[1.4.1] Kerbal Construction Time 1.4.0.69 (2018-03-24) - Unrapid Planned Assembly

[Bersia]

So, I was getting this weird game breaking glitch where on a new launch I'd fly across the world and have no speed, money, or science. I thought it was maybe KAC since the time was also 99years into the future and KCT stated infinity on everything. I got rid of KAC and re-installed KCT and then on my first flight I got a CTD. I think I got my logs properly uploaded. Can anyone tell me if it is a KCT issue or something else?

Weird glitch of 000000: https://www.dropbox.com/s/85pu9avr0h4nr0r/20160506193200_1.jpg?dl=0 (I removed KCT and reinstalled it, plus I got the same glitch after removing it completely last night. Minus the popup of course.)
Glitch with KCT popup log: https://www.dropbox.com/s/rhbvl2ntiar6vm3/OLD_KCT_Popup_output_log.txt?dl=0 

Newest CTD Log: https://www.dropbox.com/s/c7g6uawynxcfccn/output_log.txt?dl=0

[magico13]

20 hours ago, tacobowler said:

Forgive me if this is answered somewhere, but I can't find it for the life of me.  I am assuming that i can use my upgrades in this mod to speed up the time it takes to research new technology (i.e. 150 days for Landing), but using anything on the R&D doesn't seem to be making any noticeable difference.  Am I missing something or is there just no way to substantially increase research time?

Add points into the second R&D rate which should be called Development. The Getting Started Guide goes over all the different upgrades. Research gives you science for building ships (the science you get from performing hands-on research work with the parts) and development is for developing new technologies (unlocking nodes).

18 hours ago, Bersia said:

So, I was getting this weird game breaking glitch where on a new launch I'd fly across the world and have no speed, money, or science. I thought it was maybe KAC since the time was also 99years into the future and KCT stated infinity on everything. I got rid of KAC and re-installed KCT and then on my first flight I got a CTD. I think I got my logs properly uploaded. Can anyone tell me if it is a KCT issue or something else?

Your second log is the same as your first one. The CTD appears to be fairly random. I've had CTDs with nearly Stock versions of 1.1 64bit and I have barely played it. I think 1.1 is just kind of inherently buggy.

I can't really help with the first one without a log from it. Sounds like a mod got bugged when the scene was loading. I noticed several errors logged in the second log, like DangerAlerts not being able to find its icon and Vanguard Technologies not being built for this version of KSP and thus not loading. DecayManager has a bunch of errors when the flight scene starts loading, but they go away after it gets a little further along.

With as many mods as you have, though, there are a lot of possible sources of issues.

[Andem]

FYI, the experimental version can eat kerbals, they get listed as dead in the save file after you put them aboard for launch.

[magico13]

12 minutes ago, Andem said:

FYI, the experimental version can eat kerbals, they get listed as dead in the save file after you put them aboard for launch.

I'll need logs for that one. Haven't seen that myself and noone else has reported it.

[Torgo]

Something I've noticed in the new version: The KCT Build List window doesn't stay active on the KSC screen like it used to in 1.05. Every time you go back to the screen, it needs to be re-opened. It does pop up if you hover your cursor over the icon on the KSP screen, but it may or may not stay active without a mouse click on the icon. I've experienced both. 

I tested this in a clean install made from a copy of KSP from my Steam app folder, and only the manual install of KCT is done. This is the 64 bit version of 1.1.2.1260

I can go into the various sections of KCT (VAB, SPH, Tech, etc) , and it remains active after closing them. I am not seeing any errors, and the program appears to be functioning perfectly. I did notice some weirdness in the log, though. Here's a link to it. 

I also have KCT installed in a very heavily modded install, and outside of the needing to re-open the window, it's running great. 

To duplicate -

• fresh install of KSP 1.1.2.1260
• install KCT 1.1.3-pre4 from zip
• start KSP, start new game (happens with default or modified career settings)
• go through initial KCT setup (happens with default or modified settings), spend upgrade points
• At main KSP screen, click KCT icon to open Build List
• Build list opens as it should
• Enter any building (either via icon or clicking on building)
• Exit that building (performing actions or not performing actions has no affect on outcome)
• Back at main KSP screen, KCT Build List window is not active.

This isn't the end of the world, just a couple of annoying extra mouse clicks and movements I would rather not make. 

[Andem]

27 minutes ago, magico13 said:

I'll need logs for that one. Haven't seen that myself and noone else has reported it.

Well, that's a problem... I don't have the logs. The kraken has been eating them ever since I updated to 1.1.2. Have reinstalled twice. Stock and Modded.

[magico13]

8 minutes ago, Torgo said:

Something I've noticed in the new version: The KCT Build List window doesn't stay active on the KSC screen like it used to in 1.05. Every time you go back to the screen, it needs to be re-opened. It does pop up if you hover your cursor over the icon on the KSP screen, but it may or may not stay active without a mouse click on the icon. I've experienced both. 

This isn't the end of the world, just a couple of annoying extra mouse clicks and movements I would rather not make. 

I know why this started, but haven't had success getting it to stop, which is one of the main reasons I haven't done the release yet. Basically I added some code to close the GUI when you enter the R&D center, Astronaut Complex, and the other non-scene change buildings, but then that messed up the button state as it thought it was active when it was hidden. So instead I made it also turn the button off. Now for whatever reason it's stuck in a weird state where if you hover over the button it brings the window up and when moving off the button it stays open (when it should close). A single click closes the window when it should be keeping it open. I've looked through the code multiple times and can't figure out how that's happening, so I need to sit down and mess around with it for a while until I get something that works correctly but I haven't had enough free time to do that yet (and with the final for the class I TA being Monday morning, I keep getting nonstop questions via email that I have to answer and I'm going to lose most of tomorrow to office hours and other obligations).

Basically, it's a problem I know about and it's half the reason I haven't done the release yet.

5 minutes ago, Andem said:

Well, that's a problem... I don't have the logs. The kraken has been eating them ever since I updated to 1.1.2. Have reinstalled twice. Stock and Modded.

I haven't heard of logs being deleted before. They should be in Kerbal Space Program\KSP_Data\output_log.txt or Kerbal Space Program\KSP_x64_Data\output_log.txt depending on if you're running 32 bit or 64 bit (assuming you're on windows, if you aren't then check this post)

[autumnalequinox]

3 hours ago, magico13 said:

Stock versions of 1.1 64bit and I have barely played it. I think 1.1 is just kind of inherently buggy.

 

I too am discovering this.  It reminds me a bit of the older x64 Kerbal only with less frequent crashes.  Most crashes are straight-to-desktop with log files revealing nothing..  the game humming along, no unity errors, and then "crash!".  I have a feeling it may be a few updates before we get semi-stability.  

Note:  I've heard people say it's problems with Unity 5 but I play another game that was recently ported to U5 from 4 that definitely uses x64 and is very stable (a physics heavy voxel type game).

[NathanKell]

@magico13 I have had 0 problems with the test build, short of the minor window annoyance (which IMO should not prevent a release, we'll deal ).

[magico13]

1 hour ago, NathanKell said:

@magico13 I have had 0 problems with the test build, short of the minor window annoyance (which IMO should not prevent a release, we'll deal ).

It was only part of the reason. The other was finding the time and getting at least one useful thing in. In this case, the useful thing is that you can now modify the effective cost formula on a per part or per module basis. So if you want everything with ModuleCommand to take longer then you can do that, or if a specific part is weirdly priced and doesn't fit the formula well then you can fix it.

 

Edit: Hopefully final pre-release now available. Want to get at least one external test of the Part Variables functionality before release. Some more explanation of that feature here.

Edited May 8, 2016 by magico13

[NathanKell]

@magico13 woot! I'll give it a try tomorrow!

[Squiggsy]

hello! Thanks for your update to 1.1, unfortunately upon starting up KSP for the first time since adding KCT I got an error message telling me something had gone wrong and the game was likely to be unplayable, the output log is below:

Output Log

[magico13]

11 minutes ago, Squiggsy said:

hello! Thanks for your update to 1.1, unfortunately upon starting up KSP for the first time since adding KCT I got an error message telling me something had gone wrong and the game was likely to be unplayable, the output log is below:

Output Log

What parts of Real Solar System are you running? Is it a full RO game? It looks like a full RO + RP-0 game (is RP-0 updated to 1.1 already?). Seems to be an issue related to KSCSwitcher and it's totally possible KCT isn't compatible with the latest KSCSwitcher yet. It's definitely a problem on KCT's side, not KSCSwitcher's. I'll try to get it fixed tonight.

Edited May 8, 2016 by magico13

[Squiggsy]

3 minutes ago, magico13 said:

What parts of Real Solar System are you running? Is it a full RO game? It looks like a full RO + RP-0 game (is RP-0 updated to 1.1 already?). Seems to be an issue related to KSCSwitcher and it's totally possible KCT isn't compatible with the latest KSCSwitcher yet.

Oh god thats embarrasing I forgot I was testing RP-0 out in 1.1 since it is almost entirely MM Configs. so yes it was a full RO+ RP-0 game. I just installed the development build from Github and replaced the other version for it and didn't get any of the complaints, RP-0 seems mostly working as well but I got a crash to desktop after a couple of minutes before I got to test things out properly, so there are certainly some stability issues.

[magico13]

8 minutes ago, Squiggsy said:

Oh god thats embarrasing I forgot I was testing RP-0 out in 1.1 since it is almost entirely MM Configs. so yes it was a full RO+ RP-0 game. I just installed the development build from Github and replaced the other version for it and didn't get any of the complaints, RP-0 seems mostly working as well but I got a crash to desktop after a couple of minutes before I got to test things out properly, so there are certainly some stability issues.

I think something changed in KSCSwitcher ever so slightly (my guess is that a previously static variable is no longer static) so KCT's attempts to integrate with it are getting messed up. The fix should be just to figure out what changed and change my code appropriately, but depending on how much changed it could take a while. I won't get a chance to take a look at it for another 3 to 4 hours though.

[Squiggsy]

40 minutes ago, magico13 said:

I think something changed in KSCSwitcher ever so slightly (my guess is that a previously static variable is no longer static) so KCT's attempts to integrate with it are getting messed up. The fix should be just to figure out what changed and change my code appropriately, but depending on how much changed it could take a while. I won't get a chance to take a look at it for another 3 to 4 hours though.

thanks for all your hard work! Yes having removed KSCSwitcher did indeed fix the issue to that is where the problem lies.

On a side note, apart from every single part claiming to not be supported by RP-0, it actually seems to work perfectly...

[TrooperCooper]

 

Just installed the new RO with RSS and other mods as well. Tried KCT of course and ran into this issue that broke it:

 

Exception loading ScenarioModule KerbalConstructionTimeData: System.IO.FileNotFoundException: Could not load file or assembly 'MagiCore, Version=1.0.0.0, Culture=neutral, PublicKeyToken=null' or one of its dependencies.

File name: 'MagiCore, Version=1.0.0.0, Culture=neutral, PublicKeyToken=null'

  at KerbalConstructionTime.KCT_DataStorage.OnDecodeFromConfigNode () [0x00000] in <filename unknown>:0

  at KerbalConstructionTime.ConfigNodeStorage.IPersistenceLoad.PersistenceLoad () [0x00000] in <filename unknown>:0

  at ConfigNode.LoadObjectFromConfig (System.Object obj, .ConfigNode node, Int32 pass, Boolean removeAfterUse) [0x00000] in <filename unknown>:0

  at ConfigNode.LoadObjectFromConfig (System.Object obj, .ConfigNode node) [0x00000] in <filename unknown>:0

  at KerbalConstructionTime.KerbalConstructionTimeData.OnLoad (.ConfigNode node) [0x00000] in <filename unknown>:0

  at ScenarioModule.Load (.ConfigNode node) [0x00000] in <filename unknown>:0

  at ScenarioRunner.AddModule (.ConfigNode node) [0x00000] in <filename unknown>:0

 

 

[NathanKell]

@magico13 the current KSCSwitcher repo is https://github.com/KSP-RO/KSCSwitcher/

@regex was kind enough to do the 1.1 update; I'm not super familiar with what had to change.

 

I am trying the part thingie now, I'll let you know how that goes. I do have a question about module resetting--is that accessible to MM in any way? Also, and I apologize because I should remember this, but are presets MM-accessible? Ideally I'd generate the part modifiers by MM...

[regex]

ohgodimessedup

Do we have an error log?  I'll check into it, I can't remember exactly what I had to do to make it work.

Just saw the log file, I'll try to address that soon, thanks.

Edited May 9, 2016 by regex

[magico13]

3 hours ago, TrooperCooper said:

 

Just installed the new RO with RSS and other mods as well. Tried KCT of course and ran into this issue that broke it:

 

Exception loading ScenarioModule KerbalConstructionTimeData: System.IO.FileNotFoundException: Could not load file or assembly 'MagiCore, Version=1.0.0.0, Culture=neutral, PublicKeyToken=null' or one of its dependencies.

File name: 'MagiCore, Version=1.0.0.0, Culture=neutral, PublicKeyToken=null'

  at KerbalConstructionTime.KCT_DataStorage.OnDecodeFromConfigNode () [0x00000] in <filename unknown>:0

  at KerbalConstructionTime.ConfigNodeStorage.IPersistenceLoad.PersistenceLoad () [0x00000] in <filename unknown>:0

  at ConfigNode.LoadObjectFromConfig (System.Object obj, .ConfigNode node, Int32 pass, Boolean removeAfterUse) [0x00000] in <filename unknown>:0

  at ConfigNode.LoadObjectFromConfig (System.Object obj, .ConfigNode node) [0x00000] in <filename unknown>:0

  at KerbalConstructionTime.KerbalConstructionTimeData.OnLoad (.ConfigNode node) [0x00000] in <filename unknown>:0

  at ScenarioModule.Load (.ConfigNode node) [0x00000] in <filename unknown>:0

  at ScenarioRunner.AddModule (.ConfigNode node) [0x00000] in <filename unknown>:0

 

 

You need to also install MagiCore. It should already be in the zip file. It goes directly in GameData. Alternatively you can grab it off of CKAN.

1 minute ago, NathanKell said:

@magico13 the current KSCSwitcher repo is https://github.com/KSP-RO/KSCSwitcher/

@regex was kind enough to do the 1.1 update; I'm not super familiar with what had to change.

 

I am trying the part thingie now, I'll let you know how that goes. I do have a question about module resetting--is that accessible to MM in any way? Also, and I apologize because I should remember this, but are presets MM-accessible? Ideally I'd generate the part modifiers by MM...

Currently nothing is MM capable. I don't think it would be too hard to make it MM capable but I haven't actually taken an in-depth look at MM enough to know how easy it would be.

It makes sense to at least make the module resetting MM compatible. I never thought about mod authors including their own resets.

As I mentioned before, I haven't done much with MM, but I am curious as to how creating the part modifiers with MM would be different than just putting them into the RP-0 Preset directly (not trying to be condescending or anything like that, just curious if MM has some useful feature for making that easier, other than the replacement feature)

[NathanKell]

@magico13 Ah, hmm, I'm not sure if MM supports using variables for key names rather than key values. But consider this workflow in RO:

1. We create a general engine config file. It has the stats for that engine (thrust, propellants, reliability info, etc).

2. That config is applied all parts that have engineType = [that type]

If your stuff were MM accessible, we could put the modifier code in the global engine config, and it could place the partname (for every part to which MM applies it) in your modifier.

The use case here is solids: They should be much quicker per dollar. So in each of our global solid configs we could set up a multiplier of 0.5. Then each part to which RO applies that engine config would register something in your node.

 

Do recall that MM is Turing-complete now, so we can do some crazy stuff

[magico13]

9 minutes ago, regex said:

ohgodimessedup

Do we have an error log?  I'll check into it, I can't remember exactly what I had to do to make it work.

Just saw the log file, I'll try to address that soon, thanks.

I've got to make some changes anyway to use the actual site name rather than the internal one, so don't fret too much. It'll have to wait until tomorrow though on my end.

2 minutes ago, NathanKell said:

@magico13 Ah, hmm, I'm not sure if MM supports using variables for key names rather than key values. But consider this workflow in RO:

1. We create a general engine config file. It has the stats for that engine (thrust, propellants, reliability info, etc).

2. That config is applied all parts that have engineType = [that type]

If your stuff were MM accessible, we could put the modifier code in the global engine config, and it could place the partname (for every part to which MM applies it) in your modifier.

The use case here is solids: They should be much quicker per dollar. So in each of our global solid configs we could set up a multiplier of 0.5. Then each part to which RO applies that engine config would register something in your node.

 

Do recall that MM is Turing-complete now, so we can do some crazy stuff

That's the kind of thing that I imagined MM would have. I'll look into making Presets MM capable. As a possible workaround can you make a custom module with MM that doesn't do anything, then use the Module Variables? I'm not sure if MM can create "empty" modules or just apply existing ones. KSP probably wouldn't be too happy about a nonexistent module.

I don't want to push this release back any more (1.1 has been out for what, almost 3 weeks now). But I'll try to get something added this week if I get the chance. What's the tentative plan for RP-0 release?

[NathanKell]

@magico13 don't hold on our account; as soon as KCT goes up, RP-0 does (a simple port of 1.0.5 RP-0 so people can continue saves).

[NathanKell]

@magico13 for the record while I'm happy to test the multipliers, I will not be holding RP-0 back so as to implement them--I want to get it up so people can continue their 1.0.5 careers, and then make use of that feature, as well as launch costs, in a save-breaking rework of our entire pricing model.

[magico13]

3 minutes ago, NathanKell said:

@magico13 for the record while I'm happy to test the multipliers, I will not be holding RP-0 back so as to implement them--I want to get it up so people can continue their 1.0.5 careers, and then make use of that feature, as well as launch costs, in a save-breaking rework of our entire pricing model.

I want to make sure the bugs with KSCSwitcher are worked out on my side and I have to take some time to rewrite my .netkan file because of MagiCore. I can't do all that tonight since I have to be at work early tomorrow to proctor an exam, so the release will happen tomorrow evening. Sound good?