Search Unity

  1. Get all the Unite Berlin 2018 news on the blog.
    Dismiss Notice
  2. Unity 2018.2 has arrived! Read about it here.
    Dismiss Notice
  3. We're looking for your feedback on the platforms you use and how you use them. Let us know!
    Dismiss Notice
  4. The Unity Pro & Visual Studio Professional Bundle gives you the tools you need to develop faster & collaborate more efficiently. Learn more.
    Dismiss Notice
  5. Improve your Unity skills with a certified instructor in a private, interactive classroom. Learn more.
    Dismiss Notice
  6. ARCore is out of developer preview! Read about it here.
    Dismiss Notice
  7. Magic Leap’s Lumin SDK Technical Preview for Unity lets you get started creating content for Magic Leap One™. Find more information on our blog!
    Dismiss Notice
  8. Want to see the most recent patch releases? Take a peek at the patch release page.
    Dismiss Notice

What happened to Unity 2018 in Physics? It breaks my entire game with child rigidbodies...

Discussion in 'Physics' started by Marcos-Elias, Aug 12, 2018 at 10:15 PM.

  1. Marcos-Elias

    Marcos-Elias

    Joined:
    Nov 1, 2014
    Posts:
    69
    I recorded a detailed video showing how Unity 2018.2 affected my project, comparing the same features both in 2018.1b11 (where everything works) and 2018.2.1f1 (broken physics). Tested with 2018.2.2 and .3 later, same problem.



    What's happening? Kinematic rigidbodies should NOT be affected by the forces, I learned... It's impossible upgrading my project to all newer Unity versions. It worked in almost the same way with Unity 4, 5.x, 2017.x up to 2018.1b11. But final releases of 2018.x broke everything. I tried many changes, including leaving all child kinematic rigidbodies with the same interpolation settings... Nothing seems to work. Tried also Physics.autoSyncTransforms both to true and false, same bugs...

    Is that a bug or something was changed?! It's hard to believe that this critical bug still exists in a "final" release, if it is a bug...

    1 - Moving kinematic rigidbodies inside a moving non-kinematic rigidbody have a strange behaviour, they are left behind. Watch people in my bus, shown at 8:35.

    2 - Each time we put the vehicle on the ground from some height, the child kinematic rigidbodies are moved a little bit to the top on y (check at 6:19 to 7~8min). Unlike the people on a local waypoint system inside the bus, buttons don't have any script moving them, they have a collider and a rigidbody just to recognize mouse clicks.

    Both problems don't happen with Unity 2018.1 beta 11. Tested with 2018.2.3, bug still exists besides some fixes in physics.

    This bug seems related to this one, I'm not sure:
    https://issuetracker.unity3d.com/is...ild-object-to-be-misplaced-after-the-movement

    My project is too big/too complex to send a sample project, I may try exporting only a bus and some part of the map to the bug report, but I cannot do it right now.
     
  2. Peter77

    Peter77

    Joined:
    Jun 12, 2013
    Posts:
    2,437
    If it works with 2018.1b11, but stops working with 2018.2.1f1, then it's most likely a bug and submitting a bug-report is sounds like the appropriate thing to do.
    https://unity3d.com/unity/qa/bug-reporting

    If you do not submit a bug-report, which Unity Technologies can use to reproduce the issue and test their fix against, it's rather unlikely that they're going to work on a fix, according to my experience.

     
  3. Edy

    Edy

    Joined:
    Jun 3, 2010
    Posts:
    1,143
    My guess is that the parent transforms are somewhat affecting the transforms in the child rigidbodies and these take the new values as an update, which should not happen. Child rigidbodies should be independent of their parents as the Rigibdody component controls the Transform.

    When you modify the RB's transform explicitly the RB takes the new values as an instant update to its position. How and when this happens is what has been worked around in recent versions (as with the introduction of Physics.autoSyncTransforms and Physics.SyncTransforms).

    I'd experiment with rigibdody hierarchies. Try to find some case where the parenthood relationship is affecting the child RB's transform in some way in one version, but differently in other. Then you could send a bug report with it.
     
    Last edited: Aug 13, 2018 at 11:47 AM