Mod called resonate orbit calculator combined with a mech jeb execute maneuver node input is probably going to be the most precise you can possibly get.

Just in case any new player needs it:
At that point I would edit the persistent.sfs file.
Set all satellites' SMA (Semi Mayor Axis) to same value (and delete the fraction part),
which effectively gives you zero drift.
I always do that when using RemoteTech or CommNet, to not get anxiety from waiting for transfer windows.

What is interesting: getting transfer time to a couple millions and then reloading brings it back to around 10k years. It changes the stored fraction b yabout 0.002 seconds in m orbit. So its truncated when saving/loading it appears. The precision float is there when its running, but seems like storing/loading is less precise than when its in memory.

Originally posted by blowupguy:

Mod called resonate orbit calculator combined with a mech jeb execute maneuver node input is probably going to be the most precise you can possibly get.

Actually I found that mech jeb 2 can do it automatically and automatically correct itself.

Here's the thing if the drift is not substantial it could take potentially hundreds of years in game for the orbits of your satellites to make the network non-functional. So I strongly suggest instead of trying to get a perfect orbit three satellite network with no drift just throw up six or seven satellites in two different orbits and a third orbit with a few satellites and a polar orbit. Stop obsessing about it and move on with your life maybe after a thousand years in game you'll have to correct the drift of one over your satellites here and there it's not a big deal and you guys are breaking your brains trying too hard to fix a problem that is nearly impossible because this is a game. In my current game once I realize this I have four satellites in a 600 km orbit four more in a 1 million km orbit, and then a further four more in a 1 million km polar orbit. There is nowhere you can go around kerbin and not have a connection. My next step was to put a few satellites around mun and minmus and now they are permanently connected to the network and I have not to worry about anything and I don't have to obsess over keeping everything perfectly aligned

Its not as simple as that.

Given a desired phase angle of 180°, the time until transfer at which point both satelites meet depends on the orbital period difference divided by the shorter orbital period. Which means the time until transfer is doubled for each doubling of the orbital period.

Low, short periods diverge much faster. Could be couple of dozen years or hundred at which point you don't just have an error, but they actually meet each other. This is easily achieved through reasonable amount of time-warps for reaching bodies within the solar system, which can take many years.

Another problem is the one I've explained. In a long orbital period, I achieved transfer (180° phase) in around 20,000,000 years. Which is reasonably long. But after saving the game, it droped to around 28,333 years. Depending on how you play, with time-wrap you can achieve considerable fraction of this realistically, which wouldn't make a transfer, but noticable error you had to correct. Imagine doing that on 50 satelites or more.

And I'm not talking about modded, interstellar travel yet. Just vanilla.

Originally posted by lalala:

Its not as simple as that.

Given a desired phase angle of 180°, the time until transfer at which point both satelites meet depends on the orbital period difference divided by the shorter orbital period. Which means the time until transfer is doubled for each doubling of the orbital period.

Low, short periods diverge much faster. Could be couple of dozen years or hundred at which point you don't just have an error, but they actually meet each other. This is easily achieved through reasonable amount of time-warps for reaching bodies within the solar system, which can take many years.

Another problem is the one I've explained. In a long orbital period, I achieved transfer (180° phase) in around 20,000,000 years. Which is reasonably long. But after saving the game, it droped to around 28,333 years. Depending on how you play, with time-wrap you can achieve considerable fraction of this realistically, which wouldn't make a transfer, but noticable error you had to correct. Imagine doing that on 50 satelites or more.

And I'm not talking about modded, interstellar travel yet. Just vanilla.

Just use mech jeb + resonate orbit calculator.

Yes, just was replying to IronSides

Originally posted by lalala:

Yes, just was replying to IronSides

what i'm saying is with enough sats and it really doesn't take many it matter NOT where ANY or ALL of them are in their orbit they will somehow find connections. you have to use ZERO brain power setting it up.... just more time launching more sats.

here is a super easy less than 20 relays here never had anything out of connection yet.

simple setup with 100% connection.... https://steamcommunity.com/sharedfiles/filedetails/?id=3817364165

You seem to assume that “just launch more sats” is some clever solution nobody else thought of, rather than recognizing that some players are deliberately trying to solve the network efficiently instead of brute-forcing it with redundancy.

You’re still answering a different question. The point was never whether “just launch more sats” can brute-force reliable coverage — obviously it can.

I was specifically asking how to improve orbital precision, because in KSP that can itself be the goal. Maybe I’m optimizing for elegance or efficiency, maybe I’m on a harder difficulty with a limited budget, maybe I want fewer launches, or maybe I simply enjoy solving the precision problem itself. Those are just a few examples.

What’s frustrating is the repeated implication that everyone else somehow failed to think of the “simple” solution of adding more satellites, when the actual discussion is about deliberately solving a different problem.

I did answer,
This is an imperfect game there has been over the YEARS, multiple proofs that no matter how you try to perfect the orbits they will decay or change. You seem to love being super efficient and use realism the only answer is MODS that are un-realistic, therefore either "cheat" the orbits or try endlessly too perfect them. I'm just trying to save you time and stress but do your thing.

And it's not like a I haven't played the ♥♥♥♥ out of this game and maybe know a thing or 2, I just don't go about in the way some weird perfectionist players do My stuff has FORM AND but over function its what I'm known for.

You did answer, but you’re still answering a different question than the one I asked.

I’m aware KSP has limitations and that perfectly stable orbits may not remain perfect indefinitely. My question was specifically about how far the stock game lets me push orbital precision, because experimenting with that is part of the fun for me.

“Use more satellites” or “use mods/cheat the orbit” are valid ways to avoid the problem, but I’m deliberately interested in solving it within those constraints. I’m not stressed by spending time on it — that’s the gameplay I chose.

I appreciate that you’re trying to save me effort, but there really isn’t a need to keep convincing me that I should pursue a different goal.

Originally posted by IronSides:

I did answer,
This is an imperfect game there has been over the YEARS, multiple proofs that no matter how you try to perfect the orbits they will decay or change. You seem to love being super efficient and use realism the only answer is MODS that are un-realistic, therefore either "cheat" the orbits or try endlessly too perfect them. I'm just trying to save you time and stress but do your thing.

And it's not like a I haven't played the ♥♥♥♥ out of this game and maybe know a thing or 2, I just don't go about in the way some weird perfectionist players do My stuff has FORM AND but over function its what I'm known for.

Since you updated your comment, after my reply:

I’m not questioning how much you’ve played or whether you know the game well. I’m saying that you keep treating my chosen objective as if it’s a mistake that needs correcting.

I’m deliberately interested in how precise I can get within the stock game. That may be “weird perfectionist” gameplay to you, but that’s fine — KSP is a sandbox.

You don’t need to convince me that I should enjoy solving a different problem.

Originally posted by IronSides:

I did answer,
This is an imperfect game there has been over the YEARS, multiple proofs that no matter how you try to perfect the orbits they will decay or change. You seem to love being super efficient and use realism the only answer is MODS that are un-realistic, therefore either "cheat" the orbits or try endlessly too perfect them. I'm just trying to save you time and stress but do your thing.

And it's not like a I haven't played the ♥♥♥♥ out of this game and maybe know a thing or 2, I just don't go about in the way some weird perfectionist players do My stuff has FORM AND but over function its what I'm known for.

Dude mech jeb is not a cheat, it’s just a precision guidance program.

Who said mech jeb was a cheat? freezing things into perfect orbits is a cheat, mech jeb is realistic... after docking manually a thousand times, it's nice to have it do it for you.