Parts & VesselsDecoupled/Undocked part behaves like it's still connected to the craft (drifting the camera, keeping thrust control...) because of copied part.
Video Driver: 4.6 (Compatibility Profile) Mesa 23.1.9-manjaro1.1
VRAM: 12 GB
Severity: Low (Spicat Edit: High: mission ending)
Frequency: Reproducible
Description:
Spoiler
I attached 4 satellites meant for a comms network to a hub using stack decouplers. Symmetry does not work on hubs, so I attached them individually and added the decouplers as a group to a stage action for simultaneous release. Upon release, 1 became its own craft and drifted away as intended, but the other 3 stayed clustered around the main craft, their engines showing in the resources list, and all responded to control input. They also tied the camera to the center of mass of the entire group of linked parts, so as the partially separated craft moved away from the main craft, the view also drifted.
I tested this several times, including with mods disabled, then tested again using individually placed radial attach points around a central tube to eliminate the hub as a factor. The issue continued exactly as before.
Attached is a craft file with one demonstration of the issue. Fly straight up, ditch the shell, and attempt to deploy the dummy satellites. You should find one becomes independent and the other 3 will stay tied to the main craft.
Requesting a Windows user test and verify this since Linux is not officially supported yet.Spicat Edit: Not relevant, work on Windows
Mods used (Spicat Edit: Not relevant, work in stock):
Video evidence (from Spicat with the craft file provided):
Spoiler
Spicat Edit:
Steps to reproduce:
In the vab, put a pod
Attach a launch clamp to it (optional but it's to see the bug better)
Attach a hub on top
Next to the craft (in a different subassembly), put a tank and attach a decoupler
Take the tank and attach it to the hub so that the decoupler is the part separating the hub and the tank - Important
Copy the decoupler (thus, with the tank)
Attach the copied part to another hub connection
Launch and decouple
Result: When the copied part is decoupled, he thinks that it's still part of the craft, so the camera follow and the part still reacts to the input.
Video evidence (with all the steps):
Steps to reproduce (Docking port version):
In the vab, put a pod
Attach a clamp to it (it's not mandatory like above)
Attach a docking port to the pod (or any other part, a truss in the video below)
Attach an engine to that docking port (to detach it)
Copy the docking port (with the engine on it) and attach it next to other one
Go to the launchpad
Undock the docking port (the one that was copied)
Video evidence:
Spoiler
Additional Information:
The version with the docking port also work if you attach another docking port instead of an engine but it triggers another bug (You can see what does the version with only docking port in it):
Bug Report
MoonfirePDX
Reported Version: v0.1.5 (latest) | Mods: See description for full list | Can replicate without mods? Yes
OS: Manjaro Linux + Steam Proton | CPU: AMD Ryzen 7 5800X | GPU: AMD Radeon RX 6700 XT | RAM: 32 GB
Spicat Edit: This bug work with no mods and on windows (see below)
Initial bug report (hidden for visibility)
Specs:
Severity: Low (Spicat Edit: High: mission ending)
Frequency: Reproducible
Description:
I attached 4 satellites meant for a comms network to a hub using stack decouplers. Symmetry does not work on hubs, so I attached them individually and added the decouplers as a group to a stage action for simultaneous release. Upon release, 1 became its own craft and drifted away as intended, but the other 3 stayed clustered around the main craft, their engines showing in the resources list, and all responded to control input. They also tied the camera to the center of mass of the entire group of linked parts, so as the partially separated craft moved away from the main craft, the view also drifted.
I tested this several times, including with mods disabled, then tested again using individually placed radial attach points around a central tube to eliminate the hub as a factor. The issue continued exactly as before.
Attached is a craft file with one demonstration of the issue. Fly straight up, ditch the shell, and attempt to deploy the dummy satellites. You should find one becomes independent and the other 3 will stay tied to the main craft.
Requesting a Windows user test and verify this since Linux is not officially supported yet.Spicat Edit: Not relevant, work on WindowsMods used (Spicat Edit: Not relevant, work in stock):
Craft file:
Test.json
Video evidence (from Spicat with the craft file provided):
Spicat Edit:
Steps to reproduce:
Result: When the copied part is decoupled, he thinks that it's still part of the craft, so the camera follow and the part still reacts to the input.
Video evidence (with all the steps):
Steps to reproduce (Docking port version):
Video evidence:
Additional Information:
The version with the docking port also work if you attach another docking port instead of an engine but it triggers another bug (You can see what does the version with only docking port in it):
Link to comment
Share on other sites
7 answers to this question
Recommended Posts