[1.1] RemoteTech v1.6.10 [2016-04-12]

[Drew Kerman]

@Steven - Cheat menu location would be fine, but really anywhere in game the player could toggle it on/off would be fine too

@damerell - you're still missing the point or I haven't explained it well enough. I'll assume the latter, in which case let me try again. RemoteTech is a mod that implements signal delay in the game, in other words it allows you to play with signal delay by using its flight computer to control the vessel. That should be the player's choice - if they want to play with signal delay they can use the supplied flight computer, if they don't they just control it as normal. However RemoteTech over-reaches in its design and enforces signal delay by also applying it to the flight controls themselves, which if you are choosing to play with signal delay using the flight computer, you would have no reason whatsoever to touch anyways! No one would ever want to control a ship under signal delay by using the actual controls (although I admit it would make for a fun reddit Challenge or something). So overall it's a rather pointless and frustrating bit of "realism" from a gameplay standpoint.

And yes, RemoteTech also forces you to use signal delay when it's activated when selecting part actions, but you can get around that by assigning them to an action group and disabling signal delay (for action groups only) through Action Groups Extended. I'd still rather RemoteTech allow you to switch it off itself, but it's still a solution that works.

[Dunbaratu]

@damerell - you're still missing the point or I haven't explained it well enough. I'll assume the latter, in which case let me try again. RemoteTech is a mod that implements signal delay in the game, in other words it allows you to play with signal delay by using its flight computer to control the vessel. That should be the player's choice - if they want to play with signal delay they can use the supplied flight computer, if they don't they just control it as normal.

There's no reason to limit fuel when the player could just choose to shut the engine off after a limited time.

There's no reason to have limited thrust engines when the player could just use a high thrust engine and choose not to throttle it up all the way.

You're pretending that the flight computer was added as the means to *implement* delay, when you know perfectly well it was added as the means of *dealing with* the delay that RT implements, not implementing it in the first place.

I have no problem with saying that once RT's limits are active in a saved game they are active fully for that saved game. Enforcing that is sort of the point of RT. My annoyance is with not having an option to selectively choose which savegames are subject to RT's limitations and which aren't.

RT is like a hard mode for the game. But it should be like the stock game's hard mode options like no-revert - a hard mode option you pick per savegame, not something that once you use it in one saved game it destroys your ability to ever use any other saved games that aren't set up for it.

Right now the only way to do that is to maintain entirely separate installations of KSP itself, which seems like overkill just to work around the lack of an option to de-enforce RT, and requires some ModuleManager magic if you want to still be able to test craft designs that contain RT antenna in them, but without RT's functionality.

[Drew Kerman]

Well since no else seems to be in support of this, I'll just fork my own version.

Steven if you were the one who actually fixed the bug like you said you were going to do a couple weeks/months back, would you mind sharing what I need to undo? Otherwise I guess I can just peruse the Diffs...

Edited September 12, 2015 by Gaiiden

[Peppie23]

jo jo, small update today:

General

• Added a new value to RTSettings to keep the throttle on connection loose (ThrottleZeroOnNoConnection=True or False)
  
• Clean up vessel target handling (thx to geoffromer)
  
• Added Asteroid Day antennas (thx to phroggster)
  
• Added NovaPunch antennas (thx to blnk2007)
  

[rottielover]

" Added NovaPunch antennas (thx to blnk2007) " - Question, is that on a request basis? Can someone add the AIES dishes and Antennas? (is that already done and I missed it?)

[Borskey]

I'm having some trouble with the "integrated 3.0 Km omnidirectional always on" antennas that are supposed to be built into the probe cores.

This antenna doesn't seem to function properly in my game. For example of how they are supposed to work- in this video https://youtu.be/0sUjsI9SpB4?t=297

He builds a simple probe rocket with no added antenna, launches it, and then loses signal after a short distance. It's a demonstration for why you need an additional antenna to get further than 3km.

However, If I try to launch this exact same rocket in my own game(just probe core+SRB), my probe has no connection at the launchpad and I can not stage or do anything.

My issue is, I'm building a space station, and I wanted to use little probe RCS tugs to re-arrange and dock all the different modules. They should always be within 3 km of the space station, which will always have a connection to KSC, so I shouldn't need anything other than the integrated probe core antenna. However, they are not working as they should- just like how the probe SRB doesn't work for me.

Do I really have to add DP-10 antennas to my orbital construction RCS tugs?

Edit: Nevermind, looks like you need the "Remote Tech" node unlocked for the integrated antenna to work. Didn't realize.

Edited September 13, 2015 by Borskey

[Lilienthal]

Problem with flight computer and holding direction

Hi all,

for quite a while I am having problems with using the flight computer to point the craft a certain direction (i.e. facing a maneuver node) on RSS/RO/RP0.

The behaviour I am seeing is that the controls (pitch, yaw) are always either +100% or -100%, thereby overshooting a lot and wasting heaps of precious propellant.

Is that intended behaviour? Do either people see the same?

I imagine this could be solved by implementing something like:

pitch_control = if (pitch is already moving in the right direction) {0.1 * (pitch_actual-pitch_target)} else { 0.2 *(pitch_actual-pitch_target)}

That at least has worked for me in kOS.

This probably would need to go into: https://github.com/RemoteTechnologiesGroup/RemoteTech/blob/develop/src/RemoteTech/FlightComputer/FlightCore.cs Function SteeringHelper, but I am not quite sure.

What do you think? Would that help? Is already someone working on that?

Tbh, I can kind of read C# (used C++ 10 yrs ago), but don't know how to compile etc. If anyone wants to pair up with me, I am happy to give advice on the math side.

Best,

Gustav,

PS: Sorry if I missed an already happening discussion, couldn't find a recent one.

[Poodmund]

Does the inclusion of the:

ControlAntennaWithoutConnection = True

... line in the setting config file now render the RemoteTech XF redundant? Does it do the same thing?

[Farix]

Does the inclusion of the:

ControlAntennaWithoutConnection = True

... line in the setting config file now render the RemoteTech XF redundant? Does it do the same thing?

I believe that line is from RemoteTech XF itself.

[zamboni]

Hiya all.

I recently noticed a problem that I *thought* was due to several mods I installed a day or so ago, but the problem looks to be with RT, and there all along.

I was on the launchpad and tried to send some science, but the transmission was instant, and didn't yield any type of science for it (and the data collected was cleared). The communication was apparently was being sent through the launch clamp but never got through. I spent time tonight deleting the several mods that I installed, and I had the same behavior. It wasn't until I looked at the RT .cfg files that I realized what is happening. Here is a screenshot of the debug after tying to send some science back:

I believe it may have been like this all along, because I am pretty sure now all of the science I have transmitted from the launchpad was without using the clamps (I usually just strapped a new science part on a mk1 capsule sitting on the launchpad itself and then did the science that way and then did a quick recovery).

What I think is happening: Although the clamps are fitted with an "antenna" in the .cfg file:

@PART[launchClamp1]:FOR[RemoteTech]
{
	%MODULE[ModuleRTAntennaPassive] {
		%OmniRange = 5000
	}
}

There is no associated transmitter parameter. So when it tries to transmit, but there is no parameter there it gets sent to nowhere, and the data is cleaned out. I was able to fix this by inserting a random parameter:

@PART[launchClamp1]:FOR[RemoteTech]
{
	%MODULE[ModuleRTAntennaPassive] {
		%OmniRange = 5000

	%TRANSMITTER {
			%PacketInterval = 0.15
			%PacketSize = 3
			%PacketResourceCost = 0.1
		}
	}	
}

Science is now sent and credited accordingly.

[damerell]

@damerell - you're still missing the point or I haven't explained it well enough. I'll assume the latter, in which case let me try again. RemoteTech is a mod that implements signal delay in the game, in other words it allows you to play with signal delay by using its flight computer to control the vessel. That should be the player's choice - if they want to play with signal delay they can use the supplied flight computer, if they don't they just control it as normal. However RemoteTech over-reaches in its design and enforces signal delay by also applying it to the flight controls themselves, which if you are choosing to play with signal delay using the flight computer, you would have no reason whatsoever to touch anyways!

I think it's fairly clear that the design was to enforce signal delay. How you choose to deal with that - the flight computer, MechJeb, kOS, SmartParts, or even just putting up with it (the .16s to Minmus is perceptible but hardly overwhelming, and the delay to a local command centre might be longer but not prohibitive, so the idea that if signal delay applies you'll never touch the flight controls doesn't hold up) - is up to you.

I see absolutely no problem with Steven's per-savegame idea. I see no problem with a debug menu cheat to turn off signal delay. I can't, however, see any point in making one particular mode of control immune to signal delay because one particular user wants it.

The player's choice is simple. If they want to play with signal delay, turn on signal delay.

[Peppie23]

So, summer vacation are now over and i'm back.

Here are the current flight plan for the next feature release v1.7.0 (dated for 31.10)

- RemoteTech_Settings into the savegame folder & RT option Window

- Flightcomputer overhaul = add RCS Translation & Rover Computer

- API needs a way to get connection status of individual antennas

- Feature: Better status for local controlled ships

- Enable other mods to push custom commands to the flightcomputer

- Adding maneuvers from the maneuver node to the flightcomputer

I'll do my best to keep the defined release date for this feature branch.

[Dunbaratu]

- Enable other mods to push custom commands to the flightcomputer

What would it take to for you to add generic text input fields to this? I read through the github issue, and it looks pretty old, but it only mentions simple fixed one-shot events, events tha are invisible in RT, and events where the mod tells RT what to do. But for us kOS devs what would be really nice is if you allowed us to go the other way around.

Instead of "kOS sends a request to RT's flight computer" it would be "RT's flight computer sends a request to kOS".

The idea being to allow RT users to embed kOS commandline strings into the RT flight computer, that RT tells us to execute when the event fires off. The only sticking point being we'd need some way to get the user to pick which kOS CPU part its talking about. when the craft has more than one.

is that general idea too much of a change or is it pretty simple?

Edited September 14, 2015 by Steven Mading

[Lilienthal]

Great Stuff.

So, summer vacation are now over and i'm back.

Here are the current flight plan for the next feature release v1.7.0 (dated for 31.10)

- Flightcomputer overhaul = add RCS Translation & Rover Computer

I'll do my best to keep the defined release date for this feature branch.

Does this mean that the problems I mentioned in post #3508 will be solved? (i.e. the problem that the RCS attitude control can only do +100% or -100%, oscillates a lot and uses heaps of propellant.) - Or is this not a common problem?

Thanks,

Gustav

[Peppie23]

Hi Steven,

... but it only mentions simple fixed one-shot events ...

Not only. You can create a one-shot event to the flight computer how SirDiazo would need this for his ActionGroup Mod. Small preview: http://i.imgur.com/TgwybAk.jpg (The Receive Data - will trigger a callback method from another mod)

The second way to create an event is, for example, like our "Hold prograde". So, for another example: The Autopilot from mechjeb. If this event is triggered we will call your defined callback until you will tell us that your event is done OR the user will stop this event from the flight computer (via cancel command - that we also trigger to you)

... The idea being to allow RT users to embed kOS commandline strings into the RT flight computer ...

If we can't bind kOS and RT with this functionality its up to you (or the other kOS Devs) to help me to do this. Create a Task on our github page and tell me what i should do, or make a pull request with this feature. i'm absolutely not familiar with kOS in any case.

- - - Updated - - -

Hi Lilienthal,

... RCS attitude control can only do +100% or -100%, oscillates a lot and uses heaps of propellant.

hmm na not really, i only thought about an functionality to create a command like: rsc thrust prograde for 5s or 5m/s (what ever). BUT i'm not opposed to add another slider to control the rcs thrust, so yeah maybe

I have no Infos about a problem that the flight computer attitude program would drain so much monoprop for the rcs thrusters, but in any case our SteeringHelper would need an overhaul as well ^^ but not for 1.7

[Dunbaratu]

Hi Steven,

Not only. You can create a one-shot event to the flight computer how SirDiazo would need this for his ActionGroup Mod. Small preview: http://i.imgur.com/TgwybAk.jpg (The Receive Data - will trigger a callback method from another mod)

The second way to create an event is, for example, like our "Hold prograde". So, for another example: The Autopilot from mechjeb. If this event is triggered we will call your defined callback until you will tell us that your event is done OR the user will stop this event from the flight computer (via cancel command - that we also trigger to you)

Yes I saw that but as far as I can tell neither one allowed the user to type in a string argument for the event in the RT Flight Computer user interface. It's the ability to get user-supplied type-in data that you then send to us when you tell us the event is happening that I'm referring to. That would allow us to support people typing in generic kOS commands into the RT flight computer's future events.

If we can't bind kOS and RT with this functionality its up to you (or the other kOS Devs) to help me to do this. Create a Task on our github page and tell me what i should do, or make a pull request with this feature. i'm absolutely not familiar with kOS in any case.

I know next to nothing about RT's internals but a few of the other kOS devs do. I'll bring this up with them.

I have no Infos about a problem that the flight computer attitude program would drain so much monoprop for the rcs thrusters, but in any case our SteeringHelper would need an overhaul as well ^^ but not for 1.7

I hear rumors that both kOS and RT cropped their SteeringHelper code from the same original source. We've just undergone a massive overhaul to ours (not released yet) that helps repair a lot of its ugly behaviors.

[Svm420]

1.7.0 sounds like everything I have wanted from RT for a long time. Can't wait.

[blnk2007]

" Added NovaPunch antennas (thx to blnk2007) " - Question, is that on a request basis? Can someone add the AIES dishes and Antennas? (is that already done and I missed it?)

EDIT: There is already an AIES config.

Edited September 15, 2015 by blnk2007

[Panel]

I've been wondering for a while what is going to happen to the mod when 1.1 is released. Will it overwrite the game's mechanic?

[Peppie23]

... Will it overwrite the game's mechanic?

We'll see what squad brings in with ksp 1.1

[federicoaa]

Hi, I'm trying to get a full list of commands for the computer but I can only find basic info.

I want to know, for example, how to do staging and actions groups.

Thanks.

[Peppie23]

Hi, I'm trying to get a full list of commands for the computer but I can only find basic info.

I want to know, for example, how to do staging and actions groups.

Thanks.

Hi federicoaa,

do you mean kOS ? We have no command list ... or i don't understand you ^^ To stage or toggle the action groups just hit your defined keys for it

[federicoaa]

Nono, but I actually found some videos of people using RT and I think I know now, but the manual does not explain very well I think (there should be more examples)

I did not know how to make, for example, that the ship deploy panels (lets say action group 1), I thought you had to input something, KOS like.

But I discovered that you set the delay and just press '1' and it's done. It's so simple it's not intuitive

[BananaDealer]

Hmmm... I seem to have ran into some connection issues since the latest RT update. When using third-party antennae (like the ones in the Tantares mod, or the Modular Rocket Systems one), which are supposedly RT-compatible. Basically, probe craft list as "No Connection" even while on the launchpad pre-launch. I'm gonna try to reproduce with stock parts, if possible.

[Gryphon]

Hmmm... I seem to have ran into some connection issues since the latest RT update. When using third-party antennae (like the ones in the Tantares mod, or the Modular Rocket Systems one), which are supposedly RT-compatible. Basically, probe craft list as "No Connection" even while on the launchpad pre-launch. I'm gonna try to reproduce with stock parts, if possible.

Make sure they have the correct modules being added. There are plenty of examples of this as ModuleManager patches included in the RemoteTech directory.