[d20] Anyone ever tried the 3d6 variant?

[GoatBoy]

This is mainly directed towards the 3d6 variant for d20 rolls as presented in D&D 3.5's Unearthed Arcana, but any d20-based system could apply.

I'm just wondering about anyone's experience with it, and any thoughts they have. I prefer reducing randomness in games, and I'm also wondering what a game system based on a 3d6 roll would be like.

[AstralFire]

In my experience:
- Critical hits become less intuitive to figure out, slows down gameplay.
- Almost all non-ToB sources of melee damage rely on increasing the risk to increase the reward - this is much worse for iterative attacks, since they rely on being long-shots that'll hit on a high roll for lots-o-damage.
- Skills become less of a ridiculous gamble.

I recommend using it if you have ToB and not otherwise; the first two points can make things too weird for conventional melee builds.

[J.Gellert]

It does what it's supposed to do, but it's not worth the trouble in my experience.

And I am not certain what it's supposed to do is all that great, anyway. Small differences in skill become a lot more obvious. If you are not serious about increasing your attack bonus, or your saves, or your spell DCs, you might as well not even bother.

[AstralFire]

Making small differences matter is part of the point. The issue is that Core D&D has so many mechanical things layered around that it can cause complications, and it's really better making a system from ground up to involve it.

It probably should work out just fine in Star Wars Saga and 4E.

[Autolykos]

I tried GURPS with both 3D6 and D20, and I prefer 3D6 for this (but that may well be because for GURPS using 3D6 is the default and the whole system is designed for this). However, the main effect of increased randomness is to reduce the players' and GM's ability to plan ahead - which can be good to increase tension (especially in horror games) or bad when the game is very tactics-heavy. Failing at an easy but important step in your plan can make your whole day suck. If this happens regularly, your game has begun its rapid slide into slapstick (which does not have to be bad...).

[Psyx]

Using 2d10 gives a nice bell-curve for skill checks, and it's what I prefer.

However, for combat and saving throws, it just works better to use a d20.

[GoatBoy]

Very helpful feedback, thank you all.

I was just considering a new homebrew RPG system combat system, where there were multiple options for attacking and defending, and where one round lasted one second to give the feeling of short, furious battles.

I am guessing that less randomness would be better in a more complex system, whereas more is beneficial for D&D's "hit or miss, once per turn" melee.

Edit: This post finally took me out of "pixie," hooray.

[AstralFire]

One round to one second is probably subdividing too much. I can't imagine trying to have a chase at that speed. Grats on your halfling status.

[Greenish]

I tried GURPS with both 3D6 and D20, and I prefer 3D6 for this (but that may well be because for GURPS using 3D6 is the default and the whole system is designed for this). However, the main effect of increased randomness is to reduce the players' and GM's ability to plan ahead

3d6 reduces randomness from d20.

[The Big Dice]

3d6 reduces randomness from d20.

It reduces randomness, but that has the benefit of letting a player play the odds with more success. When you can rely on getting 12 or less 75% of the time, you're more inclined to go for things. It also makes your "50/50 DC" much more important. That's the number you'll hit on a 10, for those unfamiliar with the concept.

[Toliudar]

I've used the 3d6 variant for skills only, and liked it a lot.

[Iceforge]

It reduces randomness, but that has the benefit of letting a player play the odds with more success. When you can rely on getting 12 or less 75% of the time, you're more inclined to go for things. It also makes your "50/50 DC" much more important. That's the number you'll hit on a 10, for those unfamiliar with the concept.

Surely you mean "on a 11"?

At least for a 50/50 DC

To hit something on a natural 10 or greater, is a 55% chance of success.

On 11 is 50/50.

Observe

DC 11
Fail Numbers: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10 = 10 numbers
Win Numbers: 11, 12, 13, 14, 15, 16, 17, 18, 19, 20 = 10 numbers
= 50/50

Likewise DC 10
Fail numbers: 1, 2, 3, 4, 5, 6, 7, 8, 9 = 9 numbers
Win Numbers: 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20 = 11 numbers
= 55/45 in the rolling parts favour

[Callista]

If you have people who are slow at math, then it really slows things down. You mightn't think it's such a big deal, but if they happen to be playing the level 17 TWF ranger with ridiculous numbers of attacks (or similar), they're going to be rolling a lot of dice and doing a lot of adding on their fingers while everybody waits.

[Greenish]

It reduces randomness, but that has the benefit of letting a player play the odds with more success.

Indeed. I was responding to someone claiming that it increased the randomness and thus reduced the ability to plan ahead, ie. play the odds.

Or so I understood Autolykos' post.

[Morquard]

This sounds like an interesting, but how does that work on higher levels, with classes that have 3/4 BAB for example? They usually have to roll higher to even hit, and a 13 and higher e.g. is alot more unlikely with this than with a d20.

Doesn't that result in fights where either the full-bab classes almost always hit, because the AC is low enough to give the 3/4s a good enough chance. Or its balanced for full-BAB but then the rogue has problems.