The game of the future.

[Spacemonkey]

Recently i've just started playing Darwinia, and i have to say it's one of the most awesome games i've ever played. Now I know I should have bought the game years ago and now many of you will think i'm then spawn of all evil for not doing so, but I only bought it yesterday.

So anyway, one thing I really like about this game is the fictional world of Darwinia, and after reading Chris' latests blog where he mentions the power of procedurally generated content.

So anyway last night in bed, I came up with an awesome idea. What if the fictional world of Darwinia was real? A simulation of sorts, where digital lifeforms can live in a digital world, doing their own stuff, based on a simple set of rules.

This has already been done before, with the game of life. But what i'm thinking about is the game of life on a much much larger scale.

We have a randomly generated world, sort of like the world in Darwinia, except there are no map borders, the "map" contains an entire world, the digital lifeforms can explore the world in any direction they wish. Now the digital lifeforms, like the ones in the game of life, are defined by a simple rules like gravity (keeps them stuck to the surface), hunger (they need to eat), death (they can die) etc.

Now each lifeform can perform tasks like planting a farm which would produce food, or mining the ground for minerals. Take for example the peasant in Warcraft 2, this guy can build farms and buildings, except the player controls him, well in this game, the lifeform will build what is required by his local community, if many of his community are hungry, then he will build a farm, if his community is falling in numbers, then he can mate with another lifeform and produce baby lifeforms.

Now the reason why this most likely hasn't been done on such a large scale, is because firstly, i would probably require too much processor power for it to run on a single system, and secondly, no game programing company would want to make it as it would give no financial return. So instead, we need a game element in the simulation, where players like us can interact with the digital lifeforms, by controlling our own group of lifeforms. We can trade with them, fight them in wars etc. The processing power for the game will then come from all the payers computers which are connected to the game, players computers can even lend processing power to the simulation while the player is away from his computer.

This has all been done before, but only in a small scale, and what the AI can do has always be constrained by the game designers, A simulation like this would remove those constrains, allowing the AI lifeforms freedom to do what they like.

What will this achieve? I'm not to sure my self, it would be a long process, the we would need to keep on improving the digital lifeforms, giving them extra abilities, eventually they may start even learning new things. Now i'm not a programmer, I don't know whats possible with our current technology, but it's something to think about, I have lots more to say on this, but i'm running out of time, so i'll add more later.

- Spacemonkey

[DueAccident]

I didn't read all of it, as I'm in a bit of a hurry, but it seems like an interesting idea, though I think it'd get boring pretty quick if you didn't have some objective to aim for. Perhaps small things such as 'Increase your population by 25%, to 1000' or something, and then you gradually progress towards a utopia!

[KingAl]

Sounds like a combination of the Darwinia premise and MMORPG. The essential element here is designing AI not to fulfill the purposes of a game, but still able to be controlled. At what point does designing 'controllable' AIs overrule the ability for them to be otherwise autonomous?

DueAccident: An overpopulated utopia?

[Shwart!!]

Could be interesting, although a true AI would be difficult to accomplish. It could be simulated by allowing the little guys to 'learn' from their experiences in specified paths (i.e. an army gets flanked, some survivors learn to keep a better watch out; a farm on rocks fails while a farm on fertile ground succeeds, they learn to plant on fertile ground). This would be similar to artificial intelligence, but would still be somewhat limited. Could be interesting, however. Especially if the land was limited, but enough lifeforms working at a shore point with enough resources would cause an extension by 'procedurally generated content'. This new land would reflect the core programming of the creatures, like: A mainly mining nation would get a lot of resource-rich mountains, while a militarily strong nation would get more ideal land for barrackses and the like. A nation that gets attacked frequently could get an easily defendable plateau. While there would always be a chance of something different, this would get the players' individual worlds to look quite different. They could maybe even make a nation of content creators, and other nations could hire a few lifeforms to create a specific type of ground for them.
Thus, the world would quickly grow to fulfill the dreams of the players involved, and the game designers would be able to focus on game dynamics and not on making new worlds for players to explore. A few computer-run nations could be there for people to observe, and they could watch the expansion over time of a purely isolated world.

I would find something good of this nature fascinating.

[Spacemonkey]

Thanks guys, you're coming up with some really good idea.

I agree that a true AI would most likely to be hard to program, but the learning idea would work.

How I think it could work is at the start of the simulation, many small AI communities are placed around the world, the terrain they live on will be different, some areas will have better food production etc. At the same time the world will be open to human players, who can also start their communities. As each community grows, they will need to trade with, or attack, other AI communities as well as human players.

However I think most of the fun will come from observing the AI, on such a large scale, it will be hard to predict what will happen, which should make things very interesting.

[Darksun]

There are quite a few artificial life projects out there, check some of them out.

[Xocrates]

That reminds me of a "game" I once played. basically you were able to spawn soldiers and set their priorities (like "approach the enemy" or "Attack the enemy") and design the scenario so they would learn how to fight and move. The ones with the best results would be kept and the weaker ones eliminated.

After training, you could set several armies against each other and they would fight by themselves as you watched the result.

The game was rather frustrating because as were teaching them one thing, they were forgetting another. But is was interesting.

Unfortunetely I can't remember the name of the thing

[Montyphy]

There is only one game and that is...






















































THE GAME!

You lose.

[el_cascador]

You might be interested in Spore
it's a game by Will " The Sims" Wright
I believe it will be realeased this year

[sirvoks]

Well, Spore interests me - I'm going to pre-order it whenever thats possible.

[mibias]

i have seen the 30 min. long demonstration of spore 3 times now... I can't wait for that game to come out

[lunquewill]

That reminds me of some of Sid Meier's games - Civilization and Alpha Centauri or Bruce Balfour's Outpost.

[Spacemonkey]

You might be interested in Spore

Spore looks good, but it's still designed as a game, it won't actually do anything unless someone is playing. However it's an awesome concept.

[DueAccident]

THE GAME!

You lose.

Goddamn..

[sirvoks]

You might be interested in Spore

Spore looks good, but it's still designed as a game, it won't actually do anything unless someone is playing. However it's an awesome concept.

That's correct. It's made like that for a reason - money