Limb-oriented skill system - You see any pitfalls?

[Antsan]

I am working on the design for my first roguelike right now.
So I had the idea to make a limb-based skill system that takes into account the physical composition of a creature. This should go beyond what's DF is doing but it should be easy to understand in principle, although it might be complex under the hood.

So far I've got only a system that may be appropriate for battle-related skills (because those should use the whole body).

The basic idea is to have limbs attached to each other in a hierarchical order, like a non-binary tree. The root of this tree should be the center of mass (one hand below your belly-button if you are a human). The nodes of this root are the different bodyparts where the connections between this nodes are joints which have a certain degree of freedom (between 0 inclusive and 100 exclusive in percent) - the joint to my upper arm is way more flexible then my elbow.
For every bodypart there now is a dextirity and a strength expressed through a number between -100 and 100 exclusive in percentages (so actually its between -1 and 1). Every attack is done through a special limb - a punch is delivered through your hand, a kick is done with your foot while you can choke someone with your upper or lower arm.
The chance-to-hit for an attack is calculated thorugh the limb you use to deliver the attack. The basic chance-to-hit is the dextirity of that limb. This chance is modified by the following rules:
Go one limb up, multiply the chance-to-hit with the joint rating between the bodyparts. If this is positive, this is the percentage that the original rating is moved into the direction of 100%. When this is negative, it's the percentage the chance-to-hit is moved into the direction of -100%.
This is done recursively to the center of mass.
The damage dealt through an attack is calculated similarly, except that the chance-to-hit of the next-upper bodypart is multiplied with (1 - joint-flexibility) instead of the joint-flexibility.

This seems calculation-intensive on first sight, but the calculated stats for the different bodyparts only change, when the strength or dexterity of a bodypart above the bodypart in question change or when the joint-flexibility of a joint above the bodypart in question change.

This system allows for different fighting styles which use different bodyparts in their attacks. You could even see weapons as just additional limbs with their own joint-flexibility, strength and dextirity. The negative ratings for strength and dexterity could be used to model pain or blockades.


Now this looks all shiny and nice to me, but there surely are some really nasty implications which I overlooked.
Some opinions on this?

[Etinarg]

I've used a tree-based appraoch for creature structures in a former project, too. From a technical point of view, it was quite mighty. Gameplay wise, I could not make much us of it, but that's not a fault of the system ...

Downsides:

- Some coding overhead for all actions involving the creatures body
- Quite some additional work in configuring such creatures
- Added complexity makes debugging more difficult

If you go IVAN++ then definitely, go for it. But unless your game makes really heavy use of the limb structure, it will first of all just slow down development, and will make all tasks involving creature bodies a bit harder. Make sure the gains in gameplay outweight the drawbacks.

I don't want to say it's good or bad. It's additional work, sometimes quite some, and it should pay off for you. Worst case is, to get frustrated about all the additonal work and give up in 6 months ...

[Krice]

It doesn't matter how you implement something, but it better be good in the gameplay. Some things may feel like great ideas in theory, but they just don't function properly in the game. I hope it will be fine.

[Etinarg]

I forgot to say in my last message that I found, there are only a few body structures which are commonly used:

- No limbs (snakes, worms, some fish)
- Naga type (2 arms, no legs)
- Humanoid (2/4 arms, 2 legs)
- Centauroids (2 arms, 4 legs)
- 4/6/8 legged creatures.

If you want to separate wings/tentacles from arms, you get a few more, but it doesn't bring something substantially new.

It's really difficult to work out a benefit of a tree structure over a list structure, or just a few sets of flags. The only real advantage that I see, is that a tree-structure can also implement joints, which list or set structures don't naturally keep track of.

[Antsan]

It doesn't matter how you implement something, but it better be good in the gameplay. Some things may feel like great ideas in theory, but they just don't function properly in the game. I hope it will be fine.

Thats my question exactly - is it worth considering as actually good gameplay or is it just an idea-blob among many others?

@Hajo:
I am using Common Lisp, this should make bugtesting really easy for rule-systems that can be modeled functional. And this idea seems to be born to be implementet in functional style.

If you go IVAN++ then definitely, go for it. But unless your game makes really heavy use of the limb structure, it will first of all just slow down development, and will make all tasks involving creature bodies a bit harder. Make sure the gains in gameplay outweight the drawbacks.

The limb-structure was intended to be used heavily. I even had the idea of managing mutations in this manner, maybe even involving such things as blood-systems (for humans this might be just one system that is attached to all bodyparts, but more fantastic creatures may use more), nervous systems (for distributing pain through the body onto several limbs when only one is hit, which may be positive or negative in effect) and such things, but those are most likely to be overkill anyway.
On the other side this could get some kind of mutant-centered game, where you build up your body by yourself (at least partially), so character development would be on a completely different level. This way I could cut out anything that cannot be modelled throught the limb-system and is not absolutely crucial for the character concept to work.

I forgot to say in my last message that I found, there are only a few body structures which are commonly used:

- No limbs (snakes, worms, some fish)
- Naga type (2 arms, no legs)
- Humanoid (2/4 arms, 2 legs)
- Centauroids (2 arms, 4 legs)
- 4/6/8 legged creatures.

If you want to separate wings/tentacles from arms, you get a few more, but it doesn't bring something substantially new.

It's really difficult to work out a benefit of a tree structure over a list structure, or just a few sets of flags. The only real advantage that I see, is that a tree-structure can also implement joints, which list or set structures don't naturally keep track of.

When I actually go the way described above a list wouldn't be enough.
The difference between different forms of bodyparts I wanted to model through tags. This way the system would be easely extensible. This way feet could be tagged as "move", hands as "grab", tentacles as "move" and "grab", the head as "crucial to survival", the heart as "at-least-one" and so on...

[Etinarg]

I feel uncertain on giving advice. For once, I was very enthusiastic about the tree-based limb system of my former project when I started it, but I really disliked it in the end. Lisp might make dealing with tree structures easier (at least recursion is the best way to traverse trees), so it might help you.

There are settings where such a system is needed very much - think of mecha games, all those big robots with their limbs, joints, and the weapons built into the different body parts. There such a system definitely is a key feature. This also can be done for organic creatures, too, just the words are different.

I can see that this will work in the right type of game.

[Z]

IMO if you center your game about that limb system, it might be OK. If it has not much importance for the gameplay, it is not worth it. 6 bodyparts is enough for a human. (IVAN has 7 and they can be transmuted into different materials, its system could be extended by allowing mutations of bodyparts etc., and be fun. Even if there is use for being able to polymorph your right ear into a bat ear, it might be better gameplay-wise to only allow polymorphing your head into a bat head, to gain echolocation at the cost of normal sight. That way you have an interesting choice.)

And, if you use this system for limbs, and similarly complicated systems for all other parts of the game, don't assume that you will ever finish it, especially if you are not extremely skilled/experienced. This is harder than it seems.

[Anvilfolk]

My opinion is sort of what everyone has said.

It needs to have a visible impact on gameplay. If you model this incredibly detailed character body and then in the end the combat that the user sees is pretty much what they see in all other roguelikes (you hit the bat or you don't hit the bat), then what's the point?

IVAN makes use of the limb system in a gameplay-affecting way. You can lose limbs. If it's an arm, you can't use weapons there. If it's a leg, you can carry less and move with more difficulty. Then you can pray, and get limbs made out of different materials, which is just awesome. I hated praying and getting a limb made of expensive fabric. but then, if you got mauled, you might be able to sell it for good money (never tried it, but I guess you could).

So, either you make an IVAN++, which is a lot of work... or you develop on your suggestion, which I'm inclined to support: a game based on mutations. Either make your own creature (à la Spore?), or let there be a LOT of mutations/limb-cutting going on.

A mechanic could be based on food: if you get a limb cut off, maybe you need to eat a lot to regain the "supplies" in order for it to regrow. Or for something else to grow (even if it's not been cut off). Maybe you never figure out what's gonna grow. Maybe there's toxic waste barrels all over the place that you can eat from, and they'll mutate you. You just don't know how. Could make for some funny stuff: "An arm grows out of your groin!".

Or maybe a limb just grows, and you have the option of turning it into either a foot, hand or other appendage.

Anyway, just ideas

[Fenrir]

Just to add to AnvilFolk's ideas, perhaps the player could graft the severed limbs of enemies to his own body, Frankenstein style.

[Vanguard]

I think most of the potential pitfalls here are in the ways the idea could be executed, rather than in the concept itself.

How would the player select which attack to use, for example?  Navigating a menu for each attack is tedious, most roguelike players like that they can attack with a single keypress.

How would strengthening various limbs work out?  What would happen if a character who improved everything equally fought a guy with one huge hulk-arm, and every other limb worthless and atrophied?

[purestrain]

It works only if you repetetive have to press keys to move you left and right leg to get a cell ahead...

Sorry; maybe its manageable if you have limbs, but don't event want to try to simulate muscles and so on. Lets just have some limbs with hit-percentages and location based wounds (which result in "Narben" (don't know the term; emboss? grain? scarred?) and are some sort of status symbol ;-)), that would be enough.

[Etinarg]

How would the player select which attack to use, for example?  Navigating a menu for each attack is tedious, most roguelike players like that they can attack with a single keypress.

In my old project that implemented a limb system, there was a list of possible attacks. Players chose one entry as default, and this one was taken for the usual "bump into enemy" type auto-attacks.

So attacks still took one keypress only, a switch of the attack took one or more keypresses though. People didn't switch often, but that was because there was usually a "best" attack for a battle, depending on the enemy. So people only switched once, fought the battle and then let it be until the next battle.

It can be done. Problem is how to make it interesting.

[Antsan]

Wow, whole lotta posts!

Would like to answer directly to you, but this would disrupt train of thought I had while thinking about your posts.

Bodyparts actually get an additional mass or volume and a material (the material would have density and thus you wouldn't have to store mass and volume). The damage rating wouldn't be calculated in terms of strength alone, but in strength' percentage of mass. Defense would be calculated by dexteritie's percentage of volume (in some way similar as described in my first post).
You could alter bodyparts now in two ways: Changing mass/volume or the material. An example for this: muscles way more then fat, so muscles provide more actual strength then fat.
Now any bodypart has to carry all the bodyparts below it, so the more mass outer bodyparts have the more the inner have to carry. This would eat away at the strength and maybe dexterity of those bodyparts (because they have to be used to manage swinging those big limbs). This would prevent overuse of massive limbs. Adding limbs in processs of mutation wouldn't be a nobrainer.
At the same time big organisms would become more dangerous to small creatures. On the other side they most likely will be slower when not very much thought has been put into them.
So one ressource to handle would be mass.

Another idea is to use concentration - your brain can only handle this much things at a time, so you wouldn't be able to get everything out of your resolution-enhanced eyes and your bat-like ears, because your brain cannot handle this much at a time. Usage of some bodypart (wether it is for input or output) would enhance the ability to use this bodypart (while eating away on the ability of things you didn't use in a long time).
The player could create sets of actions to concentrate on at a time. This way he could switch from one use of his abilities to another at the appropriate times. Smelling and listening for searching, sight and your big arm in your belly for battle.
I think the game should be playable without using those sets - they would make things easier, but shouldn't be absolutely required to survive.
Next ressource to handle: concentration.

To keep the mass and functionality of your bodyparts intact, you have to eat. Mechanical bodyparts could be cept intact through other means: oil, an electric energy source, something like this.
The bigger and stronger you get, the more food you need to stay this way. You will need it to grow. With the concept of materials we even get "You are what you eat".

This system should heavily depend on learning by doing. Player choices would be made on strategics, what to eat and at which mutations you aim. Frankenstein-style would be nice. There should be a difference to "normal" mutations, though. Maybe faster to get but eating more concentration? Any ideas?

Handling attacks: Hajos idea seems appropriate. Maybe we could expand this a bit (with optionality to the game in mind): Let users write behaviour-scripts for his game. "No go explore (use this pattern) until something unexpected happens (soething, thats not handled in the pattern)." Crawl comes into mind.

Opinions on this?

[Etinarg]

Now that finally make the race "amorphous blob" playable. A creature that can shape it's body depending on the requirements of the situation, but needs time to shift shapes successfully. Such a system is interesting because it provides a high number of combinations to explore, and if done right, there is no single "best" configuration, but a lot that excel at different tasks.

Maybe the player does no polymorph himself, but can summon helpers, and in the process he also creates the body structure for his summons?

A few ideas:

Damage usually is the result of impact, that is speed multiplied by mass. Many martial arts people try to maximize speed over mass, and are successful with that.

Also, animals have horns/rams, claws, spikes and teeth to fight. You might want to equip the creature(s) with such, too.

Add an "internal structure" attribute, that will serve two purposes: Once it determines how much mass outer limbs can have, until this limb fails to function. And in battle, internal structure will be a measure for the functionality for the limb chain. (E.g. if it's an arm, internal structure is the bones, and a broken bone renders an arm very much useless).

I feel uncertain about the concentration idea. Our brain handles a lot of tasks automatically for us, not only the low level stuff like heart beats, and breathing, the middle level stuff like walking or throwing, but it also filters a lot of sensoric input, and only the things that classify as "important" reach our conscious thinking. I think we can handle a lot of different input. What we are bad at, is trying to perform more than one task or activity at a time. In such cases our success rate at each drops quite a bit.

Thus instead of limiting input, I'd limit control. A human brain in a octopus shaped body will have troubles to control 8 tentacles in a sensible way. So maybe the more limbs there are to be controlled, or the more limbs are involved in an activity, the lower the rate of success becomes?

Also, a semi-fluid body at times will allow to pass through small openings, but at the price of loosing all equipment. Lots of ideas to explore there for sure. Good luck

Edit: spelling.

[Antsan]

Now that finally make the race "amorphous blob" playable. A creature that can shape it's body depending on the requirements of the situation, but needs time to shift shapes successfully. Such a system is interesting because it provides a high number of combinations to explore, and if done right, there is no single "best" configuration, but a lot that excel at different tasks.

Thus we arrive at a skill system above the one described before: One to measure the abilities of mutating in certain sections. How good are you at using different materials to alter yourself? Are you doing it fast, are you doing it reliable? What are the kinds of mutations you do? Additional Limbs, natural weapons, brain-extensions?

Maybe the player does no polymorph himself, but can summon helpers, and in the process he also creates the body structure for his summons?

I think with this idea I would stick to some dystopical post-apocalypse setting. Thus you would use radioactive substances to do mutation. The concentration system (see details below) could be used to handle wanted and casual mutation: When you concentrate on mutating yourself (when you eat mutagenic stuff willingly), you will have more detailed options in describing the mutations you want to have, while the detail in casual mutation (caused through background-radiation maybe) will most likely be a little bit lower.

Damage usually is the result of impact, that is speed multiplied by mass. Many martial arts people try to maximize speed over mass, and are successful with that.

This is actually a really nice idea, but I don't see how I could model it right now. I'm open for suggestions.

Also, animals have horns/rams, claws, spikes and teeth to fight. You might want to equip the creature(s) with such, too.

Attributes like "pointy" or "sharp" should be implemented as tags - the could modify the damage rating.
Another advantage of those bodyparts would be, that they are not normally connected to the bloodsystem, so you've got not this much risk in them being hit - they may shatter, but this won't eat away on the rest of your body, because you won't loose any blood.

Add an "internal structure" attribute, that will serve two purposes: Once it determines how much mass outer limbs can have, until this limb fails to function. And in battle, internal structure will be a measure for the functionality for the limb chain. (E.g. if it's an arm, internal structure is the bones, and a broken bone renders an arm very much useless).

This sounds nice.
"Internal structure" could be even done through bodyparts consiting of more than one material, one describing structure, the other describing ability.

I feel uncertain about the concentration idea. Our brain handles a lot of tasks automatically for us, not only the low level stuff like heart beats, and breathing, the middle level stuff like walking or throwing, but it also filters a lot of sensoric input, and only the things that classify as "important" reach our conscious thinking. I think we can handle a lot of different input. What we are bad at, is trying to perform more than one task or activity at a time. In such cases our success rate at each drops quite a bit.

This sounds good.
But remember: We're not talking about normal usage of your senses, this is using 3D-hearing with construction of environment in your conciousness just by sound. Doing this with sound and eyes and smell at the same time should be nearly impossible. When using up concentration the system shouldn't consider normal use for alertion in emergency, just for concentrated use, like really determining an objects position solely on sound.

Thus instead of limiting input, I'd limit control. A human brain in a octopus shaped body will have troubles to control 8 tentacles in a sensible way. So maybe the more limbs there are to be controlled, or the more limbs are involved in an activity, the lower the rate of success becomes?

Sounds nice. The more bodyparts involved in an action (whether it is input or output, see above) the lower the success of this action (statistically).

Also, a semi-fluid body at times will allow to pass through small openings, but at the price of loosing all equipment. Lots of ideas to explore there for sure.

This will get hard. Influencing the flow of a liquid (maybe based on where the temperature is between melting and boiling point) by an AI or the player sounds nice.
A high ability to change the aggregate state of your body (which had to be surely highly homogenous) would make nice shapeshifters.