Search Issue Tracker
Third Party Issue
Votes
0
Found in
2020.3.35f1
2021.3.5f1
2022.1.6f1
2022.2.0a17
2023.1.0a2
Issue ID
UUM-7396
Regression
No
GameObjects with "AddForceAtPosition" are not rotating to the force direction when Force has a positive angle
How to reproduce:
1. Open project "PhysicsForce.zip"
2. Enter Play Mode and press the Pause button near the Play button
3. In the Hierarchy window select "Cube (1)" GameObject
4. In the Inspector window change "Force Test" Component "Y Angle Rotate" to 30
5. In the Hierarchy window select "Cube (2)" GameObject
6. In the Inspector window change "Force Test" Component "Y Angle Rotate" to -30
7. Unpause the Game
8. Observe the result
Expected result: Both GameObjects are moving towards the force direction
Actual result: GameObject with positive angle (Cube (1)) is moving not towards the force direction
Reproducible with: 2019.4.37f1, 2020.3.31f1, 2021.2.17f1, 2022.1.0b13, 2022.2.0a8
Add comment
All about bugs
View bugs we have successfully reproduced, and vote for the bugs you want to see fixed most urgently.
Latest issues
- "One or more data file missing for baking set NewScene Baking Set. Cannot load shared data." error in Player when a specific project is built
- Choosing new HDR Colour using RGB values breaks colour on Intensity Selectors
- Rendering/Decal Layer Mask options are different inside Prefab Mode and outside Prefab Mode when the project is upgraded to Unity 6
- Incorrect Realtime GI Light Probes baking when more than one Light Probe Group is used and "Baked Global Illumination" is enabled
- Hovering on the three-dot menu also highlights the tab header near it when not all tabs fit the bar
Resolution Note:
This is a complex issue inside PhysX (even with the latest PhysX 5). Nvidia is aware and looking into it but for now here are a few workarounds:
- Do not use the Patch Friction type, it only picks 2 out of 4 contacts and the result is never going to be symmetrical.
- Use the Two-Directional Friction type. It works on all 4 of the produced contacts and has the best chance of producing a symmetrical result.
- Bump us the velocity (and potentially position) iteration count to increase the convergence of the friction constraints