CINEVERSITY

    Maxon Logo
    • Login
    • Search
    • Recent
    • Popular
    • Users
    • Groups

    Bullet vs Simulation System RBD Follow Position

    Question & Answers
    sim vs dynamic rigid
    2
    13
    874
    Loading More Posts
    • Oldest to Newest
    • Newest to Oldest
    • Most Votes
    Reply
    • Reply as topic
    Log in to reply
    This topic has been deleted. Only users with topic management privileges can see it.
    • H
      history-growth last edited by Dr. Sassi

      Hi all,

      trying to set up a simple scene where a sphere with RBD is following a path with Follow Position. When I'm doing this with the new Simulation System I get a somewhat erratic behaviour where the sphere goes through the geometry and gets massive amounts of energy - spinning around like crazy.
      The same setup with the old Bullet systems generates a predictable and clean result.

      Am I missing something crucial with Sim System? Or is this a bug?

      Have a look here:
      https://www.dropbox.com/scl/fi/m81ep0rp56x4tt2stfsrf/RBD-vs-sim-system-follow-position.mp4?rlkey=ukoc9z30o7crjte7rgvdq9lfy&dl=0

      1 Reply Last reply Reply Quote
      • Dr. Sassi
        Dr. Sassi last edited by

        Hi history-growth,

        I can understand the expectation; it would be great to just use the gained experience with the Bullet system, but alone, the recoil slider would change the outcome, and it would stray apart.

        Yes, as most dynamic systems themselves deliver different results, even if nothing else has changed, the synchronicity between two different systems is certainly not a given.

        The values have a different range, and all parameters, in concert, will deliver different outcomes.

        It took me a while to get them closer, but I never thought that there would be the idea of using the same values. The parameters show different scales.

        After all, as with most things in visualization, it is physically plausible, not physically correct. It can come close, but speed is the main target, so far, I understand it.
        The oldest example would be Global illumination. To be physically correct, we would need a computer farm to really get one light source right, perhaps just for a second-long animation. My two cents.

        I can't see it as a bug. But feel free to get a second opinion from tech support. I'm always open to learning more if I miss something. Either way, thanks for asking; it made me think about it for a while today.

        All the best

        1 Reply Last reply Reply Quote
        • H
          history-growth last edited by

          Hi Dr Sassi!

          Just to clarify: I do not really expect to use the same values in Simulation System as in Bullet - but I do expect to find a stable simulation. I did several tests with different values, and Damping values - each time with sort of the same outcome where the sphere rapidly intersects the plane geometry and gained a lot of momentum.
          I'm trying to understand if this behaviour is expected - does Simulation System requires much more precise values for a RBD sim to work? Or is this a bug?

          1 Reply Last reply Reply Quote
          • Dr. Sassi
            Dr. Sassi last edited by

            Hi history-growth,

            If you would like me to look more closely into this, I need the file. Videos or images have a very limited option to show needed data. I'm happy to look into it.

            All the best

            1 Reply Last reply Reply Quote
            • H
              history-growth last edited by

              Please do! You can find this example here:
              https://www.dropbox.com/scl/fi/c8uib05tqfkh0z70g9sky/follow-position-RBD-vs-Simulation-System.c4d?rlkey=xpy40udf9q6ovykvu0ep85yqi&dl=0

              1 Reply Last reply Reply Quote
              • Dr. Sassi
                Dr. Sassi last edited by Dr. Sassi

                Thanks for the file, history-grow.

                Please check with tech support if you think the problem is a bug.
                https://www.maxon.net/en/support-center

                I can see that it doesn't work well with Spline-based animation in your example.

                Please have a look here:
                CV4_2024_drs_23_SIdy_01.c4d

                The results are different, as mentioned above, but I do not see a Keyframed animation of the problems from your file.

                Screen Shot 2023-12-11 at 10.48.04 PM.jpg

                All the best

                1 Reply Last reply Reply Quote
                • H
                  history-growth last edited by

                  Thank you!
                  Can it be that Simulation System is better suited for keyframed animations rather than the Align To Spline tag?
                  Perhaps I should report this to Maxon...

                  1 Reply Last reply Reply Quote
                  • Dr. Sassi
                    Dr. Sassi last edited by

                    Hi history growth,

                    I must check if procedural/expression-based animation is part of the "Force" evaluation.
                    I can write a report and see what the reply is like. The manual is very silent about that, and I haven't heard anyone specifying it during product meetings.
                    In case you like to voice your concerns as well:
                    https://www.maxon.net/en/support-center

                    All the best

                    1 Reply Last reply Reply Quote
                    • H
                      history-growth last edited by

                      Appreciated Dr Sassi!
                      Would be interesting to know the answer.

                      1 Reply Last reply Reply Quote
                      • Dr. Sassi
                        Dr. Sassi last edited by

                        This post is deleted!
                        1 Reply Last reply Reply Quote
                        • Dr. Sassi
                          Dr. Sassi last edited by Dr. Sassi

                          Hi history-growth,

                          As promised, I keep you informed.

                          Please try the freshly released 2024.2, which should take care of everything.

                          I tested your file and titled the planes to stress test it. I think it works very well now.

                          CV4_2024_drs_23_SIbl_01.c4d

                          Please let me know what you think.

                          All the best

                          1 Reply Last reply Reply Quote
                          • H
                            history-growth last edited by

                            Hi Dr Sassi. I came here to write the same thing.

                            2024.2 seems to have figured this out. Very good news!

                            Although I'm wondering if Sim System is stable enough for RBDs for production work if these kinds of bugs exists... only one way to find out.... 🙂

                            1 Reply Last reply Reply Quote
                            • Dr. Sassi
                              Dr. Sassi last edited by

                              Thanks for the reply, history-growth,

                              Have a lovely rest of the weekend.

                              1 Reply Last reply Reply Quote
                              • First post
                                Last post