Seeking Harmony In Virtual Interplanetary Conflict

The later half of Wednesday was consumed by winning the upper hand in a game of Space Empires V that, surprisingly, took the entirety of Thursday to finish.

For time to fly so quickly indicates that there was definitely some gaming bliss here. However, I picked up many reminders of things I should avoid in developing my own game, too.

1. In monolithic games, bugs are oft hard to quash

There’s been many, many patches done to Space Empires V since its release in 2006, but I’m surprised to see that there were still a few things that were there even when I was playing the demo over a year ago.

For example, the “Point Defense Cannons” are supposed to be rewarded from “Military Science” level 1, but show up even without it researched.  It really sticks out when I’m noticing I’ve suddenly access to point defense weaponry about 16 times more effective than I should be using.  The tech tree could really use a re-org to avoid issues like this. (I should really give the game a try with Captain Kwok’s balance mod.)

Though crashes in v1.77 were rare, manually transferring ordinance and supplies from deployed fighter groups seemed to invite them.  Thank goodness the game has a “save every turn” option.

2. Providing Multiple Means To The Same End Are Perilous To Balance

The tech tree in Space Empires V is very ambitious, but it is somewhat tricky to balance because many of the technologies are different ends to the same means.

The issue with Fighters dominating the game is a tricky one to conquer for this reason.  Technically speaking, Fighters do the same job as attack ships, except they’re smaller and can’t warp.  By the time you factor in their greater speed and evasion, a couple dozen fighters can dominate entire fleets of full-sized ships for a fraction of the cost.

The trouble is that Fighters are largely redundant.  They’re just more efficient ships – they generate more offense and defense per maintenance and development costs than normal ships.   In that case, what’s the point of having ships at all?   A quick fix would have been giving fighters a fatal flaw, such as making point defense cannons more effective against them.  However, so long as fighters are redundant, the core trouble remains: how can you balance two different things to perform the same role?

Ultimately, fighters need some kind of unique role – a purpose of their own for being there.  For example, perhaps fighters could only attack other fighters, missiles (for point defense) and hostile ground units.  In space, having fighter superiority means your missiles are the only ones getting through.  On the ground, troops (another kind of unit) would be required to capture planets, but Fighters would be a useful tool for clearing out hostilities.  In this scenario, bringing along a carrier for fighter superiority carries a completely different purpose, and one that does not knock the balance out of whack.

It’s not just an issue with Fighters, it’s something that can be seen in many places across the Space Empires V tech tree.  In another example, by bringing Meson Blasters and Incinerator Beams together on the same role from different tech trees, one or the other was doomed to inferiority.  If Monolith facilities harvest all three resources, why have tech trees for facilities that take the same amount of space to harvest only one thing at a time?

In retrospect, perhaps the best approach to organize the diverse technologies in Space Empires V would have been specialization trees decided at the start of the game.  The way the tech tree is organized, an empire could specialize in Propulsion and end up with some high-energy beam weapons.  Perhaps choose three categories from Theoretical Science that your race knows, and all others are off-limits for your race to research.   Each category would include its own strengths and weaknesses, and there would be a general pool for basic needs of all races.  It wouldn’t be as open-ended, but it would nonetheless be more interesting than the mishmash in the game now.

3. Micromanagement In A Turn-Based Environment Leads To Poor Pacing

It really shouldn’t have taken me all day on Thursday to finish the game I had already won on Wednesday.  All I did was roll my fighter carriers over the enemy planets, system by system, pausing only two or three times to upgrade and refill their fighter supply a few times.

The reason why it took so long – perhaps a dozen hours or so – was in the details.  I had to do a lot of fussing around with build queues on planets all over my empire.  Building and dispatching colony ships to claim the enemy space.  Upgrading designs as new technology came available.  Rifling through cargo storage, planet-by-planet, recycling obsolete fighters.

Each system I took was just another batch of planets I now had to keep running right.  However, it wasn’t just excessive micromanagement which bogged down the game, it was partly an issue with the GUI improvements and code optimization that needed to be done.

The GUI should have included some quicker means to perform this kind of empire maintenance, perhaps a “scrap all obsolete designs” button.  There was an “update queues” button, but it needed to include (or have the option of including) updating things currently being built – without, many building queues set to repeat were left unaffected.

A need for code optimization was evident in several places.  I run the game at desktop (1280×1024) resolution on high detail on a pretty up-to-date computer, and this is actually quite laggy – maybe I should turn things down a bit.  Non-graphically, my 2.4 Ghz CPU (the one of the four cores being used) generated a few seconds of lag every time I pulled up the colonies list, accumulating several minutes of wasted time per turn.  The unit list was even worse – it could take half a minute or more for all my fighters to be found and the unit list populated – I had to avoid using that control entirely.

Even better than improvements to make micromanagement easier would be options to eliminate the unnecessary micromanagement entirely, and this is the job for what Space Empires V calls “ministers” (what Master of Orion 3 called “planetary governors”).  This is the AI taking control of selected portions of your  empire to eliminate a lot of micromanagement.

The trouble with the ministers is that they’re all pre-programmed and it makes doing exactly what you’d do impossible.  Perhaps I’m spoiled – a lot of what Space Empires V offers the player is in customization, and here too is somewhere where complete customization would have been nice: the ability to customize a minister’s behavior to do exactly as I would, automatically.

For example, I’d love to customize a planetary AI whose sole function is to simply put one facility upgrade in the queue if the queue is empty and an upgrade is available, thereby keeping my planets upgraded while not filling my queue with upgrades that are about to be made obsolete anyway.   In the game I just finished, I could not easily find nor count the number of facilities that needed upgrading – many of these upgrades would have paid for themselves thousands of times over.

In another example, scrapping obsolete planetary units (such as fighters or troops) and rebuilding them one by one (oldest design first) to automatically replenish unit levels to presets.  This would have shaved hours off my game, kept my maintenance costs from spiraling out of control by making it unnecessary to set queues to simply repeat, and probably would have made the enemy AI more challenging too.


Space Empires V emerges as an extremely ambitious home-grown product which probably would have been a bit better if they consolidated their tech tree into clear upgrades and streamlined some of the more obvious choices.  For example, what should I upgrade?  Everything.  When?  Whenever I have the resources and means to spare and there’s not a better version coming before the upgrade would be complete.

Still, in examining of the game with a game design mindset, I can really sympathize with 4X Game designers.   These endeavors are major, extremely sophisticated on several levels, about the game design equivalent of developing a whole motor engine where many game designers are content with developing a spark plug.  As far as engine analogies go, Space Empires V has a lot more horsepower than most, even if it isn’t perfect.  I would try a Space Empires VI.

It definitely gave me some food for thought as to my own game.  Things such as:

  • A feeling of accumulation really keeps a player riveted, but ultimately leads to a one-sided game.  In allowing the players to accrue power, new challenges should be introduced.  The tricky bit is doing it in such a way that the players get to feel good about having accrued power: if the bar simply moves away, why reach for it?
  • AI can be developed to control individual units in order to keep AI routines simple and modularized.  However, without a single leading AI to give the units direction, they’re unable to work well for a greater whole, and are oblivious to the mistakes of others.
  • Exactly what is the goal of the player, anyway?  In a 4X game, it’s to become the most powerful, ultimately dominant, empire so you can win the game.  In an RPG, it’s often a matter of accumulating, but there’s rarely ever a game to win.  In terms of the real goal, having fun, either approach has longevity flaws.   In a 4X, there reaches a point where domination is assured and the rest of pointless.  In an RPG without end, monotony is inevitable.

I hope to spend Friday thinking heavily about shaping my BYOND endeavor to suit my chosen answers to these questions.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: