magico13

Glad to hear it! Let me know if you've got any questions or run into any trouble!

Thank you for that! I know what the issue is now, you don't have KSP through steam so when it sets the default directory to the steam directory it fails. I can reproduce it on my end quite easily now. I'll get you an update asap. Edit: Alright, this one should work properly now! http://www./download/y6z4s49bswd121p/DataSwitcher-0.2.18.jar

Yep, as futrtrubl mentioned you just left click the StageRecovery icon in the toolbar. You'll get a message in the notification system (like you do when things are recovered or you complete a contract) which will give you some info about the current ship in the editor. That's all there is within the editor at the moment, no actual interactive GUI.

The development builds have had a lot of bug fixes and a few new features, so they're worth checking out. I'm planning on making it the new release version when I get a chance to do some proper testing myself. You'll be glad to know that the latest build should fix the issue with having to press rollout and then launch even when rollout is disabled, but I didn't test that build in-game. I also added the reconditioning formula to the KCT_Formulas.cfg file. I might make a tweak to that before release though to add one more variable (the maximum reconditioning value as defined in the settings) and have it use min(formula, max setting). Edit: I made that tweak. I need to write up some documentation on how to modify those formulas and what variables are exposed. Edit: And the first draft of that documentation is up on the Github page: https://github.com/magico13/KCT/wiki/KCT_Formula.cfg-and-Math-Parser-Documentation

Was this with the development builds or the official release? I thought I had fixed most/all the "out of sync" issues in the dev versions. Initial build times are high, but start to level out/drop substantially once you start reusing and recovering parts. The inventory drops things quite a bit, but you're also rewarded for using identical parts throughout builds. A large majority of the mod is configurable. There are three build time modifiers: the overall modifier which is applied linearly to the build time calculations, the "build effect" which handles how subsequent builds with the same parts get faster (by default, the fourth launch of an identical rocket will take half the time), and the "inventory effect" which handles how much things decrease when you use parts from the inventory. The recent development builds actually let you completely redefine the entire build time in the KCT_Formulas.cfg file if you want more control, including letting you switch from a cost based time to a mass based one if you want (or a mix if you're feeling adventurous). I should also add the reconditioning and rollout time formulas to that file, thanks for reminding me. I'll fix the issue where turning rollout down to 0 but keeping reconditioning on requires you to press rollout and then launch, thanks!

Hmm, there's only one input lock and it doesn't look like it's actually doing anything. And usually the buttons get greyed out when a lock is active. I guess I have more questions than answers now 1) Is it only after reverting from a simulation? What if you launch a ship, then revert to editor? 2) What happens when you click the buttons? Does the texture change at all? Do the tooltips pop up when you hover over the buttons? (ie, "save" "load" "launch", etc) 3) If it does happen after normal launches too, when you remove KCT does it still happen? Also, I don't see the KCT icon anywhere in that screenshot. Are you using the toolbar mod with it set to show up on there? (KCT is obviously installed and running though). I am currently somewhat baffled by what is happening. I don't think it's directly KCT related but you've got so many mods it's hard to diagnose!

When I get home I'll try with a similar setup. I think my Windows 7 system uses Java 7, but my Windows 8 uses Java 8. It works fine on both for me. The log.txt should be in the DataSwitcher folder that gets made when you start DataSwitcher. If that's not being made, you might have antivirus software that's blocking it? When you click the "+" button, the window doesn't appear, correct? Or is it that when you press the "Create" button, nothing happens? You could try running it through the command line to see if it outputs anything there. Just hold shift and right click on the folder where you have the DataSwitcher jar file. Select the "Open Command Window Here" or similar action. A black box will appear. Just type (without quotes) "java -jar DataSwitcher-0.2.17.jar" (using whatever the file is named, which would be DataSwitcher-java8.jar for the new file). It should open the main window like normal, and you might see some text printed to the command window when you try to create a new instance. If you can take a screenshot of that text (assuming there is some) that would be great!

That is an impressive mod list! It's hard to say for sure what's happening. There are a lot of errors with FAR after you loaded a ship in the editor. When you revert back to the editor there are some errors with Kerbal Engineer, the PQS system (particularly with Outer Planets it seems), some more with FAR, Procedural Wings. Action groups extended seems to have it's usual round of NREs. No input locks were active according to the log when entering the editor (and that's what would cause that issue). Some mods that I know use input locks in the editor: KCT, Kerbal Engineer, MechJeb, I think FAR does too. Anything that has a GUI in the editor probably uses input locks there. I would suggest opening the debug menu (Alt-f12 usually) and clicking "Show Input Locks" then sending me a picture of that. It should say whether any locks are active, and if so what they're named. If we're lucky, the mod author named their Input Locks with something descriptive (all of KCT's start with "KCT" and then have semi-descriptive names)

Sure! I didn't realize anyone other than me ever used this. I should be able to do that pretty easily. I should take some time to update this a bit. I've thought of a few ways to make it a bit better. Edit: So it seems to work fine for me with Java 8, but just to be sure I opened it up in eclipse using Java 8 and rebuilt the jar. I'll link to that below. It's otherwise identical to the one in the OP. If you're still having issues, please let me know what operating system you're using (I had issues with it in the past on Linux) and maybe send me the log.txt file since it might contain an error message. http://www./download/6zb64hzdk5z6mlu/DataSwitcher.jar

That's what I want to do, either as part of this or as a separate mod. There used to be a mod called TTNeverUnload that would extend the physics range for specific craft using a part, so it should be possible to do purely through a plugin. A quick press of a button to "flag" the ship to stay loaded, and another quick press to "unflag" it and put it back to the normal physics range. I just haven't had time to play around with the code for it yet. For StageRecovery I'd want to set it up to be automatic. Anything under sufficient parachute would stay loaded (optional config settings obviously), everything else would unload (including powered landers, since I'm not going to also write a full autopilot mod that could physically land any ship). I'd also provide the option for people to flag a ship to stay loaded (if they were using mechjeb for instance). But it all comes down to time and priorities. Unfortunately school + work take the number one slot for both of those, followed by KCT, followed by SR. Conceivably the simple "press button to not unload" mod would be faster to write, so maybe I'll play around with that this weekend.

Can't KSP TOT use RSS' settings for the planets? Then you could really easily calculate gravity assists. I don't follow it very closely, so I'm not 100% sure it works with RSS.

Unfortunately, if StageRecovery is recovering them then KSP has already deleted them from the game. Disabling StageRecovery will just make them get deleted without you getting any of the resources back. If that issue wasn't present in KSP, StageRecovery wouldn't need to exist You have a few options though. The first is to make sure to stay within the physics range of the pod you're dropping (stay within 2km to be safe, physics range is about 2.3km). Just circle your plane around until the thing lands. The second is to use a mod that extends the physics loading range (and thus can cause more lag by causing more parts/ships to be loaded at once). BDArmory doubles it to roughly 5km, and the Lazor system can extend it to very high values (I think up to 100km). The last option is to use a mod called Flight Manager for Reusable Stages (FMRS). It uses save file trickery to let you manually control everything you decouple, then merges all the saves together into one. It's my intention to eventually find a way around this issue, either by letting you declare certain vessels to not unload or by physically landing vessels before KSP deletes them. Kerbal Construction Time has been my main development focus lately, but I intend on spending some time on StageRecovery after I release the next KCT version.

A, sort of B, B. For clarification: 1) I typically let my kOS script handle launches, and it actually rotates and then pitches, so that would actually be B, but I manually do A. 2) I get pretty low before landing, kill off horizontal, then the last <500 meters is entirely vertical 3) I have never once used docking mode, but I also don't typically dock medium or smaller ships using RCS at all.

I play with a lot of mods in 0.90 and the only real issues I have are crashes from hitting the memory limit after a few hours of gameplay. There's definitely a serious memory leak somewhere, cause loading a save I'm at 2 gigs and 2 hours later I rarely go below about 3 gigs. While 0.90 is pretty stable for me, I am afraid of bugs in 1.0. It's not because of current experiences with KSP, it's because I've seen them have pretty bad bugs on releases with substantially fewer new features. They're trying to do so much, and I'm sure it'll be really awesome when it works right 85% of the time, but that 15% that they didn't encounter in the limited testing is what scares me, and is why I wish there was one more "beta" release where they could have all those features and let us bug test, and then have 1.0 as the ultimate bug free release version.

You can, btw, just go into the "difficulty settings" before you start the save and add in some extra funds and/or science. I've been playing Career with a Sandbox-esque feel to it. I started out with some extra science to unlock the early nodes and only take contracts that fit with what I want to do. And then I only take contracts to be able to fund my side missions. I had an Eeloo transfer window coming up but didn't have the funds to upgrade the R&D center to the level where I could get an antenna that could reach Eeloo. Nor was there a contract for it. But I sent a probe anyway and once it leaves comms range it'll be controlled by a kOS script. That was funded by some random contracts. My multimillion funds Duna Station paid for itself and then some, despite not having a single contract for a Duna station (just the Explore Duna and Explore Ike contracts). I managed to complete contracts for a Kerbin station, Minmus base, and Solar station all with the Duna Station (yes, I landed a massive OKS station on Minmus). I enjoy the freedom to do whatever I want, but like the artificial limitations that make me have to get a bit more creative than I otherwise would be. I'd definitely still have fun in Sandbox mode (it was the only thing in existence when I started playing and it was plenty of fun+challenge back then). I'd just set my own limitations (with the aid of mods) rather than what the Stock game says I should be limited to. The two modes are really not much different to me. I don't think of Sandbox mode as "do whatever you want in God mode", I think of it as "do whatever you want but it's still going to require patience, planning, and skill".

Not all the features work 100% as intended (since they depend on science or funds), but everything is designed to be functional in Sandbox mode as well as Career/Science. I typically do my tests in Career mode, so there may be some things that don't behave correctly in Sandbox mode. If you encounter anything, let me know and I can fix it! Build times will still be applied and simulations around other celestial bodies will still require you to visit there first. The upgrade system depends on the tech tree for it's progression, so in Sandbox mode you are just given 45 upgrades at the start. You can increase or decrease that number at any time in the KCT settings (you could, for instance, start out at the 15 that career mode starts with, then give yourself more when you complete personal challenges). The 45 you get is about equal to what you can get with the Tier 2 R&D center (so not quite a complete Tech Tree, but pretty far along). Also, at any time you can enable or disable features as your playstyle warrants

Yeah, but "46.4.57.19:80 - havok.blizzy.de:http" looks a lot like Blizzy's Toolbar

Those were actually all plans for it originally. Maybe I'll get at least the parser to let you enter a date in normal time rather than seconds. I obviously never got around to finishing that feature. Edit: Aaaannnnnd done! Well, somewhat. You can now enter a time in "colon formatted" format, so yy:dd:hh:mm:ss. You don't have to use higher terms (like years), but must always go down to seconds. So 10:3:16:23 would be day 10, hour 3, minute 16, second 23. You can also now set an "absolute time" like before, or can use time relative to the current time (by clicking the "From Current UT" button). Both ways account for Kerbin days/years vs Earth days/years. Absolute time uses the KSP convention of starting at Year 1, Day 1. So if you put 2:3:0:0:0 it'll go to Year 2, day 3, rather than Year 3, day 4. Relative time doesn't need to worry about that problem, so if you put in 2 years, it'll advance time by two years, which is year 3 if current UT is 0 (a new game). It doesn't automatically fill in relative time based on when the ship would finish, but you can always do that yourself. It's also backwards compatible with directly putting the UT in as seconds.

I personally make a backup of my save folder every time I stop playing (which is once a week when I live stream) and only keep 1 or 2 backups at a time. I should just download/write a mod for it

I'm still making a few changes every now and then. Mostly small graphical fixes or improvements to some existing systems. Recently there were tweakscale related issues that I fixed and issues with the Upgrades window being out of sync. I've gotta do some testing on the system I just put in last night for resetting modules when recovering to storage. I moved most of the settings for it out into a config file so that users can add more modules (presumably from mods) that absolutely need to be reset, along with the parameters they should have reset. So things like ModuleEngines and ModuleEnginesFX or ModuleParachute are in there by default to turn off engines and repack chutes, but there isn't anything for RealChutes for instance and people might want that to be reset. A lot easier to change a config file than to have a new release after every new mod gets added. The only hardcoded modules are ModuleScienceData and Log. The first gets totally removed (it's literally science data, which gets recovered and so needs to be removed) and the second gets all of the values removed (that's the trip logger I think). Adding the RealChutes module should be as easy as adding something like the following (but with the right parameters) to the KCT_ModuleTemplates.cfg MODULE { name = ModuleRealChute state = STOWED } I also might play around with the math parser some more. If I change from a character by character parser to a block by block one (with each block being separated by an operator, such as "+") I can probably speed things up. There's also a limitation that nested min(x,y) and max(x,y) functions must be placed in the second slot [the "y" slot, ie min(x, max(a,), not min(max(a,,y)] which I can probably fix by checking the current "depth" when the comma is encountered. The current method splits the string by commas, takes the first split as "x" and everything else as "y". Checking the depth would split it into exactly two parts at the appropriate place, meaning you could nest anything you wanted in any slot as many times as you like. By depth, I mean "how many parenthesis are there" where "(" increases the depth by 1 and ")" decreases the depth by 1. So, starting at depth=0, the first "(" in min(max(a,,y) sets the depth to 1, the second one encountered at max( sets it to 2, a comma is encountered, the sets it back to 1, a second comma is encountered, and the last ) sets it to 0, which lets us know we're done. The appropriate comma would be ONLY in depth 1. The one encountered in depth 2 isn't the correct one and is ignored. Any additional ")" later would be depth -1 or lower, but the parser would have stopped searching already when it got to 0. Technically I think we start at depth 0 and ignore the first "(", and stop when depth goes negative, but it's equivalent. I might not get to work on things for a few days, but I might push one or two more small changes here and there. I see no reason why a release shouldn't happen by this weekend.

Definitely! It's by no means a required feature for normal gameplay. Also, the worst I can imagine happening is that some module that needs reset doesn't get reset, in which case you can still scrap the vessel or replace the part with a new version. I'm also setting it up so you can add additional modules that should be reset to a config file along with the values they should have when reset. Currently engines, science equipment, and parachutes are reset.

I considered having science be serial like the VAB/SPH and I also considered requiring you to complete pre-req nodes before further nodes would progress. In the end, I preferred the current system. The R&D center is big. It's got a lot of people in it working on independent things all at the same time. More advanced technologies require more time still, but can also be worked on at the same time (though, realism should require them to be at a reduced rate while their pre-req nodes are still in progress). The VAB is also big, but so are rockets, and you can realistically only build a small amount of independent things at once. If you find the research speeds to have too large of a jump, you can very easily change the formula that defines it in the KCT_Formulas.cfg file. The current formula for the node completion rate is "2^([N]+1) / 86400" where [N] gets replaced by the number of upgrades (essentially, 2^(N+1) science per 24 hour day, 86400 seconds are in 24 hours). You can change that to be whatever you want, and can even use to represent the science cost of the node (which could be used to make them all take the same amount of time by having higher cost nodes have higher rates). The next version of KCT supports min() and max() functions, along with natural logs and log base 10. For now you're limited to polynomials and exponentials, with max rates defined by a different setting. Next version you can even define the entire build point formula for vessels and can make it mass based if you really want, rather than cost based, or some weird mix of the two. Btw, the order of operations is left to right, but parenthesis are evaluated as soon as they're encountered. So that formula would be evaluated as (0+2)^(N+1)/86400. More complicated ones need to use parenthesis or careful arrangement. 1+1*2 = 4 by this system, not 3. 1*2+1=3 however, as does 1+(1*2). Edit: Btw, if anytone's curious what's coming in the next version (and already exists in the development builds), here's the current WIP changelog for 1.1.5. v1.1.5.0 (3/X/15) New Features: - The Most Requested Feature: Recovering directly to storage. Be warned that it likely has bugs, especially with mod parts. Requires manual refuelling of ships. - KSC Upgrades now have time requirements associated with them. - Several new formulae have been exposed for editing, including the entire BP calculation formula. KCT_Formulas.cfg must be deleted for this update. - Rush builds 10% by spending 20% of the total vessel cost. Interface Changes: - Moved buttons in the build list to before the vessel name - You can CTRL+Click the arrow buttons to move a vessel to top/bottom of the list - Lines in storage are colored according to their current status - Vessels that are rolled out can be launched from the Editor (should save a scene change for KK users) - Crew Select GUI now displays Kerbal class and level - Crew Select GUI options now persist - Simulation time limit selection now persists per save. Defaults to infinite time if "free simulations" is active. - Added support for CrewQ mod (enneract) Bug Fixes: - Added a popup that occurs when KCT doesn't load save data properly. - Fixed issues with editing a ship past VAB/launchpad limits. - Possibly fixed issues with costs being incorrectly calculated in some instances - Made the launch tooltip not get stuck when pressing the launch button (enneract) - Fixed several issues with upgrade point purchases/resetting. - Fix loading a save after a crash during a simulation. - Throttled editor recalculations to reduce lag, especially with procedural parts (enneract) - Fix for Real Fuels boiling off during construction Miscellaneous: - "icons" folder renamed to "Icons" - Added min(x,y), max(x,y), l(x), and L(x) functions to math parser (l=natural log, L=log base 10)

I'm pretty sure I did. I haven't gone through and tested it (I'll do that now, I'm testing some other stuff), but I switched from ints to doubles, so that should take care of that. Also, the formulas that control those are configurable now Edit: Good thing I tested that! I'll have to make sure to warn everyone to delete the KCT_Formulas.cfg file when they update or else those will be broken. Edit2: Just bought 78 upgrades. I think the overflow is pretty much taken care of. It will eventually overflow, but it's gonna be a LOT of upgrades before it does. E3: Curiously, buying upgrades with science appears to be giving me the science. Oops! E4: I think that bug is present in the released version as well... Probably since 0.25... Also, the current dev version has a bunch of weird issues, like losing data :/ Though that's probably limited to MY dev version, not the public ones. Those ones just won't let you build anything because of an NRE. Oh, software development! E5: I think I got everything to a working state and fixed some other longstanding bugs that I didn't even know about. Pretty satisfied with that.

That's just a visual glitch. When you click to purchase the upgrade it'll charge the correct amount and the next time you start the game it should display the correct amount. I've actually already got it fixed in the development versions As for the issues with the Tracking Station, I'll check them out. It's probably a simple fix.

I personally just put my desktop to sleep if I know I'm going to be using it later that day or the next day, but I try to shutdown all the way if it's going to be a while. Then when I turn it back on everything is in a clean state. I've got an SSD and a pretty fast system, so boot times for the system and applications aren't long enough to warrant hibernating to me. But that's all just personal preference. I'm sure there will be some good posts about the merits of each.