Dice Bucket Engines - *why*?

[Ancalagon]

Hello

I was reading the thread about "Forbidden Lands" and seeing how the system worked prompted me to write this thread.

So over my gaming career, I've encountered a few system that use a "dice bucket" engine. For those unfamiliar with it, when a PC does a test of some kind (roll to hit, skill check etc) they get a number of dice - say, 3 dice because they are strong, 4 dice because they are good with a sword, and 1 dice because the sword is enchanted, for a total of 8 dice. All these dice are rolled. Each dice has to hit a certain target - say 7 or higher on a d10 - and the number of successes (say you got 3) is how well you did on the test. An easy test may require a single success, while a hard one will need several.

My personal experience comes from exalted, where rolling 18 d10 wasn't uncommon (hence the term "dice bucket" engine). I think that shadowrun used it, I know that the first star wars game did, other white wolf games did too, there must be some I don't know about, and now the Forbidden Lands is using one too.

My question is *why*. Why on earth would you do things this way? It makes each roll slow and tedious. There must be an advantage to dice bucket engines that I'm unaware of...

Please discuss

 

[ccs]

As the publisher of the game you own stock in/are in league with a dice manufacturer?

 

[Morrus]

The term is “dice pool”. If the system doesn’t make the mistake of creating too large unwieldy dice pools, adding or removing a physical die is a really easy, intuitive, tactile version of a modifier.

YMMV, of course. I like them (I use d6 dice pools in WOIN) but everybody has different tastes.


Sent from my iPhone using EN World

 

[Staffan]

As Morrus said, it makes each bonus or penalty a noticeable thing. In addition, it allows for graduations in successes. As long as the pools are kept to a manageable level, I don't find them to be much slower than d20+modifier. Of course, Exalted takes it to an extreme with dice pools of potentially 20+ dice.

It also gets easier if you have specialized dice. For example, Mutant: Year Zero uses yellow dice for dice from your stat, green for those from your skill, and black for those from gear. All have a radiation symbol on the 6, denoting a success. The yellow dice also have a bio-hazard symbol on the 1, signifying trauma taken on a push, and black dice have an explosion symbol on the 1 signifying degradation of gear on a push. You can use regular dice, but it gets a lot easier with those. Similarly, White Wolf used to release d10s with different colors on the 8, 9, and 0, denoting successes (in new World of Darkness).

In the case of old World of Darkness and older editions of Shadowrun, they also allowed for dual measure of success by varying the target number on each die as well as the number of successes needed, but I don't think I've seen that in a game released since 2000 except the anniversary editions of World of Darkness.

Star Wars d6 used a different type of dice pool, where you sum up all the dice and compare the total to a difficulty. This does take a little longer, but you pretty swiftly learn methods of making it go faster (group up dice that sum to 10).

The latest evolution of dice pools is, I think, the narrative dice system used in FFG's Star Wars system and their upcoming generic RPG Genesys. These use six different types of dice: three positive (ability, proficiency, and bonus) and three negative (difficulty, challenge, and setback), and have symbols on them instead of numbers. Ability and proficiency dice are gained based on stat and skill, and difficulty and challenge dice based on the actual difficulty. Bonus and setback dice generally reflect situational modifiers. Positive dice have successes and advantages, and negative dice have failures and threats. Failures negate successes, and if you have any successes left over you succeed in your action (extra failures do nothing). Threats negate advantages, and depending on net advantages/threats you can get good/bad side effects. For example, if you're slicing a computer system a roll that generates two net successes and three threat might mean that you did well so you got the result in half the normal time, but you also set off an alarm alerting someone to your position. If you're shooting someone and roll no net successes but two advantages, you might miss your opponent but force them to move so they are more exposed when your friend shoots them, giving your friend a bonus. There are also Triumph and Despair symbols on the Proficiency and Challenge dice, which both count as a success/failure and an extra-strength advantage/threat. It sounds complicated, but it works amazingly well in practice.

 

[Ancalagon]

The term is “dice pool”. If the system doesn’t make the mistake of creating too large unwieldy dice pools, adding or removing a physical die is a really easy, intuitive, tactile version of a modifier.

YMMV, of course. I like them (I use d6 dice pools in WOIN) but everybody has different tastes.


Sent from my iPhone using EN World

I do grant that the tactile effect is there yes. And I'm not saying that people are "bad" for using dice pools, I'm just trying to see why.

In addition, it allows for graduations in successes.

It does allow for graduation of success yes, but so do non-dicepool engine. In warhammer, where you roll a % dice, you have degrees of success as well. So I must conclude that this is not a benefit of a dicepool engine.

It also gets easier if you have specialized dice. For example, Mutant: Year Zero uses yellow dice for dice from your stat, green for those from your skill, and black for those from gear. All have a radiation symbol on the 6, denoting a success. The yellow dice also have a bio-hazard symbol on the 1, signifying trauma taken on a push, and black dice have an explosion symbol on the 1 signifying degradation of gear on a push. You can use regular dice, but it gets a lot easier with those. Similarly, White Wolf used to release d10s with different colors on the 8, 9, and 0, denoting successes (in new World of Darkness).

I'm sure that specialty dice would help indeed, but this isn't a "benefit" of a dicepool engine.

In the case of old World of Darkness and older editions of Shadowrun, they also allowed for dual measure of success by varying the target number on each die as well as the number of successes needed, but I don't think I've seen that in a game released since 2000 except the anniversary editions of World of Darkness.

In exalted certain powers allowed you to re-roll 1s, or treat a 6 as a success so there was that... I wasn't too impressed with this mechanism, it seemed fiddly.

Star Wars d6 used a different type of dice pool, where you sum up all the dice and compare the total to a difficulty. This does take a little longer, but you pretty swiftly learn methods of making it go faster (group up dice that sum to 10).

Ah, so it's not a number of success, but a total score rolled? I didn't know that, thank you. I don't think it's a "benefit" but at least I know more about the system.

The latest evolution of dice pools is, I think, the narrative dice system used in FFG's Star Wars system and their upcoming generic RPG Genesys. These use six different types of dice: three positive (ability, proficiency, and bonus) and three negative (difficulty, challenge, and setback), and have symbols on them instead of numbers. Ability and proficiency dice are gained based on stat and skill, and difficulty and challenge dice based on the actual difficulty. Bonus and setback dice generally reflect situational modifiers. Positive dice have successes and advantages, and negative dice have failures and threats. Failures negate successes, and if you have any successes left over you succeed in your action (extra failures do nothing). Threats negate advantages, and depending on net advantages/threats you can get good/bad side effects. For example, if you're slicing a computer system a roll that generates two net successes and three threat might mean that you did well so you got the result in half the normal time, but you also set off an alarm alerting someone to your position. If you're shooting someone and roll no net successes but two advantages, you might miss your opponent but force them to move so they are more exposed when your friend shoots them, giving your friend a bonus. There are also Triumph and Despair symbols on the Proficiency and Challenge dice, which both count as a success/failure and an extra-strength advantage/threat. It sounds complicated, but it works amazingly well in practice.

Intriguing - I would have to try it to really know how it works. Is this really a dice pool engine though?

 

[Sunseeker]

Rolling dice is fun. Plain and simple. The "dice pool" concept plays upon this.

 

[Jhaelen]

I also dislike systems using 'buckets of dice', but not all dice-pool systems are like that.

It does allow for graduation of success yes, but so do non-dicepool engine. In warhammer, where you roll a % dice, you have degrees of success as well.

As you do in Runequest.
But the way dice-pools provide it, is more elegant: You don't have to make any calculations, you can tell the result at a glance. It's especially beneficial to players who are mathematically challenged.

Dice pools also scale better: you can always add more dice, and a result of zero successes remains a possiblity. Yet, by being able to achieve more overall successes, you can achieve progressively better results that cannot be achieved by characters with a smaller dice pool. Using a percentile system you are limited to a range of 0 to 100%. Trying to replicate the advantage of increasing dice-pools would result in complicated math and most likely something as unwieldy as look-up tables.

 

[Staffan]

Intriguing - I would have to try it to really know how it works. Is this really a dice pool engine though?

I'd argue that it is: you roll a number of different dice depending on stat, skill, difficulty, and circumstance to see how well you do. It's a bit different due to the iconic rather than numeric nature of it, but then again you could just as easily do the recent editions of Shadowrun using dice with four blank sides, one "Success" side, and one "Success plus" side.

I'm not saying dice pools are necessarily better or worse than more linear systems. They are different, which is good in some ways and bad in others. Some go to a bit of an extreme, such as Exalted with its humongous dice pools.

Another difference with dice pools is that they make probabilities somewhat more opaque. In D&D, it's trivial for me to know that if I have a +5 attack bonus and need to hit AC 13, I need to roll an 8+ which is a 65% chance of success. If my opponent has half cover, the AC increases by 2, so I'm down to 55%. In Star Wars FFG, I might instead be rolling 1 green ability die and 2 yellow proficiency dice, versus a difficulty of 2 purple difficulty dice - I'm feeling relatively confident because I have more good dice than bad and the good dice are also better than the bad ones are bad, but I can't figure out the probability of a hit in my head. I definitely can't judge how my chance is affected by the addition of a black setback die for cover. Even in a simpler dice pool system like Shadowrun* (which rolls regular d6es and counts 5s and 6s), it's non-trivial to figure out the chance of rolling e.g. 3 or more successes on 7 dice. You could do it, but it's not something you do in your head while shaking the dice in your hand. Again, this is something some would see as an advantage, others as a flaw.

* I'm not saying Shadowrun as a whole is simpler than Star Wars FFG - just that the actual rolls are. There's a lot more going on leading up to those rolls in Shadowrun, though.

 

[The Crimson Binome]

My question is *why*. Why on earth would you do things this way? It makes each roll slow and tedious. There must be an advantage to dice bucket engines that I'm unaware of...

At least in theory, rolling a large number of dice gives the possibility of an uncertain outcome, since even rolling a hundred dice could result in no successes. Statistically speaking, though, things will very quickly converge to the average as you roll larger numbers of dice. If you want randomness and unpredictable results, then it's better to stick with a d20 or percentile system.

If you don't want randomness, though, then a bucket of dice can really help you with that. Something that D&D doesn't really allow for is the Batman vs Army of Mooks scenario, where we expect Batman to get through every enemy untouched. For as long as you have that natural 20 as a safety valve against certainty - or even worse, Bounded Accuracy - Batman will fall to a reasonable number of mooks in just a straight fight. And of course, it wouldn't be much fun if we simply disposed of that rule, because then we would know with absolute certainty that Batman is invincible. The dice bucket engine will allow us to stat out each mook with maybe six dice, while Batman has twenty dice (or more), and we don't know that Batman will be perfectly safe the entire time... but he probably will be. It disguises the amount of uncertainty involved.

That's still somewhat of a secondary benefit, though. The major benefit is just that it gives a better distribution of probabilities around the average. If you are rolling fourteen dice and need five successes, then you are very likely to succeed by a margin of two points, but there's a smaller chance that you succeed by a margin of one or three, and an even smaller chance that you succeed by a margin of zero or four.

If you're rolling d20 + 14 against a base DC of 20, then it's equally likely that you succeed by a margin of +3 or +13. Even though you are much more likely to succeed if your modifier is +14 rather than +4, the distribution is still flat and unpredictable.

 

[Ancalagon]

[MENTION=6775031]Saelorn[/MENTION] True, but if that is a concern you can replace the d20 with 3d6 (or something like that). Probabilities around the average can be achieved without a dice pool no?

Sent from my SM-G930W8 using EN World mobile app