What would its game effects be? Are you closing off (or making puzzle-y) certain actions once the PC’s passed a threshold value, or is it just a countdown until having to re-play a portion of the story?
Realism is overrated – never include something just because it is realistic.
Two ways you might want to include illness:
Include illness as a consequence of certain actions: drinking uncooked water, eating raw meat, et cetera, gives you a certain probability of becoming ill.
Use it as a narrative device to spice up the player experience: use a drama manager to make interesting things happen, and let illness be one of them.
Which is preferable depends on whether you want more simulation/challenge, or more narrative.
But the real design problem is to make illness interesting. You want the player to say: “You know, that one game, I got ill, and then I had to… !”, and be excited about it. (The first possibility that comes to my sick mind is: “You know, that one game, I got gangrene, and then I had to cut off my leg with a knife made from an old tin can!” That would be memorable.)
I’ve also been thinking about how to include some ideas about player health and injury in my WIP too.
One thing, I would make it so if your character is injured, the whole game does not then grind to a halt and become a tedious search for first aid, as in first person shooters. I would almost rather die and restart than be forced to search all over for med kits, so I could get back to the game.
In my WIP, I’m thinking about making it so if the player is hurt or injured (such as bleeding), if they stray into certain areas they may begin to attract the attention of predators who will stalk and hunt them, as predators will stalk wounded animals out in the wild because they’re slower and make easier prey.
Other characters in the game may react to you differently when your wounded too. Some characters may be sympathetic, others uncomfortable around the sight of blood, or even scared to go anywhere near you. Just a few thoughts…
I don’t know if this makes sense from a game design perspective, but as a player I don’t mind injury / illness provided they happen to someone else. If I get sick or crippled in a game, I’m likely to simply restore from an earlier save. The narrative consequences of that are not likely to be interesting in a way that improves my experience.
One of the big improvements in recent (post-Portal) FPS games has been the removal of the fixed health bar, in favor of a system where you only die if you take more than a certain amount of damage in a short period of time. This has made me far less likely to try to optimize each encounter, saving beforehand and restoring unless I succeed with most of my health and ammo intact.
It seems like you could apply that to your game and have sleeping and eating work as a general purpose fix. Then block access to either or both at certain times, such as when hunting. If the player is seriously injured while fighting a grizzly bear, the injuries can slow him down or attract predators while he escapes. But if he gets away then it shouldn’t be too difficult to recover, or else players will simply save before they go hunting and restore if it doesn’t turn out perfectly.
Dogs or horses, however, should bear the full weight of the player’s bad decisions. I am much more likely to undertake risks or deplete precious resources on behalf of an animal companion, especially if it’s straightforward to acquire a new one. If you only get one dog, I would just restore from a saved game if he died. However, if you can tame a new one, I would try to keep him alive for emotional reasons and accept his death if it came to that.
In a world of SAVE and UNDO it’s very difficult to make procedural punishment look like anything other than a mistake. Narrative injury, wherein writing makes it clear that this cost is unavoidable, is more or less the only sort the audience will accept. Even if they do accept it, it’s a wildcard complication in difficulty assessment. Each further challenge must be tailored for every possible combination of injuries, or the player could reach a point many turns down the line where they discover the game’s either tediously complex or outright unwinnable.
Assuring that a character is precisely as injured as the story requires makes it much easier to maintain player interest and excitement.
First thing off the top of my head is, in real life, if you’re really alone, injury and illness can mean death a lot more often than most of us experience in everyday life. We depend on people helping us when we’re incapacitated.
So if it helps the story (or the simulation, or whatever) to have a friend or lover or family member take care of you when you’re sick or injured, then maybe it’s worth having. Otherwise, you might consider ending the story as soon as the PC is too beaten up to go on.
Well, I think it might be kind of cool to play with injury and illness in the kind of game you’re envisioning (as I understand); if it’s more of a sim than a majestic procession through various challenges, then it might be interesting to have illness and injury. OTOH, I play roguelikes for fun. I don’t play Dwarf Fortress, though, so there’s a limit to my masochism; which I guess means I’m with Ben, in that I like being able to recover from it.
The basic question, as Victor has basically already pointed out, is not is it realistic, but can you make it cool? Dwarf Fortress maybe is something to look to here – as I understand it, a lot of the appeal is the way things can go horribly, hilariously, awesomely wrong. (Boatmurdered!) If you can turn injury and illness into that, awesome. If not, it could be cool to make it recoverable in a cool way – one of my nicest moments in a certain vaguely roguelike boardgame came when a character just barely managed to survive a boss encounter (running away from the remaining mooks) and had to camp out for a week to recuperate. But it didn’t actually take me a week to do it.