Thoughts on stock communication system in ksp 1.1

[Guest]

This new feature looks almost exactly like the functionality you get with the AntennaRange mod. AR had selectable options to require LOS for probe control, fixing the power cost of antenna transmissions, and allowing for piggyback transmissions from several antennas to beam Science back to Kerbin. That's almost exactly what this feature does, aside from power cost fixing, and making the Planet Kerbin into a giant antenna itself.

I am absolutely fine with this implementation. I personally used AR mod back when I still ran a modded installation, and I'm glad to have the feature available for my stock game.

There is one minor nitpick, I hope the existing descriptions of the antenna parts are updated for the correct transmitter bandwidth values. It would also be nice to know how much ElectricCharge it costs to transmit 1 Mit(?) as part of the description.

Then there is one major nitpick, loss of probe control, but that has already been brought up. I don't like the idea of my very first probe being sent to an unexplored planet turning into space junk because I don't have any tools to calculate occlusion angles. Nor do I think those tools should be necessary.

I think the simple solution is to allow the player to retain flight control of a probe, so long as that probe is high enough on the tech tree. You could call it 'Computer Controlled Probes', and only have the final tier of probe cores have that ability. Then there will be more reason to research better probe cores and use them.

Other than that, I think this is a cool feature, and I'm glad it's here. Now we just need Squad to hire Claw to do dedicated bug smashing work.

Yes, but while playing in the sand, you make this:

https://theroadhouse.files.wordpress.com/2010/07/sandcastle.jpg

Now that is one very impressive sandcastle! My only fear is that the v1.1 update might be the wave that washes it all away. Is there any word on whether the v1.1 update will be savegame friendly?

I can't speak for others, but I personally would have quit the game long ago if launch times were realistic. Frankly I'm getting to the point now where the current abbreviated launch time is more than I'm willing to suffer.

MechJeb Launch Guidance is your friend. Program your ascent profile, hit go, then go make a sandwich.

There is no way Squad can pull the a support rabbit out of its hat. Proper life support, anything deserving of that name, means Kerbals die if X or Y resources are depleted over time.

Not if they use the SNACKS model, which to my mind would be the only "life support" feature that is simple and easy enough to be made stock. Granted, I don't want life support at all, but I don't speak for the developers.

[RoverDude]

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').

Can't really go into details as several options and balance factors are being weighed, but since that was one of the larger feedback items, it's been under a lot of discussion, and I wanted to make sure folks knew it was being addressed so that we can move the conversation forward. And thanks again for keeping this discussion going, it's very helpful.

[micha]

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').

That's great news; thanks for the update.

I personally prefer the idea of "the player is the automated control" (which leads to the player able to muck up a burn (think cosmic rays interfering with the electronics)) instead of some sort of automated system when occluded. Otherwise it doesn't make so much sense in-game why one can't use the automated system when not occluded (even if there are some detriments to it).

Many people have commented on this, so it's mostly rehashing, but imho, while occluded:

• Can't send any data back to Kerbin (well, duh)
  
• Can't create new maneuver nodes / see the patched conics, etc (ie, you ONLY get the nodes you pre-programmed as an aid to your burns)
  
• Turning the probe and gathering science should be allowed (it's "pre-programmed"; eg a fly-by on the occluded side of a body).
  

Perhaps even prevent burning when too far away (in time) from a maneuver node. Still need to allow centering the burn on the node, as well as allowing for long burns, but my idea here is to let the player "burn the pre-programmed nodes" but not tweak the burn before or afterwards (at least not until comms is re-established). Maybe only allow turning the engines on/off ONCE per node (but still allow varying the throttle during the burn)?

-> These ideas are mostly for an orbital-insertion burn; not sure how well these ideas would work for a landing.. then again a "pre-programmed" landing should be very difficult compared to an insertion burn.

Hence it'd still be FAR more beneficial to have comms, but even without the player is still required to perform the actual maneuver.

Can even possibly add an end-game tech "Probe AI" which is expensive and heavy (although still cheaper and lighter than a Kerballed pod) to reinstate most of these features.

Anyway, looking forward to see what you guys come up with.

[Geschosskopf]

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')

I think the autonomy of probes needs to selectable by the player, not set by default. This would give the players the freedom to set up probes to fit their idea of how their universe works. For example, if you envision your game set in the Space Race using WW2-vintage technology, then you obviously need more hands-on control than somebody whose game is set in more modern times or even the near future, where fully autonomous vehicles either come into existence or are by then the norm.

For instance, even today all you need to do is give general orders, such as or "drive my drunken self home from this bar" and the vehicle itself will handle all the details, such as picking the best route and negotiating all obstacles and other traffic en route. And this even in big cities with scads of moving parts to avoid and multiple paths to choose from. This seems a far more complex problem to solve, with vastly more decision points and opportunities to screw up, than anything that spacecraft do. Spacecraft just need to do a handful of maneuvers and only have traffic to contend with if they seek it out intentionally. Yet we've had cars able to do this for several years now and they will, in short time, become generally available, the deployment time mostly being sure they're safe to other traffic. In the less-congested air, at sea, and in space, these things have been deployed longer and autonomous drones are even available to the general public.

With this type of autonomy, there is only a need for very brief, sporadic intervals of communications if the mission changes and to transmit data back. Otherwise, they're fire-and-forget. In the absence of a stock autopilot, the only way to "simulate" this degree of autonomy is to allow the player full control of probes at all times, whether they're in communication or not. Furthermore, computers are far more precise than human pilots anyway, so probes should never be penalized on what they can and can't do compared to Kerbals.

[Dodgey]

I am honestly very excited for this. It sounds like what I thought that RT was before I realized how tedious it really is.

[NecroBones]

Personally I like what we've heard so far about this. It always bothered me that the stock game didn't really give me a reason to use anything other than the little whip antenna. The data rate and power draw wasn't much of a factor, because even if you didn't have enough power to transmit quickly, physics timewarp was enough to get the data through quickly. Not to mention, I flew a lot of missions without antennas and just brought the science straight home anyway.

I like the idea of adding more considerations when choosing antennas, and adding a mission planning aspect, but without going all of the way to the level of tedium and/or complexity that some of the mods introduce.

The mods will still exist for those who want the complexity or realism. And for everyone else, it's a mission planning aspect that can be turned off if they don't want it. Sounds great.

[juanml82]

That's great news; thanks for the update.

I personally prefer the idea of "the player is the automated control" (which leads to the player able to muck up a burn (think cosmic rays interfering with the electronics)) instead of some sort of automated system when occluded. Otherwise it doesn't make so much sense in-game why one can't use the automated system when not occluded (even if there are some detriments to it).

Many people have commented on this, so it's mostly rehashing, but imho, while occluded:

• Can't send any data back to Kerbin (well, duh)
  
• Can't create new maneuver nodes / see the patched conics, etc (ie, you ONLY get the nodes you pre-programmed as an aid to your burns)
  
• Turning the probe and gathering science should be allowed (it's "pre-programmed"; eg a fly-by on the occluded side of a body).
  

Perhaps even prevent burning when too far away (in time) from a maneuver node. Still need to allow centering the burn on the node, as well as allowing for long burns, but my idea here is to let the player "burn the pre-programmed nodes" but not tweak the burn before or afterwards (at least not until comms is re-established). Maybe only allow turning the engines on/off ONCE per node (but still allow varying the throttle during the burn)?

-> These ideas are mostly for an orbital-insertion burn; not sure how well these ideas would work for a landing.. then again a "pre-programmed" landing should be very difficult compared to an insertion burn.

Hence it'd still be FAR more beneficial to have comms, but even without the player is still required to perform the actual maneuver.

Can even possibly add an end-game tech "Probe AI" which is expensive and heavy (although still cheaper and lighter than a Kerballed pod) to reinstate most of these features.

Anyway, looking forward to see what you guys come up with.

This makes sense. About the patched conics, I guess it should still show either the trajectory as it was before occlusion and/or the projected trajectory after the burn, but would only update that (say, to correct for inaccuracies with the maneuver) once it's back in contact with the KSC.

And, as you advance through the tech tree, maybe the probes can recalculate the trajectory on its own and the best, and more expensive probes, can create new nodes anyway.

Rover wheels might be another category of sorts, as they require more precise maneuvering which can be justified to be impossible

[gogozerg]

I personally prefer the idea of "the player is the automated control"

For me that's the only way of looking at it, otherwise many things can't make sense.

In unmanned flights, the player is the AI controling locally.

I wish we could have realistic remote control of ships, but it's just not possible with KSP design where you take decisions based on live view of your ship.

Btw, RemoteTech is not very convincing, and we would need more than newtonian physics.

[micha]

In unmanned flights, the player is the AI controling locally.

I look at it more as being the probe's flight-computer executing a pre-programmed set of instructions; an AI would be able to adjust to each situation automatically. The difference only really makes sense once you introduce a game mechanic where you lose communication with the flight-computer, as the proposed Communications System will have.

Ie, a Flight-Computer must be pre-programmed (eg, burn for 20 seconds retrograde), and can't change its programming (although it can and should perform additional minor adjustments if, for example, the craft veers slightly away from retrograde heading while under acceleration) unless in communication with a controller; whereas an AI can do whatever it needs to in order to execute a broad objective (eg, enter a circular polar orbit at 200km).

[TMS]

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').

This statement worries me slightly.

A small piece of advice from the casual observer... don't go down the road of arbitrarily disabling certain functions when occluded, but leaving other functions working. It might not make a great deal of sense, long-term.

My prediction here is that you'll plump for an option where movement (rotation, pitch and translation) is locked out, but thrust is enabled (and, god help us, maybe SAS on just the higher-tier probes). The gameplay mechanic being that if you place a manoeuvre node before you're occluded, you can point your craft towards that node, lose control, but still execute the manoeuvre (as a proxy for a flight computer).

I'm not quite sure where that places the player in terms of whether they're the local AI or they're mission control (sans relativity).

Nonetheless, the important feature that needs implementing is some form of indication of if/when occlusion will occur.

[jwenting]

long as it's optional (iow, I can turn it on or off for a new save) it's a good idea.

But after playing with RT for a while I got bored of the tedious tasks of setting up massive satellite networks everywhere just so I can control my probe as it goes behind a moon...

[klgraham1013]

long as it's optional (iow, I can turn it on or off for a new save) it's a good idea.

But after playing with RT for a while I got bored of the tedious tasks of setting up massive satellite networks everywhere just so I can control my probe as it goes behind a moon...

For people who find RemoteTech tedious, AntennaRange may be more in line with the stock implementation.

[Sandworm]

long as it's optional (iow, I can turn it on or off for a new save) it's a good idea.

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.

I cannot be the only one here who dreads starting a new career-mode save. I know that no matter where I put the sliders I will eventually face a grinding cliff of procedural missions for irrational rewards. I want to play career mode, but have had too many negative experiences and now stick to science-only. Adding another slider for "life support" will only multiply this problem.

Edited July 27, 2015 by Sandworm

[regex]

I cannot be the only one here who dreads starting a new career-mode save. I know that no matter where I put the sliders I will eventually face a grinding cliff of procedural missions for irrational rewards.

You're not the only one.

For me, the entire contracts system needs to be completely redone into something coherent, maybe organized per celestial with the ability to block certain contract types from even spawning, or allowing the user to create mission proposals, or something. The appearance of the non-procedural missions seems totally random as well. Basically it's terrible.

As far as life support is concerned, I'd prefer if permadeath were the default and lack of LS killed Kerbals but without permadeath killing Kerbals is a poor option since there's no incentive to keep them alive beyond finishing their mission. We won't get a detailed LS sim, though, we'll get some "snacks" or "supplies" or something silly like that, and kerbals will just be unresponsive or something.

Where are all the consequences in this game? Where is the time and budget management? The tension? Career mode lacks all of that.

[jwenting]

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).

not really.

Can be like the "indedestructable buildings" option, or the slider to set heating effects.

Turn it on for max difficulty, maybe a medium setting where you have to be in range without occlusion, and off for if you don't want to be bothered.

[John FX]

I think my preference for any added features that is supposed to replicate reality in some way is for the game to internally calculate its reality as accurately as possible ( at least have the code there so that it could if it wanted to ) but then provide to the player increasingly simplified versions of that until you pretty much get the game holding your hand for very very new players. This would allow very experienced players to play with a very very realistic fully featured game engine and all the players from fresh new players all the way up to Whackjob could have the level of gameplay they desire or need so that the game is the most fun for them.

[mikegarrison]

not really.

Can be like the "indedestructable buildings" option, or the slider to set heating effects.

Turn it on for max difficulty, maybe a medium setting where you have to be in range without occlusion, and off for if you don't want to be bothered.

I always select indestructable buildings. The idea of damage to the buildings was pretty good, but the implementation was terrible. Even a slightly hard landing on the massive concrete launch pad suddenly causes it to blow up like it was made of nitrocellulose. Sure, I guess massive explosions are the kerbal way, but that was just too much to accept for me.

[Guest]

Can't really go into details as several options and balance factors are being weighed, but since that was one of the larger feedback items, it's been under a lot of discussion, and I wanted to make sure folks knew it was being addressed so that we can move the conversation forward.

Thank you RoverDude, a little reassurance goes a long way.

allowing the user to create mission proposals, or something.

This would be an excellent solution, choose the parent body, choose the type of mission, and the game will generate a reward based on how many times you have completed said mission type before, or what the parent body is, etc.

[Sandworm]

not really.

Can be like the "indedestructable buildings" option, or the slider to set heating effects.

Turn it on for max difficulty, maybe a medium setting where you have to be in range without occlusion, and off for if you don't want to be bothered.

That's my point. Between all the different game modes and sliders, there isn't any core game. I might sound oldschool about this, but I believe a game should have a core, something to point to as the standard mode of play. That doesn't mean a lack of freedom. Look at mincraft: a core survival mode with a couple options. KSP doesn't have such a thing. Take the three game options (sandbox/science/career) then and the stock difficulties (low/medium/high) and you have 9 game modes. Then say three heat options (off/medium/hard) ... that's 27. Life support on/off? Now its 54. Radio communication on/off? Now 108. Three resource levels? 216. Crew respawn? 432. Science/funds rewards ratios? ... Starting a new career means picking from literally a thousand different paths before you even see the VAB.

It's no wonder that the youtubers have backed off career mode and KSP generally. It's hard to start a let's play when you cannot really point to a common standard for starting a "stock" game. Too much freedom of choice becomes a burden. It really turns off new or casual players.

Reply to below: I think the youtubers want to connect with their audiences. They want to play the game in a way that their fans might. So they seek some sort of common game mode. The games that do well on youtube have that default (minecraft, subnautica, stranded deep, pre-career KSP etc). That's also why they shy away from mods. Note that some of the top reviewers have stayed far away from KSP. Totalbiscuit likes KSP, but how is he to honestly review the career mode? He'd get a few hours in, realize he started with less-than-ideal gameplay settings, and have to restart.

Edited July 27, 2015 by Sandworm

[regex]

It's no wonder that the youtuber's have backed off career mode and KSP generally.

I would have ascribed that to every new series looking exactly the same (seriously, career mode) but I suppose you have a point.

[Alshain]

The YouTubers backed off career mode because it's all the same, not because they are all too different. Everyone has seen and played the entire way through the tech tree by now. The game doesn't really get interesting till you finish the tech tree anyway.

[problemecium]

I predict that this new antenna occlusion thing will be rather like the 1.0 aerodynamics.

First, people will rage about it because it makes things harder and more complicated. The realism idealists will rebuke them at every turn.

Eventually people will adjust to the new status quo (in some cases begrudgingly), but General Discussion will for a few weeks be flooded with "My probe not work, pls help" threads.

After a few months people will accept it as the way things should be and wonder how KSP could have been considered complete without it.

[BurnPrograde]

I really hope the lines of communication are dimmed down on the map screen or are toggle-able. I dislike a cluttered map view

[mikegarrison]

Who cares about "youtubers" anyway? When I like a game I want to play it, not watch other people play it.

[Alshain]

I predict that this new antenna occlusion thing will be rather like the 1.0 aerodynamics.

First, people will rage about it because it makes things harder and more complicated. The realism idealists will rebuke them at every turn.

Eventually people will adjust to the new status quo (in some cases begrudgingly), but General Discussion will for a few weeks be flooded with "My probe not work, pls help" threads.

After a few months people will accept it as the way things should be and wonder how KSP could have been considered complete without it.

I don't think it will be as bad as the aero.

- - - Updated - - -

I really hope the lines of communication are dimmed down on the map screen or are toggle-able. I dislike a cluttered map view

Do we know for sure such lines will exist? The way they describe the system, I'm not sure they will be necessary like they are in RemoteTech. In any case we don't know that yet.