Keyframing armory properties

so now I can keyframe Armory properties in Blender, but upon building I get the following errors:

Exporting object action RecoaterAction.001
WARNING: The action “Build Plate 1Action” has fcurve channels with data paths that could not be resolved. This can be caused by the following things:

  • The data paths are not supported.
  • The action exists on both armature and non-armature objects or has both bone and object transform data.
    Unresolved data paths: {‘arm_propertylist[3].boolean_prop’, ‘arm_propertylist[1].boolean_prop’, ‘arm_propertylist[8].boolean_prop’, ‘arm_propertylist[5].boolean_prop’, ‘arm_propertylist[2].boolean_prop’, ‘arm_propertylist[4].boolean_prop’, ‘arm_propertylist[0].boolean_prop’, ‘arm_propertylist[9].boolean_prop’}

so it seems I can not use keyframes on armory properties and next make nodes using these values?

Hi, no you can’t use Blender keyframes on Armory properties. Only object Animation and armature bone animations support keyframes.

But in general, what you want to achieve could also be done only using logic nodes. What is it exactly that you like to achieve?

I’m making an animation of a machine. The (fixed) sequence of the machine is specified in a Microsoft Project file and next exported to csv file. So far, I have only worked with Blender making ‘static’ animations. For that purpose I would use a python script to read the csv file to transfer the positions and properties (e.g. constraints, visibility) to Blender.
Now I want to make the animations more interactive: the system should still run the pre-defined sequence (repeatable), but the user should be able to adapt camera positions and click on some covers to make them invisible. So I first tried to set Armory custom properties and keyframes but it seems they are not supported. So next options I see are:

  • read the complete csv file in Armory and generate timelines per periheral and property and use these (no idea how to implement but should be possible)
  • abuse the positions of dummy empties for timing of the properties (e.g. use Z position of an empty)

does that explain a bit what I’m trying to achieve?


I understand what you want to some extent. How about use the already existing animation as it is. But make some mesh in/visible when the user clicks on those. All the other animations would run as is. Just certain mesh parts become invisible with a checkbox or when clicked on using logic nodes.

For the camera movement, you could perhaps use a orbiter system using logic nodes.

Would this solution suffice or is it lacking in any aspects? A simple non-confidential example file would make things a lot easier to discuss…