Hi. So, I'm working on building a 4X game (Explore, expand, exploit, exterminate, your typical empire-builder Civilization-ish thing). It's largely inspired by Master of Orion 1 and 3, and at this point I'm sticking very closely to the tried-and-true paths of such games, mainly because it's simple to code, tweak and modify. With the legacy of a dozen abandoned half-finished projects behind me, I now tend to focus first on getting something that works, then making it neat later. Anyway, I was just curious if anyone had any ideas of things that might make life either easier or more fun for a game of this genre. Details, in other words. For instance, in Space Empires 4, it was very cool that you could use special weapons to disable the engines or shields of enemy ships, but it was not very cool that each ship had to move and fire one at a time; a big battle could take twenty minutes. Civilization's cities-make-buildings-which-make-cities-better construction system has become sorta a default fixture by now, something to revert to when you can't think of anything better. In Master of Orion 1, capturing Orion was sort of a game-long goal, and the rewards when you finally beat the Guardian were definately worth it. And so on. Basically, all 4X games essentially have the same goal: conquor the world/galaxy/universe/whatever. What makes them interesting is the system of constraints and advantages and trade-offs that the player makes. What kind of neat ideas would you want to see, that would make a game of this type fun to play? Will roving bands of pirates be a major threat? Will attention toward precisely balancing your colony construction be necessary for victory? Will you be able to build galaxy-spanning stargates, or will such devices utterly wreck the balance and fun of the game? I'm curious. And probably talk too much without actually saying anything. Tell me what you think.

I'm toying with a 4X game as well. One of the things I'm struggling with is the tech tree, and trying to fashion it in a way that produces more creativity and innovation from the player beyond just clicking through a pre-determined technology path. I think that this is something that needs some improvement in most 4X games.

I've also been frustrated with the menu systems of some games, MOO3 in particular. It's a careful balancing act between providing the player with options and choices, but making the interface intuitive and easy to use. Burying options in several levels of nested menus is annoying and frustrating, so keep your interface as intuitive and easy to navigate as possible.

If you include the option to negotiate with other civilizations, please spend as much time and energy on the AI as possible. Nothing is more frustrating than getting drug into a war that has no cause, ends just as mysteriously, and then starts up again with no warning on the next turn. If your AI can negotiate, trade, and communicate with each other and with yourself at a reasonably realistic level, you've got yourself a winner.

I don't have answers (and of course have a hobby project or 2 even in this domain as well) ... but I have good questions:

1. You should think about what level of AI players you NEED to support or CAN support when designing the game, because not needing AI players makes development a whole lot faster - but pretty much kills single player, and when designing a game that plans to use AI players, you must think about that during the design of your data interfaces (AI players need interfaces that allow looping though the information available to them and/or navigating between related pieces of information).

2. You should think about what level of diplomacy / relations you intend to support - as this and victory conditions are largely intertrwined, and this also has a significant impact on the AI subsystem. Large fuzzy diplomacy systems can be a big aid over the old silly 2 or 3 states (war / piece / friend) of most games, but they also add enourmous design thought, and potentially ruin a game if done wrong.

... more later, gotta work

One solution I've seen to the "boring tech tree" problem is a tech tree that has bits left out of it... you only end up getting a certain, random, percentage of it. This, at least, makes things more interesting. Combined with other interdependancies and different technology based on race or player decisions, I think it can be all right.

The interface I am DEFINATELY going to keep as simple and compact as possible. MOO3 annoyed the heck out of me too. "Oh, this planet finished a ship, let's build another... double-click on the star, double-click on the planet, hit the "build" tab, hit the "Yes, I want to build something" button, find the ship, and repeat the process in reverse..."

Ugh.

Anyway. I definately am going to have AI's, and i would certainly be nice to have them be as advanced as possible, I've never actually DONE any AI programming, so I'm not sure how it'll turn out... I will definately make the effort though. Maybe a system of attributes (aggression, expansionist-ness, diplomacy, mercenary-ness, etc) and goals, and victory conditions beyond "kill everyone". Alpha Centauri did a pretty good job of that, actually. I'll have to think about it.

Thanks. Any other opinions are welcome.

Greetings!

I am yet another who is currently working on a 4X game, specifically a space-strategy game. First off, I'd like to build upon what others have said so far:

1. In any game (not just 4X games), interface is VERY IMPORTANT! I can't stress that enough. If you play games as much as (or more than) you make them, you know what I mean. The 4X game with the best interface, in my opinion, is Ascendancy. Very intuitive and easy to use. I personally intend to follow its lead. :P

2. Most 4X games so far have had fairly limited diplomacy options w.r.t. the AI players. Multiplayer games, if they have some kind of chat function, are much more interesting because the diversity of relations between players is much greater. So expanding the diplomatic options for human-AI relations is definitely a good thing, IMO.

3. MOO3 sucks. I wouldn't consider it at all when making a 4X game.

Okay, now for some "new" thoughts:

1. Make the AI play by the same rules as human players. Many (if not most) games allow the AI to cheat like mad, especially in the harder levels. Trying to avoid that is a good idea, I think.

2. Have multiple ways of winning in the game. That will definitely enhance the gameplay because there are that many more strategies available.

3. Customizability. By that, I mean have a lot of game setup options that players can tinker with. That will enhance the replay value of the game.

That's about it for right now.

- Rob

Man, I almost forgot something: planetary (city, colony, etc.) management.

This is probably the hardest element to do right in 4X games. There's probably no perfect way to handle it, since different players prefer different levels of micromanagement. However, there are a few "shortcuts" around this:

1. Implement a build queue. That way, players can plan ahead and not have to worry about dealing with each planet every time it completes a new building, ship, or whatever. If they want to do that, then they don't have to put more than one item in the queue at a time. Personally, I find that games without this ability become very tedious once I have many colonies/cities - Civ2 and (sadly) Ascendancy come to mind.

2. Implement an autobuild function. This is a layer of abstraction on top of the build queue - it allows players to not worry about planetary management AT ALL, if they so choose. However, it's important to make the autobuilder pretty smart. Ascendancy has one (called "Self-Management") but it's horrendous. MOO2 also has one, but it tends to just build everything. Having a selection of different objectives (e.g. whether to maximize farming, industry, or research) can help.

3. Have a way to filter out game messages. IMO, pretty much every type of game message should be filterable. That way, players can pick and choose which events they want to deal with.

- Rob

What I'd like to see is something that I saw in the webgame Warring Factions. Basically, when you start a research project, you set certain areas which will determine what kind of blueprint you'll get. It may have different weight, different cost and different damage / speed / whatever, which makes designing your ships really interesting and also allows for totally unique designs.

Maybe this wouldn't replace a normal tech tree, but it could go along with it (for example by allowing you to research new kinds of technologies, or new features that can be found in them etc.).

It does make for a quite complex game, but I just love the individuality of it.

RobAU78:
I mostly agree with you; I intend to do a lot of toying around with the interface and diplomacy parameters. I haven't played Ascendancy, but I think I should now. However, the thing you must realize about MOO3 is that while it sucks, it does so in a very unique and particular fashion. It's lots of good ideas put together in a supremely bad fashion. One of the things I liked about it was setting ships up into combined fleets, which is something I'm going to use in my own game... after making it MUCH less cumbersome to do.

Multiple victory conditions is definately a win, I think... hmm, especially if different races are better suited to different strategies.

Build queues and autobuild orders are definately going to be included. One of the nicer things about Stars! was that you could build a bunch of colonies, set autobuild orders, and not worry too much about them while they built themselves into major production centers.

Wuntvor:
I've looked at Warring Factions, but don't really understand what you mean with the research system... it seems like you could say "Give me a better ship hull, I want speed and manuverability!" and get a design that's about the same tech level of, but very different from a hull you might get if you said "I want armor and weapons!". Is this sorta how it works, or am I being original for once? Regardless, it sounds like a neat idea...

One other thing I'm designing, and rather having fun with, is species building. I've sort of condensed it from a system of myriad attributes into four "categories" that each species chooses from: Physical, mental, social, and evolutionary. All of these can affect many different things. Physical is how the species is more or less built: are they big and strong, weak and sickly, small and sneaky, etc. Evolutionary is sorta what role they fulfill: predator, prey, parasite, hive, and so on. Mental is how they think about the universe: geek, tactician, diplomat, hippy, nomad... Social is sort of the catch-all, and the only category where the player is free to choose as many attributes as they want, each with a specific advantage or disadvantage. There are generally pairs of things: lucky/unlucky, social/isolationist, wonderful diplomats/horrible jokers, famous/anonymous, and so on. There will probably be a fairly simple point system governing advantages and disadvantages. It'll certainly require a lot of tweaking though, and I may just like the idea of making a sneaky, parasitic, fascist species who are just so charasmatic that everyone likes them anyway.
Hmm, the "social" attributes may be better implemented as a series of sliders, rather than absolute properties... That gives one more leeway.

Anyway, does this sound interesting/fun/worth playing, or overly artificial and just too silly?

Every 4x game must, must, must have a random map generator!

Oh, and moo3 did suck, but that was because of implimentation, not due to ideas. For example, the idea of customizing race via species and specifics seperately was a great idea towards tackling issues common to other 4x games. Unfortunately if you make one species totally unbalanced, it's not really going to help much...

O.K. Check this out, I'm tired and have only read the initial post, but in a civ (sorry 4x) game - have the evoluution decivilize (as Easter Island, see the book "Collapse" by Jared Diamond (as featured in this months Wired, etc...)) and thus have a result (God I can't beleive I'm saying this as I have only seen maybe 2 Star-Trek movies) have the more primitive weapons defeat the newer defense technologies. As in the Borg can't be killed by phasers by having adapted, but still are killed by ye ol bullet. I don't know if this is what you are looking for, but please, if it is not ignore it. I am heavily intoxicated and do not want my rating lowered for this. Hope it helped anyway.