I would absolutely say that this is doable. Of course, at the same time, it would also be extremely difficult from a design perspective. As other people said, if you were to implement what you're suggesting just as an actor behavior without any other design changes, it couldn't ever work, because people would just act the same. Imo the key to getting this to work is (of course) to design the %$#$ out of it, and make sure every element works precisely with the others.

You ask if players are "tainted" and would simply ignore the monsters' behavior. That depends entirely on you though, and the game you make. The core way to do so is to follow the advice of several other people, and design a reward/penalty system to encourage the behavior that you want... which also could mean several different things.

One type of reward system is similar to gninjagnome's comment, that the player receives what we might call the "Traditional" game rewards of <blank> points, loot, whatever for the behavior you want, rather than slaughtering everything. This could work, certainly, and given robust enough gameplay could come out well, but it isn't my favorite solution, because I don't feel like it addresses a deeper design goal, and instead just alters the focus.

A closely related option would be to make it tactically foolish to fight every creature you come across, which I think has the upshots of being both a little bit of a deeper mechanism, and encourages multiple types of problem solving. Essentially (and this is still what other people are saying), make the rewards for killing a creature lower than the risk for doing so. For example: The ogre is bugging you, but what if it's really powerful, and if you piss it off it'll kill you almost for sure? Instead, the player is tasked with finding another way to "lose". This requires more of the designer than the previous option.

My personal favorite stems from the idea that really, what you're saying you want is for players to interact with the game environment, rather than just pass through it on the way to killing monsters. To me, the single best way to do that is through immersion. If you can get a player to become emotionally invested in the idea of your world, it becomes something they want to explore, wonder at its marvels, and to stop and smell the roses, such as it is.

Off the top of my head, I suggest you go play Shadow of the Colossus. This is a game where you are explicitly rewarded for killing the big monsters, indeed, that's the ONLY way to progress. Yet at the same time, the worldbuilding and design of the game is so successful that you don't want to.

After I read the book, “Jurassic Park” (bought it right after I left the theater) I had an Idea for a game where you played the game warden responsible for the welfare of the dinosaurs.

You had to perform tasks suck as sedating the dinos, collecting their eggs or samples of poo, tracking and taking photos.
I remember being really into nature shows and wanting the game play to include all the elements that real game trackers have to take into account. (Such as wind direction, mating and birthing seasons, migratory patterns.)

This was like, 1993, I was teenager and making a video game was about a realistic as becoming an astronaut. But I’ve still never seen a game like that. (If anyone has, I’d love to know)

But to get to the point; Make some of your creatures an investment for the players. Let them know that they could potentially be “mindlessly slaughtering” the goose that lays the golden egg.

Even if, like the dinos in my idea, your monsters have no interest in whether the player lives, dies, or is their next meal; making the player responsible for the creatures welfare could make for some interesting game-play.

Thanks for all the responses and sorry for the late reaction, but I worked the whole weekend on the behaviour to prepare some testing.

Quote:

haegarr: Another aspect is that of profit. As long as slaying life is rewarded I'm sure that the player will do so. Rewarding a player when s/he let monsters alive is difficult. So in general you have to make it not worth the trouble. That means in a first place that indiscriminate slaying must neither grant experience points nor appreciable loot.

Quote:

GninjaGnome: If you're designing a game that uses a system where wolfs drop loot and gold when they die and players get experience points for every penguin they kill; I don’t think it will work.

This will be always a problem, "punishing" the player for killing a mob could be a problem. But on the other hand, it could be rewarding to let monsters life. Not, because they will thank you, but because they can do things you are not able to do.

@samoth & Konidias
I don't know Ryzom, but I know that a MMORPG has to be more restrictive than a single player game. So, my game will be SP at first, but I will take a look at
Ryzom to get a better understanding of how they introduce certain behaviours.

Quote:

AngryWyrm: A monster that's going to steal my loot? Kill it. A monster that causes me grief and trouble? Kill it. A monster that's going to attack me when I turn my back? Kill it.

If you will do this in my game.... it would be absolutly legitimate. Why not, if this is your way to play the game, the game should not punish you for doing
so. But there're other players who want to research other ways of playing the game, who want to discover certain behaviour etc.

@Rycross:
Kill each monster on sign could be a problem. In my game, monsters and their according "hive" has a shared knowledge of the world. They learn about the map, about obstacles and about dangers. If you kill some monster on sign, you should be fast, if it flees it will "tell" others where it encounters some dangers. This
results in more dangereous monster who will check for trouble. It depends all on your style of playing: want to hack'n'slay, than be good in it, more sophisticated is to poison a mob or hive, or support them to help you fighting an other more dangerous "hive".

@Pete
Sounds interesting, I will take a look at Abe's Oddysse.

@mrchrismnh
I think that I will target more of group behaviour than on individual behaviour, but I will keep it in mind, you never know :)

Quote:

kovalai: A closely related option would be to make it tactically foolish to fight every creature you come across, which I think has the upshots of being both a little bit of a deeper mechanism, and encourages multiple types of problem solving.

I think this should be the goal, give the player multiple options to solve a problem, but never punish him for doing other.


My game plays in a dungeon, each "level" is inhabitat by different "hives" of monsters. The monsters has to learn about the level and will probably fight each other. Depending on the environment (hostile, enough food), hives will develop in different directions. In a hostile environment more dangerous monsters will be "bred", in a none hostile environment with enough food, more health monsters will be bred and the hive will expand.

Well, if you think, that is nice, but why should the player even take notice ? The trick is, that different monsters have certain resources and can generate certain resources. Crafting is a major goal in our game, but you are not able to "generate" resources, but you can "farm" them. To farm, you have to manipulate the "hive" by using their behaviour to your own benefit. Fight hives to provoke tougher opponents to gain certain resources from them (scales, hard leather).
Or there will mushrooms you can't really use, but certain creature are able to eat and digest them to produce a secretion they will get to their nest. As player you could harvest this nest to get the secretion to create a healing potion from it.

Still, if you want, you can kill everything on sign, or at least try it, my goal is, to give the player really multiple options to play the game.

--
Ashaman

I've seen put it into action before with relatively poor education. The problem is not the mechanic, but how it is implemented.
You have to "tutorialize" the player enough so that he understands that you have distinct "enemy" AIs (from what I understand, they aren't exactly enemies, but mostly living creatures with relative emotions, including a wide array of aggressivness levels).
Be aware, however, that if you teach the player too much, the mechanic will loose its interest. So basically, tell them just enough for them to want to explore that mechanic on their own, but by all means, don't spoil the details.
And don't make this your core mechanic, just a supporting one.
It did work out great for C-man's oblivious dark, as far as I can recall. And it *was* in Anthelos 3.

If you're willing to go that extra mile, it would be interesting to see these "enemies" interact with one another. Part of where The Elder Scrolls IV failed was in tuning down their AI system (but surely for a gameplay reason). See how far you can go with the mechanics, and how it improves gameplay. If it doesn't, then your prototype will answer why this mechanic is that rare, although, it is my opinion that the reason we don't see more of this from the industry is merely because of the funds it would require to define such clever AIs, versus how much it would actually profit the companies.

Work it like bullfighting. Give the player more experience and mob-specific "lore points" for interacting with a critter without having to kill it. If you go toe-to-toe with a gorilla and win some kind of minigame that de-escalates the situation, it gives you a lot more XP than shooting the thing would have, and it also gets you some "Gorilla Points" that can either unlock gorilla powers, or help you befriend gorillas, or prevent gorillas from attacking you in the future.

Mix that into a system where some monsters just straight-up can't be killed without a serious party, and a guy who's able to get across Rhinoland without getting trampled has an edge over everyone else.

Iron chef's idea is actually neat.
If you can prototype a mini-game for every AI you want to interfere with and list a series of upsides to using it in a core game, you can end up with something very neat, although the thing is it would take a lot of design space and it would not be recommended to work with a lot of outsider elements besides your AI interactions.

what about the monster's companions?
imagine Metal Gear Solid, or Assassin's Creed. if you are caught attacking a hostile/guard, more guards will come, the alarm will sound, etc. At first it can be fun (specially if you like the combat system), but at some point you realize it's bothering and time consuming, so it's better to keep the stealth.

of course these are not RPG's, and you pretty much get nothing out of the corpses, so it still makes sense.

if you could have a wolf's pack hidden somewhere in the trees until the curious one is attacked, make the fight pretty tough, and have their corpses only give you "useless small chunk of wolf meat" and veeeeeeery little XP, it could work.

Quote:

Qrymus You have to "tutorialize" the player enough so that he understands that you have distinct "enemy" AIs (from what I understand, they aren't exactly enemies, but mostly living creatures with relative emotions, including a wide array of aggressivness levels).

I think that I will start with one tutorial, just to show the player, that the monsters behaviour is more complexe than he may think at first.
After this, I will give the player some hints in the form of diaries or notes he can find. It should be part of the game to learn more about the behaviour
and use it in one of many ways.

Quote:

If you're willing to go that extra mile, it would be interesting to see these "enemies" interact with one another. Part of where The Elder Scrolls IV failed was in tuning down their AI system (but surely for a gameplay reason).

I think that this is the problem of trying to simulate human or human like behaviour which is just too complexe for my hobby budget. My goal is to improve none human, more animal like, behaviour.

Quote:

Iron Chef: Give the player more experience and mob-specific "lore points" for interacting with a critter without having to kill it.

But how ? I think the problem is, that I can't decide about "clever" interaction. I.e. you are in search of a water/food source. One way would be to craft salted meat, lay in down on the ground and wait for some creature who get interested in it, take it up, eat it and will be hungry/thirsty within half a minute. Then you follow the creature to a hidden food source. I gave the player the possiblity to do things like this, but how should I rate his "cleverness" ?

My idea is, to make the crafting and behaviour system depending on resources and you benefit from interaction by getting access to rare resources, which will not be accessable by simple hack'n'slay.

Quote:

Zerox: if you could have a wolf's pack hidden somewhere in the trees until the curious one is attacked, make the fight pretty tough, and have their corpses only give you "useless small chunk of wolf meat" and veeeeeeery little XP, it could work.

This is just an other way of punishing the player for not interacting in the way the game designer wishes. Instead of giving the player just one goal like gathering XP, I think that it would be better to give him at least two goals, like gathering XP and gathering resources. Both ways should lead to success, it is just the decision of the player which way to go.