• entries
45
111
• views
49860

# Getting Dusty In Here Again...

535 views

You know, with journal updates so infrequent of late I just might have to turn in my GameDev "Most Prolific Poster" badge. [rolleyes]

Life is quite busy! I'm trying to chew through database programming material for a new job, which is in healthcare. Although it's not the most exciting thing in the universe, I work with some really cool people. I'm also, thanks to the tender loving folks at the IRS, enjoying having more than $0.37 USD in my bank account... just don't tell anyone that I now program in Visual Basic, okay? o_O The local lottery here is up to$250 million. The only thing I can think about is how many games I could make with that. (Of course, with next gen budget balloonage, probably two... anyways...)

Straylight Update

I finally have the cash to buy game tools! I picked up the Torque Shader Engine, Character Viewer and Torque 2D. I'm getting into learning the viewer because I really want to see what my models look like in their engine. But I can't seem to get WinCVS to download TSE! I'm going to fight with it a bit more before I try switching to an older version of WinCVS.

The real thing I want to talk about for a moment is this, though: What are some credible ways to build a galaxy?

One major problem I find in trying to come up with new game design ideas is the search for validation. You might have an awesome idea with great potential. Or you might just be off your freakin' rocker.

This is one of those areas. How do you even approach an idea as intimidating and potentially overwhelming as trying to generate the illusion of a seemingly boundless play space? Our galaxy contains between 100 and 400 billion stars, more than anyone could ever explore in any lifetime. How can you create the sense that not only can the player strike out in any direction as often as they like, but that there'll be something interesting to do when they get there?

I've got several ideas here. The first, as I've written about before, is "Go Anywhere Gameplay." Whatever you are as a character, that's what generates the gameplay. It could be a ship filled with mutinous pirates, or the cybernetics, AI and nanotech in your own body.

But that by itself is not enough. There needs to be the sense of story, there need to be interesting places to go, and worthwhile characters to meet/defeat.

I'm considering a number of issues here:

• Simple one, but how do you name everything? I don't like the results I've seen of some of the random technologies like Markov lists, so I'm experimenting with phonemes. Ideally, I can both screen for inappropriate names (or even legally verbotten ones, like "Skywalker"), and create grammars that match the disposition of the race (harsh sounding words for a harsh culture, IOW)

• What does the world look like and what is a planet good for, let alone a solar system? This is a hard design conundrum because it basically asks what the player can do and what they're supposed to be doing. Should every planet be a threat of some kind? Should there always be resources? How many "duds" can there be, even if it's realistic to find a barren, radiations blasted wasteland without life or minerals of value?

• How do you (or do you?) keep the player focused? If the play space is too large, they may wander down so many paths that they get swamped and can't even remember the main goal.

I think one of the biggest secrets lies in giving the illusion of boundless space but making the player so involved that they don't need to find its edges. I'll explore this more in coming posts.

If your having problems with getting the feel of a galaxy/universe, you should go and chat with Yasenaya. His project is uber 1337!

*takes pills*

Quote:
 This is one of those areas. How do you even approach an idea as intimidating and potentially overwhelming as trying to generate the illusion of a seemingly boundless play space? Our galaxy contains between 100 and 400 billion stars, more than anyone could ever explore in any lifetime. How can you create the sense that not only can the player strike out in any direction as often as they like, but that there'll be something interesting to do when they get there?

This, to me, is what would make or break the game and the need for the 'boundless galaxy'. The game Elite II: Frontier prided itself on an accurate modelling of the galaxy and all its immensity. There were thousands, if not millions of systems to explore - the problem was, however, that aside from the main centre of the game area, the rest of space was empty and as a result it was totally and utterly boring. As we all know, boring play is pointless play and players will have no incentive to visit the systems, meaning there's little point putting them in there to start with. I guess what I'm saying is that without adequate and consistent content for the vastness of space (which would result in thousands of man-hours to create, most likely) there's probably very little point including them - especially when you consider that players are likely to stick to a core of home systems because they have familiar cargo routes and/or factions. In my opinion the game would benefit from fewer, but more richly crafted sectors to visit.

A perfect example of this principle in action is the Freelancer game from Microsoft - there's only 40 or so visitable systems, but they're very distinct, have a rich vein of content and are worth exploring in their own right - yet the game is still totally freeform and never feels cramped.

Quote:
 The game Elite II: Frontier prided itself on an accurate modelling of the galaxy and all its immensity.

Elite/Frontier is overrated. Truely. You can't imagine the number of tricks David Braben has been using to recreate an illusion of a full galaxy. But if you're interested in reading some, somebody has been reverse-engineering the galaxy generator of Frontier

That being said, planets in Frontier do not move. Or actually, they do.. but in camera space. I think DB used that model to hide the precision problems. Forget about using floats or doubles at that time. In addition, planets have perfectly circular orbits. Many orbital parameters are simply ignored, or choosen randomly from statistical tables, and not calculated realistically. We are far, very far from Celestia.

And you know the best ?

Nobody has noticed that (except a few geeks).

I think my point is becoming clear:

Quote:
 How do you even approach an idea as intimidating and potentially overwhelming as trying to generate the illusion of a seemingly boundless play space?

The keyword here is illusion. You should trick the player. You have no choice actually. As Wavinator said, using a procedural system you can generate billions of worlds. Not all of them are going to be unique or interesting. Neither are they in reality. So, instead of trying to generate billions of interesting worlds..

Try to come up with a few hundred interesting worlds only,

And make sure the players find them :)

How to do that ? Well i'm still researching. But here are a few ideas:
- manually design the most important worlds (like the Solar system, or Alpha centauri).
- use a statistical model to randomly make some interesting features on some worlds,
- group the interesting worlds in clusters
- make the player start in an interesting world
- link the interesting worlds together, via jumpgates, trade routes, or other design elements used in your game
- make missions for the player that lead him to another interesting world
- make interstellar travel between non-interesting worlds harder (i plan to do that by making fuel harder to gather in unpopulated areas)
- probably a lot more

You've probably already seen it, but just in case:

http://www.gamasutra.com/features/20010302/oneil_01.htm
http://www.gamasutra.com/features/20010810/oneil_01.htm
http://www.gamasutra.com/features/20020712/oneil_01.htm

Good luck on your project, I was just browsing it and it sounds extremely interesting.

Quote:
 Original post by exoskeletal_ninja If your having problems with getting the feel of a galaxy/universe, you should go and chat with Yasenaya. His project is uber 1337!

[cool] Most agreed!

@IQpierce: Thanks for the links. I did see them awhile ago, but it's a good reminder!

Quote:
 Original post by evolutional I guess what I'm saying is that without adequate and consistent content for the vastness of space (which would result in thousands of man-hours to create, most likely) there's probably very little point including them - especially when you consider that players are likely to stick to a core of home systems because they have familiar cargo routes and/or factions. In my opinion the game would benefit from fewer, but more richly crafted sectors to visit. A perfect example of this principle in action is the Freelancer game from Microsoft - there's only 40 or so visitable systems, but they're very distinct, have a rich vein of content and are worth exploring in their own right - yet the game is still totally freeform and never feels cramped.

Very good points. What this comes down to ultimately is what value / utility any area within such a game has, how quickly you can access it, what its payoff is versus the work to get it, and how varied it must be both visually and functionaly (in gameplay terms) in order give you that value. When I think of trying to solve this problem I think of a stat heavy but interface light approach because one of the easiest things to vary are numbers. (The trick, I think, is to make you ignore the fact that this is all that's happening, by wrapping it in meaning-- either in the form of strategy or story.)

On Freelancer, I have to disagree a bit. I must say that the universe did feel depressingly cramped to me, even once I broke free of the yoke of the story. I couldn't get out of my ship except to see cutscenes, trade didn't have enough nuance and strategy for me, the warp travel & disruption wasn't interesting enough, and I really wish that space had more of a feel of life to it. I mostly play it for a good, straightforward combat game, and as long as I remind myself that combat is the game's soul, it's enjoyable.

Quote:
 Original post by Ysaneya Elite/Frontier is overrated.

Well, it will be when you get done. [lol] But, seriously, when you consider that the game was on floppies...

Quote:
 Try to come up with a few hundred interesting worlds only, And make sure the players find them :) How to do that ? Well i'm still researching. But here are a few ideas: - manually design the most important worlds (like the Solar system, or Alpha centauri). - use a statistical model to randomly make some interesting features on some worlds, - group the interesting worlds in clusters - make the player start in an interesting world - link the interesting worlds together, via jumpgates, trade routes, or other design elements used in your game - make missions for the player that lead him to another interesting world - make interstellar travel between non-interesting worlds harder (i plan to do that by making fuel harder to gather in unpopulated areas) - probably a lot more

I think this is a great approach, and I think it will work. But the one problem I have with it is it may violate what I think is a fundamental theory of game design:
You should never, as a standard feature of gameplay, allow the player to do something that would be boring, frustrating or annoying.

It is a very good idea to chunk the interesting content together and offer attractors that keep the player homing in on it. But if you create areas that are hard to get to, you may be creating a subconscious signal to players that they should work harder to get to them, and expect a reward as a result. They may come to think of hard to reach systems like FPS gamers think of hard to reach jumps, or secret doors-- as hiding places for the power ups and super weapons.

I think most players feel more joy at accomplishing something that's difficult than something that's easy. If this is so, then when they finally work their hides off to get to, say Wolf 359 and find that there's nothing but a couple of rocks and gas clouds, they may be bummed.

That said, you probably won't have this problem because you're going MMO. Even in the wastelands, I'm sure you'll find players doing everything from chatting to jousting to prove who the best clan is.

In a co-op or single player RPG context, though, this problem is much more pronounced. I'm going to post another journal entry about how it might be better solved.

Quote:
 That said, you probably won't have this problem because you're going MMO. Even in the wastelands, I'm sure you'll find players doing everything from chatting to jousting to prove who the best clan is.

I'm not that confident. In a universe made of millions of systems and assuming a good success (say, 1000 players online at any time), the probability to find another player in an isolated system is close to nil.

Maybe the solution is to reduce the universe size ? Eve still has a few thousand systems.

Most players, me included, probably never visited more than a hundred systems in Frontier. Yet i'm pretty sure that if only 100 systems were available in Frontier, the game would have been less acclaimed and popular. A coherent, huge universe, helped to maintain the illusion, even if there was nothing to do in all these empty systems.

I'm looking forward to reading your thoughts and possible solutions to this problem.

Quote:
 Most players, me included, probably never visited more than a hundred systems in Frontier. Yet i'm pretty sure that if only 100 systems were available in Frontier, the game would have been less acclaimed and popular. A coherent, huge universe, helped to maintain the illusion, even if there was nothing to do in all these empty systems

Quote:
 Frontier is overrated

I agree on both points. I consider that Frontier was overrated in the fact that it quickly got boring. Sure, there was a whole universe out there but there was NO incentive to visit any of it.

Even the hyperspace mis-jump was pointless; it was merely a prompt to reload an old game. Instead, I think you NEED pockets of interest through the galaxy. Sure, a lot of it may be random emptiness but you still need the incentive to explore further in the hope that you WILL encounter something of interest that will give you a couple of hours, at least, of your time.

Quote:
 I'm not that confident. In a universe made of millions of systems and assuming a good success (say, 1000 players online at any time), the probability to find another player in an isolated system is close to nil.

The awesome thing is that you have what famous science fiction author Larry Nevin called the power of "ballonium."

Unless you're hewing to strict, rigid models of physics (which don't even permit rapid interstellar travel), I think you have wide latitude in bringing players together.

For instance, what if:

- Superphotonic travel is based on some quantum superpositioning concept involving linked drives (adapting this from an idea by Dan Simmons, author of the hugely successful Hyperion, btw). The human race has only managed to ever produce one drive, even though it manifests itself in various instances. As a side effect, jumping temporarily brings players into proximity with each other for periods of time. The rules governing this probability could be random or strategic.

If that's too hairy, what about:

- Warp travel by all ships is dependent strictly on a massively expensive mobile jump carriers. These singularity casting mobile stations are homes to clans (given at start), and like dreadnaughts in World War I so expensive that they're NEVER risked in combat. Rather than fixing your network of wormholes around stars or planets, the jump carriers provide a mobile nexus that can send players anywhere (at exponentially expensive energy costs).

The carriers themselves are always on the move, continuously scavenging for resources. This ensures that players travel throughout the galaxy, but because of the magic of singularities, are never far from each other. Allied carriers could link to each other much more cheaply, and enemy carriers could be raided for supplies. Planets would provide temporary resource bases, and could provide interesting and unique gameplay but without long-term investment and management (or even data storage!)

Mixed with #1, maybe jump carriers are the only way a ship can travel between the stars. Ship drives get connected to a carrier when they join a clan. Being expelled from a clan would mean that you need to negotiate to join another. (And there could always be those rogue / nefarious AI clans that, in the mold of Mos Eisley, take all comers and do everything...)

Just a thought...

## Create an account

Register a new account