Boris-Barboris

works on my machine

It may execute in time but change something in the stock module/gameobject ordering. Could the number "8" be occupied by something?

Dear Sarbian, I've got a guy who got AtmosphereAutopilot broken by MFI 1.2.8 (I reproduced myself as well), apparently by this commit I do not understand the exact reason why the control is borked, but I have a pretty strong gut feeling that the stock update order is broken by it. Control/model artifacts I see suggest that the old schedule of physics update: Control input taken from input devices and written to vessel's control state -> onPreAutopilot/onAutopilot/onPost.../onFlyByWire -> control surface module applies the actuation -> aerodynamics forces are applied -> Unity Phys

@Tahvohck and everyone else. It's ModularFlightIntergrator's new version. Rollback to 1.2.7.0 or updade to 1.2.9.0+, AA has no problem with it.

@Tahvohck First video indeed looks like some conflict, as if the control output from the AA is not matching the input to control surfaces, maybe there is an additional mod in the control chain. AA model decided that your craft is extremely unstable and overmoderated allowed AoA to such a small value that you could not recover from a nosedive. I don't think I'll have a look before friday, so if you could try to deduce the mod in question it could help immensely.

There was a landed-state detection bug but it was fixed in 1.5.14, are you guyes saying it's back in recent KSP versions on 1.5.16 AA?

@danbadone and others with the same problem, try https://github.com/Boris-Barboris/AtmosphereAutopilot/releases/tag/1.5.17rc1

I'll try to re-compile the dll against the new ksp libraries, maybe that'll be enough.

that's quite an unfortunate crash, with no error or trace... not sure I'll be able to debug it.

FAR mod does not patch control surfaces of those modded shuttle parts. Talk to the part pack authors, and maybe FAR thread.

In "Flight model" UI panel, is "has_csurf" true for these shuttles?

It's OK, AA's model is capricious and borks itself sometimes. Sometimes toggling it off and on a couple of times helps.

Looks like a bug in control surface logic. Try AAP's control surfaces (in my signature).

nah, sorry

I had no such intention. Possible, but I never tried and had no interest.