Well if its not some ram starving 3D game , the game development itself takes about 3 days. The networking is the the killer part , so mutch bugs can happen with computers communicating at different speeds, different connections etc. then comes the pain of setting up a server. So that takes about 3 weeks. I dont know any networking so I am not sure how badly I am off, I know it shure takes more then 3 days.
Thinks I was being serious!! <_<
It is going to take a whole lot more than a weekend. Most likely a whole lot more than 3 months too.
The networking component should not be too hard.The tricky bit is in synchronising the signals so that the graphics are seemless.
I am actually considering running the whole system from a SQL server and having the player's PCs handle all the graphic from queries sent to the server over the internet at a rate of a couple per second. It is going to take a pretty hefty server to cope with that kind of traffic. The individual PCs can calculate how to move sprites from where they are now to where they will be according to the X,Y coordinates from the server.
I had the idea of a mMO once. Let me explain:
Instead of a Massive Multiplayer Online game, you do a minimally Multiplayer Online game. You make a program with a server and client. Most people have a spare computer to devote to being a server.
So Joe Shmoe buys the mMO, and sets up a persistant world. He then tells his friends, and they start playing on their own server with something like 10 players.
Which has the benefits of:
1. No servers to buy and maintain for the developers.
2. No worries about hacking (what sort of friends do you have that would hack their own personal server? Anyway, just don't let your hacker friend play)
3. Less anonymous playing. You can be the star of the world, since there aren't hardly any of you playing. The one thing about MMOs that frustrate me is how hard it is to gain any sort of clout.
I like saving the world and having everyone recognize me. This way you can have that and still have the persistant world thing going on.
4. One player can be the DM instead of playing a character. If you make some really good DM tools, this ends up bridging the gap between D&D players and RPG players.
5. No overcrowding. There just aren't enough people to overcrowd. Yet, since you know everyone that's playing, you can easily coordinate some grand adventure.
6. You can get rid of the spawning monster generation model. You can have monsters that die and don't dissapear. You can have monsters that are actually born. You can have players actually chop down trees and the trees don't grow back. If someone torches a forest, the whole forest stays down, or slowly recovers.
Basically, whenever you do something, it permanently would effect the world. It would be quite possible to kill every last ooglblat in the world, making them go extinct. Or kill every single person in the world. Or maybe even start a family with an NPC and make your own town.
The developers don't have to worry about gameplay balance (is this action overpowered/underpowered) since all the players are friends, and can make up their own "house rules", and tweek the stats to their own style.
Bandwidth requirements are likely within the realm of most standard cable modems since you only have a handful of players. I see a game like this being much closer to tabletop RPGing experience and less of the level treadmill that most MMOs necessarily are.
I've had this idea since way back when I got Ultima Online. Only problem is any sort of online game takes in the range of 1000s of manhours.