Galactic Empires, Part Two: Expanding into Space

by Shannon Appelcline

March 7, 2002 – Next week, if I've paid all my karmic balances correctly, we'll be releasing a new game at Skotos called Galactic Emperor: Hegemony. It's a strategic game that's tested and works well, and we're also going to be doing our best to expand it into a roleplaying arena.

I think each of those topics is worthy of discussion in and of itself, so I've decided to split my discussions of Galactic Emperor: Hegemony into two parts. This week I'll talk about the strategic game that is, and next week I'll ponder on the roleplaying game that could be.

Clearly a strategy game is a very different thing from the online multiplayer roleplaying games that I usually cover in this column. However, I think much of the process of game design is quite similar. You make decisions, you implement them, and players react in certain ways, which you must in turn react to. As you move through several iterations, hopefully your game improves.

So, let me tell you about the strategy game that would become Galactic Emperor: Hegemony, both as we originally discovered it, and through some changes that we've suggested since.

Beginning with Space Empires

The heart of Galactic Emperor: Hegemony is what's billed as "Play by Web" game named Space Empires. It was designed, developed, and engineered by just one programmer — a gentleman by the name of Richard Wolfe. It's a simple CGI-based game with a combination HTML/Java interface. It's a game of combat where you have a three-week period in which you try and take over the galaxy.

Spaces Empires has very simple mechanics and a very simple interface. That's much of the beauty of the game. You and eleven other players start out in a galaxy of 150 worlds. On your homeworlds, each of you have 250 starting ships. Surrounding your homeworlds, and creating an initial firebreak for the impending hostilities, are 138 neutral planets, each protected by a tiny little garrison of between 2 and 20 ships. The object, clearly, is to take as many of those other planets as you can with your fleet of ships. Ships can do two things: move and fight.

Movement is limited by distance (initially 6 parsecs) and speed (initially 3 hours per parsec). All this ship movement happens in real-time, but it's a very slow real-time. Planets tend to be at least 3 parsecs apart which means that transport from one planet to another would take at least 9 hours at the start of the game.

Combat is mostly based on raw numbers of ships, though there are a few additional factors like defensive bonus, home planet bonus, and a player's battle bonus.

To add a little complexity to the game, you also start off with a small sum of money, around $200 or so if I recall, and that money can be used for a variety of things: building factories which create new ships every day ($5 each); increasing the range of your ships ($5 per .1 parsec); increasing the speed of your ships ($20 per 5 minutes/parsec increase); and increasing battle ($10 per point). Every planet you control also adds to your daily wealth.

You can also buy spy probes ($10) to figure out the contents of a planet (wealth, factories, and ships) and a spy shield to block that ($30); and a death probe to totally nuke a planet ($200) and a death shield to protect it ($200).

And that's pretty much the game in a nutshell.

Christopher Allen found Space Empires late in 2000. I played a trial game in December of that year, then a regular game in 2001. I found it very addictive, as did several of my coworkers. It was a rarity, a well designed free game on the Internet, so we started talking about adding it to the Skotos site. Clearly that process took a while.

As for what made Space Empires a good game ...

Success Through Complex Simplicity

In Trials, Triumphs & Trivialities #39, Ah, Sweet Simplicity of Life! I mentioned that one of the chief criteria in creating a good game was simplicity, but last week in Trials, Triumphs & Trivialities #62, Galactic Empires, Part One: Failing at Succession I noted that that really meant a small number of systems, not simplicity within those systems — as the latter would lead to boredom, and thus game failure.

I think Space Empires starts out on a really good food by keeping the number of game systems low, but also having some variety in each of them. Here's how I'd lay all the game systems out:

1. Major System: Monetary System.
   • Wealth from Planets.
   • Starship Related Purchases.
   • Spy Probe Related Purchases.
   • Death Probe Related Purchases.
   • Interest for Money in Bank.
2. Major System: Starship Combat.
   • Factories & Daily Ship Production.
   • Ship Speed Variability.
   • Ship Range Variability.
   • Ship Battle Variability.
3. Minor System: Spy Probes.
   • Spy Probe Purchase.
   • Spy Shield Purchase.
   • Information on Monetary & Starship Systems.
4. Minor System: Death Probes.
   • Death Probe Purchase.
   • Death Shield Purchase.
   • Destruction of Planets with Starships, Factories, and Wealth.

As you can see there's a very small number of game systems in Space Empires — just two big ones and two small ones. However, each of these systems branches out to multiple choices, and there are multiple interrelations between the systems. Thus, simple design leads to complex strategy.

For example, wealth can be used to purchase more ships (via factories), or better ships (via battle), or ships which can be more places quicker (via speed), or ships which can get to further away places (via range). Which is the best investment strategy? Some of these answers are pretty easy — for example, range usually isn't that useful until the exact moment that you want to get somewhere that you couldn't otherwise get to. But what about the other decisions, between factories, battle, and speed? Typically factories seem to be the best answer, but in the last game I played my maxed out speed helped me destroy an opponent's financial empire. There are multiple right answers, and that's what makes the strategy compellingly complex.

The interrelations between the systems just multiply this complexity. Do you spend money on spy probes to judge an opponent's strengths or save the money for some manner of ship improvement and attack blindly? Do you leave your own wealth and factories vulnerable to attack in order to stage an assault upon your opponent on the chance of crippling him? Do you spread out your factories to avoid death probes or do you consolidate them on your home planet to make them easier to defend?

I think every player has his own strategy, and maybe multiple strategies at different times in the game, and thus the game stays compelling.

Some Other Things Space Empires Did Right

Hopefully the above helps to explain what makes Space Empires a good game. Quite simply, Richard Wolfe did his homework and thought carefully about how to design systems, keep them simple, and still make them interesting. Yet, there are a number of other things which contribute to the success of the game, and raise it up, in my book at least, from good to great.

Diplomacy. The one big system that I didn't mention above was diplomacy. It's kind of a touchy-feely system, with no actual mechanics, and thus to me it isn't quite a game system, but rather an extension of the multiplayer online medium.

Quite simply, after you've played a single game of Space Empires it becomes clear that you need friends to win — or at the least that you need to be carefully diplomatic. As you probably determined from my initial description of the game, every player starts out exactly the same. Thus if I attack Joeblow, the player nearest me, I should have approximately a 50% chance of success, modified by how our relative strategies work out.

However, if I attack Joeblow and convinced Fredbob on the other side to attack at the same time, our chances of success are going to be much higher. As the game expands and opponents begin to fall, the maxim remains true.

A game that was just good would probably have presumed that players would figure out ways to get ahold of each other and let it go at that. Space Empires, on the other hand, makes that Diplomacy prominent. It's one of the buttons on the standard GUI. There's also support for the transfer of message in-game. Giving diplomacy that prominence and making it that much easier ensures that diplomacy will occur in games, and thus adds yet another layer of complexity.

Regular games also have an added feature: you can give money to other players. This is yet another example of a designer interrelating game systems when possible.

After a regular game you can also choose to rate your allies and opponents, creating an EBay-like reputation system, letting you know who to ally with in future games, and who not to.

Partially Blind Games. Space Empires also takes advantage of another power of the online game medium, one I described in Trials, Triumphs & Trivialities #34, The Power of the Medium: Individualized Output. At any time you can see the state of the whole galaxy — who rules what — but for all those planets you don't control you don't get to see the information on their wealth or factories. You also can't see your enemies' fleet movements. This creates a delightful feeling of uncertainty throughout the game that you almost never find in a tabletop strategy game. You actually have to make guesses about what your opponents are doing, and try and figure out how to react best to those strategies, and this is yet another factor that adds complexity.

If you were making decisions based upon known facts, your path might be much clearer. But making decisions based upon presumptions, guesses, and careful analysis of your enemies is a tricky thing.

Clear Cause and Effect. When I was talking about Galactic Emperor: Succession last week I mentioned that one of our problems was that we muddled cause and effect. Players didn't know what reactions they could expect from their actions. In Space Empires it's very obvious. Your success against your opponents is based upon your ship numbers, your ships' battle prowess, and a few simple additional factors like defensive bonus. I know if I launch 100 ships at a planet I can generally figure to take out 90-100 enemy ships, and if I take out less it's probably because they're a lot better at battle than me, and I need to take that into account in the future. Simple.

Appealing to Bartle's Players. I've talked about Richard Bartle's categories of players a few different times — most clearly in Trials, Triumphs & Trivialities #25, Telling Stories, Classifying Worlds. Over in the forums, Christopher Allen suggested that one of the reasons for Space Empire's success was that it actually appealed to a number of different player types: "For the socializers among us the diplomacy and roleplay [are] fun. For the killers winning against real opponents is fun. For the achievers the high scores and stats are fun. For the explorers, wondering what was behind the next star, or what if I tweak my stats this way, is fun."

Limited Creativity. Finally, the regular version of Space Empires does something really cool: it lets players name stars as they conquer them. It's a really simple form of dynamism — a very low-level type of creativity — but to me it adds immensely to the game, because I can feel like I'm really creating something, really putting my stamp on the world, even if it's an ephemeral world that only lasts three weeks.

The Limitations We Saw

By now you probably have a simple picture of the game. Let me expand upon it just a little bit, to make sure we're all on the same page:

In Space Empires every player has a fleet of ships. He can move any number of ships from the planet they're on to any other planet within their range, which takes a certain amount of real time. At a game's start a player typically divides his fleet of 250 up into smaller fleets and jumps them out to the nearest planets, creating a spherical empire out from his home planet.

Along the way a player captures worlds which provide additional wealth. He also builds factories on planets he controls which provide additional ships. Ships are built and wealth is generated once a day at a set time.

Players typically form either alliances or non-aggression pacts so that when they finally start running into other players they can try and make sure they're in an advantageous position. Most games seem to end with 1 or 2 or 3 or 4 players in control of most of the galaxy, possibly with a few stragglers still hanging on by their fingernails.

The largest battles tend to be fought around homeworlds and around factory worlds. In the second half of the game things tend to get bloodier with death probes going off and nuking valuable resources.

When we at Skotos first played this game we thought it worked great, but we also saw a few problems in the design that could limit the game's appeal:

• The Starting Problem. Time is very valuable in Space Empires because of the real-time aspect of the movement. Thus if you lose time you can be at a serious disadvantage. This is a fairly big problem at the start of the game. If you logged in as soon as the game started and began moving your ships right away you'd be in the optimal position. If you waited over a day, you were fairly certain to lose the game.
• The Production Problem. Similarly, since wealth and ships got produced once a day you'd lose advantage if you didn't log in right after that production time, to get things moving and spend wealth.
• The Frequent Login Problem. Finally, you were also at some disadvantage if you didn't log in frequently. Ships would arrive at a destination and sit there, unmoving. As your empire (hopefully) expanded to fill the galaxy this could become an increasingly large problem, as ships would waste time at every stop along the way.

I've played three games of Space Empires to date, the third of which is just finishing up as I write. One of those games was a regular game, and the other two were trial games. During the regular game I was obsessive. I checked the game status hourly. I had my wisdom teeth out during the middle of that game, and for a while was having to wake up in the middle of the night to take additional narcotics for the pain. I checked in on the game then too, my dark office, lit only by the glow of my CRT. I was obsessive and I kicked butt, scoring the highest score ever on the game, while one of my allies scored #2. My ally's score held for a few months. My own has just been beaten now, in the last few weeks, after standing for over a year.

I don't think this was all a coincidence.

So, when we first started talking with Richard Wolfe about Space Empires we suggested some ideas to resolve what we saw as the game's only limitations, and thus make it more fair to a less obsessive player than I.

Here's what we came up with:

• Standing Orders. The simple ability to list, for each planet that you control, a minimum garrison and the planet which ships should go to if that garrison was exceeded. This allowed the creation of supply lines, as you could point ships from your factory planets all the way to your front line, and also ensured that your defenses stayed steady. This was implemented about a year ago, and seemed to resolve much of the frequent login problem. There are still issues, like the fact that your supply lines collapse if your enemies take one of your planets, even temporarily. And you can't really save up forces for an automated attack from your front. Still, it resolves 80% of the problem.
• Starting Orders. The simple ability to list where ships should go when the game starts. Recently enhanced by the ability to purchase speed or range before those ships head off. A 100% win for the starting problem.
• Gradual Production. A system to average out that daily production of wealth and ships over the entire day. It was intended to resolve the production problem, but in actuality that had already been 80% fixed by standing orders. The only remaining production-related problems were that you could get an advantage if you bought range, speed, or battle right after production and you could get an advantage if you staged a front-line attack with a large force right after production (i.e., this was the remaining 20% of the production problem). Unfortunately adding gradual production seemed to exacerbate one of our older problems. It increased the need to login, because there were always new ships to move and new wealth to spend.

I don't have much more to say on these issues. They were mainly intended to offer a very real world example of the type of problem analysis that should go into the design of any game and also the type of problem solving that should go into fixing things. I'd also make the following larger points:

• Many solutions will only get you 80% of the way there; that's often enough, and trying to fix the last 20% might be much, much more complex (or destructive to your original game).
• All problem solving will have weird interrelations. Some fixes will magically fix other problems. Some fixes will magically break other things.

So What's All This Have to Do with Skotos?

Although I hope you've all found this game design analysis interesting, I'm sure there's some of you who are wondering, "What's that gotta do with Skotos?" After all, Space Empires is a strategy game while Skotos is mainly an RPG community.

That's what I plan to talk about next week in "Building a Hegemony", the column originally planned for this time: why we decided to add Space Empires to the Skotos community and what we're planning to do to make it an even better fit.

See you then, and hopefully in Galactic Emperor: Hegemony before that.

Recent Discussions on Trials, Triumphs & Trivialities: