Parts & Vessels Parts Are Moving Slightly Out of Place During Flight (Affecting SAS)

[zaqwsx156]

Reported Version: v0.2.0 (latest) | Mods: BepInEx, Commlines, Community Fixes, Flight Plan, Hide Orbits, K2D2, Maneuver Node Controller, Node Manager, Orbital Survey, Patch Manager, SpaceWarp, UITK for KSP 2 | Can replicate without mods? Yes 
OS: Windows 10 Home 19045.3803 | CPU: i7-13700KF | GPU: NVIDIA RTX 3060 12GB | RAM: 32 GB

 

I'm not sure if it's always happening, but it seems to happen every time I launch the specific rocket in the video, as well as happening with other rockets occassionally.

The vessel will suddenly have many of it's parts slightly offset from each other mid-flight, without apparent reason.

I've also found that this causes the craft to no longer be able to correctly/accurately find SAS points (prograde, etc.). It is off by a degree or two, but there's no indication that the craft is attempting to correct this (no movement on the Pitch/Yaw/Roll indicators).

EDIT: On vessels affected, it seems to happen after a non-physics time warp.

EDIT 2: I believe I've identified what causes the issue. It seems to be associated with the Tiny Reaction Wheel (TRW-0625). I isolated it after carefully removing each part from the vessel, flying it above 70k, then doing a few short time warps at 0 thrust. Issue doesn't occur when  the Tiny Reaction Wheel is not on the vessel.

 

Included Attachments:

 

OrbitalRelayProbeMk8.json

Edited January 4, 2024 by The Space Peacock
Identified Seeming Root Cause

[ocueye]

Reported Version: v0.2.0 (latest) | Mods: none | Can replicate without mods? Yes 
OS: windows 11 (64-bit) | CPU: 11th gen intel core i5-1150 | GPU: NVIDIA GeForce RTX 3060 | RAM: 16103 Mb

 

I switched menus and parts on my ship shifted a bit even colliding with each other and causing the craft to spin uncontrollably.

Video Evidence:

 

 

[Vl3d]

I can confirm - it happened to me while going to Moho. Everything got very badly misaligned! F5-F9 does not fix it.

Edited December 31, 2023 by Vl3d

[zaqwsx156]

Happened to me again. Different vessel setup.

[The Space Peacock]

@ocueye merged your report

[zaqwsx156]

Commenting for visibility:

Edited original message with what I discovered about the issue. Temporary fix until this bug is squashed is to not use Tiny Reaction Wheels.

[Anonymous Bug Reporter]

Reported Version: v0.2.0 (latest) | Mods: none | Can replicate without mods? Yes 
OS: Windows | CPU: Intel i9 | GPU: NVIDIA 3060 | RAM: 32 GB

I have had this bug show up twice now since the new science update, basically, a part's location will move some arbitrary distance away from the vehicle/ship- while still being "rigidly attached" to the vehicle, despite not touching it or even being within 30 meters of it. I had a lander leg which appears to be 50 meters from the vehicle. When I landed on the mun, it acted like a rigid lander leg that was attached to the vehicle - despite being far away. It was supporting the ship's weight, too. 

Included Attachments:

[c_bunny]

Reported Version: v0.2.1 (latest) | Mods: none | Can replicate without mods? Yes 
OS: Manjaro Linux | CPU: AMD R7 5700X | GPU: AMD Radeon RX7800xt | RAM: 32GB

The parts on my space station around Jool have shifted into incorrect locations. The space station (with docked landing craft) appeared correct when I first got into Jool orbit and performed science missions. I later launched a craft from Kerbin to land on Laythe with the intention of returning the Laythe lander to the space station, however after reaching Laythe orbit I switched to control the space station and found the parts had shifted out of place. 

The parts appear to still be "logically" connected, I can transfer fuel and Kerbals around. If I rotate the station all the parts rotate around a common point. The part physics appears to respect the new location, it is impossible to move the station since the thrust vector no longer points through the center of mass. 

I think the same bug was reported for version 0.2.0 but the issue persists in version 0.2.1.

Included Attachments:

ForScience!.json

Pollander.json

[DibzNr]

@c_bunnyMerged your report with this post

@zaqwsx156Unarchived your report