Dice Mechanics

Which is your fave? d20? d100? 3d6? Dicepools?

And why?

I enjoy the d100 the most.
I like how it's easily quantifiable with regards to percentages.
"You have a 20% chance of success", etc.

by the way would you happen to know what game OP's pic is from? I found a dice exactly like it and I don't know where it's from. I got it from one of those assorted dice bowls in an LGS.

I like d20's because favoritism. I can't give a rationally built reason why.

When DM'ing is i give players specific dice for their weapons, and i have noticed that makes people far more attached/engaged to their dice, and when they get new equipment they get excited for new dice.

I like d100 too.

>by the way would you happen to know what game OP's pic is from?
Looks a lot like Arkham Horror dice to me, user.

I need some D6 that are basically illegible through their patterns.

d100 is objectively the best cause you can basically make distributions of any function you'd ever need, but it's fun to use the d6s and d20.

I saw someone recommend 3d20m (roll 3 d20, use the median roll, ignore the other two)
It's results are an interesting medium between 3d6 and 1d20

I like dicepools with underrolls because I like how it sounds.

Roll and Keep from L5R is probably my favourite.

Can you explain it some more for me? Crab Bro's story thread has been getting me interested in the setting and I'd like to hear more of the system.

Dicepools Collect Successes hands down. Rolling a handful of dice and sorting certain numbers out is just fun by itself, it has a visceral and intrinsically motivating thing about it that no other resolution mechanic has.
The system needs to be well designed though. It has to hit a sweetspot for the number of dice and the mechanics have to be intuitive. If the increase in the likelyhood of success feels right when you get +1 or +2, you really don't need to know the exact percentages.

Least favourite are Dicepools that are shittily designed and d100.
Shitty Dicepools are unintuitive and don't work right. LotW is a good example: Even the shittiest possible character has always at least a >90% chance to roll doubles, but bump that difficulty up by one and it goes down by 70 to 75%. And it that doesn't significantly change for better characters either. So not only does even a slight increase in difficulty fuck up the numbers, rolls produce mostly the same outcomes to a point where you don't really need to roll at all most of the time. Sure, there's the dimension of additional pairs, but it's the same here. There's very little variance to the system.

D100 just don't feel fun to roll for me. Sure, it gives me the greates amount of "control" as DM, but you don't really need that amount of granularity most of the time anyway. The D20 does exactly the same job while being at least a bit more satisfying to roll.

You have a Stat and a Skill. That's the amount of dice you roll.
You pick as many dice from that roll as your Stat and add them together. That's your result.

I love either d10 dice pools with target numbers or D100 roll under

Can confirm have a set. Arkham Horror dice

I refuse to play any systems that require nerd dice.

I dislike the d100.

It's pointlessly granular (seriously, which GM is going to be messing around with 1% difference on DCs?), and leads to the same result as just rolling a single d10 90% of the time.

Some systems make it work, by using the second digit as hit location for example, but then you are really just rolling twice in a d10 system for hit and location.

D20
using d100s ends up using multiples of 5 anyway, and rolling just one die is a lot more convenient

1 die instead of 2 seems like a trivial thing, but over a long period of time, tiny things like those can be quite significant

I have a guilty pleasure for dicepools like FF Star Wars.

I'm a bit of a dice junkie and rolling giant handfuls of dice at once is the greatest feeling ever.

I like dice pools. I think the biggest reasons is that they allow more refined control over effort, and that they're usually associated with more robust wound systems.

When it comes fights against mooks, lieutenants and then the big bad, it feels like the narrative meshes better with the mechanics of the combat.

that is the dark eye (DSA) dice mechanics

>rolling just one die is a lot more convenient
a lot more? seriously? so GURPS' 3d6 is even more inconvenient?

D100
simple and to the point

at a guess something cthulhu esque.
looks like the elder sign.
maybe arkham horror

I prefer 2d10 percentage dice, or d100.

I like dice pools rolling vs target numbers, especially d10/12/20 dice pools. I like the mechanics, it feels nice to roll a chunk of dice and it speaks to the die collector in me.

Either 3D6 or Target Number systems, like Savage Worlds.

what's the appeal of 3d6?

>I need some D6 that are basically illegible through their patterns.
You want Q-Workshop. Very pretty, great for gits and such, but most are terrible to play with because of too much ornamentation.

Smart distribution of randomness. Hitting 10 is exactly 50% and the distribution is more of a bell curve rather than D20's flat distribution.

As someone who loves LotW, I think you're missing the point of what it's trying to do. Then again, the system does an awful job of explaining itself, so I can't exactly blame you.

The dice pool in LotW is relatively reliable and static, but dice increases still does give you a lot better chances. Looking at the chance of getting one big set is missing the point- Multiple small sets can be just as effective.

It also needs to be realised that unlike most dicepool systems, increasing your dicepool is not the primary means of progression in LotW. Your static modifiers that you add to your sets are the 'real' power increase between ranks, the dicepool is just a nice, general indicator of your current strength that also gives you more chances for better options.

LotW is a weird as fuck system but when you get it working and understand what their intentions were (after a long and arduous process), the dicepool system works wonderfully.

I know, I was fascinated by it and spend a good three weeks fucking around with math and concepts because I wanted to understand the inner workings of the game. I don't doubt that it works well for what it wants to be, a Wushu Protagonist is either incredibly skilled at what he does (hence having a >90% chance to achieve a 3-Normal if you have +10 in a skill and a 100% chance to achieve a 2-Easy), but in the end the numbers boil down to very little variance. They tried to alleviate that a bit by using the face of your set too, which helps in combat. But at that point you are pretty much hitting each other with 50% most of the time.

I ran a campaign with it, because I really wanted to like the system. I just don't feel it's intuitive in the slightest and doesn't work out all that great at the table, even though it has a bunch of ideas that sound brilliant on paper.

so? why is a bell curve better than "flat distribution"?

>what's the appeal?
>why it's better?
Nice goalpost changing.

It's a meme, mostly. People approach them as if 3D6 is some kind of weighted D20, when they actually work completely differently. In reality it doesn't matter if you are rolling a 3D6 or a D20 if the DM wants you to have a 50% chance to succeed. The 3D6 is just "obscured" compared to the D20, which gives people a smug sense of superiority because they don't actually know how it works but it's definitely better than a D20.

it seems insignificant, but it really helps speeds things up and keeps the mood flowing at my table

i love d10 with another D10 kept beside you in case you need to go D100.

Retard detected. Games like GURPS fix the success rate to you, not your opponent.

I don't understand why people like D100, It's literally the exact same as rolling a D20.

That's definitely Arkham Horror's dice

>lots of numbers that rarely ever get used
>no bell curve and the mechanics generally don't reflect one
>unironically wanting to roll d20s

>swingy as fuck
>modifiers don't matter, especially in the shitty new edition with its "bounded accuracy"
>crit fail autism ("lmao u kill urself")
>nat20 autism (see: the critical roll roastie screeching at an event that has a 5% chance of happening)
>being associated with crusty grognards, autistic munchkins, cancerous furries/ weebs, and skyrim addicts

>d100 is objectively the best cause you can basically make distributions of any function you'd ever need,

Yeah but basically no game actually does that without autistic math so it's essentially a giant d20. d100 is not "objectively" the best, you autistic fuck, just because you watch lindybeige jacking off over runequest doesn't mean you know shit about shit. 3d6 is just as good if not superior.

It's basically a pointlessly complicated dice mechanic John Wick made up so he could point at it and say "Look! Look! I had an original idea! now play my seven seas setting that is as boring and cliche as the old Sid Meier's Pirates game with some additional bullshit, and ignore that the mechanics fall apart everywhere behind my single vaguely-clever idea!"

>m-muh bell curve!

70% are 70%, no matter what dice you roll. All it does is heavily change the worth of modifiers, depending on the base number.

>just because you watch lindybeige jacking off over runequest doesn't mean you know shit about shit
I never understood why lindybeige likes RQ so much, everybody I knew thought several design decisions made it borderline unplayable.

>as boring and cliche as the old Sid Meier's Pirates
While i might actually agree with some of your bait post, this triggers me. For starters, it's Pirates! not Pirates.

such as?

I consider d100 and d20 the same mechanic. It's just a flat distribution. They pretty much need some manner of strong-but-sharply-diminishing returns baked in, otherwise you get the expert who is actually not much better than some random guy (5e) or the actual expert who is such an expert that he throws +56 per die eventually (pathfinder).

I like 2d6 systems myself. The dice are small enough that only the most significant modifiers are added, which forces systems to focus on the big things and not get nitpicky. The diminishing returns and bell curve probabilities are baked in. Results of 2 and 12 are rare enough (~3%) that you can still use them for critical results if you care to.

I thought oWoD (haven't played new world of darkness) had a very intuitive setup. Stat+Skill = this is how many dice you roll. Difficulty = what you need to roll to keep a dice. How many dice you keep = degree of success. It was a bit cumbersome in actual play because you were rolling a lot and counting a lot, compared to d20, but on a platform like roll20 it would be easy to deal with.

You can get completely fucked over by rolling poorly on your advancement checks several times in a row, it can permanently halt your characters progress if the DM doesn't intervene.

d100 and d20 make you feel heroic while
dice pools make you feel like a genius. 2d6 gives you a feeling of being a squishy mortal fuck whose about to get murdered horribly.

While oversimplified, you have a good point.

Dice mechanics aren't really able to be evaluated in a vacuum. They have strengths and weaknesses, but often something that might be a weakness in one context is a strength in another. It all depends on finding a dice mechanic suitable to the tone you're trying to establish for your system.

Then again, there are raw garbage dice mechanics, like Cthulhutechs nonsensical fucking poker dice.

Dice pools. There's nothing more satisfying than rolling 40 dice all at once, sorting out the success, then roll all those again.

You're a retard. Unless you adjust the modifiers or skill increases to match a bell curve, you are not getting a bell curve. You are just getting d20 with more granularity.

>70% are 70%, no matter what dice you roll.

Not everything has a 70% chance of happening. Not all of that is reconciable with linear modifiers.

>There's nothing more satisfying than rolling 40 dice all at once, sorting out the success, then roll all those again.

Yeah except when you have to pick them all up afterwards. Or when the rest of your group kicks you out for being a twat who takes 50 minutes to take his turn.

I like the bell curve of 3d6.

I don't have any particular favorite, but I like when the basic mechanic does something interesting. Such as
>ORE's sets of quality and speed
>UA's percentiles with flip-flops and normal-matched-crit granularity
>MAJYCK ROONS LAHD

The official Shadowrun dice. Near impossible to read under normal lighting conditions.

>Then again, there are raw garbage dice mechanics, like Cthulhutechs nonsensical fucking poker dice.

I like it a lot in theory, but in practise it's so complicated. Two methods of checking successes is fine, three is too many.

The nice thing about the roll and keep system is that it offers two tracks for advancement, which works well to keep powergaming in check.

A guy who has 6k5 is still able to compete with the guy who has 12k5; the latter does better on average, but especially when it comes to combat he's not going to onepunch his way through everything.

I love the Edge of the Empire dice system.

I love non-binary results. Success/failure with advantage/disadvantage works really well narratively and makes things really interesting, imo.

the irony is rich

I think the rest of the game is a half-assed piece of shit, but I really like the dice mechanic from the Dragon Age RPG.

You roll 3d6 -- two of the d6's are the same color, the third is a different color. You add the results of the 3d6 to your bonus in whatever you're doing and you succeed if you equal or exceed the target number.

If the roll generates doubles on any two of the three dice, you get "stunt points" equal to the number on the odd-colored die. You can spend stunt points to get different bonuses. On an attack, for instance, you could move yourself or the target of the attack; knock the target prone, disarm the target, deal extra damage, ignore some of their damage reduction, make another immediate attack, etc.

I like 3d6 for the bell curve, and I like stunt points as an alternative to D&D-style critical hits. If it was attached to a better overall game, I would love it. And if I ever create my own RPG, I'd probably use that basic system.

D10/D100 are the best combo to use, Lets you scale combat much more than a D20 alone

>well well 3d6 doesn't tell you the EXACT percentage therefore I like my game where you always improve 1% chance at a time and there is no bellcurve at all.

D7 or D13..

god i love my primes

How is it to DM? From what I read it looks like it requires a lot of improv from the DM to make things all fine and dandy. That said, it is interesting cause it makes it cinematic like the films.

On a D20, you have an equal probability to roll any number. This is swingy as fuck, and means that masterswordsmen are just as likely to get fuck themselves as an amateur is to crit as joe bumblefuck is to do average

In 3d6, average results are much more common. Furtger, how successful you are is directly tied to your skill in a subject

This is why it's different, yes. But for some games, swingier dice make more sense than reliable bell curves. It depends on what you want from the system.

Once players start collecting different Talents and equipment from their careers, they have more and more mechanical uses for the surplus advantage/threat that comes up with the dice. There really isn't a whole lot of need to "divine" the dice results when you're the GM compared to other narrative-focused systems. It's honestly one of my more favorite games to play for a balance of rules-lite gameplay while still having a decent amount of character options.

I'd say the most difficult on-the-fly improv thing you have to do is deal with the Motivations or Obligations that might come up at the beginning of the session, since you have to find a way to work them into whatever the PCs are currently doing plot-wise.

You confuse the crit mechanic (which is pretty shitty imo), with the "swingyness" of a d20.

If your badass warrior needs to roll a 5 to hit a peasant you WILL hit 80% of the time on a d20. There's nothing swingy about that.

My DMing style is very improvisational and open to player input, it works perfect for me.

The system comes with a ton of examples on how to spend advantage and disadvantage in a huge variety of scenarios. You can spend it narratively or spend it according to the rules.

Like, 3 disadvantage can cause a gun to jam when it's fired.

I'm working on a game right now that uses a modified version of this dice system and the playtests have been incredible.

>You confuse the crit mechanic (which is pretty shitty imo)
crit mechanics are standardfare in RPGs. usually, only deendeetards blame any absurd stories revolving around them on the mechanic instead of shitty GMs

also: the swinginess is due to modifiers. with bell curves, stacking modifiers will roll out smoothly. with linear dice mechanics, you might end up with >100% or

>crit mechanics are standardfare in RPGs

Yes, and when they are put there for reasons, not just because "RPGs are supposed to have crits", they are usually put there to reintroduce some swingyness into combat, for example if

>with linear dice mechanics, you might end up with >100% or

d% roll under like this:

1-n=1
2-Roll a d0-9
3-If its less than the N digit you pass, if its more than N digit, you fail, if equal to N digit go to 4
4-N = N+1
5-Go to 2
So percentages can be as specific as you want and the chance of rolling two X times is just 1 in 10^X

The fact it uses linear percentage makes easier to convert your ideas to rpg stuff

>1 die instead of 2 seems like a trivial thing,
You only 2 or more dices in a d100 1/10 of the time

Sure.

I did't confuse shit, I'm using it to illustrate one of D20's issues you tard. If you're just as likely to crit as you are to miss as you are to hit normally then that's a legitimate difference from a bell curve system.

> rolling one dice at a time

Why... would you ever do this. Why would you want percentages more specific than 1%? Who would ever keep rolling one dice at a time then checking the result to see if they should keep rolling or not?

>Who would ever keep rolling one dice at a time then checking the result to see if they should keep rolling or not?
Rolling alot of dices is usually considered a bad thing by people

>If you're just as likely to crit as you are to miss as you are to hit normally then that's a legitimate difference from a bell curve system.

Are you legitimately retarded?

If you have the warrior in my example, he has 5% of a crit, 20% of a miss, and 75% of a normal hit.

How the fuck are his chances all the same?

Any single number has the same chance to come up, but that really doesn't matter when you are dealing with DCs.

What's wrong with the rest of the game? I breezed through the book once because I heard the dice mechanic was good, and it didn't look terrible. Not a ton of content, but I only did the first book.

>dices
Also, your idea is as retarded as it is the first time I saw it. Even 100 unique results is excessive in most games that provide them, why would you ever need 1000, or 10000? What does this granularity provide?

I've actually been using L5R's roll/keep system for a Monster Hunter RPG. I've been batting back and forth with using TN vs Success, but still keeping the exploding dice.

>needing 50 minutes to count dice
how?

>Rolling alot of dices is usually considered a bad thing by people
for u

3d10 FATAL bellcurve or bust

>why would you ever need 1000, or 10000? What does this granularity provide?
converting ideas without having to think twice about it.
think about some problablility based at some idea X, and you do exactly as it came into your mind or exactly how it was at the place you got it from. no need to mess with it.

d12 best die. I wanna make a d12 system just to give it some love

a mixture of d12's and d4's, in a cup, Smashed down *hard*.

>Dice mechanics aren't really able to be evaluated in a vacuum
Game design isn't really able to be evaluated in a vacuum but that doesn't stop the existence of generic systems.

A generic system is still a context you can use to evaluate the game design used to achieve its goals.

I think the d20 would work better if it was 5d4 instead.

I've always had some frustration with people at the table to are a little slower at the math stuff for some systems, so for the homebrew I've been working on, I decided make that a non-issue. The general idea is: roll 1d12. If the result is equal to or lower than your stat (which range from 3 to 12), tell the GM the number. Otherwise you fail. The GM then compares it to the difficulty: if the number is equal to or lower than the difficulty, you fail. Otherwise you succeed.

Basically, your range for success is (difficulty +1) to (your stat). So the chances of success are overall the same as a roll-under with modifiers to the stat, but the only math the player has to do for a check is a direct comparison of two numbers, while all the modifiers are handled GM-side. To determine difficulty, basically just start at 0, and add 1 for each thing you think would make the task more difficult. As an example, climbing a steep (+1) cliff with few handholds (+1) that's still damp due to a recent rain (+1) would have a difficulty of 3. If your stat was 9, you'd have to roll less than 9, but greater than 3 (4-9, or a 50% chance of success).

I don't know if it's my favorite yet, since I haven't gotten a chance to do a real-world playtest yet, but I really like the idea of cutting out a lot of the math player-side to speed things up.

I like a combo d%/d10 thing, something I was working on a while ago. The idea is that you roll d% for your checks, but you also use the ones place as a d10 for any secondary things, like damage or how effective the result is. I've really been trying to figure out a good way to do "one roll for attack and damage" type things to speed up combat, and this was one of my favorite versions.

I mentioned it earlier in the thread, and it does something along those lines: Unknown Armies. It's d% roll under, any success is a hit. If it's unarmed, you add the pair of dice to get the damage, with melee weapons adding to it. If you hit with a gun, you deal the amount you rolled. There's also special effects that happen on matched successes (melee weapons deal damage like a firearm, for example), and 01 (outright killing someone, usually). The same sort of granularity goes in reverse for failures, matched failures, and 00.

You'd probably want to muck with the effects of each, based on how lethal you want the system to be. Also, adding in new effects based on what the ones place is would be interesting.

Been working on a homebrew game. Stats go from 2 to 4. Roll 3d6 - dice that roll under your stat are successes, dice that roll over are fails. 0 successes is "No, and...", 1 success is "No, but...", 2 successes is "Yes, but..." and 3 successes is "Yes, and..."

Don't know if it'll be good though.

3d6 because bell curve. 1d100 and 1d20 are flat curve trash. I need THICC distributions.

Seconding this, Q's dice are fucking unreadable.
Some of their work includes and

Care to explain? Not the first time I've heard good things about the system.

DnD retards have trouble processing more than one die at time. Yet prefer to autistically calculate xp for every slayed goblin instead of just giving off levels as plot progresses

...

Go read Warhammer Fantasy Roleplay 2nd edition, sounds like it's got a lot in common with your ideal that you can pilfer.

How do you predict the difficulty of tweaking things in a roll-and-keep system?

I feel like a bonus dice or penalties like rerolling your best die are hard enough to predict, let alone with variable target numbers?

math that shit up on anydice and put probability tables in your shitty homebrew

Like so

Thank you

At least it's not a D20 heartbreaker

>requiring special die just to choose your character's zodiac
I bet this system was actually made by Jews.

3d6 is best as it has a beautiful bell curve and uses standard dice.
1d100 is good as it's straight percentages, but it is annoying how you either have to have 2 different color dice or roll them separately.
d20 can go straight in the trash, 5% or more crit fail/success chance, no thanks.