Without going too much into details you have already pointed out, I’ve plan to allow players to submit content, which is played by the test group, if it suits the world its added to the servers, and playable by all others.

The world is totally dynamic, if a player wants to invent a new sword (maybe its too heavy, but he’s added a spell to make it light weight) he spends the time to make it, then he has the only version. Unless he re-creates it for another (maybe quicker, maybe better, maybe not) or teaches the technique to another, then possibly the game now has a new sword.

Also, allowing players to choose to be the leader of a monster band, who then go into the world creating havok, which prompts the other players to deal with the problem.

Way back in the day Heavy Gear had an inkling of an idea along these lines: the individual deathmatch battles counted as skirmishes in an overall global war: depending on which faction came out the winner in each matchup on a server, the battle lines for the global conflict were moved. That then affected the map locations you had open to you and the equipment you could take into a match. Unfortunately it was a flawed implementation with positive feedback loops: if memory serves, as your side lost you had less equipment open to you, which just made you lose faster. It often wasn’t as much fun since so many people were playing that you still had no influence on the larger picture, and furthermore if you logged in at 4pm wanting to play and found your faction had been losing all day, you’d be stuck with the crummier gears… The memories are fading, but I’m pretty sure they scrapped the system within a week or so of release rather than trying to fine-tune it.

Another game that is along those lines is Freespace. It had a server install, and you could leave the server running 24/7 and people could join at will, up to 32 simultaneously (if your server had the processing power). However, it was lacking in the RPG components: you could play through the main storyline with your friends, but nothing really ever changed in the game universe…

When Player A’s mom tells him to get off, he leaves his computer on (and connected to the internet), so that the other players can connect and play without him. If he’s unable to do this, in theory the world could come and go (depending on when he turns his computer on and off) but a better solution would be to have it hosted or for another player to run the game server (I have three computers that are connected to the Internet and running 24/7 – one is a server and the other two are my desktops at work and home).

Cheaters would be the major problem, and if you can’t trust the friends you’re playing with (ESPECIALLY the one running the server), then you’d be better to get the game hosted by the company (and security is one of the features they’d be offering by doing so, along with data backup in case of hardware failure and more robust uptime).

Haven’t played X2 or X3, so I can comment. There are many options for path-finding and other AI algorithms, and resource consumption is a major part of how you decide which one to use. The target hardware the game would be expected to run on would be a big part of how these design decisions would be made.

The content COULD be generated automatically (in a roguelike style perhaps) and this might even be one of the content packs offered (maybe a “Chaos dungeon” that changes every time you enter it), but there would also be scripted content that could be purchased. Say you, Mike and I chip in and buy the “Curse of the Genii” module. A mountain pass in our world that had been blocked suddenly opens up, and allows access to an Arabian style area of the world, complete with new NPCs, monsters, quests and side-quests, equipment, etc, etc. The major quest in this new content would probably be scripted (and we’d have read a bit about it before we decided to purchase it and add it to our world). We could in theory just go into the world, get a magic scimitar and create a new “desert spirit summoner” character (for Mike’s wife to play), then come back to the previous content we’d been working on, or if we were sick of that, we could move totally into the Arabian world, and work on making you the “prince of the oasis caravans” and defeating the evil Genii (then perhaps going back to familiar territory later, or even purchasing another new module, or both).

Anyway, back to the original topic again. So essentially what you’re describing is that you want not a persistent world per say, but a world that the state could be saved to come back to on a regular basis that fills in the blanks for you when you’re gone… kind of like Animal Crossing on the Wii/GameCube. If you save your game there, and then pick it up in the winter time all the people ask where you’ve been, there is now snow on the ground, that kind of thing?

I’m still hung up on the logistics. What happens If Player A invites Player B to game 1, and they are playing on Player A’s computer. Mom of Player A tells him to shut off his machine so now Player B doesn’t get to play anymore? That’s a real mood killer. What happens if this keeps repeating and Player B wants his character to move over to Player C’s game 2 but now Player A is holding Player B’s character ransom on a powered off machine. Also, with local data you end up with cheaters…. all the time.

Have you ever played X2: The Threat, or X3: Reunion? I haven’t played much of X3, but even X2 taxes my system when your in an area trying to keep track os all the AI movement. If it also had to do detailed tracking like that in multiple areas I doubt the server would be able to also be a client and have it run at any descent speed.

Though I guess if I’m reading it correctly you want to make a multi-player save-able heroes quest style game that somehow can dynamically generate content for the users depending on the number of people playing at the time?

Not sure what you mean in the 2nd paragraph. Just about everyone who has played World of Warcraft has killed “Hogger”. It’s a quest that has literally been performed (often repeatedly) by millions of players (the poor guy is probably sick of dying).

It wouldn’t be a peer matching service. You’d specifically invite your friends. So I’d buy the game and set up the world and invite you and Mike to play on it with me (maybe you’d have to buy the game as well to play, perhaps my license would cover you guys). After we’d been meeting up for adventures evenings and weekends (sometimes all of us, sometimes a subset), Mike’s wife might get sick of being left out and join the game as well, along with your cousin who likes to play RPGs. It’d be a real world group of friends playing, not random people shoved together.

I haven’t made a large or innovative game before (mostly proof of concept things), but if I was going to try something like you space empire idea, I’d try to make the smallest subset of features possible, then keep adding to it. Your players would be delighted to tell you whether its playable or not (and how to make it better). Another option is to take a game that’s similar to what you want to make, and implement it with the changes you’d like (and hope the the play balance survives, or tweak it if it doesn’t). I think a space empire version of Travian (http://www.travian.us/) would be quite interesting…

You think it wouldn’t work because of computational constraints (too much work maintaining the world the NPCs) or logistics (what if the client machine got turned off or was running other program)? In term of the computational constraints, modern PCs are VERY powerful. There would also be short cuts that could be taken (instead of simulating every NPC’s every action, deal with them at a higher abstraction when a real player isn’t nearby).

In terms of the client machine issues, if the program is turned off, it could save its state, then just resume the next time it’s turned on. I’m not really sure what part of a persistent world you think would be a challenge to run on a client machine. Ultima Online started in 1995. Affordable client machines are FAR more powerful than the servers they would have been using 13 14 years ago.

Maybe I’m just not getting the logistics, but you can’t just offer one ‘save the princess’ quest to 100,000 people. Player generated questing works sometimes a la face of mankind or something similar.

If you aren’t making an MMO then you could run each ‘world’ on a client machine as long as it wasn’t too complicated and run a peer matching service like XBox live does for console game players ( ewww ).

I’ve been thinking about trying my hand at designing a space empire management economy simulation game . . . and I have some good ideas up there in my head but no real skills to get it playable. I’ve been searching for a game to play for a while now and haven’t found anything that I really like. Though Mankind comes close (not Face of Mankind).

I still don’t get how you could run a persistent world on a client machine.

The difference being the hero of your game is that the world changes to accommodate that difference. When you’re the hero and you kill a villain (or anyone else for that matter), he stays dead instead of reappearing so you (and others) can kill him again a couple of hours later. When you’re the hero and you kill all the monsters in an area, that area becomes “safe” instead of all of them reappearing in the near future. The idea is that you can actually have impact on the world (something that’s VERY difficult to provide players in MMORPGs).

In terms of the server issues, remember this is an RPG, not a first-person shooter. If people got to the point where they didn’t want their machine to get bogged down, they could either dedicate a high-power machine to running the game, or purchase a monthly membership to have their game hosted for them. In terms of simulating the world and the AI of its inhabitants, remember that there aren’t going to be 1000′s of players on at once, which allows short cuts to be taken in the areas of the world where the players aren’t active.

The depth that you’re looking for is the sort of thing that could be offered by this. Instead of grinding, like in the MMORPGs, there would be a rich, adventure, and a world that significantly changes as you play through it (along with your characters changing, and their relationship with the world).

How it’s different from a single player game is obviously that you play with other people.

How it’s different from a MUD is the same ways it’s different from a MMORPG (as above) plus with graphics.

How it’s different from a co-op multiplayer game is the real question. It’s possible to play Neverwinter Nights and Neverwinter Nights 2 in co-op mode. As well as replaying the main adventure (except with your friends) you can also download modules (including fan created content) and play through those together as well.

What I’m suggesting is a step closer to MMORPGs than that. In NWN the module are all distinct (although characters can be imported from previous modules). In this, new content would be stitched together to make one large consistent world (so if Mike and I are trying to find the Evil Genii’s bottle in our current quest, you could go off on your own and go find the werewolf’s sister from the previous module we played and apologise to her for killing her brother – a side quest we hadn’t bothered with).

I think there would be interesting things you could do with the economy when you aren’t worried about throwing it out of whack in a MMORPG. Perhaps as you accumulate tons of magic weapons and sell them, the market gets flooded and the price of low powered magic weapons becomes cheap (and maybe city guards and goons start wandering around with them). Say you open a factory that makes healing potions, perhaps they also flood the market, propagating out to various villages and countries as traders buy up cheap potions in the village your factory is in and sell them elsewhere.

Another interesting idea is that the players could make new 1st level characters and play in the world that their higher level characters have created (even perhaps meeting them). When they got to the town where the previous group had burned down the inn, there would be a burnt shell where it used to be.

New character classes and abilities could be introduced in new content packs. Bioware makes NWN to sell the game (the creation kit is just a really nice add on). The idea here would be to sell the game as a service where new content is always being added (and hopefully players would keep buying it). Instead of having NWN and NWN2, the game would gradually evolve to have the features of NWN2 (and eventually 3, 4, and so on as long as its a viable business). People could put together their own game by buying (and using) the parts that appeal to them. If they decided they didn’t want clerics in their world, they could turn them off (and maybe add them in at a later point, ala DragonLance).

Melanie: Thanks, I was a big fan of the King’s Quest games too!