Gravity, Orbiting, etc

I could do with some help moving this thread along please. About planetary transitioning and orbits, and could also progress to your gravity and orbits conversation. Do we have anything new today from the dev’s that they can offer up for posting there. More detail.

I know very little about the planetary transitioning, and orbits. But reading this thread you guys seem to. So some help on the thread above would be most appreciated. Pick something and just comment on it, what you’ve experience in the prototype for transitioning and orbits etc.

Just to move the thread along. There will still be SC fans who have not yet seen Infinity:Battlescape.

Thank you in advance.

yes. in real N-body simulation you just do calculation for every body at every iteration. so not big deal for 10 planets at all. but if we have 10 planets and 200 ships, planets part will be 109 operations when ships part will be 20010 (ships have no mass). so real problem not planets but other objects.

but when we are using planets on rails it becomes not computation but error accumulation problem. for example space station is on rails around earth and ship that is flying close to it do not. ship calculates their speed vector every iteration so their position error will increase and it will drift away faster and faster. using influence based variant will prevent this situation for close orbits. if we have only 1 body we can easily calculate specific “rail” for every given moment for every ship. we do not even need to recalculate it every time. only when ship starts o accelerate.

p.s. sorry for my english . reading and writing skills seems different for me :smile:

Calculations every frame?

every iteration. when we have for example one iteration every 10 frames ships will fly in straight lines between them.

calculation is quite simle itself.

all i am telling is how things COULD be. i do not know what path developers will choose

Didn’t Flavien say it could be dozens of bodies?

And it’d need to be calculated on the client and server for every ship and each body. It just seems a bit much for 0.001m/s/s pulls in such an such directions that barely offset what the main thing is pulling you. Ships are so powerful that it’s super insignificant.
In KSP, with a game that has fairly sane modern TWRs, even there having N-Body simulations wouldn’t be a significant change.

I think N-body calculations would be a pointless drain on system resources. If the matches are only going to last weeks (at most), a simple calculation of each planet’s orbit around the sun would be more than adequate to put them on their own rails and off you go. Moons around planets are much the same thing: calculate their orbit around the planet, put them on an equivalent rail, and you’re done. I like KSP’s style of “Sphere of Influence,” although it’s a little too simple when it comes to interactions in the border regions. Things that should, in my opinion, be calculated for gravitational pull:

  1. The star.
  2. Celestial bodies that produce a significant portion of your delta-V (1%? 5%? I’m not smart enough in the field to know exactly what portion is reasonable). Anything below a certain threshold is just wasting resources better spent elsewhere. This includes space stations.
  3. Craft that are designed to have some significant degree of attraction (carriers, for instance, should always pull their child ships toward the deck with some degree of gravity, to prevent sliding around when docked.

I have a hard time imagining ship-to-ship gravitational action being anywhere near enough to bother calculating.

1 Like

I did a few calculations for a geostationary station in the Earth/Moon system. The Earth contributes over 99.98% of the gravitational force. I suspect no player would notice the motion on their ship. Their hand-set matching orbit would probably deviate by more than that. Active station keeping could cancel out any drift.

Now, if there was a free-floating station that was heavily influenced by multiple massive bodies then I could see the need for an n-body simulation to figure out where it was going. I have no idea how things would work out at Lagrange points.

But ultimately, this is just a tempest in a teapot. I’m sure that the INS guys will solve this quite capably without our help.

How large are the planets in the prototype (km). And in the finial release what sort or range of sizes will there be from planets and moons. What gravity will some of the large planets have.

You may get a good ballpark from the apparent respect of real planet sizes. So the gas giant is probably something like Jupiter, or smaller. Dwarf gas giants are at most “just” 4 Earth radius, and the upper bound for gas giants is basically Jupiter size (gravity quickly limits larger volume). I reckon it’s a safe bet that the current pink gas giant is somewhere in the range of our own gas giants. Somewhere between Jupiter and Neptune/Uranus.

Considering how Flavien said they were keen to add as many phenomena and variety as feasible :

I think it’s safe to assume that we’ll get as much we could hope for, given the actual time and funds.

1 Like

Thank you.

As I posted that, I wondered how useful that explanation could be when you can’t actually get a level view of all planets in the tech demo, the way that that orthographic projection above illustrates the planets.

I do think, though, that it’s safe to assume the devs will give us the best compromise between realistic and feasible. So far there’s no remarkable deviation from the inherent awesomeness of the cosmos.

This game is going to be fucking awesome :grinning:


I don’t really know where to post this, here is one silly observation.
I must say before I start that I love this game and waited for it way too much time to be sane anymore. Those two next years will be an eternity !

In this video, at 3:57, Flavien Brebion (I can’t help myself, I love the accent of my fellow frenchies) shows us the effect of gravity on his ship, falling towards the moon in front of him.
But, why is he just sliding at all ?

The asteroids, the station, everything around is orbiting the moon. At this moment, he is moving relatively to the very local space rock cluster, and then he is orbiting the very same gravity anchor and at the same altitude and speed.
Then, why is he falling alone when he shuts down his ship engine ?
The only case Flavien could be sliding this fast towards the moon should be that he is not moving anymore at all relatively to the local station, and this is something that should not happen when you shut down your engine or your flight assist. On contrary, it would be a significant cost in acceleration.
But in this case, and even if everything was accelerated (as in a game simulation meaning), the ship should be moving also backward (or to the left when you look at the moon, in the case he is going on its orbit towards the right), because he is decelerating on its orbit, which is totally not happening on screen or then not at all at the same speed than its fall.

So, something seems wrong to me there.

Currently all objects are stationary and are not orbiting to make filming easier.

Also this:

“In the prototype they’re currently locked in space. We need to improve our networking code a bit before we can have them move. In most games latency hitches cause a bullet or a player to be off by a meter or 2. In our game a latency hitch can cause something to be off by 100km - it poses some unique engineering challenges!”


That’s reassuring. Thanks !

1 Like

On the subject of orbits, and planets - the latest BBC Sky at Night is all about finding a second Earth.

1 Like

Youtube version for people outside the UK: (The BBC iPlayer version is region locked)

1 Like

will the game implement EMF features? EMF makes the difference gameplay-wise, not just because you can jam your enemy’s coms, but because nature can also jam them, and that’s just 1 thing EMF can do

Kind of a tangent, but on the subject of orbiting bodies and gravity, how cool would it be for the orbiting space stations and small rocky bodies to be de-orbit-able?! Like if a player was a poor pilot of their capital ship and rammed into something, that something could get lurched out of a stable orbit and go hailing towards the planet’s surface. Or some sort of emergent game play using asteroids to bombard the enemy’s surface infrastructure? I think it would be wickedly awesome to look up and see something large plummeting out of the sky unstoppably…

If a system similar to this or this is implemented and the modifiers for stations aren’t infinitely high, you could theoretically “pull” a station, maybe even de orbit it.

Not only would it be great to see it from below. It would be great to race alongside it, trying to defend or stop it. Man I sooo wish for such things. Deorbiting asteroids or stations would still take a long time and kinda sound out of scope. It’s also hard to tweak. If the station or asteroid is in a certain orbit. It will take a certain time to hit the surface. That time can’t just be tweaked with configs like one can tweak the available resources of a team, the timelimit or the limits of available battlescapes. This would make it hard to scale it. So for instance in a big server with 500 people online it would take 2 hours for that station to deorbit. In a server with 30 people it would take the same time. Even then would it be a rather “relativel” slow moving threat.

Big missiles, bombs or stuff like that though. :wink:


Seems to be the case, that said I couldn’t be more exited. I hope there are either mods or inbuilt features to represent orbital velocity vectors like KSP.

From what I hear I novae seems really cool with mods, I really hope the guys from project RHO see this and make an Orion class battleship… would be absolutely, positively too much fun.