Planetary Orbits?

Well, the math for orbital dynamics is 'open source', if you will, so I don't think this is really an area to be concerned about...

Rough circular orbits are easy to do, and it'd probably be just as easy to make them vary with eliptical orbits. Those of you wanting true physical orbits are kinda crazy though.

First, the true physical orbits really aren't gonna add much to the game. So the moon is now abit wobbly because of a nearby planet... are you ever gonna notice? it'd be just as nice looking, and alot easier to understand if they were locked to pre defined orbits untill player intervention (ie, engines )

Second, it's actually alot easier to look at the playing field and see all the pre-defined orbits than to try and figure out why x moon or planet will suddenly shoot off in a random direction, if the ui would even be able to predict that, because it's orbit gets to close to another planet or orbital body. Not needed in a game where you're already worrying about why your artillery shells are arcing the way they are from said moon onto a planet. It might be neat but ultimately the number of scenarios youd see physically simulated orbits vs pre baked circles just makes it pointless.

If they can make game mechanics where asteroids can be moved I believe it'd be just as easy to make the planets move to, just have a tooltip saying the orbital bodies max, and the predicted number of engines to move it (actual moving speed's irrelevant, you can't try till you have the number of engines required... for simplicities sake)

We're really not though. The math for all of this exists. And I really don't get where people think that your computer is going to be overworked doing this because it really won't.

Without using the good physics formulas you cannot have eccentric orbits for example. Nor can you easily adjust for when things start getting pushed around and blown up. You kind of need the higher math for what they're trying to accomplish here.

What blows my mind is that people somehow think their computers or the developers are somehow incapable of handling college level math. This is simply not true.

The problem is that with the formula of real world physics it is really difficult to create and maintain stable orbits. Not to mention float imprecision slowly tearing away at your stable orbits in long games.

In my opinion celestial objects should just be stationary. Unless, of course, you build engines on them, in which case they should move. Constantly moving planets, asteroids and moons would not only cause a performance impact, it also confuses the player constantly, since they would constantly have to reorient themselves and have to find their asteroid base again.

Either ways, planets should still rotate and have night/day and moons should still orbit their parent.

Keplerian motion is no less stable than stationary planets, and at least creates the illusion of planetary motion without having n00bs going "Herp derp I accidentally ejected my planet from the solar system.". For spaceflight, units could just automatically travel on Hohmann transfers without actually running the physics behind it, since all adding real physics would do is force the AI to numerically arrive at a Hohmann transfer whenever it needed to move things around.

Sure, we won't have gravitational slingshots or moons with their own moons, but this is a war game, not Space Probe Simulator 2012.

I find myself wavering towards that viewpoint at times, esp. after I look hard at some of the problems (particularly how to make it all comprehensible and intuitive), but the bodies might as well be flat playing boards you jump between if it all remains static.

The mathematics of general relativity are also "open source", should that be implemented to get all the mechanics truly correct then? It's not about computers or developers being incapable, but about what's playable and intuitive in the end.

Uber's "we're not shooting for realism, we're shooting for awesome" has more than a few implications. It means planets won't take years of in-game time to orbit their host star, for example. It also means the scale of things (like solar systems) won't be true to life. So since there are going to be necessary concessions to realism to keep the game fun, orbital physics will probably also receive a few tweaks and simplifications.

Yes, you could integrate the forces twice to get how everything moves, but it's a complicated enough system that you have to do it numerically, which means every tiny step in time doing the force calcs, doing the integrals again, and you have to do that some amount of steps into the future so that you can warn the players of crazy things, which when a player starts moving things they then throw all that prediction out the window and we need to start again on a new prediction set.

Yes the math is not that hard, but it turns into a lot of math, where any inaccuracy is gonna catch up with you at some point for what seems like very little gain over Kepler orbits . You still get ellipses, you get all the normal orbits (geostationary, LEO, GPS orbit, Hoffman transfers between things), what you don't get are Lagrange points, or interplanetary transport networks, or random moons doing weird things.

We don't need physics dropping KEW's on us, we'll do it plenty ourselves. I would love to see this as something in a mod or in a well thought out expansion, but not now, when there are better awesome things for the devs to do first, like game play.

Wait, so, what the hell was wrong in the first place with basic circular orbits and 'fake' gravity wells for each one? All you'd need is a patched conics trajectory system to calculate how units get from A to B, which was shown in the concept video. Nobody should care about gravitational attraction since it will seriously never be an issue in here. When they want an asteroid to smash into a planet, it'll go to the planet and smash into it. Who gives a damn if it properly accelerates according to the specific gravity of that planet?

The more I read this conversation the more retarded it is. Some people seem to think this game is going to be a space simulator or something is about all I can get out of it.

But I want some players to be fighting over a planet and me on my neighbouring planet build a giant invasion force and base on my moon. They are squabbling over their planet when *ominous music starts* a new moon moves into the sky and a reign of terror begins as my units and artillery rain down upon them. I don't want a space simulator i just want the end game option to move things around the solar system, nothing would be more awesome than the solar system to bend to my will. It's all about the awesome factor and strategic value of moving parts of the map to better positions.

yes.

Exactly how I invisioned it. Simple orbital dynamics makes this so much simpler than having your moon you just moved completely effect the orbital bodies between your moon and the target planet, which would just be ludicrous! asteroids would be getting caught in new orbital paths and complete chaos would reign... not that it's actually sounding bad now XD but it would be very very confusing during gameplay, and that level of confusion would throw alot of players off.

Sounds great. And you know most of the large bodies in our solar system move relatively slowly to one another.

Combine that with vastly scaled down distances between them, all the natural stuff can make a stately dance around each other, and it's not until you start strapping massive rockets to things do they really get moving.

Over the course of a couple hour game there's no reason why a well designed map couldn't work brilliantly, just enough built in movement to feel 'alive', but not so much you have to keep track of complex orbital mechanics to plan your strategies.

And now I present to you the coolest 2-4 player skirmish map ever
http://upload.wikimedia.org/wikipedia/c ... Orbit5.gif

That looks like a very good map idea indeed.

I would love to see planetary orbits and rotation effect such things as the efficiency of solar power plants (ie facing sun ~ 100%, away from sun ~ 50%) and the effectiveness of orbital bombardment (ie a fast rotating planet is hard to accurately bombard)

I think that adding actual orbital paths is adding a layer of complexity that is not needed. I *think* it would be better having each orbital body in a set location and that it can be moved or made collide with thrusters, allowing you to move it to a superior position.

I am saying *think* because I can not really tell, the only thing I can speak from is personal experience with previous RTS games. It is worth absolutely worth testing but I doubt orbital paths would add anything to the gameplay.

Lets say you have two planet game, each player starting on their own planet and these planets only come within shelling range of one another once every 5 minute. The result would be that both players would arm like crazy for 4 minutes and then total chaos would break out for 1 minute followed by another 4 minutes of complete base building with little to no player interaction. This is not a type of match I would enjoy playing. I like RTS games with constant player interaction.

I would not like to see this since it adds an unnecessary layer of complexity to the game. If something like this where to be in the game I would like to see it fully realized and not just a "neat thing because it is realistic and would look cool".

Disagree. If you can't wrap your head around night and day that's your problem. I fail to see how that's unnecessary or complex.

Though the 'quickly rotating being hard to bombard' is kind of dumb, I'll agree with that.