The most far-reaching bad descisions in RPG design?

[Sith_Happens]

Doesn't that relate back to the decision to make "the dungeon" the base unit of adventuring?

The two things are related, yes, but still distinct from each other. To put it in the proper format for this thread, the problem I'm describing is one of Designing Common Problems to Only Be Solvable by Specific Character Archetypes.

In the case of the poor, poor D&D Rogue, this problem overlaps with that of Designing a Character Archetype Around Something that Must Be Done Solo. See also: the Decker Problem.

[Flickerdart]

You can say that about almost anything though. If the players don't want to be treasure hunters why put the hoard in? If they don't want to be dragon slayers why put the dragon in?

A dragon fight has dozens of checks where both success and failure bring about meaningful results - in this case, either the dragon's death, or your own.

[Kurald Galain]

The two things are related, yes, but still distinct from each other. To put it in the proper format for this thread, the problem I'm describing is one of Designing Common Problems to Only Be Solvable by Specific Character Archetypes.

Yes, that's pretty annoying.

Another is designing problems to be only solvable by a particular check. For example,

• The dungeon contains a locked door that must be picked.
• (note, the problem here lies in the scenario writer deciding that the party MUST cross the door and MUST do so by picking the lock)
• Problem 1: parties without a rogue cannot do this
• Solution: make sure every skill can be used by everyone, even untrained
• Problem 2: parties that roll badly still cannot do this
• Solution: design a new system where all skill DCs are low, and skills can be infinitely retried
• Problem 3: you basically don't have a skill system any more. But at least that railroady dungeon is passable by all!

Yeah, that's pretty far-reaching.

[huttj509]

Yes, that's pretty annoying.

Another is designing problems to be only solvable by a particular check. For example,

• The dungeon contains a locked door that must be picked.
• (note, the problem here lies in the scenario writer deciding that the party MUST cross the door and MUST do so by picking the lock)
• Problem 1: parties without a rogue cannot do this
• Solution: make sure every skill can be used by everyone, even untrained
• Problem 2: parties that roll badly still cannot do this
• Solution: design a new system where all skill DCs are low, and skills can be infinitely retried
• Problem 3: you basically don't have a skill system any more. But at least that railroady dungeon is passable by all!

Yeah, that's pretty far-reaching.

I played in a module where the very beginning had a magically sealed door that required a Dispel Magic to bypass (DnD 3.5). No, chipping through wouldn't work.

The min character level listed for the module was 4. At least we had a cleric...and a GM who added some scrolls at a vendor in town.

[Arbane]

I played in a module where the very beginning had a magically sealed door that required a Dispel Magic to bypass (DnD 3.5). No, chipping through wouldn't work.

The min character level listed for the module was 4. At least we had a cleric...and a GM who added some scrolls at a vendor in town.

Ah, yes, the ever-popular plot with a Single Point of Failure.

I said 'popular', not 'fun'. Mystery plots are especially prone to this...

[Cikomyr]

A clever GM should just accept whatever idea his players throw at him to go around this "single point of failure", and make it seem like the players are genius who figured out his clever plans.

A clever ploy is often more.fun and entertaining than having the right skill on your character sheet.

[Frozen_Feet]

Brute force doesn't work against active agents. Any obstacle where failure means you try again without consequence is a boring obstacle. This is the one thing Fate gets right - unless there are interesting consequences for both passing and failing a check, it's not a check that needs to exist.

The way D&D originally got around this was that you could only attempt a specific manner of getting around an obstacle (like picking a lock) once per character level. It was there to tell a player they had to do something else. You had to find another way around, you had to leave the dungeon and go somewhere else etc. Failure or unwillingness to try meant you either were found by wandering monsters or ran out of light and food.

Unfortunately, a lot players missed the memo. And a lot of GMs, too, based on the on-going discussion of rogues.

You all must've noticed. Everyone's talking about how the GM must specifically introduce content for the rogue and prod them in the right direction. There's little talk of the rogues and their players proactively seeking ways to use their skills and prodding the GM in the right direction. And most of that talk is snarking about how you can replace a rogue with a log of wood.

2. There are other ways to get past or around the sorts of obstacles that the Rogue class is specifically designed to deal with, in which case the class is functionally useless.

This is just false. Or rather, it's trivial to make false, through either the player or the GM ensuring that doing it the rogue's way is less costly, more rewarding or just generally better.

I mean, you can say "why steal, when you can just buy stuff?" But what if you want to have both your money and said stuff?

And it doesn't even have to be better, really. Just different, with different outcomes. "Multiple functions can achieve this goal" =/= "All but one fuction are useless".

To put it in the proper format for this thread, the problem I'm describing is one of Designing Common Problems to Only Be Solvable by Specific Character Archetypes.

In the case of the poor, poor D&D Rogue, this problem overlaps with that of Designing a Character Archetype Around Something that Must Be Done Solo. See also: the Decker Problem.

Both of these are actually valid design choices, and can lead to increased strategic depth if done correctly. Majority of problems caused by the first one are solved with non-linear scenario design and allowing multiple ways for a game to proceed. Maybe the non-wizard can't use or disenchant that wand, but they could sell or throw it to a volcano instead. Maybe the non-rogue can't pick that lock, but maybe they can bribe one of the guards to let them in. So on and so forth.

The problems caused by the latter stem largely from treating "Don't split the party!" as gospel, being overly afraid of player metainformation and general inability to play turn based games with actual turns. Sometimes honest-to-god bad design is involved, like with AD&D psionics where the entirety of psionic combat is ruled to be much faster than actual combat, but most of the time it's playgroup inability manage time between participants.

[Segev]

Actually, a more accurate comparison would be "It doesn't take any special skill to pick a lock. All you need is a hammer to smash it."

Your attempt at an reductio ad absurdum argument fails because it's not addressing or refuting my point.

The fact that you state this means we're clearly not even agreeing over what the point is.

My point is that when you start to say, "You don't need any special skill to crack open this chest/door/whatever, because all you have to do is remove the pin in the hinges with a thin spike and a hammer," you're ignoring that the knowledge and skill you're describing is not necessarily something your character's know. You are, in fact, ignoring that having the know-how and practiced muscle memory to perform those acts is what having the relevant skill(s) represents.

Now, yes, if you're just literally smashing the door to flinders or ripping the chest apart with brute strength, that's probably just a strength check.

But when the joke item "the Portable Rogue(tm)" is having all of its uses described, a fair number of them are things which would require, even by DMs I've had who would buy into the notion that it obviates the need for the rogue, an ability check or an attack roll (e.g. throwing it/rolling it down a hallway successfully).

I would argue, in fact, that the "portable rogue(tm)" is more than likely an item which looks easy to make (and may well be), and seems easy to use, but is in fact part of a real rogue's tool kit. It's something he uses to effect his skill checks.

The portrayal of the consummate professional dungeon-delving rogue often features a number of odd-looking behaviors, from dropping grass or dust to see where wind is blowing to testing and probing for traps, etc.

I would, as a general rule, have such "rogue replacements" actually constitute tools which grant circumstance bonuses to the skills that you normally have a rogue to effectuate.


THAT is why my examples include, "You don't need any special skill to make gunpowder: just mix sulfur, saltpeter, and charcoal!"

You're right, that is how you make gunpowder. However, that is not something that is common knowledge to D&D characters; it requires skill in Alchemy to know that and to effectively do it.

If you start to substitute "the player described in detail how it's done, and said his character took each individual step" for "the character has the skill and used it in the game's mechanics," then where do you stop?

This isn't a rhetorical question; it's something that every table needs to answer. Because there is the opposite, equally ludicrous extreme: "I hang my clothes out to dry." "Roll your Housekeeping skill, or you don't know how to dry out your laundry." "..."

My general answer to it comes down to what you feel should be a challenge that characters are able to be experts at resolving. Those things should require reliance on CHARACTER skills/powers/etc., not on player knowledge. Those things which are just interesting ideas and approaches but which shouldn't pose a challenge? Let the player describe it and move on.


But again, my POINT that I was getting at is that a lot of the "we don't need no steenkeeng rogue" arguments rely on the fact that players at the table can conceive of how to describe the expert actions that would be represented by the rogue's skillset. And then assume that not only does "just any" character have the ability to know of those behaviors, but the skill to execute them with minimal need for rolling (perhaps an Ability check, at best).

In reality, it is better - unless you really do just want to remove the rogue as a mechanical entity from your game - to treat the knowledge and capacity to adequately perform these behaviors as part and parcel of the various skills the rogue rolls against when trying to overcome these obstacles.

It's part of his Detect Traps and Disable Device routine to probe ahead with 10-foot poles and "portable rogues(tm)." It's part of his Open Locks skill to know when it's better to just remove the pin from the hinges...and to know how to do that effectively, without risking damage to the contents that might come about by just smashing it.

Yes, the brute force "just smash it" should use the break objects rules. But most of the rest of it sounds an awful lot like obviating character skills through application of player knowledge.

As well say that, because the spells' incantations and gestures are written out for fluff purposes in some rulebook, you can have your Fighter say [insert incantation here] and move his arms and hands in [insert precise gestures here], and that he should therefore be able to do magic. After all, that's all magic is, right? Who needs a sorcerer, with his limited number of spells known and limited number per day?

Who needs the Fighter with his Bull Rush feat? My character's got a strength score and is Medium sized, but really heavy in his armor. I'll just charge into the enemy's space and barrel into him with my bulk. Oh, yeah, and the football player at the table even describes the best way to posture myself to transfer the energy properly. So I do it like that. Why would I need a feat for it?


Does that make my point clearer? I get the impression that you believed me to be arguing something different, from your response.

[Frozen_Feet]

The way I solve it, if a game doesn't have specific rules for a thing, if the player can convincingly describe an action, they get to do it, no roll needed. Player knowlede is character knowledge in such situations. But of course, this can cut both ways:

Player: "I make gunpowder mixing sulphur, saltpeter and charcoal!"
Me: "Where are you going to get sulphur?"
Player: "I... uh..."
Me: "Roll dice."

Many exploitative uses of real life knowledge actually fail if you pay attention to how superficial player knowledge often is. And BOOM, the mechanics are back on the scene.

[Flickerdart]

And most of that talk is snarking about how you can replace a rogue with a log of wood.

Isn't replacing themselves with a log a time-tested ninja rogue technique?

[Frozen_Feet]

Indeed it is.

[Segev]

Isn't replacing themselves with a log a time-tested ninja rogue technique?

Well-played.

[1337 b4k4]

It's part of his Detect Traps and Disable Device routine to probe ahead with 10-foot poles and "portable rogues(tm)." It's part of his Open Locks skill to know when it's better to just remove the pin from the hinges...and to know how to do that effectively, without risking damage to the contents that might come about by just smashing it.

I've also seen DMs (and tend to do this myself) take the old Thief skills and dial them up to 11. Sure, anyone can probe ahead with the 10' pole. But only the Thief can discover that the treasure chest is trapped because only the Thief knows what to look for. Likewise, when the trap gets missed (because someone stepped without thinking for example) only the Thief gets an extra throw after the trigger roll, to notice the trap and avoid setting it off at the last second. And yeah, you might be able to jam the trapdoor with a spike, but only the Thief knows how to completely disable the device and since you didn't find the trap in the treasure chest, good luck disabling it.

Same with the other skills. Anyone can climb walls with grapling hooks and gear, the Thief is the only one that has a chance to do it bare handed with minimal to no hand holds (think the difference between the beginner wall at a rock climbing gym, and the guys that are climbing the negative inclines). Anyone can sneak and move quietly, but when the Thief moves "silently" it's absolute silence. Anyone can hide behind things or around corners, but only the thief can be standing in a dark corner, or just outside the edge of the light and go completely unnoticed.

[Sith_Happens]

And it doesn't even have to be better, really. Just different, with different outcomes. "Multiple functions can achieve this goal" =/= "All but one fuction are useless".

If any character can solve the problems that only rogues are supposed to be able to solve, and non-rogue classes have other problems that they're designed to be able to solve on top of those, then the rogue class becomes useless (or, more accurately, pointless).

[Segev]

If any character can solve the problems that only rogues are supposed to be able to solve, and non-rogue classes have other problems that they're designed to be able to solve on top of those, then the rogue class becomes useless (or, more accurately, pointless).

Well put.

To expand on it a bit:

What can a rogue do that no other class can?

What can a fighter do that no other class can?

What can a wizard do that no other class can?

What can a cleric do that no other class can?


If you cannot answer each of those with something positive, then the question you next need to ask is: does this class deserve to exist as its own thing?

[SimonMoon6]

Okay, those would work if you play D&D at a sufficiently high level. But even then, could you afford preparing them so many times that you can cast them at every door? Resting to regain spells would be a huge problem if getting it done in limited time is an issue.

Of course, the standard dungeon design allows for unlimited resting time because the menacing menaces politely stay behind their doors until the doors are opened. So, in a dungeon, yes, you have unlimited resting time.

And if you're not in a dungeon, then the point is probably moot.

[Segev]

Of course, the standard dungeon design allows for unlimited resting time because the menacing menaces politely stay behind their doors until the doors are opened. So, in a dungeon, yes, you have unlimited resting time.

And if you're not in a dungeon, then the point is probably moot.

Actually, the standard design involves something that all too many groups tend to ignore as "annoying:" wandering monsters. They explicitly exist to prevent this kind of nonsense.

[Lord Raziere]

Well put.

To expand on it a bit:

What can a rogue do that no other class can?

What can a fighter do that no other class can?

What can a wizard do that no other class can?

What can a cleric do that no other class can?


If you cannot answer each of those with something positive, then the question you next need to ask is: does this class deserve to exist as its own thing?

Problem with this thinking is that the wizard and cleric can clearly do all the things that Rogue and Fighter can without any problem, so the conclusion is obviously to get rid of the Rogue and Fighter classes.

but since I imagine that people DON'T want to use spell-casting classes for mundane classes, the question should be:

how can we limit these classes to do things that others can't and make them less flexible? this doesn't mean the wizard has to be only blaster mage. it just means that the wizard and cleric have to be broken up into more specialized classes so as to be with equal roles with the fighter and rogue. a utility wizard shouldn't have access to blaster wizard powers and vice-versa, a healbot cleric shouldn't have the power to be a CoDzilla and vice-versa. an illusionist should only have the power to be an illusionist, and so on. these problems only arise because we allow one or two classes to have too much flexibility and thus screw all the other classes out of the abilities they could potentially get if we weren't so focused on a few classes being incredibly flexible.

[Feddlefew]

Actually, the standard design involves something that all too many groups tend to ignore as "annoying:" wandering monsters. They explicitly exist to prevent this kind of nonsense.

Wandering monsters only really work if there's enough dungeon to wander around in, and no one is wondering why so-and-so hasn't been heard from for a few hours. If the later applies, then it's not a wandering monster so much as an organized assault on the party's encampment. If neither applies, and there's no reason for something new to come wandering INTO the dungeon, I usually look for later encounters that might leave their current locations and stumble onto the PCs. The end result is still that the same creature(s) is fought, just in a different location.

[Eldan]

The way I solve it, if a game doesn't have specific rules for a thing, if the player can convincingly describe an action, they get to do it, no roll needed. Player knowlede is character knowledge in such situations. But of course, this can cut both ways:

Player: "I make gunpowder mixing sulphur, saltpeter and charcoal!"
Me: "Where are you going to get sulphur?"
Player: "I... uh..."
Me: "Roll dice."

Many exploitative uses of real life knowledge actually fail if you pay attention to how superficial player knowledge often is. And BOOM, the mechanics are back on the scene.

So, if, say, one of your players is an engineer with historical interest who does their research, they get to advance technology a few centuries on their own?

[Flickerdart]

So, if, say, one of your players is an engineer with historical interest who does their research, they get to advance technology a few centuries on their own?

"The future is already here, it's just not distributed."

One man, no matter his genius, will only be able to scrounge enough resources for a handful of crumbling prototypes. The pivotal inventions of the ages have become more and more interdependent as time goes on.

[SimonMoon6]

Actually, the standard design involves something that all too many groups tend to ignore as "annoying:" wandering monsters. They explicitly exist to prevent this kind of nonsense.

Yeah, but being of a decent level, the party can always just teleport back home to their nice safe hotel room.

They can then take on Door Number Two the next morning after a full 24 hours of sleep.

[Sith_Happens]

Problem with this thinking is that the wizard and cleric can clearly do all the things that Rogue and Fighter can without any problem, so the conclusion is obviously to get rid of the Rogue and Fighter classes.

Seeing as we're talking about design intent I think this can be ignored.

[Segev]

Yeah, but being of a decent level, the party can always just teleport back home to their nice safe hotel room.

They can then take on Door Number Two the next morning after a full 24 hours of sleep.

Then they're probably taking on dungeons taht are too low-level for them. Standard high-level dungeon design always included rather BS anti-escape or anti-quick-entry mechanisms, precisely because of this.

Problem with this thinking is that the wizard and cleric can clearly do all the things that Rogue and Fighter can without any problem, so the conclusion is obviously to get rid of the Rogue and Fighter classes.

You're right; I worded those incorrectly. It should be: "What should $class be able to do that no other class should be able to do (as well)?"

[Flickerdart]

The distinction doesn't need to be as binary as it currently is. GMs should never have to create situations where only one party member matters.

Instead, it's much better to have classes that have heightened expertise in one area, but are reasonably useful everywhere. Take the niche of the fighter - he is the master of combat, and good at both taking hits and dishing them out. Everyone else can do that too, though, so it's not "well we're in a fight so the fighter can go fight and the rest of us will roll Perform (Thumb Twiddling) checks." Outside of battle the fighter guy can participate in social encounters by intimidating people or swearing upon his knightly honour (but not as well as the dedicated face, who has many other tools at his disposal) or remove obstacles by bashing them (though not as well as the thief, who has the skills necessary to take care of this stuff quickly and cleanly).

It should never be about "what can this class do that no one else can" because every situation where nobody else can contribute is a bad situation. It should always be "what worthwhile thing is this class the best at" and "how can this class still contribute to areas outside its expertise."

[Segev]

It should never be about "what can this class do that no one else can" because every situation where nobody else can contribute is a bad situation. It should always be "what worthwhile thing is this class the best at" and "how can this class still contribute to areas outside its expertise."

Agreed. The trouble comes in when "what this class is best at" is, at least by consensus, "worthless" and "replaceable."

[veti]

Yeah, but being of a decent level, the party can always just teleport back home to their nice safe hotel room.

They can then take on Door Number Two the next morning after a full 24 hours of sleep.

Two words: time pressure. If the party goes away for 24 hours, the world (and the dungeon's inhabitants) are not going to be standing still all that time. There's no guarantee Door Number Two is still there the next day - or the passage it's standing in, for that matter, so welcome to the False Destination teleport table... Serve 'em right if they teleport back into the middle of a strategically-placed gelatinous cube.

So, if, say, one of your players is an engineer with historical interest who does their research, they get to advance technology a few centuries on their own?

It's not quite that easy. You need a certain degree of infrastructure in place to make a real anachronism.

But in general, I would ask them to justify where they got the in-character knowledge from. (And so that "magic" is not a universal answer, one of my consistent houserules is that magical divination can only tell you something if there's somebody, on some plane somewhere, who already knows it.)

[1337 b4k4]

Two words: time pressure. If the party goes away for 24 hours, the world (and the dungeon's inhabitants) are not going to be standing still all that time. There's no guarantee Door Number Two is still there the next day - or the passage it's standing in, for that matter, so welcome to the False Destination teleport table... Serve 'em right if they teleport back into the middle of a strategically-placed gelatinous cube.

It's amusing how much D&D (especially so early D&D) relies on the sort of quirks and oddities that Gary Gygax loved so much. Sure you could teleport out of your dungeon, but given by the time you're teleporting you're at least level 9 (15 if you're taking your party with you, and no level allows you to both teleport your party in and then back out again) and thus have about 9 levels of dungeon above you, filled with twisting turns, false walls, teleporters, trap elevators and slopes, could you even guarantee that you knew where you were to get back to door #2?

[Flickerdart]

It's amusing how much D&D (especially so early D&D) relies on the sort of quirks and oddities that Gary Gygax loved so much. Sure you could teleport out of your dungeon, but given by the time you're teleporting you're at least level 9 (15 if you're taking your party with you, and no level allows you to both teleport your party in and then back out again) and thus have about 9 levels of dungeon above you, filled with twisting turns, false walls, teleporters, trap elevators and slopes, could you even guarantee that you knew where you were to get back to door #2?

This is what beacons are for. A canny spellcaster will leave arcane marks which is difficult to detect and remove. It's a cantrip, so leaving decoys and then choosing which one to return to isn't super hard.

[Telok]

This is what beacons are for. A canny spellcaster will leave arcane marks which is difficult to detect and remove. It's a cantrip, so leaving decoys and then choosing which one to return to isn't super hard.

Recall too that in AD&D Cantrip was a first level spell and a 12th level wizard had four first levels spells a day, he might be lucky and have looted a scroll or two with a couple extra spells. In 3.5 D&Dthe wizard may have 7 first level spells and access to a magic-mart to buy partially charged wands or scrolls. What worked in older editions doesn't work in newer editions with casting bloat and the ability to easily buy and make magic items.

It works even less after 3.5 D&D when the casters have infinite magic uses. When using magic is as accessable and usable as swinging a sword then it is either only allowed to be as powerful as swinging a sword or everyone must be a magic user because the sword users can't keep up.