Radio Signal Ping

[AEROSPACE guy]

KSP has got it right in terms of physics, except for one thing. That is, that if we send a command to an unmanned probe, the response is immediate. But it should take some time for the signal to reach the craft. So there should be a programmable flight computer I'm KSP 2 so that it can be preprogrammed to do all of its manuvers. This would enhance the realism of ksp

[Geonovast]

Moving to KSP 2 discussion, since this is a request for a feature in KSP 2.

Also, welcome to the forum, @AEROSPACE guy!

[adsii1970]

43 minutes ago, AEROSPACE guy said:

KSP has got it right in terms of physics, except for one thing. That is, that if we send a command to an unmanned probe, the response is immediate. But it should take some time for the signal to reach the craft. So there should be a programmable flight computer I'm KSP 2 so that it can be preprogrammed to do all of its manuvers. This would enhance the realism of ksp

I am not responding as a member of Squad, TakeTwo, or any other organization. The comments reflect my personal opinion.

With that said, just so no one thinks I am speaking officially for the makers of KSP2, I think this is something which could be implemented fairly easy within the game. But, rather than making it a hard setting, I'd recommend a slider to set the delay. This way, someone who wants a realistic experience could say, slide it all the way to the extreme point on the slider bar. Those who don't want a delay could slide it to the opposite end of the slider. And those like me, who wouldn't mind a little delay could set it according to our tastes.

But to be honest, we do not know exactly what we will be seeing in KSP2 - there's simply not enough to say for sure without a degree of speculation. I guess we will have to wait to see what any update and development notes tells us more about the game.

[DStaal]

There have been a couple of threads on this.  The pros of course are increased realism, and increased challenge.  The cons that the challenge is in an awkward place, and you'd have to give the player additional tools to deal with it - tools which would either themselves be a large challenge to use, or if they're simplistic may not meet the needs.  (While making some of the play easier.)

I can see the point, but I'll admit I'm mostly on the side against - I don't really think it adds anything valuable to gameplay for most players.  I can see it being a difficulty option - but you'd need a way to automatically execute at least maneuver nodes.

[sgt_flyer]

Well, for pure gameplay purposes, the comms delay would be difficult to play out. 

It could be justified in severals ways though -

1- you could tell that your manual manoeuvers of the probe is simply a preprogrammed action sent some time before the manoeuver begins

2- the kerbals are more advanced than us in rocket AIs, so the probe is capable to manoeuver on it’s own in order to reach it’s mission goals  (though, being a kerbal AI,  it can mistime it’s manoeuver or perform the wrong one ;))

 

nevertheless, as they plan mod support for ksp 2, there would more than likely be a mod for that  (as there actually is some mods for that in current KSP ^^)

 

 

 

 

Edited September 19, 2019 by sgt_flyer

[DStaal]

4 minutes ago, sgt_flyer said:

1- you could tell that your manual manoeuvers of the probe is simply a preprogrammed action sent some time before the manoeuver begins

2- the kerbals are more advanced than us in rocket AIs, so the probe is capable to manoeuver on it’s own in order to reach it’s mission goals  (though, being a kerbal AI,  it can mistime it’s manoeuver or perform the wrong one ;))

How is either of those different than *not* having com delay?  (How do you know it's not the case in the current KSP, without mods?)

[sgt_flyer]

7 minutes ago, DStaal said:

How is either of those different than *not* having com delay?  (How do you know it's not the case in the current KSP, without mods?)

I meant to provide possible ingame justifications on why the current no delay comms gameplay exists. - as you said yourself - comm delays would likely not make a good gameplay experience. (At least, not for most people)

as for KSP1, the remote tech 2 mod can provide comms delay. (Though, it also comes with a flight computer in order to preprogram the manoeuvers and execute them at a specific timing) 

Edited September 19, 2019 by sgt_flyer

[Incarnation of Chaos]

An interesting idea since KSP2 has been confirmed to have out of focus acceleration would be whenever a craft is in focus; you have no delay. This is because you're either directly controlling it; or simulating the programs that would be running the commands. But whenever a craft is out of focus; you can still place manuever nodes but the craft will only execute them with a delay equal to what the light lag would be. This means that you can still run an entire mission the same way KSP1 allows you to; but if you want to send a massive interstellar ship on a semi-automated expedition that you then have to contend with signal delay.

This would ofc be a toggle if implmented

[Laie]

17 hours ago, AEROSPACE guy said:

[Comm delay] would enhance the realism of ksp

Indeed it would, but be careful what you wish for. Have you ever played with Remote Tech? It does model comm delay, and gives you a flight computer to program ahead of time.

Using it for a few missions is a worthwhile experience in my opinion. But I for one would hate it if I had to use it all the time.

16 hours ago, adsii1970 said:

I'd recommend a slider to set the delay.

I see no benefit in that. Delay is either short enough as to barely matter, or so long that you have to pre-program your actions.

--

As a side note, our current science system doesn't play well with comm delay. You can't really tell which biome you'll be over at what point in time, so either your flight computer has a "biome sensor" and knows when to run the experiments, or you have to run them in short intervals and hope for the best. Neither solution makes for good gameplay, though. For me that's the worst aspect of using comm delay.

[Guest]

2 hours ago, Laie said:

You can't really tell which biome you'll be over at what point in time, 

That's the solution right here, make it possible to pre-plan science gathering. Though TBH, the whole science system in KSP sucks. I hope KSP2 remakes into something less grindy and more logical.

[Guest]

21 hours ago, AEROSPACE guy said:

KSP has got it right in terms of physics, except for one thing. That is, that if we send a command to an unmanned probe, the response is immediate. But it should take some time for the signal to reach the craft. So there should be a programmable flight computer I'm KSP 2 so that it can be preprogrammed to do all of its manuvers. This would enhance the realism of ksp

There are mods for KSP1 that do this. It drastically changes the gameplay. Basically, probes are right out unless you're comfortable programming in KOS. So KSP wouldn't be about building and flying spacecraft anymore, it'd be about programming them, and that would have a much more limited appeal. I probably wouldn't bother and I am a programmer.

So no, this isn't going to happen. I expect it will be moddable, however, so realism fans will be able to make an overhaul that has them waiting eight years for signals from their interstellar probes.

[Guest]

Don't be so certain. You don't need kOS at all, if we had a way to sequence MechJeb commands, this would be a quick and easy way to program a probe. A simple implementation is already possible by scheduling multiple maneuvers, it would only be necessary to expand the system to cover other commands. For instance, a "landing gear node", which would be a point on the flight plan at which the probe would toggle its landing gear.

Setting up a flight plan and having it automatically executed could be the only way to fly unmanned probes. You want real-time improvisation, pack a Kerbal. It would help make probes different from crewed ships. In KSP1, there's no real reason to use probes, and if life support is added, there'd little reason to use Kerbals if probes have no real disadvantages. Indeed, if you could set up an automatic "launch, rendezvous and dock" plan, and it could be executed with the ship being out of focus, it could also prove handy for resupplying LKO stations.

[Guest]

26 minutes ago, Dragon01 said:

Don't be so certain. You don't need kOS at all, if we had a way to sequence MechJeb commands, this would be a quick and easy way to program a probe. A simple implementation is already possible by scheduling multiple maneuvers, it would only be necessary to expand the system to cover other commands. For instance, a "landing gear node", which would be a point on the flight plan at which the probe would toggle its landing gear.

Setting up a flight plan and having it automatically executed could be the only way to fly unmanned probes. You want real-time improvisation, pack a Kerbal. It would help make probes different from crewed ships. In KSP1, there's no real reason to use probes, and if life support is added, there'd little reason to use Kerbals if probes have no real disadvantages. Indeed, if you could set up an automatic "launch, rendezvous and dock" plan, and it could be executed with the ship being out of focus, it could also prove handy for resupplying LKO stations.

That wouldn't work on any body with topography and/or an atmosphere, i.e any of them. It would simply not be possible to plan the nodes precisely enough -- you'd crash into a mountain or screw up your descent because of aerodynamics. 

The only way you could make signal lag actually work, in terms of gameplay, is if you introduced new autonomous probe cores that you could fly by taking command of them as if they were crewed -- which would make signal lag something of a cosmetic thing; it wouldn't matter at all in the Kerbol system, and you could always time-warp through it for interstellar missions (if you even cared about the Sci at that point).

My money is on this not happening. It would either break the gameplay or have to have workarounds that trivialised it like that. 

There's another thread about this btw, maybe these ought to be merged? @Snark 

 

[Guest]

There are 2 ways of implementing signal delay in KSP:

1) you can't control the probe and have to use an autopilot to play in your place.

2) you can't control the ship and you have to write your own autopilot with a KoS analogue.

In neither case you need to actually have the signal delay in place.

Unless you have a team of engineers planning and programming the missions for you like actual NASA there's no way of having a realistic signal delay.

[AEROSPACE guy]

We should only have this feature in a high difficulty level, and not in the easier levels so players can decide whether they want to have signal delay or not

[Guest]

2 hours ago, Brikoleur said:

That wouldn't work on any body with topography and/or an atmosphere, i.e any of them. It would simply not be possible to plan the nodes precisely enough -- you'd crash into a mountain or screw up your descent because of aerodynamics. 

Once again, not true. MechJeb has an autoland feature, you could simply tell the probe "at this point of orbit, land", and the UI would show you where the landing site will be, considering atmospheric drag, planetary rotation, and everything else. It would not work in every situation, but it'd be up to the player to plan where the probe should come down, so that the automated system can cope with it. You could also be creative, and if you don't have the autoland (for example, because your probe core is too primitive), plan it manually, and have a lithobraking step at the end. It only requires node creation to be a bit more precise than in KSP1. For a body with atmosphere, it's even easier, just aerobrake and schedule a chute opening.

Your mistake is that you keep thinking up reasons for direct player control to be needed. There's nothing in KSP that can't be automated by something like MechJeb. The question is only what to automate, and how to do it in a way that keeps the player relevant. It is already possible, with just one mod, to fly a complete mission without touching the controls even once, and instead, inputting data into MechJeb's various modules. This is exactly what I am proposing, only with the ability to queue commands, and integration into the node system (mostly because it's familiar).

[Guest]

3 minutes ago, Dragon01 said:

Your mistake is that you keep thinking up reasons for direct player control to be needed. There's nothing in KSP that can't be automated by something like MechJeb.

I never said it's not possible to do it. Of course you could add autolanding features or any amount of other autonomy (that's not automation anymore btw, it's autonomy which is something that's a lot more complicated). 

But it would drastically change the gameplay, and it would change it in a way that I'm pretty confident in saying would only appeal to the tiny minority of hardcore realism enthusiasts playing KSP. In other words, it's not gonna happen.

[DStaal]

1 hour ago, AEROSPACE guy said:

We should only have this feature in a high difficulty level, and not in the easier levels so players can decide whether they want to have signal delay or not

In theory I like this.  In practice - to have it in *any* difficulty level you have to have the needed autonomy features built in as well.  That is a lot of extra work for the devs to get into place for a set of features that's somewhat divisive to have in the game in the first place.  (There's still a fair amount who feel that MechJeb is 'cheating'.)  Then the question is when do you allow the autonomy features to be used?  Just with signal delay?  All the time?  Just in the easiest and hardest difficulty modes?

And having the autonomy features would change the gameplay quite a bit - you're no longer seat-of-your-pants flying, you're programming steps.  (Even if the interface is simple and easy to use.)  Does that make it easier?  Harder?  Better?  Worse?  I'm not sure.

All of this results in a lot of work and effort balancing *two* new mechanics into the game - and if it's on a difficulty option, that balance is only going to be for a subset of the players.  Is that worth the development time?  Or is it better left to mods, which don't have to worry as closely about the balance of the rest of the game?

My feeling is that it's better left to mods, at least unless the autonomy features are already something you're developing for other reasons.

[Guest]

Yeah. Since you can stick a probe core on anything, this would change KSP gameplay from "build rockets and fly them" to "build rockets, program them,^* press "Launch," then watch them fly." I think I can safely say that most of us would not find that appealing. I know there are some that would -- hence, kOS and other mods -- but they are a pretty small fraction of the player base.

^*Pre-planning manoeuvre nodes and setting up launch and landing sequences using packaged auto-take-off and auto-landing scripts is still programming, even if kOS or something like it isn't involved.

Edited September 20, 2019 by Guest

[Guest]

29 minutes ago, Brikoleur said:

Yeah. Since you can stick a probe core on anything, this would change KSP gameplay from "build rockets and fly them" to "build rockets, program them,^* press "Launch," then watch them fly." I think I can safely say that most of us would not find that appealing. I know there are some that would -- hence, kOS and other mods -- but they are a pretty small fraction of the player base.

^*Pre-planning manoeuvre nodes and setting up launch and landing sequences using packaged auto-take-off and auto-landing scripts is still programming, even if kOS or something like it isn't involved.

Well actually a kOS equivalent would be a great addition to the game for everyone, especially if with a better language and an additional graphical programming interface.

Even if you only use it to manage solar panels, propellers or fairings

 

[Guest]

Just now, Master39 said:

Well actually a kOS equivalent would be a great addition to the game for everyone, especially if with a better language and an additional graphical programming interface.

Oh I would love stock kOS! I've done a fair bit of stuff with the KAL and it's severely limiting, it would be extremely nice to have an actual programming tool to give my craft more complex behaviours. I'm not categorically against stock MechJeb either, for that matter. 

I just do not under any circumstances want to have them be mandatory. It should still be possible to do pretty much everything by just manually flying those craft -- as I said, I don't want KSP to turn into a game of programming missions, hitting the launch button, and watching them play out, and signal lag would do that.

(I think we're in agreement about this point actually...)

[Guest]

4 minutes ago, Brikoleur said:

Oh I would love stock kOS! I've done a fair bit of stuff with the KAL and it's severely limiting, it would be extremely nice to have an actual programming tool to give my craft more complex behaviours. I'm not categorically against stock MechJeb either, for that matter. 

I just do not under any circumstances want to have them be mandatory. It should still be possible to do pretty much everything by just manually flying those craft -- as I said, I don't want KSP to turn into a game of programming missions, hitting the launch button, and watching them play out, and signal lag would do that.

(I think we're in agreement about this point actually...)

Exactly my point, once you have automation having actual delay is pointless.

[DStaal]

1 minute ago, Master39 said:

Exactly my point, once you have automation having actual delay is pointless.

And without automation, the actual delay makes the game impossible.    So... What's the argument for delay again? 

[Guest]

35 minutes ago, Master39 said:

Exactly my point, once you have automation having actual delay is pointless.

Again, not true. For one, it would mean you have to plan ahead, scheduling commands while knowing that if something goes wrong, you may not be able to reprogram the probe in time. Even better, lower-tech probes might have a limited space for commands. You may need to decide whether you want the "fold antenna" command or the "do more science" command (anyone remember the Galileo antenna?). "Build the rockets and program them" is a valid way to play the game. It won't be mandatory, because probes themselves won't. I don't think the devs could be convinced to go with a "probes first, Kerbals later" tech tree, so signal lag would do nothing as long as you keep Kerbals aboard all your interplanetary vessels. 

Also, something like stock kOS is very highly likely. The devs wanted a LUA-based... something in the game, and it doesn't seem to be related to writing plugins. If this was supposed to be a major part of the gameplay, though, I'd prefer something like MechJeb with a more streamlined GUI.

Oh, and autoland is automation, at least in its simplest form. It's a pretty simple process, if your craft is designed right. Landing on an airless body is essentially a gravity turn in reverse. Doing it manually with maneuver nodes would be quite possible if KSP had a GUI to allow viewing the descent path in more detail.

[Guest]

9 minutes ago, Dragon01 said:

Even better, lower-tech probes might have a limited space for commands.

Why would this be better? More realistic, sure, but better?

9 minutes ago, Dragon01 said:

"Build the rockets and program them" is a valid way to play the game. It won't be mandatory, because probes themselves won't.

I'm currently doing a no-probes career. It's a different way to play. However it does effectively cut out much of the gameplay -- there's no point building CommNet because I just need to transmit back Science, and packing a powerful-enough antenna is enough for that since I'll just have to wait a bit for a connection.

There is no way Star Theory would do something as boneheaded as implementing a large set of systems (probes, CommNet) and then gating it behind a programming requirement, thereby ensuring that only a tiny fraction of the player base would ever use them.

There will be no time delay.