![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
Crunchy game mechanics
May. 10th, 2023 10:46 pmI'm not really a fan of crunchy RPG systems so the following isn't really a proposal. Just some musing on realism.
Most Role Playing Games use dice for determining whether something is a success. The general principle is skill have numbers. If you want to do something, you pick a skill, roll an appropriate assortment of dice, and either add to skill and compare with a target, or compare with skill (usually roll under). It works, but for modelling realism it's a bit simplistic.
The main problem is that the mechanism is typically the same for all stats. It works for a game but if you think about it it's a little unrealistic. Lets look at some common skills:
The party reaches a door. It's blocked by a boulder. The hefty barbarian goers up to it, rolls a die, and fails. Another character, half his strength has a go, rolls really high, and succeeds. Narratively this makes no sense. Strength should really just be a raw parameter, like height and weight; the amount a character can lift. Multiple party members can combine strength.
They enter an archery contest. This is something that a highly skilled character should win, but, in the case of a severe fluke, an unskilled character might hit a bullseye and a skilled character might just miss. This can be modelled with a single die, although simple addition doesn't really model the probability curve very well.
Modelling this mathematically, we'd have a bell curves. A skilled archer would have a tall narrow curve. And unskilled person with a bow will have a low flat curve. A blindfolded idiot with no idea where the target even is will have a completely flat distribution. Of course the problem here is that to do this accurately we need to use something like the Box-Muller function which is not something we can do with dice. Something that might work a little better is rolling multiple dice - a number depending on skill level, and taking the highest. The downside to this is that for smaller dice, e.g six sided dice, you'll roll a lot of high numbers which lacks granularity. Could roll multiple d20s perhaps but that feels rather clunky.
So the next challenge they face is knowledge. The Call of Cthulhu mechanic really works pretty well here. You have a knowledge stat - 0-100. Roll under it with percentage die to determine whether you know a specific piece of information. There will be some obscure trivia that people might know, and well known facts that might, for reasons have slipped the expert's mind.
The other one I want to look at is magic. I always find magic is a little too mechanical. It should be unpredictable!
A skilled mage, throwing a fireball will be able to go for exactly the level they require, at a precise target. An amateur will have much less predictable results. Possibly much more powerful than expected, and possibly with a lot less range. Feather fall - get it wrong and you end up falling too fast, or falling upwards. Giant insect - if it works too well, then it gets messy. Too weak might result in giant insects that don't feel like obeying commands. The idea that comes to mind is that you have a target range for each spell, and you roll a die (perhaps the mage can select the type of die). The skill is the modifier that can be applied after the roll. This is one that I feel will need a lot of play-testing and balance. The other option is the FFG Star Wars mechanism where dice have symbols that indicate not only success and failure, but also "advantage" and "threat".- that is you can succeed but something nasty happens, or fail but something good happens
There are other traits that I have no idea about. How do we do perception? Ideally we don't want the players to know if they fail their perception roll. The mere act of asking for a perception indicates there's something to perceive. But hidden rolls are unsatisfying. Rolls should be based on active player choice, not passive events.
Then there's charisma based skills. How should that work? There has to be something more organic than "You succeed. The other character likes you".
So that's some random musing on making things far too realistic. Perhaps a good idea for a computer based system, where the machine does all the calculations, and things can be as crunchy as they need to be.
Ideally though it would be nice to rely more on "input randomness" than "output randomness". That is, the randomness comes before player choice. An example might be that you roll a die, and have that many action points. Some board games do things this way. This is the typical mechanic with card games. It might be applicable to TTRPG.
There are a whole bunch of other game mechanics that might apply. Bluff, risk, (liars dice combines both and a simple version can be used for conflict resolution).
So what's the conclusion? There isn't one. This is just some random thoughts about balancing realism and crunchiness of a role-playing game.
Most Role Playing Games use dice for determining whether something is a success. The general principle is skill have numbers. If you want to do something, you pick a skill, roll an appropriate assortment of dice, and either add to skill and compare with a target, or compare with skill (usually roll under). It works, but for modelling realism it's a bit simplistic.
The main problem is that the mechanism is typically the same for all stats. It works for a game but if you think about it it's a little unrealistic. Lets look at some common skills:
The party reaches a door. It's blocked by a boulder. The hefty barbarian goers up to it, rolls a die, and fails. Another character, half his strength has a go, rolls really high, and succeeds. Narratively this makes no sense. Strength should really just be a raw parameter, like height and weight; the amount a character can lift. Multiple party members can combine strength.
They enter an archery contest. This is something that a highly skilled character should win, but, in the case of a severe fluke, an unskilled character might hit a bullseye and a skilled character might just miss. This can be modelled with a single die, although simple addition doesn't really model the probability curve very well.
Modelling this mathematically, we'd have a bell curves. A skilled archer would have a tall narrow curve. And unskilled person with a bow will have a low flat curve. A blindfolded idiot with no idea where the target even is will have a completely flat distribution. Of course the problem here is that to do this accurately we need to use something like the Box-Muller function which is not something we can do with dice. Something that might work a little better is rolling multiple dice - a number depending on skill level, and taking the highest. The downside to this is that for smaller dice, e.g six sided dice, you'll roll a lot of high numbers which lacks granularity. Could roll multiple d20s perhaps but that feels rather clunky.
So the next challenge they face is knowledge. The Call of Cthulhu mechanic really works pretty well here. You have a knowledge stat - 0-100. Roll under it with percentage die to determine whether you know a specific piece of information. There will be some obscure trivia that people might know, and well known facts that might, for reasons have slipped the expert's mind.
The other one I want to look at is magic. I always find magic is a little too mechanical. It should be unpredictable!
A skilled mage, throwing a fireball will be able to go for exactly the level they require, at a precise target. An amateur will have much less predictable results. Possibly much more powerful than expected, and possibly with a lot less range. Feather fall - get it wrong and you end up falling too fast, or falling upwards. Giant insect - if it works too well, then it gets messy. Too weak might result in giant insects that don't feel like obeying commands. The idea that comes to mind is that you have a target range for each spell, and you roll a die (perhaps the mage can select the type of die). The skill is the modifier that can be applied after the roll. This is one that I feel will need a lot of play-testing and balance. The other option is the FFG Star Wars mechanism where dice have symbols that indicate not only success and failure, but also "advantage" and "threat".- that is you can succeed but something nasty happens, or fail but something good happens
There are other traits that I have no idea about. How do we do perception? Ideally we don't want the players to know if they fail their perception roll. The mere act of asking for a perception indicates there's something to perceive. But hidden rolls are unsatisfying. Rolls should be based on active player choice, not passive events.
Then there's charisma based skills. How should that work? There has to be something more organic than "You succeed. The other character likes you".
So that's some random musing on making things far too realistic. Perhaps a good idea for a computer based system, where the machine does all the calculations, and things can be as crunchy as they need to be.
Ideally though it would be nice to rely more on "input randomness" than "output randomness". That is, the randomness comes before player choice. An example might be that you roll a die, and have that many action points. Some board games do things this way. This is the typical mechanic with card games. It might be applicable to TTRPG.
There are a whole bunch of other game mechanics that might apply. Bluff, risk, (liars dice combines both and a simple version can be used for conflict resolution).
So what's the conclusion? There isn't one. This is just some random thoughts about balancing realism and crunchiness of a role-playing game.
![[livejournal.com profile]](https://www.dreamwidth.org/img/external/lj-userinfo.gif){kind=small}
