The Primogen library my company is constructing is basically an MMO construction kit.

That said, it's best to consider as more a production environment, since 'kit' implies that it prepares a complete product. We take NO interest in the balancing of game specific quests.

The issue of game-specific balance can't really be approached if you want to allow any flexibility of gameplay in your kit's product.

In our top-end library, we have a simulation layer that can be used to generate standard 'pickup / delivery / recon', 'manufacture', 'attack' quests, with monetary or non-unique (system craftable) item rewards. Basically, it just generates 'work' for players and NPCs. It works for 'stuff to do', but it doesn't loan itself particularly well to story-based quest generating, stories that it can generate (through its random events pools) are fairly obviously pieced together, but again, they work for the situation they're instanced for (and they are generated on the fly). All this stuff can be automatically balanced to some extent - by tweaking the input values to the system (spawn and decay rates, event tables) which is a nice subtle way of shifting the economic balance of the game world. I have to stress that although this system works to generate 'chaff' quests (disposable adventures!), it shouldn't be relied on exclusively - instead it can be used to provide an unstable backdrop to the game world and to regulate that game world if required.

Epic feeling stuff is something that ends up having to be carefully scripted, or it just doesn't work. Unique items, again, really have no place in a random system - a +1 longsword quest may be random, but the +5 Holy Avenger quest certainly shouldn't be. It would only feel cheap. Such items and events really need a special environment that isn't just a humdrum part of the world - the quest has to interact with the environment in a more complex way than a treasure chest buried under a tree.

I agree with hplus - player driven quests should be IN GAME. Bloodspear (our MMO project to be developed with Primogen) has a very complex player quest and social system. I'm really quite proud of it. However, for many games this won't work - Bloodspear's a relatively small community platform, and it's very much a social experiment. I don't envisage taking any instance of the game world over around 750-1000 concurrent players. There are two ways we deal with quests - through the social system (which I won't go into, but it's basically a cascade), or through a quest broker. Quest brokers are NPCs that allow you to specify a quest, and provide a reward. On receipt of the item (or 'appropriate' confirmation of the deed - it IS possible to cheat them - some more easily than others), the broker provides the reward, taking a cut. Because of the small community size, well-regarded quest givers gain a reputation as such - particularly important for those wishing to maintain their status on the social tree.

Bloodspear also has player-developed content, within very tight restrictions, as vetting everything (as has been mentioned) is impractical.

Any Guild, free company (approved player guild), Lord, Baron or Regent (all potentially headed by players) may develop their own heraldry - this is a choice of colours and motif. Heraldic rules are enforced, so nothing too garish results. Also there can only be ONE instance of any heraldic pattern in a kingdom.

Any player positioned world objects (building components, ground-rooted objects) on land allotted to them can be arranged as the player likes. This means a player can potentially create a labyrinth of hedges, or a castle, providing they have a large enough allotment. There is no option to build anywhere - players are forced to settle in towns and villages that WILL owe their allegiance to one of the kingdoms (balancing factors) in the game. We can therefore track an overall balance of power as players move around and tweak the economy and events to keep things interesting.

I see I've started rambling again. In terms of where to go in building an 'MMO kit'...

Well, I'd start with AD&D and a host of other PnP RPGs. GURPS is good as a basic concept, too, it's adaptable, and that's what you're looking for. Next up you need mechanisms to generate and balance some activity in the world. These could be simple spawn systems or a more complex simulation. You need to be able to counterbalance the effect your players have on this system. If it starts producing gold like nothing, you'll have hyper-inflaction in your game. If it doesn't produce enough, you'll either end up with players leaving because they can't afford kit, or a barter system arising (with items changing hands on ebay etc). Whilst it's fun and convenient to have rats and bunnies drop a few gold coins, it's not realistic, and you can eventually expect a non-realistic economic shift. Next you need a progression system. Skills or levels. Again you need to ensure your game world supports either appropriate difficulty beasties (somewhere) or *some* (appropriate difficulty again) use for the skills you're providing. 'Weapon proficiency - laser blaster' is not much use in a medieval historic setting. Killing the same dreary bunny rabbit is no fun when you're a lvl90 barbarian.

Scripting is an art, and designing quests that test (in a fun way) any character eligible to do them is exceedingly difficult. Experienced DMs / GMs sometimes have a knack for it, but even then often human moderation is required to either up the ante or lessen the pressure. These aren't really doable for the majority of quests in an MMO, where you have hundreds of people wanting to do hundreds of different things. Cycling one piece of shit quest in to replace another (although dynamical(?!)) would still result in shit quests. Quality checking is essential - you'd need to gather appropriate feedback from the test server, and I'd bet you'd find an awful lot of rejections.

I didn't ever bother looking at an MMO construction kit. I didn't want to waste my time learning a piece of software that might not do what I want, wouldn't be easily expandable, couldn't be licensed on if successful and otherwise might not measure up to what I could get produced myself, especially when I had such complex mechanics in mind.

Quote:Original post by ronkfist
yea that's what I meant with "making your game interesting in other ways", so that you don't only have to "kill, get drop and repeat".

Quests are basicly the same... instead of having the freedom of chosing what u wanna kill and where, they are even worse by making u kill a specific thing on a specific location. I don't see how you will change that by letting the community create their own quests.

That's not a quest you're describing, it's a static spawn. I agree, static spawns are utter crap. But they're easy from a development and debugging / balancing point of view (they don't spawn anywhere near newbie areas, and / or have a constrained range of movement from their spawn point).

If you're responsible for tracking said beastie (perhaps only through the aftermath of attacks it makes on settlements) that kill the moor beast quest gets a lot more interesting, especially if it takes a different route (past different hostiles) each time.

@Cold_Flame: I purposely limit input from players to in-game mechanics. There's no 'dev tool' involved, everything is done from the standard client front end. We use (individually identifiable) 'superclients' for actual world editting, but these simply don't go to players, the reasoning being that anything a player does HAS to be counteractable by the system.

Allowing one side to 'win' in a warfare situation is untenable in most situations - the resulting misery for the losers is not fun, and will cost you subscribers. Allowing the balance of power to teeter (and using various social and economic tricks to keep the population relatively balanced across the nations involved) makes for an interesting and unpredicatable setting between two limits that cannot be reached.

The same is true of fixed epic plots for game settings - not really useful since by their nature they finish. If you plan to run the game for a set time and a set time only, then this is fine. For the relatively long period of time an MMO player life cycle covers this isn't practical without an army of scripters at your disposal.

I guess the point I'm trying to make is that there should be set mechanics for the players to be able to carve out a real, tactile history for themselves in the game world (and not rely on invulnerable cheesy NPCs to provide 'plot' in meaningless chunks of terrible dialog). Letting players set tasks for each other (and attempting to ensure this happens by deliberately crippling economies at key points) is a simple way forward to cause plots to be generateed at the player level, not the game. Of course, you have to make sure that everyone knows that a certain amount of lateral thought and *gasp* communication and *shock* socialisation will be required to progress through the game.