Help me understand power levels, coming from 3.5

[Kafana]

I used to play and DM a lot of 3.5 D&D, where optimized characters would outclass any commoners and would win 999 out of a 1000 times any skill matchup they were skilled at, from levels 6 and beyond.

Coming to 5e, I see that this power level is greatly reduced, where a 6th level character has a slight edge over a commoner when it comes to skill and especially ability checks (e.g., +3 from proficiency +3 from ability vs a +1 from the commoner). This means that, when it comes to skills and abilities, perhaps looking at roleplaying and skill-based encounters, the adventurer is slightly favored against a peasant. However, when it comes to fighting a monster, the adventurer has 6dX of HP + supernatural damage and cc abilities, while the peasant can't hope to defeat anything.

I realize that the adventurer is specialized in adventuring, as opposed to the peasant, but what I find weird is that the peasant (or say a barbarian PC) might have a similar chance to identify a magic phenomena as a wizard who actively studies them. Is the DM meant to impose restrictions of what a character can know, where a player can argue why his barbarian might know about the elder gods based on the PCs background?

[Arathryth]

I used to play and DM a lot of 3.5 D&D, where optimized characters would outclass any commoners and would win 999 out of a 1000 times any skill matchup they were skilled at, from levels 6 and beyond.

Coming to 5e, I see that this power level is greatly reduced, where a 6th level character has a slight edge over a commoner when it comes to skill and especially ability checks (e.g., +3 from proficiency +3 from ability vs a +1 from the commoner). This means that, when it comes to skills and abilities, perhaps looking at roleplaying and skill-based encounters, the adventurer is slightly favored against a peasant. However, when it comes to fighting a monster, the adventurer has 6dX of HP + supernatural damage and cc abilities, while the peasant can't hope to defeat anything.

I realize that the adventurer is specialized in adventuring, as opposed to the peasant, but what I find weird is that the peasant (or say a barbarian PC) might have a similar chance to identify a magic phenomena as a wizard who actively studies them. Is the DM meant to impose restrictions of what a character can know, where a player can argue why his barbarian might know about the elder gods based on the PCs background?

So a couple of things, I'm not sure where you're getting the +1 for the commoner from, but your run of the mill peasant has no character levels, so no proficiency bonus.

I also played 3.5 for most of my D&D "career", so I understand how jarring the concept of bounded accuracy can be. The idea is for the game to be more streamlined, so the modifiers are smaller. But the DM is supposed to adjucate over the course of the game whether PCs or NPCs can even make certain checks, whether they make a straight roll (plus relevant skill modifier), or whether they have advantage/disadvantage. For example, a Warlock with an Acolyte background might have advantage on a roll to identify a demonic ritual, while the rest of the party has disadvantage because they just don't have familiarity with things of that nature.

[Millstone85]

Coming to 5e, I see that this power level is greatly reduced, where a 6th level character has a slight edge over a commoner when it comes to skill and especially ability checks (e.g., +3 from proficiency +3 from ability vs a +1 from the commoner).
So a couple of things, I'm not sure where you're getting the +1 for the commoner from, but your run of the mill peasant has no character levels, so no proficiency bonus.Actually, monsters and NPCs do have a proficiency bonus, starting at +2 for a CR0 creature. The commoner presented in the MM is proficient wth clubs but not with any skill or saving throw.

[Pelle]

I think that the main thing is that in 5e, you only call for checks when the result is uncertain, and failure would be interesting. It is not a simulation, like in 3.5. The PCs are not commoners, so you usually don't have to make a check for a commoner. In 5e the DM just decides that this commoner has very specialized knowledge about the magical phenomena, or not, you don't use skill ranks et al to model that.

Yes, if it makes sense that the barbarian knows about elder gods based on background or similar, the DM/player can decide he knows or decide on a method for randomly determining it. There are no rules for what the characters know in 5e. Int(Arcana, Religion etc) is for recalling what you already know.

[Unoriginal]

Your typical by-the-book Commoner has +0 in all stats and no skill proficiency. They have 5% chances of succeeding a DC 20 Intelligence check. One success every 20 times, statisticaly.

A PC Barbarian with 10 in INT would also have 5% chances of succeeding this same check. If the Barbarian has 8 or 9 in INT, like many players choose to give them, they straight up cannot succeed that check.

A lvl 6 Wizard with 20 in INT has 25% of chances to succeed that DC 20 ability check. 1 success every 4 tries. That's five time as likely as the chances of someone with 10 INT.

If that Intelligence check is influenced by a proficiency in Arcana, and the lvl 6 Wizard happens to have said proficiency, then their chances becomes 40%. More than one success every three checks. Eight times more likely than the Commoner's or the 10 INT Barbarian's chances.


I don't know how it can be argued that their chances of success are anywhere close.

If you got an INT 20 Barbarian with proficiency in Arcana, they'd have the same chances, but at this point they're the equal to the Wizard in both intellect and specialized training (non-specialized training being covered by the stat, not by the proficiencies).

Keep in mind that contrarily to 3.X, DC 20 is supposed to be hard all game for everyone (only the skillmonkey classes escape that, and it's only if their Expertise come into play). To put into perspective, a DC 20 Strength check is what you need to break a metal chain bare-handed.

[Unoriginal]

I think that the main thing is that in 5e, you only call for checks when the result is uncertain, and failure would be interesting. It is not a simulation, like in 3.5. The PCs are not commoners, so you usually don't have to make a check for a commoner. In 5e the DM just decides that this commoner has very specialized knowledge about the magical phenomena, or not, you don't use skill ranks et al to model that.

Some of the adventure modules use the Commoner statblock for "normal people who got specialized training and are good at their job", like the Dinosaur trainers of Chult (+5 in Wisdom (Animal Handling) checks) or the circus performers you can meet in Waterdeep.

Said checks would still only occure, as you said, if there is a doubt on the success/failure of the action they perform, if the failure is interesting, and if the DM decides to let the dice decide the outcome. Rolling to see if a dino handler manages to calm a wounded dino or if the animal goes in a rampage can be interesting, when the PCs are involved or just nearby.

[Grod_The_Giant]

5e's weakest point is probably its skill system, where low modifiers meet the large flat distribution of the d20 meet single-roll success or failure. You basically have to find ways to work around it-- removing rolls whenever possible, treating proficiencies more like permissions*, leaning into group checks, tailoring results to the characters**, and so on. I'm strongly in favor of using a houserule to make skill checks more consistent-- Advantage on proficient checks, doubled proficiency bonus on checks, rolling 2d10 instead of 1d20, etc.


*ie "you're proficient in Survival, you can follow the tracks without a roll."
**ie "the Barbarian who passes the Arcana check remembers the critical piece of info from an old myth; the Wizard read a book about it and knows the context as well."

[Tanarii]

.I realize that the adventurer is specialized in adventuring, as opposed to the peasant, but what I find weird is that the peasant (or say a barbarian PC) might have a similar chance to identify a magic phenomena as a wizard who actively studies them. Is the DM meant to impose restrictions of what a character can know, where a player can argue why his barbarian might know about the elder gods based on the PCs background?The latter.

Int checks arent state-of-the-character checks. You dont have a quantum state character thay may or may not know something until you make a check.

They're for recalling information your character already knew, under pressure.

They may seem the same from a player perspective, where you roll to get a tidbit of info from the DM. But they aren't the same. Your character must have known something in the first place to even be able to make a check your character to recall it.

A peasant that has never learned about nor experienced magic shouldnt be making checks like that.

A barbarian PC is a different matter. Theyre adventurers, and assumedly have in-universe experience. They probably wont be making book-learning recall checks.
But that could easily be the case for any non-PC, including those proficient in Arcana. But they should certainly have as much commonly known Arcana Lore as the next PC. They just might suck at recalling it.

[Astofel]

The latter.

Int checks arent state-of-the-character checks. You dont have a quantum state character thay may or may not know something until you make a check.

They're for recalling information your character already knew, under pressure.

While I don't necessarily disagree, how do you determine DC? Typically I'd expect a DM to set the DC based on how obscure the information is, but if you're only rolling to recall information you already know, obscurity doesn't really matter. Personally speaking, I'd be inclined to set the DC for obscure information lower using your method, since interesting and obscure tidbits tend to stick in my mind more easily.

[DeTess]

While I don't necessarily disagree, how do you determine DC? Typically I'd expect a DM to set the DC based on how obscure the information is, but if you're only rolling to recall information you already know, obscurity doesn't really matter. Personally speaking, I'd be inclined to set the DC for obscure information lower using your method, since interesting and obscure tidbits tend to stick in my mind more easily.

On the one hand I agree, on the other hand, the obscure little tidbits are also more important to get exactly right. Dragons being big and scary and magical is fairly common knowledge, the exact meanings of the color coding is less so, and getting the breath weapon of the green and red dragon reversed is rather problematic (representing a failed roll to recall the exact details of the fun little draconid factoid).

[Kadesh]

While I don't necessarily disagree, how do you determine DC? Typically I'd expect a DM to set the DC based on how obscure the information is, but if you're only rolling to recall information you already know, obscurity doesn't really matter. Personally speaking, I'd be inclined to set the DC for obscure information lower using your method, since interesting and obscure tidbits tend to stick in my mind more easily.

The assumption is that you can make an Intelligence based check to learn about something/anything: but unlike 3.5 there is no calculation to determine how much you know about a creature, and a DM has more recompense to say 'that isn't something you'd potentially have even learned', unless of course you are specifically making about learning that: Lore Bard Knowledge Cleric, or Tome Warlock, or Hunter Ranger, may be doing a lot of reading up about it and have a wider baseline for knowledge, as opposed to a pig ignorant barbarian who makes a knowledge check about an enemy they have never seen, rolling a 20, and learning a load about it.

3.5 had a hard cap on what an untrained check wpuld give, but made no bones about whether the character trained in demon fighting would know devils and angels to the same standard, or how a Dragon Hunter would have the same ability to know about the Inevitables. 5e just put the limit in the hamds of a DM, and can be resolved by you backstorying it.

At the risk of everyone writing in they are a loremaster of everything, every character has a specific background which allows a DM to essentially one stop shop judgement call. An Acolyte, Sage or potentially Hermit, sure. Guild Merchant or Urchin? Less so. The background system is excellent for allowing you to build atypical characters: barbarians may cater to being idiots, not having abilities keying from Int, but if you want to play a Barbarian who can think, you can take one of the appropriate background withiut the pain of Cross Classing Skills: and that way, as you get a bonus on Character level simply for being a higher level, you get better without having that lack of ability score investment.

And yes, the general DC set is as follows: DC30, nigh impossible, DC10, pretty easy, most should be able to achieve this if they have to think about it. Given the bounded accuracy, a proficient character with the maximum in the ability score has a 10% chance of achieving that DC30. If they have Expertise, or an Ioun stone of Mastery (or both: and the Stone has an attunement requirement, of which you have 3 Max) it gets easier but they both have their requirement (Dex/Int 13 for a rogue multiclass and delayed advancement of feats/Stat increases, class features, spell progression etc, or being a certain race, and delaying your primary/secondary stat focus).

[Tanarii]

On the one hand I agree, on the other hand, the obscure little tidbits are also more important to get exactly right. Dragons being big and scary and magical is fairly common knowledge, the exact meanings of the color coding is less so, and getting the breath weapon of the green and red dragon reversed is rather problematic (representing a failed roll to recall the exact details of the fun little draconid factoid).
"Red Drgons prefer to to exchange jokes with their prey before attack, unlike Copper who prefer a show acknowledging their obvious superiority."
- Barbarian fails his Int check and becomes a snack.

[Millface]

Basically, in 5E low level characters aren't gods, they're people who are slightly better at these kinds of things than the average Joe, and it stays that way until level 11, where you will once again feel far, far superior to commoners. Keep in mind that it's not just your prof bonus that sets you apart, but your stats. The village smithy who works the forge all day might have a strength of 12, whereas your human fighter starts with 16. That, plus proficiency, makes you +4 more likely to succeed, or 20%. I think that sounds fair at level 1. By level 5 you have 18 Strength and +3 Prof, so +6, or 30%. On top of your class abilities.

That's the second kicker: 5E is far less about stats and pluses and far more about your class abilities. In my opinion this is great. A 5th level 3.5 fighter rolls 20s and that's it. A 5th level 5E Fighter has Action Surge, Second Wind, Indomitable, and possibly even maneuvers from the battlemaster. None of which a commoner can do. It's about a slightly better than average person becoming great by learning a class instead of a massively better than average person being great because they were born that way.

[Beelzebubba]

While I don't necessarily disagree, how do you determine DC? Typically I'd expect a DM to set the DC based on how obscure the information is, but if you're only rolling to recall information you already know, obscurity doesn't really matter. Personally speaking, I'd be inclined to set the DC for obscure information lower using your method, since interesting and obscure tidbits tend to stick in my mind more easily.

I'm inspired by the rumor-mongering mini-game in early modules - the charts of rumors contained a mixture of truth, exaggeration, and falsehood, and persistence got the characters many rumors which they could cross-reference to get a closer approximation to the truth.

In that vein, I think of the DC as 'how effective the information is to me at the time', mixed with 'how confident I am in the knowledge'. I also like to use the 'variable success' kind of rule, where a low roll is a partial success, because that drives the game forward - and in the world of knowledge, that appears in many ways.

Try these for a low roll:

- They know four facts, and two of them that you provide are false.
- They heard an exaggerated version that is so outlandish they're not sure if it's true or not.
- They heard a thing that was true, but from a doubtful source, so they are taking a risk in trusting it.

It's hard to do this all the time, but every once in a while makes knowledge checks more fun than a binary 'truth' or 'not'.

[Knaight]

The skill system basically precludes competence outside of combat or spell casting, outside of a few niche cases - the short version is that 5e kept the idea of the skill monkey, and represented that by them actually being better at their skills, which at high level can actually get really solid. Everyone else suffers from this though.

The other thing is that the ability check system is clearly made for the combat system first and foremost with the rest sort of bolted to the side, and while it works just fine for that combat system (where more attacks, more damage, magic items, class features, etc. add a lot more than the gain in Proficiency ever will), which causes issues. This also has some occasional oddities even in the context of combat skills.

That said the criticisms of 5e characters being flimsy in combat are, to put it mildly, completely ridiculous unless you've got a specific set of expectations built to where "only" being able to handle a dozen soldiers by yourself in a single straight fight is somehow pathetic.

[PhoenixPhyre]

I'm going to counter @Knaight's post above.

A fundamental shift in philosophy between 3e and 5e is that 5e's ability checks are not an attempt to simulate anything. They're an action-resolution mechanism, nothing more. That means that an ability check (no, it's not a skill check, those don't exist) is only to be rolled if both

a) failure is likely
b) failure is interesting.

This means you're rolling a lot fewer checks, but each check is meaningful and no one should be able to go off-d20 without specific features or spells (because otherwise condition a is false). Many, if not most, of the things you'd roll for in 3e are supposed to be automatic successes (no check necessary). Climbing most surfaces, swimming most waters, jumping (within certain bounds based on strength score), etc. are all no-check-needed success. This puts the baseline for 5e characters even without proficiency much higher. Couple that to the much lower DCs (20 is as high as most published DCs go) and you have characters succeeding where 3e characters would fail without specialized builds/equipment.

On the other hand, since checks are for things where failure is likely, reliability in 5e (being able to do that every time) comes from class features/feats. Specifically expertise (rogue 1/bard 2/feat), which allows a level 5 character to auto-succeed at any task of DC 10 or lower in a secondary stat (+3 MOD + 6 expertise means minimum roll is 10, with no auto-fail on a 1) and a level 17 character to auto-succeed on a DC 16 check (+3 MOD + 12), which accounts for the large majority of checks (since DC 20 checks are supposed to be rarer than DC 10/15 ones). Rogues also get reliable talent at level 11 which pushes them up to a minimum of 14 (proficiency but no stat mod) and more likely 15-20 or 25+ (for something they have expertise in).

Being an expert in something, by 5e design, requires class levels and class features. That's consistent with the rest of the design--just having more total levels doesn't make you overwhelmingly better in specific areas. That's what class features are for.

From a subjective standpoint, I never felt more useless than playing a PF character. And I was playing a skill-monkey-esque type (oracle specialized in skills using INT -> CHA abilities). The DCs were high, I was rolling all the time, and things that I could do without rolling in 5e took rolls. It felt like the point was to either drive your skill bonuses high enough to obviate the rolls or not try the things at all. This makes it totally binary, which is stupid (in my eyes). Why even have a dice mechanic if you'll never roll for anything important because you'll either fail all the time or you'll succeed all the time?

[Grod_The_Giant]

3.PF has its own set of issues with skills, no question. But "only roll when failure is interesting" is a universal rpg guideline, and is useful however good or bad the skill system is.

[PhoenixPhyre]

3.PF has its own set of issues with skills, no question. But "only roll when failure is interesting" is a universal rpg guideline, and is useful however good or bad the skill system is.

That's true. But 3.PF doesn't believe in that good principle. It instead only obeys the first principle (roll only when failure is in doubt). And since the set DCs tend to be either trivial with any slight investment or require maximum specialization, you end up with a binary result. Either you don't try because you can't succeed or you trivially succeed. This, in turn, basically turns the skill system into another "you must be this tall to ride" sign (as well as a resource tax) instead of a way of introducing interesting complications and adjudicating player actions. It's either an almost insurmountable barrier or it's trivial, locking a bunch of character concepts out of the whole subsystem (poor fighters).

5e instead accepts that failure can be interesting. And so it prioritizes keeping things in the "possibly interesting" range as much as possible. Being able to break out of this (expertise, reliable talent) can be then held as perks, signs that you've grown powerful, rather than a necessity to interact with the system. You're expected to fail at ability checks in 5e. But that's OK, because you're not making ability checks for things that aren't interesting.

5e also allows the DM more flexibility to make the mechanics fit the fiction (instead of vice versa). If there's something that's just bound to work (from a fictional point of view), it just does. This is contrasted to still having to bow to all the mechanical hoops--after all, you can't use a lot of skills while untrained, but even someone not practiced in certain things can still come up with a faultless plan occasionally (by playing on the situation). Even an uncharismatic curmudgeon can convince someone of something if it's what they wanted to believe already and given the right argument. 3.PF locks that away behind gates.

So to the point that I was responding to, 5e characters, despite the mechanical indications, are more competent in the main than 3.PF characters. Yes, the ceiling is lower (mostly), but the floor is so much higher that everyone's succeeding more at things. 5e presumes competence (and has you roll only for things that are difficult already, despite the monikers "easy", "medium" and "hard"), while 3.PF presumes incompetence until you spend resources on skills and items (and magic) and has you (by default) roll for everything.

[Celcey]

I think something important to consider as well is that +1 in 5e is not the same as +1 in 3.x. In 5e a +1 actually makes a tangible difference to your ability to do something, which is why +X items are a much bigger deal and don't really go above +3.

[Tanarii]

3.PF has its own set of issues with skills, no question. But "only roll when failure is interesting" is a universal rpg guideline, and is useful however good or bad the skill system is.
Really? Because the 3r I remember was "roll any time you won't auto succeed", with everything having an effective DC. As if skill checks were the underlying structure of doing anything in-universe, not a resolution mechanic when one was needed.

In fact, that's the entire logic behind "Take 10" and "Take 20". Everything is a roll, but put some control in the players hands to auto succeed for things that make sense to succeed on.

[PhoenixPhyre]

Really? Because the 3r I remember was "roll any time you won't auto succeed", with everything having an effective DC. As if skill checks were the underlying structure of doing anything in-universe, not a resolution mechanic when one was needed.

That's my experience as well. It's a completely different philosophy of what checks are for, and causes 99% (in my experience) of the disconnect with skills for people coming from 3.PF.

[Tanarii]

That's my experience as well. It's a completely different philosophy of what checks are for, and causes 99% (in my experience) of the disconnect with skills for people coming from 3.PF.
Honestly, I'm willing to bet if went back and looked at e 3e PHB or DMG there's be some commentary on when to roll.

But the DC tables going down to DC 0, take 10 rule for when you're not under pressure, and take 20 rule for when you have time, all belie that.

If they'd wanted to make it about DCs not underlying every task and action in the in-game universe, those things would have been structured more like 5es versions of them (passive checks and automatic success rule) instead.

(Edit: despite my tone, you better believe I found 3e's "revolutionary" skill system absolutely and jaw dropping-ly amazing at the time. :smallwink: It's just that it engendered a different mindset from what 5e is trying to do with its ability checks.)

[PhoenixPhyre]

Honestly, I'm willing to bet if went back and looked at e 3e PHB or DMG there's be some commentary on when to roll.

But the DC tables going down to DC 0, take 10 rule for when you're not under pressure, and take 20 rule for when you have time, all belie that.

If they'd wanted to make it about DCs not underlying every task and action in the in-game universe, those things would have been structured more like 5es versions of them (passive checks and automatic success rule) instead.

Take 10 and 20 always felt like a hacked-on attempt to cover the "only roll when it's interesting" case when they realized they were rolling for everything, but only a band-aid, not a true fix.

Relevant MM Twitter chain on the philosophical differences:

https://mobile.twitter.com/mikemearls/status/1041057506628255744

[Astofel]

Yeah, I haven't had much experience with 3.5, but with Pathfinder and also Pathfinder 2e I find that I'm asked to roll for things that 5e typically wouldn't. I was playing the PF2 playtest last night, and we had to cross a river then climb a cliff. The module asked us to roll Athletics for both of these things, despite the lack of pressing circumstances (there was a time limit to get where we were going, but it was on the order of days rather than hours or minutes). I couldn't help but wonder what the point of rolling was, failure was only an annoying short-term delay rather than any sort of consequence.

Also when it came to climb the cliff, the druid with 8 strength and me, the sorcerer with 8 strength and no training in Athletics both managed to pass the DC 19 check needed to climb the cliff while the fighter and ranger, both with decent strength and training in Athletics both failed, so that kind of thing is obviously not unique to 5e.

[Knaight]

I'm going to counter @Knaight's post above.

A fundamental shift in philosophy between 3e and 5e is that 5e's ability checks are not an attempt to simulate anything. They're an action-resolution mechanism, nothing more. That means that an ability check (no, it's not a skill check, those don't exist) is only to be rolled if both

a) failure is likely
b) failure is interesting.

This means you're rolling a lot fewer checks, but each check is meaningful and no one should be able to go off-d20 without specific features or spells (because otherwise condition a is false). Many, if not most, of the things you'd roll for in 3e are supposed to be automatic successes (no check necessary). Climbing most surfaces, swimming most waters, jumping (within certain bounds based on strength score), etc. are all no-check-needed success. This puts the baseline for 5e characters even without proficiency much higher. Couple that to the much lower DCs (20 is as high as most published DCs go) and you have characters succeeding where 3e characters would fail without specialized builds/equipment.


Sure, but all of this also applies to say, Fate, or Burning Wheel. Yet in those games good skills can be a whole lot better than bad skills, and the curved distributions innate in either 4d3 (which is what Fate uses, just shifted) or the binomial distribution that dice pool systems tend to follow make that really come through.

The 3.x skill system drives me up a wall, but as someone who generally plays stat-skill systems 5e just isn't that impressive.

[PhoenixPhyre]

Sure, but all of this also applies to say, Fate, or Burning Wheel. Yet in those games good skills can be a whole lot better than bad skills, and the curved distributions innate in either 4d3 (which is what Fate uses, just shifted) or the binomial distribution that dice pool systems tend to follow make that really come through.

The 3.x skill system drives me up a wall, but as someone who generally plays stat-skill systems 5e just isn't that impressive.

I'm not sure this is responsive. This whole thread is in the context of comparing 3.5 to 5, and that was the context I was replying to. Bringing in other systems is rather off-topic. Specifically, what I posted was a distinction between 3.5e and 5e--that 5e realized that we should roll only when it makes sense (rather than only when it's mechanically impossible to fail). That's the philosophical shift between those two editions. And it's a major one.

Whether or not it's better than some other random system (which has many more salient differences and so cannot be compared directly without cherry-picking) is totally irrelevant to the thread.

[Sigreid]

Just to respond to something up thread about the barbarian vs. wizard arcana check. IMO it is possible for the party to come across some obscure arcane phenomenon and there be a slim chance that the wizard hasn't come across it in his studies, but the barbarian remembers something relevant that he heard from the tribe's story teller/witch doctor/shaman/whatever. It won't happen often, but it's not out there for it to happen occasionally.

[Kadesh]

Just to respond to something up thread about the barbarian vs. wizard arcana check. IMO it is possible for the party to come across some obscure arcane phenomenon and there be a slim chance that the wizard hasn't come across it in his studies, but the barbarian remembers something relevant that he heard from the tribe's story teller/witch doctor/shaman/whatever. It won't happen often, but it's not out there for it to happen occasionally.

Exactly, and that can be represented by a Barbarian rolling a Nat 20 with a - 1 Mod on Arcana and a Prodigy Wizard rolling a 1 with a +17 to the check, and failing the DC19 test, but in 3.5, that just wasn't a thing.

If the DC was low enough for the uninvested Barbarian to pass, a Wizard was usually rocking somewhere around +35 to the check.

[Tanarii]

Just to respond to something up thread about the barbarian vs. wizard arcana check. IMO it is possible for the party to come across some obscure arcane phenomenon and there be a slim chance that the wizard hasn't come across it in his studies, but the barbarian remembers something relevant that he heard from the tribe's story teller/witch doctor/shaman/whatever. It won't happen often, but it's not out there for it to happen occasionally.
IMO using a check to determine that is like determining if your character learned how to swim Rapids when making a (unusual circumstances) Strength (Athletics) check to cross a raging river. Or if they happen to know a particularly important and persuasive argument they can make against a particular individual when trying to convince someone with a Charisma (Persuasion) check.

It's a way to play. But it's a carryover from 3e and 4e Knowledge checks. State-of-your-character isn't inherent in the 5e Intelligence check rules to recall Lore.

[Sigreid]

IMO using a check to determine that is like determining if your character learned how to swim Rapids when making a (unusual circumstances) Strength (Athletics) check to cross a raging river. Or if they happen to know a particularly important and persuasive argument they can make against a particular individual when trying to convince someone with a Charisma (Persuasion) check.

It's a way to play. But it's a carryover from 3e and 4e Knowledge checks. State-of-your-character isn't inherent in the 5e Intelligence check rules to recall Lore.

Actually, it's a carry over from the West End Games Star Wars system where attributes cover broad areas of education and ability and then adding dice to skills indicates a focus in a particular area of that broad swath of training, inborn ability and experience. It works for me because I'm often surprised at what people know and can do that they didn't realize they knew or could do until the situation came up. I mean, I couldn't sit here and just rattle off everything I know. There typically has to be a trigger to find a particular bit of info.

[Knaight]

I'm not sure this is responsive. This whole thread is in the context of comparing 3.5 to 5, and that was the context I was replying to. Bringing in other systems is rather off-topic. Specifically, what I posted was a distinction between 3.5e and 5e--that 5e realized that we should roll only when it makes sense (rather than only when it's mechanically impossible to fail). That's the philosophical shift between those two editions. And it's a major one.

These comparisons routinely involve more absolute statements about both system, not strictly relative comments - and these more absolute statements are better contextualized with a broader viewpoint.

[PhoenixPhyre]

These comparisons routinely involve more absolute statements about both system, not strictly relative comments - and these more absolute statements are better contextualized with a broader viewpoint.

Not really, because that "broader viewpoint" usually just ends up being "I hate D&D and you should too."

Everything in this thread has been about comparisons, implicitly or explicitly. Even from the OP (from the title even). And your first post (to which I was responding) didn't mention other systems. It talked about D&D only, making what were totally incorrect statements about 5e's power levels.

[Knaight]

Everything in this thread has been about comparisons, implicitly or explicitly. Even from the OP (from the title even). And your first post (to which I was responding) didn't mention other systems. It talked about D&D only, making what were totally incorrect statements about 5e's power levels.

Every statement I said was correct - 5e still uses a d20, and +0 to +11 on a d20 still isn't that impressive a difference, regardless of how a system contextualizes it. 5e can present it however it wants; that doesn't change the actual behavior of the system math.

[Kadesh]

Every statement I said was correct - 5e still uses a d20, and +0 to +11 on a d20 still isn't that impressive a difference, regardless of how a system contextualizes it. 5e can present it however it wants; that doesn't change the actual behavior of the system math.

You're going to have to explain to me the point of using dice in a game that has the dice where you can remove any aspect of dice rolling.

[PhoenixPhyre]

Every statement I said was correct - 5e still uses a d20, and +0 to +11 on a d20 still isn't that impressive a difference, regardless of how a system contextualizes it. 5e can present it however it wants; that doesn't change the actual behavior of the system math.

...you just don't get it. What you're rolling for (and when you're rolling that d20) makes a huge difference in how the mechanics translate to the fiction layer. A much bigger case than rolling different dice ever does.

Consider two (extreme) cases:

Case one:
Everything you do (each step, etc) requires a separate roll. Failure means you fail (fall down, etc). In this case, having a chance of failure in the 1% range is too much--the system better allow you to "take 20" (auto-pass) some of those results. Since you can't auto-pass everything, even if the numbers are huge you still are going to fail at simple tasks and feel weak. This is exacerbated if there's a separate skill for each task and, even if you're amazing at something you specialize in, but you suck horribly at things you're not specialized in. Yes, you can balance on a cloud, but you can't swim an inch without drowning.

Case two:
Only "heroic" things need to be rolled for. Things that Hercules would struggle with. Everything else is an automatic success. In this, having a 1% chance of success can be enough, and a +1 (+5% chance) is huge. But even if you fail, you're still regularly doing things worthy of an epic hero.

3e is much closer to case 1 than 5e is. 5e is much closer to case 2. Are there systems that are nearer to either one? Maybe. But that's irrelevant in this context.

Context matters. Philosophy matters. It matters much more than raw mechanics. For most players, mechanics are pretty darn irrelevant. Statistics (the shape of the bell curve) is especially irrelevant--you don't roll enough comparable dice in any one session to see the shifts unless you're throwing shadowrun-size dice pools around. Add in the fact that humans are notoriously bad at statistics, and I'd go so far as to say that the exact dice mechanics are one of the least important things about a rule-set. The philosophy of when you should roll those dice is always more important and more meaningful.

[PhoenixPhyre]

And to be a bit more general here--

5e takes the PoV that

* Rolling dice is fun. Mechanics that obviate the need to roll dice at all should be exceptions, not the norm.
* Uncertainty is fun. If you can predict the outcome of a task with 100% certainty, there's no tension. No dramatic question. Just fast forward.
* Failure is (or can be) fun. Pure power fantasies can be fun, but they get boring pretty fast. Having to adapt to things going wrong is fun for a lot of people.
* Success also is fun. People should have a reasonable chance to succeed, as well as a chance to fail. Many, if not most, things don't need a roll at all--they just happen.

Put these together and you're rolling dice for things that matter, about which there's uncertainty, and for which failure can be fun.

This means that 5e requires systems where going "off dice" are rare and powerful. This does not mean that 5e characters are weak. If they are, you're rolling for things that don't matter and using the system wrong (just like if an AW GM starts rolling dice, he's doing it wrong).

[johnbragg]

Something no one has mentioned is how the bonuses interact with the advantage/disadvantage mechanic.

With a +0 modifier, a DC 15 check means a 75%/25% failure/success split.
With a +5 modifier, a DC 15 check becomes a 50%/50% failure/success split.

Bring in Advantage (for any number of circumstantial reasons) and you have:
With a +0 modifier, a DC 15 check with advantage becomes a 56%/44% split.
With a +5 modifier, a DC 15 check with advantage becomes a 25% failure/ 75% success

[PhoenixPhyre]

Something no one has mentioned is how the bonuses interact with the advantage/disadvantage mechanic.

With a +0 modifier, a DC 15 check means a 75%/25% failure/success split.
With a +5 modifier, a DC 15 check becomes a 50%/50% failure/success split.

Bring in Advantage (for any number of circumstantial reasons) and you have:
With a +0 modifier, a DC 15 check with advantage becomes a 56%/44% split.
With a +5 modifier, a DC 15 check with advantage becomes a 25% failure/ 75% success

That's a good point. Also to note that while advantage/disadvantage can move the success chance around, advantage =/= success and disadvantage =/= failure. It gives a nice spectrum of possible results. A high enough bonus can compensate for disadvantage, or a clever plan (for advantage) can compensate for a low modifier. Or they can stack (in either direction).

[Knaight]

...you just don't get it. What you're rolling for (and when you're rolling that d20) makes a huge difference in how the mechanics translate to the fiction layer. A much bigger case than rolling different dice ever does.

Consider two (extreme) cases:

Case one:
Everything you do (each step, etc) requires a separate roll. Failure means you fail (fall down, etc). In this case, having a chance of failure in the 1% range is too much--the system better allow you to "take 20" (auto-pass) some of those results. Since you can't auto-pass everything, even if the numbers are huge you still are going to fail at simple tasks and feel weak. This is exacerbated if there's a separate skill for each task and, even if you're amazing at something you specialize in, but you suck horribly at things you're not specialized in. Yes, you can balance on a cloud, but you can't swim an inch without drowning.

Case two:
Only "heroic" things need to be rolled for. Things that Hercules would struggle with. Everything else is an automatic success. In this, having a 1% chance of success can be enough, and a +1 (+5% chance) is huge. But even if you fail, you're still regularly doing things worthy of an epic hero.

3e is much closer to case 1 than 5e is. 5e is much closer to case 2. Are there systems that are nearer to either one? Maybe. But that's irrelevant in this context.
5e is marginally further towards the latter - in play, as demonstrated by both WotC modules and records of people playing 5e still sees a lot of rolling, for things which aren't hugely impressive. Just look at the definitions used, where 5e calls a DC 10 check "easy", and there's a pretty high failure chance on that for all but the highest level adventurers, or those with expertise.

There's also the matter of how frequency of rolls is system philosophy, but the definition of difficulty standards is absolutely mechanics. Had 5e defined its mechanics differently (e.g. DC 0 being easy, DC 5 being moderate, DC 10 being hard, DC 15 being heroic; this would actually fit your case 2) some of its behavior would look very different, though it would still run into many of the same issues.


Context matters. Philosophy matters. It matters much more than raw mechanics. For most players, mechanics are pretty darn irrelevant. Statistics (the shape of the bell curve) is especially irrelevant--you don't roll enough comparable dice in any one session to see the shifts unless you're throwing shadowrun-size dice pools around. Add in the fact that humans are notoriously bad at statistics, and I'd go so far as to say that the exact dice mechanics are one of the least important things about a rule-set. The philosophy of when you should roll those dice is always more important and more meaningful.
This is where it goes off the rails. You don't need to throw that many dice around to see how the mechanics play out in broad strokes, and while it takes a lot of rolling to pick up on really subtle differences things like what the baseline success rates are and how much better one character is than another become immediately apparent. Every time a group rolls for the same thing in 5e the extent to which the random factor dominates skill becomes apparent.

[laman132]

If knowledge checks aren't about what you know, but what you can recall in the moment, can I reroll knowledge checks infinitely out of combat?

[Wub]

I'm of the philosophy that a commoner's nat-20 would get different results from a wizard's nat-20 when it comes to knowledge. A commoner might be able to recall an old hedge-wizards ramblings from when they passed through town. A trained wizard, however, might have read an entire research paper on the subject and thus have much more information to go on.

Likewise, a mighty leap for a commoner could be a regular leap for a barbarian.

[GlenSmash!]

In 4 years of playing and DMing 5e I've never once had a Player playing a Barbarian (much less a player playing a commoner, if that even is a thing) describe an approach to a situation such that an Intelligence (Arcana) check was merited.

Ability Checks are not fundamental to the D&D universe. They are a DM tool for resolving uncertainty when adventurers do adventuring stuff.

[Kadesh]

If knowledge checks aren't about what you know, but what you can recall in the moment, can I reroll knowledge checks infinitely out of combat?

Ask your DM.

[Tanarii]

If knowledge checks aren't about what you know, but what you can recall in the moment, can I reroll knowledge checks infinitely out of combat?
Recalling things checks are the same as any other check. The automatic success rule comes into play. You can take ten times as long and automatically succeed, provided it's possible to succeed and there's no consequences for failing a single check other than time. If you knew it and can eventually remember it given time, you can just cut to the chase. So effectively yes, under those circumstances.

With a state-of-my-character "knowledge" check, there is a consequences for failure. You never knew it. So one roll, and automatic success cannot be used.

[Solusek]

5e's weakest point is probably its skill system, where low modifiers meet the large flat distribution of the d20 meet single-roll success or failure. You basically have to find ways to work around it-- removing rolls whenever possible, treating proficiencies more like permissions*, leaning into group checks, tailoring results to the characters**, and so on. I'm strongly in favor of using a houserule to make skill checks more consistent-- Advantage on proficient checks, doubled proficiency bonus on checks, rolling 2d10 instead of 1d20, etc.


*ie "you're proficient in Survival, you can follow the tracks without a roll."
**ie "the Barbarian who passes the Arcana check remembers the critical piece of info from an old myth; the Wizard read a book about it and knows the context as well."

Those are some really great ideas. It would be nice if they were in the book so I could count on more DM's doing that kind of thing. Too often DM's just say "everyone roll history" or "everyone roll arcana" to find out if anyone knows something, and don't give it a second thought. The weakness of the skill system really shows itself in such situations.

[Malifice]

So a couple of things, I'm not sure where you're getting the +1 for the commoner from, but your run of the mill peasant has no character levels, so no proficiency bonus.

Your run of the mill peasant is most likely a Commoner NPC, and as a CR'd monster they most certainly do have a proficiency bonus.

From memory they arent proficient in anything by default, but most would be assigned Proficiency in their chosen professions tools/ skill.

[Spriteless]

Hey, Kafana! Are you still here?

First level characters are more powerful in 5th ed. Everyone gets Spring Attack for free. Lots of cantrips and 1st level spells have utility besides damage. Everyone has cool setting related powers from their backgrounds, like a fighter who was raised in a religion can always find lodging within branches of their church, or if you worked for a guild you can have a letter that says you are a legitimate craftsman never mind the thieves' tools at the bottom of your luggage. You have a lot of control over how the civilized world sees your character that is supported by the rules. Charm Person was nerfed, so legitimate standing is more useful than being an enchanter/bard.

Mid level characters seem a little less powerful. You don't get as many attacks in around, and the numbers don't go up quite so high. Since AC doesn't get that high, low level monsters are still quite a threat, if there are enough of them. Flexible spell casters still do more to change the world than more mundane characters, unless those characters have been cultivating their background boons. Since that option is also available to spell casters, that's not compensation exactly, but it is something.

But the rules stop covering high powered characters before they did in previous editions. There isn't much about strongholds or followers or business ownership in the DMG, if you want that kind of thing, err, I think Matt Colville is gonna sell a 3rd party supplement, or you could grab rules from previous editions, or from Kingsmaker. You have the rules writers' official permission to fix their mistakes and fill in their gaps. Whether this counts as a fix is up for debate, in this very thread, kind of rudely. But it is the same old argument over and over, ha ha.

[laman132]

Recalling things checks are the same as any other check. The automatic success rule comes into play. You can take ten times as long and automatically succeed, provided it's possible to succeed and there's no consequences for failing a single check other than time. If you knew it and can eventually remember it given time, you can just cut to the chase. So effectively yes, under those circumstances.

With a state-of-my-character "knowledge" check, there is a consequences for failure. You never knew it. So one roll, and automatic success cannot be used.

Then what is the point of giving explicit DC's for knowledge checks in, say, published adventures, if they're out of combat tidbits? There are loads of examples. For example, from Dragon Heist, page 116



Cassalantar Lore
A successful DC 15 Intelligence (History) check reveals the following information about Cassalanter Villa and its inhabitants:


Why not just say "characters with proficiency in History know the following bits of information"? It isn't in the middle of combat, and players could just go back to their house and puzzle it out till they remembered. The published adventures seem to be indicating that knowledge skills are, indeed, 'state-of-character' checks. It isn't like they shy away from that wording, from page 94



Painting of a Dwarf Miner
A painting of a dwarf miner priced at 5d10 gp can be found each day a character spends searching art shops. A character proficient with painter's supplies can create a painting that qualifies as this key.


It doesn't ask for checks, because proficiency implies competence enough, and there are no costs of failure. I'm confident there are other examples, too, though this is the only book I have on me right now.

[Galactkaktus]

I'm of the philosophy that a commoner's nat-20 would get different results from a wizard's nat-20 when it comes to knowledge.

Raw it probably would be different since skill checks doesn't critt the commoner would thus get a lower result on his check.