Surprising the GM, or, Random Content in Dungeons

GMMichael

Guide of Modos
Would the dynamic of the game change if the GM takes on more of a player-like role? I know I know, the GM isn't supposed to be an adversary, but that's almost implicit in the mindset. However, if the GM has a more player-like attitude*, doing more responding than instigating, would the group know the difference? Would it show in the after-game chatter?

In a sense, the GM's priorities might shift/alter, like this:
Standard GMingPlayer-like GMing
Entertain playersEntertain players
Present challengesFoster immersion
Enforce rulesEnforce rules
Foster immersionPresent challenges
...with immersion becoming more important, since random tables have the potential to break immersion, while generating challenges (of more than one type) on their own.

*An attitude that stems from having little-to-no control over what appears next in the dungeon/game (as a result of random generators). Or: "let's defeat the next challenge together."
 

log in or register to remove this ad

hawkeyefan

Legend
Would the dynamic of the game change if the GM takes on more of a player-like role? I know I know, the GM isn't supposed to be an adversary, but that's almost implicit in the mindset. However, if the GM has a more player-like attitude*, doing more responding than instigating, would the group know the difference? Would it show in the after-game chatter?

In a sense, the GM's priorities might shift/alter, like this:
Standard GMingPlayer-like GMing
Entertain playersEntertain players
Present challengesFoster immersion
Enforce rulesEnforce rules
Foster immersionPresent challenges
...with immersion becoming more important, since random tables have the potential to break immersion, while generating challenges (of more than one type) on their own.

*An attitude that stems from having little-to-no control over what appears next in the dungeon/game (as a result of random generators). Or: "let's defeat the next challenge together."

I think that looking at examples of games along the lines that you’ve described would show some key differences. If we think of Standard GMing as something like D&D or similar and then Player-like GMing as something like Dungeon World or similar.

DW and other Powered by the Apocalypse games have the “play to find out” approach. The GM is involved in setting up the situation and the participants involved, but how things will play out is as uncertain for him as it is for the players.

I think your list of priorities shows that although both games may foster immersion and both present challenges to players, which of those things is prioritized would be the key difference. There would be other differences, too, and also many similarities, and of course everyone’s game is different, so there’s no universal answer....but where a game’s priority is placed is the big factor here.
 

DW and other Powered by the Apocalypse games have the “play to find out” approach. The GM is involved in setting up the situation and the participants involved, but how things will play out is as uncertain for him as it is for the players.

I have yet to run a DW game, but my GURPS games often feel this way too. Not just in the sense of "I don't know whether they will win this battle," but rather, "I have no idea how they will react to this NPC and depending what they do, this session could go in too many directions for me to prepare for." I love those sorts of sessions.

Learning how to play now with VTTs has been tough in this regard. Roll20 seems to work best if your have all of your maps and tokens prepared in advance. That's devilishly hard to do if you have no idea what might happen in the session. I'm finding myself loading up visuals to support the campaign session rather than battlemaps.

In a sense, the GM's priorities might shift/alter, like this:
Standard GMingPlayer-like GMing
Entertain playersEntertain players
Present challengesFoster immersion
Enforce rulesEnforce rules
Foster immersionPresent challenges

One could quibble endlessly about the exact details of these lists, but it's definitely interesting to think about. I think I'd put immersion above challenges for myself, generally. I also don't so much "enforce" the rules as much as being a rules resource (since I usually know them better than my players). We regularly toss the rules out the window, though, when they aren't doing what we need them to do.
 

pemerton

Legend
I think that looking at examples of games along the lines that you’ve described would show some key differences.

<snip>

I think your list of priorities shows that although both games may foster immersion and both present challenges to players, which of those things is prioritized would be the key difference. There would be other differences, too, and also many similarities, and of course everyone’s game is different, so there’s no universal answer....but where a game’s priority is placed is the big factor here.
One could quibble endlessly about the exact details of these lists, but it's definitely interesting to think about. I think I'd put immersion above challenges for myself, generally. I also don't so much "enforce" the rules as much as being a rules resource (since I usually know them better than my players).
I agree with @hawkeyefan that it makes sense to tackle this by looking at actual games and the actual play of those games.

For me, what makes a game immersive is that the stakes and consequences are compelling, and that I can experience them directly through the play of the game. This works on either side of the table.

Here's one example that I pick because I hope it's easy to communicate: when we play Prince Valiant we use dice rather than coins (most of us use evens as successes; one uses 4+ on d6s). When there is a joust on, the typical pool is 6 to 10+ dice. When you pick up a pool of that size and start shaking them in your hands, it's like the thunder of hooves as the two knights charge towards one another! And then when the dice hit the table, there's that moment of suspense just before the impact and then you can see who's won and whose been defeated and perhaps unhorsed.

Here's a different example, that also involves jousting in Prince Valiant. Sir Lionheart, who claimed to be the greatest knight in Britain, was blocking a bridge to the north of Castle Hill. He would cede the brige only if defeated in a joust. Naturally the PCs rode out to challenge him! One of the PCs was only a squire, but - after the knight PCs had both been defeated - was able to manipulate Sir Lionheart into knighting him so that he too could joust. (Prince Valiant resolves social conflilcts just like jousting and other conflicts, ie via opposed dice pools.) The player knew that Sir Lionheart's jousting pool was big (having been at the table to see the other PCs bested) and knew that his own pool would be modest (the character emphasises the social side of things a bit more than the martial). But he had, in hand, a "Storyteller's Certificate" that allowed one use of a fiat ability from a list that includes both Knock an Opponent Senseless and Kill a Foe in Combat.

So as we were rolling our pools he was also deciding whether to use his certificate - yes, he would - and which option to go for. He didn't want to kill Sir Lionheart, but he was scared that if he only knocked him out then when Sir Lionheart regained consciousness he would insist on continuing the fight on foot, and perhaps to the death - and the player didn't have a second certificate!

Now I know some people assert that these sorts of meta-abilities break immersion, but for me at least it wasn't like this at all. The player's thinking about how to play the game at this point correlated pretty directly to the PC's experience: I can't defeat this knight except through some uncontrollable luck. I wouldn't want him to die - he's a valiant knight - but unless some misfortune kills him he'll either win the joust, or else insist in continuing on foot and win at that point. And just as the PC wants but doesn't want Sir Lionheart to unluckily die, so the player reluctantly but resignedly plays his certificate - as we narrated it, the PC's lance shattered on Sir Lionheart's shield and a splinter went through his visor slit and pierced his eye and brain, killing him.

That was immersive, and the whole situation was surprising to everyone at the table: the PC who started the campaign as a squire, with the lowest standing and cheapest gear of all the PCs, was knighted, and defeated the best knight in all Britan, and as a result took possession of his fine steed, excellent armour and bejewlled sword.

For me what spoils immersion is (i) a sense that the game is already written, and (ii) a sense that the action is all happening offscreen. This is as much true for me GMing as playing - because as a GM if what I'm doing is (i) bringing my prior authorial decisions to bear and/or (ii) manipulating fictional elements in my head that no one else even knows about, then there's no sense of immediacy, of being there in this mutually-generated situation and experiencing it as it unfolds. Which is what I love about RPGIng.
 

hawkeyefan

Legend
I have yet to run a DW game, but my GURPS games often feel this way too. Not just in the sense of "I don't know whether they will win this battle," but rather, "I have no idea how they will react to this NPC and depending what they do, this session could go in too many directions for me to prepare for." I love those sorts of sessions.

Learning how to play now with VTTs has been tough in this regard. Roll20 seems to work best if your have all of your maps and tokens prepared in advance. That's devilishly hard to do if you have no idea what might happen in the session. I'm finding myself loading up visuals to support the campaign session rather than battlemaps.

I've found D&D to be very challenging online. At least, any kind of D&D that uses maps and minis. Since the pandemic, I've set aside my 5E campaign in favor of letting others run some stuff, and I've run a few sessions of the new Alien RPG from Free League. I had some maps in the form of ship schematics for reference, but we otherwise played theater of the mind, and it was a lot of fun.

If I run anything else online, I think it would either need to be theater of the mind, or else something I could prepare for well in advance. The way I tend to DM is to craft maps on the fly with a wet-erase battle mat, and there's not really a quick version of that for online play. I also usually let the players drive where things go, so most of the time, having like 3 or 4 maps prepared for a session is, for me, committing strongly to a specific path for an adventure, which I try to avoid.
 


pemerton

Legend
If I run anything else online, I think it would either need to be theater of the mind, or else something I could prepare for well in advance.
The only online I've done is one session of Cortex+ LotR. We used Zoom. There are no visual aids required, but we used Chat so I could post the Scene Distinctions and periodically post the current state of the Doom Pool. And we used share screen for building and resolving our dice pools. (Cortex+ is dice-pool intensive, and you need to see each individual roll, not the total.)

I don't think the sort of pre-generated content I frequently see referenced for VTTs (maps, minis etc) is a good fit for the sort of approach being discussed in this thread.

I've run a few sessions of the new Alien RPG from Free League. I had some maps in the form of ship schematics for reference, but we otherwise played theater of the mind, and it was a lot of fun.
You shoud post/start a thread about this! I ran some Traveller sessions inspired by reading about that new system (I think you've seen at least one of the posts I just linked to), and am interested to hear more about it from the perspective of someone who does PbtA as well as D&D.
 

GMMichael

Guide of Modos
I have yet to run a DW game, but my GURPS games often feel this way too. Not just in the sense of "I don't know whether they will win this battle," but rather, "I have no idea how they will react to this NPC and depending what they do, this session could go in too many directions for me to prepare for." I love those sorts of sessions.

Learning how to play now with VTTs has been tough in this regard. Roll20 seems to work best if your have all of your maps and tokens prepared in advance. That's devilishly hard to do if you have no idea what might happen in the session. I'm finding myself loading up visuals to support the campaign session rather than battlemaps.
This is how my games usually go: any direction. You'd have to be a pretty serious railroader (or using an adventure module?) to know which (exact) direction the session will go.

Just a heads-up on Roll20: the drawing tools are pretty easy if you put in a little practice. If your PCs go off-map, whipping up a black-and-white (or two-color) map is doable. If you need a token, Bing the picture, save to your desktop, and drag it onto your map. I haven't had much luck with their in-journal art locator. Also, theatre of the mind = no maps needed :)

For me, what makes a game immersive is that the stakes and consequences are compelling, and that I can experience them directly through the play of the game. This works on either side of the table.

For me what spoils immersion is (i) a sense that the game is already written, and (ii) a sense that the action is all happening offscreen. This is as much true for me GMing as playing - because as a GM if what I'm doing is (i) bringing my prior authorial decisions to bear and/or (ii) manipulating fictional elements in my head that no one else even knows about, then there's no sense of immediacy, of being there in this mutually-generated situation and experiencing it as it unfolds. Which is what I love about RPGIng.
So, put more control into PC hands, or remove it from the GM's hands via random generators, and the GM can share the experience? That's where my head is going.
 

pemerton

Legend
So, put more control into PC hands, or remove it from the GM's hands via random generators, and the GM can share the experience? That's where my head is going.
I would say players rather than PCs. We're not talking here about the in-fiction events, but the real-world process of how the shared fiction is created.

When you talk about random generators you seem to be focused mostly on the establishment and introduction of fictional elements at the framing state (eg who do you meet? what do you see?) If you look at games that are deliberately designed to surprise the GM - eg PbtA games that @hawkeyefan has mentioned, or the Prince Valiant experience that I posted about - they often focus as much or more on action resolution as the place where the GM gets surprised.

There are a number of technical aspects of this that could be unpacked, but probably the most important is that, in these systems the outcome of action resolution (i) is not pre-determined, and (ii) can sometimes be something that the player who declared the action for his/her PC didn't expect, and (iii) can sometimes be something that the GM who set the scene for the action declaration didn't expect, and (iv) can sometimes be both (ii) and (iii). These games also take it as given that whatever results from action resolution must be incorporated into and honoured in subsequent framing.

Conversely, if you look at games where the GM is rarely or never surprised, the above elements are missing: outcomes of action resolution are pre-determined; or in some other way cannot be different from what the GM expected; or do not have any significant impact upon subsequent framings. You can see a whole lot of examples of this in the currently active "How was your last session?" thread on this forum.
 

Li Shenron

Legend
All that being said, I think doing an adventure that is completely randomized on the go would be difficult. The biggest issue is simply physical dimensions. Trying to fit randomly sized rooms into a dungeon without time to work with would be a pain (it would be easier in a more vaguely defined environment like a city). I don't know that that particular element would be worth it. But if you had the physical dungeon drawn up and you simply randomized the contents of the rooms on the go, that's completely doable. Or what might be more fun would be to pre-map a variety of room-complexes (let's say they all take up a 200'x200' area) and then during play randomly roll which room complex is next when you get to the edge of the map.

I think it's a great idea!

I love randomness and used random tables to "surprise myself" while designing a dungeon or whatever, but I've never done some like that in real time.

My preference would be to use environmental constraints (obvious ones such as "only cold-themed monsters in a cold environment") to keep a feeling of consistency, but otherwise in a sandbox campaign I might even have no CR restrictions 😈
 

Remove ads

Top