allista
-
Posts
2,208 -
Joined
-
Last visited
Content Type
Profiles
Forums
Developer Articles
KSP2 Release Notes
Everything posted by allista
-
Hangar v2.3.2-BETA1 for KSP 1.0.2 is released You can download it at GitHub for testing I do not recommend to use it with your main game It is far from tested and SHOULD still have some issues with the new API, especially with new aerodynamics and engine models. But remember: the more you test it and report bugs, the sooner I'll make a proper release. Also, it would be best if you use GitHub's bug tracker to report bugs instead of this forum, as it's much easier to access from the phone. Thank you all for your patience and support! - - - Updated - - - And to explain the delay to some extent, here's my current schedule: work-work-work, a lot of experimental work planning a vacation (29.06->15.07); and that means getting visa with all the bureaucracy finally marrying my wife officially in two weeks 16.07+ a conference on bioinformatics 27.08+ expedition to Kamchatka I know, I'm a bad, baaad developer
-
Good idea, but I'm afraid I don't have time to work on such integration. It should be a huge work. But KAS may certainly use my code and count on my help if they decide to incorporate some of the Hangar's features. - - - Updated - - - Next week I have to make a mid-term report about 2 years of my work in the lab >_< So until then no programming; only number-crunching, statistics, slides and speech-writing - - - Updated - - - If you're using KSP-1.0.2 -- wait for the update. If 0.90, the hatch should work. Arm it and aim carefully: it is best to attach it to a flat or slightly convex surface, so that the distance from the surface to the middle of the hatch is minimal. Imagine that the hatch has a very short hook at its center.
-
Thank you, Kablooey, your explanations are perfectly correct. To "clarify" things a bit: TCA exposes only the P and I coefficients because it actually uses a PI controller (without the differential component) as a band-frequency filter for engine's thrust limiter to prevent tremor and oscillations. And the values of the limiters are computed by optimization of the objective function F=|total_toque - needed_torque| So sensitivity of the control response should be tuned via the main Steering Gain or the separate yaw-pitch-roll gains. - - - Updated - - - Currently there's no way, except to switch them into the manual mode; but that's not what you want I suppose) Apparently, we need a fourth engine mode in TCA: Balance Only -- group of engines that is optimized to always provide zero torque. That's not easy with the current calculation scheme, but I have the general idea of how to do it. What is strange, is that when you lower the steering gains your craft looses steadiness. These controls only affect the steering, not the balancing. Then again, without some steering help from the SAS a balanced craft experience negligible torque, but it still may have a non-zero angular velocity. - - - Updated - - - Now I see. That is indeed a useful option. Do you have an idea of how this should be implemented in terms of user interface? I mean, we only have one Main Throttle. Of course, if I add the Pitch->altitude control, we may use the MT for the linear thrusters group, but if a user wants to use MT for altitude, then what? Add a separate slider and control this groups' thrust with a mouse? - - - Updated - - - Thank you for the report. I see the problem and will work on a fix. It is strange, though, that it was working in 0.90, because I haven't changed calculations even a bit in the 1.0.2 compatibility release. So the problem is in some obscure KSP-API changes that I missed.
-
Oh boy! Now THAT is the real pain
-
And after that you can never be sure that you'd hunted down all the bugs and none are lurking under the hull somewhere It's like driving a car fixed after a serious crash. It works, but you don't trust it like before...
-
So bloody true!
-
So far it does not look very good, I'm afraid. The version I've compiled against KSP-1.0.2 (by simply fixing naming inconsistencies) does not work, raising many new NREs in previously tested places. Some of them I've already fixed, but I don't know how many more there are.
-
I never tested as I upgraded from 0.90 directly to 1.0.2, but you may try. I doubt they changed a lot (or even anything) in the API in two revisions. The KSP-AVC may be complaining if you have it installed, but you may fix by changing the KSP_VERSION_MIN in GameData/Hangar/Hangar.version file.
-
It is on the CKAN which should update its database soon (I've just published the update). If not, please report back -- CKAN is prone to many errors during automatic updates.
-
TCA v2.2.1 compatible with KSP-1.0.2 is released. All download links are in the main post of the thread.
-
The good news is: KSP-1.0 is awesome! They finally added all the essential functionality (like aerodynamics and reentry heating and versatile engine models) that modders' community striven to maintain until now. The "bad" news -- the API was changed significantly (Squad's code became more streamlined and effective in many ways as I've seen so far), there are many new part parameters from the new physical models and all. While a simple recomplilation may be a matter of a few hours, proper adaptation of the Hangar will take time. The time I don't currently have in enough quantity to work effectively. So the best course of action, it seems, would be to release a "working" recompilation ASAP, and then to work on a bug-report basis.
-
Brilliant! And I've always thought that it's impossible: several times tried to switch with [] between a ship and its last stage both returning from orbit. Always saw the "Cannot switch while in atmosphere" message. O_o
-
Not while in the atmosphere, at least in .90 and before.
-
I'm not sure I understand your proposal. Could you post a picture or a .craft file of the vessel in question and describe which engines you want to control differently and how in particular? So basically you propose the same mechanism that exists now with the MainThrottle that controls vertical speed, but instead of the MT you want to use Pitch value. If I have understood you correctly, that's pretty easy to implement. I gave some more thought to this idea and see only one, but serious issue: KSP does not allow you to switch vessels while flying in the atmosphere. So even with multiple TCA instances you simply cannot switch from your hovering carrier to another vessel.
-
I haven't upgraded my KSP installation yet, so I don't know how much time will it take to fix all compatibility issues. Depending on the extent of the API changes I may be able to update TCA in a couple of days, up to a week maybe. Sorry for the delay, too much work(
-
I haven't upgraded my KSP installation yet, so I don't know how much time will it take to fix all compatibility issues. Depending on the extent of the API changes I may be able to update Hangar in a couple of days, up to a week maybe. Sorry for the delay.
-
Have you removed the previous version, or installed by overwriting? If the latter, you may have a conflict with the older version, so move away the old Hangar folder and install anew. Otherwise, could you publish the whole log, not just these lines? - - - Updated - - - That would not be hard to implement, so yes. Along with the saved tank ratios. But not until summer, I'm afraid. Busy work schedule.
-
Thank you, Enceos, for playing host in my absence! Your answers are very accurate. The workflow is described in Hangar's wiki in the Asteroid Hangars section and its subsections. Thank you for the report! I'll try to publish a fix in the next few days. Theoretically this could be done (not easily though), but I really don't see why do it. This only differs from the present inline hangars in not-renderred hull of empty hangar. But meaningfully this could only be applied in editor; in flight, even when you empty the "Pod" it should stay there, right? So, as I see it, we're talking about in-editor eye-candy that costs a lot of effort.
-
Thanks! I don't quite understand. What do you mean by "can't restore satellites"? And what parts exactly? - - - Updated - - - Perfect! Thanks a lot!
-
It seems I have finally came up with a fix. But before I release it, I want to ask those of you who had problems with vessel launching in orbit at Minmus and other planets to test it thoroughly. Just replace the Hangar.dll in Hangar/Plugins folder with the new version: https://www.dropbox.com/s/qhcewyikuz7ijla/Hangar.dll?dl=0
-
Fine parts they are, and there will be no problem converting separate cargo bays, but I can't make a compound hangar from several bays and back door as they show in the screenshots: these will still be several separate hangars.
-
Thanks for the reports, all of you. I know about the problem and I'm still looking for the solution. What causes it is the Krakensbane, the moving frame of reference. When you launch a ship but before the physics kicks in all the rigid bodies of the parts have 0 velocities, but all the rigid bodies of the mother-ship have the velocity of the reference frame. So even when I synchronize the orbits and match orbital velocities, for the PhysX there still is a collision. In Kerbin's orbit the frame-speed is low, about 1.5m/s; you may notice it when, after the launched ship is switched to, it sometimes bumps a little to the hangar's wall. But in Minmus orbit it may be ~100m/s, thus the explosion. I'm very sorry for the inconvenience, but I cannot revert to the old launch system now, so we have to wait until I find out how to deal with it. What may help, though, is a quick-save/reload, or switch form and to the ship. At least Badsector states so.
-
Well... I have to say, I'm almost convinced But this will be a serious overhaul which will take time... Ah, well, not almost, apparently
-
I suspect it is a conflict with another mod. Nextjr recently reported that Kerbal Foundries causes such bug. But to know for sure, we need to look at the logs. The mod that cause the issue does not, probably, make proper null checks in its handler of onNewVesselCreated event. So you need to check the KSP_Data\output_log.txt (on win): search for any exceptions that occur after the initializing flight mode line and you may find the culprit. If not, then send me this log and I'll see what I can do. Note, that the log is erased and written from scratch on each game launch.
-
No, I'm not. But I saw your description today and right now I'm testing the fix. And it seems to be working.
