Hacker News Reader

Comment by Zikes | Hacker News Reader

Setting aside the ideas for the plot, could games like Eve, No Man's Sky, or Elite: Dangerous accomplish the same thing if they used actual astronomical data?

View on HN

They can't. In order:

Eve is a murder/capitalism simulator and has almost no physics in the first place. (There's 3-dimensional positioning, but you can fly through planets and shoot through other ships as much as you like; also the entire world is static.)

No Man's Sky is 90% an art piece with some added murder simulation.

Elite: Dangerous, also a murder/capitalism simulator, only unlike Eve it's more about the immediate visceral.

Notably all three of those pretend space ships are, like WW2 planes, subject to air drag, have a terminal velocity and have no automation on-board whatsoever, and astronomic physics aren't even considered in the first place. These games are literally WW2 fan fics in "space". (Or really, ether, as old-timey people thought of what's between planets.)

If you're looking to teach physics and astronomics, Kerbal Space Project does that quite well, even though it also ignores relativity and speed of light considerations.

More importantly, none of these games are even remotely able to tackle the topic of civiliations and saving them, which won't be done with a gun, but with the word. The closest thing i can think of to that are the Mercenary games: https://en.wikipedia.org/wiki/Mercenary_(video_game)

Edit: Not saying that the three mentioned are bad games, just that they can't be taken seriously as space games.

You're assuming the goal is to teach astrophysics, rather than astronomy. If we're just trying to teach the space equivalent of geography, the physics and plot don't matter so much as the physical locations and nomenclature.

Astronomy is the study of the laws of stars, not of their names, regardless of how close "nomy" sounds to "name".

https://simple.wikipedia.org/wiki/Astronomy

Additionally Sagan was most certainly conflating it with astrophysics, as is commonly done, and evidenced by his mention of "stellar evolutionary events".

That's barely even astronomy. At the university level, astronomy coursework is basically math.

Slight correction but Elite: Dangerous has the capability to do better flight mechanics.[1]

The game is often quite boring, but I joke with people that it makes an amazing space station docking simulator. I actually can't wait to try that with an Oculus.

[1]http://gaming.stackexchange.com/questions/224726/does-elite-...

Only in the barest sense, sadly: "Each ship has a top speed, this can be increased with better thrusters and power distribution, but they still cap out whether you have flight assist on or off."

I was pretty disappointed about this. The predecessor games to Elite Dangerous, Frontier and Frontier First Encounters, had a Newtonian flight model (and an optional "time acceleration" feature to artificially shorten intra-system flight time).

Same, it's why i retracted my original backing. I do understand the issue they faced, since they wanted to make a multiplayer game. But they sacrificed their original fanbase on the throne of that.

realistic physics is mega boring in space games. The first elite was much more fun with arcade physics. Frontier combat was rubbish, you just yoyo'ed past each other with 5 seconds of shooting between oscillations. That's with a human pilot driving.

Realistically a space battle would be, does the evader want to evade and has more thrust? => they evade, else, the attacker approaches and both get to shoot until one is dead. There would be no real combat other than weapon size and armour plating. I don't think there would be space dogfights as the onboard computers would just be able to track each other too accurately / apply optimal game theoretic evasions given the constraints. Boring ... might as well play spreadsheets.

They are boring when the player's tools and gameplay are not adapted to a situation involving relativistic distances. In space warfare sensors, delay of signal reception, interpretation of sensor input, projection of positioning both human-based and automatic, as well as weaponry built and programmable to relate to all of these things become much more important than:

See, point, shoot.

Rogue System looks promising as a semi-realistic combat space sim: http://imagespaceinc.com/rogsys/

Pioneer also exists as an open source Frontier remake, with the same sort of flight model and universe: http://pioneerspacesim.net/

Independence War 2 - the best space simulator I have ever played

I think it wouldn't hurt to use actual stars and such in these games, if only to give people a sense of our astral "neighborhood".

I don't know if it would make for good gameplay though.

In my game (https://technomancy.itch.io/bussard) I consulted star maps of nearby regions of the galaxy (http://www.projectrho.com/public_html/starmaps/mapindex.php) and placed systems accurately.

However, I felt that using their astronomical names/numbers would be unrealistic--if these stars ever ended up being colonized, they would be given colloquial names, so I made up my own names for most of them.

There was an old 8-bit game called The Halley Project that aimed to teach you constellations by having you navigate through the solar system: https://en.wikipedia.org/wiki/The_Halley_Project

> If you're looking to teach physics and astronomics, Kerbal Space Project does that quite well, even though it also ignores relativity and speed of light considerations.

KSP "only" has a sphere of influence model. That's why there aren't any Lagrange points anywhere in the solar system.

Also relativity and speed of light are pretty unnecessary in a chemical-rocketry simulator.

Category of "murder/capitalism" is an interesting one. Do you think that the people who wrote Elite, for example, were so steeped in the values of colonialism or piracy that they were oblivious to the murder/capitalism subtext? If not, it must be human nature, I suppose.

Not so much that he (he's made several games like that) is oblivious to it, but more that it's the most easy thing to simulate in games, and also what people expect. You can do better, as games like Untertale or KSP demonstrate. But it takes a special kind of person.

Also, Kerbal Space Program.

Star Citizen perhaps, since they out of the games you've mentioned hew closest to established astronomy.

The rest just invent whatever through the magic of procedural generation, which is cool but not coherent with using actual astronomical data.

But that's my point exactly. If those games did use actual astronomical data (and I'm told by foo123456 that Elite: Dangerous actually does) then could those games accomplish Carl Sagan's goal of teaching astronomy in a compelling video game setting?

Those games' established plots are apparently compelling enough, but would they drive learning if their current environments were swapped with something based on reality?

Star Citizen's intent seems to be to ship personal servers (quake style). Maybe this will be possible, I'm sure some people would be interested.

ehrm, Elite: Dangerous actually use real data.

Yeah, the entire Elite series has always used real star coordinates and names, at least for those close to Earth.

Except for the original game, which used procedurally generated galaxies, and didn't feature Earth. This was the only way to get the game to fit into the 32K RAM of the original BBC Micro.

https://en.wikipedia.org/wiki/Elite_(video_game)

You're right! It wasn't until the second game, Frontier: Elite, that they started using real stars around Sol.

Reply on news.ycombinator.com