have you tried not playing dnd?

>>89540598
Nobody said anything about playing dnd.

I don't like the physical objects, and I don't believe the large range is really necessary or even particularly useful. Why would you want to use them over other dice anyway? Except the d4 I mean, those are horrible.

>>89540670
The results varying largely from roll to roll might be useful in a game in which you have a set range of target numbers to avoid things getting stale, like games in which you roll under an attribute.

>>89540592
You may have fundamentally misunderstood the distinction between physical dice and the "d20" rpg framework which uses 6 attributes, classes, levels, spells or abilities at a per-day rate (and otherwise mostly free), tends to span a wide range of levels of play in a single system, and rolls d20s plus mods against a DC with binary success.

>>89540748
I am aware of the distinction, it just seemed to me like both the action resolution using the die and the design framework were issues for people.

>>89540720
I don't agree at all. For a binary pass/fail roll all those extra numbers give you is a finer resolution: 5% increments rather than 10% or 17%. But unless you're making huge numbers of them the difference between a 70% chance of success and a 75% chance of success is too small to feel much different.
If the degree of success or failure mattes then the large range will be more important, but all the systems I can think of that do so use multiple dice because the large swing is generally not desirable there.

>>89540812
You raise a good point and I think it makes for at least one way issues with d20 games could be fixed which is having a non binary resolution system. As for games using the d20 that rely on rolling under I can name Dragonbane and the Dark Eye, the latter accounting more for the swing through its skill system.

>>89540967
I don't see how roll-under helps anything.
The large swing makes it relatively favourable to extreme results, which is probably most suitable for a comedy oriented system. There are ways to reduce the effect of the swing, but the simplest one is just to use multiple dice.

>>89540592
>So how would you go about making an rpg using the d20?
I'd ask the players to roll it. If the number was big enough, they'd succeed.

>>89541108
>the simplest one is just to use multiple dice.
not the guy you're replying to, but that's what the dark eye does
it's 3d20, roll under, and everything rolls under multiple attributes
if I remember right, you can trade extraneous success for failure up to a point. so if you succeed by like 10 on two rolls and fail by 1 on the third you can still pass kinda thing so nobody's too MAD

>>89540592
I wanted to use d20 exclusively for my homebrew and spend months "fixing" it, but at the end it all comes down to things everybody already know:
a) the range is too wide. Gradiality doesn't worth swinginess;
b) it is too hard to make modifiers feel fair and balanced;
c) modifiers are the best for d20, but it is hard to keep track of them no matter how keen you are.

>>89540592
For 1d20 resolution a 10% chance of a heckin' nattyrino on any roll is too high

>>89540592
>I figure that if there is such a big issue with systems that use it might as well take the opportunity to talk about them and maybe even iron them out. So how would you go about making an rpg using the d20?
I don't think there's anything wrong with a d20 as a randomizer.
But character bonuses need to matter more relative to the range of possible outcomes.

If the ogre barbarian has a +5 to a Strength Check and the halfling wizard has a -1, the Wizard still ends up with having a good chance to over power the guy that's 5 times his size.

>>89542191
I think if you replace "critical rolls" with critical success/failure being based on margin of success. Like doubling the target number, or failing it by half.
Though that still leaves you with many players who are conditioned by DnD to play it wrong.

>>89540592
I'm just not a fan of the idea that doing alright is just as likely as failing miserably is just as likely as great success. The flat distribution just dosn't feel great.
2d10 or 3d6 systems feel better because there is a greater feel of reliability, you know your own abilities pretty well and can approach tasks with some level of confidence. Then when you happen to roll especially high or low it feels more impactful because it happens less.

>>89540967
>>89540812
I think non-binary degrees of success is probably the best use for the large result range of the d20. You can look at Pillars of Eternity for an example, where attacks can miss, glance, hit, or crit based on degree of success or failure.
This also means "natural 1/20" doesn't exist as a concept, which may or may not be a positive but certainly seems to be a sticking point for detractors of the d20 system.

>>89542303
I forgot to add, degrees of success based on success+x (or failure-x) are especially suited to a d20 game with modifiers, because they make modifiers more meaningful whereas in The D20 System they can feel granular and arbitrary.

>>89542285
I agree. I like 3d6.
For me a 1/108 chance of a hecking natirino is about where I want to see it. You might get one or two a game. Not 1/10.

>>89540592
>So how would you go about making an rpg using the d20?
no crits
NO crit fails
when assigning DC, recognize a quarter of all rolls will be 5 or less

>>89540592
>Making the D20 work
Why? Plenty of other games already work and I don't need to work hard to make them do so.

>>89542007
>not the guy you're replying to, but that's what the dark eye does
Yes, and it's a nightmare. Rolls can take a minute to resolve and GMs generally don't know what they're doing once they add difficulty modifiers.
>if I remember right, you can trade extraneous success for failure up to a point. so if you succeed by like 10 on two rolls and fail by 1 on the third you can still pass kinda thing so nobody's too MAD
Not quite, you can use your skill points for compensation. Every 3 skill points you have left increase your quality of success. But rolling lower doesn't do anything except if you get double or triple 1s (=crit success)

>>89540592

The twenty-sided die works just fine. Modifiers you add and subtract from the rolled result apply a clear modification to the dice result. Yes, rolling a single die makes for a flat-line in probability that doesn't look pretty, but it is practical.

What about using the d20 doesn't work?

>>89542283
> the Wizard still ends up with having a good chance to over power the guy that's 5 times his size.
This is a meta issue, though. If you’re treating it like a simulation and rolling for things that shouldn’t be left up to chance, then there are far bigger issues than physical disparity between wizards and barbarians.

File: d20.jpg (978 KB, 3147x3147)

978 KB JPG

>used for more than 2000 years
>but some neckbeards don't like it because of meme association with a game most of them don't even play
Yeah, I'll continue to just not listen to you.

>>89542285
> I'm just not a fan of the idea that doing alright is just as likely as failing miserably is just as likely as great success.
Neither is D&D given that this is not what a d20 roll represents under most circumstances. You either pass or you fail. If you pass by rolling exactly your target number, or if you pass by rolling triple your target number, it doesn’t matter. It’s still just a “pass”. Likewise it doesn’t matter how much you fail by - if it’s 1 or 5 or 10 or 1000, it’s still just a “fail”.

The only consistent exception in modern D&D (that is, 5e) is a critical hit and automatic miss on an attack roll, in which case rolling a natural 20 allows you to use double your damage die for rolling damage; while rolling a natural 1 means you simply miss (but due to the way 5e’s math works, even if it wasn’t an automatic miss, you’re unlikely to hit your target anyway).

That’s it. If you’re rolling an Athletics check to jump over an obstacle and roll a 20, all it means is you’ve rolled 1 better than a 19, it doesn’t mean that you have jumped so well you can launch yourself into orbit. If you roll a 1, all it means is you’ve rolled 1 worse than a 2, you haven’t brained and killed your character by running headfirst into the obstacle. You might actually still pass if your modifier is high enough because a 1 is not an automatic failure in skill checks.

>>89547293
I dont think you understand. The anon was talking about curves.
The likelyhood of getting 1, 10, or 20 is equal on a d20.
The likelyhood of getting 2, 8, 11 is not equal on a 2d6

>>89547742
>The likelyhood of getting 1, 10, or 20 is equal on a d20.
>The likelyhood of getting 2, 8, 11 is not equal on a 2d6

And none of that matters. What matters is your chance of doing what you wanted to do. If you're trying to jump over an obstacle, then what matters is whether you've succeeded or failed at jumping over the obstacle.

You can add in some complications or gritty details - do you just about clear it, do you clip it and injure your leg, do you leap over it with such panache that you actually manage to accelerate when landing - but at the end of the day that's just just adding some details to the central question: did you succeed, or did you fail?

>>89547742
To continue on a theme, I don't understand the appeal of curves when most 3d6 systems I'm aware of are still basically trying to hit a range of numbers. All it does is make average rolls extremely likely and extreme rolls extremely unlikely, but I think there's a disconnect where advocates of 3d6 forget that ROLLS are not the same thing as RESULTS. That being, the question of whether you succeeded or failed at what you were trying to do.

>>89540748
No they haven't. The majority of posters in /tg/ think that the polyhedral is more important than the resolution mechanics.

You could make a post about 2d20 or Numenera and get several posts of
>ew d20
>omg it's D&D
Even though they're both completely wrong.

I've seen people cry that all 3d6 games are GURPS, but d20 is far more common.

>>89548077
>>89548189
Results are connected to Rolls.
Example: your bonus to acrobatics is +7. Using a d20 means you have a 5% chance of getting 8. and a 15% chance of 10 or less. Dc 11, so 15% chance to fail.
With 3d6 likely hood of getting a 10 or less is 0.5%.
In other words the likelyhood of your relatively good at acrobatics guy slipping on a simple check is 30x as likely with a d20 as it is with 3d6.
Of course it works the other way around too and getting great results is similarly affected.
I personally prefer swingy dice, but in more gritty games swinginess is not appreciated.

>>89540592
>how would you go about making an rpg using the d20?
I wouldn't. I can think of some cool ideas like 3d20-take-middle, but honestly there is no reason to use the d20 outside of some tables and maybe I dunno, some ultra-killotron weaponry or something.

>>89548077
Are you the resident retard that consistently refuses to understand basic probability?
The number generator you use affects the odds of doing what you want. In particular it affects how modifiers affect base chances and interact with one another.

>>89540592
Drop AC and HP. Make combat interesting with dodge, block and parry.

File: Reality.png (38 KB, 1147x411)

38 KB PNG

>>89548516
No, I understand basic probability just fine, I just don't understand why you're all so obsessed with getting a number generator that tends to produce average rolls instead of extreme rolls more often when that's not what you're ever trying to do. You're not trying to consistently roll the average number, you're trying to get a total result that's at least or better than a target number, which means that it's totally irrelevant whether you use 1d20 or 3d6 or anything else since your math is going to be centered on the probability of hitting that range, not the probability of getting the average result.

Like, say that I've decided that a game is most fun when a typical roll intended to be "challenging" has about a 65% chance of success, i.e., I want the players to succeed more than half the time while still having good odds of failure.

With no modifiers, on a d20, that's 8 or better. In 3d6, it's about a 10 or better. The fact that on 1d20 I have a 5% chance of rolling an 8 while on 3d6 I have a 12.5% chance of rolling a 10 DOES NOT MATTER, because I'm not trying to hit those specific numbers, I'm trying to roll those numbers of better, and the odds in both cases are about 65% (slightly lower than that for 3d6).

You people are totally in the wrong for looking at bell curves and claiming them to be the most beautiful thing ever conceived of by man, because you're not rolling in such a way that the bell curve matters when you're trying to roll at least or better than a target number. The graph you actually want to be looking at is pic.

>>89548360
In 3d6 systems the numbers are different, a roll of 17 or less to succeed makes the character one of the greatest acrobatics masters of all time, not "relatively good" at acrobatics. But if you've only got a single d20 you can't modify the chance of critical failure below 5%. Just a problem with using one die, and these limitations get written off by DND fans as "teehee swing"

>>89540592
I don't like the d20 or d20 system. There are several reasons for this but there is no reason to rehabilitate this die when I can just use 3d6 or a WoD style dice pool system and get all the verisimilitude and balance I need at once.

File: sdgdfg.jpg (97 KB, 1592x712)

97 KB JPG

>>89548077
>What matters is your chance of doing what you wanted to do.
yes
and that varies drastically when you sum multiple dice vs reading one die

>>89548717
I didnt mention roll under its roll over. I was just how the difference works. I know the confusion happened because most games that use 3d6 use roll under.

>>89548360
Why would the modifier be +7 in both a 3d6 and 1d20 system? Why would the target DC be the same in both systems? The number generator is different so they shouldn't be.

Let's set the target DC to 16 in a d20 system. On a flat roll that's a 25% chance of success. The same chance of success in 3d6 is about a 13.

If I give someone a +7 bonus in d20, their odds of passing a DC 16 check rise to 60%. The roughly equivalent bonus in a 3d6 system against their DC 13 is a +3 (for 62.5%).

>>89548707
>No, I understand basic probability just fine
Yeah, I thought it was you.
Again, you're ignoring that games generally involve more than one die roll, and various situational differences that affect those rolls differently. The issue is not and never was, "What are the odds of making a typical check?" but "How to typical modifiers affect the odds of a check?" Which we have to consider in light of diminishing and accelerating returns: the difference between a 50% success rate and a 55% success rate is barely noticable, but the difference between 90% and 95% is huge, never mind 95% to 100%.
And besides all that I would rather have a system that accounts for degrees of success or failure rather than binary pass/fail. And the swing of the d20 is far more noticable and wonky there.

>>89548189
Because it's simpler and easier to design your game based on the lovely 3d6 curve that rolls give you than trying to back engineer a curve into the game when it has a flat d20 for rolls. You're talking about autistically complicated math at a certain point and, like, why? Why not just avoid that headache and use a system that already solves the d20's core issue of being absurdly swingy and unbelievable?

>>89548717
>But if you've only got a single d20 you can't modify the chance of critical failure below 5%

You can't modify the chance of critical failure at all because THERE IS NO SUCH THING AS CRITICAL FAILURE.

A natural 1 is an automatic miss on an attack roll. Nothing more. Further, due to the math behind 5e a natural 1 is unlikely to result in a success on that attack roll anyway. Most attack modifiers end up being about +8 or +9 in most games (since D&D games tend to end around 10th level), meaning a natural 1 + modifier of 9 = 10, which is going to be too low to hit the AC of every single monster or NPC a 10th level character is going to be fighting. They don't hit on a 2 or 3 either, it's merely a formality that 1 is an automatic miss.

But on every other roll, such as skill checks, a 1 is just a 1. It is not a "critical" anything. If you've got a +13 modifier then you've rolled a 14. if the DC was 14 or lower, you pass. If the DC was 15 or higher, you fail.

Learn the game before you complain about the game.

>>89548842
First, 5e is not the only, and far from the first d20 system. Second, while the rules often do clarify that you cannot critically succeed or fail on skill rolls by rolling a 1 or 20, arguably there are more D&D games where this rule is ignored than enforced, and the community of 5e heavily, heavily incentivizes using this house ruling, to the point that WotC was planning to make the change official. I don't know if they pulled the trigger on that, I don't play or have any interest in that dumpster game, but it's really not absurd to point out that in many, many d20 games and systems, a 1 is a critical failure and a 20 is a critical success.

>>89548831
>And besides all that I would rather have a system that accounts for degrees of success or failure rather than binary pass/fail.
That's completely possible when using a d20, but just adding in stuff like, "if you fail by X or more" or "if you pass by X or more". This is not an objective problem with the d20, this is your personal, subjective gripe about how it's been implemented in the d20 System as a whole (and not even every iteration of it; check out SpyCraft 1.0 sometime).

>>89548833
>Because it's simpler and easier to design your game based on the lovely 3d6 curve
How in God's green Earth can it possibly be simpler to design a game around a curve with varying probabilities for each point on that curve then a straight line where each point on the line represents a straightforward 5% increase? Now you're just straight-up lying.

>Why not just avoid that headache and use a system that already solves the d20's core issue of being absurdly swingy
Because it's NOT SWINGY, you moron! It only appears to be swingy when you decide that you care about the odds of hitting a specific number, but you are never trying to do that! You're trying to hit a range of numbers. You're not trying to roll a 15, you're trying to get a result that is AT LEAST a 15 but can and likely will exceed it.

>>89548891
>First, 5e is not the only, and far from the first d20 system
Yeah but critical failure was also NOT a thing in 4e, 3e, or 2e.

> arguably there are more D&D games where this rule is ignored than enforced
Don't care. We cannot meaningfully argue about every single person's house rules, otherwise this conversation literally cannot advance as we each of us keep bringing up anecdote after anecdote. We deal with the rules as written or we both shut up, there are no viable alternatives.

>>89548792
This is the most retarded post in this thread. More than the "curve doesnt affect anything" guy.

>>89548948
>flat probability
>value could be a 1 or a 20, no way to tell what you'll get
>waaaahh stop bullying d&d, it's not swingy!!!!!
You're retarded. You are trying to staple a believable probability curve onto the d20 by inventing a convoluted series of modifiers and target numbers. That you cannot understand why this is vastly more complicated than simply using the GURPS 3d6 curve, I do not know. It is mind-boggling to me that you insist upon this when it's very clear that your method is insane, and nobody has ever designed a good system with this idea of yours as the baseline.

>>89548948
>That's completely possible when using a d20
I never claimed otherwise. But it's extra work over just using different dice, for no benefit I can see.
>this is your personal, subjective gripe about how it's been implemented in the d20 System as a whole
I don't know what d20 system you're talking about. I don't play any d20 based games, because I don't like the dice or see any advantages to them.
This thread is not about whatever specific system you're trying to white knight for. Presumably some form of D&D, but I haven't played any of them. The thread is about using d20s. I don't like them. You've not even tried to give any reasons why it's useful.

>>89540592
I actually enjoy the D20 beyond its association with D&D, for its 5% increments. This allows me to make tables with up to 20 outcomes, and to easily make tables with outcomes of varied probabilities.
While it's true that D100 would allow for finer probability variance between outcomes, and to even have a greater number of outcomes on a table, I find myself leaning towards increments of 5% anyway, so I may as well just use a D20.
The D10 just doesn't suit how I prefer to make my tables, and every other die doesn't divide 100 into integers.

>>89549317
>I just can't fathom probability unless it's in whole numbers divisible by five or ten
I guess this is the use case for the d20, having a completely wrinkle-free brain.

>>89540592
The issue isn't the d20, the issue is none-curved distribution of probability.
2d10 is superior to 1d20 in nearly ever conceivable way.

>>89549494
It's not about fathoming, it's about convenience.

>>89540592
2d20 roll under, 1 die under is success, 2 die is great success.

>>89549317
>Doesn't use curved distribution
Having 3-18 on 3d6 let's you make a list where some numbers are naturally more likely to happen.

>>89549584
Good for them.

>>89549584
It also makes you roll 3 dice for each resolution instead of 1.
If you're into that, cool, but I'm not.

>>89549676
>t. baby with small hands
Can you not add?

>>89549764
Can you not read?
-> >>89549542

>>89540598
Have you tried paying dnd?

>>89549800
Apparently you can't since you'd see I didn't reply to that post, anon.

>>89549887
If you don't read the other replies to a post you're responding to, how do you know you aren't repeating a concern that was already covered?

>>89549909
I did, anon. When I posted >>89549542 wasn't even on the page for me. You do know pages don't auto refresh, right?

>>89549960
>When I posted, 89549542 wasn't even on the page for me
>>89549542: 13:10:00
>>89549584: 13:13:33
It took more than 3 minutes to type ~23 words?
Holy shit. Holy fuck.

>>89550010
>Anon cannot fathom reading an entire 50+ post thread in less than three and a half minutes
Actual brain damage.

>>89550073
*In more than
God I'm tired today.
>>89550010
If I started reading the thread at 13:09:59 I wouldn't have seen your post anon.
Please learn how this website works before you start posting and revealing your summer status.

>>89550073
You saw my first post just fine.
What, did you read the entire thread from the beginning after seeing my post? That would really be retarded.

>>89550092
So in addition to being unable to read, you also can't type.

>>89550111
I didn't see that post, are you actually stupid and don't think people read threads?

>>89550125
>He's doubling down
Posts made after you open a thread dont populate on your browsers page unless you refresh. Please learn how the website works. You're making a fool of yourself.

>>89550130
>I didn't see that post
You would have had to seen it to post >>89549584 as a response to it.

>>89540598
Posted suggested alternative or shut up.

Posting generations is also a shut up. Post a specific system.

>>89548948
>How in God's green Earth can it possibly be simpler to design a game around a curve with varying probabilities for each point on that curve then a straight line where each point on the line represents a straightforward 5% increase? Now you're just straight-up lying.
NTA, but I find that a normalish curve better models additive modifiers.
Forgive the food analogy, but if you're trying to cook where you're trying to hit say 10 then good ingredients (+2) and good equipment (+2) will boost your odds from 62% to 95% in a 3d6 based system, whereas in d20 you're only going from 55% to 75%.
This makes modifiers and slight edges in skills a lot more meaningful where you're near that 50% part of the curve, whereas in d20 based systems a +1 by itself doesn't mean much, you have to stack a ton of modifiers to get something closer to a guaranteed result.
This also makes for a lot more variety between rolls without the actual numbers having to vary a lot - having to hit 12 (37%) vs 13 (26%) feels a lot more different than in D20, where difference of something be 45% vs 40% likely to succeed doesn't really feel meaningful.

>>89550152
>he doesn't refresh the page before responding to a post

File: Useful site functions.jpg (61 KB, 965x221)

61 KB JPG

>>89550235
You actually don't even need to F5 the entire page.

>>89548948
>>89550218
to elaborate on this, if you want to relate to real life skill we can trivially take chess as an example where the elo rating system (and other systems like glicko) assume a normal curve in player performance and do quite well in predicting results. We also see normal curves in basically any other statistical measurement whether it's seeing how high people can jump or presumably how well they swing a sword. If a roll is meant to check player skill (say jumping) then assuming they have a baseline height of whatever and a normal distribution of results around that height is natural for checking success.
I also thought of an example regarding stacking modifiers - in Go you give extra stones as a handicap to the weaker player, and these are directly proportional to the difference in rating (although you use less at a higher rating) - each stone helps shift that normal and can radically change the odds of the match rather than just shifting them by 5% at a time

>>89549584
I think D20 is crap for resolving skill rolls but for a table a more flat distribution makes perfect sense - why would you write up an entry for something that's only going to happen 0.5% of the time? 5% is already plenty rare for an event to happen during a campaign, you're not going to roll a table that many times (and if you are, having something come up 20 times as often as something else isn't really a good thing

>>89550218
What you're actually demonstrating there is the lack of granularity in 3d6 systems and how you should use different step sizes for bonuses depending on your resolution mechanic.

>>89550332
No it isn't.

>>89550188
If you had any reading comprehension you'd see >>89549584
Isn't a reply to >>89549542 but a reply to >>89549317. Please use your eyes.

>>89550332
How would you use different step sizes for bonuses when rolling to cook a meal to model the fact that having multiple things helping you is going to account for a big swing in the chance for success?
And it's not that 3d6 lacks granularity, it's that it's pushing it towards the ends where it's more meaningful - having a difference between .5%, 2%, and 4.5% is giving you things that are successively twice as likely to fail with the high end being basically guaranteed. D20 has no equivalent of "basically guaranteed (99.5%)" or even "pretty much basically guaranteed (98%)", you either have guaranteed success or "mostly guaranteed but not really (95%)" with a strong downwards trend from there.

>>89550332
Earthdawn did that and it's one of the worst parts of it.

>>89550332
No, it's about diminishing/accelerating returns. See >>89548831
This applies both in terms of making a decent game (modifiers stay relevant regardless of your base success chance) and realism (an advantage won't normally give you a flat 5 percentage point increase in success).

>>89550422
>having a difference between .5%, 2%, and 4.5% is giving you things that are successively twice as likely to fail with the high end being basically guaranteed. D20 has no equivalent of "basically guaranteed (99.5%)" or even "pretty much basically guaranteed (98%)", you either have guaranteed success or "mostly guaranteed but not really (95%)" with a strong downwards trend from there.
No decent system and certainly no decent GM would have you roll for shit that you're pretty much basically guaranteed to succeed.
>>89550443
My assumption was that you choose one resolution mechanic with an appropriate range of bonuses for your system and stick with that. Earthdawn seems plenty weird and terrible to play, but I only know it second-hand from an SA goon's writeup.

>>89550325
NTAYRT
>I think D20 is crap for resolving skill rolls
"Skill" rolls are retarded to begin with.
>why would you write up an entry for something that's only going to happen 0.5% of the time?
There are certainly benefits to this, granting an incredibly rare boon or bane to throw a wrench in things can cause some excitement, for people who are into that. I wouldn't do it, personally.
>you're not going to roll a table that many times
I make sure any tables I create will have prominent use, whether it's a random hit location from a hasty attack, the effectiveness of an attack/defense, an item (and its quality) when harvested, the type and quality of a loot item, the quality of a crafted item, etc.
If a table is only going to be rolled from every once in a while, I don't really see the point in having it.
>having something come up 20 times as often as something else isn't really a good thing
I 100% agree with this.

>>89550599
>No decent system and certainly no decent GM would have you roll for shit that you're pretty much basically guaranteed to succeed.
This is a retarded take, and my assumption is that you've never run anything if you think it's reasonable to have to know and gauge the individual abilities of every character at the table. I will set a DC. Maybe your character can hit it, maybe he can't. It's not my job to know whether you can hit any given DC.

>>89550689
>It's not my job to know whether you can hit any given DC.
It literally is your job though.

>>89550689
>Retard doesn't know about Passive Checks and taking 10.

>>89550712
Every edition after 2e was a mistake.

Another aspect to d20 being bad is the fact that level increases in skills aren't particularly meaningful. Getting an extra 5% chance to beat something really isn't much
>>89550599
I've never had anyone skip rolls in combat outside of very rules-light games. And even a 99% chance to succeed is going to fail every hundred rolls or so, it's a safe bet but still a bet

>>89550625
>There are certainly benefits to this, granting an incredibly rare boon or bane to throw a wrench in things can cause some excitement, for people who are into that. I wouldn't do it, personally.
I think this makes more sense if you're writing a game or a splatbook rather than doing tables for a specific campaign.
For things like magical items or locations I prefer "tiered" tables where you go through various subtables which can get you rare variants. This gets you uniqueness without that uniqueness being contingent on hitting a particularly lucky roll. I also just think they're fun to go through.
>whether it's a random hit location from a hasty attack, the effectiveness of an attack/defense
Surely this is more of a system thing than a table?

>>89548993
> >value could be a 1 or a 20, no way to tell what you'll get
But you’ve missed the part where there’s also no reason to care. You’re not trying to roll exactly 20 or avoid exactly 1. All that matters is the odds of success or failure and from that the number range you’re trying to roll in. If it’s a flat d20 roll and the target number is 15, then there is no difference between a 1 and a 14, and no difference between a 15 and a 20. There’s just the 70% failure chance and the 30% success chance.

>>89548998
> But it's extra work
No, it really isn’t. Look, I’ll prove it.

3d6, target number is 14, roll has a +4 modifier. Now give me the odds of:
- Resist between 7 and 10 (“major failure”)
- Result between 11 and 13 (“marginal failure”)
- Result between 14 and 17 (“success”)
- Result between 18 and 22 (“great success”)
And do all of the math IN YOUR HEAD. No calculator, no Anydice, no scratch paper, it has to be just in your head.

Now, compare/contrast: 1d20, modifier of +6, target number of 17. Give me the odds of:
- Result between 7 and 12
- Result between 13 and 16
- Result between 17 and 21
- Result between 22 and 26
And again, do it all purely in your head.

The odds are 30%, 20%, 25%, and 25%, respectively. And I know this because each target point has a 5% chance and there are 20 possible results, so all I have to do is multiply the number of points in the range by 5).

You cannot, with a straight face, argue that the odds are easier to calculate with 3d6 then they are with 1d20.

>>89540592
>>89540592
Surely there must be a D20 app that players can download.

>>89551130
>You cannot, with a straight face, argue that the odds are easier to calculate
I would never claim that's remotely relevant either. You don't need to calculate those odds. Ever.
With 3d6 you subtract the number you rolled from the target. With your system you do that, then an extra step based on remembering your specific reduction function. It's explicitly more work.

>>89550712
No, it fucking isn't. It's my job to set DCs. It's not my job to know the skill proficiencies of every fucking PC in every fucking skill. That's the players' job.
>>89550734
What the fuck does this have to do with what I said, you fucking dipshit? By the way, not every system has this, brainiac.

>>89550787
>I've never had anyone skip rolls in combat outside of very rules-light games. And even a 99% chance to succeed is going to fail every hundred rolls or so, it's a safe bet but still a bet
Literally had one of these fucking retards say not to have your players roll when they're fighting an incorporeal enemy. Just announce that they missed and skip the action. You're talking to someone who never ran a successful game in their life. They're not living in reality.

>>89551130
when are you calculating these odds?
Had to think a bit because normally you're doing roll under in 3d6 but I'm pretty sure it's 62% to succeed (same as rolling 11 or under) and then a 15% chance to hit 18 (rolling 14, same as rolling 7 or under)
If you're even a bit familiar with the system it's not that hard to roughly memorize the odds for everything, you've basically just got 7 numbers that aren't 0%, 50%, or one of the other numbers mirrored
50%, 62%, 75%, 85%, 90%, 95%, 98%, 99.5%

>>89551130
>All that matters is the odds of success or failure and from that the number range you’re trying to roll in.
Incorrect.

>>89551130
>calculating the odds
Is this some DnDshit, shouldn't those odds translate to general levels of ability and difficulty of the task, instead of making it easier for pussy ass spergs to calculate their exact mathematical chances before making a move?
Anyways doesn't change that the lowest chances of a natural (not necessarily a critical) are either 0% or 5% with nothing in between.

File: Tagmar - Tabelas de Resol(...).pdf (297 KB, PDF)

297 KB PDF

Table of results.

>>89551130
I don't think anyone will try to argue that 3d6 is easier to calculate for. It's more about that you're more likely to roll mid numbers muh bell curve and makes the chances of getting extremely high, and extremely low numbers unlikely. Therefore making it more predictable in general. It's easier to notice this when you need a high or a low number. Trying to roll a 20 with that +4 modifier would be almost impossible. Which makes modifiers and skill points more valuable.

>>89551255
> With 3d6 you subtract the number you rolled from the target.
To…what? If my target is 14 and I rolled a 12, that leaves 2. But the odds of rolling 14 or better on 3d6 are 16.2%, not 2% or 20% or whatever. So you subtract the number you rolled from the target number, and…?

Either you’ve forgotten a step or you’re not as good at math as you think you are, and I’m inclined to think it’s the latter.

>>89551497
So in other words it requires rote memorization rather than following a formula that can be easily explained as “number of points in the range x 5”.

>>89551631
If saying it only made it so.

>>89551725
True, but why do you need odds lower than 5%? And since the lowest result is not necessarily an automatic failure, why does it matter if the lowest result is a 5% chance?

> Is this some DnDshit, shouldn't those odds translate to general levels of ability and difficulty of the task, instead of making it easier for pussy ass spergs to calculate their exact mathematical chances before making a move?
They can’t. If you actually played the World’s Greatest Roleplaying Game(tm), you’d know that the DM is never under any obligation to tell players any target DCs. The DM says something like, “make an Athletics check”, but doesn’t tell the players the target number for success or failure is 15, he only tells them if they’ve succeeded or failed after the roll is made.

However, being able to quickly and easily calculate odds is useful from a game design standpoint, both at the level of individual DMs, and at the level of publishers trying to make challenging adventures.

I don't think I've ever seen anyone actually take into account the player's odds of success on the fly based on their roll bonuses, D20, D100, 2D10, or 3D6. Most handbooks include some level of difficulty curve of DC which assumes a certain level of skill investment to reasonably achieve the DC. But what usually happens is the GM looks at the chart of DC's or modifiers and decides what seems roughly appropriate for the task at hand, then the one or two people who have the highest odds of success roll. The math is already done for you, you aren't stopping the game to run calculations in the middle of a session. And players shouldn't be told the roll DC upfront in scenarios like this (there are some where it's relevant like AC/To-Hit or opposed rolls where you do tell them they have to beat X number, but the default is they do not know the number they have to roll to succeed at the task)

>>89552586
>why do you need odds lower than 5%?
Are you fucking retarded, or just being obtuse? You're just now realizing this is the rub? This is the entire reason people are saying they don't like d20. Because when you're extremely good at something, you don't have a 5% chance of failure at it, you have a fucking 0.05% chance or some shit. Can you stop posting retarded shit?
>being able to quickly and easily calculate odds is useful from a game design standpoint, both at the level of individual DMs, and at the level of publishers trying to make challenging adventures.
This is very easily solvable with a quick DC guide. And no, the exact statistical probability of getting a particular DC is not really that important at all, though it can easily be provided in a table. What's actually important is the relatively likelihood of something happening (this is very easy to hit, this is pretty easy to hit, this is nearly but not totally impossible to hit).

>>89552414
> and makes the chances of getting extremely high, and extremely low numbers unlikely.
I just don’t see the benefit to this when you combine it with a binary pass/fail system that does not care about a very high or very low number being rolled, or even with a gradient result system that, again, doesn’t care about specific numbers being rolled, just a range of numbers. In that case all that matters is the odds of landing within that range.

The fact that small modifiers are more impactful also doesn’t seem to matter because ultimately all a modifier is, is a percent increase to your chance of success or failure, and the d20 system’s modifiers do the exact same thing.

Although something I didn’t touch on, the fact that modifiers in 3d6 also increase odds of success in a difficult to calculate way. Against a target number of 14 or better, your odds of success are 16.2% for a flat roll, 25.93% for a +1 (improved by 9.43), 37.5% for a +2 (up 11.97), and 50% (up 12.5) for a +3.

In d20, the modifiers provide simple and easy +5% bonuses per +1 modifier to your odds. +1 is 5%, +3 is 15%, and so on.

>>89552679
> Because when you're extremely good at something, you don't have a 5% chance of failure at it
Depends on what it is, now doesn’t it? Also are you forgetting that the entire rolling system generally presumes that you are rushed or under pressure and thus not necessarily able to be as good at the thing as you’d be in a more sedate, white-room situation? Also-also, since skills can’t critically fail, very easy tasks are basically automatic passes anyway. Very easy tasks have a DC of 5. Even moderately competent people will have a modifier of +4. Natural 1 + 4 = 5.

You need to stop thinking of natural 1s as failures. They’re not. They’re just the worst you could have done in circumstances when you are rushed or under pressure. It is possible that this will result in a failure, but it’s also possible that it couldn’t.

You also need to start remembering that the game has a leveling system built into it that provides extremely high modifiers over time. My 15th level 5e Thief at the end of her adventuring career had a +16 bonus to Dexterity (Acrobatics) checks. On her worst, clumsiest day after a night of hard drinking resulting in a hangover straight from the Nine Hells, her absolute lowest result was still a 17 with a natural 1. She could with zero effort still do what ordinary people can only do 20% of the time. She had a 50-50 chance of rolling a 26, i.e. a feat of acrobatics that is utterly beyond even the greatest natural talent on the planet (DC 25) and approaches something considered to be outright impossible (DC 30). She could in fact do things considered to be outright impossible slightly more than one-third of the time.

>>89540661
have you tried not playing dnd, though?

>>89552773
Just to note, Rogues have Expertise so they choose skills they get double Proficiency bonus on (+10 at level 15 instead of +5 in this case). I'm also just guessing you have a magic item as there's no way to break 20 in a stat otherwise which would leave you at +5 from your DEX mod, unless I'm forgetting miscellaneous bonuses or rules. And you'd need at least a 22 DEX to get the +6 you'd need to have +16 in this case.

The major point however is that you're getting an extra 33-50% bonus to your modifier on Skill Checks compared to characters who do not have Expertise, because the most all classes not named Bard, Rogue (Expertise), or Barbarian (gets to break Stat Cap) will ever have is +5 from a Stat and +6 from Proficiency. +11 still makes a DC 25 check notably difficult, and a DC 20 a bit of a gamble.

>>89552948
>Also are you forgetting that the entire rolling system generally presumes that you are rushed or under pressure and thus not necessarily able to be as good at the thing as you’d be in a more sedate, white-room situation
No, that's bullshit. Trained characters are paradoxically more impacted by a stressful situation than untrained ones.

>>89553384
Manual of Quickness Action, yes, although the +15 I’d otherwise have does not significantly alter the point I was making. As for Expertise, while Rogues and Bards have easiest access to it, there are still multiple other avenues through race or class choices - and more importantly through fear choices, which are technically optional rules but which I’m more willing to allow in conversation since unlike critical misses or critical successes on non-attack rolls, I have *never* encountered a table that didn’t allow them. And on top of that the upcoming 5.5 edition appears to also be set on standardizing them.

Nevertheless, even with a more natural +10 at that level, that would still mean such a character would be able to, on their worst day, automatically pass DC 5 and 10 checks, pass DC 15 checks 80% of the time, DC 20 checks 55% of the time, and on rare occasions pull off utterly impossible feats at DC 30.

The point being, the people who go on about “you shouldn’t be failing 5% of the time!” Seem to forget that thanks to modifiers and the fact that you can’t auto-fail skill checks, you’re *not*. Not at the simple, easy tasks. You’re only failing at things that are actually *hard*, and even then you’re a lot less likely to fail then someone who’s a neophyte.

>>89554210
No they’re not. They’re *less* impacted, precisely because of the training meaning that they’re relying less on the d20.

Also, you’re forgetting that because of their training, they can auto-pass certain tasks that an untrained character has a chance of failure at. If you’ve got a +9 modifier, it doesn’t matter how rushed or under pressure you are, you can auto-succeed on DC 5 and 10 checks.

>>89554361
Yes they are.

>>89554692
Explain, then. How does having a +5 modifier make you more affected by a d20 roll's result than a +0 modifier?

>>89554796
Passive checks.

>>89540670
>Why would you use a die that breaks flat probability down into easy to understand 5% chunks? The world may never know
The d20 is exceptionally handy from a game designer perspective. If you want something to have a 65% chance of happening you make it so the target number is 8 or higher. If you decide you'd rather a 60% chance you change it to a 9 or higher. That's why use it.
As far as dislike... It's a fucking die. They don't hate it, they hate D&D, a system which has made the d20 synonymous with itself.

>>89554846
What if you don't want things to happen in 5% increments?

>>89554811
The passive score for someone with +5 is 15, and for +0 is 10. The person with the 15 is still more likely to notice things then the person with a 10. On account of it being five points higher.

So either explain harder, or accept that you're a moron.

>>89554846
You do, if you're working off flat percentile. Fewer die sides = less control. Most people do not give a shit about anything smaller (I could not be bothered to worry about the difference between a 10% chance and an 11% chance on a roll)

>>89555025
Then use a different system and recognize that this is a matter of personal subjective taste, not general objective superiority.

>>89555080
I mean, he is retarded, but a lot of people who don't understand math think they have a problem with the dice in D&D when what they really have a problem with is bounded accuracy.

>>89555103
meant for >>89555025

>>89555025
Then you roll d100 and you have a Coke and a smile and shut the fuck up.

>>89555114
Of course it isn't about general objective superiority. But one extremely insistent anon in this thread can't seem to wrap his fucking tiny brain around the fact that no, a d20 system and a 3d6 system don't basically, essentially do the same thing but with different resolution mechanics and DCs. One provides much more granularity and variance in probabilities, and that is a perfectly valid reason to prefer it.

>>89555103
>>89555124
There's not much difference between having a 10% chance vs. an 11% chance. There is a fucking massive difference between having a 95% chance vs. a 99.5% chance.

>>89555268
Who taught you math? Go kick them in the nads, whoever they were.

>>89555260
The variance in probabilities for any one result still doesn’t matter in practice when your goal is a range, though. A 30% chance of success is a 30% chance of success no matter what dice you’re rolling and what modifiers you’re adding.

The 3d6 naturally allows for probabilities of less than 5% for extremely low or high numbers, fine. But in practice in actual games how often are you really setting difficulty targets with a less than 5% chance of success?

>>89555352
any solution that involves multiple dice will create a bell curve. There's a reason I said 'flat probability'.

>>89548831
>Yeah, I thought it was you.
Hate to break it to you anon, but the "you" you're talking about is dozens of people. Probably only half of /tg/ is as mathematically incompetent and illiterate as you are.
That leaves quite a lot of people who actually understand resolution mechanics and probabilities and who don't cry tears of blood over meaningless polyhedral shapes like you.

>>89552679
>Because when you're extremely good at something, you don't have a 5% chance of failure at it, you have a fucking 0.05% chance or some shit.
You just wouldn't call for a roll in that situation you hilarious idiot. Go play a game even once in your life to see how it's done. Pro-tip, it's not how you describe it:
>I lift my foot to take a step
>roll for it
>I extend my foot as part of a step
>roll for it
>I lower my foot as part of the step
>roll for it

>>89542018
>the range is too wide. Gradiality doesn't worth swinginess;
there is no such thing as a 'too wide range' unless the problem is 'too much gradiation'. A 60% chance is a 60% chance, it doesn't matter if you're rolling a d10, d20, d30 or d100, rolling under or rolling over, there's a number on it that a roll will beat 60% of the time. 'Swinginess' doesn't exist, if you're mad that you rolled a 2 on a check you had a +11 on and 13 was lower than the DC that doesn't mean 'the range is too big'- you're missing the point of what the die roll represents.

>>89555260
>One provides much more granularity and variance in probabilities
It doesn't though. 50% is 50%. 75% is 75%. You can adjust the resolution system around the dice any way you want. Just look at Numenera as an example.

If you think everyone else in the world has a tiny brain because they all can't stop disagreeing with you, I have some very troubling news for you on where the tiny brain resides.

>>89542191
Seeing as few to no games actually include then by raw you don't need to include them. "ReAlPlAy" podcasters aren't standing behind you making you act like them.

File: poor bait.png (115 KB, 625x626)

115 KB PNG

ITT a retard is mad that he can fail rolls even if his character is good at them

>>89555555

>>89555511
Low IQ bullshit.

>>89555507
6e is doing that though.

>>89555537
they aren't, unless they changed their mind again. I'm pretty sure they got it hammered into them that a 5% chance to succeed at convincing the king to let you bang the queen while he watches is probably not a good thing, especially if the players know it will always work.

>>89555537
Even if 6E does, nobody is forcing you to play it.

>>89555080
What else would you call it when the lesser character goes from having no chance of success to having one while the greater character has no chance of failure to having one?

>>89552586
>To…what?
To get a distribution where extreme results are less likely than average ones so using degrees of success isn't ridiculous. This was never anything remotely to do with calculating odds, which you were told several times. Calculating odds isn't particularly useful beyond a very rough idea of "X is (much) more likely than Y".
>>89554846
That sounds like a really terrible way to design a game.
Don't tell me what I do or don't like, faggot. I don't hate D&D, because I've never played it. I do hate d20s, because I've rolled them and I dislike it.

>>89555598
>To get a distribution where extreme results are less likely than average ones
Any system, just use modifiers or clarify what extreme results are? You could do that with a coin flip you retarded mong.
>I don't hate D&D
An amusing but very uninteresting lie.

>>89555533
Yup, that's also why he spams the thread. Being low IQ was implied by the "retard"

>>89555822
>Any system, just use modifiers or clarify what extreme results are?
What the fuck are you talking about?
I'm not asking you if it's possible to make such a system. I'm telling you that's one of the things that using multiple dice lets you do easily without resulting to extra steps like, "if you pass by X or more then ... Y or more then ..." and so on.

>>89555851
>I'm telling you that's one of the things that using multiple dice lets you do easily without resulting to extra steps
Just to be clear, I am not the same poster you are acting unhinged at.

>>89551255
>With 3d6 you subtract the number you rolled from the target
It lets you.. subtract your result from something? You're not being complete and coherent.

Are you saying your degree of success is the target minus your 3d6? What do you think that means?

>>89555851
And to add on even more:

How are you determining that target number in the first place without any statistics, probability, or odds? Since you insist they don't matter and are subtracting that step.

>>89555969
Sorry if not being able to follow random non-sequiturs makes me "unhinged".
>subtract your result from something
From the target number. The number that determines whether you pass or fail the skill roll. Why are you pretending not to know this?
>>89555995
I did not say you shouldn't ever use any probability theory, I said you don't actually need to calculate precise probabilities. Players only need a very rough idea of, "X is likely to work, Y is even more likely, Z is less likely but still better than 50:50" and so on. Which is easy in all but the most convoluted of systems.
A referee barely even needs to know that: they need a rough idea of how to map the difficulty of a task to a target number, which could be done with a table and no understanding of probability whatever.

>>89540592
A lot of people are autistic about MUH BELL CURVE without having any good reasoning as to why, and they're projecting their issues with D&D onto flat probability rolls.

>>89552773
>I just don’t see the benefit to this when you combine it with a binary pass/fail system that does not care about a very high or very low number being rolled
1) Some systems do care about this, having success with a margin of 5+ mean something is useful
2) The important thing isn't the actual roll but the way you calculate it, how it interacts with modifiers/stats/difficulty. The reason you use dice isn't to resolve something with a known 65% chance, the reason you use dice is to resolve "does my character with these skills, stats, and modifiers succeed in task with this task".
If all you cared about was resolving percentage rolls then obviously d100 would be far superior to d20
>The fact that small modifiers are more impactful also doesn’t seem to matter because ultimately all a modifier is, is a percent increase to your chance of success or failure, and the d20 system’s modifiers do the exact same thing.
It's not really that small modifiers are more impactful, it's that small modifiers are more impactful specifically when you're near the 50% mark. A +1 can turn a 50% into a 62% but if you're already at 98% or 0.5% it's not going to do as much for you.
>Although something I didn’t touch on, the fact that modifiers in 3d6 also increase odds of success in a difficult to calculate way.
Again, you're acting like you need to get percentages for certain things when you really don't. You're also doing this calculation wrong (in both 3d6 and d20), what you care about in these cases is not the delta in probability but the change in EV. in D20 every +5% bonus is not a +5% bonus to your expected value, going from 5% to 10% is giving you 200% EV while going from 50% to 55% is only giving you 110% EV.
In the other hand in 3d6 at the extremes you're getting 400% EV at the low end and still 125% in the middle.

>>89555352
>A 30% chance of success is a 30% chance of success no matter what dice you’re rolling and what modifiers you’re adding.
If all we cared about was the 30% chance the dice wouldn't matter, the point is how we arrive at that 30% chance.
Difficulty targets aren't percentage based, they're target numbers. With a curve that number is valid for a larger range of skills rather than quickly becoming either impossible to hit or impossible to fail.

>>89555995
Looking at tables in the source books to see what's appropriate for the challenge, with appropriate modifiers?

>>89549541
How is it objectively superior though?

>>89555598
> Calculating odds isn't particularly useful beyond a very rough idea of "X is (much) more likely than Y".
Which is the entire point of DCs and rolling dice in the first place, you mongoloid. The reason why we’re rolling dice is because we want there to be a chance of failure. The odds represent how big a chance of failure that is. And when you’re playing - scratch that, when you’re RUNNING a game where player characters have different levels of skill at different tasks and can attempt tasks with different levels of difficulty, you want a resolution mechanic that is quick and easy to calculate the probabilities on so that you know, when you’re setting difficulties, if what you’re setting is too likely or too unlikely to happen.

Hence, while 1d20 isn’t really any mathematically better than 3d6 because 30% is 30% no matter how you get there, it IS better in that the straightforward 5% increments make it absurdly easy to calculate odds.

>>89555851
> I'm telling you that's one of the things that using multiple dice lets you do easil
So what, characters in your sacred 3d6 system never have modifiers to their rolls? All checks are just straight 3d6s because everyone is as competent as everyone else at everything that they try?

>>89556588
> what you care about in these cases is not the delta in probability but the change in EV
I don’t play Pokémon much these days and I never played it enough to care about EVs even when I did.

More seriously while I get what you’re saying, I don’t get why you’re saying it. No, the only thing that matters is the raw probability of passing or failing a given check. Even in systems where you can achieve great successes or suffer catastrophic failures, the first thing a player wants to know is “did I achieve what I wanted to do?” Since major effects are unlikely, in practice players still simply try to and count on achieving basic success.

>>89555598
> With a curve that number is valid for a larger range of skills
Yeah you’ve still lost me here. I’ve subtracted number rolled from target number and gotten a 2. So what does that 2 DO? “2” isn’t a distribution. Is that my result or something? You seem to be presuming knowledge of a system that I clearly don’t have because I like to play actually good games.

>>89557299
>Which is the entire point of DCs
What? None of that rant is remotely relevant to what I said. CALCULATING EXACT ODDS IS NOT IMPORTANT.
>never have modifiers to their rolls
They do, which is another reason - which we've already covered several times - why a non-uniform distribution is better: because a straight percentage point difference does not mean the same thing regardless of what it's added to.
>sacred 3d6
I generally prefer 2d6, but that's not remotely the point.
>>89557321
What the fuck are you even quoting?
>So what does that 2 DO?
The amount you beat the target number by determines the degree or quality of your success. Did you never take a fucking test in school? You might just barely pass a test, or you might ace it.
I'm not presuming knowledge of any system. I'm not talking about any particular system, and I don't know why the fuck you'd think I was.

>>89556683
No table can possibly have every single circumstance. You’re going to need to make up shit on the fly, something you’d know if you actually ran games on the regular.

So again, how are you determining that target number in the first place without any statistics, probability, or odds?

>>89557359
Your table isn't supposed to list every possible thing that could ever happen, it gives you a rough idea of how difficulty of a task corresponds to target numbers.
How much would calculating probabilities even help you? Do you know the odds of every possible real world scenario to compare them to?

>>89557299
>the first thing a player wants to know is “did I achieve what I wanted to do?”
No, it's "how well did I do?"

>>89557351
> CALCULATING EXACT ODDS IS NOT IMPORTANT.
It is in fact the only thing that matters, as otherwise we’d be playing a diceless, random-less system.

> why a non-uniform distribution is better:
No, it isn’t. It, at its best, can ape what the d20 can do, but without the intuitive math behind it that rhe d20 enjoys.

>What the fuck are you even quoting?
Theeee previous posts, you mongoloid? Some Anon, I assume you, said “with 3d6 you subtract the number rolled from the target”.

Okay, so I’ve done that. I’ve subtracted a rolled 12 from my target of 14. That leaves 2. Now what is that 2 actually DOING? When I asked that some Anon, again I assume you, responded “ To get a distribution where extreme results are less likely than average ones so using degrees of success isn't ridiculous.”

“2” is not a distribution. If you meant something else, or thought you were responding to some other point, then you need remedial English as well as remedial math, and the confusion here is on you, not me.

>>89557391
YOU DON'T NEED TO CALCULATE ODDS TO ROLL DICE HOW FUCKING RETARDED ARE YOU HOLY FUCKING SHIT.

>>89557381
No, but I know the odds of every single possible Difficulty Class for every single possible roll of a d20 with every single possible modifier, thanks to how easy it is to calculate odds. Sometimes you’re not trying to simulate the real world, you’re directly trying to challenge a specific character’s specific skill level. Again, something you’d know if you played games on the regular.

>>89557384
No, it’s “did I succeed?” THEN it’s “how well did I do?”.

>>89557407
But you do need to calculate odds to know what the result of that roll will actually achieve, you innumerate bastard.

>>89557419
Asking how well you did already answers whether you succeeded or not.

>>89557434
Yeah, it’s two questions in one, but “did I succeed?” Is the most important and first question. If I throw a punch, sure, I’d like to knock the guy out in one blow, but the first thing I want to know is if I managed to hit at all. If I’m climbing a wall, yeah I’d like to do it fast, but the first thing I want to do is make any kind of progress at all. If I’m trying to make a sword, sure I’d like for forge Masamune, but first and foremost I’d like it to not break while I’m trying to forge it.

THEN I care about the degree of my success, AFTER that success has been established.

>>89557419
>No, but I know the odds of every single possible Difficulty Class for every single possible roll of a d20 with every single possible modifier, thanks to how easy it is to calculate odds. Sometimes you’re not trying to simulate the real world, you’re directly trying to challenge a specific character’s specific skill level. Again, something you’d know if you played games on the regular.
1) knowing what's challenging for a specific skill level is easy with 3d6 as well
2) adjusting the DC of a task based on the character instead of being consistent with how they're handled is fucking gay. You might as well just give everyone two coins and make them fail if they get double tails.
>>89557428
Why? When are you using the specific percentage odds to mean something in your game?

I like 3d6 bell curves because it clusters around a few average numbers with variance without completely removing luck from the scenario. And even within the usual range of 9 to 12 there's still variance. The actual math of a system will be done beforehand and I should never have to "calculate the odds" in the middle of a game like some of you psychopaths seem to be doing.

>>89557489
> knowing what's challenging for a specific skill level is easy with 3d6 as well
But not as easy as it is for a d20.

> adjusting the DC of a task based on the character instead of being consistent with how they're handled is fucking gay.
Oh? As part of describing a saloon you offhandedly say that there’s a game of poker going on because it feels in-character for a saloon to have a card game going on. A player, unexpectedly, decides he wants to join in. What, objectively, is the difficulty you should be setting?

Are you really claiming that it’s bad DMing to, instead of assigning it an arbitrary “DC 20”, deciding that you want to have the game be challenging for the player and so set the DCs based on their skill?

>>89557489
Well for example when a player sits down to a poker game apropos nothing.

>>89557567
> and I should never have to "calculate the odds" in the middle of a game like some of you psychopaths seem to be doing.
Of course you should, unless you’re running the most railroady game ever. Even if your players are retards - and they are, they always are, don’t worry mine are too - but even if they’re retards, there’s more of them then there are you. They can and will come up with unexpected plans or try and do unexpected things that’ll force you to come up with stuff on the fly. And sometimes you can just get by with arbitrary difficulties - “oh this is moderate, DC 15, this is hard, DC 20, this is mind-numbingly easy, DC 5, WHAT THE HELL HOW DID YOU STILL FAIL” - but there are times when you’ll want to challenge a specific character’s skill level, and so being able to quickly calculate odds is useful.

Really all this thread is showing me is how rarely people who don’t play D&D actually play games. There really is a benefit to being the World’s Most Popular Roleplaying Game(tm): you actually get practical experience.

>>89557605
Decide what skill the NPCs at the poker table should have and roll based on that, don't say "this should be about 65%", you're basically ignoring all the character development.
And, again, _you do not need to know the exact numbers_, in either system.
If you want to challenge you trivially know what numbers are hard to hit in both systems, acting like having to memorize what number 85% corresponds to is difficult just makes you look like a retard

>>89557659
how is deciding the odds and calculating a DC based on that instead of being consistent with the rules not railroady? I'm sure it's very fun deciding to skill into lockpicking and having all the locks in the world scale to meet that level

>>89557693
> Decide what skill the NPCs at the poker table should have
And why is it wrong to say, “they should have the skill needed to challenge this player”? As long as they consistently have that skill going forward if it comes up again, I don’t see the issue.

>>89557605
>Oh? As part of describing a saloon you offhandedly say that there’s a game of poker going on because it feels in-character for a saloon to have a card game going on. A player, unexpectedly, decides he wants to join in.
Obviously you resolve it by pausing the session, have all players who want to play poker join in, while the rest either eats dinner or tries to sabotage your minigame.

>>89557717
What? If you're starting from percentage odds and calculating backwards from there you're not saying "they should have the skills"
The non railroady thing is to just roll for random NPC skills on parlor games (unless you already know them for some reason) and let the players have the consequences of either being up to those the challenge or not.

>>89557605
>What, objectively, is the difficulty you should be setting?
Depends on how good the other players are. Roll Gambling and compare degrees of success. Done.

File: 17477739-5f45-4f86-ad40-5(...).jpg (98 KB, 690x862)

98 KB JPG

>>89557708
...If you've never GMed a game of anything at all in your life (and yes, you really did just out yourself) you shouldn't talk about dice mechanics at fucking all.
Just fuck off back to GURPSGEN already and preen about how you pwned the libs, I mean D&Ders.

>>89557708
You’ll note that my example one post up was for NPCs, not objects. I’ll completely agree that, for example, locks in a given area should usually be based around the wealth of the area (wealthy people can afford better locks) and not player skill, but if a player is sitting down to a random poker table that I hadn’t otherwise prepared for, it makes sense to me to scale the NPCs there to the player so as to provide a meaningful challenge, else otherwise I’m just handing the player free money. As long as, once I’ve established the NPC’s skill level, they STAY at that level going forward (unless there’s a good reason for them to have gotten better or worse), I don’t see the issue.

Now you can ask something like, “but the PC is a 10th level rogue, why does this podunk town have a poker player that can challenge him?”, but to that I’d respond, “why does this podunk town have a 10th level rogue in it?” By simply being in the town we’ve established that even a tiny hamlet can have at least one high-level character good at gambling in it; why not two? I’ve built numerous NPCs backwards like this over the years, it’s a great way to introduce rivals, contacts, allies, etc.

>>89557605
D2 is easier than a D20. So we should flip coins to determine pass/fail if we just want the simplest option.

>>89557717
You wouldn't be looking at the dice system then you'd be looking at the players skills. A challenging opponent would be somewhere near the players skill level.

>>89557766
No, because a d2 isn’t granular enough, despite its obvious benefits in predictability, especially as modifiers are added.

The d20 is the perfect combination of granularity (twenty possible results with meaningful distinctions between results), predictability (all results have equal odds), and simplicity (everything increases by a straightforward 5%).

File: disdainforplebs.png (137 KB, 400x400)

137 KB PNG

>>89557659
>Of course you should, unless you’re running the most railroady game ever.
Other way around. I don't need to "calculate the odds" *unless* I'm running a railroady game, because it is only in a storyshitter-game that I'd not just present the scenarios and situations as they come, regardless of the chances the players have to succeed or not.

>>89557763
the amount of money a master gambler can win is still limited to what the other players in the game are willing to give up, you're not breaking the economy and you're rewarding the player for actually selecting and using his skills.
If you're going to make NPCs to match the player then the extra point he put into it just didn't matter, it's the same as with the locks.
And in a decent game being a high level rogue is not the only way to be good at gambling, you'll occasionally find just normal people who are good at it.

>>89557659
If your universe scales around the players, you're running the most railroady shit imaginable. The fact that you try to project that onto others is fucking insane.
>Verification not required.

>>89557786
You're a fucking idiot for many of the reasons previously stated, but I still wanted to just poke my head into the thread and tell you that a d100 is superior by every stated metric. That is all.

>>89557102
>Doesn't know why curve is good
Take power away from random number generator and put more emphasis on PC skills and abilities. Plus greatly reduces amount of "muh natty 20/1".

>>89557842
I mean I do agree, but I feel that the d100 is TOO granular. There isn't enough of a meaningful distinction between 50% and 51% odds to make it worthwhile.

>>89557798
>because it is only in a storyshitter-game that I'd not just present the scenarios and situations as they come
That's literally what was described, you illiterate mong. Your players walk into a saloon, you describe the contents of the saloon and offhandedly mention a poker game going on in the corner because it feels like something that would be happening in a saloon, and apropos nothing a player has decided that his character wants to join in on the game.

At that point you can choose to set an arbitrary DC, or you could choose to customize the difficult to the player's skill level so as to make the situation challenging for them. Neither is the correct move, it depends on what you want to happen with this unplanned-for poker game, but if you decide on the latter it sure is nice to be able to look at the player's +9 to gambling checks and know that if you want to challenge them (65% pass rate), you set a DC 17, isn't it?

>>89557777
Near, but how near?
D&D 5e has an answer for that, it defines "challenging" as a 65% pass rate. In other words you succeed notably more often then you fail, but failure is still a real possibility. Much higher odds and you might as well not even bother rolling because success is to certain; much lower, and players start to go out of their way to try and avoid rolling at all because failure is too strong a possibility.

A 65% pass rate is easy to calculate in 5e: it's the PC's skill bonus minus 2. But that's only something you'd know if you knew that the 65% pass rate was the goal and, more importantly, WHY it was the goal. Thus: knowing the odds is useful.

Especially because, since you know why a 65% pass rate is desirable, you now have a better idea about when and where you should deviate from it, and by how much depending on why you're deviating from it.

>>89558041
By scaling the skill of the other gamblers based on player skill you're doing the exact same shit as scaling locks, how can you think this is any different? The actual player skill and how they've but their character is meaningless.
And calculating something like 65% for "challenging" in 3d6 is still trivial, it's just target number 11.

>>89557823
But we're not talking about the universe. We're talking about one poker table in one town at one point in the campaign. And technically we're not even talking about that, that was just an example; what we're talking about is circumstances under which it is both believable and desirable to adjust a challenge according to player skill. Obviously it's not EVERY circumstance, but equally obviously it's not no circumstance, either.

>>89557810
>If you're going to make NPCs to match the player then the extra point he put into it just didn't matter
By that logic there's no point in having NPCs ever scale at all. An ancient red dragon should be as difficult as a kobold, because if the monsters they're facing are growing in threat as the players grow in level then there's no reason for them to gain levels.

Like I said at least twice before, I don't see the issue of setting NPC skill based on player skill as long as that skill level remains consistent going forward. If the player shows up, gambles, and loses, but then wanders off and gains 10 ranks in gambling and comes back to the same gambler who hasn't done anything in particular to get better, then the gambler shouldn't be better and thus the player should have an easier time beating them, possibly to the point where their victory is certain.

>And in a decent game being a high level rogue is not the only way to be good at gambling
It's not in D&D either. Kind of missing the forest for the trees here, buddy.

>>89558112
>scaling but only once.
Skyrim, the TTRPG

>>89555448
>You just wouldn't call for a roll in that situation you hilarious idiot.
Keep repeating this retarded shit, you can do that if you want but I don't and never will want to have to know all of my players' proficiencies at all times, individually gauging every action they want to take against their level of skill and saying yes anon, you can just do that automatically, but that anon needs to roll for it. I will set DCs as is my actual responsibility, and whether that DC is possible to hit or not is up to the players. Just blow your condescending oversimplification out your ass.

>>89557999
>and put more emphasis on PC skills and abilities
Not particularly. One or two points in a modifier counts for more in 3d6 instead of 1d20, but you also tend to get much, much smaller modifiers, making the whole thing broadly a wash.

For that matter if you want power taken away from the random number generator, let me introduce you to 3rd Edition D&D, where it was trivially easy, using nothing but basic core options, to make characters that could trivialize the entire skill system by largely negating the impact of any d20 roll they'd make due to having high bonuses.

Like if I wanted to make an expert weaponsmith, then at 1st level I could be a dwarf (+2) with 4 ranks in Craft (weaponsmithing) (+4) and skill focus in Craft (weaponsmithing) (+3) and an 18 in Intelligence (+4) and get some masterwork tools (+2). That's a modifier of +15. On my worst day I can churn out any simple (DC 12) or martial (DC 15) melee or thrown weapon or a crossbow of any kind with literally no effort. With only the slightest of effort (rolling a 3 or better to hit the DC 18) I can make exotic weapons. If I take 10 then I can make exotic weapons easily, or I can make masterwork crossbows or simple or material melee or thrown weapons. The d20 roll is only relevant if I'm trying to make an exotic, masterwork weapons - literally some of the hardest things to do. And even then I succeed most of the time.

The problem you have isn't with the d20, it's with the comparatively smaller bonuses of 5e compared to 3e. But the thing is that this is your subjective opinion, not an objective fact. The smaller bonuses of 5e are a result of extensive playtesting revealing that most players, or at least most players who took part in the playtest, actually prefer smaller bonuses where the d20's roll isn't rendered moot.

>>89555995
How are you so retarded that you need this explained to you this many times?

>>89558216
>I will set DCs as is my actual responsibility, and whether that DC is possible to hit or not is up to the players

"Why can my players never open locks? The DC is only 30! They have a +10 bonus, so I know it's possible! I didn't fail, they failed! They all failed! Just like they did before!"

>>89558258
"This custom monster I've made only has an AC of 21! The players are 5th level, they should be hitting that easily! Why are my players so incompetent?!"

>>89557299
>you want a resolution mechanic that is quick and easy to calculate the probabilities on so that you know, when you’re setting difficulties, if what you’re setting is too likely or too unlikely to happen.
If you are back-engineering your DCs this way, you are doing it wrong. All you need to know is the relative difficulty of the task. You absolutely do not need to know the literal exact probability that the specific player will be able to hit the DC.

>>89558041
>choose to customize the difficult to the player's skill level so as to make the situation challenging for them
So literally the opposite of what you state, then? The idea that the skill of the other players at the poker table depends on the skill of one of the player characters is just pure storyshitting, you absolute fucking idiot. The fact that you need to dishonestly cut out the most important qualifier when quoting says it all.
>regardless of the chances the players have to succeed or not.
Literally the only reason you've given for being able to tell probability on-the-fly is so that you can railroad and storyshit, presumably so you don't have to straight-up take control of the characters to make them do what you want.

>>89558041
>d100 is TOO granular
>meaningful distinction between 50% and 51%
So don't go from 50% to 51%, go from 50% to 55% or from 50% to 56% or anything beyond or in-between. With a d100 you have that option, with a d20 you simply don't. There is no scenario where d100 is not superior to d20 in by *every* stated metric. *Objectively*.

d20fags should be fucking killed

>>89558267
"Argh! They all made their saves AGAIN? How is this possible, I set the DC to 13! The chart says that DC 13 is appropriate for 3rd level characters! How can the cleric, monk, paladin, and wizard ALL keep succeeding on their Wisdom saving throws?!"

>>89558237
>The smaller bonuses of 5e are a result of extensive playtesting revealing that most players, or at least most players who took part in the playtest, actually prefer smaller bonuses where the d20's roll isn't rendered moot.
this is part of that point of 3d6, you can have small bonuses that feel meaningful while at the same time not immediately running out of runway and having automatic success.
A +1 with 3d6 at the 50% mark is giving you a significant change but has diminishing returns for something you're already at 90% for.

>>89558258 >>89558267 >>89558288

"Where did all my players go?"

>>89558237
>Fix the d20 rng to pc skill imbalance by just throwing bigger and bigger modifiers
Horrible take anon. Please go study some actual game design and not Neutronium Golem shenanigans.

>>89557359
Have you never opened up a GM's guide? You've never seen the tables that list DCs, show examples of tasks of the given difficulty, etc? They're extremely common, so it would surprise me if you've run a game and never seen one. Those tables never show actual probabilities of anything, btw.

>>89557798
This

>>89558041
>but if you decide on the latter it sure is nice to be able to look at the player's +9 to gambling checks and know that if you want to challenge them (65% pass rate), you set a DC 17, isn't it?
No, that's fucking retarded

>>89558291
Yeah, but why should I go through the trouble of memorizing what those diminishing returns are when I can also just know that a +1 bonus means a flat 5% increase in probability on a d20 roll?

More to the point, why do I even want diminishing returns? If you've gotten better at something then surely you should be actually, notably BETTER at it, right? I don't want to improve my chances of making an average roll, I want to improve my chances of making exceptional ones.

>>89558280
>So don't go from 50% to 51%, go from 50% to 55% or from 50% to 56% or anything beyond or in-between.

Generally speaking I find that a 5% increment is perfectly suited for my needs. So I can either have a d100 and modifiers and DCs that are divisible by 5, or I can have a d20 and have simple +1 bonuses. At that point, why not just use a d20?

Also there's no way in Hell I'm rolling an actual physical d100, it's either d% (2d10 with one of the d10s marked as the 10s digit) or the d20 (since I'm set to 5% increments). Rolling and having to read one dice is infinitesimally easier than rolling and reading two dice, but it IS easier.

So, d20 it is.

>>89558280
>The idea that the skill of the other players at the poker table depends on the skill of one of the player characters is just pure storyshitting

No, it's not. Storyshitting would be taking a look at the setting and likely make-up of the saloon and deciding the NPC skills based on that - we're in a podunk town? Then they'll be podunk players. It would be trying to make the events fit the setting as part of some kind of narrative coherence.

This is me deciding that if a player wants to earn some significant amounts of gold (a useful resource) without having to go into combat (thereby risking other resources), then I want him to still have to work for it, and so deciding to ensure that his opponent is challenging. I'm then working backwards from that very gamist, not story, determination in order to build the NPC.

>>89558258
Stop posting you fucking dipshit

>>89558363
It's never retarded to be able to intuitively figure out odds, Anon. You'd have to be retarded to think that.

>>89558370
>I'm then working backwards from that very gamist, not story, determination in order to build the NPC.
Oh, and the whole scenario literally can't be a railroad, since the very basic premise is that the PC has done something that the DM had not planned on, but then rather than contriving to stop him and force him back on track, the DM has indulged him and is trying to come up with a fair way of challenging him, coming up with a new scenario on the fly.

This is the textbook definition of NOT railroading a player. Unless "railroad" has become like "Mary Sue" or "Fascist" or whatever and become a word that has no actual meaning other than "thing I don't like".

So, yeah, upthread I mentioned that someone needs remedial English classes. I'd like to repeat that now.

>>89558258
Yes, that certainly is a ridiculous situation you've imagined. It's also completely disingenuous and thus irrelevant.

>>89558412
You write like an autistic faggot.

>>89558389
It's retarded to base DCs off of the odds of a specific PC to have a %chance to succeed.
I'd argue doing that is even worse than storyshitting.

>>89558419
It's a hyperbolic but still valid example of why your stance is so retarded. If you don't know your players' modifiers, or at least have a rough idea of them, then you can't possibly know what DCs they are capable of passing or failing, and thus run the risk of constantly putting them into situations that they can't actually handle, or which they might trivialize.

I mean at a more basic level, say you put your 1st level players in a dungeon full of kobolds. Kobolds are known for trapmaking and stealth so you decide that the traps are going to be Hard (DC 20) to spot and disarm. But your players, it turns out, all thought that someone else in the group took Perception, and so as a result none of them are actually proficient with it. The highest Perception in the group is +3 from the Cleric's natural Wisdom modifier. They've only got a 15% chance of spotting a trap. And none of them have proficiency in thieves' tools because none of them felt like being Rogues or having backgrounds that would justify it, so they can't disarm them. That's not Hard, that's a virtual death sentence, or at the very least is going to require them to disarm traps by running into them, setting them off, and then hoping that they either make the saves or have the healing spells to recover afterwards. And then the kobolds kill them because they're low on spells, low on healing, and low on hit points

What on paper should have been a fairly easy dungeon for a typical party is instead a meat grinder because you failed to take your players' actual capabilities into account. Even if you don't know the specific modifiers you should generally have an idea of what they can and can't do and make sure to set DCs accordingly. This is a failure on YOUR part, not theirs. While it is not your job to keep them alive, it IS your job to make sure that they're having fun, and it's not fun to be put into a situation of near-certain death because your DM didn't care whether or not you could handle it

>>89558548
As long as you've established that it's a kobold dungeon and they're enthusiastic for traps there's nothing wrong with the dungeon being difficult. They'll either have to take it more carefully or go back and figure out alternative ways of tackling the dungeon.
Again, if you're constantly adjusting based on PC skills you're undermining them choosing them in the first place.

>>89558447
>It's retarded to base DCs off of the odds of a specific PC to have a %chance to succeed.
No, it's not, not in the circumstance I described, and saying otherwise doesn't make it so. There are circumstances where it makes sense to build a challenge off of specific player capabilities, and when those circumstances crop up, especially in an unexpected way, it's good to know the probabilities of the system you're using so as to be able to make challenges on the fly.

>>89558433
I don't care. I'm right, and that's more important. The circumstance I described is literally the opposite of railroading since it's premised on a DM indulging a player who's gone off the rails rather than one who forces the player to remain on them.

>>89558548
>He didn't read the DMG section on balancing DCs vs PC level

>>89558641
You're "indulging the player" by ignoring his skill point allocation.

>>89558548
Yes, the way modules have been written since time immemorial is wrong. My poor players might have a bad time :( because they built a completely fucktarded party, so it's my responsibility to change everything to fit the characters.

The obvious solution, retard, is to have multiple avenues of completing the goal that are all at an appropriate level for the party. If something is locked behind a potentially-missible door, it is not required for completion of the quest. The players will learn pretty fucking quick. If they still can't succeed somehow, they're really fucking stupid. Are you in the business of coddling your retarded players and letting them succeed even when they objectively made poor decisions?

>>89558624
>Again, if you're constantly adjusting based on PC skills
You're not, no on is, and no one suggested that.

For the fourth time now (I've been counting): I have said that there is nothing wrong with setting a challenge based on player skill as long as that specific challenge does not change difficulty later to account for new skill (unless there's a good reason for it to be justified - the gambler was met while on a worldwide gambling tour and now eight months after first meeting him he's gotten better, for example). And it should only be done in circumstances where it would make sense to specifically try and challenge a player, such as a player taking advantage of an opportunity for downtime to personally challenge themselves (i.e., you didn't plan on a poker game, it was only supposed to be set dressing, but the player specifically sat down at the game with the intent to make significant money and so is essentially literally asking for a challenge).

>>89558711
>Again, if you're constantly adjusting based on PC skills
>You're not, no on is, and no one suggested that.
>
>For the fourth time now (I've been counting): I have said that there is nothing wrong with setting a challenge based on player skill

>>89558652
>You're "indulging the player" by ignoring his skill point allocation.
I'm not ignoring it, I'm taking it very much into account to ensure that when a player pretty much literally asks for a challenge, I know how to challenge them.

>>89558650
PC level is not the whole story. Consider a party consisting of a Fighter, a Rogue, a Wizard and a Druid. This is essentially the classic setup, right? Warrior, rogue, mage and priest, should be able to handle anything.

Yeah except since what ordinarily would be a Cleric is instead a Druid, none of them are proficient in Charisma saving throws and all of them have Charisma as a dump stat; even the Rogue has to prioritize Dexterity, Constitution, and Intelligence first (and in that order), making Charisma at most a 12 (assuming Default Array). Meaning that when they're, say, 7th level and so nominally should be more then capable of dealing with DC 15 saving throws, they're instead hyper-vulnerable to effects that target Charisma. An Umber Hulk, nominally a CR 5 creature that should be pretty easy for a group of 7th level adventurers, instead becomes a dire threat due to its Confusing Gaze ability both keying off of Charisma and affecting the whole party at once.

I'm not saying you don't send Umber Hulks against such a party. I'm saying you should do so with the knowledge that it's going to be harder tor them to deal with than you might expect and balance the rest of the enconters in the dungeon accordingly.

Or, y'know, act like this retard, >>89558665, don't know what your players are actually capable of, and then act like they're the morons when they have a difficult time.

>>89558727
Yeah, nice try, except I am NOT SUGGESTING CONSTANT ADJUSTMENT, YOU ILLITERATE MORON. Key word here is "constant", fucktard. You set the DC based on player skill ONLY when it make sense, and once you set it you DO NOT CHANGE IT, unless it makes very good sense to do so.

>>89558665
And then there's the other end of the spectrum: since you don't know and insist you shouldn't have to be bothered actually knowing your PCs capabilities and chances of success, you end up consistently designing dungeons that are too EASY.

You make your big bad an enchanter and then when the enchanter first meets the PCs, they absolutely roll him without him successfully charming or dominating a single one of them, because the party consists of a monk, a paladin, a cleric and a wizard - all proficient with Wisdom saving throws, two of them having Wisdom as a main stat, and the Paladin boosting his allies' saving throws just by being near them.

Wouldn't have happened if you'd just thrown a Strength save in there, but you didn't know that was necessary because you find the idea of knowing what your players are capable of morally repugnant.

>>89558447
Why do people always hate on 4e? You have to keep DCs just about 50% per the DMG.

>>89558280
>There is no scenario where d100 is not superior to d20 in by *every* stated metric. *Objectively*
If False Precision didn't exist you would be correct. It does exist and humans fighting over increments less than 5% is an impossible waste of time, so the d20 is *Objectively* superior to the d100 by every metric, except
>ability to make useless and incorrect adjustments to the odds.

There's a ridiculous amount of samefagging and stupidity in the thread but >>89556165 is still the dumbest and funniest post in thread. The other retarded trolls need to step up their game.

>>89558943
>Why do people always hate on 4e?
I have my reasons mostly related to what happened to the Forgotten Realms when a literal country was literally-at-the-time dropped on where my character at the time was from (though 5e made me feel a little better when it was retconned that Unther was actually phased out rather than crushed...and when it came back it instead was literally, and definitely literally this time, dropped on top of Tymanther, the country that had fallen onto it), in addition to a lot of other bullshit, but they're not relevant to this thread.

>You have to keep DCs just about 50% per the DMG
I think 4e is calculated at about 65%, actually, though its math was so wonky on release that I don't know if that's the case...or if that was the goal.

>>89558942
Again, so every module with flat DCs ever written is a piece of shit, because to be a good DM you NEED to fiddle with the DCs to meet YOUR understanding of the appropriate level of challenge, right?

>>89558312
They are extremely common. In the 5e DMG it's on page 238. But in that same entry they are only referred to as "typical" DCs, and a specific note is made that you may want to change DCs based on character ability.

Which reminds me,

>>89558665
>Yes, the way modules have been written since time immemorial is wrong.
Every module since time immemorial notes that you'll likely have to adjust it to take into account the abilities of your players' characters. In fact my favorite module, Red Hand of Doom for 3.5, is full of sidebar notes from the designers about various points in the module, why they made some of the decisions they did, and what you should do if your PCs find the circumstance too easy or too hard.

My favorite one of those sidebars concerns a battle that's meant to be impossible to win; the PCs are supposed to help refugees escape and then flee themselves and discusses what you should do if the PCs try and stand and fight (essentially, try and give them every opportunity to change their mind as the battle happens, but if they're determined to fight to the death, let them, then pick up the adventure later with a new group of adventurers). It also notes that if the PCs start actually succeeding, you should up the danger level of the monsters they're facing by drawing on the reserves of the Red Hand Horde (the exact size of the Horde is detailed in the book, it's got literally thousands of monsters and the PCs at this point are only 7th level). If they beat even that then you should probably upgrade the monsters (turn hill giants into fire giants, for example) and adjust challenges throughout the book accordingly, while also containing a wry note that if this is necessary, "you might not be running the adventure for appropriately-leveled characters".

>>89559092
See >>89559149. Almost every module notes that you should, if you feel the need, adjust it to fit your personal style and preferences and to better fit the party that your PCs have made. I don't think I've ever run a module exactly as printed out of the box. I don't know why anyone would, that sounds boring.

>>89540592
d20s are great. They're easy to work with and they have a big enough range that you don't have to be overly stingy with modifiers for fear of smacking your head on the ceiling. Much of the hate against d20s is due to their use in D&D, with some superstitions/misunderstandings about probability and such. For example, people try to compare 3d6 to 1d20 without adjusting for granularity. A +1 on 3d6 is, on average, roughly equivalent to a +2 on 1d20. And if you move in steps twice as large on the d20 (with each shift of 1 point on 3d6 equaling a shift of 2 points on 1d20), your probability of success will never be more than 4.5% different in the two systems (and that's at the extreme end). And if you truly need to have some special critical that happens less than 5% of the time, it's easy enough to have a critical confirmation roll: roll a d10 and if you get a 10, that's a critical. That'll give you give you roughly the same chance as you would of rolling an 18 on 3d6. And sure, it's an extra die roll, but that just means that 5% of the time you'll have an extra die you need to roll (or 10% of the time, if you're doing it for critical failures as well), as opposed to rolling 3 dice 100% of the time, as you do with 3d6.

>>89558943
I generally prefer d20s to d%, but the latter does have some advantages. For any individual roll, I don't think the difference in granularity is worth troubling over. You'd have to roll so many times to statistically determine the difference between a d% score and the closest approximation on 1d20, that I don't think it's worth troubling over. But over the long haul, it can be a different story, and d% lets you increase scores in smaller increments, which is pretty cool. It can allow you to dispense with a secondary tracking system for experience and to simply advance your percentile scores directly.

And there are a number of special case results you can easily read on d%. When you roll 1/10 your target score (an 8 or under if your target is in the 80s). When you roll doubles. When you roll multiples of 5 or of 10. Or some combination of these. You could maybe even play around with being able to switch the tens and ones in some circumstance, or rerolling the ones die alone. Or using a d8 or d12 for the tens die to boost or decrease your chances of success.

So there is a bit more flexibility there. But you're also having to roll two different dice and read them in order, which is easy enough, but still not as quick and easy as reading a single number off a single die.

>>89558641
>No, it's not, not in the circumstance I described
It does because your circumstances are retarded. Garbage in, garbage out. Simple as.

>>89558943
My post had nothing to do with 4e, dumbass.

>>89559484
>It does because your circumstances are retarded.
You think it's retarded to indulge a player who's decided to go off the rails? Choo-choo, there are no brakes on your railroad?

>>89559506
It actually does, but probably not intentionally so I won't hold it against you that you didn't realize it, and you'd have to have actually read the rules for 4e to understand why it does anyway.

>>89559506
Anon, that's the only system that uses a d20 where that's intentionally designed in. Do you ever play games?

>>89559442
>It can allow you to dispense with a secondary tracking system for experience and to simply advance your percentile scores directly.
This has always been one cool way to do d%.

>>89558833
You're using PC skill into account so that you can perfectly cancel it out - the result is that the skill doesn't matter at all.

>>89559262
A +1 in 3d6 shifts more in the middle than the ends, which means that getting extra advantages for things that are close to 50% can make them more reliable while at the same time not hitting "impossible to fail" as quickly.
And degrees of success are useful as a tool, with d20 those can swing wildly while in 3d6 most of your results are going up be pretty reasonable.

>>89559775
>You're using PC skill into account so that you can perfectly cancel it out
If I was cancelling it out I'd be setting a DC that had a 50-50 chance of success or failure, so that all that mattered was the raw dice. By setting a 65% pass/fail rate on a check that the PC has a modifier for I'm ensuring that reason he's got a greater than even chance of winning is precisely because he's invested into a skill.

File: 3d6 v d20 with double-siz(...).jpg (53 KB, 890x603)

53 KB JPG

>>89559868
>A +1 in 3d6 shifts more in the middle than the ends, which means that getting extra advantages for things that are close to 50% can make them more reliable while at the same time not hitting "impossible to fail" as quickly.
While true, it's easy to exaggerate the difference of 3d6 relative to 1d20. See the image and tell me that the difference between the red and blue lines is that profound. The minimum and maximum chance of success is the only place they strongly deviate, and like I said, that's easily addressable with an additional die roll on min/max.

>>89559928
The "base" on that graph is "10 or under", by the way. For both systems that gives a 50% chance of success. Then 11 or under on 3d6 compares to 12 or under on 1d20, and 12 to 14, and 13 to 16...

>>89559868
>with d20 those can swing wildly
No, they don't, because you're trying to roll inside of a range, which can be set to have any percentage chance of occurring that you like.

Like in 3d6, what do you want as ranges of success? 3 is critical failure, 4-8 is failure, 9-12 is success, 13-17 is great success, 18 is critical success?

You can do that with essentially the exact same odds in d20 with:

1 + 1 on a d10 (or 1-2 on a d20): Critical failure
1-5: failure
6-15: success
16-20: great success
20 + 10 on a d10 (or 19-20 on a d20): critical success

The fact that 6-15 is ten points on a d20 while 9-12 is only four points on 3d6 is meaningless. 50% of all rolls on a d20 will fall between 6 and 15. 50% of all rolls on 3d6 will fall between 9 and 12. In neither case do we care about the specific number rolled except when trying to "activate" a critical, and even then the odds of rolling a 1 on a d20 and a 1 on a d10, or a 20 on a d20 and a 10 on a d10, are about the same as the odds of rolling a 3 or an 18 on 3d6.

Even if you want to make specific numbers matter, it's trivially easy. Say you want rolling 9 on 3d6 to be a marginal success? Just have rolling a 6-7 on 1d20 also be a marginal success. The odds are essentially identical.

>>89559891
>he's got a greater than even chance of winning is precisely because he's invested into a skill.
What if he hasn't invested in the skill? Or if he's invested a little bit but not a lot? You're still going to pick the 65% number. That's fucking gay and retarded.

>>89559928
The difference is major on the ends, what do you niggers not understand? 95% and 99.5% are not the same psychologically or actually

>>89559993
This. It's like saying a d20 is more swingy than a d12, because you have an 83% chance to roll a 10 or under on a d12, whereas on a d20 it's a coin flip. But the target numbers mean different things on the different dice, because they have different ranges and different medians.

>>89560028
So the last sentence of the post you responded to (which you apparently decided to ignore) was:
>The minimum and maximum chance of success is the only place they strongly deviate, and like I said, that's easily addressable with an additional die roll on min/max.

That additional die roll thing it's referring to was talked about in the post I made before that one:
>And if you truly need to have some special critical that happens less than 5% of the time, it's easy enough to have a critical confirmation roll: roll a d10 and if you get a 10, that's a critical. That'll give you give you roughly the same chance as you would of rolling an 18 on 3d6. And sure, it's an extra die roll, but that just means that 5% of the time you'll have an extra die you need to roll (or 10% of the time, if you're doing it for critical failures as well), as opposed to rolling 3 dice 100% of the time, as you do with 3d6.

>>89560017
>What if he hasn't invested in the skill?
Then why's he sitting down to gamble? In my experience players generally try to avoid making checks that they don't have positive modifiers in. This is like asking what if the Fighter casts Time Stop; there are certainly situations where it's possible, but it's so niche that I don't think it's worth considering.

>Or if he's invested a little bit but not a lot?
Then his little bit (but not a lot) will be what allows him to make the roll at better than 50-50 odds.

>>89560028
>95% and 99.5% are not the same psychologically or actually
I would personally hesitate no longer at a 95% chance on a dice roll than I wold at a 99.5% chance. The psychological impact isn't there except in your diseased little mind.

>>89559993
>d10 confirmation die
Why the Gygax-brained "use every die type because it exists" design, instead of rolling the d20 again for a confirmation roll, 1-2 for critical failure and 19-20 for critical success?

>>89560121
Because rolling a d10 to confirm was suggested by a previous Anon, but you'll note that in my post I do allow for using a 20 in exactly the manner you describe.

Maybe in the future we try actually reading posts before responding to them?

>>89559891
Setting it at 65% is just as much cancelling it out as 50%, the actual skill doesn't matter either way

>>89557470
"Did I succeed" is answered by 3d6 roll under immediately without any need for calculation or speculation, it's going to be instantly and inherently apparent if your Hiking-14 character succeeded at a roll especially since normal people play GURPS with a VTT like Foundry, that allows you to simply click the relevant button on your sheet to make an attempt at something. I'm really quite sick of you and people like you, and I hope you are struck by a terrible misfortune soon.

>>89560093
>I would personally hesitate no longer at a 95% chance on a dice roll than I wold at a 99.5% chance. The psychological impact isn't there except in your diseased little mind.
One of them is ten times as likely to fail, you are a retard.

>>89560320
You're forgetting that the player usually doesn't know the DC that they're rolling for. From the player-side, they see that they've rolled a dice, added a modifier, and then gotten a result. The 65% pass chance on the DM side is to ensure that there's a good chance of success while also having a real chance of failure, on a spontaneous task that the DM did not initially anticipate having to deal with.

You could also just arbitrarily choose a system-specific "moderate" or "hard" DC or whatever, but since a good DM knows their players' skills that's not really any different. In 5e a DC 20 is considered "hard". If I know my player has a +7 to gambling checks then that's not really any different from just deciding that he's gonna have a 40% chance of success.

>>89560352
>especially since normal people play GURPS
I don't think the phrase "normal people play GURPS" has ever been a true one, no matter the context.

>it's going to be instantly and inherently apparent if your Hiking-14 character succeeded at a roll
Doesn't a roll-under mechanic necessarily limit how much your character can actually mechanically improve at a thing? For that matter, doesn't it essentially limit the range in which they can improve without adding in some kind of gimmick, like rolling 4d6 drop 1 or something?

I mean this seems like it's destined to create characters who are broadly and increasingly competent at mundane tasks but who eventually peak in how much they can improve. They can never do truly extraordinary things.

Also, not sure that I like players always knowing automatically if they've succeeded or failed on a roll. One of the benefits of the d20 System is that a player could get a roll result of 24 and still not know if that's actually successful.

>>89560378
Only in comparison to the other. But a 95% chance still succeeds nineteen times in twenty. I like those odds.

>>89560460
Okay, you genuinely have no idea how GURPS works, got it. Go away Mr. /tg/'s Worst Troll.

>>89560516
>Okay, you genuinely have no idea how GURPS works

I mean I also don't know how VTNL or Iron Claw works, this should be less surprising to you then it is.

>>89560545
Than* you fucking mudskin ESL nigger. Just fuck off back to your favela now, real humans are discussing how shit your favorite jewish pedophile game is.

>>89560460
Just because the player doesn't realize you're basically ignoring their skill level doesn't mean you're not doing it
fuck you

How are you faggots this autistic.

>>89560352
>it's going to be instantly and inherently apparent if your Hiking-14 character succeeded at a roll
Okay, so how much did I succeed by? Are you finally going to tell me what that 2 means when I subtracted my roll result of 12 from the difficulty of 14 like you told me to upthread? Because I've only asked you like three times now and you're the one who brought it up in the first place.

>>89560582
Ah. We've reached the gesticulating helplessly and ad hominem part of the conversation. Please, by all means, continue.

>>89560591
Yes, but I'm a practical sort of person, so I feel the important question is "why does it matter", alongside "what is the practical difference between doing that, and choosing a system-recommended DC, when either way I'll already know the PC's pass/fail rate"?

I set the odds of passing to 65% by using ([modifier - 2] + 10) as the DC, I set the odds of passing at 40% by deciding it's a Hard task and the book tells me that's DC 20 (or it's Moderate and thus DC 15 at 65%, or Easy and thus DC 10 at 90%, etc), either way I'm still just straight-up determining the PC's chance of passing.

>>89560636
The second way you're not invalidating player build choices
Seriously, fuck you, I don't think you understand the point of using rules instead of rolling arbitrary dice for everything

>>89560606
I'm not the person you were previously talking to you dumb animal, this isn't reddit or RPGNet, you are talking to an anonymous mass of people.

And to answer the question of how the elementary fucking core mechanic of GURPS works, since you developmentally stunted children have never spent a single second considering anything not printed by WotC, you roll 3d6 and compare it to your modified skill rating. A Hiking-14 character who rolled a 12 succeeds by a margin of 2. This margin may or may not have any meaning depending on what rules the GM is using and how the character is built, but it will always matter in a Quick Contest or any other situation where rolls are opposed. You increase skills in GURPS past 16, and past even 18, to be able to do increasingly difficult things more reliably. An archer with Bow-16 can reliably hit you, an archer with Bow-25 can reliably hit you in the eye.

>>89555352
>Who taught you math? Go kick them in the nads, whoever they were.
>
>The variance in probabilities for any one result still doesn’t matter in practice when your goal is a range, though. A 30% chance of success is a 30% chance of success no matter what dice you’re rolling and what modifiers you’re adding.
>
>The 3d6 naturally allows for probabilities of less than 5% for extremely low or high numbers, fine.


> But in practice in actual games how often are you really setting difficulty targets with a less than 5% chance of success?
We're talking GURPS, right? The player arguably sets the DCs by what they attempt, unless the GM has improvised modifiers they want to add.

But also, the DM could tell you what the total task modifier is if you dont know the rules, sure, but he's not going to memorize all the skill levels on your sheet.

As for the flat % for a single roll, those rolls dont exist in a vacuum. They exist in relation to the other rolls and bonuses and rules, amd your bonuses exist in relation to the other player characters and the npcs. Thats where the game part cones from.

>>89548707
>65%
What makes you think there will be a single target success rate, rather than a predefined list of task modifier examples which can be stacked and used as benchmarks for new situations as needed?

Similarly, what makes you think all the PCs will have the same bonuses?

One PC might have the expertise to auto succeed at some tasks that another PC can't even attempt.

The game is often a lot more than a single die roll in a vacuum, often difficulties will be based on explicit predefined modifiers, and character design will determine bonus allocation across a large potential range. Not everyone is looking for ICRPG.

>>89560636
>I set the odds
I very much prefer those to be benchmarked out explicitly in the book, and when theres something we can't just look up the penalty or DC for we use the existing benchmark options to place it. Often I will set such a thing by player consensus asking them if they agree its around the same difficulty and some other benchmark on the list. If I'm just deciding the odds on a whim, I'm being a dbag and should stop.

>>89560746
> And to answer the question of how the elementary fucking core mechanic of GURPS works
I think you vastly overestimate how many people are actually familiar with your game’s system, even on this board. GURPS is and has always been a meme first and a game second.

>>89561238
GURPStards unironucally think they are people keke

>>89561238
GURPS is something every person outside of the D&D containment zone has played before, or at least knows the basic mechanics of. According to the most recent /tg/ census GURPS is the third most popular game on this board after 5e and Pathfinder 1e. More people like GURPS than 3.5 and 4e put together. More people like GURPS than OSR shit. And that's just the people who said it was their favorite game, I didn't even answer that way in the thread. The reality is that people under the age of 25 like 5e, and that's it. Nobody else likes it, everyone else is playing GURPS, Genesys, CoC, or WoD. 135 respondents and 11 of them liked 5e, and by the way, several of the people who answered 5e said that they only chose it for lack of any other choice. D&D might have had some wind in its sails over streamershit but that's all dead now. People who actually play games prefer to play more than just WotC's dogshit.

>>89561050
>why
Because he's a fucking retard. Apparently every fucking thing should have a 65% chance of occurring. I don't know what the fuck this guy is on.

>>89561701
He's the guy who rages in every single thread where someone says something negative about D&D, he specifically likes to call anyone who posts HYTNPDND "/tg/'s Worst Troll". Think it's bumptranny but it could just be some horrible new autistic troll.

>>89560352
>normal people play GURPS with a VTT like Foundry
They probably shouldn't because it makes retards think they should play with all autism options on, instead of reading GURPS How to not Be a Sperg

>>89540592
It’s just as good, if not better, for a roll-under system as a d100

>>89561382
Lmao. Tell another one!

>>89560746
>I'm not the person you were previously talking to
Weakest lie in the thread.

>>89559631
You are incredibly stupid if you do not understand why deciding to grant a 65% chance of success no matter how much the PC invested in Gambling or whatever because you want to 'challenge the PC' is a shit idea that robs them of their investment. It means if he invested fuck all it's 65%, if he invested normally it's 65%, and if he invested deeply it's still fucking 65%.

>>89562615
Odds guy is a retard
>need to be able to quickly calculate the odds
>why not use the modifiers in your system based on the level of difficulty?
>huurr I want to set this specific players odds of success at exactly 65%, regardless of his skill level

>>89561382
> According to the most recent /tg/ census GURPS is the third most popular game on this board after 5e and Pathfinder 1e

By how much? If there are 100 board members, and 50 play D&D, and 49 play Pathfinder, and 1 plays GURPS, then GURPS is in third place too, but I wouldn’t call it reasonable to expect anyone but that one guy to know how it functions. Or why it functions. Or even that it functions. In fact with only 1 player I’d question if it even functions.

>>89562615
>It means if he invested fuck all it's 65%, if he invested normally it's 65%, and if he invested deeply it's still fucking 65%.

For the one check, in one particular instance. I’ve made it clear five times now that this isn’t a general thing the DM should do, just a thing that can be done and is useful to do in a pinch.

>>89562684
> >why not use the modifiers in your system based on the level of difficulty?
Because those don’t actually tell me th e odds.

I feel like you’re still missing the forest for the trees here. Again, consider something like a custom monster. You need to have a rough idea how how likely your PCs are to make saves against it, or dodge its attacks, or hit it. You wouldn’t throw a monster with AC 25 and a spell save DC of 4 against 1st level players, would you? At least not if you expect them to ever hit it, or to ever fail against its saves. The 5e DMG has guides but they are expressly just guides, not set-in-stone rules, because D&D’s been around long enough to know that every group is different and you need to tailor the gaming experience to your specific group.

Also here’s the funny thing, you’re not even objecting to the 65% pass/fail rate, since for a +7 (which is what I’ve been assuming) that is DC 15, which is also exactly what th e DMG sets as a Moderate challenge.

So you’re not even really objecting to the odds, you’re just objecting to the fact that I know both how and why I arrived at them, rather than just blindly following what’s printed in the book, since you’re morally opposed to doing that I guess

>>89563109
It's something a DM shouldn't do at all.

File: file.png (67 KB, 524x261)

67 KB PNG

>>89540592
I wouldn't.

>>89562914
Latest survey indicates:

5e has 11 fans, Pathfinder 1e had 9, GURPS and Genesys tied in 3rd with 7 fans.

oWoD is next with 6, then WFRP 2e and CoC 7e tied with 5 each, Shadowrun 5e and Savage Worlds each had 4, and then in 10th place is Cyberpunk 2020 with 3 fans. All versions of D&D combined to 23, and added to Pathfinder 1 & 2 it comes to 33/135, with 2 fans of OSR shit, bringing the grand total portion of DnD-drones on /tg/ to 25.9%.

>>89561382
> The reality is that people under the age of 25 like 5e, and that's it

I’m 36 in six days. I’ve been playing D&D since 1999. I’ve been playing with the same particular group of people since 2005. Two of them are my age, one is two years older, one is four years younger. We all like 5e. In fact of the editions I’ve played (2e, 3e, Pathfinder, 5e), I certainly my personal favorite. I honestly and unironically describe it as feeling and playing like the 3e we would have gotten in 2000 if WotC had already learned all the lessons of 3e and 4e and Pathfinder, due to the smaller numbers, tighter math, emphasis on character ability over magic items, and just generally being fun to play. Plus it’s the only edition where I rolled up a Thief and had a character who actually felt like a THIEF, and not a dungeon-delving assassin who moonlights in larceny.

So, yeah. The point is, you’re wrong about only people under 26 liking 5e.

>>89563260
Listen man the way you clearly don't have a solid grasp on the linear passage of time means you definitely have the mental maturity of someone under 26. In fact you probably shouldn't be in this site.

>>89563249
Hang on, wait, your survey only had about 100 respondents? Less? With, I’m presuming, no way to verify that there were definitely distinct respondents and not people voting for the same thing multiple times?

Actually even that is secondary to the main question, less than 100 respondents out of how large a total board population? What was the participation percentage? Any statistician will tell you that you need a minimum of either 10% of the population, or 1,000 respondents if 10% is too infeasibly large (because you have a total population of more than 10,000), to consider a poll even close to valid.

>>89563321
I’m not the one who can’t grasp basic statistics, which is more than can be said for like half the respondents to this thread, including every GURPSfag.

>>89563260
If we had gotten 5E in the year 2000 I would have just abandoned D&D altogether. At least 3E had some good content.

>>89563333
Sir you literally said "the third edition we'd have gotten if they'd learnt the lessons of shit that came after third". Your understanding of statistics, or lack thereof, is irrelevant to me making fun of you being a retard.

>>89563345
> At least 3E had some good content.
3e had some good fluff. I hesitate to say any of its mechanics were fundamentally good given how easy it was to break the system; sometimes without even meaning to. And at least with 5e you don’t end up with classes as thoroughly useless as the CW Samurai, Healer, lr Fighter. Even the 5e Ranger primarily suffers from being boring rather than actually bad; it’s still mathematically capable of contributing to combat at the expected rate for every level.

>>89563540
If the 5E Ranger contributes mathematically then so does the CW Samurai and Fighter and especially the Healer.

>>89563109
>For the one check, in one particular instance. I’ve made it clear five times now that this isn’t a general thing the DM should do, just a thing that can be done and is useful to do in a pinch.
Alternatively, you could set the DC at a medium level of difficulty and how he does will be dependent upon how much he has invested. Crazy, I know.

>>89550235
Correct, I don't. Such concerns are beneath Gods like myself.

>>89563323
>With, I’m presuming, no way to verify that there were definitely distinct respondents and not people voting for the same thing multiple times?
What? Why would people do that? No one would vote two or thirty times, or link the random thread poll into their General and discord.

The fact is that 100% of /tg/ plays GURPS.

>>89563249
what survey? what was your methodology? where can we see the data?

>>89564745
The survey as always was conducted by the nonpartisan think tank Anon's Butthole

>>89563540
>And at least with 5e you don’t end up with classes
Instead I end up with a game where it's been 9 years and they've yet to produce a single class that is not boring as fuck.

>>89564995
Artificers are fun because you get to watch the DM tear hair out dealing with the dickhole who decided to stack 27 AC on himself

>>89565011
wdym "tear his hair out"? bypassing AC is the easiest thing in the game lol

>>89565011
https://dnd.arkalseif.info/classes/artificer/index.html

>>89565033
yes, that's an artificer. why did you post this link?

>>89565047
Because that's what an actually fun artificer looks like.

>>89565061
it looks like shit

>>89565070
Because you don't know how they work.

>>89565077
I know everything. Don't talk back to me.

>>89563109
>adjusting the difficulty on the fly to give the player an exact 65% chance of success
literal rubberbanding in a tabletop game

>>89564995
I’d like to repeat that 5e is the first edition I’ve played where, when I made a thief, I actually felt like a THIEF, and not a dungeon-delving assassin who moonlights in larceny. Something about Cunning Action coupled with Fast Hands and Second-Story Work, I think. I’ve had more fun playing a Thief in 5e than I did in 2e or 3e, anyway.

5e Rogues in general are like the antithesis of boring. All their abilities are available all day every day and most of the subclasses are built around giving you stuff to do with your bonus actions. If you run out of interesting things to do, you either lack imagination, or you just aren’t trying.

>>89565033 >>89565061
What exactly is fun about the 3.5 artificer? It never seemed particularly interesting to me. Beguilers were fun. Factotums were fun. Binders were fun. Artificers…were just half-casters that could make magic items to ape full casters, while being mechanically worse than them. Why not just be a full caster?

>>89567052
3.5's Artificer is fun because it's the most tweaking-friendly class in the entire game and because it keys directly into 3.5's magic items it has an enormous amount of things to do. Crafting magic items is engaging on its own and it's better at it than anyone else.

>>89548077
>And none of that matters. What matters is your chance of doing what you wanted to do. If you're trying to jump over an obstacle, then what matters is whether you've succeeded or failed at jumping over the obstacle.
Alright, I'll help you out little nogame faggot. Out of the goodness of my heart.
In your typical flat DC, success or fail system like D&D you need to roll a dice, add a modifier, and hope the total is equal or above a specific number.
So lets say you have a modifier of +5 and needed to hit 10, so you need to roll a 5 or higher. If you are rolling a single d20 then you have a 80% chance to make it, but if you are rolling 2d10 then you have a 94% chance to make it. The 2d10 system does indeed make it much more likely that you will succeed on a low DC task.
In the same way if I needed to roll an 18 or higher to succeed then I have a 15% chance using a d20, but only a 6% chance with 2d10. The flat distribution of the d20 makes it much more likely to hit very high DCs.
So the choice of dice is actually important for the reliability that you will hit a DC. Some people like a nice reliable system where 5% of the time they fail at a DC 10 task even through they have +8 to the roll. If I have that much of a bonus I would expect to be very unfortunate to fail at something so easy.

>>89540592
I suggest a compromise system that should solve everyone's issues: a 1d20+2d4 system

File: 1568212554020.png (1.16 MB, 1423x429)

1.16 MB PNG

>>89540592
Lessers envy and fear the greatest

>>89568225
Talk me through it. Whats the benefits?

File: d8 vs 3d6 prob graph.png (25 KB, 768x655)

25 KB PNG

>>89548729
>>89548707
See the pic here >>89559928

Most of the difference between the distribution in your images comes down to the fact that 1d10+5 or 1d8+6 is more equivalent to 3d6 than 1d20 is. 3d6 has a range of 15 vs. d20's 19, but since the highest and lowest values on 3d6 are so improbable, they don't weigh into things much, so for the sake of comparison, the effective range of 3d6 is actually quite a bit lower. Look at the image here to see how extremely closely 1d8+6 parallels the probability curve of 3d6. It's only the extreme stuff at the end of 3d6 which significantly deviates.

>>89569801
The important point here is that the distributions in >>89569801 much more closely resemble each other than the distributions in >>89548729 / >>89548707 because the granularity is similar. So most of the difference you're seeing between the d20 and 3d6 in those latter graphs is because of granularity and not bell curve.

File: 3d6 and 1d10+5.png (32 KB, 1130x402)

32 KB PNG

>>89569801
And here's 3d6 vs. 1d10+5 done on anydice.com.

File: d100 vs d20.jpg (189 KB, 1502x1546)

189 KB JPG

>>89569801
>>89569828
>>89569841
To further illustrate, this is how the flat distributions of 1d20 and 1d% compare if you adjust for granularity/scale and if you don't. The comparisons in the graphs in >>89548729 and >>89548707 are equivalent to the second/lower graph I just posted, so of course they're gonna look very different.

>>89567499
> If you are rolling a single d20 then you have a 80% chance to make it, but if you are rolling 2d10 then you have a 94% chance to make it.
If you’re rolling 2d10 you probably don’t have a +5 modifier. If you’re changing the RNG, why wouldn’t the modifiers also change to account for it?

> So the choice of dice is actually important for the reliability that you will hit a DC
If the modifiers stay the same. Again: why would they? Presumably if in the d20 system someone with a +5 has a 20% chance of failing a DC 10 task, it’s because the game designers *wanted* a DC 10 task to have a 20% fail rate. If those same designers had been working with 2d10, then they wouldn’t have handed out a +5 modifier, they’d have handed out a +2.

>>89567499
>The 2d10 system does indeed make it much more likely that you will succeed on a low DC task.
Unfathomable stupidity

After reading this whole thread I found the D20 apologists arguments more convincing.
3D6 guys you need more convincing arguments.
Call me retarded if you want but if you cant make an argument that can convince a simpleton then what does that say about your arguments? Put more skill progression in Rhetoric next time.

>>89569801
>>89567499
>>89570169
Let's look at the standard deviations for these different dice systems...
3d6 = 2.96
2d10 = 4.06
2d6 = 2.42

1d20 = 5.77
1d14 = 4.03
1d12 = 3.45
1d10 = 2.87
1d8 = 2.29

So the modifiers for 3d6 should be roughly the size of the modifiers for 1d10, and half the size of those for 1d20. You don't expect a +5 bonus to mean the same thing on a d10 as on a d20, and you shouldn't expect it to mean the same thing on 3d6 as a d20.

The modifiers for 2d10 should be roughly equivalent to equivalent to those for a d12, and only about 70% the size of those for a d20. So a +3 on 2d10 would round to a +2 on 1d20.

And the modifiers on 2d6 should approximate those for 1d8.

>>89570825
The main point here is that any argument that says that you're more likely to succeed when rolling 2d10+5 verses when rolling 1d20+5 and thus 2d10 is better (or 3d6 or whatnot), is getting the math right while completely misunderstanding the intentions underlying that math.

>>89570825
>and only about 70% the size of those for a d20. So a +3 on 2d10 would round to a +2 on 1d20.
Whoops. Reverse those, obviously. +2 on 2d10 rounds to +3 on 1d20.

>>89571878
Yeah.

>>89569841
These graphs might look the same but they're pretty meaningfully different. In d10 + 5 the range of characters that can interact with a task at all in the system is drastically lower, you're getting a lot of characters that either autofail or autosucceed. Just flipping from impossible to 10% isn't satisfying. The increased granularity with d20 helps but I think the same problem is there.
Any while the graphs look similar if you squint, the differences are meaningful. During the "middle range" where things feel pretty coinflippy (probabilities around 25-75%) a modifier in 3d6 gets you more - the slope is steeper. 8-9, 9-10, 10-11, and 11-12 are all 12%ish changes, vs the just 10% with d10+5.
>>89570169
Modifiers aren't handed out in a vacuum, they're part of your system as a whole.
>>89570729
A lot of it has been people upset at a guy who wants to ignore player character skills and just set a flat 65% chance at doing some task, which is retarded but unrelated to d20 vs 3d6, you could do the same in any system

File: file.png (61 KB, 1711x394)

61 KB PNG

>>89572187
>>89569841
d8+6 actually matches the curve at the crucial middle part more closely, shows off how meaningful modifiers are in that range

File: #####temp.jpg (97 KB, 1550x663)

97 KB JPG

>>89572187
>>89572202
I'm not trying to suggest that the bell curve of 3d6 is insignificantly different from a flat distribution, but rather to show that the vast majority of the difference between a 3d6 and 1d20 distribution, like these >>89548707 and >>89548729 were posting, comes down to granularity. The difference between the black and yellow lines in the image in this post is a combination of granularity and bell curve. But the vast majority of that difference disappears when we mostly take away the granularity and just look at bell curve (when comparing the black and yellow lines).

>>89572584
>But the vast majority of that difference disappears when we mostly take away the granularity and just look at bell curve (when comparing the black and yellow lines).
*when comparing the black and blue lines

>>89563765
Or consider the task's difficulty relative to existing difficulties you remember, and pick a static number relative to those benchmarks.

>>89572584
Sure, that's fair. I think it's reasonable to say that people want more of a range than d10 though, and 3d6 gives you that range while still having a predictably you don't get with a single die.
We could do 4d6 as well but games don't really need that much precision.

>>89563109
>Custom monster
Notably different scenario than skills.

But, I wouldnt make up the monster numbers on the spot if I'm improvising a monster. I'd grab them off a table. I made myself a table for each of the 4e monster roles, and one 'average' based on the blog of holding numbers.

In a hurry, it worked great when I still ran 5e. Frankly I may have to make a 3e version.

Grab the table that matches the monster's role, go down the rows until I hit the CR I want. The stats are there. Then I just need to give it the appropriate abilities.

>>89550152
>He uses 4chan without browser extensions
My condolences. I mean, I have my auto post loading set to 1800 seconds, but stock 4chan is just so unpleasant to use.

>>89572732
Or decide that you want to challenge the player character at this particular juncture, and decide the percent chance of success there. Then from there build an NPC backwards that can be a rival, informant, contact, or otherwise recurring character, who is now notable for being approximately equal to the PC in skill.

By the way I’m gonna take this time to remind y’all that this conversation spun out of what turned out to be a GURPSfag claiming that it was unthinkable and pointless for a DM to have an idea of what his players’ characters are actually capable of and being incapable of thinking up a reason for why you would ever need that information, when it seems to me that it’s plainly obvious that if you don’t know your PCs then you’ll never be able to reliably challenge them.

And relying on difficulty tables in books isn’t reliable either, as I later pointed out with my example of a party that appears balanced on paper (fighter, rogue, wizard, druid) but actually has a glaring weakness which means that a nominally “balanced”, even supposedly easy, encounter (an umber hulk) is actually far more dangerous then it should be due to that weakness.

You don’t like that I occasionally try and challenge a specific character, fine. But let’s not forget that the underlying point was that a responsible GM knows, at least roughly, the stats of his players’ characters, so as to make sure that he’s not constantly throwing things that are too easy or too hard at them. Occasionally, sure, but that should be due to deliberate GM choice. Not laziness.