/smg/ System Mechanics General

Discussion thread for Dice Mechanics, Game mechanics, game theory, and anything else that makes games tick. Who knows? something brilliant may come of this.

How do you feel about Degrees of success, or systems that attempt at diversifying the Pass/Fail method? I rather enjoy the FFG Star Wars method, where Pass/Fail is simply the y-axis to the system, with the Threat/Advantage giving us an X-axis. It's a simple concept really, sure you may succeed at your goal, but was it actually useful and did it contribute towards your end-goal?

Other urls found in this thread:

anydice.com/program/9458
anydice.com/program/9459
anydice.com/program/945c
twitter.com/SFWRedditImages

What's the glaring flaw in this idea?
1. Combat doesn't use a map, any position and movement is abstracted like an old-school vidya JRPG, which I think is necessary for the unusual turn order mechanic.
2. At the start of the round, each character declares their action, in ascending order of Intelligence + Perception. Let's say a tie goes to the PCs at this step.
3. Then each character's speed stat plus a speed modifier from the type of action determines the order in which the actions resolve. All declared actions are locked in at this step, and if an action becomes impossible it just doesn't happen, unless the character has an ability that says otherwise.
4. Any effect of an action such as damage only counts after the action happens, which means if there's a tie in turn order, no one goes first and it's possible for two characters to kill each other simultaneously.
My gut says it won't work, even though I know it's ripping off an existing system, but I can't consciously see why.

Who else here hates d10 dicepools? Seriously, it's boring, and it's not even convenient. Why not d6? Everyone has a fuckton of d6. Then it's way too linear, sure, target number of successes, player number of dice, but abilities can't do anything with them, except give/take away dice, or add/take away failures. About half of Veeky Forums's homebrews could be improved if they weren't d10 dicepool based.

It's fucking complicated.

I gotta agree, too complex, especially if you repeat this every round.

I want to get some feedback from you guys regarding this card-based resolution system.

Game uses a deck of tarot cards instead of dice.

Every player is dealt a new hand of four cards at the beginning each scene.

You make 'tests' by playing at least two minor arcana cards of the same suit or value and adding their values plus your bonuses. If the total beats a difficulty rating assigned by the GM, the test is successful.
You score critical hits by playing a four-of-a-kind or a straight-flush.

If you have no like cards in your hand, or you want to save the cards in your hand for later, you can either choose to fail or attempt a blind test. A blind test is a test decided by turning up and playing the top two cards of the deck.

Major arcana cards are basically dead cards as far as tests are concerned, but are discarded as a cost to cast spells.

Exhausting actions, like running while heavily encumbered, may have a fatigue cost. This means that you must discard any one card in order to perform the action.

You may draw back up to your maximum hand size by performing the 'Breath' minor action during your turn (you get one minor action, major action, and move action per turn.) You can choose to discard any number of cards before drawing.

You can also 'Commune' with another player as a minor action, allowing each of you to give the other a card from your respective hands.

I was thinking of using suits to determine which hit location you strike; heart for torso, diamond for head, spades for legs, and clubs for arms.

Lowest card among those played for an attack test determines damage, but number of cards played on an attack test adds a bonus.

Too complicated?

Bump

d6 dicepools are just fine.
I think the problem is that there isn't much design space left in them to explore. Just about everything you can do with a d6 dice pool has been done already.
There's nothing wrong with that; it means that you can just focus on differentiating the rest of your system.
But then people have a creative itch they need to scratch, and only novelty will do it...

I like degrees of success, but FFG goes way too far. There only needs to be success, success & GM makes something happen, and failure (GM makes something happen). Then the GM can determine what fits the story best instead of having to wedge he dice roll into the story no matter what.

Look into an ORE game (personally I'm most familiar with Reign) for an example of this being done well. Reign has the benefit of combat being resolved in one roll (hence "ORE", One Roll Engine), which makes combat forgivably quick, even if you do have to calculate who goes when every time.

If there's one other big advantage it has on your system, it's that speed is based purely on your die result. Having players look up a speed modifier based on their action is going to add a whole lot of slowdown to your combat. Try to keep as many of your calculations on the table and out of the book as possible.

If you're looking for inspiration though, give it a look. I think it does a pretty solid job of accomplishing what you're trying to do.

just simplify it - i'm fiddling with some homebrew where it uses kind of an MtG stack - declare actions in reverse speed order, resolve in speed order/

I had a stupid idea where the resolution mechanic was you rolled large pools of d6s and kept 2 for your total, or 3 if you were a specialist in the field (Fighters kept 3 for combat-related rolls, for example). Characters were able to activate abilities if they spent a die with a face value of 4+ out of their pool.
Say, for fighting magicians, people with magical armor, or werewolves:
>Silver Blade:
>Cost: 3 4+, 2 5, or 1 6.
>Ignore any Defenses with the Mystic tag.
So if you were a Fighter and had a rolled pool of 6, 5, 4, 3, 2, 1, you could take a score of 15 against an enemy, or you could spend your 6 on Silver Blade and take a score of 12 instead.
Or a skill like this:
>Furious Blow
>Cost: None
>When you make an attack, add +3 to your total but remove the highest-rolling die. You must declare the use before you roll.
This has the interesting effect of boosting general damage output considerably while sacrificing the ability to use Skills consistently.
Am I retarded or does that not sound cool?

mathed out some dice rolls where you roll Nd6 and keep the top X a while back and now i don't have access to a mathematica license

On the other hand, this kind of system is such that it's less the GM'S Purview to decide every outcome, and more his job to pick the best situation presented by the players. whenever a major roll is made, we basically throw out ideas as the best way to interpret the roll, and the GM takes these suggestions into account. Narrative Dice System indeed, it achieves something no other RPG has: Roleplay doesn't come to a screeching halt the moment combat starts.

Neato. I ran some numbers myself but I haven't really sat down to figure how to extensively design a system based on Keep X yet.
Oh, and Furious Blow should be +4 or +5 damage, not +3. Man, what was I thinking?

Try Anydice.com. honestly, I should gather a couple of resources to post. Links to a couple of sites on game design, dice statistics, and maybe a trove with stuff like the Kobold Press Game design books. What says you Veeky Forums? Should I attempt to make /smg/ a fixture of Veeky Forums?

Sure, why not?
I'll just post stuff whenever I have ideas for my Keep X system.

fucking sweet, thank you.

Be sure to click on the button at the top to create a link, and share that with us, that's the best part of Anydice.

Well here' s the stuff for the pic in anydice.com/program/9458
and here wold be taking the lowest
anydice.com/program/9459

This has the potential to be really interesting, but I'd design it with a maximum of 6 dice ever being rolled at any given time. Maybe you could have an interesting Crit mechanic if you get all of a single number, or if you manage to get every number different on the die.

Anydice is pretty excellent for a whole lot of dice math, especially if you know a little basic programming.

Here's a simple Xd6 keep Y program:
anydice.com/program/945c

The base pool I had considered was a pool of 1 for untrained, 2 for intuitive, 3 for general, 4 for specialization, 5 for expert, and 6 for mastery.
The crit mechanic I had considered was any sixes get 'confirmed': Roll 4 or higher on any of the dice that rolled sixes and all sixes from before now count as 7s, or two sixes for spending on skills.
Say you roll 6, 6, 2 on a pool of 3. Now you roll 2d6. If either of them turn up as a 4, 5, or 6, you now have a pool of 7, 7, 2.

Yep. Anydice will be the meat and potatoes of /smg/ if it ever takes off. Any other resources, books, sites, or what have you that I can add? What are the essentials for system theory and design?

Dude, FFG just took Apocalypse World and made it more autistic. The roleplay comes to a screeching halt with literally every roll as you gather the dice pool, read the stupid symbols, and then figure out how it can apply to the situation. It's literally unnecessary, and most games don't stop the roleplay once combat starts - and nothing about the combat promotes roelplay aside from having characters get knocked out at the drop of a hat.