Jump to content

allista

Members
 View Profile  See their activity

  • Posts

    2,208

  • Joined

  • Last visited

 Content Type 

Everything posted by allista

  1. allista
    With model wielding through MODEL nodes it does seem possible (yet I'm not totally sure). But it is sooo much easier to just make two separate parts for that scenario Especially now with the HangarStorage-HangarPassage framework.
  2. allista
    TCA-1.5.0-BETA.2may be downloaded from the qfeys repository This is a BETA release which is intended to be used for testing. Read the ChangeLog and BACKUP YOUR SAVES.
  3. allista
    No-no-no! SAS is the Stability Augmentation System -- the software inside a probe-core or a qualified kerbal who pushes the levers. Reaction wheels and RCS-engines are too just actuators. And TCA tries to actuate SAS commands through thrust limits. So if you have SAS-capable ships and SAS itself enabled (with default 'T' key), each of devices (wheels, RCS, TCA+engines) that provide torque will react on its command.
  4. allista
    Two advices: 1) TCA v1.4.1 and below is useless without SAS. It is the SAS that tries to balance the ship, TCA only actuates its commands. 2) We're currently working on the next release which is a complete rewrite and will address these problems. The first beta release is published above so you can compare, but the development moves on and current dev-TCA already uses a very different engine-controller mechanism than even 1.5.0-beta.0. So the advice, unfortunately, is to wait for the next release.
  5. allista
    1) It was not only exploity, but also dangerous, as upon launch the rover could intersect with the lander and the whole thing would explode. The check in editor is a precaution. 2) This was discussed already (a long time ago). There is a problem with making anisotropic resize of model space: it makes transformations from submodel spaces into world space also anisotropic. This means that if you have an animated model where parts of it move non-perpendicular to the direction of anisotropy, these parts become distorted as they move. Too see this for yourself, you may open the RoverLander.cfg and in HangarPartResizer module node set the SizeOnly flag to false. Try then to increase the aspect (say, to 2) of the lander and open the doors, looking at the doors and hydraulic cylinders.
  6. allista
    I meant github version of BetterBouyancy About resizing: See the ChangeLog. In v2.0.2 I've fixed in-editor check of a stored vessel's dimensions on hangar resize. Hangars again remove stored vessels if they don't fit anymore.
  7. allista
    If you resized the Rover Lander, then the jump is caused by it intersecting with the launchpad before physics kick in. You see, a vessel is placed on the launchpad before modules are activated, so it is placed in it's default size and only then is resized by resizer module. For now the only way to prevent it is to use launch clamps. Another thing that happens upon launch is storing of constructs that were saved in editor. This process is accompanied by very fast switching between the hangar and launched-and-stored vessels. But this cause the camera to jitter, not the vessel.
  8. allista
    Sorry, I will try to look into the matter ASAP, probably this evening. It's a working week, you know. - - - Updated - - - I've failed to reproduce the bug, then looked into your logs and found the culprit: NullReferenceException: Object reference not set to an instance of an object at [B]BetterBuoyancy[/B].BBCrewEVAWatcher.HandleCrewEVA (.Vessel v) [0x00000] in <filename unknown>:0 at EventData`1[Vessel].Fire (.Vessel data) [0x00000] in <filename unknown>:0 at ProtoVessel.Load (.FlightState st) [0x00000] in <filename unknown>:0 at AtHangar.StoredVessel.Load () [0x00000] in <filename unknown>:0 at AtHangar.HangarMachinery.TryRestoreVessel (AtHangar.StoredVessel stored_vessel) [0x00000] in <filename unknown>:0 at AtHangar.HangarWindow.LaunchButton () [0x00000] in <filename unknown>:0 at AtHangar.HangarWindow.HangarCotrols (Int32 windowID) [0x00000] in <filename unknown>:0 at UnityEngine.GUILayout+LayoutedWindow.DoWindow (Int32 windowID) [0x00000] in <filename unknown>:0 at UnityEngine.GUI.CallWindowDelegate (UnityEngine.WindowFunction func, Int32 id, UnityEngine.GUISkin _skin, Int32 forceRect, Single width, Single height, UnityEngine.GUIStyle style) [0x00000] in <filename unknown>:0 It's the same bug that haunted ScienceAlert: BB handles poorly the onVesselCreate event. Again, this should affect ExtraplanetaryLaunchpads as well. Sorry it took me so long to react: I have busy schedule these days. I'll write to ferram about it. - - - Updated - - - No need to write to ferram it seems. He found the bug and fixed it in the code already. You may try to grab the dev snapshot or wait for the next release.
  9. allista
    Sorry, I cannot find anything in KSP.log. Unfortunately, in most cases it's useless. And you should have output_log.txt under {KSP}\KSP_Data\ if you're on Win. If on Linux, search for the Player.log in $HOME/.config/unity3d/Squad/Kerbal\ Space\ Program/
  10. allista
    Well, I can't argue with that I was just pissed of by NetKANians who've changed the rules with redistributables without explicitly stating it anywhere, and they change the metadata you provide without informing you and break your installation; but it's your reputation that suffers in the end and it's you who end up fixing the mess. I agree that they're trying to do the right thing and trying hard, but I still certain that it's their responsibility to make their installation process work (provided that the manually installed package does) either by tweaking their rules or by contacting the developer before changing anything, so that he/she could figure out the problem. If that was the case, the fix would be there two weeks already and all would be happier.
  11. allista
    Oh, boy! Looks ugly, I'm sorry for the mess. The first thing you can do is to publish your KSP_Data/output_log.txt, so that I could see what really happened. Also, were you using some previous issues without this issue?
  12. allista
    Everyone who has trouble with CKAN, please, try to install the new version ( 2.0.2.2) and tell me if it's working.
  13. allista
    Hangar uses the latest KAE, and local copy is just the way to NOT have a dependency on a particular path and user's environment. It's a normal practice for c# assemblies to be redistributables (e.g. TweakScale provides its own redistributable for the same purpose -- to be included in mod's distribution). Even more: KAE has an election system that always chooses the latest of the local versions and disables the rest. I don't know about KSP-AVC -- it does not check for dlls, only for the .version file, and native KAE distribution does not provide one. CKAN may be trying to do a good thing, but they create virtual mod conflicts and contradict the plugin developers in the way they want to provide their content. It's not their place to decide how a plugin should be linked and packaged. I would not have problems either if the NetKAN maintainers had not messed with my .netkan file. It was their decision to incorrectly state that the Hangar "conflicts" with KAE. It does not.
  14. allista
    I've released a compatiility update with no requirement for the KAE .dll to be in the Hangar/Plugins. I also updated the Hangar.netkan file, but until they pull in my request and the NetKAN bot indexes it you won't see the change. So I suggest you to just grab the update from KerbalStaff (in fact, v2.0.2 should also work if you install it manually). But to be honest: this is the last time I'm dealing with CKAN. They add redundant restrictions and mess up with the metadata files you provide without telling you. I have my hands full with the development to spend effort on support for yet another buggy third party addon manager
  15. allista
    Be damned the CKAN and their ideas of mod dependencies! Hangar does use KSPAPIExtensions itself, but as many other mods, it provides its own copy of it. Which is a normal and intended usage pattern for KAE. But in the CKAN they decided that if you have one copy of a dll, you cannot install another one. So, apparently, if one mod pulls in KAE as a dependence, the other cannot bundle KAE dll. I will try to sort it out somehow, but really, it's CKAN's problem UPD: Even more! The "wise" NetKAN maintainers ADDED the conflicts statement: "conflicts": [ { "name": "KSPAPIExtensions" } ], to my Hangar.netkan! No wonder you get the message. Bastards!
  16. allista
    TCA-1.5.0-BETA.0may be downloaded from the qfeys repository This is a BETA release which is intended to be used for testing. Read the ChangeLog and BACKUP YOUR SAVES.
  17. allista
    Sorry for the delay. What you've described does look like Hangar's bug, but to understand anything I, as always, need logs of the gamerun during which the bug happened. Please, publish your logs, so I could examine them. If you don't know how to do this or where the logs are, look through several of the last pages of this thread: the routine was explained in detail here several times.
  18. allista
    While Zenka works on the full-fledged new release I took the liberty to port the 1.4.0 to KSP-0.90 and made some improvements on the way. You can download the new version from GitHub: TCA-1.4.1 (KSP 0.90) UPD: I've fixed the link in the KSP-AVC .version file. If you're using it, please, replace the GameData/ThrottleControlledAvionics/ThrottleControlledAvionics.version with the new version.
  19. allista
    And I was just thinking to take over development (or at least maintenance) myself, as I desperately need the TCA for Spaceplane Hangars I'm developing. In either way I may help with both programming and modeling (if the need in the latter arises).
  20. allista
    I've figured out what is happening. It's a KSP bug that should occur whenever a mod that provides custom Agents for Contracts System is removed. After that if you try to do anything that triggers the game to be saved (enter VAB/SHP, exit Contracts building and so on), a NullReferenceException occurs in Contracts.Contract.Save because the agent does not exist anymore. The only way to fix this is to manually remove all the contracts (including finished ones) with such agents from the corrupted savegame. I've explained what to do in a simple HOWTO. UPD: I've filed a bug report to KSP-support.
  21. allista
    Your logs say "out of memory". Don't know how you've managed to run out of it on 64bit system, but then again, it is said that Win64-KSP is buggy beyond measure and Win's 64bit support is crippled.
  22. allista
    Indeed! Thanks!
  23. allista
    Understood. Will work on this first thing after the NY.
  24. allista
    Glad it's helpful Hm... I think I know the reason and the solution: Stock docking ports are surface-attachable by default. This means that they stick to any (this is just a bad implementation by Squad) collider you point to with the mouse. Inline hangars have their trigger colliders running along their axis, so a docking port is probably sticking to the surface of a trigger, instead of attaching to the docking port node. To overcome this you may use EditorExtensions to temporarily disable surface attachment. I personally use EE from the first day of playing KSP and haven't really noticed the problem until now, so thank you for bringing my attention to it. Will add it to Known Issues.
  25. allista
    It's not that hard, but it takes some time to explain, so it's better to write proper docs for modders with screenshots and all. And if you're not in a hurry I would rather do it after the NY. Right now there's too much work and presents hunting
×
×
  • Create New...