Advertisement

Uber-Universe or New Map Capability

Started by November 23, 2004 08:54 AM
26 comments, last by Numsgil 20 years, 2 months ago
Quote:
Original post by Edward Ropple
Oy. And I thought MOO3 was bad.

More than a million stars is what I personally would call crazy. No player will ever be able to explore them all. Even so, I think that's probably the maximum you could get away with for a high-level 4X game.

What you suggest would be really, really more suitable to a single sector of space or something. I'd put a max of maybe 10,000 to keep it playable (unless this is something designed for, like, 100+ players).


No one expects you to. And the game isn't 4X, just somewhat similar, probably more so than any other genre. (It's more 4X tahn say RPG or FPS or racing)

4X is about taking over other countries, and expanding your borders, and becoming the most powerful. This is more about expanding your borders and exploring for the sake of exploration, leaving the day to day decisions to the computer.

The size of the universe is not really open for discussion. I want to make it big, and I can make it big, so I see no reason not to. What is open for discussion is wether you should have one really really big one, or several somewhat smaller but still really big ones.

Anyway, I've never heard of a player complain that the game world is too big except in RPG's, where you are expected to go from A to B. If you have free reign to go wherever the heck you want, wouldn't you prefer more room to explore? Unless you're a perfectionist, who cares that you only explored .00000000001%?
[size=2]Darwinbots - [size=2]Artificial life simulation
Oh, moo3 was largely terrible [though the map was nice, and the empire management ai was above average...], but when there's so few 4x games; you take what you can get.

Back on topic. Why not make your game flexible enough to allow either? Some people might like the smaller maps, and some might like the large "standard" map.
Advertisement
Quote:
Original post by Numsgil
Quote:
Original post by Edward Ropple
Oy. And I thought MOO3 was bad.

More than a million stars is what I personally would call crazy. No player will ever be able to explore them all. Even so, I think that's probably the maximum you could get away with for a high-level 4X game.

What you suggest would be really, really more suitable to a single sector of space or something. I'd put a max of maybe 10,000 to keep it playable (unless this is something designed for, like, 100+ players).


No one expects you to. And the game isn't 4X, just somewhat similar, probably more so than any other genre. (It's more 4X tahn say RPG or FPS or racing)

4X is about taking over other countries, and expanding your borders, and becoming the most powerful. This is more about expanding your borders and exploring for the sake of exploration, leaving the day to day decisions to the computer.

The size of the universe is not really open for discussion. I want to make it big, and I can make it big, so I see no reason not to. What is open for discussion is wether you should have one really really big one, or several somewhat smaller but still really big ones.

Anyway, I've never heard of a player complain that the game world is too big except in RPG's, where you are expected to go from A to B. If you have free reign to go wherever the heck you want, wouldn't you prefer more room to explore? Unless you're a perfectionist, who cares that you only explored .00000000001%?


I know many people who complained that MOO3 was too big. A simple enough reason not to is that there is a large segment of players who WANT to be able to explore everything.

You seem to be one of the people who does things because they can. Setting aside the system constraints (and save files of truly obscene proportions), the playability comes into question. What is the point of such a huge playing field if you will never see most of it?
http://edropple.com
Quote:
Original post by Numsgil
PROBLEM STATED
I'm at a cross-roads in my space-exploration game design. I can either create a realistic (200 Billion + stars) galaxy where every time the player starts a new game, it's in the same universe, or I can make smaller globular clusters (~ 1 Million stars) of roughly 1000*1000*1000 LY, and have up to 2^16 different maps for the player to explore.

Uhh, if everything is randomized based on a seed, then in the first scenario, won't I still get a new universe of 200 billion+ stars each time I play if I change the seed?

Personally, the larger the better is my opinion. Why not? And your solution of allowing a randomized starting system as your "Sol" system is pure elegance and simplicity combined. Perfect! You've solved your own problem.

The only thing I'd suggest is making verson 1.0 be one system only. That's right. Sol, 9 planets, and their moons. Make everything you want in your final version WORK, and then for version 2.0 expand your scope to include the Milky Way. I don't know. You mention something to the effect that there are no greater challenges to programming for 1000 stars or 1 billion, but it just seems like you'd have a more realistic goal to reach if you were setting your sights on something more seemingly "accomplishable", for lack of any effort on my part to go to www.dictionary.com. Heh.

I've been interested in something like this for a long time and I can't wait until you have something playable. Keep us posted!

Take care, and good luck.
Florida, USA
Current Project
Jesus is LORD!
Sorry, ^ was me.
[size=2]Darwinbots - [size=2]Artificial life simulation
I can see a potential problem with exploring lots of star systems - keeping track of what's going on in them. You'll need to revise the model by which non-player empires evolve, again using something maths based rather than an AI following fundamental rules.

Otherwise every time you find a new civilization you'll need to spawn a new AI to manage it, and the demand on the CPU will just skyrocket as both you and they grow. With a mathematical system you can find an empire, leave it, come back again later and calculate what state it's in without having to do thousands of AI routines in the interim.

There's also the problem of the player's own empire, although given realistic limits on how much a player is likely to be able to explore in a given time frame this might not be a huge problem, but it is conceivable that a player might spend ages expanding his empire to something like a million stars - and even if he ignores them, the computer still has to keep track of what's going on in them.

Advertisement
Quote:
Original post by Sandman
I can see a potential problem with exploring lots of star systems - keeping track of what's going on in them. You'll need to revise the model by which non-player empires evolve, again using something maths based rather than an AI following fundamental rules.

Otherwise every time you find a new civilization you'll need to spawn a new AI to manage it, and the demand on the CPU will just skyrocket as both you and they grow. With a mathematical system you can find an empire, leave it, come back again later and calculate what state it's in without having to do thousands of AI routines in the interim.

There's also the problem of the player's own empire, although given realistic limits on how much a player is likely to be able to explore in a given time frame this might not be a huge problem, but it is conceivable that a player might spend ages expanding his empire to something like a million stars - and even if he ignores them, the computer still has to keep track of what's going on in them.


Yes, this is the one main disadvantage to my game. I think, at least for now, the NPC empires will largely be stagnant. The only interaction they'll have with the player is when the player interacts with them, and things like war etc. will be handled using math instead of simulation.

In my mind, it's a worthwhile tradeoff, since the game is about exploration and not management, I can have the computer deal with upkeep with relatively simple mathematical models, keeping the player none-the-wiser (hopefully anyway.)

As for the player expanding too large, I'm hoping to find a way to keep the simulation of colonies as straightforward as possible. Since the colonies' only affect on the player is money (so far atleast), it shouldn't be too hard to just make a model for it to follow. However, I can definately see a possible problem of having a human empire that's just too large to cope with. I'll have to play around when I get to alpha testing to see how large the player's empire has to get before that happens.
[size=2]Darwinbots - [size=2]Artificial life simulation
First: Just make a cap a player can choose for number of stars. Have it go as high as you please. Don't make the player choose how "big" it should be in spatial directions; do that behind the scenes.

Second: Not only will the CPU demand be insane, have you considered how big a memory hog this would be--and, for the love of God, disk space? I'm thinking "gigantic," here. Assume each star has a ten-character name; that alone, for a galaxy of 200 billion stars, is TWO TERABYTES. Compress that and you save some space, but still, that's just a *wee* bit over the top. Add in the numerous values necessary to accurately represent each star, and you'll rapidly go insane and get some serious out-of-memory errors.
http://edropple.com
Quote:
Original post by Edward Ropple
First: Just make a cap a player can choose for number of stars. Have it go as high as you please. Don't make the player choose how "big" it should be in spatial directions; do that behind the scenes.


Okay... but why not? I imagine most of us think in terms of spatial coordinates.

Quote:

Second: Not only will the CPU demand be insane, have you considered how big a memory hog this would be--and, for the love of God, disk space? I'm thinking "gigantic," here. Assume each star has a ten-character name; that alone, for a galaxy of 200 billion stars, is TWO TERABYTES. Compress that and you save some space, but still, that's just a *wee* bit over the top. Add in the numerous values necessary to accurately represent each star, and you'll rapidly go insane and get some serious out-of-memory errors.


Yes, that would be true if I actually stored everything on disk, and kept track of each star, however I don't. I'll put this in caps so that people who are reading impaired can understand it.

MY GAME USES PROCEDURAL GENERATION!!! -or- if you don't know what that means:

MY GAME MAKES STUFF FROM A RANDOM FUNCTION SEEDED WITH THE SAME NUMBER EVERYTIME INSTEAD OF SEEDING IT WITH THE CLOCK!

Sorry, I didn't mean to shout. Just wanted to make sure everyone understood. [wink]
[size=2]Darwinbots - [size=2]Artificial life simulation
I read what you wrote--I still think you're wrong.

Find a way how not to keep the majority of stars (every one an alien empire has control of, every one a player has control of) in memory at any time. You'll shortly find that you'll be running low on available memory really quickly, even on fairly buff machines.

Also, somehow I doubt your uber-l33t procedural generation routines will be able to come up with the alterations the player and AI empires/players have put onto planets they control, etc. If the game is on such a scale as you're talking about, this alone could run to tens of megabytes. Even with procedural generation, there's a hell of a lot of stuff you need to put on disk.
http://edropple.com

This topic is closed to new replies.

Advertisement