Let's say that I'm DMing Arden Vul for three different groups. The first group A is your optimized party, the second group B is your unoptimized party, and the third group C is also unoptimized but I use GM perogative to start them at level 5. What will happen? Well, likely B will stick around in Lankos Basement fighting rats, while A can advance deeper into the dungeon but not as deep as C. None of them can go down to the deepest levels and assault the factions there. The sandbox balances itself: players will seek a danger level that's appropriate. Is there a risk that party C will start farming rats in Lankos Basement like a deranged WoW player? Not really: tabletop game time is too precious to waste and most players realize this in my experience. I think this solves your cursed problem: players can go where they want and they are unlikely to be WoW-level degenerate about it.
When I think of difficulty in OSR games, I don't think much about monster combat effectiveness. Say I make a simple dungeon: room 1 has a bunch of orcs, room 2 is empty and room 3 has some treasure. How well the players optimize the party determines how likely they are to defeat the orcs, but optimizing the party isn't that hard (as you note). I can add more orcs to room 1, or make them stronger. This will make the players more likely to fail. But it doesn't affect how their choices impact the risk of success: their best strategy is still to optimize the party, same as before. Rolling 10 on a d10 is "more difficult" that rolling 6 on a d6, but is it really when skill doesn't matter?
Now if I add a chess puzzle to room 2 then "real" difficulty comes into play. How hard do I make the puzzle? Or if I add a NPC to negotiate with: How well can the players convince the NPC? This is the player skill I'm interested in.
The true difficulty in OSR games mostly come from faction play, in my experience. And per our previous discussion on your previous post that's a lot GM fiat. But a adventure where the faction descriptions are less forgiving ("The vampires will lie and stab the PCs in the back, the goblins will adapt to every threat, the kobolds have an excellent order of battle", etc. ) will truly be more difficult than an adventure with "easy" factions ("The orcs are brutish and easy to fool, the ghouls are prone to infighting, the myconids will worship any proven magic user as a god", etc).
Finally the ancient red dragon. Examples like it are common. I don't buy either of your takes: If party B from earlier waltzes up to the dragon of Arden Vul, the dragon can easily kill them. But in my game it won't. Instead it will dominate the PCs and try to make them do it's bidding. This can be a good deal for the PCs! Powerful dragons make great patrons. If the PCs play smart (here: difficulty!), they can profit mightily from the relationship. If the players insult the dragon or attack it, the dragon will eat them instead (bad play is punished).
Replying twice to elaborate in a different direction
Say that the party is 5th level. If they go do 5th level content, say they have a 70% chance that everyone survives ~4 sessions to make it to 6th level.
But instead of doing 5th level content, they could instead do 3rd level content. If you check out the treasure tables and xp guidelines, each PC level takes about twice as much XP as before, so content appropriate to 5th level characters should be about twice as lucrative as content for 4th level characters, and would be four 4x as lucrative as content for 3rd level characters.
So, rather than do 4 sessions of 5th level content, they instead choose to do 16 sessions of 3rd level content. 3rd level content is *a lot* less dangerous, so they have a ~95% chance of surviving.
And that's **if** we're assuming that the treasure is appropriate, which in basically all of the content I've ever reviewed, it isn't. Values are often off by between 5x and 20x, so 2nd level content will have treasure appropriate for 5th level characters.
With that in mind, consider the actual party. A well-optimized 2nd level party might be able to operate as a bog-standard 3rd level party, while a poorly optimized 2nd level party might be operating as a 1st level party. Now this well optimized 2nd-level party is absolutely farming 1st level content and the difficulty and risk:reward feels totally jacked up.
Your math is true, but this just doesn't happen at my table. If my high-level players are trying to play it safe by doing low-level quests, I guess I would mock them for it, and then advance the plot of high-level rivals and enemies to stop them from dillydallying (noted that almost all OSR games have the players gain power too fast to make this threat strong enough). What are the trolls doing while the players clear level 3?
> Your math is true, but this just doesn't happen at my table.
People (and tables) are all different! Some people are fun optimizers and don't understand or empathize with the tension created when playing a game well isn't fun
----------
> If my high-level players are trying to play it safe by doing low-level quests, I guess I would mock them for it, and then advance the plot of high-level rivals and enemies to stop them from dillydallying (noted that almost all OSR games have the players gain power too fast to make this threat strong enough).
If you'd advance the plots of high level rivals when you think they're wasting time, but advance them at a different rate when they're playing correctly, that's *exactly* the sort of dynamic difficulty I'm talking about :D
----------
> What are the trolls doing while the players clear level 3?
Not getting simulated, broadly. Same as the other 12 factions the players haven't met. Same as when my first party TPK'd because they went too deep too fast and months passed in-game.
"Despite their strength, the varumani have shown little interest in acquiring new territories within the Halls. Partly this is due to their declining numbers, but it is also partly a product of their temperament. While individually fearsome warriors, collectively they are better followers than leaders (a trait bred into them by the rudishva). This means that they are unlikely to expand except under unusual circumstances."
I try to be a good blorber so the trolls will be plotting no matter the players action. But if the high-level PCs are wasting time then the trolls can plot unopposed, which gives them the initiative.
I may have a lower threshold for starting simulation of the trolls. I too don't do it if a low-level party does a TPK and months passes. But if a level-8 party is slouching around on level 3 in my game, then that will be noticed by the goblins, Setites and/or beastmen, and they will send word down and then I start simulation. A level-8 party is a powerful chesspiece, the trolls would want to talk to them.
Also you forget the mocking!: "This is the fifth session you spent on level 3! Are you guys too chicken to see what's deeper down in the dungeon? It's where all the cool stuff are..." ;)
> What will happen? Well, likely B will stick around in Lankos Basement fighting rats, while A can advance deeper into the dungeon but not as deep as C. None of them can go down to the deepest levels and assault the factions there. The sandbox balances itself: players will seek a danger level that's appropriate.
Why will they seek that? If we assume that players are fun-optimizers then *maybe*, but I don't think that's a safe assumption. If there's a place they can earn lots of XP for little risk, they're then now they're having to weigh in-game power acquisition against fun, and IME the reward mechanisms are strong. My party of 8th level dudes are still clearing out level 3, for instance. The same thing has historically happened in other sandbox settings (like hexcrawls) and other games.
----------
> But it doesn't affect how their choices impact the risk of success: their best strategy is still to optimize the party, same as before. Rolling 10 on a d10 is "more difficult" that rolling 6 on a d6, but is it really when skill doesn't matter?
There's still plenty of combat-time optimization to do: positioning, spell slot usage, target optimization, etc. Some games (draw steel, 13th age, etc) have a lot more of this than others.
That game is fun for a lot of people (including my table), but when fights are too easy then bad play (sloppy positioning, split fire instead of focused fire, using too few/too many spells, etc) still results in easy wins, so the play doesn't matter. Similarly, when the encounter is too difficult, then good play still loses, so it doesn't matter. Only the sweet spot of difficulty (which is narrower or wider depending on the game; a larger range of fights are appropriate in BX than Pathfinder 2e because of how the scaling works) are the choices impactful.
----------
> Now if I add a chess puzzle to room 2 then "real" difficulty comes into play. How hard do I make the puzzle? Or if I add a NPC to negotiate with: How well can the players convince the NPC? This is the player skill I'm interested in.
I don't think that's any more or less real than the sorts of choices we make in combat
----------
> The true difficulty in RPGs mostly come from interacting factions, in my experience.
I think this wades into no-true-scotsman stuff, but having played a very wide range of games I think there's plenty of difficulty in these games that don't come from interacting factions
----------
> If party B from earlier waltzes up to the dragon of Arden Vul, the dragon can easily kill them. But in my game it won't.
probably both the optimized ground and unoptimized group would brick to fighting a dragon, so that's a spot where it's inappropriate for creating meaningful combat choices. That's not necessarily bad; it's still fine to have things that the party is intended to negotiate, run, etc rather than fight (which agrees with the original seth quote)
I find that most "combat optimization" in RPGs are just a bad board game in practice. If I want tactic combat I play XCOM. I'm the first to agree that my taste is strange. (I'm still working on my mythical one-roll OSR combat system, so that my table waste as little time as possible on combat.)
I think Seth is fighting a straw man. My game is blorb. No power scaling. But I still don't need to curbstomp players with dragons, per my example.
I don't have much to add to your other points: I think I just don't value tactical combat so I don't care much about it. Another hypothethical:
- Say that Alice makes a dungeon called Orc Cave, and releases it in public domain. Orc Cave is a typical OSR dungeon of 50 rooms, and obviously it has a bunch of orcs in it.
- Bob thinks Orc Cave isn't difficult enough. He remakes it as Orc Cave 2.0 with twice as many orcs in it.
-Carol remakes Orc Cave it into Anti-Optimization Cave, which replaces all orcs with Carolites: Carolites are a homebrew monster which ignores armor, cause fear in elves and has an aura that turns all healing of non-Carolites into damage instead (i.e Carolites punish the optimization you describe).
-Dave remakes Orc Cave into Random Cave, which replaces all orcs with Davters: Davters are another homebrew monster which have 1 hp and bad AC, cannot surprise or be surprised, and their only attack is a 1in6 chance of self-destruct that kills anyone (no save!) within 100ft (so basically there's a 1in6 risk of TPK when you fight these guys no matter the build and tactics).
You might find it important which of these Orc Cave variants we play. I wonder "Are there any good chess puzzles in Orc Cave?" and "Are the factions interesting?".
> You might find it important which of these Orc Cave variants we play. I wonder "Are there any good chess puzzles in Orc Cave?" and "Are the factions interesting?".
Yeah - the variant of orc cave we play has meaningful impact on the at-the-table experience, especially given how much table-time these games (which varies by system) gives combat. If we were playing a one-roll-combat system, I could understand the argument, but we're not :D
----
re: good blorbing
Blorb has "no paper after seeing rock"; but I don't think that implies that the entire game world must be totally specified prior to chargen (or meeting the players) or similar. Instead, there's the tiers of truth, where prep > rules/mechanics > neutral improv. I posit that you can intentionally not specify large amounts of the campaign world and leave them to tier 2/3 truths, and when the players decide to go to those places, you put on your prep-hat in-between sessions and prepare appropriately-difficult-and-rewarding content.
Rather than saying that hex 2392 has and 8-room lair that's appropriate content for 2nd level characters, and then your 5th level party decides to finally go there and blows it out, we can just say that hex 2392 is a lair, and when your 5th level party goes there, you serve up a 4th, 5th, or 6th level lair. That's still following a rule (when my players go somewhere undefined, i serve up level-appropriate content) or neutral improv (make something up that won't help or harm the player characters too much)
Let's say that I DM Orc Cave one night and Anti-Optimization Cave the next night, using OSE.
1. The first night we spend 90 minutes on combat. Players roll dice and say "I hit". The PCs position themselves around a corridor chokepoint, one PC kicks an orc down a pit trap. One PC dies, the player quickly rolls a new one: it's a cleric in plate.
2. The second night we also spend 90 minutes on combat. Players roll dice and say "I hit". The PCs position themselves around a corridor chokepoint, one PC kicks an Carolite down a pit trap. One PC dies, the player quickly rolls a new one: it's a fighter with a two-handed sword.
Is it really that different? ;)
I like the OG hex 2382, and I hate the proposed change. What's stopping you from making the entire hex crawl "every hex has a level-appropriate challenge with a 90% chance of success"? Where's your blorb then? If I can muster enough effort to write "hex 2392: lair" I can muster the effort to write "hex 2392: lvl2 lair" and the world will be more alive thanks to it. Like, the point of making a hex crawl is to make a living breathing world where things fit together. If the players blow through it then so what? Just like how a level 1 party can interact with a dragon, a higher-level party has plenty of ways to interact with lower-level monsters.
> Let's say that I play Orc Cave one night and Anti-Optimization Cave the next night, using OSE.
>
> 1. The first night we spend 90 minutes on combat. Players roll dice and say "I hit". The PCs position themselves around a corridor chokepoint, one PC kicks an orc down a pit trap. One PC dies, the player quickly rolls a new one: it's a cleric in plate.
>
> 2. The second night we also spend 90 minutes on combat. Players roll dice and say "I hit". The PCs position themselves around a corridor chokepoint, one PC kicks an Carolites down a pit trap. One PC dies, the player quickly rolls a new one: it's a fighter with a two-handed sword.
>
> Is it really that different?
Nope, not different at all, presumably because you've constructed the hypothetical deliberately to not be different. Hopefully we agree that I can also intentionally construction a hypothetical that would be different, and so cherry-picking hypotheticals isn't fruitful :D
I totally understand that you don't find the sorts of decisions people make in combat to be interesting (as evidenced by the reductive example). I do, my players do, and lots of other people do as well.
While it's possible that a cave full of orcs produces the same physical experience as a cave full of carolites, i'd posit that that's not *usually* what happens, and you can verify that by writing simulations. I have https://github.com/beaurancourt/tunnel-fight if you want to play around in that space
----------
> What's stopping you from making the entire hex crawl "this hex has a level-appropriate challenge with a 90% chance of success"?
Well, nothing is "stopping me", per-se, but blorb does warn against it: "A campaign that’s built on all T2 and T3 truths isn’t as engaging as one that has some solid T1 framework in there (in a cloud, bones of steel), but as you patch holes (as T3 instructs you to) feel free to patch them with mechanics and general solutions (i.e. T2 truths). That’s you building a DM’s toolbox."
eg, rather than specifying everything up-front and then lots of our content is either too-difficult or too-easy and there's tons of room to farm and be degenerate, which collapses the in-session interesting decisions, we can specify some-or-most of the content, and leave a lot of it dynamic so that more of our table-time is spent in the appropriately-difficult-zone. This can be as simple as having a big shelf full of modules and grabbing one that's within the appropriate level range
Say that the optimized party has a 90% chance of defeating 8 orcs, and a 90% chance of defeating 4 Carolites. Then fighting 8 orcs will be mostly the same as fighting 4 Carolites. There might be some minor differences (clerics won't use heals when Carolites are close, etc) but the choices available to the players are very similar and there's often an obvious best option. The players should still stack high-dex, high-hp dudes in the front and have polearms in the second rank, they should focus-fire wounded enemies for action economy, etc. If you think these two fights would feel very different at your table then I can't argue against it, but at my table it would be a lot of dice rolling and "I hit" in both cases.
I try to not view content as "too-difficult" or "too-easy". Is the dragon in Arden Vul too-difficult for a level 1 party? The world is the world, by blorb. If a single goblin lives in a hole and the level-8 party wants to interact with him then I'll use blorb to either figure out something interesting for him to say or fast-forward the interaction to the likely result. There's interesting decisions to be made talking to a goblin in a hole (but maybe chiefly: "Should we waste our time talking to a goblin in a hole?" Now we're back to "mock the players" and "what are their rivals and enemies doing?" from the other thread.).
I think there is another aspect to this as well. I don't think you can actually escape dynamic difficulty, you can only try to keep it stable-ish. I know for a fact that my rulings are not entirely consistent. I'll make a decision one night, and three months later, I'll rule differently because I'm either tired and made a mistake, have gained new information, or just am tired of my group being in the same room after 2 hours of play and I want to keep things moving. That's going to have an effect on difficulty, even if I'm trying to stay consistent.
Something I'm coming to understand is that a lot of GMing and game design is about approximation. I work in a medical laboratory, so the reference that I'm going to use is quality control charts.
When working with lab tests you have three kinds of samples: Calibrators, Controls, and Patient Samples.
A calibrator is a fixed value. If your machine shows something else, your machine is broken or there is a misalignment somewhere.
A Control is a range of expected values, and that range will fluctuate over time. If things run a little high one morning, that's not a big deal. If it runs high over multiple days or shifts it's a problem.
Then you have patient samples, where are chaotic little monsters with no respect for the rules. They can be, and often are, anywhere on a chart.
Getting back to game design and GMing, I don't want the extremes of a calibrator or the chaos of a patient sample. I want to establish a comfortable range of possibilities. If the party steps outside of that range occasionally, that's fine. Sometimes it's fun to take on a challenge you know you shouldn't be able to, or to absolutely curb stomp a handful of low level NPCs. What I want to avoid is constantly being outside the intended ranges.
Keeping content within the ranges is the job of the dungeon master and adventure designers. Sometimes that means details need massaged a little. Other times it falls into place pretty easily. The simulation is, at least a little, flexible. If the party makes it clear they just want to talk, the dragon will probably let them. If the party cracks jokes at the dragon's expense to it's face? Well, different story.
> Getting back to game design and GMing, I don't want the extremes of a calibrator or the chaos of a patient sample. I want to establish a comfortable range of possibilities. If the party steps outside of that range occasionally, that's fine. Sometimes it's fun to take on a challenge you know you shouldn't be able to, or to absolutely curb stomp a handful of low level NPCs. What I want to avoid is constantly being outside the intended ranges.
I think that very efficiently articulates what I'm trying to say. It's *technically possible* to create a static-difficulty game where the challenge consistently feels correct for a particular player, but I think it's exceptionally unlikely to actually occur in practice.
There's a lot of different goals you can have when deciding what happens in your game world, during the large-space of GM fiat you're given (what exists where, how NPCs speak and behave, etc). You try go for realism where what happens is *always* what you think would actually happen. You can try to go for gamism, where what happens is what you think will generate interesting player-side decisions. You can go for narrativism, where what happens is what will generate the most story-worthy result. There are many other such aims!
I think leaning too far in one direction has diminishing returns; it takes more and more effort, and increasingly reduces the other goals. I *lean* toward wanting interesting decisions, but that's definitely not objectively correct in any sense.
I remember writing a golbin example on reddit a while back. The players travel through a 6-mile hex and anger a goblin tribe. Later, they travel back through that same hex. Do they encounter the same goblins?
In pure-simulation, you probably need to roll for an encounter with a low %chance, if you get it, roll to see what sort of encounter (maybe it's bandits instead of goblins or whatever), and then if it happens to be goblins, you can try to estimate how many different goblin tribes live in this ~30 square mile area, and then use that to give a %chance that it's the same tribe you angered, and then if it is, roll a reaction to see if they're still upset, and so forth.
In pure-gamism, we want there to be impact to our choices, and choosing leave a loose-end and then come back should probably have the consequences that you now have to deal with angry goblins on your way back; that strengthens the impact of the choice to leave them, and strengthens the choice about the return route.
In narrativism, it would be *absurd* to not use this payoff for what is so obviously a good plant.
Then, there's a whole spectrum in between; maybe we still roll for a random encounter like good simulationists, but if we get one, it's going to be the goblins. Or maybe it's not automatically goblins, but we make a rulings that there's a higher-than-normal chance that it's goblins or whatever. I just think that "playing it straight" is leaving a lot of value on the table and will produce less interesting games than sacrificing a tiny bit of simulationism
> " now, because of a few hours spent making meaningful character creation/build choices, way more hours at the table is spent playing wrong-difficulty content. " <
I see your point here and understand where you are going. Your advice seems accurately calibrated to your argument. All good so far.
My question though is what determines "wrong difficulty"? Is it because the challenge and rewards are out of balance? Is it because the players are bored?
Because it seems the problem as you have defined it is "game gives higher rewards than difficulty justifies so players can farm easy stuff". But then you aim all your solution mechanics to character development instead of stocking appropriate levels of treasure.
It seems like a workaround rather than a solution.
Which I think is what you meant by your #2 which you say you've never seen. So do I understand you correct or if not what did I get lost?
> My question though is what determines "wrong difficulty"? Is it because the challenge and rewards are out of balance? Is it because the players are bored?
So say that we design so that we aim for a 1:4 monster:treasure XP ratio, which is still well below what I see in modules.
1000xp worth of defenders (50 bugbears) might be guarding 4000g worth of treasure, spread around a bugbear lair or dungeon wing or whatever.
This is appropriate content for a ~third level party, but inappropriate for a 5th or 6th level party, the bugbears just get fireballed.
You might imagine that the 6th level party thinks this is a waste of time, and so would go instead adventure in areas with stronger defenders and more treasure. In my experience, they *don't*; if the bugbears are around to get fireballed, then fireballing them is worth the easy 5000xp and we just do that handful of sessions in a row (sacrificing fun for power). I've seen this in literally every gaming context i've played, from ttrpgs, to mmos, to single player games
Let's say that I'm DMing Arden Vul for three different groups. The first group A is your optimized party, the second group B is your unoptimized party, and the third group C is also unoptimized but I use GM perogative to start them at level 5. What will happen? Well, likely B will stick around in Lankos Basement fighting rats, while A can advance deeper into the dungeon but not as deep as C. None of them can go down to the deepest levels and assault the factions there. The sandbox balances itself: players will seek a danger level that's appropriate. Is there a risk that party C will start farming rats in Lankos Basement like a deranged WoW player? Not really: tabletop game time is too precious to waste and most players realize this in my experience. I think this solves your cursed problem: players can go where they want and they are unlikely to be WoW-level degenerate about it.
When I think of difficulty in OSR games, I don't think much about monster combat effectiveness. Say I make a simple dungeon: room 1 has a bunch of orcs, room 2 is empty and room 3 has some treasure. How well the players optimize the party determines how likely they are to defeat the orcs, but optimizing the party isn't that hard (as you note). I can add more orcs to room 1, or make them stronger. This will make the players more likely to fail. But it doesn't affect how their choices impact the risk of success: their best strategy is still to optimize the party, same as before. Rolling 10 on a d10 is "more difficult" that rolling 6 on a d6, but is it really when skill doesn't matter?
Now if I add a chess puzzle to room 2 then "real" difficulty comes into play. How hard do I make the puzzle? Or if I add a NPC to negotiate with: How well can the players convince the NPC? This is the player skill I'm interested in.
The true difficulty in OSR games mostly come from faction play, in my experience. And per our previous discussion on your previous post that's a lot GM fiat. But a adventure where the faction descriptions are less forgiving ("The vampires will lie and stab the PCs in the back, the goblins will adapt to every threat, the kobolds have an excellent order of battle", etc. ) will truly be more difficult than an adventure with "easy" factions ("The orcs are brutish and easy to fool, the ghouls are prone to infighting, the myconids will worship any proven magic user as a god", etc).
Finally the ancient red dragon. Examples like it are common. I don't buy either of your takes: If party B from earlier waltzes up to the dragon of Arden Vul, the dragon can easily kill them. But in my game it won't. Instead it will dominate the PCs and try to make them do it's bidding. This can be a good deal for the PCs! Powerful dragons make great patrons. If the PCs play smart (here: difficulty!), they can profit mightily from the relationship. If the players insult the dragon or attack it, the dragon will eat them instead (bad play is punished).
Replying twice to elaborate in a different direction
Say that the party is 5th level. If they go do 5th level content, say they have a 70% chance that everyone survives ~4 sessions to make it to 6th level.
But instead of doing 5th level content, they could instead do 3rd level content. If you check out the treasure tables and xp guidelines, each PC level takes about twice as much XP as before, so content appropriate to 5th level characters should be about twice as lucrative as content for 4th level characters, and would be four 4x as lucrative as content for 3rd level characters.
So, rather than do 4 sessions of 5th level content, they instead choose to do 16 sessions of 3rd level content. 3rd level content is *a lot* less dangerous, so they have a ~95% chance of surviving.
And that's **if** we're assuming that the treasure is appropriate, which in basically all of the content I've ever reviewed, it isn't. Values are often off by between 5x and 20x, so 2nd level content will have treasure appropriate for 5th level characters.
With that in mind, consider the actual party. A well-optimized 2nd level party might be able to operate as a bog-standard 3rd level party, while a poorly optimized 2nd level party might be operating as a 1st level party. Now this well optimized 2nd-level party is absolutely farming 1st level content and the difficulty and risk:reward feels totally jacked up.
Your math is true, but this just doesn't happen at my table. If my high-level players are trying to play it safe by doing low-level quests, I guess I would mock them for it, and then advance the plot of high-level rivals and enemies to stop them from dillydallying (noted that almost all OSR games have the players gain power too fast to make this threat strong enough). What are the trolls doing while the players clear level 3?
> Your math is true, but this just doesn't happen at my table.
People (and tables) are all different! Some people are fun optimizers and don't understand or empathize with the tension created when playing a game well isn't fun
----------
> If my high-level players are trying to play it safe by doing low-level quests, I guess I would mock them for it, and then advance the plot of high-level rivals and enemies to stop them from dillydallying (noted that almost all OSR games have the players gain power too fast to make this threat strong enough).
If you'd advance the plots of high level rivals when you think they're wasting time, but advance them at a different rate when they're playing correctly, that's *exactly* the sort of dynamic difficulty I'm talking about :D
----------
> What are the trolls doing while the players clear level 3?
Not getting simulated, broadly. Same as the other 12 factions the players haven't met. Same as when my first party TPK'd because they went too deep too fast and months passed in-game.
"Despite their strength, the varumani have shown little interest in acquiring new territories within the Halls. Partly this is due to their declining numbers, but it is also partly a product of their temperament. While individually fearsome warriors, collectively they are better followers than leaders (a trait bred into them by the rudishva). This means that they are unlikely to expand except under unusual circumstances."
I try to be a good blorber so the trolls will be plotting no matter the players action. But if the high-level PCs are wasting time then the trolls can plot unopposed, which gives them the initiative.
I may have a lower threshold for starting simulation of the trolls. I too don't do it if a low-level party does a TPK and months passes. But if a level-8 party is slouching around on level 3 in my game, then that will be noticed by the goblins, Setites and/or beastmen, and they will send word down and then I start simulation. A level-8 party is a powerful chesspiece, the trolls would want to talk to them.
Also you forget the mocking!: "This is the fifth session you spent on level 3! Are you guys too chicken to see what's deeper down in the dungeon? It's where all the cool stuff are..." ;)
> What will happen? Well, likely B will stick around in Lankos Basement fighting rats, while A can advance deeper into the dungeon but not as deep as C. None of them can go down to the deepest levels and assault the factions there. The sandbox balances itself: players will seek a danger level that's appropriate.
Why will they seek that? If we assume that players are fun-optimizers then *maybe*, but I don't think that's a safe assumption. If there's a place they can earn lots of XP for little risk, they're then now they're having to weigh in-game power acquisition against fun, and IME the reward mechanisms are strong. My party of 8th level dudes are still clearing out level 3, for instance. The same thing has historically happened in other sandbox settings (like hexcrawls) and other games.
----------
> But it doesn't affect how their choices impact the risk of success: their best strategy is still to optimize the party, same as before. Rolling 10 on a d10 is "more difficult" that rolling 6 on a d6, but is it really when skill doesn't matter?
There's still plenty of combat-time optimization to do: positioning, spell slot usage, target optimization, etc. Some games (draw steel, 13th age, etc) have a lot more of this than others.
That game is fun for a lot of people (including my table), but when fights are too easy then bad play (sloppy positioning, split fire instead of focused fire, using too few/too many spells, etc) still results in easy wins, so the play doesn't matter. Similarly, when the encounter is too difficult, then good play still loses, so it doesn't matter. Only the sweet spot of difficulty (which is narrower or wider depending on the game; a larger range of fights are appropriate in BX than Pathfinder 2e because of how the scaling works) are the choices impactful.
----------
> Now if I add a chess puzzle to room 2 then "real" difficulty comes into play. How hard do I make the puzzle? Or if I add a NPC to negotiate with: How well can the players convince the NPC? This is the player skill I'm interested in.
I don't think that's any more or less real than the sorts of choices we make in combat
----------
> The true difficulty in RPGs mostly come from interacting factions, in my experience.
I think this wades into no-true-scotsman stuff, but having played a very wide range of games I think there's plenty of difficulty in these games that don't come from interacting factions
----------
> If party B from earlier waltzes up to the dragon of Arden Vul, the dragon can easily kill them. But in my game it won't.
probably both the optimized ground and unoptimized group would brick to fighting a dragon, so that's a spot where it's inappropriate for creating meaningful combat choices. That's not necessarily bad; it's still fine to have things that the party is intended to negotiate, run, etc rather than fight (which agrees with the original seth quote)
I find that most "combat optimization" in RPGs are just a bad board game in practice. If I want tactic combat I play XCOM. I'm the first to agree that my taste is strange. (I'm still working on my mythical one-roll OSR combat system, so that my table waste as little time as possible on combat.)
I think Seth is fighting a straw man. My game is blorb. No power scaling. But I still don't need to curbstomp players with dragons, per my example.
I don't have much to add to your other points: I think I just don't value tactical combat so I don't care much about it. Another hypothethical:
- Say that Alice makes a dungeon called Orc Cave, and releases it in public domain. Orc Cave is a typical OSR dungeon of 50 rooms, and obviously it has a bunch of orcs in it.
- Bob thinks Orc Cave isn't difficult enough. He remakes it as Orc Cave 2.0 with twice as many orcs in it.
-Carol remakes Orc Cave it into Anti-Optimization Cave, which replaces all orcs with Carolites: Carolites are a homebrew monster which ignores armor, cause fear in elves and has an aura that turns all healing of non-Carolites into damage instead (i.e Carolites punish the optimization you describe).
-Dave remakes Orc Cave into Random Cave, which replaces all orcs with Davters: Davters are another homebrew monster which have 1 hp and bad AC, cannot surprise or be surprised, and their only attack is a 1in6 chance of self-destruct that kills anyone (no save!) within 100ft (so basically there's a 1in6 risk of TPK when you fight these guys no matter the build and tactics).
You might find it important which of these Orc Cave variants we play. I wonder "Are there any good chess puzzles in Orc Cave?" and "Are the factions interesting?".
> You might find it important which of these Orc Cave variants we play. I wonder "Are there any good chess puzzles in Orc Cave?" and "Are the factions interesting?".
Yeah - the variant of orc cave we play has meaningful impact on the at-the-table experience, especially given how much table-time these games (which varies by system) gives combat. If we were playing a one-roll-combat system, I could understand the argument, but we're not :D
----
re: good blorbing
Blorb has "no paper after seeing rock"; but I don't think that implies that the entire game world must be totally specified prior to chargen (or meeting the players) or similar. Instead, there's the tiers of truth, where prep > rules/mechanics > neutral improv. I posit that you can intentionally not specify large amounts of the campaign world and leave them to tier 2/3 truths, and when the players decide to go to those places, you put on your prep-hat in-between sessions and prepare appropriately-difficult-and-rewarding content.
Rather than saying that hex 2392 has and 8-room lair that's appropriate content for 2nd level characters, and then your 5th level party decides to finally go there and blows it out, we can just say that hex 2392 is a lair, and when your 5th level party goes there, you serve up a 4th, 5th, or 6th level lair. That's still following a rule (when my players go somewhere undefined, i serve up level-appropriate content) or neutral improv (make something up that won't help or harm the player characters too much)
Let's say that I DM Orc Cave one night and Anti-Optimization Cave the next night, using OSE.
1. The first night we spend 90 minutes on combat. Players roll dice and say "I hit". The PCs position themselves around a corridor chokepoint, one PC kicks an orc down a pit trap. One PC dies, the player quickly rolls a new one: it's a cleric in plate.
2. The second night we also spend 90 minutes on combat. Players roll dice and say "I hit". The PCs position themselves around a corridor chokepoint, one PC kicks an Carolite down a pit trap. One PC dies, the player quickly rolls a new one: it's a fighter with a two-handed sword.
Is it really that different? ;)
I like the OG hex 2382, and I hate the proposed change. What's stopping you from making the entire hex crawl "every hex has a level-appropriate challenge with a 90% chance of success"? Where's your blorb then? If I can muster enough effort to write "hex 2392: lair" I can muster the effort to write "hex 2392: lvl2 lair" and the world will be more alive thanks to it. Like, the point of making a hex crawl is to make a living breathing world where things fit together. If the players blow through it then so what? Just like how a level 1 party can interact with a dragon, a higher-level party has plenty of ways to interact with lower-level monsters.
> Let's say that I play Orc Cave one night and Anti-Optimization Cave the next night, using OSE.
>
> 1. The first night we spend 90 minutes on combat. Players roll dice and say "I hit". The PCs position themselves around a corridor chokepoint, one PC kicks an orc down a pit trap. One PC dies, the player quickly rolls a new one: it's a cleric in plate.
>
> 2. The second night we also spend 90 minutes on combat. Players roll dice and say "I hit". The PCs position themselves around a corridor chokepoint, one PC kicks an Carolites down a pit trap. One PC dies, the player quickly rolls a new one: it's a fighter with a two-handed sword.
>
> Is it really that different?
Nope, not different at all, presumably because you've constructed the hypothetical deliberately to not be different. Hopefully we agree that I can also intentionally construction a hypothetical that would be different, and so cherry-picking hypotheticals isn't fruitful :D
I totally understand that you don't find the sorts of decisions people make in combat to be interesting (as evidenced by the reductive example). I do, my players do, and lots of other people do as well.
While it's possible that a cave full of orcs produces the same physical experience as a cave full of carolites, i'd posit that that's not *usually* what happens, and you can verify that by writing simulations. I have https://github.com/beaurancourt/tunnel-fight if you want to play around in that space
----------
> What's stopping you from making the entire hex crawl "this hex has a level-appropriate challenge with a 90% chance of success"?
Well, nothing is "stopping me", per-se, but blorb does warn against it: "A campaign that’s built on all T2 and T3 truths isn’t as engaging as one that has some solid T1 framework in there (in a cloud, bones of steel), but as you patch holes (as T3 instructs you to) feel free to patch them with mechanics and general solutions (i.e. T2 truths). That’s you building a DM’s toolbox."
eg, rather than specifying everything up-front and then lots of our content is either too-difficult or too-easy and there's tons of room to farm and be degenerate, which collapses the in-session interesting decisions, we can specify some-or-most of the content, and leave a lot of it dynamic so that more of our table-time is spent in the appropriately-difficult-zone. This can be as simple as having a big shelf full of modules and grabbing one that's within the appropriate level range
Say that the optimized party has a 90% chance of defeating 8 orcs, and a 90% chance of defeating 4 Carolites. Then fighting 8 orcs will be mostly the same as fighting 4 Carolites. There might be some minor differences (clerics won't use heals when Carolites are close, etc) but the choices available to the players are very similar and there's often an obvious best option. The players should still stack high-dex, high-hp dudes in the front and have polearms in the second rank, they should focus-fire wounded enemies for action economy, etc. If you think these two fights would feel very different at your table then I can't argue against it, but at my table it would be a lot of dice rolling and "I hit" in both cases.
I try to not view content as "too-difficult" or "too-easy". Is the dragon in Arden Vul too-difficult for a level 1 party? The world is the world, by blorb. If a single goblin lives in a hole and the level-8 party wants to interact with him then I'll use blorb to either figure out something interesting for him to say or fast-forward the interaction to the likely result. There's interesting decisions to be made talking to a goblin in a hole (but maybe chiefly: "Should we waste our time talking to a goblin in a hole?" Now we're back to "mock the players" and "what are their rivals and enemies doing?" from the other thread.).
This may be a bit of a ramble.
I think there is another aspect to this as well. I don't think you can actually escape dynamic difficulty, you can only try to keep it stable-ish. I know for a fact that my rulings are not entirely consistent. I'll make a decision one night, and three months later, I'll rule differently because I'm either tired and made a mistake, have gained new information, or just am tired of my group being in the same room after 2 hours of play and I want to keep things moving. That's going to have an effect on difficulty, even if I'm trying to stay consistent.
Something I'm coming to understand is that a lot of GMing and game design is about approximation. I work in a medical laboratory, so the reference that I'm going to use is quality control charts.
When working with lab tests you have three kinds of samples: Calibrators, Controls, and Patient Samples.
A calibrator is a fixed value. If your machine shows something else, your machine is broken or there is a misalignment somewhere.
A Control is a range of expected values, and that range will fluctuate over time. If things run a little high one morning, that's not a big deal. If it runs high over multiple days or shifts it's a problem.
Then you have patient samples, where are chaotic little monsters with no respect for the rules. They can be, and often are, anywhere on a chart.
Getting back to game design and GMing, I don't want the extremes of a calibrator or the chaos of a patient sample. I want to establish a comfortable range of possibilities. If the party steps outside of that range occasionally, that's fine. Sometimes it's fun to take on a challenge you know you shouldn't be able to, or to absolutely curb stomp a handful of low level NPCs. What I want to avoid is constantly being outside the intended ranges.
Keeping content within the ranges is the job of the dungeon master and adventure designers. Sometimes that means details need massaged a little. Other times it falls into place pretty easily. The simulation is, at least a little, flexible. If the party makes it clear they just want to talk, the dragon will probably let them. If the party cracks jokes at the dragon's expense to it's face? Well, different story.
> Getting back to game design and GMing, I don't want the extremes of a calibrator or the chaos of a patient sample. I want to establish a comfortable range of possibilities. If the party steps outside of that range occasionally, that's fine. Sometimes it's fun to take on a challenge you know you shouldn't be able to, or to absolutely curb stomp a handful of low level NPCs. What I want to avoid is constantly being outside the intended ranges.
I think that very efficiently articulates what I'm trying to say. It's *technically possible* to create a static-difficulty game where the challenge consistently feels correct for a particular player, but I think it's exceptionally unlikely to actually occur in practice.
There's a lot of different goals you can have when deciding what happens in your game world, during the large-space of GM fiat you're given (what exists where, how NPCs speak and behave, etc). You try go for realism where what happens is *always* what you think would actually happen. You can try to go for gamism, where what happens is what you think will generate interesting player-side decisions. You can go for narrativism, where what happens is what will generate the most story-worthy result. There are many other such aims!
I think leaning too far in one direction has diminishing returns; it takes more and more effort, and increasingly reduces the other goals. I *lean* toward wanting interesting decisions, but that's definitely not objectively correct in any sense.
I remember writing a golbin example on reddit a while back. The players travel through a 6-mile hex and anger a goblin tribe. Later, they travel back through that same hex. Do they encounter the same goblins?
In pure-simulation, you probably need to roll for an encounter with a low %chance, if you get it, roll to see what sort of encounter (maybe it's bandits instead of goblins or whatever), and then if it happens to be goblins, you can try to estimate how many different goblin tribes live in this ~30 square mile area, and then use that to give a %chance that it's the same tribe you angered, and then if it is, roll a reaction to see if they're still upset, and so forth.
In pure-gamism, we want there to be impact to our choices, and choosing leave a loose-end and then come back should probably have the consequences that you now have to deal with angry goblins on your way back; that strengthens the impact of the choice to leave them, and strengthens the choice about the return route.
In narrativism, it would be *absurd* to not use this payoff for what is so obviously a good plant.
Then, there's a whole spectrum in between; maybe we still roll for a random encounter like good simulationists, but if we get one, it's going to be the goblins. Or maybe it's not automatically goblins, but we make a rulings that there's a higher-than-normal chance that it's goblins or whatever. I just think that "playing it straight" is leaving a lot of value on the table and will produce less interesting games than sacrificing a tiny bit of simulationism
edit: boom, found the post. god reddit's search is bad https://old.reddit.com/r/osr/comments/1pgxhqu/odd_d6_only_damage_how_do_you_make_individual/nsxu1dk/
> " now, because of a few hours spent making meaningful character creation/build choices, way more hours at the table is spent playing wrong-difficulty content. " <
I see your point here and understand where you are going. Your advice seems accurately calibrated to your argument. All good so far.
My question though is what determines "wrong difficulty"? Is it because the challenge and rewards are out of balance? Is it because the players are bored?
Because it seems the problem as you have defined it is "game gives higher rewards than difficulty justifies so players can farm easy stuff". But then you aim all your solution mechanics to character development instead of stocking appropriate levels of treasure.
It seems like a workaround rather than a solution.
Which I think is what you meant by your #2 which you say you've never seen. So do I understand you correct or if not what did I get lost?
> My question though is what determines "wrong difficulty"? Is it because the challenge and rewards are out of balance? Is it because the players are bored?
So say that we design so that we aim for a 1:4 monster:treasure XP ratio, which is still well below what I see in modules.
1000xp worth of defenders (50 bugbears) might be guarding 4000g worth of treasure, spread around a bugbear lair or dungeon wing or whatever.
This is appropriate content for a ~third level party, but inappropriate for a 5th or 6th level party, the bugbears just get fireballed.
You might imagine that the 6th level party thinks this is a waste of time, and so would go instead adventure in areas with stronger defenders and more treasure. In my experience, they *don't*; if the bugbears are around to get fireballed, then fireballing them is worth the easy 5000xp and we just do that handful of sessions in a row (sacrificing fun for power). I've seen this in literally every gaming context i've played, from ttrpgs, to mmos, to single player games