🎉 Celebrating 25 Years of GameDev.net! 🎉

Not many can claim 25 years on the Internet! Join us in celebrating this milestone. Learn more about our history, and thank you for being a part of our community!

The problems of space sim design

Started by
3 comments, last by Dramolion 6 years, 9 months ago

So this is apropos of nothing — I'm not working on a space sim and looking for ideas or anything like that, but this is an issue that's been rolling around in my head a lot lately, and I thought it would be an interesting topic of discussion and a fun thought experiment of sorts!

I'm an avid fan of both DCS World — a hyper-realistic flight simulator — and Elite Dangerous — a pseudo-realistic "hard sci-fi" space sim. But when I think about which one holds my attention longer, which one provides more challenge, and which one engages me more as a player, it's no contest: DCS World. But the problems with Elite Dangerous seem, to me, to have less to do with the specific design decisions of that game, and more to do with the limitations of the genre.

In DCS World (and any proper flight sim), even during low-intensity activities like taking off and flying from waypoint to waypoint, there's a lot to think about and keep track of. You need to watch your heading, altitude, and airspeed. You're constantly trimming and re-trimming your flight surfaces to account for the aerodynamic changes that come with any sort of minor course adjustment. And when you do have downtime, there's always something relatively interesting to look at, i.e. the terrain passing below your plane.

In a space sim like E:D, it's a totally different story. Making a long journey between waypoints becomes a matter of setting a course, letting go of your instruments, and waiting. Since space is so big and bodies are spread so far apart, you're mostly just looking at empty space for the entire journey, or the same, unchanging backdrop of stars.

And a similar problem happens in dogfights. In DCS World, you have to manage your radar (which is usually a pretty complicated task in itself), and maneuver for position compared to your opponent, all while being very careful about the current flight characteristics of your plane. If you pull a turn too hard for too long, not only do you risk blacking out, but you could bleed off too much energy and send your plane into a stall. It's a lot to manage all at once and provides a very high degree of challenge even for experienced pilots.

But newtonian physics in a vacuum are quite a bit easier to manage. Elite: Dangerous has attempted to add in some systems to keep things a little more interesting than they would be otherwise, like maximum speeds and turn-rates that vary from ship to ship, the ability to target individual modules on the enemy's ship, and mid-fight power management between your engines, weapons, and shields. But these battles still end up mostly being a game of point at your enemy and pull the trigger. This can be challenging in the same way that first-person shooters are challenging, in that it takes practice to aim quickly and accurately, but there's nowhere near the level of depth that something like DCS World provides. Even in more casual flight sims like War Thunder, flying your plane in a way that effectively manages your potential and kinetic energy plays a huge role in dogfights and has a direct impact on your ability to maneuver and aim effectively.

So the challenge I set before you is: How would you design a space sim to overcome some of these issues and provide an experience that rivals the depth, complexity, and challenge of an equivalent flight sim? What sort of mechanics would you introduce to make dogfights more exciting? How would you make the simple activity of flying long distances in space as engaging as flying long distances in a realistic flight sim?

Advertisement

In the X series, they give you other things to do while flying with your auto-pilot on, mostly menu-based stuff like managing your fleet, adjusting prices for your goods factories, giving commands to your other ships, checking your stats, etc.

In addition to that, they give you the "Singularity Engine Time Accelerator" later in the game, that basically speeds up the game's time tick so that you don't have to wait for too long while still maintaining all the stuff happening around you in "in-game realtime".

If your dogfights aren't exciting enough, you definitely got something wrong in the design.
That's why play testing is so important even in the early stages of development.
It's way easier to change things early in development than later on.

 

I'll start with this one first:

7 hours ago, Danny McGee said:

How would you make the simple activity of flying long distances in space as engaging as flying long distances in a realistic flight sim?

Neither are realistic.  That is, the big "realistic" flight sims tend to add more of the real life controls to the simulation, but they are far from real life.  

You touched on it with this line in your statement: Since space is so big and bodies are spread so far apart, you're mostly just looking at empty space for the entire journey, or the same, unchanging backdrop of stars.  

DCS World that you mention is good because they have long distances, but even so they terribly compress the time and distance between takeoff and the actual combat theater.  They've got a bunch of D-day simulations, and gloss over the fact that flying out was nearly an hour, then the combat, then nearly an hour flying back to land. The Gulf War simulations are similar, even though jets were faster the distances were longer, still taking over an hour traveling out and an hour traveling back.  A little cheating by shortening the distances make the simulations that much better.

People generally don't want to play a simulation needing two hours sitting around watching instruments with only 20 minutes of 'fun' gameplay in the middle.

Flying in space suffers the same problem. Our closest neighbor is Proxima Centauri at 4.25 light years. Voyager 1 is the fastest-moving object we've launched because it was basically bounced off two planets. It's been flying for 40 years, and is not quite 20 light-hours away. If it were heading to Proxima Centauri -- which it isn't -- it would take about 70,000 years to reach it. If you use a faster-than-light technology in your game, and traveled at 2x light speed, you're still looking at two years of traveling.  

If you wanted to reduce it to a funner time scale, perhaps 20 minutes of travel, then you're looking at speed of nearly 100,000 times the speed of light to reach our nearest star.  If you're looking at other stars you'll need many more times faster to cut it down to fun times.

Realistic time and distances are out if you want a fun game.  You'll want to go with unrealistic distances, unrealistic speeds, and/or unrealistic time to make it fun.

 

8 hours ago, Danny McGee said:

What sort of mechanics would you introduce to make dogfights more exciting?

If you've got a ship that can withstand travelling 100,000 times the speed of light there is no point in dogfighting. 

In the games that handle dogfighting, the things that make it more exciting and fun are the limitations of the craft.  Each has limits on how fast they can turn and bank, limits on how quickly they accelerate or slow, limits on how much stress the craft and the pilot can endure. Each has limits on weapons; firing rates, effective distances, and comparative velocity based on the enemy craft. 

The fun is finding ways to win as scenarios are increasingly more challenging. Finding yourself outmanned and outgunned, but in a significantly more maneuverable craft, you can still win with a challenge.  Or finding yourself paired evenly with the same number of equally-equipped craft, finding ways to win. 

8 hours ago, Danny McGee said:

How would you design a space sim to overcome some of these issues and provide an experience that rivals the depth, complexity, and challenge of an equivalent flight sim?

I personally think that if you're making a game about dogfighting you should dump all (or nearly all) the boring stuff in between. Travel time and being bored between waypoints is not why players come to play.  I play games to escape from reality because I have some time to kill. I don't want to find another diversion to keep my mind occupied while I'm waiting for my game to fly between waypoints.

The fighting itself using constrained craft against other constrained craft is fun by itself.  The complexity should not be through adding more levers and switches in the cockpit. You don't want tracking complexity (too many things to follow) or comprehension complexity (trying to figure out which commands do what, or translate what you see into something meaningful).  What makes games fun are decision-making complexity.

That is, players start out with few decisions necessary to win. As players understand each decision, they can find ways to make more complex decisions using the same tools in order to reach more complex results.  The trick in design is to provide a small number of versatile tools, then to craft scenarios that require an ever-increasing skill in using those tools.

An oft-cited example is Portal, which had amazing design. When you start you have very few choices, you cannot manipulate the world and walk around when the portals open.  About the time you master navigating portals, you are given the option to create a single end of the portal. When you master that, you're offered the ability to create both ends of the portal.  After that you get increasingly more difficult scenarios where you must place portals at strategic locations, then use them at strategic locations and strategic times, until you reach the point where you can fall off a ledge, place a portal, fall through the portal and shoot to place another portal while mid-air, then go through that portal to your intended destination.  Few tools, but deep decision-making complexity with a well-designed learning curve.

The various Legend of Zelda tools are another example.  You get a small number of tools, such as arrows, boomerangs, bombs, hookshot, etc., and as the game progresses you must learn how to use each one. The game grows with more decision-making complexity as each tool is mastered, until at the end of the game you are quickly switching between tools and making rapid decisions that make the game fun.  It is not tracking complexity (having so many objects on screen that you can't follow them) or comprehensions complexity (having so many controls or so much information it is hard to parse it all) that makes the game fun.

I expect my flight simulator and space game to give me a small number of options regarding weapons and maneuverability. I expect all the information on screen to be easily understood, or to be intentionally removed such as radar going out for a scenario difficulty. Then as a player I can use those limited options to make complex decisions.

There are plenty of objects in space to fly around making travel more interesting, some gaseous-cloud in space might be only one-tenth of the density of our atmosphere, but that doesn't mean a space-ship can fly through it unharmed.

 

This topic is closed to new replies.

Advertisement