Thoughts on stock communication system in ksp 1.1

[tater]

The entire relay concept in RT made me instantly abandon it, it makes no sense.

[rcp27]

That's why they were unable to communicate with the probe while it was actually flying by Pluto.

http://in.reuters.com/article/2015/07/06/space-nasa-pluto-idINKCN0PG24B20150706

That isn't the reason they were unable to contact it during the actual flyby. The reason it had to be out of communication during the flyby was that the both the sensors and the antenna on it are directional, and in order to orient the probe to collect its data that resulted in it orienting itself so as not to be able to communicate with mission control. Once the data collection was complete it reoriented itself for communications and transmitted the data.

On the subject of occlusion by intermediate celestial bodies, if the occluding body is not either the sun, the moon or Pluto/Charon, the distances involved are such that it is almost certain that the radio transmissions would diffract around those bodies, so effectively occlusion is not an issue.

None of these effects, ether the directionality of sensors, the limitations on the CPU capabilities of probe cores or the diffraction of radio waves is considered in any KSP mod I am aware of, let alone stock.

[mikegarrison]

most people don't think about the speed of light in their daily lives (though many of us here do, I'm sure)

So what? Most people don't think about delta-v in their daily lives either. Should KSP not model delta-v?

- - - Updated - - -

That isn't the reason they were unable to contact it during the actual flyby. The reason it had to be out of communication during the flyby was that the both the sensors and the antenna on it are directional, and in order to orient the probe to collect its data that resulted in it orienting itself so as not to be able to communicate with mission control. Once the data collection was complete it reoriented itself for communications and transmitted the data.

All right, that sounds believable.

[Tarmenius]

So what? Most people don't think about delta-v in their daily lives either. Should KSP not model delta-v?

I wasn't actually defending instantaneous light. My only point was that the simplification is a low-cost design choice, unlike the choice to disable probes completely when they lose communication with Kerbin.

Personally, if they decided to make light (and therefore signals) travel the appropriate speed, I don't know how I'd feel about it. At interplanetary scales, light feels a lot slower than you might think. And now I'm curious... which mod models this?

[gogozerg]

But... ground control doesn't race against the clock. They've issued commands to New Horizons to take all those pictures of Pluto long before the five hours or so before those pictures had to be taken. Had something been in on the way and the signal gotten blocked, NASA would have known with a lot of advance. As for moons, known (largish) moons and their orbits are well known by now.

Yes, space probes often fail. But not because the people in charge of those missions failed to account for signal delay or occlusion.

As you said... ...They've issued commands to New Horizons to take all those pictures of Pluto long before... ...NASA would have known with a lot of advance...

NASA is playing a different game.

[Varsi]

I wasn't actually defending instantaneous light. My only point was that the simplification is a low-cost design choice, unlike the choice to disable probes completely when they lose communication with Kerbin.

Personally, if they decided to make light (and therefore signals) travel the appropriate speed, I don't know how I'd feel about it. At interplanetary scales, light feels a lot slower than you might think. And now I'm curious... which mod models this?

Using Remote Tech mod you really need to use the built in flight computer to execute pre-planned maneuvers outside Kerbins SOI. Direct controls start getting a bit laggy already at Minmus. The fun thing with RT flight computer is that you can also cancel commands (staging and such) but the command had better have some delay if you expect the cancel-command to get there before the command is executed.

[Thomas988]

Yes, we will have to deal with some form of antenna range and object occlusion in (as yet) unspecified degrees. It was announced last week by Ted.

Thanks, Horseman! This sounds really darn cool!

[juanml82]

As you said... ...They've issued commands to New Horizons to take all those pictures of Pluto long before... ...NASA would have known with a lot of advance...

NASA is playing a different game.

The question is: how do you model that in-game? Because translating that directly seems like building an autopilot

[mikegarrison]

I wasn't actually defending instantaneous light. My only point was that the simplification is a low-cost design choice, unlike the choice to disable probes completely when they lose communication with Kerbin.

Personally, if they decided to make light (and therefore signals) travel the appropriate speed, I don't know how I'd feel about it. At interplanetary scales, light feels a lot slower than you might think. And now I'm curious... which mod models this?

Remote Tech models this.

[tater]

There are no good ways in KSP to model programming things, and light lag given the inability to deal with anything out of focus. Craft should be able to do what they are told to do, even if the player is busy doing something else. Not just a KAC warning for the player to watch the craft, then effectively manually send the EXEC command in RT, but to have the craft automatically execute any maneuver node it comes upon.

I suppose a work-around would be to have the game check the craft vs the planned node, and determine how much of that node it can actually execute using the current stage, then calculate the resultant patched conic for that. Next time the player selects that craft after the node has passed, it uses the new conic. If you make a node that the craft can actually fulfill, then the planned node (dashed line) is the conic the game will use. The craft would show an inaccurate path in the tracking station, I suppose, until you selected it, then it would retroactively switch (sort of like catching up ISRU after being out of focus).

[mikegarrison]

There are no good ways in KSP to model programming things, and light lag given the inability to deal with anything out of focus. Craft should be able to do what they are told to do, even if the player is busy doing something else. Not just a KAC warning for the player to watch the craft, then effectively manually send the EXEC command in RT, but to have the craft automatically execute any maneuver node it comes upon.

I suppose a work-around would be to have the game check the craft vs the planned node, and determine how much of that node it can actually execute using the current stage, then calculate the resultant patched conic for that. Next time the player selects that craft after the node has passed, it uses the new conic. If you make a node that the craft can actually fulfill, then the planned node (dashed line) is the conic the game will use. The craft would show an inaccurate path in the tracking station, I suppose, until you selected it, then it would retroactively switch (sort of like catching up ISRU after being out of focus).

Yes, this dependency on being "in focus" is a big problem for implementing anything like realistic programmed behavior. Bottom line is that the game concept just does not really allow for robot ships. There are mods that try to work around this somewhat, but none of them feel 100% successful. And when playing stock, there is essentially no difference between manned and unmanned except that manned is not quite totally as bricked as unmanned if you lose all power.

[LostOblivion]

I think the ideas shared by RoverDude sound pretty good for a stock system, but I think a DSN should be more than just having Kerbin "be" the antenna. If stock is to emulate a DSN, it would make sense to model actual tracking stations around Kerbin and that you have to be in touch with at least one of them (as IRL).

Perhaps this could be a completely other building progression. Tier 0: No DSN. Tier 1: Full Kerbin coverage, but only so far. Tier 2: Much farther.

No matter how Squad solves it, it'll be nice having RemoteTech use the stock system as a base for its own additional constraints.

[juanml82]

I think the ideas shared by RoverDude sound pretty good for a stock system, but I think a DSN should be more than just having Kerbin "be" the antenna. If stock is to emulate a DSN, it would make sense to model actual tracking stations around Kerbin and that you have to be in touch with at least one of them (as IRL).

Perhaps this could be a completely other building progression. Tier 0: No DSN. Tier 1: Full Kerbin coverage, but only so far. Tier 2: Much farther.

No matter how Squad solves it, it'll be nice having RemoteTech use the stock system as a base for its own additional constraints.

I understand a proper ground system will have full coverage (and just in case, the ground stations are connected to each other with submarine optic fiber, not through satellites in GEO), so while it would be nice to have two additional tracking stations at different places in Kerbin (so they work like a sort of easter egg), I think the game doesn't need to check and simulate if the craft is in direct line of sight with one of the three ground stations

Edited July 30, 2015 by juanml82

[monstah]

You guys remember the ionosphere, right? Radio signals bouce off it from within, that's why we had radios long before satellites.

In-Kerbin communication should never be a problem. I'm happy with the solution Squad is using.

[rkman]

If stock is to emulate a DSN, it would make sense to model actual tracking stations around Kerbin and that you have to be in touch with at least one of them (as IRL).

IRL a probe always has line of sight with one of the ground stations (aside from occlusion by planets/moons), so in practice modeling the connection to individual ground stations comes down to the same thing as simply modeling a connection to the planet.