Jump to content

Communications in KSP 2


GoldForest

Recommended Posts

So I had a thought. How are communications in KSP 2 going to be handled? Then I had another thought. Time delay. Will there be that in KSP 2? 

To be honest, I would be fine with time delay since it is a real life problem, but only if it was done in a way that it didn't take years for interstellar messages to make the journey.

It takes anywhere from 5 to 30 minutes iirc for any message to get from Earth to Mars. And that's a one way trip.

Signal delay in KSP won't be that bad obviously. 1/10th scale means 1/10th the way time, so 1 to 10 minutes from Kerbin to Duna or somewhere in that area.

My main concern, interstellar communications. Science is still a thing. Now, it could be that colonies add to science instantly, but until a colony is made, all info would need to be sent back to the KSC on Kerbin. So that could take literal months to years in game time.  

The time delay isn't the only problem. Signal strength and the length of travel. A light year or two is a long time. 

So, are they going to add light based communications?  And by that I mean laser light. 

Another possibility, though highly unlikely to be put in game, subspace communication. Subspace is science fiction, yes, but I'm not talking about the concept of another dimension, but more just the FTL communications. FTL communications would be a great help if time delay is added into the game. 

Obviously it would ramp up with difficulty:
Easy - Instant messages
Normal - Slight delay
Hard - Medium delay
Very hard - realistic delay 

Just my thoughts on the matter, what do you guys think?

Edited by GoldForest
Link to comment
Share on other sites

I like the idea of time delay in general to get immersion. But we have the option to fast forward the time. So i think people would skip the time with that option to get the information/science/message faster. 

Also another thing came to my mind with a time delay hardcore mode: 

Remember the scenes from the Control Room of the "Curiosity"-Team back then in 2012? People were looking at their screens because the delay between earth and mars not allows them to do anything. they just followed the messages the rover did sent during descent. everything was an automatic sequence. It would be so cool if could also prepare something like that and then we start the mission and in the final phase we just can watch our rover to succeed and land, or we see it crashing because of a malfunction or a mistake in the sequence. 

But as i said it would only make sense in a super hardcore realism mode with time delay. i personally really enjoy to have everything in my own hands when it comes to land a vessel or rover ;)

Greetings from Berlin

Link to comment
Share on other sites

16 minutes ago, Mousambi said:

I like the idea of time delay in general to get immersion. But we have the option to fast forward the time. So i think people would skip the time with that option to get the information/science/message faster. 

Also another thing came to my mind with a time delay hardcore mode: 

Remember the scenes from the Control Room of the "Curiosity"-Team back then in 2012? People were looking at their screens because the delay between earth and mars not allows them to do anything. they just followed the messages the rover did sent during descent. everything was an automatic sequence. It would be so cool if could also prepare something like that and then we start the mission and in the final phase we just can watch our rover to succeed and land, or we see it crashing because of a malfunction or a mistake in the sequence. 

But as i said it would only make sense in a super hardcore realism mode with time delay. i personally really enjoy to have everything in my own hands when it comes to land a vessel or rover ;)

Greetings from Berlin

Sounds more like KOS but with extra steps. 

Although, a probe control room like what that one mod adds would be interesting. But that's kind of off topic.

I'm sure if they added time delay there would be settings that could be changed, although I can see someone turning the setting so that signals are basically snail mail just for laughs. 

They do science, send it, do more science, send it, repeat a few dozen times. Then a few years later they get a science boost. 

Edited by GoldForest
Link to comment
Share on other sites

I'm kind of curious if there will be interstellar relays. Since the largest relay can cover most of the kerbol system, and the assumed distances involved, that's a lot of relay stations to drop to keep comms between the interstellar colonies and Kerbin. 

For the comms time delay, leave it as is for intrastellar comms, but for interstellar comms, there should be a delay.

Link to comment
Share on other sites

I assume everything will be handled locally by the colony or, before that, by the interstellar craft, which will each act as a "mini-ksc".

 

Delay is interesting, but easy to overcome if they follow KSP1

In KSP1, a probe will retain control if its in comm range of a manned capsule (with the assumption that the crew now controls the probe, instead of KSC)

So it would be a matter of sending a manned craft tagging along :)

 

Of course, that adds a whole different set of problems if Life Support, even a rudimentary one, is added

Link to comment
Share on other sites

Time delay for science is skippable with timewarp so doesn't change much.

The real deal with the time delay is if it's applied to probe control and I think that at the end of the day it wouldn't be realistic at all because IRL probe are automated,

Having a stock Kos and using it to land probes would be more realistic than trying to control a probe manually with time delay.

Link to comment
Share on other sites

You can have signal delay in KSP1 with certain mods (ex. RemoteTech).  If you haven't tried this already, do so.  Then you'll see why signal delay is a bad idea as a stock feature.

If you don't have time for that, here's the short version.  Without signal delay, the assumption is that probes are highly autonomous.  When you fly a distant probe in realtime, you're playing the role of its AI, just as you play the role of pilots with crewed ships.  So, signal delay has to assume probes are too stupid to fly themselves or you'd still be able to do that in the game.  And the lag also means you can't fly the probe from KSC, either, EVEN IF YOU HAVE A GOOD CONNECTION.  Thus, the only way to control distant probes is to install the kOS mod and write your own autonomous AI routines for the probe.  So, if you're not proficient at writing kOS scripts, you basically can't use probes.

As for interstellar lag in KSP2, there are only 2 solutions as I see it.  Either 1) completely ignore it, or 2) make communications totally impossible between stars.  The latter would basically put each interstellar outpost in its own universe, so it would have its own separate stocks of money, Science! points, and rep, and basically run as a separate game from the Kerbol system.  

Link to comment
Share on other sites

3 hours ago, Geschosskopf said:

When you fly a distant probe in realtime, you're playing the role of its AI,

And yet when you go into a communications shadow you can't control your vessels :)

My pick is that this will be an option with signal delay.

Also, with advanced tech nodes, engines and such I'm tginking thsy there will also be an advanced AI that functions as a kerbal - otherwise why have anything othet than an OCTO2?

Peace.

Link to comment
Share on other sites

5 hours ago, theJesuit said:

And yet when you go into a communications shadow you can't control your vessels :)

As with all so-called "realism" features, when added to KSP, they're half-measures that annoy the folks who never wanted them and don't satisfy those who clamored for them.  A pox on them all.  Let us have no more "realism" features because they're all bad for everybody's enjoyment.

 

5 hours ago, theJesuit said:

Peace.

Good. I bring sacred 100% perique tobacco for the peace pipe.  It'll put hair on your teeth :D

 

Link to comment
Share on other sites

1 hour ago, Geschosskopf said:

 

Quote

Peace.

Good. I bring sacred 100% perique tobacco for the peace pipe.  It'll put hair on your teeth :D

 

Peace is how I most often sign of my posts :) I look forward to my teeth being much warmer for the cold spring we are supposed to be having.

As long as don't have to vape.  That's just wrong.

Peace.

Link to comment
Share on other sites

On 9/11/2019 at 6:24 PM, Geschosskopf said:

As for interstellar lag in KSP2, there are only 2 solutions as I see it.  Either 1) completely ignore it, or 2) make communications totally impossible between stars. 

I can think of (3). Have it, but only allow Sci interstellar transmissions. Additionally, have an Advanced Autonomous Probe Core part that doesn’t require signal for control - like an AI probe. So you can still fly it normally and transmit back your findings, but they will arrive at KSC after a delay.

This would be mostly flavour. I think it would be better to just to not have lag. Not allowing comms at all would mean no interstellar probes and that would be a shame.

Link to comment
Share on other sites

7 hours ago, Brikoleur said:

I can think of (3). Have it, but only allow Sci interstellar transmissions. Additionally, have an Advanced Autonomous Probe Core part that doesn’t require signal for control - like an AI probe. So you can still fly it normally and transmit back your findings, but they will arrive at KSC after a delay.

I don't think this is actually workable gameplay mechanic, and it would definitely poke more holes in KSP physics ;)  First of all, the devs would have to decide once and for all what the speed of light in KSP really is, the same as here or some considerably lower number inferred from other aspects of KSP physics.  They have to do this to know how long to delay transmissions from interstellar ships.  And therein lies the rub.

To justify this sort of system, the speed of light must be presumed to be so high that all comms within the Kerbol system have effectively zero delay.  But the faster light goes, the greater the in-game interstellar distances have to be to create any meaningful delay between transmitting and receiving Science!  A "meaningful" delay is something longer than hours or even a few days, which a player can just warp past without disrupting his other plans.  It has to be months at least, perhaps even years or decades.  But this all means that sub-light interstellar ships would take on the order of centuries to get anywhere, which is bad for the gameplay mechanic of exploring other star systems.

But OTOH, if you still want to have delay while keeping interstellar travel times down to a few decades at most, then the speed of light in KSP has to be significantly slower than here, scaled to the toy-size KSP solar system.  Which logically means there'd have to be signal delay within the Kerbol system, so you'd need kOS just to send a probe orbiter to Duna.  Either that or you have to make light's fundamental properties very different from here.  Like it maybe it's really fast (so no signal delay within the Kerbol system) but it does in fact use lumineferous aether, which is produced by stars and mostly stops at the heliopause, so the speed of light between stars is much lower than between planets.  Because, after all, you'd still need to have instantaneous comms within the other star systems.

But OTGH, if light slows down between stars, then interstellar ships could be end up being faster than light while travelling between stars, despite being much slower than light within solar systems.  Which means it would be faster to return Science than to transmit it.

That's a whole lot more hand-waving than I want to put up with just to incorporate a feature of very dubious gameplay value :) 

 

7 hours ago, Brikoleur said:

This would be mostly flavour. I think it would be better to just to not have lag. Not allowing comms at all would mean no interstellar probes and that would be a shame.

I agree 100% that it would be better all around to not have lag.  Lag just doesn't fit in with the overall concept of "building a civilization".  For even interplanetary, let alone interstellar, colonies to function together as a cohesive whole, you have to have fast communications.  Otherwise, you're not creating a civilization, you're creating a bunch of independent city-states, like the ancient Greeks and Phoenicians did, or a bunch of independent nations (once they become sufficiently self-sufficient) like the Age of Empires ultimately did.

Link to comment
Share on other sites

On 9/10/2019 at 2:27 AM, Mousambi said:

I like the idea of time delay in general to get immersion. But we have the option to fast forward the time. So i think people would skip the time with that option to get the information/science/message faster. 

Also another thing came to my mind with a time delay hardcore mode: 

Remember the scenes from the Control Room of the "Curiosity"-Team back then in 2012? People were looking at their screens because the delay between earth and mars not allows them to do anything. they just followed the messages the rover did sent during descent. everything was an automatic sequence. It would be so cool if could also prepare something like that and then we start the mission and in the final phase we just can watch our rover to succeed and land, or we see it crashing because of a malfunction or a mistake in the sequence. 

But as i said it would only make sense in a super hardcore realism mode with time delay. i personally really enjoy to have everything in my own hands when it comes to land a vessel or rover ;)

Greetings from Berlin

That sounds too much like most of it is basically just literally tv watching simulator the game. Now with increasing boredom as you have to wait longer and longer for anything to happen.

Link to comment
Share on other sites

Signal delay, combined with automated execution of maneuver nodes, is an interesting option. RemoteTech shows how it  could  work, albeit feels clunky at times, because it is bolted on. Integrated into the game would be a lot better!

As @Geschosskopfpoints out, the game does need to include a certain amount of automation, or a lot of things are simply not possible. But the same automation would make the game too easy for many, without time delay (auto land/launch for instance). Especially in career mode.

Of course we don’t know how career mode works in KSP2, and if (optional) signal delay is included those concerns  would obviously be addressed. I would like to see it!

Link to comment
Share on other sites

1 hour ago, Kerbart said:

Signal delay, combined with automated execution of maneuver nodes, is an interesting option. RemoteTech shows how it  could  work, albeit feels clunky at times, because it is bolted on. Integrated into the game would be a lot better!

As @Geschosskopfpoints out, the game does need to include a certain amount of automation, or a lot of things are simply not possible. But the same automation would make the game too easy for many, without time delay (auto land/launch for instance). Especially in career mode.

Of course we don’t know how career mode works in KSP2, and if (optional) signal delay is included those concerns  would obviously be addressed. I would like to see it!

Signal delay and automation are totally separate things.  One has no effect on the other.  For instance, in KSP1, you can install either MJ or RT, you don't need both at once.

If you think signal delay is a good idea in KSP, then you've either never actually used it OR you're a proficient kOS coder and enjoy writing probe-flying scripts at least as much as you enjoy building and flying rockets.  But If you're the later, we wouldn't be having this conversation because you'd surely recognize that 1) not everybody is good at writing kOS scripts, 2) not everybody enjoys writing code enough to want to do that instead of flying rockets, and thus 3) that signal delay is a VERY BAD thing to force on majority of players, who aren't good or willing coders.  This is because the ONLY gameplay impact of signal delay is to force you to write a bunch of code to do some fairly complex math.

Seriously, within the Kerbol system, signal delay has no meaningful effect on Science! transmission.  Whatever the speed of light, the delay between transmitting and receiving the points is so small as to be inconsequential, even with OPM installed.  ALL that signal delay does is turn probe cores into model airplane RC sets, nothing but a receiver and some servos.  No autonomy at all, no brains.  Like WW2-vintage optically guided weapons.  If the probe is too far from Kerbin to be flown hands-on by somebody with a joystick at KSC, then players have to write programs so the probe can fly itself, because the only way to implement signal delay is to make direct hands-on control (like you have without signal delay) be so far behind events as to be useless. 

True, this is how it works in real life.  Teams of folks highly skilled in several separate Dark Arts spend years writing complex programs for distant space probes or even boosters that never even reach orbit.  But putting this into the game means an individual player has to master all these Dark Arts alone.  That is, knowing the proper orbital mechanics equations to use for the job, knowing enough about programming in general to convert these equations into self-correcting flight control algorithms, and knowing enough about the specific language in use to do so in that language.  And then have the time to do this and repeatedly debug the software, just to get the rocket to function as desired.  All of which is in addition to just building the rocket itself.

Of course, some folks groove on this, which is why kOS exists and why RT has the option to include signal delay.  But for the few folks who both can and WANT to do this, the signal delay option isn't even necessary to have because they'd write the same programs with or without it, just because that's what they enjoy doing.  For everybody else, this is sheer torture getting in the way of their fun, which is just flying rockets.

Fortunately, the unmodded game provides an easy solution.  You can control probes in real time no matter their distance because you, the human, are pretending to be the AI of a highly autonomous robot.  Thus, you can ASSUME there's however much signal delay you want and it doesn't matter, because your probes are modern, autonomous vehicles capable of flying themselves (because the highly skilled, multi-talented team behind the scenes at KSC built them that way), as opposed to ancient, dumb, WW2-vintage tech you should be embarrassed to send even to LKO.  Then you don't have to write all the code to do this, and don't need to feel ashamed by this.

So really, when you boil it all down, having signal delay does nothing good for anybody.  Those who like writing control algorithms will do so whether there's actually delay or not, while those who don't enjoy coding (or lack sufficient training even to attempt it) will be forced into it anyway if signal delay exists.  And as mentioned, it has no impact on Science! at least within the Kerbol system.

The exact same arguments against signal delay also applies to probe control at interstellar distances.  It just shouldn't be stock and there's really no need even to make it a mod.  The difference with interstellar is the impact on Science!  But as I outlined in a post above, putting a significant delay on Science! transmission pretty much kills the idea of building a cohesive space empire or even making an interstellar trip within a reasonable amount of gametime.  Thus, it's best not to have it at all.  This leaves players free to enjoy the game however they want.

Link to comment
Share on other sites

8 minutes ago, Geschosskopf said:

Signal delay and automation are totally separate things.  One has no effect on the other.  For instance, in KSP1, you can install either MJ or RT, you don't need both at once.

They are separated in the same sense that driving a racecar and having good eyesight are totally separate things. Yes, they are separate. But that does not make them unrelated. Signal delay does create the need of some automation. And luckily, RT provides that (at least when I played with it), so there is indeed no need to have both.

8 minutes ago, Geschosskopf said:

If you think signal delay is a good idea in KSP, then you've either never actually used it OR you're a proficient kOS coder and enjoy writing probe-flying scripts at least as much as you enjoy building and flying rockets.  But If you're the later, we wouldn't be having this conversation because you'd surely recognize that 1) not everybody is good at writing kOS scripts, 2) not everybody enjoys writing code enough to want to do that instead of flying rockets, and thus 3) that signal delay is a VERY BAD thing to force on majority of players, who aren't good or willing coders.  This is because the ONLY gameplay impact of signal delay is to force you to write a bunch of code to do some fairly complex math.

The KOS language and me don’t agree with each other, so I stay away from that. But RT is very playable without it; you can schedule burns with a fairly high degree of flexibility. No need to write code for that. Where you do need advanced automation is when you want to land a probe, especially when you can’t use parachutes for that. That’s where some kind of “autoland” option would come in. If that’s in place there wouldn’t be a need to write code for it, either.

Finally, I used the word option. So I don’t understand why it would be “forced” upon players. The very definition of optional is that it’s... well, optional. It’s a choice. Which is pretty much the opposite of “being forced.”

Link to comment
Share on other sites

The more I think about it the more I wonder if he gets why Rice played Texas.

 

hes basically asking for a way to automate landing because he’s not going to be able to do it in real time with “realistic delays”

just between earth and moon the average delay is 1.25 seconds. One way. Which is why remote landing even on the moon is not easy. Seriously why do you think we actualy used airbags on some Mars landers? Granted airbags would not have saved Schiaparelli. 

Link to comment
Share on other sites

3 hours ago, Kerbart said:

They are separated in the same sense that driving a racecar and having good eyesight are totally separate things. Yes, they are separate. But that does not make them unrelated. Signal delay does create the need of some automation. And luckily, RT provides that (at least when I played with it), so there is indeed no need to have both.

Umm....  No.  Signal delay creates the need for AUTONOMY, which is a totally different thing from automation.  Automation was invented in the Neolithic (irrigating fields with a waterwheel instead of peasants with buckets, for example) if not even earlier.  Automation is a dumb device doing a pre-set, hardwired routine.  It makes no decisions, it makes no in-process adjustments.  It just does the same thing over and over regardless of inputs.  Autonomy does what automation cannot.  It works via algorithms instead of fixed mechanical (or hard-coded) linkages.  These algorithms accept a range of input values and crunch the numbers to arrive at the  same desired result by different routes.  They can even monitor their own progress and adjust themselves in-process to help end up with the desired result.

 

3 hours ago, Kerbart said:

The KOS language and me don’t agree with each other, so I stay away from that. But RT is very playable without it; you can schedule burns with a fairly high degree of flexibility. No need to write code for that. Where you do need advanced automation is when you want to land a probe, especially when you can’t use parachutes for that. That’s where some kind of “autoland” option would come in. If that’s in place there wouldn’t be a need to write code for it, either.

The preplanned maneuvers in RT are holdover from ancient times when the base game did not allow a ship to have more than 1 node at a time, and also did not remember a ship's nodes when you switched focus to another ship or KSC.  Nowadays, you can make long chains of nodes for the same ship and the game saves them all so they're still there (maybe years/decades later) when you return focus to that ship.  Given that even the shortest interplanetary trip is far longer than the signal delay, you have no problem at all manually plotting all the nodes you need far in advance and setting them to autoburn.  The end result is EXACTLY the same as, if not better than (due to more precise execution of the burns),  doing all these burns manually in realtime.  There's no rush to create the nodes so there no need to compensate for the signal delay.  IOW, this whole thing is a scam, making you think signal delay has any meaning at all for such simple things as individual maneuvering burns, when it really doesn't.

But, as you say, doing anything more complex than a single maneuver node can handle requires not automation but autonomy.  Landing is just one example.  Launch,  rendezvous, and docking are others.  For all of these, you have a choice.  #1 is that you can accept what the game gives you in allowing realtime, direct control of probes regardless of their distance from home, which basically means that you as the player are the probe's autonomous, self-correcting control algorithm.  In this case, signal delay could equally well be zero or infinity for all the difference it makes to gameplay because the probe is flying itself without need of input from home.  #2 is to disable direct, realtime control of probes beyond LKO and then write flight control algorithms to do the same job you'd do yourself if you were controlling the probe in realtime.   In which case, signal delay could ALSO range from zero to infinity without it making the slightest difference on the probe, which is, again, flying itself without input from home.

From this you can see that the ONLY way signal delay has any impact AT ALL on gameplay is if it prevents the direct, realtime control of probes like we have without signal delay  If it does that, then it condemns all players to doing nothing with probes but flyby, impact, or simple orbital missions.  Everything else requires either a pilot (and thus having to bring him home) or the underlying ludicrous assumption that your space program somehow missed out on the Silicon Revolution which itself, in real life, was largely a spin-off of, and synergistic with, the space program.   Which makes no sense unless you're setting your space program in the 1950s.

 

3 hours ago, Kerbart said:

Finally, I used the word option. So I don’t understand why it would be “forced” upon players. The very definition of optional is that it’s... well, optional. It’s a choice. Which is pretty much the opposite of “being forced.”

C'mon, man.  The whole concept of signal delay for probe control is totally bogus outside the 1950s.  It's an insult to your Neolithic umpteenth great-granddaddy who invented the waterwheel, not to mention any society capable of landing a booster on a barge at sea.

 

51 minutes ago, [email protected] said:

hes basically asking for a way to automate landing because he’s not going to be able to do it in real time with “realistic delays”

That was my impression, too.

Link to comment
Share on other sites

13 hours ago, Geschosskopf said:

Of course, some folks groove on this, which is why kOS exists and why RT has the option to include signal delay.  But for the few folks who both can and WANT to do this, the signal delay option isn't even necessary to have because they'd write the same programs with or without it, just because that's what they enjoy doing.  For everybody else, this is sheer torture getting in the way of their fun, which is just flying rockets.

Fortunately, the unmodded game provides an easy solution.  You can control probes in real time no matter their distance because you, the human, are pretending to be the AI of a highly autonomous robot.  Thus, you can ASSUME there's however much signal delay you want and it doesn't matter, because your probes are modern, autonomous vehicles capable of flying themselves

This.

It's impossible IRL to control probes in real time with signal delay, it would be unrealistic to have to do so in KSP.

Having a stock KoS with a real programming language (Lua or Python are totally fine) would be more realistic than any implementation of signal delay.

Link to comment
Share on other sites

33 minutes ago, Master39 said:

This.

It's impossible IRL to control probes in real time with signal delay, it would be unrealistic to have to do so in KSP.

Having a stock KoS with a real programming language (Lua or Python are totally fine) would be more realistic than any implementation of signal delay.

This. 

Should be made into an option. It will be too hard for newbies to learn coding, even if I love the idea.

Link to comment
Share on other sites

36 minutes ago, Xd the great said:

This. 

Should be made into an option. It will be too hard for newbies to learn coding, even if I love the idea.

I fully agree! Even without it being optional, n00bs can just strap on a Kerbal on their craft so they wont have to deal with any com delays.

Terminology aside, a certain degree of automation/autonomy would be very welcome in KSP2, particularly since we'll be dealing with space/planetary infrastructure, supply and logistics (since we get LS), and interstellar travel (for which the devs already mentioned they have included some sort of autonomy). Fingers crossed.

Edited by Yakuzi
Link to comment
Share on other sites

45 minutes ago, Xd the great said:

Should be made into an option. It will be too hard for newbies to learn coding, even if I love the idea.

Not only that, the implementation of coding would be all we need to enjoy the realism of signal delay, actual signal delay would just be useless at that point.

 

If it were for me I would totally replace the action group system with a graphical programming interface, it would be as easy to use as the action group UI and even at the lowest level it would enable a lot of quality of life automation.

Link to comment
Share on other sites

Make signal delay an option that is turned off by default. So hard core players happy, and regular players are happy. 

Make a stock kOS like script system. Turned off by default. So hardcore players are happy, and regular players are happy.

But if it's not done so.... well, there will be mods anyway right ?

I bet kOS will be running on KSP2 a few weeks after the release. Maybe even before if moders get an access to the beta.

Link to comment
Share on other sites

20 minutes ago, Challyss said:

Make a stock kOS like script system. Turned off by default. So hardcore players are happy, and regular players are happy.

It doesn't even need to be off by default, it can just be an additional option on probe cores or SAS modules.

Link to comment
Share on other sites

The thing is if they actually implemented that someone would write the code and put it up as a mod. Then anyone could just download and install the bot and let the bot play the game for them.

 

at which point I  have to wonder why they just skip the “buying the game and install it” and just watch a twitch streamer. It’s cheaper and twitch playes on iPads 

Link to comment
Share on other sites

×
×
  • Create New...