hold_sync or equivalent for batch updating

I’d like to perform a batch update of the state of a scene graph. Display events are generated synchronously when you naively change properties (such as a transform). This causes a refresh for each change, which can get quite slow and results in “swimming” where partially updated state is displayed.

ipywidgets has a method called hold_sync, but using this with a K3D drawable doesn’t seem to help (e.g. I place all the objects I want to update in a Group, and use hold_sync on the group while making the changes). Assume that all meshes[] have been added to the group:

for i in range(50):
    with group.hold_sync():
        for j in range(len(meshes)):
            meshes[j].transform.translation = [0,float(i)/10,2*j]

I’m not that familiar with ipywidgets or how this should be done.

[hellcats]

My plan was to use K3D to visualize simulation and dynamical graphic data within a notebook environment (and not just to make movies). I was attracted to the ease of use and dynamic scene graph capabilities of K3D. I like that changing a property results in immediate display. But I also need to make big changes (like stepping forward a time step in a simulation with hundreds or thousands of objects), and for that I need some kind of batch capability. Maybe that isn’t a goal of K3D, or maybe the serialized/synchronous behavior is tied to the architecture of ipywidgets. I spent just a little time trying to read the code, but I don’t really understand the gestalt of the system. I’ll check back from time to time, but for now I’ll probably look for an alternative.

[hellcats]

@artur-trzesiok Thanks for looking into this issue - I am pretty happy with it now and will continue with my evaluation. I tried to follow the code changes you made to implement this, but I don’t know javascript or how ipywidgets works. I still have to learn a lot.