Can one animate the movement of objects to different layers in 2.55 yet?

After some Internet research, I had the impression that the ability to animate the movement of an object to different layers has not yet been ported from 2.49 to 2.55. (In other words: to make something appear/disappear from the layers in rendered view)

Does anyone know whether it is possible and, if so, how does one do it?


Keyframe the layer the object is on in the Object / Relations panel.

Thanks Richard!

I used to be on this forum quite a lot but have not been on it very regularly for the past year or two (mostly because I moved back to Engineering), but a quick two points:

  • I can’t remember seeing your name a lot before (unless you changed your Avatar), but you seem quite knowledgeable. Thanks for sharing it.
  • Did you do the Jing animation now - or was it done sometime earlier?


Did you do the Jing animation now - or was it done sometime earlier?

Just now. Only took a minute but shows the answer much clearer than writing it down. I wish more people would use it.

Thanks - definitely worth doing it. I’ll investigate Jing a bit.

I’m using it now and I have one problem: it seems that on the FIRST frame where a make the something visible, it is still at the original position (say zero). Only on the next frame it is on the proper place.

In my case it is a helmet that is parented to a bone, so it is way off the guy’s head the first time.

Do you know of any fixes for it? (Other than putting it as close to that position while on the other layer)

This is what I mean…

On the previous frame (the moment it becomes visible):

And on the immediate next frame (current frame):

I don’t think animating layers is working quite the same as in 2.49 at the moment because when you keyframe a layer change the object is on both layers as the same time at that frame. You might want to try having your objects on the same layer and animate visibility and renderability instead.

Thanks for those tips guys, I’ve been searching how to do this for a long time. (Since I’ve been using 2.5X)

it would be nice if someone with more knowledge could give some hints
about the order of the processing of animations to rendering.

As far i noticed the last step in a processing of a frame is not
the rendering process! Really?
The computation of drivers(from fcurves) seem to be after the “normal” animation
and this example -about the animation of layer- too is processed after
the animation part. That could be the reason why the object is
switched in time from one layer to another, but the parenting, was
already processed, when the object appears from the non-rendered
layer - and for the non-rendered layer it looks like blender skips
all objects (which make sense, … and speeds up everything, cause
the animations need not be calculated for objects on non-rendered
Animating the visibility (eye-icon) seems not to save animation-time,
the invisible object will always animated too.

The eye icon is only visibility in the viewport of the AI.

There probably is a limit there somewhere, the “system” lets you add a driver to just about anything… bake start and end frames for instance… to the point where it can get a bit nonsensical…

If you just want to animate visibility wouldn’t scale be the easiest solution?

Yes, I noticed that too. Another thing I noticed that the visibility on both layers do not necessarily follow the same (inverted) order when you step forward or backwards in your animation.

Thanks, I took your advice on the scaling: I merely scaled it down to zero on the awkward frame.