PDA

View Full Version : Would these mechanics make any sense?



tedcahill2
2018-05-15, 09:39 PM
Like in Shadowrun I'm using a Attribute + Skill dice pool system for a little custom project I'm doing. Here's how things will interact together, and does this make sense?

Attributes will range from 1-5 for humans, or up to 7 non-human races. Skills will range from 0 to 7 (not set on 7 for max skill rank, but it's where I'm starting). When making a skill check (which includes attacks and defenses and all that) you will succeed on a 7+ on a d10. You can only count a number of successes up to the linked attribute.

Example: Jaxx has a STR of 5 and an athletics skill of 4. He roll's 9d10 and score 6 successes. Because his STR is only a 5 he can count 5 of those successes.

Now that part is simple, and I think will work fine, and is logically sound. Now what about magic and gear? I don't really want to have a system that just adds dice on dice on dice and you're rolling 25d10. So I was thinking that magic and gear can have augments ranging from +1 to +3. Any combination of spells and gear can never exceed a bonus of +3.

Instead of that bonus simply giving you another 1-3 dice into your dice pool, what if it changed the target number for a success. So if you have a +1 strength spell cast on you, and you roll a STR related skill check, your target number to count as a success is 6+ on a d10 instead of 7+.

I'm trying to rationalize it working that way, that a magic strength buff makes STR tasks easier, but doesn't add dice? Does that make any sense? Statistically does it make any sense for it to work that way?

Another option: instead of magic and gear adding +Xs it would be simplified to either having a STR buff or not. If you have a strength buff you succeed on 6+.

JNAProductions
2018-05-15, 09:46 PM
Statistically, you've got a 40% chance of success per die on a 7+. Increase by 10% per point of magic bonus, to a max of 70%.

And it makes sense, if magic enhances natural abilities, but does not add new ones. Because magic items make it more likely for you to succeed on tasks you can succeed at (say, Strength 4 and you need 3 successes) but does NOT allow you to succeed at something you can't normally succeed at (Strength 4 and you need 5).

Cosi
2018-05-15, 09:57 PM
I think you should step back and think about the outputs you want the system as a whole to produce. It's very hard to say where having (what sounds like) Shadowrun 5e style Limits will do what you want without defining what it is that you want. How big of a range do you want? How competent is a random person at an arbitrary task? How often should a trained expert beat them? What should decide competitions between experts? How easy should it be for an expert to succeed on a task that is impossible for an untrained person? Do you want to have multiple jumps like that? I think those are the questions you should ask, and they are probably heavily dependent on what the game you are trying to make is.

Also, I think for dicepool games d6s are probably better than d10s. People have lots of d6s in a way that they do not have lots of d10s. This is somewhat mitigated by you capping dicepools at less than twenty, but I still think d6s are better unless you have some reason for preferring d10s.

Bansheexero
2018-05-15, 10:03 PM
A system I've been working on uses alterations to Target numbers, but I also have natural 10's count as 2 successes. It's a d10 system. You claim that you keep x number of successes based on an attribute, if you roll a natural max on a die does that have any additional effect? If not, it could, then you can alter not only the range that triggers that effect, but it's magnitude as well (like it counting as 3 successes). I say this because changing a target number can really shift probability in practice. Even in the craziness of 2nd edition Exalted, there were only like one or two charms in it's entirety that affected target numbers.

John Campbell
2018-05-16, 12:39 AM
What you're describing sounds a lot like the core mechanic of the older Shadowrun editions - which is more complicated than the newer editions, but a much more elegant and generally functional mechanism.

Up through 3E, an ordinary test in Shadowrun had three different axes of variability: number of dice, target number, and number of successes.

Number of dice was used to represent skill at a test. If you were making a skill test, you'd get your skill rank in dice, so 1 die for a beginner, 6 or more for an expert. In a raw attribute test, like a Body test to resist damage, you'd get the attribute rank, so a fragile character might get only 1 die, while the cybered-up troll brick might get a dozen. (And then there were pools that you could pull from to add dice to a specific test, but that's outside the scope of the current discussion.)

Factors that makes you better or worse at doing stuff change the number of dice you get. Usually this just means your relevant skill rank, but other things can do it. For instance, many shamans are better at handling some types of spells or spirits, and worse at others, because of their totem affinities. They get bonus dice on the former, and lose dice on the latter. Or magic weapons actually make you better at fighting, and add bonus dice to your melee tests.

Target number was used to represent the difficulty of the task. A trivial task, the easiest things you'd have to actually make a roll for, would be TN 2. (1s are always failures.) A really insanely difficult task, like purchasing a stealth tank on the black market, might be TN 31. (Shadowrun dice explode - if a die comes up 6, you roll it again and add the new result, and keep going as long as you keep rolling 6es - so it's, at least in theory, possible to hit any arbitrary target number on a d6.)

Things that make the task easier or more difficult would change the TN. For example, the base TN for shooting something at short range with a pistol is 4. But if you've got a smartlink on it, that's a -2 bonus, so it drops to TN 2. But maybe other factors like lighting conditions or cover might raise it again, so you'd be rolling your Pistols dice against maybe TN 5.

And then the number of dice you get successes on represent your degree of success. Usually only one success was required, but often more successes would improve the quality of your success - you'd do more damage with an attack, or reduce the asking price of that black market stealth tank, or finish whatever you're doing more quickly, or so on. Occasionally some test would require you to succeed really well to succeed at all, and require more than just 1 success for even basic success. For example, 1 success is not enough for the Control Thoughts spell to work; you need successes equal to half the target's Willpower. This was unusual, though.

More usually, this would come up in opposed tests - say, someone's shot at you with their pistol, and gotten 4 successes on the attack. You want to dodge, so you roll your Combat Pool dice (against TN 4, usually), and you've got to get 4 successes to match theirs in order to successfully dodge. If you only get 3, you didn't dodge, and you've got to soak a gunshot wound. Though your 3 successes do subtract from their 4, so they don't succeed as well and you have less damage to try to soak.

Florian
2018-05-16, 01:42 AM
Hm, so basically Roll And Keep, just with generating "hits" at this point. Be advised that this puts an extreme emphasis on attributes over skills, because every XP point spent to advance an attribute will simultaneously advance all skills as well as raising the keep value of the checks.

Why D10 instead of D6? Having played both, Shadowrun and Vampire, D6 are easier to handle.

On magic gear, IŽd say: Don't go into numeric/mechanical magic gear at all. Nothing is more boring than a D&D character only decked out with the most cost-effective and function +X stuff in any slot. Don't go there.

Cespenar
2018-05-16, 05:01 AM
I'd say just give extra successes instead.

Floret
2018-05-16, 09:56 AM
For limiting possible dice, saying "everything above x dies converts like this" (or simply being careful with how many bonuses you make possible) works.
Beyond that, magical gear might allow you to raise your limits, to modify the number of necessary successes, generate additional ones, or modify the number of dice you get.

More "out there" suggestions might be to have it unlock exploding dice (highest number gets rerolled, potentially infinite successes); or even (especially with a fixed TN) modify dice type, allowing you for example to roll d12 instead of d6. (At a success of 5+ mathematically beautiful, btw. Especially when you also include exploding dice.)

I second everything Cosi has to say, btw.

One thing I personally would avoid is to modify the target number. Because
A, I hate it with a passion when systems do that,
B, it makes it just so much harder to see at a glance what is a success and what isn't,
C, you make the GMs job just that much harder by giving them yet another number to modify when calling for tests - either to be recalled from yet more tables explaining what modifier is supposed to be what, and never quite managing to clearly draw the lines anyways, or having to make it up on the spot,
And finally D cause you don't even need yet another thing to modify in the first place. Substracting and adding dice, as well as modifying how many dice are needed is really enough. Noone can calculate that much math on the fly.

A thing I notice about the way you drew up the numbers is that noone will be able to ever get more than half their dice as (counting) successes. Humans even less so. I would be wary of limiting characters that much - having to discount successes is one thing, this intensity could make it a big source of frustration.

tedcahill2
2018-05-16, 10:37 AM
A thing I notice about the way you drew up the numbers is that noone will be able to ever get more than half their dice as (counting) successes. Humans even less so. I would be wary of limiting characters that much - having to discount successes is one thing, this intensity could make it a big source of frustration.
So under this system 10 hits are needed to complete a task with a nearly impossible difficulty. I did not want someone, no matter how highly skilled, to be able to achieve near impossible tasks with any consistency.

So using attributes as a limiter felt necessary to both control the tasks that could be consistently achieved as well as creating a dynamic of "It doesn't matter how highly trained in sneaking you are, you are still physically limited by your agility score. In this case your training helps make the most of your natural limits, but at the end of the day, to get better at sneaking you need to raise your agility, thus raising you limit.

Additionally, this is a fantasy game, so chances are you'll want to attempt some heroic things. Enter the edge stat (name may change). Edge is a special statistic, and functions like a luck stat or inspiration points. For example, if you spend a point of edge when attempting a skill check you can ignore your limits for that one check.

tedcahill2
2018-05-16, 10:39 AM
Hm, so basically Roll And Keep, just with generating "hits" at this point. Be advised that this puts an extreme emphasis on attributes over skills, because every XP point spent to advance an attribute will simultaneously advance all skills as well as raising the keep value of the checks.

Why D10 instead of D6? Having played both, Shadowrun and Vampire, D6 are easier to handle.

On magic gear, IŽd say: Don't go into numeric/mechanical magic gear at all. Nothing is more boring than a D&D character only decked out with the most cost-effective and function +X stuff in any slot. Don't go there.

If I stick with a static threshold for hits I'll probably go with 5+ on d6's. The reason I considered using d10's was for some granularity if I wanted to change the target threshold up or down.

tedcahill2
2018-05-16, 10:41 AM
I'm actually kind of liking the idea that magic items don't add additional dice, nor to they change the target threshold.

I think I could make magic feel truly powerful if it actually added successes to your rolls instead of just dice.

So gloves of strength +1 would add 1 automatic success to any strength based skill you use, maybe also increasing the normal limit as well.

Florian
2018-05-16, 02:31 PM
I'm actually kind of liking the idea that magic items don't add additional dice, nor to they change the target threshold.

Delay such design decisions until you really tackle your magic system.

Ive already mentioned Splittermond as a system, which uses skill-based magic. Broadly speaking, spells fall into two categories, burst and sustained mode. A third level fire spell could either enhance a normal weapon to be a flame blade, or create a fireball, cost is the same. Magic items in this system are either geared towards providing the spell (a sword with a permanent flame blade on it), or are geared towards enhancing spells (a sword that will always do max energy damage, when a x blade spell is cast on it).

meschlum
2018-05-20, 11:40 PM
Whee! Math!

Ahem.

Let's look at the basic outputs of this system, shall we? Nothing but humans, no magic, everything bland.

Dice Pool 1: 60% chance of failure, 40% chance of 1 success.

Dice Pool 4: 12.96% chance of failure, 34.56% chance of 1 success, 34.56% chance of 2 successes, 15.36% chance of 3 successes, 2.56% chance of 4 successes, possibly capped by attribute.

Dice Pool 7: 2.8% chance of failure, 13% chance of 1 success, 26% chance of 2 successes, 29% chance of 3 successes, 19% chance of 4 successes, 9.4% chance of 5 successes (capped) (7.7% chance of exactly 5, 1.7% chance of 6, 0.16% chance of 7).

Dice Pool 10: 0.6% chance of failure, 4% chance of 1 success, 12.1% chance of 2 successes, 21.5% chance of 3 successes, 25.1% chance of 4 successes, 36.7% chance of 5 successes (capped).


So with a target of 1 success, most people (dice pool 4) will usually succeed but failure is still an option (1/8 chance, which is enough to be risky in the long run. Missing one attack? Fine. Not being able to build a mundane object, swim across a river, etc.? Dangerous).

A target of 2 successes is impossible for anyone with an attribute of 1, and succeeds a bit more than half the time for most people (dice pool 4). It's likely to work but still a risk for experts (dice pool 7), with about 1/6 chance of failure, and fairly easy for grandmasters (dice pool 10), with 1/20 chance of failure. This is a hard challenge to adjudicate, because it's something normal people can do but fail often and experts can manage but find difficult nonetheless in the long run. A normal person has around a 2.4% chance of beating a grandmaster.

A target of 3 successes could be beyond the capability of many people (attribute below 3), and is basically a step up from difficulty 2: most people either can't do it (low attribute) or find it difficult (1/6 chance of success), experts can do it about half the time, and grandmasters have a 1/6 chance of failure. It's a nice cutoff in some ways for challenging tasks - it's possible for most, but the skill differences affect how it's viewed. It still means that a normal person has a 3% chance of beating a grandmaster - higher than before!

A target of 4 successes will rarely be possible for most people due to attributes (and succeeds 1/40 if it can be done), so is the sort of task that you can manage without training if you're naturally capable, but for which no amount of training can overcome attribute limitations. Hard to imagine, but so it goes. Experts have a bit more than 1/4 chance of success, grandmasters around 60%. So it's a sort of step up in that it's now the grandmasters who have around 50% chance of success, though finding appropriate tasks is difficult. Also, someone with no skill and an attribute of 4 can beat a grandmaster around 1% of the time. At least it's lower than before!

A target of 5 successes is impossible for most, around 1/11 chance for experts, and around 1/3 for grandmasters. The type of task that even a grandmaster is unlikely to manage, though it's feasible, and that experts will rarely pull off.


Edit: also note that almost the entire difficulty spectrum happens at levels which are challenging to most up to challenging to grandmasters. So you won't have any gradation in difficulty for ordinary tasks - a lock is either mundane or master class, a race is either open to everyone or restricted to Olympic athletes. A bit of an exaggeration, but still has to be kept in mind when defining what tasks to roll for.


Mechanically, increasing an attribute is always better than increasing a skill - the maximum number of successes increases, with identical changes to the dice pool. Plus, there are typically a lot more skills than attributes, so increasing an attribute makes you better at many things where increasing a skill makes you better at only one. You could try to balance this with different costs (making skills cheap to increase), but it's still a numbers game. If there is an attribute that is associated to 8 skills (e.g. D&D Knowledge skills), skills have to be incredibly cheap to be worthwhile. And if there is an attribute associated to only 2 useful skills (e.g. dodge and melee), there is no reason for anyone not to be hyper trained in those skills as it's really cheap!

This type of problem has turned up in a lot of games, and typical approaches are either to make attributes more expensive (some games make it impossible to raise attributes, others just make it cost more) or to use skills as the cap (e.g. ShadowRun 3 where attributes set the maximum easily achievable skill). There are also systems that ignore attributes entirely (Fate), or do the same to skills (Amber).

Floret
2018-05-21, 03:32 AM
This type of problem has turned up in a lot of games, and typical approaches are either to make attributes more expensive (some games make it impossible to raise attributes, others just make it cost more) or to use skills as the cap (e.g. ShadowRun 3 where attributes set the maximum easily achievable skill). There are also systems that ignore attributes entirely (Fate), or do the same to skills (Amber).

One thing I once did in a homebrew I alluded to above: A fixed TN, attributes applying to more tests, but skills giving you better dice (D6 for attributes, d12 for skills in my case because maths). Makes the choice between improving one area more reliably, or lots of areas less so (Though with the same potential).