What everyone needs to understand here is that you can't think, conceive, a game in a 1:1 world scale the same way you think a game with a more conceivable size. The biggest 3D game for the moment is Fuel if I'm not wrong, that game has an area of less than 200 square kilometers and is consider to be huge, 3 times bigger than the second biggest map (without taking MMO into account). With those games, development is likely to go this way: They grab a blank piece of paper and vaguely draw the map and determinate the size taking into account the movement speed of the players and within the hardware limitations; then they distribute the assets in a practical way and start sketching how that map will look (with an artistic view), that goes to the modelers and the map itself becomes what we end up playing on.
The game with the biggest map ever made is Daggerfall, graphically based on sprites and it's about 650 square kilometers. The hand made part is the overall shape and the distribution of some of the cities, the rest is procedurally generated within some parameters.
Planet earth (outerra) has almost 150 million square kilometers and just taking the land into accout (30% of the planet surface), there is no point of comparison here, the sheer size is absurd.
To me, the best example Outerra could follow is the Daggerfall approach, you can't manually fill the world with worth-a-while content, you need to give tools to allow players to fill it themselves and, of course, enough mechanics and complexity to make it entertaining in the long term. No one will be capable of manually placing each spawn points for animals/enemies, no one will be capable of distribution towns/smaller settlements and or ruins/dungeons. A similar balance between "manually" setting parameters and procedural generation à la Outerra's map generator has to be found for the content too.
Animals should be given a behavior (aggressive, non-aggressive, herbivorous, carnivorous, etc) and a pre-determinate natural environment they inhabit (ie: hares living only in the forest).
Some build up areas should be handled with procedural generation too. Made some building models, assign some parameters for the disposition and the conditions where said areas can occur (ie: x amount of square meters of terrain with less than y variation on of terrain height), so the buildings aren't stuck between two mountains or sitting unnaturally standing out of a cliff.
On the player side, there needs to be tools to set up buildings with different qualities and capabilities, make it as complex and depth as possible, otherwise it will likely become boring after a while. A game that did this sort of right (or at least with the right intentions) is Planet explorers, you have a good degree of freedom when building and a variety of things you can add to your base to help you defend it. Imagine that for food gathering with methods such as hydroponic farms, metallurgic facilities that require some more interaction from the players than just putting a raw material in a small box and waiting for it to become steel, energy generators that require taking into account the environment where you are setting up your facilities (ie: Making solar panels less efficient with a weather like you will have in England and more efficient in the Mohave), and so on in all the branches of the player activity.
This creates its own set of issues from the gameplay design perspective. I'm not sure if anyone has played Dayz here (it's an ARMA mod that eventually became a standalone game about surviving in a quite realistic zombie apocalypse), but most servers become a kill-on-sight fest, no one really cooperates and the majority of the players keep to themselves, you end up surviving from the people rather than from the elements (rain, cold, etc) and the zombies. To keep a game from going that way you need to have a common treat and enough reasons to set up a base and look for cooperation. Slightly forcing people to team up giving a background story to tie them together wouldn't be a bad idea per se, but that could end up creating over powered teams abusing all the smaller ones. To me, it would be better to let people create their own groups and even give them tools to assign authority ranks and access levels (to limit the newer members from using/wasting resources without older member's approval); again, it would be necessary to make the resource generation task not completely automatic, so someone needs to actively do something with each machine every now and then to keep it working, that would motivate people to come together and work as a team.
I don't know enough about the game (anteworld) to make a specific suggestions about this, but the common treat that everyone could be worried about is crucial to avoid becoming yet another Dayz. The zombies should do that on that game, but they are way to stupid and irrelevant and no one is really worried about them, they all focus on killing people and looting them. If people needed help to deal with the zombies, that wouldn't happen.
Well, end up writing way more than what I originally had in mind, it became more of a general thought and concerns about game mechanics than just how to handle the map, but I feel everything is closely related.
Author
Topic: do the devs wana shrink the world size for anteworld? (Read 11344 times)