Now-defunct-thread-that-should-not-appear-in-google-search.

[KerbMav]

The only issue I really have with that is that it doesn't make any sense to send one command to the spacecraft and then build the sequence in behind it.

But it saves the player the hassle to delete and recreate (with changes he suddenly wants to implement) his whole sequence and cilph from programming something to reload and edit an uploaded sequence.

[Cilph]

Ultimately if you want more programming ability, you'll need kOS.

I have three things to report. One, I added a focus switch dialog to the map view. It's still a bit tricky because the game does not make the targets initially and currently it causes some harmless null reference exceptions (in the base game) when switching vessels. It'll also somewhat bug out tab-switching.

Secondly, someone taught me an awesome way to implement rudimentary profiling for KSP plugins and gave a neat optimization to start with. I'll be developing my own toolkit for this and use this to reduce or cache calls to reduce the performance impact.

Third: Job negotiations went into the final phase. Might be a bit early to celebrate (I did anyway), but I guess this means I actually have a job now?

EDIT: Fourth: For you coder types, there's a new kOS integration implementation on my github. It's been there for a while, but I haven't passed it off to testing yet. Feel free to compile and try?

Edited December 5, 2013 by Cilph

[AdmiralTigerclaw]

But it saves the player the hassle to delete and recreate (with changes he suddenly wants to implement) his whole sequence and cilph from programming something to reload and edit an uploaded sequence.

It doesn't, really.

See, you fail to realize that if you want to 'suddenly change the sequence', you do that BEFORE you upload the commands in the first place. Saying 'upload commands' is like clicking the 'NEXT' tab after configuring something to install on your computer. By the time you get there, you're certain that's what you want to do. Why would you start a command train in the first place if you're not sure what all you want it to do? That's not a problem with coding how you send it, that's an error between desk and chair.

Now, if I were to compromise. The easiest trick to fixing your scenario would be a simple 'recall last upload' button. In this case, in the event of what you're suggesting actually happening. Say, you realized just after upload that you told it to burn BEFORE the command to turn the vehicle retrograde for an orbital insertion. Everything in the last sequence you uploaded can be stored in a recall slot and be a button click away. Click it, and the sequence comes up in full. Do your editing, clear the old sequence, upload the new one. (Or even streamline that into an 'upload and replace' command.)

As for the amount of work Cilph has to do. That's really up to him how exactly he codes it and how much work he wants to do. I've programmed in C and C++ as well as programmed PLCs with ladder logic myself. And while setting this kind of code up may be more work than what you suggest, you also have to consider at the same time that over several iterations, hundreds of lines of code, and days of work in manhours, we may end up there ANYWAY. Because it's just better as a control scheme.

At that point you end up resting your head in your palm, elbows on the table going 'Why didn't I just do that in the first place?'

Kind of like how my PLC instructor about chewed my head off for something. I coded what I thought was an optimization into my PLC logic. A piece of logic in the main program ladder I set up to, what I thought, would bypass the program even looking at a large subroutine when it wasn't in use, saving runtime. (Hint: It still looks at the code, it just doesn't mess with it.)

However, when she looked at it, the first thing she asked, in no uncertain terms, was 'What's this? Why'd you do this? What's this doing.' After a half-explanation/half-argument with her (Her personality is one to jump right in before I'm finished explaining my thoughts), she explained to me that even though I bypass the logic, the PLC still scans through that code. I saved nothing and only added confusion and created a headache.

The reason it added confusion is because of something she kept trying to drill into our heads over and over. When writing PLC code, you need to make it as clear and simple as possible because even though you're writing it, it's other engineers WILL have to read it and work with it:

- "If you can't read it like 'that' *snaps her fingers*, it's worthless code. Nobody will want it. They don't WANT to have to EXAMINE your code. They don't WANT to HAVE to decode it. You are selling product to them here. You have to make it as easy and as possible. If you can't do that for them, someone else WILL, and the company will pay THEM to do it instead of YOU."

And the reason my code was 'wrong' was because I took the logic bypass out of the subroutine and put it on the main ladder. In doing so, I'm making the engineers who have to read my code confused because now they see a 'bypass to nowhere' on the main code, and the subroutine appears to be missing a critical bypass. And that forces them to waste time flipping between two code sections wondering what the hell is going on. They have to 'hunt for it'. And that's a headache. And that's why it is wrong.

I know that's a bit different from what we want to see done here, but the core lesson is the same. Are we saving work by shortcutting the code? Or would this just become a headache that results in a set of fixes that ultimately cost more time and effort than 'doing it the hard way' in the first place?

Either way, it's up to Cilph. We're not coding the mod, we're just making the suggestions.

[Cilph]

I haven't even *read* these queuing suggestions you're ranting about. My head is nowhere near wanting to work on that for a while. I've also stated before that anything not posted on the github issue tracker is not guaranteed to be read by me .

What I can say is that I have reworked the flight computer to be 'more like' kOS in that it can accept, delay and execute arbitrary commands much like an abstract CPU. Might make other approaches more possible than before.

[AdmiralTigerclaw]

I haven't even *read* these queuing suggestions you're ranting about. My head is nowhere near wanting to work on that for a while. I've also stated before that anything not posted on the github issue tracker is not guaranteed to be read by me .

What I can say is that I have reworked the flight computer to be 'more like' kOS in that it can accept, delay and execute arbitrary commands much like an abstract CPU. Might make other approaches more possible than before.

*Smacks you with a rolled up newspaper*

Bad coder! No biscuit! Not all of us know how to mess with github. See? You've already violated the 'make it easy' clause.

[sonicsst]

Don't be silly. If you can post on a forum, you can post a bug report on github.

Bad poster. Don't violate the dont-bite-the-hand-that-feeds-you clause. Be nice to plugin devs, they work for free.

Edited December 5, 2013 by sonicsst

[AdmiralTigerclaw]

Don't be silly. If you can post on a forum, you can post a bug report on github.

Me, silly?

What ever gave you that idea?

[jamis]

The flight computer seems to have a very hard (read: impossible) time trying to hold a heading that is near normal ("north" in my specific case). I've tried having it hold to a maneuver node, and I've tried having it hold to a specific pitch/heading/roll, and both times it just spins wildly trying to lock on that direction. Even if I prepoint the craft in the given direction, as soon as the flight computer gets hold if the instruction to hold that heading, it wobbles a bit, and then begins spinning like crazy.

The specific heading I'm trying to lock onto right now is pitch=1.9, heading=351.2, roll=0. (I've tried various roll parameters, since those have no real effect on the direction, but they all exhibit the same behavior.)

Is this a known issue? Is it something I'm doing wrong?

[AdmiralTigerclaw]

The flight computer seems to have a very hard (read: impossible) time trying to hold a heading that is near normal ("north" in my specific case). I've tried having it hold to a maneuver node, and I've tried having it hold to a specific pitch/heading/roll, and both times it just spins wildly trying to lock on that direction. Even if I prepoint the craft in the given direction, as soon as the flight computer gets hold if the instruction to hold that heading, it wobbles a bit, and then begins spinning like crazy.

The specific heading I'm trying to lock onto right now is pitch=1.9, heading=351.2, roll=0. (I've tried various roll parameters, since those have no real effect on the direction, but they all exhibit the same behavior.)

Is this a known issue? Is it something I'm doing wrong?

Might be prudent to know more about your flight conditions.

Heading tends to be a surface-related orientation value, and changes rapidly when you're screaming along in orbit. (In example: 90 deg or Due East only really sticks when you're in a perfect flat equatoral orbit. Any other time, the orientation is going to shift from one moment to the next.)

The change in position in orbit as you move along may change what exactly it considers 'heading 351.2'. Especially if you have a highly inclined orbit. The way you describe it, it sounds like it can't actually decide what heading 351.2 is from one moment to the next.

It is also prudent to know on the computer's side if holding heading 351.2 is to hold orientation to that heading, or if that's to try and force the craft onto that heading track. (IoW: The difference between pointing at it with a space probe, and making a plane actively fly in that direction.) If it is trying to put your vehicle on that TRACK, that would explain it going cockney and spinning wildly. It's like flailing about in a pool when you can't swim.

[BlazingAngel665]

Suggestion: KAS let's you pick up antennas and even deploy it. It would be nice if the game would recognize this and allow you to transmit data. I experimented with this and it said "No antenna on this vessel can transmit" I had a clear LOS to the space center.

[KerbMav]

Suggestion: KAS let's you pick up antennas and even deploy it. It would be nice if the game would recognize this and allow you to transmit data. I experimented with this and it said "No antenna on this vessel can transmit" I had a clear LOS to the space center.

Tried to quick-save-load?

Still wondering how to equip a Kerbal with a short range omni-antenna ...

[Kerbrud]

You need a large stack probe on a 6+ crew ship, it then becomes a command center and you can fly anything from there instead of Mission Control. Science still needs a link to Mission Control though

"Large stack probe" - do you mean the biggest dish, or do you mean the "RC-L01 Remote Guidance Unit" ? Cause if it's the biggest dish, I need to change some cfg's.. The biggest dish is way too big for my ship.

[jrandom]

"Large stack probe" - do you mean the biggest dish, or do you mean the "RC-L01 Remote Guidance Unit" ? Cause if it's the biggest dish, I need to change some cfg's.. The biggest dish is way too big for my ship.

It's the RC-L01 unit -- the biggest stock probe core.

[jamis]

Might be prudent to know more about your flight conditions.

It must have been a fluke--I tried it with another (similar) ship, and it worked just fine. The flight conditions, though, were this: the ship had just finished a transfer burn from low Kerbin orbit toward Minmus. I have a network of satellites in Kerbin orbit, and another in Minmus orbit, and they talk to each other just fine, but in between--nothing. So I was trying to set up a "fine tune closest approach" maneuver on the ship heading for Minmus. To do this, I created a maneuver node, and tried to get the ship to point in the direction indicated by the maneuver node--but it would just spin wildly, very odd.

As I said, though, it worked just fine with a second ship, so I'm going to chalk it up to gremlins. If it happens again I'll try to get more info.

[BigD145]

Suggestion: KAS let's you pick up antennas and even deploy it. It would be nice if the game would recognize this and allow you to transmit data. I experimented with this and it said "No antenna on this vessel can transmit" I had a clear LOS to the space center.

KAS usually requires a save load or leave and refresh physics.

[Hodo]

You need a large stack probe on a 6+ crew ship, it then becomes a command center and you can fly anything from there instead of Mission Control. Science still needs a link to Mission Control though

I think it is actually greater than 6, or >6. Because I have had 6 Kerbals on a ship and it made no difference. But as soon as I added a 7th it worked like a charm.

Do Kerbals in a storage can count towards the six, or do they need to be in an actual command pod?

Yes, they can be in a hitch hiker pod.

[drtedastro]

is there anyplace we can 'adjust' the 6+, ok 7, down to a smaller number...? is this in a config or is it hard coded somewhere....???

[AdmiralTigerclaw]

It's six and a large probe core. (Inline)

Telltale sign that it is indeed working is that you can go to map mode and see the red dot on that new ship.

[gre8]

One question. When delay finally becomes a thing, will there be a disable delay button or we will just have to deal with it?

I'm asking because this is a great mod, but having to deal with signal delay and a scriptable auto pilot systems is a step beyond my (and probably others') interests in this game.

Not a criticism. This is an amazing mod and has already added a lot to my KSP experience; I'm just asking so I don't have to uninstall it later down the line.

[sonicsst]

Delay is already a thing. It can be disabled by changing EnableSignalDelay = True to False in RemoteTech_Settings.cfg

[gre8]

Whoa, I didn't know that. When was the delay thing updated?

[SergeantBlueforce]

Just to show of my insecurity: I hope you do not mean MY syntax?

Open the RemoteTech_Squad_Antennas.cfg file, copy an entry - you might want to pick a section that belongs to a comparable antenna/dish, meaning antenna or dish, deployable or not ... - and paste it into a new cfg file for your mod.

For example:

@PART[longAntenna]
{
	!MODULE[ModuleDataTransmitter] {}

	@MODULE[ModuleAnimateGeneric]
	{
		allowManualControl = false
	}

	MODULE
	{
		name = ModuleRTAntenna
		Mode0OmniRange = 0
        Mode1OmniRange = 2500000
		MaxQ = 6000
		EnergyCost = 0.13

		DeployFxModules = 0

		TRANSMITTER
		{
			PacketInterval = 0.3
			PacketSize = 2
			PacketResourceCost = 15.0
		}
	}

	MODULE
	{
		name = ModuleSPUPassive
	}
}

Then - in this example - change "longAntenna" to name of the modded part.

PART
{
name = [I]this one here, mind capiTilZation[/I]

Ok, but do I create an entirely new config file for the mod antennas or do I use of the existing RT configs, like the squad one or something.

[Frederf]

I am creating a com probe which purpose is to land on Duna and provide an area where I can aim and land a full science rover, analyze the density of the atmosphere, and provide communications back to Kerbin.

Problem:

(Deadly Re-entry mod, 1.25m heatshield on probe.)

While testing in Kerbin, my communotron 32 antenna is breaking under the stress on re-entry, which means that I somehow need to deactivate it during reentry to then reactivate it.

How do I get around this issue?

In my case I genuinely programmed a sequence that involved the antenna raising and lowering such that it had a command to lower the antenna, all the landing instructions, and then one to raise it minutes later and then the thing lands itself and resumes contact. You want the "Toggle" action name in an action group so you can provide both the raise and lower commands as ALT+(#) before the whole sequence even starts.

[KerbMav]

Ok, but do I create an entirely new config file for the mod antennas or do I use of the existing RT configs, like the squad one or something.

I strongly recommend a new file, you can put this into your own folder under gamedata - together with anything else you might change in the future - so if RT2 updates you do not loose your changes to the other mod.

[SergeantBlueforce]

I strongly recommend a new file, you can put this into your own folder under gamedata - together with anything else you might change in the future - so if RT2 updates you do not loose your changes to the other mod.

Would it possible to give me an example?