Jump to content

allista

Members
 View Profile  See their activity

  • Posts

    2,208

  • Joined

  • Last visited

 Content Type 

Everything posted by allista

  1. allista
    Inside KSP folder: $> wget https://github.com/KSP-CKAN/NetKAN/raw/master/NetKAN/GroundConstruction-Core.netkan $> mono netkan.exe GroundConstruction-Core.netkan $> mono ckan.exe install -c GroundConstruction-Core-1.1.2.1.ckan Yesterday I've asked about this in #ckan IRC and they say this may be a problem with the netkan.exe version the netkan-bot uses (as it always uses the latest build). I've also created an issue in NetKAN repo: https://github.com/KSP-CKAN/NetKAN/issues/5289 Could it be that the engineers just don't yet have their ConstructionSkill > 0? They would probably need a level-up for that, right? To be honest, I'm not all that good with netkan schema versioning either. Both files are simply derived from working metadata for other mods. And the exact same scheme works alright with ConfigurableContainers-Core. Edit: also, as far as I know, v1.4 is older than v1.16 because minor number 4 is less than minor number 16. There's no special meaning to the digits of a number in versions.
  2. allista
    No.
  3. allista
    Hell if I know what that means Local install from .netkan->.ckan file works fine.
  4. allista
    The last I checked (2 seconds ago) it's in the archive where it should be. So probably and installation problem.
  5. allista
    I'll try it myself then.
  6. allista
    Understandable. And I have nothing to say against CKAN stuff; they do tremendous work!
  7. allista
    No. CKAN people just haven't merged my pull-request yet: https://github.com/KSP-CKAN/NetKAN/pull/5282
  8. allista
    I'm to sure it's possible. Some Firespitter engines are not supported (see above the discussion of that).
  9. allista
    Some screenshots would help to understand the situation better.
  10. allista
    Well, realism-wise this is difficult: you should have equipment and materials with you to repartition the tank and adapt it for different type of storage. The whole idea of tank types was to allow resource switching in flight without the need for all that. Empty one resource, fill with another of the same type.
  11. allista
    Correct!
  12. allista
    Well, you could simply use the MKS archive which should bundle the -core component. Anyway, the -core does not include Patches folder and only includes Parts/DIYKit.
  13. allista
    Thank you, @FreeThinker! It does, however, solves part of the problem. Because this means we're still having a competition in terms of which patch will be applied first. CC will win alphabetically, so now if both IFS and CC are installed, parts will be mostly patched by CC. And if we're talking of parts in flight, this may cause sudden change of the governing fuel switch. CC still handles this badly. So I still think it's best to use only one full fuel switch, leaving the second (if it is needed) as the Core package only.
  14. allista
    It does require kerbals with non-zero amount of ConstructionSkill that all engineers have.
  15. allista
    Since @Terwin was so quick to test things, I've already (just now) publish the fixed version! Ground Construction v1.1.2.1 for KSP 1.2.2 (2017.02.24) ChangeLog:
  16. allista
    Yep... trajectory optimization is far from perfect I'm already working on the replacement with the proper multi-dimensional dV minimization algorithm (conjugate directions type). But it will take some time to integrate.
  17. allista
    I'll have to test this myself to figure it out. Probably, It was trying to move ApA around the moon to achieve resonance at the peak. Sometimes I don't understand how it works anymore... Issue to track: https://github.com/qfeys/ThrottleControlledAvionics/issues/35
  18. allista
    Gravity turn!!! Oh boy... The "problem" is the low altitude. TCA "thinks" you're still not in orbit and tries to raise ApA to at least 10km, before trying anything else. You see, Rendezvous program also works from surface; you can use it while your ship is landed and it will calculate the launch window and perform the classic rocket-style launch with gravity turn before actual rendezvous maneuver... Anyway, I need a more accurate definition of in-orbit state. Some not hard-coded PeA threshold. Meanwhile, you can either allow TCA to raise ApA, or do it yourself. I'm sorry for the inconvenience
  19. allista
    At least all automatic tests are passed on CKAN, so they should soon merge the PR. It seems I've already fixed the price problem (though I would love the conformation from @Terwin; link to the fix is in the comment to the issue on github). And I'm nearing my shutdown-on-keyboard threshold again So if you're waiting for CKAN anyway, I'll publish in the morning...
  20. allista
    There should be an autosave that TCA does before attempting the rendezvous maneuver. If it worked at all, the save should be there.
  21. allista
    Err... no, nothing but target is required. There's definitely something wrong with it, though you can imagine how much I tested it, and I never saw anything like this. The normal operation should be like this: initial orbit correction for resonance with the target (max. dV 100m/s) calculation of the transfer orbit and execution of the corresponding maneuver matching orbital velocity with the target at nearest approach So it should've never even tried to fire engines without a maneuver node...
  22. allista
    The complexity of a part (i.e. kit dry mass / part dry mass) is calculated as follows: complexity = 1-1/(1+x) where x = (PartCost/part.mass+part.Modules.Count*1000)*k So the main factor is the cost/mass ratio. If this kontainer is light but costs a lot, its kit/part mass ratio will be high. Edit: I see the problem now. It's caused by the same calculation bug that results in the double-price bug. Will fix this tonight. Should be like this: C3.Kontainer.02: complexity 0.4736842, KitMass 0.9473684 / PartMas 2 = 0.4736842, KitCost 5894.737 / PartCost 8000 = 0.7368421
  23. allista
    This was always the case, actually. At least for 4 months: https://github.com/allista/hangar/commit/a9a3072127eb1c479e171cf3070d00ac3a5c9051 Hangar does not require the patches, so it does not include them. Especially taking into account the conflicts they cause with other fuel switchers.
  24. allista
    Heck!!! I think I need a vacation... Of course I meant CC-full and IFS-full. @LatiMacciato, do note that, please
  25. allista
    @Terwin, thank you! @DStaal, also thank you! But this is strange: the workers at the workshop are counted at many events, including GameEvents.onVesselCrewWasModified. I will recheck it, but so far I could not reproduce it.
×
×
  • Create New...