Brad Wardell's views about technology, politics, religion, world affairs, and all sorts of politically incorrect topics.
Space is big
Published on June 24, 2005 By Draginol In GalCiv Journals

I read on a forum a concern from a user about Galactic Civilizations II -- that because the planets are now part of the main game map, the game was now so unrealistic as to be distracting for him.  The problem is, any space-based game is going to have to make some concessions in order to be fun.

Here's the problem: Space is big.  No, I mean BIG.  It's ridiculously, mind boggling, incomprehensively big.  It's just plain enormous. 

Some years ago, we released a game called Stellar Frontier. During the beta, the team tried to keep it realistic. Your ship could travel up to the speed of light.  Light travels pretty fast right?  The game took place in our own solar system.  Which meant if you wanted to travel out to say Pluto, it took SEVEN hours at maximum speed.  That's how long it takes light to get from the sun to Pluto.  Your ship traveled very fast on the game-screen, it was just that the planets were spaced really far apart.  It was one of the first massively multiplayer games and we knew that if the points of interest were so isolated then people would never..you know..get together to kill each other.  So we had to make some changes.

In Galactic Civilizations I, we didn't display planets on the main map.  So I suppose purists could imagine that the ships were symbolic (obviously not to scale) and hence since we were dealing with stars, it was okay since star to star relations could be rationalized.  But now we have planets on the map too and all sense of proportion goes out the window.

So what if we wanted to have it both ways -- what if we wanted to have planets on the map, to scale even, and distances between stars still relative.  Let's imagine that for moment.  A Jupiter-sized good planet would look something like this:

Okay, you're going to need to click on that thumbnail to see the planet because it's so tiny.  Good thing we're talking about our puny sun (Sol) and not some other star.  You see, a red giant, for example, is about 1000 times wider than the sun.  So if we were going to maintain any sense of scale, you would need to have the red giants be 1000 times bigger.  Now in this screenshot, our sun is only 296 pixels. Not too bad.  A red giant at the same magnification would be 296,000 pixels.  Scrolling across the screen at say 500 pixels per second (typical scroll speed in an RTS) it would take approximately 10 minutes to actually scroll the diameter of a red giant.

But that's okay, because compared to our other realism problems, a 10 minute scroll speed is nothing. There's a dirty little secret sci-fi has been hiding from you all this time.  Not just games like Galactic Civilizations, Space Empires, Master of Orion, etc.  But sci-fi like Babylon 5, Star Trek, etc.  You see, it turns out that most star systems are binary.  That's right, there are two stars involved in the star system.  That's not the exception, that's the norm.  Alpha Centauri is just one of three stars that make up its star system.  That's right, THREE stars for a single star system.  That's more common than a single star with a series of planets around it.  What's worse, we're not even sure our star system is truly a unitary star system.  There are many astronomers that believe that beyond Pluto there is something huge that is essentially a star that didn't quite make it. Basically our own dead star that messes around with the orbits of the outer planets a bit.

Our own star system, Sol, contains (As far as we know) 9 planets.  With Pluto being the last one (and it's barely a planet).  But that's not the end of our star system. A big pile of debris known as the "Oort cloud" marks the exit of our home star system.  And it's really far away.  And I mean really far away.  Pluto, it turns out, is only 1/50,000th of the way there.   Consider that for a moment -- if we had any sort of scale, let's say it took our ship 5 moves to go from the first planet in a star system to the last planet in a star system.  It would take another 250,000 moves just to get to the edge of that star system.  Let's say you wailed on the keyboard as fast as you could -- 4 times per second, it would take you 17 hours to get to the Oort cloud and get into interstellar space.  That would definitely affect our reviews I suspect.

Of course, we've only discussed getting out of our own star system.  Now we want to colonize the Alpha Centuari system (which is made up of Alpha Centuari, Alpha Centuari B, and Proxima Centauri as you may recall).  It's "only" 4.7 light years.   I'll spare you how many months you'd be scrolling the screen to get over to that.

You also have other scaling issues:

From a purely size point of view, the planets vary greatly in size.   But in distances, it just gets more and more painful.  Take this picture for instance, if they were spaced to scale, Jupiter would be about a 1000 feet away.  Pluto would be nearly a half-mile away.  And the next star system? 10,000 miles away.  Like I said, space is big. It's really really big.  It's so big it's just ridiculous. It's like wiping your nose in its bigness, saying to you "Hey, loser, look how big I am!" and then for no apparent reason adding a gratuitous etra "Loser!" after --  leaving you wondering whether space has some sort of..you know, inferiority complex about being so empty and all.

So from a game-play point of view, realism is just not very fun.  Besides, even if we did try to be more realistic on scaling, then we'd open the door to all the other areas we've taken creative liberties (like the fact that Earth seems to reside in random galaxies full of life and that we have a magical "Hyperdrive" to fold space and let us get around it so fast). We have to balance realism and fun. It's more fun to have the planets be on the map. It saves a click, it allows for intuitive multi-civilization ownership of a star system. It allows the visuals to be a lot more exciting. In short, it makes the game more fun which, at the end of the day, is our goal. 


Comments (Page 3)
on Mar 11, 2006
I honestly think hard science (Or at least theory.) Is far more cool and interesting with much room for stories and ideas than the plastic science most Sci-Fi throws at you. (Silly Trek.) That being said, it works better with something like a novel or a movie, and not so much a strategy game. See, I can just buy it being a tactical display, or better yet, buy that only the important parts that I'd interact with are on the map. In a novel, you have a vast infinite frontier and sea of stars to explore. Space is a big place for a threat to lurk and hide, it's foreign, massive, and imposing. And that is awesome. Give me binary star systems and time dialation to 'Scotty saved the ship by inverting the magnetic...' whatever stuff any day. But for a video game? Seriously, anyone who demands complete realism in a game where there is a minor race of evil squirrels and a shameless Pokemon ripoff needs to cool their jets and stop being so picky.
on Mar 11, 2006
There is also one other point of realism that is glosse dover for the benefit of simplicity: There is no (or very little) friction in space. Once a ship starts moving it only stops with another firing of its thrusters. Strictly speaking engines on a space craft should provide accleration not movement. this means that the longer the ship travels in a given direction, the faster it would get, decelerating again once it passes the halfway mark. It would also mean that it might take several turn for a ship to change course once it was moving. This would also make navigation of fleets (let alone whole taskforces) borderline unmanageable. Thus it is excluded, and fair enough too.
on Mar 13, 2006
If we were going to be realistic, there wouldn't be any interstellar travel at all, because matter can't travel faster than the speed of light (Star Trek, Star Wars, and Battlestar Galactica notwithstanding).
on Mar 14, 2006
f your looking for a good read along the same lines as "The Forever War" read Heinlein's "Starship Troopers" or John Steakley's "Armor". If your looking for insystem space battles read Dickson's "Dorsai"



I read Armor last year and it's by far my favourite out of that lot, though it's been about 20 years since I read any Dorsai (think I read the first couple only, I stopped reading SF for a long time. Hitting my second childhood now.)



And that guy wanting the interesting anomalies - pshaw. GC1 had space monsters that hatched out of anomolies. Only the AI actually investigated them because humans know which anomolies they came from so we'd leave them alone until some AI explorer came along and unleashed them on our colonies. I do not miss them. Um, they are gone right? I stalled in the GC2 campaign out of tedium at reresearching for every new map. Didn't feel right.


Posting on the TGN forums bears an eerie similarity to the pace of the war in The Forever War.
on Mar 15, 2006
I don´t agree that a certain level realism isn´t needed in videogames. Why bother at all with concept of spaceships ? Take cars as the vehicle to move between stars. Aliens: why bother ? Fish as pictures work as well.

I really don´t know why Stardock did choose the label "week" for turns and the "population increase" in billions in a year...

It s a good game but some decisions seem strange to me.
on Mar 15, 2006
WARNING: LONG POST COMING

You've been warned.

First off, who am I and why should you care what I think? Maybe you shouldn't care what I think, but I'm a strategy game enthusiast who still plays the original MOO as its the best game in the genre(I'm definitely looking forward to trying out this game, but as of right now its unavailable in my area and since I don't own credit cards as a matter of principle, I can't buy it online). When I say I prefer MOO though, I'm talking gameplay because I'm among those who prefer ugly fun to pretty crap.

I am an amateur programmer and modder, I tried for over a year to salvage MOO3 until I gave up on it(those familiar with the community may recognize me as the author of PrimEpic Mod, aborted due to the irredeemable economic model, particularly as it concerned the AI). Anyway enough of that, I just thought that the devs, if they read this, should know my perspective and where I'm coming from. I'm going to respectfully disagree with the realism-doesn't-matter crowd, and please don't think I'm trying to pick apart the game: from what I've seen it has a lot of promise and does many, many things well. What's more , I might be able to more fully implement a PrimEpic theme here than in any other game genre. More to the point though, I believe Stardock just might have the commitment to strategy gaming to eventually make the kind of game I'd REALLY like to play, which is why I'm taking the time to elaborate here. I should also point out that in no way do I expect these suggestions to be considered the least damn bit for this game: I'm hoping for the next iteration or two in the series.

So anyway, the central thesis as posted by Avatar Draginol regarding realism:
any space-based game is going to have to make some concessions in order to be fun.


I totally disagree ... sort of . On the interface, obviously you do have to change some things, but I really like what I've seen so far -- being able to access planets on the map, etc. As far as I'm concerned, this has no impact on realism at all though, as long as really big objects are larger than small ones, even if not to scale. Imagine yourself as a Galactic Dictator centuries/millenia into the future. You certainly wouldn't want a strategic map prepared by your advisors that had actual scale either: you'd want one that allowed you to see the picture in a way that you could see the picture at a glance and make intelligent choices on fleet movement, where to assign personnel, etc.

Where realism becomes an issue is in the game mechanics. And I believe the flaw here is whole turn-based style of gameplay. Because if you depend on that assumption, than what Draginol said here is absolutely true:

if we had any sort of scale, let's say it took our ship 5 moves to go from the first planet in a star system to the last planet in a star system. It would take another 250,000 moves just to get to the edge of that star system. Let's say you wailed on the keyboard as fast as you could -- 4 times per second, it would take you 17 hours to get to the Oort cloud and get into interstellar space. That would definitely affect our reviews I suspect.

Of course, we've only discussed getting out of our own star system. Now we want to colonize the Alpha Centuari system (which is made up of Alpha Centuari, Alpha Centuari B, and Proxima Centauri as you may recall). It's "only" 4.7 light years. I'll spare you how many months you'd be scrolling the screen to get over to that.


Consider for a moment though that turns of a fixed time period is NOT the only way to do it. I propose an alternate solution. I've been wondering for years why no game has done this, and I'm finally going public with my curiousity/befuddlement. I propose Variable Time Control(VTC). A game made with VTC would not have an End Turn button that moved the game to the next week/year/Galactic Cycle/etc, but would allow the user more control in the time advancement. The player could queue up whatever orders, make whatever fleet movements, and then hit the Advance Time(or whatever) button and input the amount of time they wanted to progress. Under this method it would matter little whether it took a month or a century to travel out of your solar system or across the galaxy or whatever: you wouldn't have to click a button tons of times, you would simply control the flow of time yourself. During key times of war/upheaval/whatever you could advance days/hours/even minutes, during other periods you might want to go decades or more. There could be various interrupt conditions you could set to stop the sim: declaration of war, assassination, new tech advance, new planetary structure, new fleet/ship construction, etc, it wouldn't take long to come up with many more.

There are many more ideas I could throw out, and I haven't so much as seen the game in action yet. The essential point though is that realism from a GAMEPLAY perspective is not necessarily at all at odds with fun for the strategy gamer: and if you are talking TBS, you are talking a strategy gamer, not usually a rapid-fire mouse-clicking spamfesting RTS enthusiast.