Thoughts on stock communication system in ksp 1.1

[Geschosskopf]

No. The "it's optional" argument is a crutch for poor decisions. If Squad wants to add core features to a released product then they must force themselves to iron out the balance prior to release. Stock KSP already saddles users with too many balance decisions (science/heating/rewardsx3). The result is, more often than not, an unbalanced mess. One should not have to grind through career mode a half-dozen times before finding all the sweet spots. That job belongs to the developers. The game needs to be fun out of the box, all the way to the endgame content.

And this is exactly why I in general oppose all attempts to add "realism" features to the stock game. Keep the stock game like it was pre-1.x and focus on fixing its innumerable bugs, glitches, imbalances, lack of coherence, and such. Leave realism features to the modders, who do a better job with them because they don't have to pull their punches and can get focused feedback directly from users. Most realism mods have been in development for years already so are much more mature than anything Squad whips up in a couple of months.

Seriously, 1.0 made both reentry heating and "realistic" aerodynamics stock. Yet DRE and FAR are still on the market and scads of people use them. Why? Because they do a better job than the stock systems. IOW, the people who whined the loudest about making these features stock don't even use the stock system but are still using the same mods as before, and the people who didn't want these features in stock got stuck with half-baked, bug-ridden systems anyway. And these bugs affect even those who use the mods. So nobody is happy with the result, meaning making these features stock was counterproductive to Squad because it irritated the entire customer base.

So yeah, I'd like to see things optional. But optional as in staying in mod form, not stock implementation.

Edited July 28, 2015 by Geschosskopf

[LostOblivion]

Realism is not even "usually" harder. People throw around "tedious" as if realism would require every single man-minute of work be modeled. In fact, we need only concern ourselves with what KSP bothers to model at all, and within that, what fidelity is possible. To the extent there is any correlation at all, I'd not be surprised if more realism within what stock KSP does would be easier, not harder. At 1:1 scale (set kerbin to some reasonable size for a small, earth like planet). All the parts are scaled in lockstep, so similar looking craft behave the same. The atmosphere would scale less, since stock is too deep anyway. Distances are a non-issue because of time warp, the only change is launch/reentry given limited time warp under acceleration. Docking is easier in many ways since you'll have more daylight if you plan properly.

Planets could have some bigger, some the current size to make various difficulties. Minmus stays small, the Mun gets scaled up to the point that 1-stage vs 2-stage landers both have rationales.

I'd argue that most of the difficulty in stock KSP is added in after the fact because of the tiny scale used, yet trying to make it feel realistic. Look at reentry, if kerbin acts like earth, nasty things happen at other worlds.

On topic, this new feature creates balance issues as mentioned above for the simple reason that the game is not realistic about life support (that's sort of understating it since KSP is infinitely unrealistic WRT life support since it has none at all ).

It's a peeve of mine to see difficulty stated as a function of realism with zero data to support it. Whenever anyone tries, they build a straw man on the order of "if you don't have to explicitly feed, sleep, and excrete with your kerbals, it's not realistic, so why bother at all?"

My point isn't realism, but difficulty and challenge. There are a lot of players who enjoy RemoteTech and AntennaRange, me included, and would like their features to be stock. My point is making the game suit more players by adding complexity to the difficulty options. If people want the game to be more difficult by having to make their own DSN, then this should be configurable.

- - - Updated - - -

As I understand the current system, if you send an unmanned ship to orbit the Mun and you need to do the circularization burn while you are over the Mun's far side, you won't be able to do it unless there is a comm sat somewhere.

This is not realistic, so realism is not the issue of this particular feature.

Exactly, which is why some people here ask for a flight computer to queue maneuvers, etc. You don't have to use the Flight Computer in RemoteTech if you think that is making the game too tedious, but it's a nice feature for those that want to take part in that kind of planning detail. I don't really use RT's FC, so I don't have any desires for this, however.

[mikegarrison]

There are a lot of fundamental problems that have come up because of the decision early on to make the KSP solar system so small. I do think that many of the problems come from the fact that the rockets have to be nerfed (mainly by making them small and way too heavy) in order to make staging even the least bit necessary, plus re-entry heating has to be artificially induced because re-entry is happening at 1/4 of the real speeds. Those both lead to problems for the whole rest of the game.

But I think that ship has long sailed. If Squad was to ever do a full-size solar system, we would obviously be looking at KSP 2.0. So much has to change -- all the physics, all the bodies, all the parts, and even the buildings would have to get bigger. Plus, all the people who love spaceplanes would probably scream because they would be so much harder (and they would look more like lifting body wedges than like F-16s).

[richfiles]

I didn't play mods for my first month or two. I switched from Sandbox to Career once I realized the science equipment was useless to me, and enjoyed it. I STILl didn't run mods. I eventually found MechJeb, and a couple part packs. I Like MJ, but I still usually launch manually. I use it for landing targeting, and ABSOLUTELY for maneuver nodes (The editor is AMAZING to have, over stock), and cause Smart A.S.S. rocks. I've also used the rover tool before. I tried Kethane, and liked it, but Mac support was lagging, and even in .25, it was throwing up compatibility warnings at launch. Clearing it out made my game more stable, so I abandoned it.

As far as other mods go... Nothing more. Never tried FAR. Never tried Deadly Re-Entry. Never tried TAC, Never tried RSS, never tried Remote Tech...

I looked at ALL OF THEM... but never decided to try then out.

I now play with stock re-entry heat, I enjoy the new aero, I am trying to set up my first ISRU base on Minmus... I DO play mods, and I've never touched the mods that once provided those features. All the people here who say "get mods" forget that not everyone has a kickass machine with tons of RAM and a sweet CPU. Not everyone is the type to even CONSIDER mods. The mod community is an AMAZING thing, and they have created TRULY AMAZING works... But you reach only a fraction of the players. Don't kid yourselves. Stock brings this to EVERYONE, and the difficulty levels ARE adjustable on all these features. And you know what? I look forward to trying those other features. I will probably wait for stock, but I WILL give it a shot.

If I don't like how stock works... Guess what? I CAN TURN IT OFF!!!

If I don't like HOW it works, but like the general idea... Guess what? I CAN GET A MOD!!!

Cool your jets! Geez! All these people whining about the stock antenna stuff being too watered down, or to much added difficulty to deal with in stock... JUST ADJUST/MOD to your little Kerbal loving' heart's content! Seriously... The attitudes here are just DUMB. I trust that the game will continue to give me enjoyment, and that I can expect to look forward to more cool spacey things that make green dudes go fast and up! Hopefully.

Speaking of mods... I want to try that one... The one with the extra planets, but not RSS... Outer Planets I think it's called. That's gonna be my reward for finishing my Simpit. Still gotta build the hardware controller for my ADI/FDAI navball!

[Randazzo]

I imagine it will be somewhat like the ISRU system or the Asteroid Day parts. Neat idea, fun at first, then it gets turned off and/or forgotten as it just becomes a chore.

Stuff that adds more work just for the sake of adding work gets old quick in my book.

Edited July 28, 2015 by Randazzo

[flyboy67109]

Remote Control Mod did this but I had to stop using it because it would constantly crash KSP. I spent more time fixing files then I did playing the game. I loved it when it worked. Not so much when it didn't. I feel it would be a nice addition.

[VektorWorlds]

I wouldn't be too concerned. With bodies with SOIs too small for synchronous orbit, just stick a few satellites in high orbits, and one or two in a polar orbit. For bodies with SOIs big enough, just make a synchronous orbit network, possibly with a few in polar orbits for polar coverage. Doesn't seem that bad to me! And with objects like the Mun, then you can just make a network for orbital craft and land the others on the kerbin-facing side. And if you don't like it, TURN IT OFF!

[gogozerg]

FYI - there will be a solution regarding occluded ships (i.e. direct control is always better, but we need something in between 'full control' and 'no control').

What is the definition of an "occluded ship" please ?

Is it based on a model with instant communications?

[RoverDude]

Occluded meaning there's a planetary body obstructing your vessel and any other vessel that can serve as a relay or control point.

So a probe on the dark side of the mun with no line of sight to Kerbin, and no other satellites to bounce off of would be occluded.

(Also for clarity) as has been stated before, signal loss is not on the table. So all control is, from the point of view of the player, instantaneous.

[gogozerg]

(Also for clarity) as has been stated before, signal loss is not on the table. So all control is, from the point of view of the player, instantaneous.

Signal loss or signal delay?

I'm sorry, but dealing with occlusions when control is instantaneous sounds absurd. Instantaneous control means local control (and you don't care about occlusions). You don't know what will happen to a probe flying 100,000,000 km away, 5 minutes after you sent a message.

KSP is nice for many things about space flight, aerodynamics, heat model and so on. Not a perfect simulation, but a rather good one that teaches something.

Space teaches us that time matters, and that simultaneity is a complex problem. More realistic communications would be nice indeed. But the game just isn't designed for it, with its live views and controls, and newtonian physics.

Edited July 29, 2015 by gogozerg

[sp1989]

I'd like to see a simple simulation for probe autonomy.

Example: WHen controlling a probe and you lose connection, you can continue to control the probe, but cannot relay any data back until connection is restored. If you switch off to a vessel that does not have control over the probe, you will not be able to switch back until connection is restored.

Ultimately I'm not at all opposed to it, I think it will have an interesting place. That said, I'm thoroughly frustrated Squad and co. are putting a sizable quantity of development effort into something that further increases the brick wall of a learning curve of this game rather than putting aforementioned dev time into making it not run like poo at times.

I'm still waiting for squad to address memory leaks for Mac

[cantab]

My two cents:

I've used Antenna Range a lot and liked it. I've tried Remote Tech somewhat briefly and rather disliked it. So I feel Squad and Roverdude's ideas are pretty good.

Why I dislike Remote Tech relates to what I would like to see and not see in stock. My biggest beef with RT is it's built on premises that are so unlike the real world. Only one ground station and doing everything by satellites, really? Sputnik was launched and tracked just fine around the world and it certainly didn't have a huge comsat network to support it! And the ground station having far inferior performance than the portable dishes, seriously? Remote Tech has been designed to force the player into a certain approach, it's not about "realism" in the slightest, and that certain approach of setting up a Kerbin comsat network just becomes gruntwork after a few saves. I feel, then, that Squad's system should be built on sound premises. That includes the assumed presence of a Deep Space Network meaning connections only need to be to "anywhere on Kerbin" (or set multiple specific ground stations if you really want), and the ground-based antennae being the best performing. Relay satellites will still be useful to connect to the far side of other planets and moons, or to boost the signal of a craft with a weak antenna. It's also good to avoid encouraging "grunt work" and to give the player multiple viable options, we have enough gruntwork in early career/science mode as it is.

My second problem with Remote Tech is the flight computer is sorely lacking. No editing an action already set, no RCS translation, no obvious indication you have additional delay enabled, I could go on. Now I'm unsure whether signal delay should be added in stock because it really makes probes in KSP an entirely new game, it turns it from being about flying to being about computer programming, and I'm not sure that's the direction KSP should take. But if Squad do feel that's worth adding, I would want to see much better probe programming tools than what RT offers. And using a more normal syntax than the weird thing kOS has too!

That's the ideas. As for the implementation, well that's where I'm sceptical. Squad have a poor recent record of implementing their new ideas. The aerodynamics does seem basically bug-free but it received two game balance patches after 1.0 that messed with people's designs. The heating on the other hand is riddled with serious bugs. So even though it's a different programmer (Roverdude) working on the antenna stuff I'm not confident.

Finally I'll be more positive and suggest something I think would be really nice: Realistic(ish) calculation of communication ranges instead of hard-coded figures. Give each antenna a transmit power and a gain, assume sensible values for bandwidth and noise floor, and it's fairly easy to calculate data rate over the link. Then impose a minimum data rate for various tasks. This would add complexity but I think make for some nice aspects. For example a Mun lander with a low-gain antenna might be able to communicate back to the big dishes on Kerbin because those dishes have lots of power and gain, but not be able to communicate with a closer satellite with another low-gain antenna.

[Rainbow-Dashie]

I love this idea, and can't wait to see it in game.

But at the same time, with the new aero model, and now this... I can feel myself slipping down the learning curve

[RogalDorn]

This will be a fun expansion. You don't need to play with it, but you can if you want to! Very Excited to lose control of most of my Space Program LOL!

[RoverDude]

Signal loss or signal delay?

I'm sorry, but dealing with occlusions when control is instantaneous sounds absurd. Instantaneous control means local control (and you don't care about occlusions). You don't know what will happen to a probe flying 100,000,000 km away, 5 minutes after you sent a message.

KSP is nice for many things about space flight, aerodynamics, heat model and so on. Not a perfect simulation, but a rather good one that teaches something.

Space teaches us that time matters, and that simultaneity is a complex problem. More realistic communications would be nice indeed. But the game just isn't designed for it, with its live views and controls, and newtonian physics.

Sorry, signal delay (it was typing early).

And I consider the time frame in which a command is executed from the time a player touches a physical keyboard to the time the in-game ship 'does something' orthagonal to local control. So the design goal is to simulate, in a reasonably believable way, how that probe would behave when occluded. YMMV.

- - - Updated - - -

I'm still waiting for squad to address memory leaks for Mac

Apples and tractors - there's probably a different thread for that topic

My two cents:

I've used Antenna Range a lot and liked it. I've tried Remote Tech somewhat briefly and rather disliked it. So I feel Squad and Roverdude's ideas are pretty good.

Why I dislike Remote Tech relates to what I would like to see and not see in stock. My biggest beef with RT is it's built on premises that are so unlike the real world. Only one ground station and doing everything by satellites, really? Sputnik was launched and tracked just fine around the world and it certainly didn't have a huge comsat network to support it! And the ground station having far inferior performance than the portable dishes, seriously? Remote Tech has been designed to force the player into a certain approach, it's not about "realism" in the slightest, and that certain approach of setting up a Kerbin comsat network just becomes gruntwork after a few saves. I feel, then, that Squad's system should be built on sound premises. That includes the assumed presence of a Deep Space Network meaning connections only need to be to "anywhere on Kerbin" (or set multiple specific ground stations if you really want), and the ground-based antennae being the best performing. Relay satellites will still be useful to connect to the far side of other planets and moons, or to boost the signal of a craft with a weak antenna. It's also good to avoid encouraging "grunt work" and to give the player multiple viable options, we have enough gruntwork in early career/science mode as it is.

Yes, stock will assume a planet wide DSN with an expanding range based on your Tracking Center upgrade level.

My second problem with Remote Tech is the flight computer is sorely lacking. No editing an action already set, no RCS translation, no obvious indication you have additional delay enabled, I could go on. Now I'm unsure whether signal delay should be added in stock because it really makes probes in KSP an entirely new game, it turns it from being about flying to being about computer programming, and I'm not sure that's the direction KSP should take. But if Squad do feel that's worth adding, I would want to see much better probe programming tools than what RT offers. And using a more normal syntax than the weird thing kOS has too!

I am not a fan of signal delay, and it is not on the table at this time. That being said, we're exploring some interesting options from a gameplay standpoint that put you (roughly) in the same place, but will not require a flight computer UI (the latter being a massive can of worms). I expect signal delay and flight computers will remain a mod thing.

That's the ideas. As for the implementation, well that's where I'm sceptical. Squad have a poor recent record of implementing their new ideas. The aerodynamics does seem basically bug-free but it received two game balance patches after 1.0 that messed with people's designs. The heating on the other hand is riddled with serious bugs. So even though it's a different programmer (Roverdude) working on the antenna stuff I'm not confident.

Well, I did do resources and that turned out reasonably well Also I'd say this system is significantly simpler than something as far reaching and complex as thermal and aero. I've been having a blast setting up and testing little relay networks in my dev build.

Finally I'll be more positive and suggest something I think would be really nice: Realistic(ish) calculation of communication ranges instead of hard-coded figures. Give each antenna a transmit power and a gain, assume sensible values for bandwidth and noise floor, and it's fairly easy to calculate data rate over the link. Then impose a minimum data rate for various tasks. This would add complexity but I think make for some nice aspects. For example a Mun lander with a low-gain antenna might be able to communicate back to the big dishes on Kerbin because those dishes have lots of power and gain, but not be able to communicate with a closer satellite with another low-gain antenna.

I assume what you're covering there is something more akin to the additive range feature in the existing antenna mods?

[ZentroCatson]

It sounds pretty good, but I'm not that much into launching probes for making interplanetary communication networks. Luckily Squad has the Kerbin network thingy! The system is a good idea for making satellites do something, but like a lot of forum people except for two dudes I want clouds!​

[CalculusWarrior]

I want clouds!​

I agree; the work of rbray and blackrack with their respective visual mods is phenomenal, and I badly want both EVE and Scatterer to be made stock. One day...

[tater]

I think that programming a probe need not be any more complicated than "make a maneuver node." The probe then executes the node when it gets there. Problem solved. Players already can "program" a node, the only thing that makes it a "program" is to have the probe execute it---following an m-node is already in game, too. There is a countdown to the node, and it knows when it is done. Seems not too complex an issue given everything already there.

[RoverDude]

It sounds pretty good, but I'm not that much into launching probes for making interplanetary communication networks. Luckily Squad has the Kerbin network thingy! The system is a good idea for making satellites do something, but like a lot of forum people except for two dudes I want clouds!​

I agree; the work of rbray and blackrack with their respective visual mods is phenomenal, and I badly want both EVE and Scatterer to be made stock. One day...

I think your princess is in another castle... or rather, your feature request is in another thread

[Errol]

...I am not a fan of signal delay, and it is not on the table at this time. That being said, we're exploring some interesting options from a gameplay standpoint that put you (roughly) in the same place, but will not require a flight computer UI (the latter being a massive can of worms)...

Very interested to see where this goes, as I too am not a fan of instant control.

[ZentroCatson]

I think your princess is in another castle... or rather, your feature request is in another thread

Well, this tread is for "thoughts" about the system. I just shared my thoughts about the comms system and that I would want clouds before the comms . Still, your work on this new system is awesome! I love the look of the new comms dishes.

[Barklight]

I agree; the work of rbray and blackrack with their respective visual mods is phenomenal, and I badly want both EVE and Scatterer to be made stock. One day...

That would be very nice, but I don't expect to see that until U5 and a much more reliably stable release of KSP x64 for Windows simply due to memory footprint. (That and Scatterer is best suited for DirectX, even though it mostly works in OpenGL.)

On topic, I like this pseudo-RT2 / Antenna Range stockification. I have been playing with Remote Tech for a long time and I like it a lot - it forces you to build a bit of orbital infrastructure before you venture out into the unknown. I think the new communications setup will be a good addition to stock, though I'll miss the challenge of planning my actions meticulously with signal delay and no immediate control to correct slip-ups As much as I love RT2's flight computer in orbital / transfer flights, the headache it gives me when landing probes on other bodies is enough that I won't miss having a flight computer.

Looking forward to being able to give the new system a spin.

[micha]

I think that programming a probe need not be any more complicated than "make a maneuver node." The probe then executes the node when it gets there. Problem solved. Players already can "program" a node, the only thing that makes it a "program" is to have the probe execute it---following an m-node is already in game, too. There is a countdown to the node, and it knows when it is done. Seems not too complex an issue given everything already there.

True; but then what's to stop people using that system everywhere. Bingo, you effectively have full perfect autopilot everywhere, something Squad has been against since forever. I also can't think of any good in-game justification why it would ONLY work while probes are occluded. Furthermore, the maneuver node system as-is is insufficient for landing burns.

Such a system would also prevent things from going hilariously wrong. When the player performs a burn there's always the chance the burn isn't perfect. Hitting the wrong key, drifting off course, getting the timing wrong, etc. And this is, arguably, what a lot of KSP is about. Sure, a lot of people play KSP closer to a Sim, but for those people there's mods.

[Red Iron Crown]

True; but then what's to stop people using that system everywhere. Bingo, you effectively have full perfect autopilot everywhere, something Squad has been against since forever. I also can't think of any good in-game justification why it would ONLY work while probes are occluded. Furthermore, the maneuver node system as-is is insufficient for landing burns.

Contradictory. It would only be good for things that maneuver nodes work well for, other things like ascent, landing, docking, etc would not have "full, effective autopilot".

[Aragosnat]

I do like the looks of the new parts. The signal range will certainly make it more of a challange. Going to suck for my 8 craft heading towards Duna right now as from the sounds of it. My level 2 tracking station might not cut it. But, only if it is not auto-disabled on active saves.

Still the idea is sound and something I would expect to happen sooner or later. Plus it means I don't have to get a mod for that funtionality.